the eye › public › books › biomed › hematopathology... · 2018-11-24 · editor domnita...

Molecular and Translational Medicine

Series EditorsWilliam B. ColemanGregory J. Tsongalis

For further volumes:http://www.springer.com/series/8176

Domnita CrisanEditor

Hematopathology

Genomic Mechanisms of Neoplastic Diseases

EditorDomnita CrisanWilliam Beaumont HospitalDepartment of Clinical PathologyMolecular Pathology Lab3601 West 13 Mile Road48073-6769 Royal Oak, [email protected]

ISBN 978-1-60761-261-2 e-ISBN 978-1-60761-262-9DOI 10.1007/978-1-60761-262-9Springer New York Dordrecht Heidelberg London

Library of Congress Control Number: 2010931434

© Springer Science+Business Media, LLC 2010All rights reserved. This work may not be translated or copied in whole or in part without the writtenpermission of the publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 SpringStreet, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarlyanalysis. Use in connection with any form of information storage and retrieval, electronic adaptation,computer software, or by similar or dissimilar methodology now known or hereafter developed isforbidden.The use in this publication of trade names, trademarks, service marks, and similar terms, even if they arenot identified as such, is not to be taken as an expression of opinion as to whether or not they are subjectto proprietary rights.While the advice and information in this book are believed to be true and accurate at the date of goingto press, neither the authors nor the editors nor the publisher can accept any legal responsibility for anyerrors or omissions that may be made. The publisher makes no warranty, express or implied, with respectto the material contained herein.

Printed on acid-free paper

Humana Press is part of Springer Science+Business Media (www.springer.com)

I dedicate this book to my mother who gave meher gene for optimism, to my brave father wholost his life fighting communists, and to myhusband, Dan, who offered his loving support,with patience and enthusiasm during work on thisbook and my entire professional career

Preface

Hematopathology: Genomic Mechanisms of Neoplastic Diseases in the bookseries Molecular and Translational Medicine addresses our current knowledge ofgenomics as applied to the pathogenesis, diagnosis, prognosis, monitoring, andtargeted therapy of hematologic malignancies. Hematology has been at the van-guard of the application of molecular technologies in diagnosis, classification, riskstratification, and use of molecularly defined therapeutic targets. These advances inmolecular technologies, diagnostics, and gene-related therapy have seen an extraor-dinary rapid pace since the completion of the Human Genome Project. Hematologyhas integrated the discoveries of genomic lesions underlying hematologic malignan-cies and applied the tools of molecular pathology, making them essential in clinicalpractice.

The scope of this book is to keep pathologists and clinicians abreast of therapid and complex changes in genomic medicine, as exemplified by the molecu-lar pathology of leukemias and lymphomas. This is a timely opportunity to not onlyupdate physicians on the complexity of genomic abnormalities but also offer anintegrated framework encompassing molecular diagnostics, the new WHO (WorldHealth Organization) classification of hematologic neoplasms with focus on molec-ular pathology, prognostic value of molecular tests, and molecular monitoring ofresponse to gene-targeted therapy.

The rapid pace of discovery, the explosion in genomic information, and the everchanging molecular technologies make it necessary to constantly update our knowl-edge and I hope that the readers will use this book as a practical resource and placeit next to their microscope, in their laboratories or clinical offices.

The first two chapters should be helpful for practicing pathologists and for clin-icians, providing overviews of molecular techniques and cytogenetics, both wellestablished and new, as used in molecular hematology. Chapter 3 is a concise reviewof the new 2008 WHO classification, which integrates molecular abnormalitiesin the diagnosis of hematologic neoplasms. The following chapters offer compre-hensive discussions of the molecular pathology of lymphoid and myeloid acuteleukemias, the mature B-cell and T-cell lymphomas, the myeloproliferation neo-plasms, chronic lymphocytic leukemia, overall representing the major diagnosticentities in neoplastic hematology.

vii

viii Preface

The new fields of targeted therapy in hematologic malignancies and microRNAsas applied in hematologic malignancies are reviewed in the last two chapters andoffer comprehensive discussions of the current state of these novel approaches.

I am extremely grateful to all the authors for their excellent contributions to thisbook; each chapter is an in-depth and thought-provoking update, as well as easilyreadable and practical.

In a specialty as exciting and rapidly evolving as Molecular Hematology, it ismy hope that this will be just the first of many editions of this book. It will beinteresting and challenging to see the progress in genomics in the next years and askthe question, Quo Vadis Hematology?

Royal Oak, Michigan Domnita Crisan

Contents

1 Molecular Techniques in Hematopathology . . . . . . . . . . . . . 1Bobby L. Boyanton Jr. and Jennifer R. Rushton

2 Classical and Molecular Cytogenetic Analysisof Hematolymphoid Disorders . . . . . . . . . . . . . . . . . . . . 39Mark A. Micale

3 Using Cytogenetic and Molecular Tests in DiagnosticWorkups with the WHO Classification – 2008 . . . . . . . . . . . 79Clarence C. Whitcomb

4 Update on the Molecular Pathologyof Precursor Lymphoid Leukemias . . . . . . . . . . . . . . . . . 103Robert B. Lorsbach

5 Molecular Pathology of Acute Myeloid Leukemias . . . . . . . . . 127Karen P. Mann and Debra F. Saxe

6 Molecular Pathology of Mature B-Cell and T-Cell Lymphomas . . 157Sophia L. Yohe, David W. Bahler, and Marsha C. Kinney

7 Molecular Pathology of Myeloproliferative Neoplasms . . . . . . 215David S. Bosler

8 Molecular Pathology of Chronic Lymphocytic Leukemia . . . . . 255Daniela Hoehn, L. Jeffrey Medeiros, and Sergej Konoplev

9 Targeted Therapy in Hematologic Malignancies . . . . . . . . . . 293Barbara Zehnbauer and Mona Nasser

10 Micro-RNAs in Hematologic Malignancies . . . . . . . . . . . . . 325Muller Fabbri and George A. Calin

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341

ix

Contributors

David W. Bahler, MD, PhD Department of Pathology, University of Utah, SaltLake City, UT, USA

David S. Bosler, MD Department of Clinical Pathology, Cleveland Clinic,Cleveland, OH, USA

Bobby L. Boyanton Jr, MD Department of Clinical Pathology, BeaumontHospitals, Royal Oak, MI, USA

George A. Calin, MD, PhD Departments of Experimental Therapeutics andCancer Genetics, University of Texas, M.D. Anderson Cancer Center, Houston,TX, USA

Muller Fabbri, MD Department of Molecular Virology, Immunology andMedical Genetics, The Ohio State University Comprehensive Cancer Center,Columbus, OH, USA

Daniela Hoehn, MD Department of Hematopathology, M.D. Anderson CancerCenter, Houston, TX, USA

Marsha C. Kinney, MD Division of Hematopathology, University of TexasHealth Sciences Center, San Antonio, TX, USA

Sergej Konoplev, MD, PhD Department of Hematopathology, M.D. AndersonCancer Center, Houston, TX, USA

Robert B. Lorsbach, MD, PhD Department of Pathology, University of Arkansasfor Medical Sciences, Little Rock, AR, USA

Karen P. Mann, MD, PhD Department of Pathology and Laboratory Medicine,Emory University, Atlanta, GA, USA

L. Jeffrey Medeiros, MD Department of Hematopathology, M.D. AndersonCancer Center, Houston, TX, USA

Mark A. Micale, PhD Beaumont Laboratory, Department of AnatomicPathology, Beaumont Hospitals, Royal Oak, MI, USA

xi

xii Contributors

Mona Nasser, MD Department of Clinical Chemistry, School of Medicine, BeniSuef University, Beni Suef, Egypt

Jennifer R. Rushton, MD Department of Pathology, Baylor College ofMedicine – BCM 315, Houston, TX, USA

Debra F. Saxe, PhD Department of Pathology and Laboratory Medicine, EmoryUniversity, Atlanta, GA, USA

Clarence C. Whitcomb, MD Department of Pathology, Miller School ofMedicine, University of Miami, Miami, FL, USA

Sophia L. Yohe, MD Department of Laboratory Medicine and Pathology,University of Minnesota Medical Center, Minneapolis, MN, USA

Barbara Zehnbauer, PhD Division of Laboratory Systems, Laboratory PracticeEvaluation and Genomics Branch, Centers for Disease Control and Prevention,Atlanta, GA, USA

Chapter 1Molecular Techniques in Hematopathology

Bobby L. Boyanton Jr. and Jennifer R. Rushton

Keywords DNA · RNA · Specimen collection · Specimen handling · Speci-men processing · Cell enrichment · Nucleic acid · Stability · Storage · Spectro-photometric · Fluorometric · Absorbance · Asymmetric PCR · Clonality ·Immunoglobulin · T-cell receptor · Antigen receptor · Gene · Hematology ·Hematolymphoid · Hematopathology · Paraffin · Formalin · Fixative ·Extraction · Purification · Phenol–chloroform · Chaotropic salt · Silica col-umn · Ethidium bromide · SYBR green · Gel electrophoresis · Capillaryelectrophoresis · Agarose · Polyacrylamide · Restriction enzyme · Sangersequencing · Chain termination · Pyrosequencing · Sequencing by synthesis · Next-generation sequencing · High-throughput sequencing · Automation · Polymerasechain reaction · PCR · Reverse transcriptase PCR · Allele-specificPCR · Nested PCR · Real-time PCR · Quantitative PCR · MethylationPCR · FRET · TaqMan · Probe · Hydrolysis · Hybridization

Introduction

The discipline of hematopathology traditionally relies upon morphologic evaluation,cytochemical stains, immunohistochemistry, flow cytometry, and karyotypic analy-sis to classify hematolymphoid neoplasms. Although these time-honored methodsstill comprise the primary diagnostic arsenal of the pathologist, the last few decadeshave borne witness to the widespread acceptance of molecular techniques to clas-sify these neoplasms. No longer considered ancillary, molecular analyses have ledto a greater understanding of the biological and clinical heterogeneity of hema-tolymphoid neoplasms, and now form the primary diagnostic criteria for manydiagnoses as set forth by the World Health Organization [1]. They also provideextremely sensitive and specific methods for prognostic marker detection and mini-mal residual disease monitoring. These techniques have evolved rapidly over the last

B.L. Boyanton Jr. (B)Department of Clinical Pathology, Beaumont Hospitals, 3601 W. Thirteen Mile Rd, Royal Oak,MI 48073, USAe-mail: [email protected]

1D. Crisan (ed.), Hematopathology, Molecular and Translational Medicine,DOI 10.1007/978-1-60761-262-9_1, C© Springer Science+Business Media, LLC 2010

2 B.L. Boyanton Jr. and J.R. Rushton

decade from Southern blot and hybridization assays to polymerase chain reactionand its variants to gene expression profiling and single-nucleotide polymorphismanalysis, and more recently to microarray technology and whole-genome analysis.Despite technological advancements, molecular techniques are critically dependentupon the nature of nucleic acids retrieved from the specimen. Results cannot becorrectly interpreted if the quantity and/or the integrity of nucleic acids are not opti-mal for the desired molecular application. As such, the purpose of this chapter istwofold. First, issues pertaining to specimen collection, handling and processing,and nucleic acid extraction, stability, and storage are reviewed. Second, moleculartechniques commonly utilized in hematopathology are reviewed. Cytogenetics, flu-orescent in situ hybridization (FISH), and microarray techniques are discussed inChapter 2.

Part I: Specimen Collection and Processing

Standard Precautions and Safety

The collection, processing, and storage of biological samples pose risks to the han-dler for the acquisition of a variety of infectious agents. All personnel handlingbiological samples should follow “standard precautions”; guidelines are avail-able from the US Centers for Disease Control and Prevention (www.cdc.gov).Additionally, the Clinical and Laboratory Standards Institute (www.clsi.org) pub-lishes literature pertaining to laboratory safety [2] and the protection of laboratoryworkers from occupationally acquired infections [3].

Patient Identification and Labeling

Specimen labeling and tracking throughout the entire testing process is paramountto ensure valid test results. Unique patient identifiers should be utilized (i.e.,full name, date of birth, medical record number). In addition, the test requisi-tion should also include (1) date and time of specimen collection, (2) specimentype and source, (3) ordering physician and contact information, (4) billing infor-mation, and (5) pertinent clinical and laboratory information. A copy of thepathology report should be included with tissue specimens to ensure accurate spec-imen identification and the ability to correlate molecular-based test results withthe histopathologic diagnosis. Every attempt should be made to obtain stainedslides for review. This will ensure that the correct tissue is submitted and thatthe tissue is representative of the intended test, and will allow for the qualitativeassessment of cellularity. Compliance with regulations protecting personal healthinformation as set forth by the US Department of Health & Human Services(www.hhs.gov/ocr/hippa) is of paramount importance and must be adhered to at alltimes.

1 Molecular Techniques in Hematopathology 3

Cell Enrichment and Selection Techniques

The ability to selectively obtain desired cell populations increases the sensitivity andspecificity of molecular-based testing and facilitates the removal of potential con-taminating substances that may be inhibitory to amplification-based methods, suchas polymerase chain reaction (PCR). This becomes vitally important with minimalresidual disease (MRD) testing, where it is not uncommon for residual malignantcells to represent a minor fraction of the cellular milieu. Basic techniques of cellselection and enrichment are discussed in this section.

The isolation of DNA or RNA from select cell populations within liquid spec-imens (i.e., whole blood, marrow aspirates, body fluids) can be accomplishedby several techniques. Perhaps the most basic approach to obtain leukocytesinvolves the preparation of a leukocyte-rich layer [4]. Following centrifugation(3,300–3,500×g for 10–15 min), the specimen will partition into three distinctlayers: an upper aqueous layer, a middle leukocyte-rich layer (buffy coat), and alower layer (erythrocytes). The “buffy coat” is easily recovered following removalof the upper layer. Another approach is selective erythrocyte lysis with hypotonicbuffer (e.g., ammonium chloride) [5]. After centrifugation, the released hemoglobinwill partition into the upper aqueous layer, while the nucleic acid of interest isretained within the leukocyte pellet at the bottom of the tube. Following decant-ing of the aqueous layer, the leukocyte pellet is washed several times with isotonicbuffer to remove any residual aqueous layer. Alternatively, density-gradient cen-trifugation facilitates the selective recovery of lymphocytes and monocytes fromother cellular constituents [6]. Several commercial products incorporating Ficoll-Hypaque or other density-gradient media into evacuated collection tubes specificallyfor molecular-based testing are available – Vacutainer CPT Mononuclear CellPreparation Tube (Becton–Dickinson, Franklin Lakes, NJ). A more recent approachuses antibody-coated magnetic beads to obtain desired cell populations from liq-uid specimens. After incubation of the liquid specimen with the magnetic beads,a magnetic field is applied, allowing unwanted cellular constituents to be removedby decanting. Magnetic bead-bound cells of interest are then washed with isotonicbuffer. Cells of interest are released from the magnetic beads by either enzymaticcleavage or competitive displacement using high-affinity monoclonal antibodies [7].Alternatively, fluorescent antibody cell sorting (FACS) is also useful for captur-ing selected cell populations. Fluorescent, differentially labeled antibodies bind todesired cellular constituents. Using a modified flow cytometer, cells of interest arerouted to separate collection chambers based upon their fluorescence profile. Thecells of interest are thusly obtained and ready for nucleic acid extraction.

Laser capture microdissection is a common technique that facilitates the selec-tion of desired cell populations from tissue sections. In short, tissue is mounted on aglass slide and covered with a translucent coating. Using a microscope, cells of inter-est are located, followed by user-defined infrared or UV laser activation that meltsthe translucent coating containing the cell(s) of interest from the slide. The selectedregions of dissected film are removed, followed by routine nucleic acid extractionprotocols [7]. Commercial systems are available from Arcturus/MDS Analytical


Technologies (Mountain View, CA), Carl Zeiss (Thornwood, NY), MolecularMachines and Industries (Knoxville, TN), and PALM Microlaser Technologies(Bernried, Germany).

Source-Specific Requirements to Ensure Nucleic Acid Integrity

The ability to reliably detect and/or quantify results for molecular-based testingrelies predominantly upon decisions made at the time of specimen collection. Thisis of paramount importance in dealing with hematolymphoid disorders where highlylabile messenger RNA (mRNA) transcripts are commonly the intended targets formolecular-based testing. The following subsections provide general guidelines forthe collection and handling of specimens pertaining to hematolymphoid disordersarising in various locations.

Bone Marrow Aspirates, Whole Blood, and Body Fluids

Ethylenediaminetetraacetic acid (EDTA) is the most commonly used anticoagulantfor molecular-based testing of hematolymphoid disorders, although acid–citrate–dextrose (ACD) is an acceptable alternative. Heparin should be avoided as itinterferes with the polymerase chain reaction [8–10], if not completely removed dur-ing subsequent extraction and purification processes. In general, body fluids are notcollected in an anticoagulant container but are commonly contaminated with ery-throcytes. Prior to DNA extraction, specimens may be temporarily stored at roomor refrigerated temperature (2–8◦C) for up to 24 or 72 h, respectively, without signif-icant DNA degradation [7]. If delays in testing are unavoidable, erythrocytes shouldbe removed prior to storage at –20◦C [7], as hemoglobin is inhibitory to PCR [11]and is readily released from erythrocytes upon thawing frozen marrow aspirates orwhole blood.

For RNA analysis, marrow aspirates and whole blood should be collecteddirectly into tubes containing an RNA stabilization agent. The PAXgene seriesof RNA stabilization tubes [PreAnalytiX; joint venture, Qiagen (Valencia, CA)and Becton–Dickinson (Franklin Lake, NJ)] is widely used and has been shownto reliably maintain RNA integrity [12–16]. Other collection systems are avail-able from Ambion (Austin, TX), Applied Biosystems (Foster City, CA), Promega(Madison, WI), Invitrogen (Carlsbad, CA), Zymo Research (Orange, CA), andGentra Systems (Plymouth, MN). If specialized collection tubes are not utilized,EDTA-anticoagulated specimens and body fluids should be placed on wet ice andimmediately transported to the laboratory. RNA extraction should take place nolonger than 4 h after collection. As with DNA, if RNA extraction cannot occurin a timely manner, the erythrocytes should be removed and then the sample frozenat –20◦C or lower [7]. Failure to comply with these recommendations may lead toerroneous test results due to either RNA degradation or altered regulation of geneexpression [13].


Dried Blood Spots (Guthrie Cards)

Dried blood spots, which are routinely collected for newborn screening programs,are an invaluable source of material for retrospective epidemiological and diagnos-tic studies. They have been pivotal in confirming the prenatal origin of leukemiain infants and young children [17–20]. To reliably obtain nucleic acid from driedblood spots, the specimen must be thoroughly air-dried and placed in a desiccant-containing sealed bag to prevent moisture accumulation and microbial growth.Samples must be placed in separate sealed bags to prevent cross-contamination [7].Dried blood spots should be maintained at –20◦C to maintain optimal nucleic acidrecovery and integrity [21].

Fresh Tissue

It is commonplace for laboratories to receive fresh tissue following or during sur-gical procedures. This facilitates the pathologist’s ability to assess the surgicalspecimen and appropriately direct intra-operative patient management, and is apivotal point from which fresh tissue can be triaged for molecular-based testing.Because one cannot always predict the future downstream testing methodologies,it is prudent to handle all fresh tissue with the mindset that RNA studies will berequired. In that regard, fresh tissue should be “snap frozen” in liquid nitrogen priorto storage at –70◦C [7]. If unavailable, the sample should be placed immediatelyon wet ice and/or in an RNA stabilization buffer, with RNA extraction taking placewithin 4 h of collection [7]. Snap frozen fresh tissue should be maintained on dryice during transportation to the processing facility. Gloves should be worn at alltimes when handling specimens, reagents, and equipment as RNases, and to a lesserextent DNases, are ubiquitous and readily present on skin. Furthermore, reagentsand equipment should be chemically treated to destroy RNase activity.

Fixed, Paraffin-Embedded Tissue

Tissue fixation and embedding has profound effects on the quality and yield ofnucleic acids that can be recovered from tissue. The extent of nucleic acid degra-dation that invariably results from fixation depends on the type of fixative, theduration of fixation, the size of the specimen and its permeability to the fixative,the degree of tissue hypoxia as determined by the time between surgical removaland fixation, and the length of storage in paraffin blocks. Degradation of nucleicacids in fixed tissues is primarily due to the cross-linking of proteins and DNA,especially for formalin-based fixatives [22–24]. This two-step process consists ofan initial, reversible reaction whereby formaldehyde induces hydroxymethylationof the amino and imino groups of nucleic acid bases; the second step involvesmethylene bridge formation between bases over the course of several days. Bothreactions are temperature dependent. Hydrogen bond disruption between base pairsis proportional to temperature so that at denaturing temperatures (>90◦C), ssDNApredominates. Formaldehyde reacts quickly with ssDNA via hydroxymethylation,preventing reannealing when the temperature is subsequently lowered. Due to the


high temperatures reached during the paraffin-embedding process, ssDNA wouldbe present in formalin-fixed, paraffin-embedded tissue that would impair restric-tion endonuclease recognition sites. The chemical-induced protein-to-nucleic-acidcross-linking is more pronounced with extended fixation time, making it neces-sary to prolong the proteinase K digestion step when dealing with formalin-fixedtissues. Acid depurination of DNA is an additional mechanism of DNA degrada-tion, if neutralization of formaldehyde is insufficient or if unbuffered formalin isused. Despite these limitations, neutral buffered formalin (NBF) still remains thebest formalin-based fixative for DNA extraction and has the advantage of preserv-ing DNA methylation patterns. Compared to NBF, extensive DNA degradation isseen with fixatives containing heavy metals, such as mercuric chloride and formalin(B-5), and dichromate and acetic acid (Zenker’s, pH 2.0) [22, 23]. Other acidfixatives, such as Bouin’s and Hollande’s also result in extensive nucleic acid degra-dation due to acid depurination [24, 25]. The chemical effects of mercury in B-5and chromium in Zenker’s are due to the formation of nucleoprotein complexeswith phosphoric acid residues and thiol groups. In fixed tissue, these complexes pro-mote resistance to digestion, and the inability to obtain extracted DNA of sufficientquality. Use of other cross-linking fixatives (i.e., glutaraldehyde and paraformalde-hyde) also results in variable degrees of DNA degradation [26, 27]. Utilization ofacidic decalcifying solutions for bone and bone marrow core biopsies also promotesnucleic acid degradation. In contrast, precipitation fixatives (i.e., ethanol, methanol,acetone) preserve nucleic acids quite well and allow the extraction of good-qualityDNA and RNA [23, 24, 28–32]. New commercially available, non-formalin-basedfixatives include Histochoice (Amersco, Inc., Solon, OH), HOPE (DCS Innovative,Hamburg, Germany), UMFIX (Sakura FineTek USA, Inc., Torrance, CA), Prefer(Anatech LTD, Battle Creek, MI), and FineFix (Milestone, Bergamo, Italy). Thesefixatives, with the exception of Prefer [33], appear to produce good-quality nucleicacid extracts [34–40].

The duration of formalin fixation is critical for the quality and yield ofnucleic acid extracts, with 12–24 h generally considered the optimal fixation time.Prolonged fixation leads to poor-quality nucleic acid extracts [22, 24, 41, 42]. Incontrast, alcohol-based fixatives allow for good-quality nucleic acid extracts, irre-spective of the duration of fixation. The effect of prolonged storage of archival fixedparaffin-embedded tissues on the quality and quantity of nucleic acid extracts is lesswell defined, but it appears that the molecular weight of extracted DNA decreaseswith storage beyond 2 years; however, DNA has been successfully extracted fromarchival paraffin blocks greater than 20 years old [43].

Part II: Nucleic Acid Extraction, Purification, and Storage

Overview

Prior to molecular-based testing, nucleic acids must be retrieved from the clinicalspecimen, by any number of manual and automated methodologies. In general, the


first step of this process involves cell lysis, which is usually accomplished withdetergents (e.g., Tween 20, sodium dodecyl sulfate) [22, 24]. The initial step shouldbe modified based upon the specimen type. Tissues (fresh, frozen, cultured cells)have a supporting connective tissue stroma, which requires proteolytic digestion tofacilitate cell lysis. Paraffin-embedded tissue should be deparaffinized prior to diges-tion, which can be accomplished by heat and/or a combination of solvent-basedreagents (alcohol and xylene) [22, 24]. The next step involves removal of proteinsfrom the lysate either by enzymatic digestion or by selective precipitation by adjust-ing the salt concentration of the lysate. Enzymatic digestion is commonly performedwith proteinase K at 56◦C – the optimal temperature for enzymatic activity [24]. Thelength of proteinase K incubation depends upon the amount of tissue being digestedand the pH of the buffering solution, although overnight incubation is usually suf-ficient. After tissue digestion, proteinase K should be inactivated by heating thesolution to 95◦C for 10–15 min. The next step involves the selective extraction ofnucleic acids from the cellular lysate via a number of organic or chaotropic salt–silica column-based methods. Finally, the purified nucleic acid is precipitated into asalt buffer (e.g., Tris–EDTA) prior to analysis or storage.

Extraction Techniques

Organic (Phenol–Chloroform)

This gold standard nucleic acid extraction method involves the hazardous hydro-carbon phenol–chloroform. In brief, the chemical properties of nucleic acids andproteins promote differential migration into the aqueous and organic phases, respec-tively [22, 44]. The pH of this biphasic solution is critical and maintenance withinthe narrow range of 7.0–8.0 is of paramount importance. Within this pH range,nucleic acids remain in the aqueous phase, while other non-essential molecules (i.e.,proteins, detergents, etc.) remain in the organic phase [22, 44]. If the solution isslightly acidified, the negatively charged phosphate groups of DNA will be prefer-entially neutralized by excess hydrogen ions, facilitating DNA migration into theorganic phase and thusly the selective extraction of RNA from the aqueous phase[44]. The addition of isoamyl alcohol to the solution also facilitates the retention ofRNA with long poly-A tracts in the aqueous phase [22]. A crucial step in organicextraction is adequate mixing of the organic and aqueous phases to allow appropriatepartitioning of the suspended molecules. The organic phase should be removed witha sterile pipette and appropriately discarded into an organic solvent disposal system.Extraction should be repeated by adding fresh phenol–chloroform, followed by thor-ough mixing until all visible protein is removed from the organic–aqueous interface[22]. The final steps involve nucleic acid precipitation from the aqueous solution,using cold ethanol and monovalent cations at 0◦C [22]. The aqueous solution shouldbe cautiously removed with a sterile pipette to avoid disrupting the nucleic acid pre-cipitate. Nucleic acid should be judiciously exposed to air to facilitate completeevaporation of the ethanol. The nucleic acid pellet should be resuspended in anappropriate buffer (e.g., Tris–EDTA).


If RNA is desired, special considerations must be followed throughout the extrac-tion process. Gloves should be worn at all times, since skin is a common sourceof RNases. All equipment and reagents must be treated with diethylpyrocarbonate(DEPC) to degrade all nuclease activity. Additionally, equipment and reagents andthe designated workstation should be dedicated solely for RNA extraction. DEPC-treated reagents and equipment should be autoclaved to inactivate the DEPC toprevent the carboxymethylation of nucleic acid [22]. A modified version of thephenol–chloroform method involves the addition of guanidine isothiocyanate – so-called TRIzol-LS R© (Invitrogen, Carlsbad, CA). Guanidine isothiocyanate (GITC)is a powerful protein denaturant and is extremely effective at eliminating RNaseactivity.

Inorganic (Chaotropic Salt–Silica Column)

Inorganic extraction is a great alternative to organic extraction, circumventing theneed to handle and dispose of hazardous chemicals. Additional advantages includecommercial availability, enhanced reagent stability, and reduced waste. Commercialkits employ the principles of anion-exchange chromatography, salt precipitation,and silica adsorption [22]. Perhaps the most popular format, because of its easeof use and extraction efficiency, is the chaotropic salt–silica column. GITC, achaotropic salt, not only inactivates nucleases but also facilitates nucleic acid bind-ing to the silica column. The silica-bound nucleic acid is purified by several washingsteps that remove contaminating proteins, lipids, and other non-essential molecules.Finally, nucleic acid is eluted from the column with a low salt concentrationbuffer [22].

Stability of DNA in Storage

Depending upon the temporal relationship between purification and analysis, thechoice of diluent may dramatically impact the integrity of the DNA sample. Distilledwater will promote spontaneous separation of dsDNA and concomitant degradationvia residual nuclease activity [45]. Tris–EDTA buffer is the most commonly utilizedstorage diluent for several reasons. First, EDTA chelates divalent cations, whichare necessary for nuclease activity. Second, the ionic concentration of the sodiumsalt facilitates DNA helix stabilization and prevents spontaneous strand separation[45]. Purified DNA can be safely stored in Tris–EDTA for up to 26 weeks at roomtemperature, at least for 1 year at refrigerated temperature (2 to 8◦C) [7], and atleast for 7 years at –20◦C or lower [7, 46, 47]. The choice of storage tube is alsocritical as standard polypropylene and polyethylene tubes bind DNA [7], therefore,specifically engineered polypropylene tubes (polyallomer) should be utilized [7].

Stability of RNA in Storage

RNA is extremely labile and degradation and/or altered gene expression beginsimmediately following specimen collection [13]; therefore, prudent oversight of


tissue handling and processing (as previously described) is of paramount impor-tance. Once purified, RNA should be stored as an ethanol precipitate at –70◦C orlower [7, 45], since RNases still retain activity at –20◦C [7]. Tubes and diluentsfor RNA storage should be nuclease free, and gloves should be worn at all times toprevent RNA degradation due to RNases [45].

Considerations for Long-Term Storage

To maintain the integrity of purified nucleic acid during long-term storage, samplesmust be stored in an appropriate buffer and at the correct temperature. Repeatedfreeze–thaw cycles will compromise nucleic acid integrity and should be avoided orminimized [7, 24, 45]. Storage freezers should not be “frost free” to prevent repeatedfreeze–thaw cycles [7]. The work of Schaudien et al. [48] has demonstrated that real-time PCR performed on purified DNA stored in 50% glycerol retains reproducibleresults even after 16 freeze–thaw cycles, an observation that supports glycerol as analternative method to preserve nucleic acid integrity.

Part III: Assessment of Nucleic Acid Quality and Quantity

Nucleic acid quantity and purity may be assessed using spectrophotometric or fluo-rometric methods. Spectrophotometers measure the absorbance of ultraviolet (UV)light. The absorbance maximum of nucleic acid and protein is 260 nm (A260)and 280 nm (A280), respectively. Quantifying nucleic acids in solution can beaccomplished by obtaining the A260 measurement. The purity of the solution maybe inferred by calculating the A260:A280 ratio. Pure DNA has an A260:A280 ratio of1.8, while pure RNA has an A260:A280 ratio of 2.0 [44]. An A260:A280 ratio lowerthan 1.8 indicates the presence of contaminants [44], which may interfere withdownstream applications. The accuracy of the A260:A280 ratio is dependent on thepH and the ionic strength of the solution. With increasing pH, the A280 decreases,while the A260 remains unaffected, causing a spuriously increased A260:A280 ratio[49]. Water is mildly acidic, which results in spurious lowering of the A260:A280ratio. As a result, buffered solutions with slightly alkaline pH (e.g., Tris–EDTA,pH 8.0) should be used as diluents and serve as a blank for spectrophotometricmeasurements. Fluorometric methods, on the other hand, use fluorescent dyes thatintercalate into dsDNA, are relatively insensitive to non-nucleic acid contaminants[50, 51], and provide more accurate quantitation as compared to spectrophotometricmethods, especially with lower nucleic acid concentrations. When dealing with verylow quantities, neither spectrophotometric nor fluorometric methods can accuratelyquantify nucleic acids. Other methods, such as elemental analysis and traceablephosphorus, have circumvented this issue and are discussed elsewhere [52–54].Commercially available systems for routine clinical work, developed by NanodropTechnologies (Wilmington, DE), offer the ability to quantify nucleic acid using only1 μL of purified sample. The ND-1000 and ND-8000 spectrophotometers facilitate


the testing of one and eight samples, respectively, at a time and can accurately quan-tify nucleic acid in the nanogram per microliter range. The fluorometric ND-3300 isa more sensitive, low-throughput option that can quantify nucleic acids down to thepicogram per microliter range.

The previously discussed methodologies can quantify and assess nucleic acidpurity; however, they are unable to assess quality in terms of molecular weight. Abasic option is the electrophoresis of purified nucleic acid in an ethidium bromide-stained agarose gel. High-quality, non-fragmented DNA will form a solitary bandnear the application well, while degraded DNA will appear as a smear through-out the lane. RNA integrity may be assessed by the electrophoresis of RNA ona denaturing ethidium bromide-stained agarose gel. Two distinct bands should bevisualized, corresponding to 28S and 18S ribosomal RNA. The 28S rRNA bandshould be about twice the intensity of the 18S rRNA band. As RNA degrades, theintensity of the 28S and 18S rRNA bands will correspondingly decrease and, analo-gous to DNA, appear as a smear throughout the lane (see Fig. 1.1). Intercalatingdyes (i.e., SYBR Green, PicoGreen) are replacing ethidium bromide due to itshazardous and mutagenic properties. Novel methods like the Agilent 2100 bioan-alyzer (Agilent Technologies, Santa Clara, CA) use a combination of fluorescentdyes, capillary electrophoresis, and microfluidics technology to simultaneouslyassess the concentration and integrity of nucleic acids. In short, purified nucleicacids migrate through a microfluidics chip and bind to intercalating dyes, with thefluorescence signal being measured as each molecule passes through the detec-tion system. The outcome is a summary of variously sized molecules and theircorresponding peak heights, reflecting nucleic acid integrity and concentration,respectively [55].

Part IV: Selected Techniques

Electrophoresis

Electrophoresis is the process by which molecules under the influence of an electri-cal field are differentially separated within a liquid or a solid matrix. The differentialseparation of molecules is based upon many factors, including the size of eachmolecule and its three-dimensional conformation, the net charge of the molecule(as dictated by pH), the pore size of the matrix being utilized, and the amount ofelectrical current utilized [56]. Because nucleic acids are negatively charged, theywill migrate toward the positive electrode (anode). The degree of migration towardthe anode is based largely upon the size of the nucleic acid molecules and thematrix pore size, while the speed at which migration occurs is primarily reflected bythe amount of electrical current applied and the matrix pore size. The compositionand concentration of the matrix dictates the pore size; the mobility of nucleic acidswithin the matrix is inversely proportional to the log of the pore size [44]. Therefore,large molecules will demonstrate limited migration and will remain closer to the


Fig. 1.1 Qualitative assessment of nucleic acid integrity by gel electrophoresis. Ethidium bro-mide stained, 2% agarose gel following electrophoresis. (Lanes 1 and 6) intentionally left blank;(lane 2) molecular weight size markers (100, 200, 300, 400, 500, 525, 700, 1,000 bp); (lane 3)genomic DNA (lambda phage DNA – Catalog # 25250-010, Invitrogen, Carlsbad, CA), demon-strating a single band (approximately 48,500 bp) near the application well, signifying the recoveryof intact, high molecular weight DNA following extraction; (lane 4) human RNA, demonstratingtwo distinct bands (28S and 18S ribosomal RNA), signifying the recovery of intact RNA followingextraction; (lane 5) degraded human RNA, demonstrating residual 28S and 18S bands and “smear-ing” of degraded RNA throughout the entire lane, representing variously sized RNA fragments.Degraded DNA would demonstrate a similar “smearing” pattern throughout the lane

negative electrode (cathode), while smaller molecules will migrate further to thepositive electrode (anode).

Agarose gel is the primary matrix utilized in clinical molecular laboratories forthe electrophoretic separation of nucleic acids, and is formed by dissolving agarosegel powder into boiling electrophoresis buffer solution, followed by pouring into acasting tray for solidification. Ethidium bromide (EtBr), an intercalating dye usedto facilitate nucleic acid visualization following ultraviolet light exposure, is usuallyadded prior to pouring the liquid solution into the casting tray. Alternatively afterelectrophoresis, agarose gels can be submerged into a solution of EtBr to accom-plish the same end result. Agarose gels with concentrations around 1% are typically


utilized to resolve nucleic acid fragments in the range of 1–20 kb because of the rel-atively large pore size created. Higher concentrations of agarose (2–3%) decreasethe pore size and are able to accurately resolve nucleic acid fragments between 100and 2,000 bp. Polyacrylamide gels consist of polymerized acrylamide monomersthat form small pores, facilitating the resolution of nucleic acid fragments in therange of 100–1,000 bp. Polyacrylamide gels are less commonly utilized due to theirfragile nature and serious risks to laboratory personnel – acrylamide monomers arerespiratory irritants and neurotoxic.

Capillary electrophoresis (CE) is a separation technique whereby proteins,nucleic acids, and other analytes are differentially separated and analyzed inthe interior of a small caliber capillary. Commercially available CE systems arereadily available with varying configurations to accommodate the various needsof molecular diagnostic laboratories. In contrast to conventional electrophoresis,CE offers numerous advantages, including standardized protocols, ease of use,increased efficiency of workforce utilization, error reduction, increased throughput,and automation.

In brief, CE separates analytes (i.e., nucleic acids, proteins, etc.) within a low-viscosity, electrolyte-containing liquid polymer that functions as a sieving matrixand facilitates current conduction within the capillary. Capillary tubes range from25 to 100 cm in length and are approximately 50 μm in diameter [44]. Capillariesare constructed of glass (silica) and externally coated with a polymer for stabil-ity. The internal capillary wall consists of neutral silanol (Si–OH) groups thatmust be ionized to negatively charged silanolate (Si–O–) groups prior to use. Thisis usually accomplished by first priming the capillary with a basic solution ofsodium hydroxide or potassium hydroxide. When the low-viscosity liquid polymeris injected into the capillary and electrical current applied, electrolytes within theliquid polymer flow from the injection site to the opposing end where signal detec-tion occurs. The sample is electrokinetically injected into the capillary, wherebythe concentration of the low-viscosity polymer establishes the sieving matrix anddifferentially sized molecules are electrophoretically separated as they too movefrom injection site to the detection end. At the detection end of the capillary, asmall portion of the external stability polymer is absent – so-called detection win-dow. The detection window is optically aligned between a laser source (argon ordiode) and a charge-coupled device (CCD) camera or filter wheel and photomul-tiplier tube (PMT). As analytes electrophoretically separate within the capillary,they are temporally detected as they pass by the detection window. In contrast toconventional gel electrophoresis, CE is more sensitive to DNA concentration andcontaminants. When relatively large quantities of DNA of a particular molecu-lar weight pass through the detection window, the signal intensity can overwhelmthe detection system and generate a high-amplitude primary peak and a secondadjacent lower amplitude peak (usually 1 bp greater in size) – so-called “shadowpeak.” Additionally, high-amplitude peaks may be spuriously detected in more thanone of the detection channels due to failure of the color compensation system tocompletely eliminate spectral overlap from the various fluorescent dyes that areincorporated into the DNA fragments. Other unwanted charged species that enter


into the capillary can interfere with the electrophoretic mobility of nucleic acidsand/or alter the intensity of the fluorescent signal. It is therefore commonplace toinclude a post-amplification purification step prior to electrokinetic injection. Thepurification step not only reduces contaminants but also incorporates formamideinto the loading buffer that stabilizes DNA and optimizes capillary electrophoreticresolution.

In the clinical molecular laboratory, CE is primarily utilized for DNA sequencingand DNA fragment sizing; however it may be used for quantitative purposes sincethe detected signal intensity is directly proportional to the amount of each frag-ment passing by the detection window. Current applications of CE to the practiceof hematopathology include dideoxynucleotide chain termination DNA sequencing[57], evaluation of immunoglobulin heavy chain and T-cell receptor gene rearrange-ments for clonality assessment [58], fragment size analysis for the detection of smallgene insertions and duplications, such as those characteristic of NPM1 and FLT3genes in acute myeloid leukemia (AML) [59], and BCR–ABL fusion transcript sizeanalysis for the discrimination of the three common fusion transcripts encounteredin chronic myeloid leukemia (CML) [60].

Restriction Enzymes

Restriction endonucleases (REs) are enzymes that cleave double-stranded DNA(dsDNA) at specific nucleotide recognition sites or restriction sites. These enzymesare ubiquitous in bacteria and are thought to have evolved as a defense mecha-nism facilitating the degradation of foreign DNA [61]. Specific methyltransferaseenzymes chemically modify, via a process termed methylation, recognition sites,thereby protecting microbial DNA from its own degradation. REs are generallycategorized into three classes based upon their target sequence, enzyme cofactorrequirements, and the position of their DNA cleavage site relative to the targetsequence. The majority of REs utilized in clinical molecular laboratories are class IIand detailed discussion of these classes is beyond the scope of this chapter. In brief,class II REs require only the presence of magnesium (Mg2+), are usually palin-dromic in nature, recognize very small DNA lengths (usually 4–8 bases), and cleavedsDNA at or near the restriction site resulting in “blunt” or “sticky” ends [56]. Sincetheir discovery in the early 1970s, thousands of REs have been characterized and aplethora of these are commercially available [62]. REs were crucial in the develop-ment of recombinant DNA technology, including the mass production of proinsulin[63], and have been widely adopted into the clinical molecular laboratory due totheir unique ability to confirm the presence of desired PCR amplification productsby knowing the predicted size of individual DNA fragments following digestion.These unique recognition sites occur at variable frequency throughout a given DNAsequence. Therefore, DNA digestion using different restriction enzymes will resultin a unique pattern of DNA fragments upon separation. Furthermore, a change inrestriction enzyme digestion pattern may result if mutations or polymorphisms occurat these recognition sites. A recognition site may be created or destroyed as a result


of such a sequence variation. The creation of a new recognition site results in twosmaller fragments in a sample with a mutation, whereas the destruction of a recog-nition site results in a larger DNA fragment in a sample with a mutation. These sizedifferences can be detected by standard gel electrophoresis or CE. These changes infragment pattern have been utilized in neoplastic hematopathology for the detectionof a variety of point mutations, including FLT3 D835 mutation in AML and JAK2V617F mutation in myeloproliferative neoplasms (MPN) [64, 65].

DNA Sequencing

Sequencing refers to a set of analytical methods to determine the sequence or theorder of nucleotide bases (adenine, guanine, cytosine, thymine) within a DNAmolecule. In 1977, two independent research groups published ground-breakingtechnological developments that facilitated the ability to determine the sequenceof DNA. In brief, Maxam and Gilbert [66] used chemicals to fragment radiolabeledDNA at specific bases, while Sanger and colleagues [67] used radiolabeled chain-terminating inhibitors. The work of Sanger et al. [67] was acknowledged with aNobel Prize in Chemistry in 1980 and established the fundamental sequencing tech-nique utilized for the Human Genome Project and clinical molecular laboratories.Over the last two decades, modifications of Sanger sequencing as well as alternativeDNA sequencing strategies have occurred. Subsequent sections on DNA sequenc-ing will focus upon Sanger chain termination methods and modifications thereof,pyrosequencing or sequencing by synthesis, and next-generation sequencing.

Sanger Sequencing

Sanger sequencing is a multistep process that begins with PCR-based amplifica-tion of target DNA, followed by removal of excess deoxynucleotide triphosphates(dNTPs) and PCR primers. The next few steps involve denaturation of the dsDNAto facilitate the annealing of a sequencing primer to the 5′-end of the desiredregion of DNA to be sequenced and the addition of a thermostable DNA poly-merase and a mixture of dNTPs and dideoxynucleotide triphosphates (ddNTPs).During repeated thermal cycling, the DNA polymerase recognizes the annealedsequencing primer at the 5′-end of the region of interest and in a 5′- to 3′-directioncreates a new strand of DNA (complementary to the template DNA) as the dNTPsand ddNTPs are incorporated. ddNTPs retain a 5′-hydroxyl group which allowsfor their incorporation into the newly synthesized DNA strand; however, ddNTPslack a 3′-hydroxyl group which prevents the subsequent incorporation of additionaldNTPs or ddNTPs by DNA polymerase. Consequently, incorporated ddNTPs ter-minate the ability of the DNA polymerase to further extend the newly synthesizedDNA strand. The rate of ddNTP incorporation is dependent upon the molar ratioof dNTPs to ddNTPs and the ability of the DNA polymerase to recognize andinsert them into the growing DNA strand. In the initial paper by Sanger et al.[67], four separate sequencing reactions were required, each containing identi-cal reagents, chain-terminating ddNTPs (ddATP, ddGTP, ddTTP, ddCTP), and a


radiolabeled dNTP, usually [35S]dATP or [32P]dATP. Following thermal cycling,each reaction consisted of a pool of DNA fragments of varying lengths correspond-ing to when each respective chain-terminating ddNTP was incorporated. The DNAfragments underwent high-resolution electrophoresis with each sequencing reac-tion occupying an individual lane in a polyacrylamide gel. Exposure of the gelto X-ray film allowed visualization of the variously sized DNA fragments dueto the incorporation of the radiolabeled dATP. The DNA sequence was obtainedby reading the “staggered stair step” banding pattern in reverse order (anode tocathode) corresponding to smallest to largest DNA fragments. Although Sangersequencing was an important milestone in molecular biology, the technique waslabor intensive, time consuming, exposed personnel to radioactive materials andX-rays, and was not adaptable to automation. Modifications of Sanger sequenc-ing emerged in the mid-1980s, whereby sequencing primers were differentiallylabeled with fluorescence dyes – so-called dye-primer chemistry [68, 69]. Similarto the original Sanger method, four separate sequencing reactions were required;however, the completed reactions could be pooled and DNA fragments resolvedwithin a single lane on a polyacrylamide gel that was coupled to a laser-inducedmulti-wavelength fluorescence detection system. Not long thereafter, fluorescencedye-primer chemistry was replaced with fluorescence dye-terminator chemistry.This technological advancement paved the way for complete automation of theDNA sequencing process by eliminating the need for four separate sequencingreactions, utilizing a single sequencing primer, and the ability to resolve the DNAfragments to a single base pair with gel or capillary electrophoretic techniques,each coupled with laser-induced multi-wavelength fluorescence detection systems.The most commonly employed fluorescent dyes were carboxyfluorescein (FAM),carboxy-4′, 5′-dichloro-2′, 7′-dimethoxyfluorescein (JOE), carboxy-X-rhodamine(ROX), and carboxytetramethylrhodamine (TAMRA) because their emission wave-lengths are spaced such that there is minimal spectral overlap, facilitating accuratedetection and base pair resolution. The adaptation of fluorescence resonance energytransfer (FRET) technology to DNA sequencing was introduced in the mid-1990susing labeled sequencing primers [70], and shortly thereafter to labeled chain ter-minator ddNTPs [71]. FRET-labeled dye-terminator chemistry for the most part isthe most commonly employed Sanger-based sequencing technology and is mar-keted as BigDye R© by Applied Biosystems (Foster City, CA). BigDye R© chemistryyields superb signal strength, is easily adapted to automated DNA-sequencingplatforms, and produces little if any differential mobility of DNA during elec-trophoretic separation [56, 72, 73]. Although Sanger-based sequencing technologiesprovide high-quality sequence information in the range of several hundred tothousand bases, there are still practical limitations that need to be considered.Despite automation, Sanger-based sequencing is still relatively labor intensive, timeconsuming, expensive, and requires specialized equipment. It also has limited sen-sitivity for detecting point mutations, approximately 20% mutant DNA within awild-type background [74–76]. Despite these drawbacks, Sanger-based sequencingis an invaluable tool for the clinical molecular laboratory and is becoming morecommonplace over time.


Pyrosequencing

Pyrosequencing, or sequencing by synthesis, was conceptualized in 1985 [77]. Theprinciple underwent modifications over the following decade [78, 79] and in the late1990s was shown to be a rapid, cost-effective alternative to Sanger sequencing [80,81]. It differs from Sanger sequencing by relying upon the detection of releasedpyrophosphate upon dNTP incorporation, rather than chain-terminating ddNTPincorporation. In brief, purified PCR amplicons are denatured to single-strandedDNA (ssDNA) templates and immobilized onto streptavidin-coated magnetic beads.The next series of steps involve four enzymes (Klenow fragment of DNA poly-merase I [82] sulfurylase [83], luciferase [84], apyrase [85]), enzyme substrates(adenosine phosphosulfate, D-luciferin), the sequencing template with annealedsequencing primer, and dNTPs (dATP, dCTP, dGTP, dTTP). Each dNTP is dis-pensed one at a time in a repetitious cyclic manner, initiating an enzymatic cascade.For each dNTP incorporated into the newly synthesized DNA strand by DNA poly-merase I, a molecule of inorganic pyrophosphate (PPi) is released which becomesthe substrate for ATP generation by ATP sulfurylase. The generated ATP is uti-lized by luciferase to emit a bioluminescent signal. Unincorporated dNTPs andexcess ATP are continuously degraded prior to the subsequent dispensation ofdNTPs. Due to the stoichiometric relationship of substrates to products withinthis enzymatic cascade, the intensity of the generated signal is directly propor-tional to the number of each dNTP incorporated. The bioluminescent signal isdetected and analyzed by the instrument in real time with the resultant genera-tion of a pyrogram which consists of a series of peaks whose temporal relationshipand height reflect the DNA sequence [86]. Detailed textual and pictorial descrip-tions of pyrosequencing are available [81, 87] but are beyond the scope of thischapter. Important properties of pyrosequencing include the ability to obtain high-quality, semi-quantitative sequence data of 20–40 bases in real time and the ability tocontrol the dNTP dispensation order. These unique properties make pyrosequenc-ing advantageous for detecting mutations within short segments of DNA and theanalysis of single-nucleotide polymorphisms (SNPs) [88]. The semi-quantitativenature of pyrosequencing facilitates determining the allelic ratio in hematopoieticchimerism or mixed clonality/heterogeneous tissue samples, the latter of which ischaracteristic of myeloproliferative neoplasms (MPNs) [76]. Dilution experimentsdemonstrate the ability of pyrosequencing to obtain assay sensitivity as low as 5%mutant allele within a wild-type background [89, 90]. More recently, pyrosequenc-ing demonstrated utility in following patients with CML to detect changes in therelative proportion of mutant clones conferring dasatinib resistance or intolerance[91]. The ability of pyrosequencing to detect epigenetic changes (i.e., DNA methyla-tion patterns) has also been described with applications to various hematolymphoidmalignancies [92–94].

Analytical drawbacks of pyrosequencing include limited base read length and“plus and minus” frameshift, all of which have been improved by the additionof E. coli single-stranded binding protein (SSB). “Plus and minus” frameshiftsare caused by insufficient activity of apyrase and DNA polymerase I (Klenow


fragment), respectively. The addition of SSB stabilizes ssDNA and protects itfrom degradation and conformation changes during the pyrosequencing reac-tion, and through various mechanisms minimizes the effect of “plus and minus”frameshift, detailed mechanistic descriptions of which are discussed elsewhere [87].Pyrosequencing technology was originally commercialized by Pyrosequencing AB(Uppsala, Sweden), later renamed Biotage (Uppsala, Sweden) in 2003 and wasrecently acquired by Qiagen (Gaithersburg, MD) in 2008. Pyrosequencing tech-nology was also licensed to 454 Life Sciences that notably developed the firstlarge-scale, high-throughput DNA-sequencing platform, thereby laying the founda-tion for next-generation sequencing. 454 Life Sciences has recently been acquiredby Roche Diagnostics.

Next-Generation Sequencing (NGS)

Sanger-based sequencing has dominated the molecular biology landscape overthe last three decades, primarily due to the desire of the international commu-nity to sequence the entire human genome. As a result, Sanger-based sequencingwas quickly adapted to large-scale, high-throughput automation allowing paral-lel sequencing of DNA in up to 384 capillaries at a time. The industrializationof Sanger-based sequencing, primarily undertaken by Applied Biosystems, Inc.(Forster City, CA), facilitated the sequencing of the human genome in 2003. Asthe molecular biology community sought to expedite the sequencing of the humangenome and that of other species, it became readily apparent that other technologieswould be required. Over the last two decades, considerable resources were investedto the development of alternative sequencing strategies, and as recently as 2005 theirutility was demonstrated. These novel sequencing strategies ushered in the new eraof high-throughput sequencing and hence next-generation sequencing (NGS) wasborn. NGS provides numerous advantages over automated Sanger-based methods,including high speed and throughput, full automation, expense reduction, and thedetermination of sequence data from amplified single DNA fragments, negating theneed for the in vitro cloning of DNA fragments. The high-throughput nature anddecreased expense of NGS cannot be overemphasized. For perspective, Sanger-based sequencing of the entire human genome took 13 years with an estimatedcost approaching $3 billion. In comparison, NGS technology sequenced the entirehuman genome in 5 months at a cost of $1.5 million [95, 96]. Despite these advan-tages, NGS does have drawbacks. Instrumentation is extremely expensive (range$500,000 to over $1 million), with individual sequencing runs costing over $5,000.With this being said, it is still several orders of magnitude less expensive thanSanger-based sequencing on a cost per base basis [97]. NGS is slightly more proneto sequencing errors due to non-uniform confidence in base calling, especially whendealing with homopolymeric tracts, and some technologies suffer from short readlengths. Furthermore, the quantity of sequence information generated per sequenc-ing reaction (range 80 Mb–3 Gb) [98] creates an enormous amount of data (range15 GB–15 TB) [95], which require unique information technology solutions for datastorage and analysis.


All NGS platforms have a common technological feature – the ability for high-throughput sequencing of clonally amplified or single DNA molecules which arespatially separated in a flow cell. Sequencing is performed by iterated cyclesof either polymerase-mediated dNTP extension or oligonucleotide ligation [95].Descriptions of the various technological strategies employed with NGS alongwith their advantages and disadvantages are beyond the scope of this chapterbut have been thoroughly reviewed [95, 97, 99, 100]. Since 2005, numerousNGS technologies have emerged and are commercially available from RocheApplied Science (454 GenomeSequencer FLX), Illumina (Illumina/Solexa GenomeAnalyzer), Applied Biosystems (Supported Oligonucleotide Ligation and Detectionsystem, SOLiD), and Helicos BioSciences (HeliScope). Other NGS platforms in thedevelopmental phase are from VisiGen Biotechnologies and Pacific Biosciences.

Although NGS was initially developed as a high-throughput means of genomicsequencing, novel applications of this technology are beginning to emerge.Examples include personalized medicine with detailed analysis of selected portionsof the human genome, transcriptome analysis or the analysis of RNA transcriptexpression, the identification of selected regions of DNA that interact with geneexpression regulatory elements, and the genome-wide characterization of mRNAs,chromatin structure, and DNA methylation patterns [97, 101]. To date, the literatureis sparse in regard to applying NGS to hematolymphoid neoplasms, but this shouldbe temporary. A recent review by Neff et al. [102] shed light on the application ofNGS to study epigenetic changes in leukemia. Furthermore, the ability of NGS tocharacterize the genome-wide transcriptome profile of normal and cancerous tissuesunder controlled conditions (e.g., presence of selected anti-neoplastic drugs) shouldshed light on mechanisms of differential RNA expression, paving the way for therapid development and employment of new anti-neoplastic agents [103–105].

Polymerase Chain Reaction

Polymerase chain reaction (PCR) is an in vitro, DNA polymerase-dependent methodfor the exponential amplification of nucleic acid. From its inception in the mid-to-late 1980s, this invention [106] has revolutionized the direction of moleculardiagnostic testing and is without question one of the most important milestonesin the field of molecular testing. Compared to older technologies used to analyzeDNA, such as the Southern blot, PCR is much more rapid, provides superior speci-ficity and sensitivity, is less technically challenging, and allows for much higherthroughput. For these reasons, PCR has become an indispensable tool in the prac-tice of hematopathology. PCR is dependent on thermal cycling, i.e., iterative cyclesof heating and cooling to allow for melting and annealing of DNA sequences,respectively. A standard PCR requires a target DNA template, two oligonucleotideprimers, which are complementary to opposite strands of denatured target DNA, athermostable DNA polymerase such as Taq, all four deoxynucleotide triphosphates(dNTPs), magnesium as a cofactor to Taq, and a buffer solution. There are threemain temperature-dependent steps in the PCR. First, the target DNA is denatured at94–98◦C for 10–60 s. The high temperature disrupts the hydrogen bonds between


complementary strands of DNA, resulting in single-stranded DNA. In the secondstep, the temperature is lowered to 50–70◦C for 10–60 s to allow annealing ofprimers to the denatured target DNA. The primer sequences must be highly com-plementary to the target DNA for stable annealing to occur. Primer extension occursin the third step at an intermediate temperature of 65–75◦C. The DNA polymerasesynthesizes a strand of DNA complementary to the target DNA by adding the appro-priate dNTPs to the 3′-ends of the primers. This PCR cycle is then repeated 25–40times. A programmable thermal cycler is used to automatically change the reactiontemperature at the appropriate times. The thermostability of the polymerase elim-inates the need to replace the enzyme after each cycle [107]. In subsequent PCRcycles, the amplification products can serve as templates, resulting in the doublingof target DNA during each cycle of PCR. This exponential amplification of templateDNA is followed by a slowing of the reaction and eventual plateau as reagents areconsumed and the activity of the polymerase is diminished.

As mentioned above, PCR is a highly sensitive and specific target amplifica-tion method. The sensitivity of the reaction is due to the exponential amplificationof template DNA, which allows for the amplification of even minute amounts ofDNA. Because amplification products can serve as templates in subsequent cyclesof PCR, these reactions are also highly sensitive to contamination by ampliconsfrom previous reactions, which may result in false-positive results. Thoughtful lab-oratory design, including the separation of pre-PCR and post-PCR areas, as wellas clean laboratory practices, can help prevent contamination by PCR amplicon orother extraneous DNA. On the other hand, the specificity of the reaction is primarilyconferred by the two PCR primers, which anneal to the target DNA only if there isa degree of sequence complementarity and only at an appropriate temperature. Theoptimal annealing temperature is dependent on the length of the primer as well as theguanosine–cytosine (GC) content of the primer. The specificity of hybridization canbe controlled by varying the annealing temperature or the magnesium concentra-tion. Increasing the temperature or decreasing the magnesium concentration resultsin a more stringent hybridization, while lowering the temperature or increasing themagnesium concentration will lower the stringency of the reaction. Furthermore,extension of a PCR primer occurs only if there is perfect complementarity at the3′-end of the primer due to the sensitivity of Taq polymerase to mismatches atthis location. To prevent the generation of nonspecific amplification products, Taqpolymerase activity may be inhibited early in the reaction by waiting to add Taqto the reaction mix until denaturation begins, by separating Taq from the reactionmix using a barrier, or by using specialized systems incorporating Taq inhibitorsthat dissociate at a high temperature. This variation on PCR is known as hot-startPCR. The specificity of conventional PCR must be confirmed by visualization of thePCR product. The size of the PCR product can be determined by standard gel elec-trophoresis or CE. Alternatively, or in addition to fragment size analysis, the PCRproduct can also be sequenced to ensure the specificity of the reaction. A variety oforganic and inorganic compounds may inhibit PCR, including hemoglobin and urea.Consideration of possible inhibitors is important in the interpretation of a negativePCR result, as is the inclusion of an internal amplification control to confirm thatamplification was not inhibited. Current clinical applications of conventional PCR


to the practice of hematopathology include the amplification of DNA for subsequentassessment of clonality and for the detection of recurrent genetic abnormalitiesincluding point mutations by sequencing or fragment size analysis as mentionedabove. Multiple primer sets can be included in a single PCR to amplify multiple tar-gets from a single specimen in a variation on PCR called multiplex PCR. Effectiveprimer design is critical for the success of multiplex PCR, and the reaction mustbe optimized to ensure functionality of the multiple primer sets in a single reac-tion. Many other variations on the conventional PCR have been developed and havecurrent applications to the field of hematopathology.

Reverse Transcription PCR (RT-PCR)

Reverse transcription PCR (RT-PCR) is a ribonucleic acid (RNA)-based PCR,which utilizes reverse transcriptase, an RNA-dependent DNA polymerase, capa-ble of DNA polymerization using RNA as a template. The resulting complementaryDNA (cDNA) strand is much more stable than the RNA target and can be used asa template in any subsequent PCR application. A single reverse transcription (RT)reaction allows for subsequent PCR analysis of multiple targets from the resultantcDNA. RT using random hexamer primers results in cDNA complementary to totalRNA, whereas the use of specific oligo-dT primers results in cDNA complementaryto mRNA only. Alternatively, gene-specific primers can be used to produce cDNAcomplementary to the gene of interest only. The availability of enzymes capableof using both RNA and DNA as templates for DNA polymerization has elimi-nated the need for an extra enzymatic step and increased the efficiency of RT-PCR[108]. As mentioned above, RNA is highly susceptible to degradation by ubiqui-tous RNases, and care is needed when handling RNA to prevent excessive lossor fragmentation. There are numerous applications of RT-PCR in hematopathol-ogy, including the detection of various fusion transcripts. The length of genomicDNA spanning chromosomal translocations usually prohibits their detection byPCR. However, since introns are removed from mRNA, these translocations can bemore readily detected in the form of fusion mRNA transcripts. Some of the fusiontranscripts currently evaluated by RT-PCR include BCR–ABL1 in CML and acutelymphoblastic leukemia (ALL), ETV6–RUNX1 and E2A–PBX1 in ALL, PML–RARA, RUNX1–ETO, CBFB–MYH11, RUNX1–RUNX1T1, and DEK–NUP214in AML, NPM–ALK in anaplastic large-cell lymphoma, and BCL2–IGH in follicu-lar lymphoma. Detection of these chromosomal translocations by RT-PCR is usefulat the time of diagnosis both for risk stratification and treatment determination aswell as following therapy for monitoring the presence of minimal residual disease.

Allele-Specific PCR

Allele-specific PCR is used primarily for the detection of point mutations andsingle-nucleotide polymorphisms (SNPs). Whereas in conventional PCR, primers


are designed to be complementary to an invariant region of the target DNA,allele-specific PCR utilizes a primer whose 3′-end includes the mutated site.Allele-specific PCR is also referred to as allele-specific oligonucleotide PCR andamplification refractory mutation system (ARMS) [109]. As mentioned above, theremust be perfect sequence complementarity at the 3′-end of a primer and the tar-get DNA for amplification to occur. Under stringent conditions, any mismatch atthis location will prevent amplification from occurring. ARMS consists of two sep-arate amplification reactions, one utilizing a wild-type allele-specific primer andone utilizing a mutant allele-specific primer. The second primer is common to bothreactions. If amplification occurs only in the mutant reaction, a homozygous muta-tion is present. If amplification occurs in both reactions, a heterozygous mutationis present. If amplification occurs only in the wild-type reaction, no mutation ispresent. One caveat of ARMS is that a polymorphism or an unsuspected mutationat the 3′-end of the primers will prevent amplification and may lead to misinter-pretation of results. Furthermore, amplification controls should be included in thereactions to exclude the possibility of PCR inhibition. Current clinical applicationsof ARMS–PCR include the identification of JAK2 V617F mutations in MPN [110]and ABL kinase domain mutations in imatinib resistance [111].

Nested PCR

Nested PCR is a variation of PCR with increased sensitivity and specificity [112].Nested PCR involves the use of two successive rounds of PCR using two primerpairs, one of which is located internally to the other. First, the outer primer set isused to amplify the target sequence. The PCR products are then amplified usingthe inner primer set, resulting in final PCR products that are shorter than the initialproducts. Nested PCR is highly specific since the second primer pair is complemen-tary to sequences within the amplicons produced in the first reaction. Each PCRconsists of approximately 25 cycles, resulting in approximately 50 total cycles ofPCR. This high total cycle number is responsible for the high sensitivity of nestedPCR. In addition, the smaller products obtained from the first PCR are more readilydenatured in the second reaction, resulting in abundant template DNA. However,because the initial PCR product is manipulated during transfer to the second PCRtube, there is a high risk of amplicon contamination. Due to its exquisite sensitivity,nested PCR has been used to monitor the presence of minimal residual disease aftertherapy for acute and chronic leukemias [113, 114].

Real-Time PCR

Real-time PCR involves the visualization of amplicon generation in real time usingfluorescence detection during the exponential phase of PCR [115, 116]. In otherwords, target amplification and detection occur simultaneously in a single tube usinga special thermal cycler, which monitors fluorescence emission and generates anamplification curve. The amount of fluorescence detected is directly proportional to


the amount of PCR product, which is directly proportional to the amount of startingtarget DNA. Therefore, real-time PCR can be used to quantify the amount of targetDNA present in a clinical sample. A real-time PCR growth curve is composed of abaseline or lag phase, a log-linear or exponential amplification phase, and a plateauphase. During the lag phase, the specific fluorescent signal is less than the back-ground nonspecific fluorescence or autofluorescence. The crossing point or cyclethreshold (Ct) is the point at which the growth curve enters the log-linear phase. TheCt is indirectly proportional to the amount of starting template such that a lower Ctvalue implies a larger amount of initial template. During the plateau phase, ampli-con accumulation slows as reagents become rate limiting and the efficiency of theDNA polymerase declines.

Real-time PCR holds several advantages over conventional end-point PCR,including faster turnaround time, higher reproducibility, wider dynamic range, andlower risk of amplicon contamination. Since real-time PCR is monitored in realtime, there is no need for post-amplification analysis, such as electrophoresis. Theelimination of post-amplification manipulation of the PCR products greatly reducesthe risk of carryover contamination and the time required to complete the analysis.Melting curve analysis can increase the specificity of real-time PCR by confirm-ing that amplification of the appropriate target has occurred. Melting curve analysisinvolves increasing the temperature of the reaction until the double-stranded ampli-con is denatured [117]. This melting results in a decrease in fluorescence anda characteristic melting peak. The melting peak of an amplicon is based on itsmelting temperature, which will be distinct from the melting temperatures of non-specific PCR products. Melting curve analysis is currently used in the practice ofhematopathology for the detection of JAK2 V617F mutations. The specificity ofreal-time PCR can also be increased by incorporating hybridization probes intothe reaction. A variety of methods can be used to generate a fluorescent sig-nal, including nonspecific intercalating dyes and fluorescently labeled primers andprobes.

Signal Detection Options

SYBR Green and ethidium bromide are nonspecific intercalating dyes that fluorescewhen bound to dsDNA. Nonspecific dyes simply indicate the presence of dsDNA,including nonspecific PCR products and primer dimers. Sequence-specific probes,on the other hand, indicate the generation of a specific amplicon. The amount offluorescence is directly proportional to the amount of specific product generated.Many sequence-specific fluorescent probes utilize the phenomenon of fluorescentresonance energy transfer (FRET). FRET occurs when a donor dye is excited byan external light source. Instead of emitting light, the donor transfers this energyto an acceptor dye when the donor and acceptor are in close proximity. Whenenergy transfer occurs, the acceptor molecule quenches the fluorescence of thedonor molecule. The acceptor dye then emits light in an amount proportional tothe amount of PCR product present.


TaqMan probes are complementary to a PCR product at a location between thetwo primer-binding sites. TaqMan probes contain a 5′ reporter donor and a 3′ accep-tor that quenches the reporter. When the probe is intact, the acceptor quenches thereporter, and no fluorescence is generated. However, upon primer extension, theprobe is cleaved by the 5′ to 3′ exonuclease activity of Taq polymerase, whichreleases the reporter from the quenching activity of the acceptor [118]. The result-ing fluorescent signal is proportional to the amount of amplicon generated. Sincethe probe is hydrolyzed, it is unavailable for future reactions and precludes theability to utilize melting curve analysis. TaqMan probes have found widespread usein neoplastic hematopathology, including the detection of various fusion transcripts.

As opposed to hydrolysis probes, hybridization probes remain intact through-out the reaction. A simple hybridization probe system consists of two fluorescentlylabeled probes, one donor and one reporter acceptor. The probes are complemen-tary to adjacent sequences of the PCR product between the two primer-bindingsequences. Upon annealing to amplicon, the donor and the acceptor are placedinto close proximity, and a fluorescent signal is generated. In contrast to hydrol-ysis probes, Taq polymerase simply displaces hybridization probes, which remainintact and available for the next amplification cycle as well as subsequent meltingcurve analysis. Variations of hybridization probes include molecular beacons andscorpion probes, characterized by a hairpin loop structure. Molecular beacons aresingle probes with a 5′ reporter and a 3′ quencher [119]. When unbound, the probeforms a hairpin loop structure, and the reporter is quenched. During the anneal-ing step, the hairpin loop unfolds, and the probe hybridizes to a complementarysequence in the amplicon. This annealing of probe to product separates the reporterfrom the quencher, and a fluorescent signal is generated. Similarly, a scorpion probeutilizes a hairpin loop structure to bring the reporter and the quencher into prox-imity. However, scorpions contain a primer covalently linked to the probe [120].To date, molecular beacon probes and scorpion probes have found clinical applica-tions primarily in the molecular microbiology lab for pathogen detection. However,this technology has the potential for clinical applicability in hematopathology, espe-cially for the sensitive detection and quantification of point mutations, such as JAK2V617F [121], and for monitoring minimal residual disease [122].

Quantitative Real-Time PCR

Quantitative real-time PCR (Q-PCR) is currently the most accurate method forquantifying DNA or RNA. Because real-time PCR analysis occurs in the earlylog phase of amplification, this method is less sensitive to differences in PCR effi-ciency between reactions. In Q-PCR, a standard curve is generated using samples ofknown template concentration. The concentrations of unknown samples can thenbe extrapolated from the standard curve (Fig. 1.2). Q-PCR is currently utilizedin hematopathology to monitor minimal residual disease post-therapy, especiallyfor ALL, AML, and CML. The amount of minimal residual disease may be usedclinically to assess the efficacy of therapy, to determine further treatment type


Fig. 1.2 Quantitative real-time RT-PCR. Real-time PCR amplification curves (top) and standardcurve (bottom) generated from a 10-fold dilution series of a known standard. Graphical dataobtained from a quantitative real-time RT-PCR assay for BCR–ABL (Ipsogen, Inc., Stamford, CT)using the LightCycler 1.2 (Roche Applied Science, Indianapolis, IN)

and timing, as well as for prognostic information [123]. Q-PCR is also used toquantify the allelic burden of JAK2 V617F in patients with a myeloproliferativeneoplasm [124].

Methylation-Specific PCR

DNA methylation is an epigenetic phenomenon critical for transcriptional regu-lation and is an essential process in human development. In humans, DNA ismethylated at CpG islands, i.e., cytosines located 5′ to guanosines. CpG islandsare present in the 5′ regulatory regions of many human genes. Aberrant hyperme-thylation has been demonstrated to be an important mechanism for transcriptionaldysregulation in neoplasia and may contribute to the development of leukemiasand lymphomas. Methylation-specific PCR is one way to distinguish methylatedfrom unmethylated DNA [125]. Treatment of DNA with sodium bisulfite convertsunmethylated cytosine to uracil while leaving methylated cytosine intact. Followingsodium bisulfite treatment, methylation-specific PCR amplifies DNA using primersspecific for either methylated or unmethylated DNA. The primer sets for methylated


and unmethylated sequences are identical except at CpG islands, where one primerpair recognizes cytosine in methylated DNA and the other primer pair recognizesuracil in unmethylated DNA. The PCR products can then be analyzed by standardgel electrophoresis or CE to identify methylated or unmethylated target sequencesor both. While a variety of differentially methylated genes have been described inhematolymphoid neoplasia, no applications have been implemented in the clinicallaboratory to date.

Restriction Site PCR

Restriction site-generating PCR generates an artificial RE recognition site. Briefly,a PCR primer is designed to have a mismatch with the template adjacent to themutation of interest. The mismatched base creates a restriction site in either thewild-type or the mutant amplicon. The RE digestion pattern can then be analyzedby standard gel electrophoresis or CE. This method can be theoretically used inhematopathology for the detection of mutations which do not result in the creationor loss of an RE recognition site. However, this variation on PCR has not gainedwidespread clinical use to date.

Asymmetric PCR

Asymmetric PCR is a variation on PCR which preferentially amplifies one strandof target DNA over the other. This outcome is typically accomplished by usingexcess amounts of the primer complementary to the desired single-stranded product.Upon depletion of the limiting primer during the exponential phase, amplifica-tion of the desired strand occurs in a linear fashion. For this reason, asymmetricPCR requires additional cycles of PCR and at times may be inefficient and diffi-cult to optimize. Linear-after-the-exponential PCR (LATE-PCR) is a specific typeof asymmetric PCR with increased efficiency due to the use of a limiting primerwith a higher melting temperature than the excess primer. While asymmetric PCRmay be used for the detection of a variety of mutations in hematolymphoid neo-plasia, clinical applications of LATE-PCR are not yet widespread in the practice ofhematopathology.

Clonality Assessment

Overview

The majority of hematolymphoid disorders can be characterized as malignant orreactive/benign by clinical history, morphology, and immunophenotyping usingancillary studies (i.e., immunohistochemistry, flow cytometry). However, up to 10%of cases may remain elusive [126]. As such, clonality assessment is an invaluable


tool to further characterize these cases. In principle, all cells comprising a malignantprocess are monoclonal because they arise, via the process of clonal expansion, froma common malignant cell; likewise, reactive/benign conditions are polyclonal asthey are composed of numerous unique cell populations. Clonality assessment reliesupon the techniques of Southern blotting and PCR to analyze the immunoglobulinand T-cell receptor genes.

Immunoglobulin (Ig) and T-cell receptors (TCRs) are encoded by gene segmentclusters that undergo genetic (somatic) recombination during the development andmaturation of B and T cells in the bone marrow and thymus, respectively. Thisrecombination essentially involves the splicing and fusion of one of numerous vari-able (V), diversity (D), joining (J), and constant (C) regions. Diversity (D) regionsare not present within Ig kappa, Ig lambda, TCR alpha, and TCR gamma chains.The recombination process is sequential in that one gene segment from each of theD (if applicable) and J regions rearrange first, followed by V to DJ rearrangement.The assembled V–J or V–D–J segments will subsequently be joined to a distinctconstant (C) region to create a unique coding sequence capable of being trans-lated into a functional antigen receptor protein [127, 128]. Immune cells yieldingnon-functional antigen receptors undergo programmed cell death via the processof apoptosis. In regard to immature B cells, the Ig heavy (IGH) chain is the firstto undergo somatic recombination. If the first allele is non-functional, the secondallele will rearrange to produce a function gene. Therefore, a single B cell canhave two different IGH rearrangements, one functional and one non-functional.Rearrangement of the kappa (κ) light chain is next and occurs only if the IGHrearrangement was successful. The lambda (λ) light chain will undergo somaticrecombination only if both alleles from the kappa light chain are non-productive,in which case the non-functional kappa light chain gene segments will be deleted.The end result is an intact coding sequence which facilitates the creation of a func-tional immunoglobulin protein receptor. In regard to T cells, the T-cell receptorgamma (TCRγ) and delta (TCRδ) loci begin the somatic recombination process.Approximately 10% of T cells will express a functional heterodimeric gamma–deltaT-cell receptor (TCRγδ). The majority (90%) of T cells will encode a non-functionalTCRγδ (gamma delta) antigen receptor and thus will rely upon the TCR alpha(α) and TCR beta (β) loci to generate a functional, heterodimeric T-cell receptor(TCRαβ) (alpha beta). It is imperative to understand that the majority of mature Tcells expressing a TCRαβ (alpha beta) phenotype still retain the non-functional γδ

(gamma delta) gene rearrangement – a unique feature that can be exploited withmolecular-based testing for clonality assessment.

The immune system requires an enormous repertoire of antigen receptors to facil-itate the recognition of an almost infinite number of antigens. Immunoglobulin andT-cell receptor antigen diversity is primarily derived from the numerous geneticsegments available within the V, D, and J regions that are randomly chosen toundergo somatic recombination (Table 1.1). As a result, the number of com-binatorial possibilities is significant, at approximately 2 × 106, 3 × 106, and5 × 103 for the Ig, TCRαβ (alpha beta), and TCRγδ (gamma delta),


respectively [126]. Additional diversity is facilitated by two mechanisms: (1) imper-fection in junction site splicing and (2) the enzymatic action of terminal deoxynu-cleotidyl transferase (TdT), which adds or deletes individual nucleotides and/orsmall oligonucleotide sequences at the V–D–J splice sites (Fig. 1.3). Following

Table 1.1 General characteristics of the immunoglobulin (Ig) and T-cell receptor (TCR) genes.Standardized nomenclature, chromosomal location of individual genes, and approximate num-ber of gene segments contained within the variable, diversity, joining, and constant regions areprovided. Compiled from multiple sources [126, 138, 139]

Antigenreceptor Gene

Genelocation Variable (V) Diversity (D) Joining (J) Constant (C)

Ig Heavy (H) 14q32 >100 30 9 11Kappa (κ) 2p11–12 50–100 0 5 1Lambda (λ) 22q11 20–70 0 7 J–C clusters

TCR Alpha (α) 14q11 50–100 0 50–100 1Beta (β) 7q34 75–100 2 13 2Gamma (γ) 7p15 14 0 5 2Delta (δ) 14q11 10 3 3 1

Fig. 1.3 Schematic of immunoglobulin heavy chain (IGH) receptor gene. Basic process of somaticrecombination at the IGH locus occurring within developing B cells. Random selection and rear-rangement of first diversity (DH) and joining (JH) gene segments, followed next by a variable(VH) gene segment completes the primary V–D–J coding sequence. Antigen diversity is derivedfrom the unique nucleotide sequences of the randomly selected gene segments. Further diversitystems from terminal deoxynucleotidyl transferase (TdT) which inserts a random number of “non-template”-derived nucleotides (designated “n”) at the variable, diversity, and joining junction sites.The finished V–D–J segment is then joined with a constant (CH) region to complete the final IGH-coding sequence for protein synthesis. The IGH variable region (VH) is further divided into threeframework regions (FRs) and three complementarity-determining regions (CDRs). Due to the highdegree of sequence homology with the framework regions, PCR-based assays can utilize differ-entially labeled fluorescent forward (FR1, FR2, FR3) and non-labeled reverse (FR4) primers togenerate fluorescently labeled amplicons that can subsequently undergo capillary electrophoreticresolution


successful rearrangement, the immunoglobulin genes within mature B cells mayundergo additional genetic alterations to “fine-tune” their antigen receptor via theprocesses of isotype switching and somatic hypermutation within the germinalcenter.

Techniques

Molecular laboratories generally rely upon the molecular techniques of Southernblotting and PCR-based methods for clonality assessment and focus on IGH, TCRγ

(gamma), and TCRβ (beta) gene rearrangements. Although still considered bymany to be the gold standard, Southern blot analysis is performed only by afew laboratories because of slow turnaround time (usually 5–7 days), expense,and labor-intensive nature and inherent technical challenges of this methodology.Southern blotting requires microgram or greater quantities of intact, high molecu-lar weight DNA which can be obtained only from fresh tissue, not fixed or fixedparaffin-embedded tissue. Additionally, the sensitivity of this method is around10%, meaning that approximately 10% of the clonal population must be representedin the tissue submitted for analysis to be reliably detected by this methodology[126]. In brief, clonality assessment by Southern blot involves subjecting highmolecular weight DNA to RE digestion, whereby clonal gene rearrangements willyield a different restriction fragment length pattern as compared to the germ lineconfiguration. Radiolabeled probes are hybridized to the digested DNA fragmentsand visualized. For B-cell clonality, probes are usually directed toward the joining(J) regions of the IGH chain or the Ig kappa light chains [126]. For T-cell clonal-ity, probes are generally directed toward the constant (C) or joining (J) regions ofthe TCRβ (beta) gene. Southern blot analysis generally fails to detect monoclonalpopulations when targeting the TCRγ (gamma) gene.

Due to the inherent challenges of Southern blot analysis, the majority of molec-ular laboratories rely upon PCR-based methods for clonality assessment. BecausePCR-based methods amplify relatively short distances (usually <1,000 bp) of DNA,this approach is advantageous for formalin-fixed, paraffin-embedded tissue, whereDNA fragmentation is universally encountered. Furthermore, the sensitivity of PCR-based method approaches 1% (i.e., 1 clonal cell in a background of 99 polyclonalcells), especially when coupled with capillary electrophoresis, an important toolfrom the viewpoint of initial diagnosis and minimal residual disease monitoring.In general, these methods employ multiple primer sets (i.e., consensus primers)that target conserved gene segments within variable (V) and joining (J) regions ofthe antigen receptor genes that have undergone somatic recombination, due to theclose proximity of the V–J or V–D–J coding sequence. These consensus primerswill amplify both clonal and polyclonal (background) cell populations, the latter ofwhich is depicted as a smear on gel-based detection platforms. Using fluorescentlylabeled primers, which are incorporated into the amplified products, backgroundcell populations will appear as a Gaussian distribution of peaks when resolved bycapillary electrophoresis (Fig. 1.4).


Fig. 1.4 Electropherogram of IGH gene rearrangement following three separate PCR-basedamplification reactions using differentially labeled fluorescent primers. Each PCR reaction utilizedthe same consensus joining (JH) region reverse primer targeting the framework 4 (FR4) regionand separate consensus forward primers targeting the framework 1 (FR1, top panel), framework 2(FR2, middle panel), and framework 3 (FR3, bottom panel) regions of the IGH gene. Each panelclearly indicates the presence of a distinct, monoclonal peak. Top and bottom panels also demon-strate the concomitant presence of normal cell populations as indicated by the Gaussian distributionof polyclonal B cells

PCR-based analysis of the IGH gene is perhaps the most commonly employedmethod of clonality assessment. The V–D–J rearrangement created after somaticrecombination can be further subdivided into discrete functional units, enti-tled framework regions (FRs), and complementarity-determining regions (CDRs)(Fig. 1.3). The three framework regions (FR1, FR2, FR3) within the VH region havea high degree of sequence similarity allowing for the creation of forward consensusprimers. The fourth framework region (FR4) is located in the JH region and allowsfor the creation of a universal reverse primer. In contrast, the complementarity-determining regions (CDR1, CDR2, CDR3) play an integral part in antigen receptordiversity and as such contain tremendous sequence diversity. Assays employingFR3 and JH consensus primers will detect 60–70% of malignant B-cell neoplasms[126]. This detection rate can exceed 80% by the addition of consensus primersets targeting the FR1 and FR2 regions [126, 129] and has been further increasedby using more extensive primer sets as reported by the BIOMED-2 consortiumproject [130, 131]. Because the FR1 and JH consensus primers generate the largestamplicons (range, 350–450 bp), there is a higher chance of false-negative resultsif significant DNA degradation is present. It is therefore imperative to incorporateamplicon size-matched internal controls to assess the degree of DNA degradationfor each sample analyzed. PCR-based methods will not detect all clonal B-cellpopulations. False-negative results are commonly encountered with post-germinalcenter B-cell malignancies (i.e., chronic lymphocytic leukemia, follicular lym-phoma, plasma cell malignancies) due to somatic hypermutation. This process


normally occurs within the germinal center and facilitates additional alterations ofthe V–D–J coding sequence of the IGH genes within cells. This process facilitates“fine-tuning” of the antigen recognition sites of the Ig receptor, however, alters theDNA sequence within the VH and JH regions such that PCR consensus primersno longer anneal to the template DNA, thereby severely reducing or preventingamplification.

PCR-based clonality assessment of the TCR is usually performed on theTCRγ (gamma) and/or TCRβ (beta) chain; however, due to the complexity ofthe TCRβ (beta) locus (i.e., 75–100 V segments, 2 D segments, 13 J segments)and the requirement for a large number of amplification primers, most labora-tories analyze only TCRγ (gamma). In comparison, the TCRγ (gamma) locusis considerably simple containing only 14 V segments and 5 J segments, withthe former divided into four discrete families (Vγ1–8, Vγ9, Vγ10, and Vγ11).As such PCR analysis requires relatively few consensus primer sets to amplifythe corresponding V–J coding sequence following somatic recombination. Thevast majority (>90%) of T-cell malignancies will be detected with this approach[132–134].

Limitations

Limitations of molecular-based methods of antigen receptor clonality assess-ment consist of DNA degradation, lineage infidelity, detection of clonal popula-tions within reactive/benign processes, oligoclonality/clonal evolution, and primer-binding site mutations due to somatic hypermutation. DNA degradation is acommon limitation and is generally due to formalin fixation as previously discussed.It is prudent to include amplicon size-matched internal controls to assess the degreeof DNA degradation with each sample to be tested. However, it should be noted thatfalse-negative results may still occur if the quantity of DNA to be amplified for clon-ality assessment is disproportionally less than the amount of DNA to be amplifiedfor the internal controls.

Lineage infidelity is a phenomenon whereby discordance exists between thedetected clonal antigen receptor rearrangement and the immunophenotype ofthe cell [i.e., detecting a TCRγ (gamma) gene rearrangement in a B-cell lym-phoma]. This commonly occurs with precursor B-cell malignancies (i.e., precursorB-cell acute lymphoblastic leukemia) and is uncommonly encountered (<10%) withmature B-cell malignancies. Although this process is not entirely understood, it isthought that immature B cells may rearrange their TCR genes prior to receivingappropriate cues from the cellular milieu (i.e., bone marrow) to direct their commit-ment to the B-cell lineage. Once committed to the B-cell lineage, the appropriateIGH and IGκ (kappa) or IGH and IGλ (lambda) gene rearrangements occur aspreviously discussed; however, the rearranged TCR gene is not deleted and can bedetected. This concept is important to understand and as such, the use of molecular-based methods for clonality assessment should be discouraged to assign cellularlineage.


The ability to detect a clonal cell population does not always equate to malig-nancy, as small B-cell clones may be detected in benign/reactive lymphoid hyperpla-sia without supporting morphologic or immunophenotypic (immunohistochemicaland/or flow cytometric) evidence of neoplasia [135–137]. This situation may occurin the setting of autoimmune disease, immunodeficiency states, and certain infec-tions. As such it is prudent to correlate molecular test results with other clinical,morphologic, and immunophenotypic data.

Another limitation of PCR-based methods of clonality assessment is oligoclonal-ity and clonal evolution. By definition, an oligoclonal pattern indicates the presenceof more than two bands following PCR and electrophoretic resolution. This typeof pattern is commonly seen in reactive processes where there are fewer distinctcellular populations (i.e., expansion of several cell clones, no longer polyclonal) aswould be expected to facilitate a focused cellular response to the pathologic process.This pattern is also seen in immunocompromised individuals who have a reducedcellular repertoire and in paucicellular specimens. The latter situation is commonlyencountered with diminutive skin specimens (i.e., punch and shave biopsies), andas such it is imperative to review the histologic preparations of the correspondingmaterial submitted for molecular analysis. It is also possible for an individual to havetwo separate, simultaneous monoclonal processes. Prior to entertaining this remotepossibility, correlation of the molecular testing results with other morphologic andimmunophenotypic information is warranted. Lastly, clonal evolution is the pro-cess whereby during the course of the disease process, additional genetic alterationsoccur within the antigen receptor genes. This will lead to either an alteration ofthe so-called tumor-specific signature (i.e., unique size of the monoclonal peak) orprimer annealing site mutations such that the so-called signature peak of the tumoris no longer detected. When the latter situation occurs, patient monitoring should beperformed with either flow cytometric analysis and/or an alternative antigen receptorrearrangement assay [i.e., IGκ (kappa), IGλ (lambda)] if informative.

References

1. Swerdlow SH, Campo E, Harris NL, et al. WHO Classification of Tumours ofHaematopoietic and Lymphoid Tissues. 4th ed. Lyon, France: International Agency forResearch on Cancer; 2008.

2. Merrick TA Beavis KG, Bhagat G, et al. GP17-A2, Clinical laboratory safety; approvedguideline - second edition. Clin Lab Stand Inst.2004;24:1–92.

3. Sewell DL Callihan DR, Denys GA, et al. RE. M29-A, Protection of laboratory workers fromoccupationally acquired infections; approved guideline - second edition. Clin Lab StandInst.2001;17:1–147.

4. Gassner C, Schmarda A, Kilga-Nogler S, et al. RHD/CE typing by polymerase chain reactionusing sequence-specific primers. Transfusion.1997;37:1020–1026.

5. Warren DJ, Andersen A, Slordal L. Quantitation of 6-thioguanine residues in peripheralblood leukocyte DNA obtained from patients receiving 6-mercaptopurine-based mainte-nance therapy. Cancer Res.1995;55:1670–1674.

6. Ferreira-Gonzalez A, Fisher LM, Lehman CM, et al. Detection of a common mutationin factor V gene responsible for resistance to activate protein C causing predisposition tothrombosis. J Clin Lab Anal.1997;11:328–335.


7. Rainen L AJ, Asthana D, Earley MC, Geiszler RL, Krieg-Schneider F, Mannhalter C, OginoS, Parish GT, et al. MM13-A, collection, transport, preparation, and storage of specimensfor molecular methods; approved guideline. Clin Lab Stand Inst. 2006;25:1–64.

8. Beutler E, Gelbart T, Kuhl W. Interference of heparin with the polymerase chain reaction.Biotechniques.1990;9:166.

9. Satsangi J, Jewell DP, Welsh K, et al. Effect of heparin on polymerase chain reaction.Lancet.1994;343:1509–1510.

10. Taylor AC. Titration of heparinase for removal of the PCR-inhibitory effect of heparin inDNA samples. Mol Ecol.1997;6:383–385.

11. Akane A, Matsubara K, Nakamura H, et al. Identification of the heme compound copurifiedwith deoxyribonucleic acid (DNA) from bloodstains, a major inhibitor of polymerase chainreaction (PCR) amplification. J Forensic Sci.1994;39:362–372.

12. Langebrake C, Gunther K, Lauber J, et al. Preanalytical mRNA stabilization of whole bonemarrow samples. Clin Chem.2007;53:587–593.

13. Rainen L, Oelmueller U, Jurgensen S, et al. Stabilization of mRNA expression in wholeblood samples. Clin Chem.2002;48:1883–1890.

14. Thorn I, Olsson-Stromberg U, Ohlsen C, et al. The impact of RNA stabilization on min-imal residual disease assessment in chronic myeloid leukemia. Haematologica.2005;90:1471–1476.

15. Vaught JB. Blood collection, shipment, processing, and storage. Cancer EpidemiolBiomarkers Prev.2006;15:1582–1584.

16. Yamamoto T, Sekiyama A, Sekiguchi H, et al. Examination of stability of bone marrowblood RNA in the PAXgene tube. Lab Hematol.2006;12:143–147.

17. Burjanivova T, Madzo J, Muzikova K, et al. Prenatal origin of childhood AML occurs lessfrequently than in childhood ALL. BMC Cancer.2006;6:100.

18. Fasching K, Panzer S, Haas OA, et al. Presence of N regions in the clonotypicDJ rearrangements of the immunoglobulin heavy-chain genes indicates an exquisitelyshort latency in t(4;11)-positive infant acute lymphoblastic leukemia. Blood. 2001;98:2272–2274.

19. Gale KB, Ford AM, Repp R, et al. Backtracking leukemia to birth: identification ofclonotypic gene fusion sequences in neonatal blood spots. Proc Natl Acad Sci USA.1997;94:13950–13954.

20. McHale CM, Wiemels JL, Zhang L, et al. Prenatal origin of childhood acutemyeloid leukemias harboring chromosomal rearrangements t(15;17) and inv(16). Blood.2003;101:4640–4641.

21. Sjoholm MI, Dillner J, Carlson J. Assessing quality and functionality of DNA from fresh andarchival dried blood spots and recommendations for quality control guidelines. Clin Chem.2007;53:1401–1407.

22. Bogner PN, Killeen AA. Extraction of nucleic acids. In: Coleman WB, Tsongalis GJ, eds.Molecular Diagnostics for the Clinical Laboratorian. 2nd ed. Totowa, NJ: Humana PressInc.; 2006. pp. 25–30.

23. Crisan D, Mattson JC. Retrospective DNA analysis using fixed tissue specimens. DNA CellBiol. 1993;12:455–464.

24. Hunt JL. Molecular pathology in anatomic pathology practice: a review of basic principles.Arch Pathol Lab Med. 2008;132:248–260.

25. Bonin S, Petrera F, Rosai J, et al. DNA and RNA obtained from Bouin′s fixed tissues. J ClinPathol. 2005;58:313–316.

26. Douglas MP, Rogers SO. DNA damage caused by common cytological fixatives. Mutat Res.1998;401:77–88.

27. Falconi M, Teti G, Zago M, et al. Effect of fixative on chromatin structure and DNAdetection. Microsc Res Tech. 2007;70:599–606.

28. Benchekroun M, DeGraw J, Gao J, et al. Impact of fixative on recovery of mRNA fromparaffin-embedded tissue. Diagn Mol Pathol. 2004;13:116–125.


29. Cox ML, Schray CL, Luster CN, et al. Assessment of fixatives, fixation, and tissueprocessing on morphology and RNA integrity. Exp Mol Pathol. 2006;80:183–191.

30. Perlmutter MA, Best CJ, Gillespie JW, et al. Comparison of snap freezing versusethanol fixation for gene expression profiling of tissue specimens. J Mol Diagn. 2004;6:371–377.

31. Shibutani M, Uneyama C, Miyazaki K, et al. Methacarn fixation: a novel tool for anal-ysis of gene expressions in paraffin-embedded tissue specimens. Lab Invest. 2000;80:199–208.

32. Urieli-Shoval S, Meek RL, Hanson RH, et al. Preservation of RNA for in situ hybridization:Carnoy’s versus formaldehyde fixation. J Histochem Cytochem. 1992;40:1879–1885.

33. Tubbs RR, Hsi ED, Hicks D, et al. Molecular pathology testing of tissues fixed in prefersolution. Am J Surg Pathol. 2004;28:417–419.

34. Giannella C, Zito FA, Colonna F, et al. Comparison of formalin, ethanol, and Histochoicefixation on the PCR amplification from paraffin-embedded breast cancer tissue. Eur J ClinChem Clin Biochem. 1997;35:633–635.

35. Goldmann T, Dromann D, Marzouki M, et al. Tissue microarrays from HOPE-fixed speci-mens allow for enhanced high throughput molecular analyses in paraffin-embedded material.Pathol Res Pract. 2005;201:599–602.

36. Lykidis D, Van Noorden S, Armstrong A, et al. Novel zinc-based fixative for high qualityDNA, RNA and protein analysis. Nucleic Acids Res. 2007;35:e85.

37. Sen Gupta R, Hillemann D, Kubica T, et al. HOPE-fixation enables improved PCR-baseddetection and differentiation of Mycobacterium tuberculosis complex in paraffin-embeddedtissues. Pathol Res Pract. 2003;199:619–623.

38. Stanta G, Mucelli SP, Petrera F, et al. A novel fixative improves opportunities of nucleic acidsand proteomic analysis in human archive’s tissues. Diagn Mol Pathol. 2006;15:115–123.

39. Vincek V, Nassiri M, Block N, et al. Methodology for preservation of high molecular-weightRNA in paraffin-embedded tissue: application for laser-capture microdissection. Diagn MolPathol. 2005;14:127–133.

40. Vincek V, Nassiri M, Nadji M, et al. A tissue fixative that protects macromolecules(DNA, RNA, and protein) and histomorphology in clinical samples. Lab Invest. 2003;83:1427–1435.

41. Castiglione F, Degl’Innocenti DR, Taddei A, et al. Real-time PCR analysis of RNA extractedfrom formalin-fixed and paraffin-embedded tissues: effects of the fixation on outcomereliability. Appl Immunohistochem Mol Morphol. 2007;15:338–342.

42. Macabeo-Ong M, Ginzinger DG, Dekker N, et al. Effect of duration of fixation on quan-titative reverse transcription polymerase chain reaction analyses. Mod Pathol. 2002;15:979–987.

43. Gillio-Tos A, De Marco L, Fiano V, et al. Efficient DNA extraction from 25-year-oldparaffin-embedded tissues: study of 365 samples. Pathology. 2007;39:345–348.

44. Smith-Zagone MJ, Pulliam J, Farkas DH. Molecular pathology methods. In: Leonard DGB,ed. Molecular Pathology in Clinical Practice. New York: Springer Science+Business Media,LLC; 2007. pp. 15–40.

45. Farkas DH, Kaul KL, Wiedbrauk DL, et al. Specimen collection and storage for diagnosticmolecular pathology investigation. Arch Pathol Lab Med. 1996;120:591–596.

46. Gustafson S, Proper JA, Bowie EJ, et al. Parameters affecting the yield of DNA from humanblood. Anal Biochem. 1987;165:294–299.

47. Madisen L, Hoar DI, Holroyd CD, et al. DNA banking: the effects of storage of blood andisolated DNA on the integrity of DNA. Am J Med Genet. 1987;27:379–390.

48. Schaudien D, Baumgartner W, Herden C. High preservation of DNA standards diluted in50% glycerol. Diagn Mol Pathol. 2007;16:153–157.

49. Wilfinger WW, Mackey K, Chomczynski P. Effect of pH and ionic strength on thespectrophotometric assessment of nucleic acid purity. Biotechniques. 1997;22:474-476,478–481.


50. Ahn SJ, Costa J, Emanuel JR. PicoGreen quantitation of DNA: effective evaluation ofsamples pre- or post-PCR. Nucleic Acids Res. 1996;24:2623–2625.

51. Teare JM, Islam R, Flanagan R, et al. Measurement of nucleic acid concentrations using theDyNA Quant and the GeneQuant. Biotechniques. 1997;22:1170–1174.

52. English CA, Merson S, Keer JT. Use of elemental analysis to determine comparativeperformance of established DNA quantification methods. Anal Chem. 2006;78:4630–4633.

53. Holden MJ, Rabb SA, Tewari YB, et al. Traceable phosphorus measurements by ICP-OESand HPLC for the quantitation of DNA. Anal Chem. 2007;79:1536–1541.

54. Yang I, Han MS, Yim YH, et al. A strategy for establishing accurate quantitation standardsof oligonucleotides: quantitation of phosphorus of DNA phosphodiester bonds using induc-tively coupled plasma-optical emission spectroscopy. Anal Biochem. 2004;335:150–161.

55. Panaro NJ, Yuen PK, Sakazume T, et al. Evaluation of DNA fragment sizing and quantifica-tion by the agilent 2100 bioanalyzer. Clin Chem. 2000;46:1851–1853.

56. Highsmith Jr. W. Electrophoretic methods for mutation detection and DNA sequencing. In:Coleman WB, Tsongalis GJ, eds. Molecular Diagnostics: For the Clinical Laboratorian.Totowa, NJ: Humana Press, Inc.; 2006. pp. 85–109.

57. Swerdlow H, Gesteland R. Capillary gel electrophoresis for rapid, high resolution DNAsequencing. Nucleic Acids Res. 1990;18:1415–1419.

58. Beaubier NT, Hart AP, Bartolo C, et al. Comparison of capillary electrophoresis and poly-acrylamide gel electrophoresis for the evaluation of T and B cell clonality by polymerasechain reaction. Diagn Mol Pathol. 2000;9:121–131.

59. Noguera NI, Ammatuna E, Zangrilli D, et al. Simultaneous detection of NPM1 andFLT3-ITD mutations by capillary electrophoresis in acute myeloid leukemia. Leukemia.2005;19:1479–1482.

60. Luthra R, Sanchez-Vega B, Medeiros LJ. TaqMan RT-PCR assay coupled with capillaryelectrophoresis for quantification and identification of bcr-abl transcript type. Mod Pathol.2004;17:96–103.

61. Yuan R. Structure and mechanism of multifunctional restriction endonucleases. Annu RevBiochem. 1981;50:285–319.

62. Roberts RJ, Vincze T, Posfai J, Macelis D. REBASE–enzymes and genes for DNA restrictionand modification. Nucleic Acids Res. 2007;35:D269–D270.

63. Villa-Komaroff L, Efstratiadis A, Broome S, et al. A bacterial clone synthesizing proinsulin.Proc Natl Acad Sci USA. 1978;75:3727–3731.

64. Kottaridis PD, Gale RE, Langabeer SE, et al. Studies of FLT3 mutations in paired presen-tation and relapse samples from patients with acute myeloid leukemia: implications for therole of FLT3 mutations in leukemogenesis, minimal residual disease detection, and possibletherapy with FLT3 inhibitors. Blood. 2002;100:2393–2398.

65. Campbell PJ, Scott LM, Baxter EJ, et al. Methods for the detection of the JAK2V617F mutation in human myeloproliferative disorders. Methods Mol Med. 2006;125:253–264.

66. Maxam AM, Gilbert W. A new method for sequencing DNA. Proc Natl Acad Sci USA.1977;74:560–564.

67. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. ProcNatl Acad Sci USA. 1977;74:5463–5467.

68. Smith LM, Fung S, Hunkapiller MW, et al. The synthesis of oligonucleotides containing analiphatic amino group at the 5′ terminus: synthesis of fluorescent DNA primers for use inDNA sequence analysis. Nucleic Acids Res. 1985;13:2399–2412.

69. Smith LM, Sanders JZ, Kaiser RJ, et al. Fluorescence detection in automated DNA sequenceanalysis. Nature. 1986;321:674–679.

70. Ju J, Ruan C, Fuller CW, et al. Fluorescence energy transfer dye-labeled primers for DNAsequencing and analysis. Proc Natl Acad Sci USA. 1995;92:4347–4351.

71. Lee LG, Spurgeon SL, Heiner CR, et al. New energy transfer dyes for DNA sequencing.Nucleic Acids Res. 1997;25:2816–2822.


72. Rosenblum BB, Lee LG, Spurgeon SL, et al. New dye-labeled terminators for improvedDNA sequencing patterns. Nucleic Acids Res. 1997;25:4500–4504.

73. Zakeri H, Amparo G, Chen SM, Spurgeon S, Kwok PY. Peak height pattern in dichloro-rhodamine and energy transfer dye terminator sequencing. Biotechniques. 1998;25:406–410,412–404.

74. Baxter EJ, Scott LM, Campbell PJ, et al. Acquired mutation of the tyrosine kinase JAK2 inhuman myeloproliferative disorders. Lancet. 2005;365:1054–1061.

75. Smith TA, Whelan J, Parry PJ. Detection of single-base mutations in a mixed populationof cells: a comparison of SSCP and direct sequencing. Genet Anal Tech Appl. 1992;9:143–145.

76. Steensma DP. JAK2 V617F in myeloid disorders: molecular diagnostic techniques and theirclinical utility: a paper from the 2005 William Beaumont Hospital Symposium on MolecularPathology. J Mol Diagn. 2006;8:397–411.

77. Melamede R, Melamede RMelamede Rs. U.S. Patent 4863849; 1985.78. Hyman ED. A new method of sequencing DNA. Anal Biochem. 1988;174:423–436.79. Nyren P. Enzymatic method for continuous monitoring of DNA polymerase activity. Anal

Biochem. 1987;167:235–238.80. Ronaghi M, Karamohamed S, Pettersson B, et al. Real-time DNA sequencing using detection

of pyrophosphate release. Anal Biochem. 1996;242:84–89.81. Ronaghi M, Uhlen M, Nyren P. A sequencing method based on real-time pyrophosphate.

Science. 1998;281:363, 365.82. Klenow H, Overgaard-Hansen K, Patkar SA. Proteolytic cleavage of native DNA poly-

merase into two different catalytic fragments. Influence of assay conditions on the changeof exonuclease activity and polymerase activity accompanying cleavage. Eur J Biochem.1971;22:371–381.

83. Segel IH, Renosto F, Seubert PA. Sulfate-activating enzymes. Methods Enzymol.1987;143:334–349.

84. Deluca M. Firefly luciferase. Adv Enzymol Relat Areas Mol Biol. 1976;44:37–68.85. Komoszynski M, Wojtczak A. Apyrases (ATP diphosphohydrolases, EC 3.6.1.5): function

and relationship to ATPases. Biochim Biophys Acta. 1996;1310:233–241.86. Boyanton Jr. BL, Luna RA, Fasciano LR, Menne KG, Versalovic J. DNA pyrosequencing-

based identification of pathogenic Candida species by using the internal transcribed spacer2 region. Arch Pathol Lab Med. 2008;132:667–674.

87. Ahmadian A, Ehn M, Hober S. Pyrosequencing: history, biochemistry and future. Clin ChimActa. 2006;363:83–94.

88. Ahmadian A, Gharizadeh B, Gustafsson AC, et al. Single-nucleotide polymorphism analysisby pyrosequencing. Anal Biochem. 2000;280:103–110.

89. Jelinek J, Oki Y, Gharibyan V, et al. JAK2 mutation 1849G>T is rare in acute leukemiasbut can be found in CMML, Philadelphia chromosome-negative CML, and megakaryocyticleukemia. Blood. 2005;106:3370–3373.

90. Mc Lornan DP, Percy MJ, Jones AV, et al. Chronic neutrophilic leukemia with an associatedV617F JAK2 tyrosine kinase mutation. Haematologica. 2005;90:1696–1697.

91. Khorashad JS, Milojkovic D, Mehta P, et al. In vivo kinetics of kinase domain mutations inCML patients treated with dasatinib after failing imatinib. Blood. 2008;111:2378–2381.

92. Bohne A, Schlee C, Mossner M, et al. Epigenetic control of differential expression of specificERG isoforms in acute T-lymphoblastic leukemia. Leuk Res. 2009;33:817–822.

93. Brakensiek K, Wingen LU, Langer F, et al. Quantitative high-resolution CpG island mappingwith pyrosequencing reveals disease-specific methylation patterns of the CDKN2B gene inmyelodysplastic syndrome and myeloid leukemia. Clin Chem. 2007;53:17–23.

94. Choi SH, Worswick S, Byun HM, et al. Changes in DNA methylation of tandem DNArepeats are different from interspersed repeats in cancer. Int J Cancer. 2009;125:723–729.

95. Voelkerding KV, Dames SA, Durtschi JD. Next-generation sequencing: from basic researchto diagnostics. Clin Chem. 2009;55:641–658.


96. Wheeler DA, Srinivasan M, Egholm M, et al. The complete genome of an individual bymassively parallel DNA sequencing. Nature. 2008;452:872–876.

97. Ansorge WJ. Next-generation DNA sequencing techniques. Nat Biotechnol. 2009;25:195-203.

98. Morozova O, Marra MA. Applications of next-generation sequencing technologies infunctional genomics. Genomics. 2008;92:255–264.

99. Hert DG, Fredlake CP, Barron AE. Advantages and limitations of next-generation sequenc-ing technologies: a comparison of electrophoresis and non-electrophoresis methods.Electrophoresis. 2008;29:4618–4626.

100. Mardis ER. Next-generation DNA sequencing methods. Annu Rev Genomics Hum Genet.2008;9:387–402.

101. Li M, Chen WD, Papadopoulos N, et al. Sensitive digital quantification of DNA methylationin clinical samples. Nat Biotechnol. 2009;27:858–863.

102. Neff T, Armstrong SA. Chromatin maps, histone modifications and leukemia. Leukemia.2009;23:1243–1251.

103. Kato K. Impact of the next generation DNA sequences. Int J Clin Exp Med. 2009;2:193–202.104. Marguerat S, Wilhelm BT, Bahler J. Next-generation sequencing: applications beyond

genomes. Biochem Soc Trans. 2008;36:1091–1096.105. Morozova O, Marra MA. From cytogenetics to next-generation sequencing technolo-

gies: advances in the detection of genome rearrangements in tumors. Biochem Cell Biol.2008;86:81–91.

106. Mullis K, Faloona F, Scharf S, et al. Specific enzymatic amplification of DNA invitro: the polymerase chain reaction. Cold Spring Harb Symp Quant Biol. 1986;51:263–273.

107. Saiki RK, Gelfand DH, Stoffel S, et al. Primer-directed enzymatic amplification of DNAwith a thermostable DNA polymerase. Science. 1988;239:487–491.

108. Myers TW, Gelfand DH. Reverse transcription and DNA amplification by a Thermusthermophilus DNA polymerase. Biochemistry. 1991;30:7661–7666.

109. Newton CR, Graham A, Heptinstall LE, et al. Analysis of any point mutation in DNA.The amplification refractory mutation system (ARMS). Nucleic Acids Res. 1989;17:2503–2516.

110. McClure R, Mai M, Lasho T. Validation of two clinically useful assays for evaluationof JAK2 V617F mutation in chronic myeloproliferative disorders. Leukemia. 2006;20:168–171.

111. Kang HY, Hwang JY, Kim SH, et al. Comparison of allele specific oligonucleotide-polymerase chain reaction and direct sequencing for high throughput screening of ABLkinase domain mutations in chronic myeloid leukemia resistant to imatinib. Haematologica.2006;91:659–662.

112. Haqqi TM, Sarkar G, David CS, et al. Specific amplification with PCR of a refractorysegment of genomic DNA. Nucleic Acids Res. 1988;16:11844.

113. Morley AA, Latham S, Brisco MJ, et al. Sensitive and specific measurement of minimalresidual disease in acute lymphoblastic leukemia. J Mol Diagn. 2009;11:201–210.

114. Farina L, Carniti C, Dodero A, et al. Qualitative and quantitative polymerase chain reactionmonitoring of minimal residual disease in relapsed chronic lymphocytic leukemia: earlyassessment can predict long-term outcome after reduced intensity allogeneic transplantation.Haematologica. 2009;94:654–662.

115. Heid CA, Stevens J, Livak KJ, Williams PM. Real time quantitative PCR. Genome Res.1996;6:986–994.

116. Wittwer CT, Herrmann MG, Moss AA, et al. Continuous fluorescence monitoring of rapidcycle DNA amplification. Biotechniques. 1997;22:130-131, 134–138.

117. Ririe KM, Rasmussen RP, Wittwer CT. Product differentiation by analysis of DNAmelting curves during the polymerase chain reaction. Anal Biochem. 1997;245:154–160.


118. Holland PM, Abramson RD, Watson R, et al. Detection of specific polymerase chain reactionproduct by utilizing the 5′-3′ exonuclease activity of Thermus aquaticus DNA polymerase.Proc Natl Acad Sci USA. 1991;88:7276–7280.

119. Tyagi S, Bratu DP, Kramer FR. Multicolor molecular beacons for allele discrimination. NatBiotechnol. 1998;16:49–53.

120. Thelwell N, Millington S, Solinas A, et al. Mode of action and application of Scorpionprimers to mutation detection. Nucleic Acids Res. 2000;28:3752–3761.

121. Sidon P, Heimann P, Lambert F, et al. Combined locked nucleic acid and molecular beacontechnologies for sensitive detection of the JAK2V617F somatic single-base sequence variant.Clin Chem. 2006;52:1436–1438.

122. Martinez-Lopez J, Lahuerta JJ, Salama P, et al. The use of fluorescent molecular beacons inreal time PCR of IgH gene rearrangements for quantitative evaluation of multiple myeloma.Clin Lab Haematol. 2004;26:31–35.

123. van der Velden VH, Hochhaus A, Cazzaniga G, et al. Detection of minimal residual dis-ease in hematologic malignancies by real-time quantitative PCR: principles, approaches,and laboratory aspects. Leukemia. 2003;17:1013–1034.

124. Lucia E, Martino B, Mammi C, et al. The incidence of JAK2 V617F mutation inbcr/abl-negative chronic myeloproliferative disorders: assessment by two different detectionmethods. Leuk Lymphoma. 2008;49:1907–1915.

125. Herman JG, Graff JR, Myohanen S, et al. Methylation-specific PCR: a novel PCRassay for methylation status of CpG islands. Proc Natl Acad Sci USA. 1996;93:9821–9826.

126. Mariappan MR AD. Molecular diagnostics in hematopathology. In: Pfeifer JD, ed.Molecular Genetic Testing in Surgical Pathology. 1st ed. Philadelphia, PA: LippincottWilliams & Wilkins; 2006. pp. 250–274.

127. Bassing CH, Swat W, Alt FW. The mechanism and regulation of chromosomal V(D)Jrecombination. Cell. 2002;109:S45–S55.

128. Macintyre EA, Delabesse E. Molecular approaches to the diagnosis and evaluation oflymphoid malignancies. Semin Hematol. 1999;36:373–389.

129. Bagg A, Braziel RM, Arber DA, et al. Immunoglobulin heavy chain gene analysis in lym-phomas: a multi-center study demonstrating the heterogeneity of performance of polymerasechain reaction assays. J Mol Diagn. 2002;4:81–89.

130. van Dongen JJ, Langerak AW, Bruggemann M, et al. Design and standardization of PCRprimers and protocols for detection of clonal immunoglobulin and T-cell receptor generecombinations in suspect lymphoproliferations: report of the BIOMED-2 Concerted ActionBMH4-CT98-3936. Leukemia. 2003;17:2257–2317.

131. van Krieken JH, Langerak AW, San Miguel JF, et al. Clonality analysis for antigen receptorgenes: preliminary results from the Biomed-2 concerted action PL 96-3936. Hum Pathol.2003;34:359–361.

132. Greiner TC, Raffeld M, Lutz C, et al. Analysis of T cell receptor-gamma gene rearrange-ments by denaturing gradient gel electrophoresis of GC-clamped polymerase chain reactionproducts. Correlation with tumor-specific sequences. Am J Pathol. 1995;146:46–55.

133. Greiner TC, Rubocki RJ. Effectiveness of capillary electrophoresis using fluorescent-labeledprimers in detecting T-cell receptor gamma gene rearrangements. J Mol Diagn. 2002;4:137–143.

134. Theodorou I, Bigorgne C, Delfau MH, et al. VJ rearrangements of the TCR gamma locus inperipheral T-cell lymphomas: analysis by polymerase chain reaction and denaturing gradientgel electrophoresis. J Pathol. 1996;178:303–310.

135. Elenitoba-Johnson KS, Bohling SD, Mitchell RS, et al. PCR analysis of the immunoglobulinheavy chain gene in polyclonal processes can yield pseudoclonal bands as an artifact of lowB cell number. J Mol Diagn. 2000;2:92–96.

136. Lee SC, Berg KD, Racke FK, et al. Pseudo-spikes are common in histologically benignlymphoid tissues. J Mol Diagn. 2000;2:145–152.


137. Nihal M, Mikkola D, Wood GS. Detection of clonally restricted immunoglobulin heavychain gene rearrangements in normal and lesional skin: analysis of the B cell compo-nent of the skin-associated lymphoid tissue and implications for the molecular diagnosisof cutaneous B cell lymphomas. J Mol Diagn. 2000;2:5–10.

138. Pan L CE, Knowles DM. Antigen receptor genes: structure, function, and genetic analysis oftheir rearrangements. In: Knowles DM, ed. Neoplastic Hematopathology. Philadelphia, PA:Lippincott Williams & Wilkins; 2001. pp. 307–328.

139. Viswanatha DS, Larson RS. Molecular diagnosis of hematopoietic neoplasms. In:McPherson RA, Pincus MR, eds. Henry’s Clinical Diagnosis and Management byLaboratory Methods. 21st ed. Philadelphia, PA: Saunders Elsevier; 2007. pp. 1295–1322.

Chapter 2Classical and Molecular Cytogenetic Analysisof Hematolymphoid Disorders

Mark A. Micale

Keywords Myelodysplastic/myeloproliferative disorders · Leukemia ·Lymphoma · Fluorescence in situ hybridization · Array comparative genomichybridization (array CGH) · WHO Classification of Tumors of Hematopoieticand Lymphoid Tissues · Hematolymphoid neoplasms · Philadelphia chromo-some · Chronic myelogenous leukemia (CML) · Burkitt lymphoma · BCR/ABL ·PML/RARA · Acute promyelocytic leukemia · Chromosome microar-ray analysis · International System of Human Cytogenetic Nomenclature(2009) · Illegitimate V(D)J or switch recombination · ALU sequences · LINEelements · Error-prone non-homologous end joining · Translin-binding consen-sus sequences · Scaffold-associated regions · Centromere enumeration probes(CEPs) · Locus-specific identifier (LSI) probes · Whole-chromosome paint(WCP) probes · Dual-color · dual-fusion (DCDF) LSI probes · InterphaseFISH · Paraffin FISH · Myeloproliferative neoplasms (MPNs) · JAK2V617F mutation · FIP1L1-PDGFRA · ASS gene · Imatinib mesylate(Gleevec) · Postpolycythemic myelofibrosis · Essential thrombocythemia · Primarymyelofibrosis · HMGA2 gene · Chronic neutrophilic leukemia · Normalkaryotype · Chronic eosinophilic leukemia/idiopathic hypereosinophilicsyndrome · Myeloid and lymphoid neoplasms with PDGFRA rearrange-ments · Polycythemia vera · Hypereosinophilia · CHIC2 gene · Myelodysplasticsyndrome · Chromosome 5q deletion · MDS associated with isolated del(5q) · Lossof the Y chromosome · MDS-FISH panel · Monosomy 5/del(5q) · Monosomy7/del(7q) · Chromosome 11q deletion · Chromosome 13q deletion · Acute myeloidleukemia (AML) · Therapy-related- or t-AML · AML with t(8;21)(q22;q22) –RUNX1/RUNX1T1 · AML (promyelocytic) with t(15;17)(q22;q12) –PML/RARα · ZBTB16/RARα · NPM1/RARα · NUMA1/RARα · RARα

gene · All-trans-retinoic acid (ATRA) · AML with t(9;11)(p22;q23) –MLLT3/MLL · MLL (myeloid lymphoid lineage or mixed lineage leukemia)

M.A. Micale (B)Beaumont Laboratory, Department of Anatomic Pathology, Beaumont Hospitals,3601 W. Thirteen Mile Rd, Royal Oak, MI 48073, USAe-mail: [email protected]


40 M.A. Micale

gene · t(9;11)(p21;q23) · t(11;19)(q23;p13.1) · t(11;19)(q23;p13.3) · MLLgene break-apart probe · AML with inv(16)(p13q22) or t(16;16)(p13;q22) –CBFβ/MYH · acute myelomonocytic leukemia (AMML) · Core binding factorbeta subunit (CBFβ) · AML with t(6;9)(p23;q34) – DEK/NUP214 · Multilineagedysplasia · FLT3-ITD · AML with inv(3)(q21q26.2) or t(3;3)(q21;q26.2) –RPN1/EVI1 · AML (megakaryoblastic) with t(1;22)(p13;q13) –RBM15-MKL1 · Acute myeloid leukemia with myelodysplasia-relatedchanges · Therapy-related myeloid neoplasms · Alkylating agent · TopoisomeraseII inhibitor therapy · Acute myeloid leukemia · not otherwise speci-fied · Acute myeloid leukemia with minimal differentiation · Acute myeloidleukemia without maturation · Acute myeloid leukemia with matura-tion · Acute myelomonocytic leukemia · Acute monoblastic and monocyticleukemia · Acute erythroid leukemia · Acute megakaryoblastic leukemia · Acutebasophilic leukemia · Acute panmyelosis with myelofibrosis · B-lymphoblasticleukemia/lymphoma with recurrent genetic abnormalities · Acute lymphoblas-tic leukemia (ALL) · Childhood ALL · B-lymphoblastic leukemia/lymphomawith hyperdiploidy · Hyperdiploid ALL · Hypodiploid ALL · B-lymphoblasticleukemia/lymphoma with hypodiploidy · B-lymphoblastic leukemia/lymphomawith t(12;21)(p13;q22) · TEL/AML1 (ETV6/RUNX1) · TEL (ETV6) gene · AML1(CBFA2 or RUNX1) gene · B-lymphoblastic leukemia/lymphoma witht(9;22)(q34;q11.2) · BCR/ABL · B-lymphoblastic leukemia/lymphomawith t(v;11q23) · MLL rearranged · B-lymphoblastic leukemia/lymphomawith t(1;19)(q23;p13.3) · ETA/PBX1 (TCF3/PBX1) · B-lymphoblasticleukemia/lymphoma with t(5;14)(q31;q32) · IL3/IGH · Children’s OncologyGroup (COG) · T-lymphoblastic leukemia/lymphoma · T-cell receptor(TCR) genes · chronic lymphocytic leukemia/small lymphocytic lym-phoma · MYB gene · ATM gene D13S319 locus · LAMP1 gene · p53gene · CLL FISH panel · Plasmacell myeloma · IgH gene rearrange-ments · Monoclonal gammopathy of undetermined significance · plasma cellleukemia · Monosomy 13/del(13q), t(11;14)(q13;q32), t(4;14)(p16.3;q32),t(14;16)(q32;q23) · FGFR3/IgH · MAF/IgH · C-MYC gene · IgH/CCND1 · CCND1gene · Plasma cell myeloma FISH panel · Non-Hodgkin lymphoma · API2-MALT1 · MALT lymphoma · Burkitt lymphoma · t(8;14)(q24;q32) · Mantle celllymphoma v t(11;14)(q13;q32) · Diffuse large B-cell lymphoma · Complexkaryotype · Follicular lymphoma · t(14;18)(q32;q21) · BCL2 gene · IgHgene · BCL6 gene · t(2;8)(p12;q24), t(8;22)(q24;q11) · Splenic marginal zonelymphoma · Extranodal marginal zone B-cell lymphoma of mucosa-associatedlymphoid tissue · Anaplastic large-cell lymphoma · t(2;5)(p23;q35) · Anaplasticlymphoma kinase (ALK) gene

Cytogenetic Analysis in the Diagnosis of HematolymphoidDisorders

Non-random chromosomal abnormalities are a common feature of many hema-tolymphoid disorders and are a key component of their pathogenesis. As such,

2 Classical and Molecular Cytogenetic Analysis of Hematolymphoid Disorders 41

routine chromosome analysis is critical in the laboratory workup of most knownor suspected myelodysplastic/myeloproliferative disorders, leukemias, and lym-phomas. Such studies can provide (1) diagnostic confirmation; (2) informationuseful for classification, staging, and prognostication; (3) information to guideappropriate choice of therapy; and (4) evidence of remission or relapse. In lymphnode evaluation, cytogenetics can differentiate a reactive process from a malignantcondition.

With the continued evolution of genetic laboratory methodologies, highly sen-sitive techniques have become commonplace in the laboratory workup of hema-tolymphoid disorders, including fluorescence in situ hybridization (FISH) andpolymerase chain reaction (PCR). These technologies do not, however, providethe genome-wide coverage afforded by classical cytogenetics. Array compara-tive genomic hybridization (array CGH) promises the opportunity to study thesemalignancies in a genome-wide fashion and at a level of resolution not previ-ously achievable by conventional cytogenetics. Using cytogenetic and moleculardata, along with morphology and immunophenotype, hematolymphoid neoplasmscan now be classified into clinically relevant categories that greatly improve tumorclassification.

While earlier disease classification schemes included primarily clinical fea-tures, morphology, and immunophenotype, the recent advances in cytogenetic andmolecular genetic analysis have greatly refined this process. Earlier attempts toclassify myeloid and lymphoid neoplasms into meaningful subgroups resulted inthe French–American–British (FAB) scheme proposed in 1976, which was basedprimarily on tissue morphology. Later revisions of the FAB system took intoaccount immunocytochemical reactions of neoplastic cells; however, bone mar-row morphology continued to be the backbone of classification. The “RevisedEuropean–American Classification of Lymphoid Neoplasms” (REAL) in 1994extended the basis for classifying lymphoid neoplasms to include morphologic,immunologic, and genetic features; clinical presentations and disease course; andpostulated normal cellular counterpart [1]. In 1997, the World Health Organization(WHO) released its first edition of a classification scheme developed jointly bypathologists, hematologists, and oncologists for hematologic malignancies [2]. Thisclassification recognized specific disease entities based on a combination of mor-phologic and cytogenetic features. More recent editions of the WHO Classificationof Tumors of Hematopoietic and Lymphoid Tissues have included information aboutimmunophenotype and molecular abnormalities. The latest edition (4th edition)published in 2008 [3] categorizes hematolymphoid neoplasms based on clinical andbiological features, morphology, immunophenotype, cytogenetic abnormalities, andmolecular genetic mutations. While certain conditions have specific immunopheno-typic, cytogenetic, and/or molecular features, some myeloid and many lymphoiddisorders demonstrate chromosomal abnormalities that may be observed in a num-ber of entities. Nevertheless, these “non-specific” abnormalities can still provideimportant prognostic information that may guide choice of treatment. In addi-tion, the classification of hematolymphoid neoplasms based on the multiple criteriadescribed above has led to elucidation of involved genes and pathways, which hasbeen critical for the development of “molecularly targeted” therapeutics.

42 M.A. Micale

Classical Cytogenetic Analysis of Bone Marrow and LeukemicBlood

Historical Perspectives

Despite the original description of chromosomes by Professor Walther Flemmingin 1876 [4], it would be some 80 years later before the correct number of humanchromosomes in a cell (46) was elucidated by Tjio and Levan [5]. Since then,the field of cytogenetics has witnessed several eras that have ushered in new andexciting discoveries that have dramatically improved the diagnostic capabilities ofcytogenetics laboratories and made it an important specialty of medical genetics.The first cytogenetic abnormality associated with a specific type of malignancy wasdescribed by Nowell and Hungerford in 1960 [6]. The marker chromosome, namedthe Philadelphia chromosome for the city where it was identified, was associatedwith chronic myelogenous leukemia (CML). Utilizing better banding techniques,Janet Rowley at the University of Chicago later identified this marker as a derivativechromosome 22 originating from a reciprocal translocation between chromosomes9 and 22 [t(9;22)(q34;q11.2)] [7]. Burkitt lymphoma was the first lymphoid neo-plasm in which a characteristic chromosomal abnormality [t(8;14)(q24;q32)] wasidentified [8].

The “banding era” of the late 1960s and 1970s resulted in improved visualizationof the human chromosome complement through the formation of unique bandingpatterns for each chromosome. The development of quinacrine banding, Giemsabanding, C-banding using barium hydroxide and reverse banding using acridineorange permitted delineation of individual chromosomes which improved the capa-bility of cytogenetics laboratories to more accurately define numerical and structuralchromosomal abnormalities.

The introduction of in situ hybridization methodologies in cytogenetics utilizedDNA probes labeled with biotin and detected by sequential hybridizations withstreptavidin–horseradish peroxidase and diaminobenzidine followed by visualiza-tion using standard bright-field microscopy. A slight modification of this enzymaticISH procedure, known as chromogenic in situ hybridization (CISH), utilized flu-orescently labeled DNA probes. This technique, known as fluorescence in situhybridization (FISH), initially used single-fluorophore DNA probes and appliedthem to standard chromosome preparations for chromosome enumeration. As moresingle-copy FISH probes became commercially available, the diagnostic utilityof FISH in the clinical cytogenetics laboratory increased. This technique becameeven more powerful as multicolor FISH probes became commonplace, permittingthe identification of characteristic hematolymphoid chromosomal rearrangementssuch as the BCR/ABL fusion gene associated with the t(9;22)(q34;q11.2) in CMLand the PML/RARA fusion gene associated with the t(15;17)(q22;q12) in acutepromyelocytic leukemia. Additionally, because FISH did not require dividing cells,chromosomal abnormalities could be identified in non-dividing cells, includingthose in paraffin sections where tissue architecture is retained. As time went


on, more sophisticated molecular cytogenetic techniques were developed. Theseincluded comparative genomic hybridization (CGH) [9], spectral karyotyping [10],and Fiber FISH [11]. While these techniques extended the diagnostic capabilities ofFISH, their technical complexity precluded their routine implementation in manycytogenetics laboratories.

With the approach of the 21st century, clinical cytogenetics found itself atyet another crossroads, defined by a powerful new diagnostic assay that trulyblurs the line of demarcation between molecular genetic and cytogenetic analy-sis. Chromosome microarray analysis, or microarray CGH, is akin to a multiplexFISH experiment utilizing thousands of individual DNA probes arrayed to a glassslide. The “microarray era” has witnessed a substantial improvement in the diag-nostic capability for identifying small (less than 5 Mb) unbalanced constitutionalchromosomal rearrangements and has found great utility in the workup of childrenwith developmental delay, mental retardation, autism/autism spectrum disorder, andmultiple congenital anomalies [12, 13]. Recent literature has also demonstrated thatthis technology (using SNP arrays) will have a significant impact on the cytogeneticworkup of hematolymphoid disorders, permitting detection of molecular mecha-nisms of tumorigenesis such as copy-number neutral loss of heterozygosity thatcannot be identified using other cytogenetic methodologies [14, 15].

Specimen Collection and Storage

Bone marrow is the tissue of choice for chromosome analysis in most hematolog-ical disorders including myeloproliferative neoplasms, myelodysplastic syndrome,chronic lymphocytic leukemia, and acute leukemias. Collection of 1–2 ml of bonemarrow aspirate is adequate in most cases; however, a smaller sample may beacceptable if the marrow is hypercellular. If a bone marrow aspirate cannot beobtained, a bone core biopsy can be processed; however, the success rate for obtain-ing cytogenetic data on such a specimen is lower than that for a marrow aspirate. Inpatients with a white blood cell count greater than 10,000 billion/l and at least 10%circulating blast cells, a peripheral blood specimen cultured without phytohemag-glutinin (PHA) can be studied. PHA will stimulate division of nonmalignant cellswhich can potentially interfere with the analysis of spontaneously dividing neoplas-tic cells. For lymphoma, sampling an involved lymph node is the method of choice.Cytogenetic analysis of bone marrow in lymphoid malignancies will yield positiveresults only if the bone marrow is involved as well; however, lymphoid-associatedchromosomal abnormalities can sometimes be identified in bone marrow specimenswithout any overt morphological evidence of lymphoma involvement.

Immediate heparinization of a newly obtained bone marrow aspirate is critical,as clotting can make it difficult to process the specimen and may, in extreme cases,render the sample useless for cytogenetic study. Processing a clotted bone marrowspecimen involves mechanical disaggregation of the clot and overnight treatmentwith 0.1 ml of heparin (stock solution 1,000 U/ml). In our experience, this procedurehas proved successful in obtaining enough cells for tissue culture in most cases, with

44 M.A. Micale

more limited success in older bone marrow specimens. The newly obtained bonemarrow aspirate should be transported in a sterile container containing preservative-free sodium heparin, tissue culture medium such as RPMI 1640 supplemented withheparin, or Hank’s balanced salt solution (HBSS) containing heparin. A small sam-ple of lymph node should be placed in sterile tissue culture medium or HBSS andtransported to the cytogenetics lab as soon as possible. Every attempt should bemade to transport the specimen to the lab without delay so that cultures can beinitiated. If any delay in transport is expected, the specimens should be placed insterile tissue culture medium to maintain cell viability. Samples can be stored at 4◦Covernight and for no longer than 3 days. An important concern is the overgrowth ofnormal cells in specimens that have been subjected to delay prior to culture initia-tion. Thus, with longer delays comes an increased chance of a false-negative result.Specimens with high white blood cell counts and acute lymphoblastic leukemiaspecimens are particularly vulnerable and should be processed without delay.

Specimen Processing and Tissue Culture

Successful tissue culture of bone marrow specimens requires an optimal cell den-sity, which for a bone marrow culture is approximately 106 cells. Extremely low andextremely high cell densities can compromise tissue culture outcomes. To determinethe proper dilution of the original bone marrow suspension to ensure optimal celldensity, two common methods are utilized. A hemocytometer can be used to per-form a cell count on the original specimen, with the results used to determine theproper dilution of the original sample to 106 cells/ml per culture. The second methodis cruder and utilizes the patient’s white blood cell count to determine the numberof drops of bone marrow suspension to add to 10 ml of tissue culture medium.

An experienced cancer cytogenetics laboratory will more often than not identifyone or more chromosomal abnormalities, either by conventional analysis or FISH,in a bone marrow specimen with abnormal morphology. One exception to this ruleare the chronic myeloproliferative neoplasms such as polycythemia vera, which areoften characterized by molecular genetic changes such as the JAK2 V617F mutation.Success in obtaining positive cytogenetic results is highly dependent on choosingthe appropriate culture conditions for the bone marrow, leukemic blood, and lymphnode specimen. Providing clinical information and a suspected diagnosis (if possi-ble) can aid greatly in determining the type and number of cultures to be established.Table 2.1 provides an overview of various culture regimens.

While a short-term culture of 24–48 h is initiated in most studies along with addi-tional cultures as described above (given an adequate specimen volume), a directmethod is also used in some laboratories. In this method, cells are treated withColcemid for 1 h followed by incubation in a warm hypotonic solution (0.075 MKCl) for 15 min and fixation with 3:1 methanol:glacial acetic acid. The directmethod often yields suitable metaphase cells for analysis and can provide a resultwithin 24 h; however, short-term (24 h) cultures have two major advantages overdirect preparations. First, the metaphase quality obtained with the direct preparationis not as good as that obtained in a short-term culture. Second, in some cases, clonal


Table 2.1 Culture conditions used for hematolymphoid disorders

Clinical findings/suspected diagnosis Types of cultures to be initiated

Myelodysplastic syndrome andmyeloproliferative neoplasms

ST + CM

Acute myeloid leukemiaAnemiaBicytopeniaChronic myelogenous leukemiaa

EosinophiliaPancytopeniaThrombocytosis

ST + CM

Leukopenia or neutropenia ST + CM (adult)ST + GCT (child)

B-cell lymphocytic leukemiaB-cell lymphomaChronic lymphocytic leukemiab

Hairy cell leukemiaLymphadenopathyLymphocytosisLymphomaLymphoproliferative disordersMantle cell lymphomaMonoclonal gammopathyNon-Hodgkin lymphomaPlasmacytomaPlasma cell leukemiaPlasma cell myeloma

ST + GCT + LPS + PWM (if adequatespecimen volume)

T-cell leukemia/lymphoma ST + GCT + PHA

ST – short-term culture (unstimulated 24-h culture)CM – conditioned medium (48–72-h culture). Preparation: 1 ml supernatant from HTB-9 laddercarcinoma cell line culture (obtained from ATCC) added to 9 ml complete mediumGCT – giant cell tumor culture supplement (48–72-h culture). Preparation: 1 ml supernatant fromTIB-223 human lung histiocytoma cell line culture (obtained from ATCC) added to 9 ml completemedium.LPS – lipopolysaccharide (3–4-day culture)PHA – phytohemagglutininPWM – pokeweed mitogen (3–4-day culture)a For peripheral blood, if WBC <50.0, set up buffy coat; if WBC>50.0, set up whole bloodb If peripheral blood, LPS + PWM only; if post bone marrow transplant, ST + GCT

rearrangements are detectable only in cultured preparations, such as the diagnos-tic t(15;17) in acute promyelocytic leukemia. If the specimen is extremely limitedin quantity, it may be necessary to initiate only one culture, which is usually ashort-term unstimulated culture.

The protocol for studying peripheral blood is similar; however, transport mediumshould not be added to the blood sample. Transport in a sodium heparin vacutainer(alternatively lithium heparin) is necessary. The culture conditions described abovewould also be appropriate for peripheral blood samples. Mitogens are added tostimulate the growth of B or T cells as clinically indicated (Table 2.1).

46 M.A. Micale

Culture Harvesting, Slide Preparation, and Staining

Specimen harvesting involves incubating the culture in a mitotic spindle inhibitor,such as Colcemid (0.05 μg/ml) to collect metaphases, then adding a hypotonicsolution (0.075 M KCl), and incubating at 37◦C for 10–15 min. This is followedby several fixation steps using chilled 3:1 methanol:glacial acetic acid. With eachsuccessive fixation, the cell pellet is being “cleaned up,” ensuring an optimal cellsuspension with little or no background when slides are prepared. After 4–5 fixationsteps, the cells are resuspended in fixative and can be stored at –20◦C.

Preparing slides is as much an art as it is a science, and each lab will haveslight variations on their technique. The overall goal is to prepare slides with well-spread chromosomes that can be recognized as individual “metaphase spreads.”Precleaning microscope slides with 95% ethanol can facilitate uniformity of chro-mosome spreading and enhance the quality of metaphase preparations. Optimalhumidity (45–55% ) and ambient temperature (70–75◦F) are also important, neces-sitating in some cases the use of an environmentally controlled chamber for slidepreparation.

In the metaphase stage of the cell cycle, chromosomes are condensed suchthat individual chromosome morphology can be recognized. Chromosome mor-phology is characterized by size, centromere position, and banding pattern. Thebands observed in metaphase chromosomes are prepared by processing slides usingvarious different methodologies and staining solutions, including quinacrine mus-tard and fluorescence microscopy (Q banding), Giemsa or an equivalent stain(G banding), hot alkali followed by staining with Giemsa or acridine orange(R banding), chromosome denaturation prior to Giemsa staining (C banding) tovisualize heterochromatic DNA, and staining with silver nitrate (AgNOR band-ing) to visualize nucleolar organizing regions in the short arm of acrocentricchromosomes.

Guidelines for Microscopic Analysis of Bone Marrowand Leukemic Blood

In most cases, normal and neoplastic cells will coexist in the specimen. The goal,therefore, is to identify those neoplastic cells that potentially carry one or morechromosomal abnormalities. Care should be exercised when making any clinicalpredictions based on the proportion of normal to abnormal cells in a given speci-men, as this can be influenced by cell culture conditions as well as sampling error.The microscopists performing the cytogenetic examination must be aware that insome conditions, particularly ALL, it is those metaphases with poorer morphologythat may be representative of the neoplastic clone. Care should, therefore, alwaysbe taken to examine a variety of metaphase cells of differing quality. In addition,a case which is found to be cytogenetically normal may still harbor significantmolecular abnormalities. Approximately 40–50% of AML cases demonstrate a nor-mal karyotype but possess one or more acquired mutational changes [14] or have


submicroscopic abnormalities detectable by higher resolution techniques such asarray comparative genomic hybridization [16].

The typical oncology chromosome study requires the examination of 20metaphase cells. A clonal abnormality as defined in ISCN 2009 [17] consists oftwo or more cells with the same chromosome gain or structural rearrangement, orthree cells with the same chromosome loss. If a single-cell abnormality is identified,the process cannot be defined as clonal; however, if it is a characteristic abnormal-ity associated with a specific hematolymphoid disorder or is observed in a patientthat demonstrated it as part of an abnormal clone in a previous study, an extendedworkup is indicated. Sometimes, an apparently balanced rearrangement not knownto be associated with any hematolymphoid disorder will be observed. This observa-tion necessitates the examination of a PHA-stimulated peripheral blood culture todetermine if the abnormality is constitutional in nature.

The Karyotype and Cytogenetic Nomenclature

The karyotype is a pictorial representation of the 46 chromosomes present in eachcell (Fig. 2.1). They are classified by their size and centromere position into sevengroups. Within each of these groups, individual chromosome homologues are pairedwith each other based on their similar banding pattern generated by G, Q, or Rbanding.

Fig. 2.1 A normal G-banded bone marrow karyotype

48 M.A. Micale

The development and refinement of a specific and descriptive way to describekaryotype abnormalities has been integral in the growth of cytogenetics over the last30 years. The original attempt to achieve standardization of chromosome nomen-clature was the document A Proposed Standard System of Nomenclature of HumanMitotic Chromosomes presented at the Denver Conference in 1960, while the newestrevision of An International System of Human Cytogenetic Nomenclature (2009)has just been published [17]. Each revision in between has addressed the enhancedmethodologies for studying chromosomes that were developed since the previousrelease. Techniques such as high-resolution banding, FISH, and most recently arrayCGH have resulted in a refinement of chromosome morphology, necessitating anexpansion of chromosome nomenclature in each revision.

The reader is referred to ISCN 2009 [17] for an in depth description of humanchromosome nomenclature. Only a few basic features will be presented here.Chromosomal abnormalities are of two types, numerical and structural. The normalmodal chromosome number is 46, designated as a diploid cell. If more or less than46 chromosomes are present, the cell is referred to as being aneuploid. If more than46 chromosomes are present, the cell is hyperdiploid; if less than 45 chromosomes,the cell is hypodiploid. Gain of a chromosome is referred to as trisomy, while lossof a chromosome is referred to as monosomy. These are described in the karyotypedesignation by a “+” or “–,” respectively.

Structural abnormalities are designated by the type of abnormality present andthe breakpoints involved. The breakpoints will lie either within a chromosome bandor at the junction between two chromosome bands. A chromosome band is a por-tion of a chromosome clearly distinguishable from adjacent segments which maybe lighter or darker depending on the banding technique. There are specific “land-mark” bands that help to distinguish one chromosome from another. Each band is,at successively higher levels of resolution, further divided into subbands. The bandsand subbands are numbered outward from the centromere.

Descriptions of structural rearrangements commonly observed in neoplas-tic disorders are provided in Table 2.2. Table 2.3 lists several examples ofcommon chromosomal abnormalities and their description using ISCN 2009nomenclature.

The Molecular Mechanisms Responsible for ChromosomalRearrangements in Neoplasia

Many cancers are associated with specific chromosomal abnormalities that dis-rupt normal cellular processes leading to malignant transformation. Much workhas focused on the molecular mechanisms that lead to the visible chromoso-mal rearrangements observed in a variety of human constitutional chromosomedisorders [18]. The mechanisms that underlie those pathogenetic rearrangementsobserved in various hematolymphoid disorders may not be dissimilar. A chro-mosome translocation appears to be initiated by a DNA double-strand break


Table 2.2 Overview of structural chromosomal rearrangements observed in neoplasia

Type of structuralchromosomal abnormality Morphological change observed

Translocation Exchange of two chromosomal segments distal to thedesignated breakpoints in two chromosomes; can bebalanced or unbalanced

Derivative A structurally rearranged chromosome generated from eventsinvolving two or more chromosomes or multiple eventswithin a single chromosome

Inversion 180◦ inversion of chromosome segment between twodesignated breakpointsPericentric – breakpoints are in p and q armsParacentric – both breakpoints in same arm

Deletion Loss of chromosome segmentTerminal – Loss of segment distal to single breakpointInterstitial – Loss of segment between two breakpoints

Isochromosome Chromosome arms are identical with a single centromereIsodicentric chromosome A mirror-image chromosome with two centromeresRing Chromosome with one breakpoint in each arm followed by

reunion of two endsMarker Chromosome that appears to be mitotically stable but cannot be

classified by conventional banding studiesDouble minutes Acentric chromosome fragments often in multiple copiesHomogenously staining region Chromosome region that stains uniformly

(Both are cytogenetic manifestations of gene amplification)Add Chromosome with “additional” material of unknown origin

attached to the long or short arm

Table 2.3 Common hematolymphoid chromosomal abnormalities and their description usingISCN nomenclature

t(9;22)(q34;q11.2)t – denotes translocation(9;22) – translocation between chromosomes 9 and 22(q34;q11.2) – breakpoints are in the long arm of both chromosomes at bands 9q34 and 22q11.2

der(22)t(9;22)(q34;q11.2)der(22) – denotes derivative chromosome 22 originating from the t(9;22)(q34;q11.2)

inv(16)(p13q22)inv – denotes inversion(16) – inversion involves chromosome 16(p13q22) – the inverted segment lies between the breakpoints 16p13 (in the p arm)

and 16q22 (in the q arm); the segment between the two breakpoints rotates 180◦

del(5)(q13q33)del – denotes deletion(5) – deletion involves chromosome 5(q13q33) – deleted segment is interstitial between bands 5q13 and 5q33 in the long arm

i(17)(q10)i – denotes isochromosome(17) – isochromosome involves chromosome 17(q10) – chromosome arms are composed of two identical complete chromosome 17 long arms

50 M.A. Micale

followed by “misrepair.” Proposed mechanisms that result in recurrent, balancedtranslocations in hematolymphoid disorders include the following: (1) illegitimateV(D)J or switch recombination, (2) presence of repetitive sequences such as ALUsequences or LINE elements at “broken” ends, and (3) error-prone non-homologousend joining [19]. The joining of the two “broken” ends appears to be facilitated bythe presence of specific DNA sequences at these breaks, including ALU sequences,translin-binding consensus sequences, and scaffold-associated regions [20]. In addi-tion, the formation of recurrent translocations also requires a clustering within thenucleus of the two involved chromosomes. This has been demonstrated for the BCRand ABL1 genes associated with chronic myelogenous leukemia and the BCL6 andMYC genes that rearrange with IGH in B-cell disorders [19]. The mechanism(s) thatfacilitate this clustering within the nucleus are not known. Finally, the identifica-tion of recurrent hematolymphoid translocations including the t(9;22)(q34;q11.2),t(15;17)(q22;q12), and t(14;18)(q32;q21) rearrangements by the polymerase chainreaction (PCR) in apparently healthy individuals raises the possibility that the mech-anisms required for malignant transformation may be far more involved than whatis currently appreciated [21].

Molecular Cytogenetic (FISH) Analysis of Bone Marrowand Leukemic Blood

Basic Principles of Fluorescence In Situ Hybridization

Conventional cytogenetic analysis of bone marrow or leukemic blood culturespermits a genome-wide assessment of chromosomal abnormalities; however, itis sometimes hampered by low mitotic index, poor chromosome morphology,considerable karyotypic complexity, and normal karyotypes. Fluorescence in situhybridization (FISH) can overcome these problems by targeting specific nucleic acidsequences in a highly sensitive and rapid manner. The powerful diagnostic capabili-ties of FISH are rooted in its relative ease of use in the clinical laboratory, enhancedsensitivity over conventional banding studies, and ability to probe for one or morespecific genomic regions of interest in either dividing or non-dividing cells, as wellas in in situ tissue preparations permitting identification of cytogenetic changes in aspecific cell lineage. By utilizing fluorescently labeled DNA probes to detect geneticaberrations that are generally beyond the resolution of conventional chromosomebanding studies, FISH in a sense merges conventional cytogenetic analysis withmolecular genetics.

FISH is based on the principle that a single-stranded DNA molecule will rec-ognize and bind to its complementary sequence on a metaphase chromosome orin an interphase nucleus. The overall hybridization is similar to in situ hybridiza-tion using radioisotope-labeled probes. The major advantage of FISH, however, isthe utilization of a DNA probe labeled with a fluorescent dye, which results in a


highly sensitive, simple, and rapid assay. Both the probe and the target DNA aretreated with heated formamide solution to denature double-stranded DNA, followedby probe application to target DNA and incubation at 37◦C. During the incubationprocess, annealing of the probe to the target sequence occurs through complemen-tary base pairing. A fluorescence microscope equipped with appropriate filters isthen used to detect the hybridized probe on the target material, appearing as bright-colored signals. Multiple probes labeled with different colored fluorescent tags canbe applied simultaneously on the same target to detect one or more specific regionsof the genome.

FISH analysis can be performed on either metaphase chromosomes derived fromcultured cells or non-dividing cells, allowing identification of chromosomal aber-rations irrespective of cell cycle stage. This latter technique, known as interphaseFISH, is a powerful cytogenetic tool that can be applied to a wide variety of clinicalspecimens to enumerate chromosomes and identify chromosomal rearrangements.When viable specimens are not available, interphase FISH can be performed on abone marrow or a blood smear, disaggregated cells from a paraffin block, touchpreparation from a lymph node, or cytospin cells fixed on a microscope slide. FISHcan also be performed on a paraffin-embedded tissue section. While this techniquehas the advantage of maintaining tissue architecture, its inherent disadvantagesinclude nuclear truncation artifact and overlapping cells that may make analysisdifficult.

Clinical Indications for FISH Testing in HematolymphoidDisorders

Common indications for FISH testing in hematolymphoid malignancies includethe following: (1) confirmation of chromosomal abnormalities detected by conven-tional cytogenetics and establishment of FISH signal pattern for follow-up study, (2)detection of chromosomal abnormalities when clinical and morphologic findings aresuggestive of a specific chromosomal abnormality [e.g., t(11;14) in mantle cell lym-phoma], (3) characterization of genetic aberrations using a panel of disease-specificFISH probes for risk stratification and therapeutic management, such as in acutelymphoblastic leukemia (ALL), chronic lymphocytic leukemia (CLL), and plasmacell myeloma (PCM), (4) detection of cryptic or masked translocations when chro-mosome analysis is inconclusive or yields a normal karyotype [such as the t(12;21)in ALL or t(4;14) in myeloma], (5) detection of lymphoma-associated translocationsin paraffin-embedded tissue sections, (6) quantitation of minimal residual diseaseand detection of cytogenetic remission and relapse through analysis of a large num-ber of both dividing and non-dividing cells, (7) monitoring cross-sex bone marrowtransplantation patients for engraftment status (chimerism), and (8) rapid detectionof PML/RARA gene fusion in acute promyelocytic leukemia, where quick diagnosisis required for prompt treatment.

52 M.A. Micale

Types of FISH Probes Routinely Used in HematologicalDisorders

There are primarily three types of probes used in clinical FISH testing: cen-tromere enumeration probes (CEPs), locus-specific identifier (LSI) probes, andwhole-chromosome paint (WCP) probes (Fig. 2.2).

The CEPs recognize a highly repetitive alpha-satellite DNA sequence locatedat the centromere of each chromosome. These probes are labeled in one color andgive a large, bright signal, useful for chromosome enumeration in both interphaseand metaphase cells. The LSI probes hybridize to single-copy DNA sequences in aspecific chromosomal region or gene. These probes can identify fusion gene prod-ucts generated from a reciprocal translocation, chromosome inversions, and genedeletion or amplification. On metaphase cells, the LSI probes give two small, dis-crete signals per chromosome. The gain of LSI signals within a nucleus is consistentwith duplications or amplifications, while the loss of LSI signal indicates a deletion.The design of LSI probes targeting specific translocations has evolved consider-ably, minimizing the false-positive and false-negative rates. Dual-color, dual-fusion(DCDF) LSI probes are designed to span both sides of the breakpoints in two dif-ferent chromosome regions/genes involved in a reciprocal translocation, resulting

Fig. 2.2 Examples of FISHprobe designs commonlyused in hematolymphoiddisorders and their resultinghybridization patterns(reproduced with permissionfrom [58])


in remarkably improved specificity. To assess the rearrangement of a gene that maybe associated with multiple translocation partners, a dual-color break-apart (DCBA)LSI probe has been designed. The DCBA probe is a combination of two differentlylabeled probes that bind to sequences that flank the 5′- and 3′-ends of the break-point within the involved chromosome region. The separation of the two colors isindicative of rearrangement. WCP probes are cocktails of unique sequence DNAprobes derived from flow-sorted chromosomes, chromosome-specific libraries, orchromosome-microdissected regions that recognize specific sequences spanning thelength of a chromosome. In normal metaphase preparations, this gives the effectthat both chromosome homologues are “painted.” WCP probes are useful to iden-tify marker chromosomes and to detect cryptic translocations; however, their utilityin interphase nuclei is limited. An overview of commercially available FISH probesuseful to characterize hematolymphoid disorders is provided in Table 2.4.

Table 2.4 Commercially available FISH probes used to characterize hematolymphoid disorders

DiseaseChromosomalabnormality Gene(s) involved FISH probe(s)

CML t(9;22)(q34;q11.2) ABL, BCRASS

BCR/ABL fusionBCR/ABL fusion + 9q34

AML t(8;21)(q22;q22) RUNX1T1 (ETO), RUNX1(AML1)

RUNX1T1/RUNX1 fusion

inv(16)(p13q22)/t(16;16) MYH11, CBFβ CBFβ or MYH11 breakapart

MYH11/CBFβ fusiont(v;11)(v;q23), del 11q23 MLL MLL break apartMonosomy 5/del 5q33-34 CSF1R CSF1R/5pMonosomy 5/del 5q31 EGR1 EGR1/5pMonosomy 7/del 7q D7S522/CEP7del 20q D20S108Trisomy 8 CEP8

AML-M3(APL)

t(15;17)(q22;q12)t(v;17)(v;q12)

PML, RARA PML/RARA fusion,RARA break apart

MDS Monosomy 5/del 5q33-34 CSF1R CSF1R/5pMonosomy 5/del 5q31 EGR1 EGR1/5pMonosomy 7/del 7q D7S522/CEP7del 20q D20S108Trisomy 8 CEP 8del(11)(q23) MLL MLL break apartdel(13)(q14) RB1 RB1/13q14

D13S319/13q14MPN Trisomy 8 CEP 8

Trisomy 9 CEP 9del(4)(q12q12) CHIC2 CHIC2/ 4qter

PDGFRA/FIP1L1 PDGFRA/FIP1L1 fusiondel 20q D20S108

B-ALL Trisomy 4, 10, 17 CEP 4, 10, 17t(12;21)(p13;q22) ETV6 (TEL), RUNX1

(AML1)TEL/AML1 ES fusion

t(v;11)(v;q23) MLL MLL break apartt (9;22)(q34;q11.2) ABL, BCR BCR/ABL fusion

54 M.A. Micale

Table 2.4 (continued)

DiseaseChromosomalabnormality Gene(s) involved FISH probe(s)

t(1;19), t(17;19) E2A TC3F/PBX1 fusionT-ALL t(5;14)(q35;q32) TLX3 (HOX11L2) TLX3 break apart

del(9)(p21) p16 p16/D9Z37q35 rearrangement TCRbeta TCRbeta break apart7p14-15 rearrangement TCRgamma TCRgamma break apart14q11.2 rearrangement TCRalpha/delta TCRalpha/delta break

apartCLL del(11)(q22.3) ATM ATM

Trisomy 12 CEP 12del (13)(q14.3) Micro-RNA genes

(miR-16-1,miR-15a)D13S319/13q14RB1/13q14

del(17)(p13) TP53 TP53del(6)(q23) MYB MYB

PCM Monosomy13/del(13)(q14)

D13S319/LAMP1RB1/LAMP1

Trisomy 5,9,15,19 CEP 5, 9, 15, 19del(17)(p13) TP53 TP53t(11;14)(q13;q32) CCND1, IGH CCND1/IGH fusiont(4;14)(p16.3;q32) FGFR3, IGH FGFR3/IGH fusiont(14;16)(q32;q23) IGH, MAF MAF/IGH fusiont(V;8)(V;q24) MYC MYC break apart

NHL t(V;14)(V;q32) IGH IGH break apartMCL t(11;14)(q13;q32) CCND1, IGH CCND1/IGH fusionFL t(14;18)(q32;q21) IGH, BCL2 BCL2/IGH fusionBL t(8;14)(q24;q32)

t(2;8)(p12;q24)t(8;22)(q24;q11.1)

MYC, IGHMYC, IGKMYC, IGL

IGH/MYC,CEP8MYC break apart

DLBCL t(3;14)(q27;q32),t(2;3)(p12;q27),t(3;22)(q27;q11.2)

BCL6, IGHIGK, BCL6BCL6, IGL

BCL6 break apart

MALT t(11;18)(q21;q21),t(14;18)(q32;q21)

API2, MALTIGH, MALT

API2/MALT1 fusionMALT1 break apart

ALCL t(2;5)(p23;q35),t(V;5)(V;q35)

ALK, NPM ALK break apart

CML, Chronic myelogenous leukemia; AML, acute myelogenous leukemia; APL, acute promyelo-cytic leukemia; MDS, myelodysplastic syndrome; MPD, myeloproliferative disorder; ALL, acutelymphoid leukemia (B or T cell); CLL, chronic lymphocytic leukemia; PCM, plasma cell myeloma;NHL, non-Hodgkin lymphoma; MCL, mantle cell lymphoma; FL, follicular lymphoma; BL,Burkitt lymphoma; DLCL, diffuse large-cell lymphoma; MALT, extranodal marginal zone B-celllymphoma of mucosa-associated lymphoid tissue; ALCL, anaplastic large-cell lymphoma; BMT,bone marrow transplantation

Advantages and Disadvantages of FISH

FISH analysis has both advantages and disadvantages over conventional cytoge-netic analysis. FISH can (1) be performed on metaphase cells or interphase nuclei


(non-dividing cells) and on fresh or fixed tissue samples, (2) target genetic aberra-tions that pinpoint candidate genes involved in leukemogenesis, (3) simultaneouslyassess chromosomal aberrations, cellular phenotype, and tissue morphology utiliz-ing paraffin-embedded tissue sections (paraffin FISH), (4) provide in a rapid fashionhighly specific, sensitive, and reproducible results that are interpreted objectively,(5) simultaneously assess multiple genomic targets, (6) provide superior resolu-tion (interphase FISH > 20 kb, metaphase FISH > 100 kb) compared with standardkaryotyping (>10 Mb), and (7) detect specific cryptic chromosomal abnormalities.Limitations of FISH include the following: (1) its inability to provide a genome-wide assessment of chromosomes; (2) the necessity for clinical information or adifferential diagnosis to guide the appropriate choice of probes to be used, and(3) the requirement for a high-quality fluorescence microscope with multiple filters,a CCD camera that can detect low-level light emission, and sophisticated imagingsoftware.

Diagnostic and Prognostic Cytogenetic Markers in MyeloidDisorders

Myeloproliferative Neoplasms (MPN)

Most MPNs are not characterized by a unique cytogenetic abnormality but insteaddemonstrate molecular mutations in genes that code for cytoplasmic or receptorprotein tyrosine kinases. As such, these mutations such as the JAK2 V617F andFIP1L1–PDGFRA fusion gene do not affect differentiation but instead convey aproliferative advantage [22]. Those cytogenetic abnormalities that are identifiedare found in a variety of myeloid neoplasms, precluding their use as a marker tosubclassify the disease process. Despite the relatively low frequency of karyotypicabnormalities at diagnosis in these disorders, cytogenetic analysis is still important.It can distinguish a clonal process from a reactive myeloproliferation, it can excludechronic myelogenous leukemia characterized by the Philadelphia (Ph) chromosome,and it can be used throughout the course of the disease to identify cytogeneticprogression associated with disease progression and an increased risk of leukemictransformation. In addition, identification of a complex karyotype in the diagnosticbone marrow is associated with a poorer prognosis.

Chronic myelogenous leukemia, BCR/ABL1 positive. CML was the first hema-tological disorder to be associated with a specific chromosomal abnormality,the t(9;22)(q34;q11.2) which generates the Philadelphia chromosome (truncatedchromosome 22) (Fig. 2.3).

The molecular consequence of this translocation is fusion of the 3′ segment ofthe Abelson (ABL1) proto-oncogene on chromosome 9q34 to the 5′ segment of theBCR gene on chromosome 22q11.2, producing a chimeric 210-kDa BCR/ABL fusiongene product that has constitutive tyrosine kinase activity. At diagnosis, over 90% ofCML patients will demonstrate the t(9;22)(q34;q11.2) by conventional cytogenetic

56 M.A. Micale

Fig. 2.3 An abnormal female karyotype demonstrating the t(9;22)(q34;q11.2) which generates thePhiladelphia chromosome [der(22) chromosome]

analysis. The remaining cases present either a submicroscopic rearrangement or avariant t(v;9;22) translocation. In these cases, FISH analysis can readily detect theBCR/ABL1 fusion, and failure to do so would suggest that another MPN, such aschronic neutrophilic leukemia, should be considered. The dual-color, dual-fusionFISH (D-FISH) assay utilizing the BCR/ABL1 probe (Fig. 2.4a) not only will detecttranslocations occurring at the typical major breakpoint cluster region (M-BCR)that generates the p210 product but will also identify a breakpoint in the microbreakpoint cluster region (μ-BCR) which produces a larger fusion protein (p230)rarely observed in CML, as well as a breakpoint in the minor breakpoint clusterregion (m-BCR) producing the shorter fusion product (p190) most often observedin Ph+ ALL.

Deletion of DNA sequences proximal to the 9q34 breakpoint, which includes theASS gene, has been observed in approximately 10–30% of CML patients at diagno-sis. These deletions have been associated in some studies with a shortened chronicphase and decreased overall survival; however, other studies have reported no sig-nificant difference in those patients with a der(9) deletion with regard to responserate or overall survival [23].

Effective treatments for CML including imatinib mesylate (Gleevec),α-interferon, and allogeneic stem cell transplantation result in a decrease in thepercentage of Ph+ neoplastic cells. BCR/ABL1 FISH can accurately quantify cyto-genetic response to therapy, determine remission status, and identify relapse. With


Fig. 2.4 Commoncytogenetic abnormalitiesin MPNs. (a) A BCR/ABL1fusion in CML is demons-trated utilizing the dual-color,dual-fusion FISH assay. TheBCR/ABL1-positive nuclei(with two fusion signals) areat 3 and 6 o’clock. Thenucleus in the bottom leftcorner is negative forBCR/ABL1 fusion (two red,two green signals). Themetaphase cell in the center isalso positive for fusion[arrows identify the der(9)and der(22) chromosomes].Deletion of chromosomes20q (b) and 11q (c) is arelatively commonabnormality in MPNs

successful treatment, the D-FISH assay can monitor regression of the clone downto 1%. Much has been written about the use of BCR/ABL1 FISH analysis of periph-eral blood specimens. This is a common practice for routinely monitoring CMLpatients, as it can be performed at regular intervals without the need for an invasivebone marrow aspiration, even for patients in complete cytogenetic remission. Somestudies have suggested a similar performance of the BCR/ABL1 quantitative FISHassay in peripheral blood versus bone marrow for detection of minimal residual dis-ease; however, other studies have suggested that measuring BCR/ABL1 positivityin peripheral blood may underestimate the tumor burden [23]. Nevertheless, it isgenerally acknowledged that FISH analysis of peripheral blood utilizing D-FISH isadequate for CML disease monitoring.

Of the three diagnostic modalities (karyotyping, FISH, and RT-PCR), onlyconventional cytogenetics provides a genome-wide assessment that permits iden-tification of clonal evolution including acquisition of abnormalities such as trisomy8, isochromosome 17q, trisomy 19, and an additional copy of the der(22) chromo-some. These abnormalities herald the onset of accelerated phase or blast phase CMLwhich would necessitate modifications of the treatment plan. Thus, neither RT-PCRnor BCR/ABL1 FISH negate the importance of bone marrow cytogenetic analysis asan important management tool in CML.

Polycythemia vera. The most common cytogenetic markers identified in PV indecreasing frequency are del(20q) (Fig. 2.4b), +8, +9, 9p rearrangement, gains of1q, and del(13q). These abnormalities are observed in 15–25% of cases. Trisomy 8may be the sole change or may be found in combination with trisomy 9. A clone with

58 M.A. Micale

trisomies 8 and 9 may persist for several decades without clonal evolution or trans-formation to acute leukemia. As PV evolves to postpolycythemic myelofibrosis oracute leukemia, additional cytogenetic abnormalities are acquired [23]. This cytoge-netic evolution is apparent when comparing follow-up bone marrow biopsies withthe baseline karyotype performed on the diagnostic bone marrow specimen. Theidentification of unfavorable prognostic markers [any aberration other than del(13q)or del(20q)] appears to be the strongest predictor of a poor prognosis in secondarymyelofibrosis [24].

Essential thrombocythemia. Less than 10% of ET cases demonstrate cytogeneticabnormalities, and none are specific for this disorder. Like other MPNs, deletions ofchromosomes 5q, 13q, and 20q, along with +8, +9, and gains of 1q are commonlyobserved. One important prognostic cytogenetic marker in ET is the presence ofabnormalities involving chromosomes 7 and 17, which appear to be associated witha higher risk of leukemic transformation. Since none of these abnormalities are spe-cific for ET, the greatest benefit of cytogenetic testing in this disorder is to excludethe presence of the Ph chromosome as a cause of thrombocytosis [22].

Primary myelofibrosis. Chromosomal abnormalities are found in 40–50% ofcases and are found in greater numbers with disease progression. The presenceof either del(13)(q12–22) or der(6)t(1;6)(q21–23;p21.3) is strongly suggestive butnot diagnostic for PMF. Non-random abnormalities are similar to those foundin PV, including trisomy for chromosomes 8, 9, and 21 as well as del(13q)and del(20q) chromosomes. As the disease progresses, structural abnormalitiesbecome more common, including gain of 1q, chromosome 7q abnormalities, anddel(17p). The chromosome 7q abnormalities along with chromosome 5q dele-tions may be therapy-related changes related to cytotoxic therapy used to treatthe myeloproliferative process. The HMGA2 (high-mobility group protein A2) isdisrupted by a recurrent breakpoint at chromosome band 12q14 in some cases[3, 22].

Chronic neutrophilic leukemia. Most patients with CNL demonstrate a normalkaryotype; however, +8, +9, +21, del(11q) (Fig. 2.4c), del(12p), and del(20q) havebeen reported as clonal aberrations. As the disease progresses, clonal cytogeneticabnormalities may emerge [3].

Chronic eosinophilic leukemia/idiopathic hypereosinophilic syndrome. CEL/HES belongs to the WHO subgroup of Myeloid and Lymphoid Neoplasms withPDGFRA Rearrangement. These disorders are characterized by a persistent unex-plained hypereosinophilia and rearrangement of the PDGFRA gene. The mostcommon rearrangement of PDGFRA involves formation of a hybrid fusion tyro-sine kinase between the 5′-portion of the FIP1L1 gene and the 3′-portion of thePDGFRA gene through a cryptic 800-kb interstitial deletion within chromosomeband 4q12 [25, 26]. This event can be identified in 40–60% of CEL patients andcan be readily demonstrated by FISH utilizing a probe for the CHIC2 gene, whichlies between the FIP1L1 and PDGFRA genes and is deleted when the fusion eventoccurs. Recently, a FISH probe which recognizes the PDGFRA/FIP1L1 fusion genehas become available. A subset of patients with CES have benefited from treatmentwith imatinib mesylate, which appears to target FIP1L1/PDGFRA [25].


Myelodysplastic Syndromes (MDS)

Bone marrow cytogenetic analysis is a standard practice in the evaluation of a patientwith suspected MDS and is considered an independent predictor of clinical outcome,overall survival, and progression to acute leukemia. The extent and nature of cyto-genetic abnormalities is one of the three parameters in the International PrognosticScoring System (IPSS), along with degree of peripheral cytopenia and bone mar-row blast cell percentage that separates patients into one of the four prognosticgroups (good, intermediate-1, intermediate-2, and poor) with regard to both sur-vival and AML evolution [27]. As the disease becomes more severe, the frequencyof cytogenetic abnormalities increases. Cytogenetic analysis can also distinguish amonoclonal proliferation from a reactive process in a morphologically unremarkablebone marrow and can, through serial cytogenetic studies, identify clonal evolutionwhich accompanies progression of disease.

Conventional cytogenetic analysis can identify chromosomal abnormalities in40–70% of de novo MDS cases and in almost 95% of t-MDS at diagnosis [28], withnone specific for a particular MDS subtype except for the chromosome 5q deletion[WHO classification: MDS associated with isolated del(5q)]. Recurrent chromo-some changes in MDS include loss of chromosome 5 or 7, deletions of chromosome5q or 7q, trisomy 8, and chromosome 20q deletion. Loss of the Y chromosome isalso relatively common in MDS, but this may be an age-related artifact in manypatients. The identification of trisomy 8 and/or del(20q) in the absence of morpho-logical evidence does not provide a definitive diagnosis of MDS. Close clinical andlaboratory follow-up of such patients is necessary to identify emerging evidence ofmyelodysplasia [3]. Less frequently, structural rearrangements involving chromo-somes 3q; deletion of chromosomes 11q, 13q, and 17p; and trisomies 9 and 21 areobserved. Many of these chromosomal changes are also observed in AML, a find-ing indicative of the pathobiologic similarity between the two diseases. Complexkaryotypes are often associated with advanced disease and a greater likelihood ofleukemic transformation (Fig. 2.5).

The primary utility of FISH analysis in MDS is based on the finding that 15–20%of MDS patients demonstrate a normal karyotype, yet possess one or more clonalabnormalities of prognostic and/or therapeutic significance when analyzed by FISH[28, 29]. These patients will often demonstrate an increase in bone marrow blasts,an increase in rate of leukemic transformation, and a poorer prognosis [29]. Basedon this and other studies, most advocate the use of an MDS-FISH panel on thediagnostic specimen. The MDS-FISH panel utilized in many laboratories includesprobes to detect monosomy 5/del(5q), monosomy 7/del(7q), trisomy 8, chromosome20q deletion, chromosome 11q deletion, and chromosome 13q deletion [28].

Acute Myeloid Leukemia

Acute myeloid leukemia (AML) is characterized by excessive accumulation ofmyeloid blasts (>20%) in bone marrow, peripheral blood, and other tissues. AMLcan be de novo or can occur following exposure to cytotoxic agents including

60 M.A. Micale

Fig. 2.5 An abnormal cytogenetic clone presenting this complex karyotype was identified in an83-year-old female with pancytopenia. Her history is positive for right breast infiltrating ductalcarcinoma in 2002 with chemotherapy and radiation therapy. Bone marrow examination revealedrefractory anemia with excess blasts, type 1. Abnormalities are identified by arrows and include themyeloid markers del(5q) and monosomy 7 which are consistent with secondary (therapy-related)myelodysplasia

chemotherapy and radiotherapy (therapy-related- or t-AML). Some 10–15% ofAML cases are related to such previous cytotoxic exposure. The identification ofspecific cytogenetic abnormalities is diagnostic for specific AML subtypes and canbe powerful predictors of prognosis and response to therapy. Overall, cytogeneticabnormalities are identified in approximately 55% of adults at diagnosis, with arange of 50–80% [28]; however, only a subset of these chromosome changes areassociated with clinical, morphological, and immunophenotypic specificity for aparticular AML subtype.

In the current WHO classification scheme, the following AMLs are character-ized by a recurrent cytogenetic abnormality associated with a specific molecularrearrangement (Fig. 2.6a–d):

• AML with t(8;21)(q22;q22) – RUNX1/RUNX1T1: Identified in 5–12% casesof AML with maturation [FAB classification: AML-M2] and in 40–50% ofkaryotypically abnormal cases of AML with maturation.

• AML (promyelocytic) with t(15;17)(q22;q12) – PML/RARα (Fig. 2.6a): Acutepromyelocytic leukemia (APL) [FAB classification: AML-M3], a disease


Fig. 2.6 (a–c) Common rearrangements identified in acute myeloid leukemia (arrows point toderivative chromosomes). (d) FISH detection of chromosome 16 inversion using a break-apartprobe for CBFβ in an interphase nucleus and metaphase spread (arrow indicates break-apartsignal)

primarily observed in young adults, is characterized by the presence of abnor-mal hypergranular promyelocytes. All AML cases with the t(15;17)(q22;q12)are diagnosed as APL; however, not all cases of APL will present the clas-sic t(15;17)(q22;q12) due to the presence of (1) a complex karyotype involvingboth chromosomes 15 and 17 with additional cytogenetic changes, (2) a submi-croscopic event leading to insertion of the retinoic acid receptor alpha (RARα)gene into the promyelocytic leukemia (PML) gene, or (3) a variant translo-cation such as t(11;17)(q23;q12) with ZBTB16/RARα fusion, t(5;17)(q35;q12)with NPM1/RARα fusion, or t(11;17)(q13;q12) with NUMA1/RARα fusion. Thet(15;17) and variant translocations all have in common disruption of the RARα

gene, with the typical t(15;17) giving rise to the PML/RARα gene fusion prod-uct which causes a block in differentiation at the promyelocyte stage [3]. Theidentification of the t(15;17) and the genes involved in this rearrangement hasled to a successful treatment for APL utilizing all-trans-retinoic acid (ATRA),which acts as a differentiating agent [28]. Identification of variant translocationsis important, as some APL variants such as t(11;17)(q23;q12) are resistant tothis drug.

• AML with t(9;11)(p22;q23) – MLLT3/MLL (Fig. 2.6b): Acute myeloid leukemiawith chromosome 11q23 abnormalities generally presents with monocytic fea-tures and involves disruption of the MLL (myeloid lymphoid lineage or mixedlineage leukemia) gene. Abnormalities of 11q23 are identified in 5–6% of AMLcases occurring at any age; however, it is more common in childhood AML.The two AML subgroups that demonstrate 11q23 rearrangement most often areAML in infants and therapy-related AML (following topoisomerase II therapy).The most common translocations in childhood AML include t(9;11)(p21;q23)and t(11;19)(q23;p13.1) or t(11;19)(q23;p13.3). The MLL gene is very promis-cuous, as it is known to be involved in 73 recurrent translocations and partnerwith 54 partner genes in all acute leukemias [3, 30]. Because of this, the mosteffective method to detect MLL gene rearrangement is to utilize an MLL genebreak-apart probe that can detect involvement of MLL regardless of which partnerchromosome band/gene is involved [3].

62 M.A. Micale

• AML with inv(16)(p13q22) or t(16;16)(p13;q22) – CBFβ/MYH11 (Fig. 2.6c):Acute myelomonocytic leukemia (AMML) [FAB classification: AML-M4eo]accounts for approximately 10% of all AML cases and is characterized by anincrease in myeloid and monocytic cell lines with a characteristically abnormaleosinophil component in bone marrow. The genetic basis for AML-M4eo is thefusion of the core binding factor beta subunit (CBFβ) gene at chromosome 16q22to the smooth muscle myosin heavy chain gene (MYH11) at chromosome 16p13through either the inv(16) or the t(16;16).

• AML with t(6;9)(p23;q34) – DEK/NUP214: AML with or without monocyticfeatures that is often associated with basophilia and multilineage dysplasia. Thet(6;9) is the sole abnormality in most cases, although it can sometimes be partof a complex karyotype. The concurrent identification of the FLT3-ITD mutationoccurs in 69% of pediatric cases and 78% of adult cases [3].

• AML with inv(3)(q21q26.2) or t(3;3)(q21;q26.2) – RPN1/EVI1: AML withincreased atypical bone marrow megakaryocytes and associated multilin-eage dysplasia. Patients may present de novo or have a prior MDSphase [3].

• AML (megakaryoblastic) with t(1;22)(p13;q13) – RBM15/MKL1: A rare AML(<1% of cases) that demonstrates small and large megakaryoblasts. This dis-ease is de novo in most cases and almost exclusively seen in infants and youngchildren [3].

Of these AMLs, the rearrangements considered to be favorable with regardto response to chemotherapy, high remission rate, and long-term survival includet(8;21)(q22;q22), inv(16)(p13q22) or t(16;16)(p13;q22), and t(15;17)(q22;q12)treated with all-trans-retinoic acid. The (9;11)(p22;q23) is associated withan intermediate prognosis, while the t(6;9)(p23;q34), inv(3)(q21q26.2), andt(3;3)(q21;q26.2) are associated with a poor prognostic outcome [3].

In addition to the cytogenetically characterized myeloid leukemias describedabove, several additional AML groups are recognized. Acute myeloid leukemia withmyelodysplasia-related changes often present with severe pancytopenia. More oftenobserved in the elderly, this disease is characterized by chromosomal abnormalitiessimilar to those found in MDS. These include monosomy 5/del(5q) and monosomy7/del(7q), often as part of a complex karyotype [3].

The latency period for development of therapy-related myeloid neoplasms varieswith the type of chemotherapeutic agents used. Therapy-related AML (t-AML)associated with alkylating agent chemotherapy or radiation therapy is often pre-ceded by MDS and can develop after a period of 2–7 years, while topoisomerase IIinhibitor therapy-associated AML develops after a shorter latency period and is notassociated with a preceding myelodysplastic phase [31]. Alkylating agent t-AMLis characterized by deletions involving chromosomes 5 and 7, often as part of acomplex karyotype (Fig. 2.7). Topoisomerase-associated t-AML is associated withdisruptions of the MLL gene at 11q23, often through a balanced translocation. Whilein general the outcome of t-AML/t-MDS is poor, certain cytogenetic results includ-ing inv(16), t(8;21), and t(15;17) are associated with a better prognosis (comparableto de novo AML with favorable cytogenetics).


Fig. 2.7 This complex karyotype was identified in a patient who had undergone multiple roundsof chemotherapy and radiation for persistent follicular lymphoma over many years. The concur-rent bone marrow examination demonstrated marked erythroid hyperplasia and megakaryocytichyperplasia along with erythroid and megakaryocytic dyspoiesis and ringed sideroblasts. t-MDSwas favored, given her therapy history. This karyotype confirms t-MDS with abnormalities of chro-mosomes 5 and 7. A little over 1 month later, this patient presented with acute myeloid leukemia(t-AML), not surprising given the complexity of this karyotype

Those AMLs subtyped in the WHO classification as acute myeloid leukemia,not otherwise specified include acute myeloid leukemia with minimal differ-entiation, acute myeloid leukemia without maturation, acute myeloid leukemiawith maturation, acute myelomonocytic leukemia, acute monoblastic and mono-cytic leukemia, acute erythroid leukemia, acute megakaryoblastic leukemia, acutebasophilic leukemia, and acute panmyelosis with myelofibrosis. These diseasesare not associated with a specific cytogenetic abnormality but instead demon-strate abnormalities that are best classified as “myeloid cytogenetic markers.”These include monosomy 5/del(5q), monosomy 7/del(7q), +8, del(11q), del(20q).Complex karyotypes are often observed as well [3].

Diagnostic and Prognostic Cytogenetic Markers in LymphoidDisorders

B-Lymphoblastic Leukemia/Lymphoma with Recurrent GeneticAbnormalities

This new WHO classification defines a group of diseases characterized by recurrentnumerical and structural chromosomal abnormalities. In childhood ALL, the identi-fication of recurrent chromosomal aberrations as prognostic markers has had a major

64 M.A. Micale

impact on efforts to cure this disease, as they have permitted effective stratificationof patients into appropriate treatment regimens. Approximately 80% of ALL casesdemonstrate clonal chromosomal abnormalities, while the remaining cases eitherpresent a normal karyotype or cannot be analyzed due to a variety of factors such aspoor chromosome morphology and the apoptotic tendency of ALL blasts in culture.For this reason, FISH has become an important tool for the assessment of geneticaberrations in ALL [32].

High hyperdiploidy, defined as >50 chromosomes per karyotype, occurs inapproximately 25% of ALL cases (B-lymphoblastic leukemia/lymphoma with hyper-diploidy) and constitutes a distinct subset characterized by a favorable prognosis(Fig. 2.8). The gains are non-random, with chromosomes 4, 6, 10, 14, 17, 18,21, and X accounting for close to 80% [32]. More specifically, hyperdiploidALL with simultaneous trisomy of chromosomes 4, 10, and 17 has the leasttreatment failure and the greatest clinical outcome [33]. Enumeration of chromo-somes 4, 10, and 17 by FISH (triple trisomy FISH) can identify these numericalchanges (Fig. 2.9), providing important prognostic information when chromosomeanalysis is unsuccessful or when a normal karyotype is identified by bandingstudies. In contrast, hypodiploid ALL (B-lymphoblastic leukemia/lymphoma withhypodiploidy) defines a subgroup characterized by <45 chromosomes per karyotype.This is observed in both adults and children; however, near-haploid ALL (with23–29 chromosomes) is identified almost exclusively in children. Hypodiploid ALLis associated with a poor prognosis [3]. Care should be exercised when a high

Fig. 2.8 Pediatric bone marrow with precursor B-cell ALL demonstrating a hyperdiploid kary-otype. Note trisomy for chromosomes 4 and 17 (see text)


Fig. 2.9 Pediatric B-cellALL demonstrating trisomyof chromosomes 4, 10, and 17by FISH

hyperdiploid/near-triploid karyotype is identified to examine the pattern of chromo-some gain, as duplication of a near-haploid/hypodiploid karyotype can appear as ahyperdiploid karyotype, yet will be associated with the poor prognosis characteristicof hypodiploid ALL.

In the non-hyperdiploid ALL subgroup, four major translocations have beenobserved. B-lymphoblastic leukemia/lymphoma with t(12;21)(p13;q22); TEL/AML1(ETV6/RUNX1) is recognized in up to 30% of childhood B-precursor ALL; however,this translocation is rare or absent in infants and in adults with ALL. The t(12;21)translocation fuses the TEL (ETV6) and AML1 (CBFA2 or RUNX1) genes, normallylocalized to 12p13 and 21q22, respectively. Many studies have demonstrated thatALL patients with TEL/AML1 fusion do extremely well. This translocation can-not be detected by conventional cytogenetics due to its cryptic nature, thereforenecessitating the use of a TEL/AML1 fusion FISH probe for detection [3].

B-lymphoblastic leukemia/lymphoma with t(9;22)(q34;q11.2); BCR/ABL1 isobserved in approximately 5% of children but up to 25% of adults with ALL. Theresulting BCR/ABL1 hybrid gene product is a 190-kDa protein that, as in CML,possesses dysregulated tyrosine kinase activity and is responsible for leukemictransformation. Ph+ALL is one of the most difficult childhood leukemias to treat andis generally associated with a poor prognosis [3]. Any of the BCR/ABL1 FISH probeformats utilized in CML will also detect the fusion gene associated with breakpointswithin the minor breakpoint region in ALL.

B-lymphoblastic leukemia/lymphoma with t(v;11q23); MLL rearranged con-stitutes a subgroup characterized by translocations of chromosome band 11q23

66 M.A. Micale

causing rearrangements of the MLL gene. This is seen in 80% of infant leukemia andin secondary leukemia that arises in patients treated with topoisomerase II inhibitors.Leukemic cells containing 11q23/MLL rearrangement are usually non-hyperdiploid,have an early pre-B-cell immunophenotype, and coexpress myeloid antigens exceptfor CD10. Generally, ALL that involves MLL gene rearrangement is a clinicallyaggressive disease with a poor prognosis. Greater than 50 translocation partnerswith 11q23 have been described in ALL, suggesting that disruption or destabiliza-tion of MLL function underlies leukemogenesis in these cases. As these leukemiashave been observed in very young infants, it has been theorized that MLL generearrangement may occur in utero. The most common translocations are t(4;11) fol-lowed by t(11;19) and t(9;11). As many different variant t(v;11q23) translocationsexist, the most sensitive method for detecting MLL gene rearrangement is to utilizean MLL gene rearrangement probe [3, 30].

B-lymphoblastic leukemia/lymphoma with t(1;19)(q23;p13.3); E2A/PBX1(TCF3/PBX1) is seen in approximately 5% of adult and childhood ALLs andencodes the fusion protein E2A/PBX1. This was previously thought to represent apoor prognostic marker, but intensification of therapy in pediatric patients has over-come its effects on outcome. This translocation can be detected utilizing an E2Agene break-apart FISH probe [3].

B-lymphoblastic leukemia/lymphoma with t(5;14)(q31;q32); IL3/IGH is a rareentity accounting for less than 1% of all cases of ALL. Seen in both children andadults, the prognostic significance of this translocation is not firmly established.Conventional cytogenetic analysis can usually identify this abnormality [3].

FISH has become an invaluable tool for identifying the major genetic aberra-tions in ALL and for risk-stratifying patients with this disease. In one large study,FISH screening using probes for TEL/AML1, BCR/ABL1, and MLL gene rearrange-ments along with selected centromeric probes increased the success rate to 91% andthe detection rate of genetic aberrations to 89% [34]. Many clinical trials, includ-ing those established by the Children’s Oncology Group (COG), require all newlydiagnosed ALL cases to undergo both conventional cytogenetic testing and molec-ular cytogenetic characterization for risk stratification utilizing a panel to identifyTEL/AML1and BCR/ABL1 fusion; MLL rearrangement; and chromosomes 4, 10,and 17 triple trisomy.

T-Lymphoblastic Leukemia/Lymphoma

T-lymphoblastic leukemia/lymphoma is a malignancy of lymphoblasts committed tothe T-cell lineage. T-cell acute lymphoblastic leukemia (T-ALL) comprises about15% of all childhood ALL cases. It is more commonly found in older than youngerchildren and more often in males than females. In adults, T-ALL comprises about25% of all ALL cases. T-cell lymphoblastic lymphoma (T-LBL) is found in all agegroups and makes up approximately 90–95% of all lymphoblastic lymphomas. BothT-ALL and T-LBL demonstrate clonal rearrangements of the T-cell receptor (TCR)genes including the alpha/delta TCR loci at 14q11.2, the beta locus at 7q35, and


the gamma locus at 7p14-15. These genes rearrange with various partner genesincluding MYC, HOX11 (TLX1), HOX11L2 (TLX3), and TAL1 and lead to dysregu-lation of the partner gene when it juxtaposes next to one of the TCR gene promoterregions. While an abnormal karyotype is identified in 50–70% of cases, FISH break-apart probes to detect rearrangement of the TCR alpha/delta, TCR beta, and TCRgamma loci (Table 2.4) are available [3].

Chronic Lymphocytic Leukemia/Small Lymphocytic Leukemia

Low proliferation activity of leukemic B cells in culture and overgrowth of normalcells preclude the routine detection of chromosomal abnormalities by conventionalcytogenetic analysis in chronic lymphocytic leukemia/small lymphocytic lymphoma(CLL/SLL). For this reason, FISH utilizing a panel of DNA probes that interrogatethe MYB gene at 6q23, ATM gene at 11q22.3, chromosome 12 alpha-satellite region,D13S319 locus at 13q14.3, LAMP1 gene at 13q34, and p53 gene at 17p13.1 is a use-ful adjunct to conventional analysis in the workup of a newly diagnosed CLL/SLLcase. G banding reveals a clonal chromosomal abnormality in approximately 40%of CLL/SLL cases (with trisomy 12 being the most frequent), while FISH iden-tifies genetic aberrations in over 80% of CLL/SLL cases. With FISH, the mostcommon single chromosomal abnormality is a deletion of 13q14 found in 55–65%of cases followed by trisomy 12 in 15–25%, deletion of 11q22/ATM in 11–18%,deletion of 17p13/p53 in 7–8%, and deletion of 6q in 5–6% [35–37]. Identificationof these abnormalities is of prognostic value, with isolated 13q deletion or a normalkaryotype predicting a better prognosis, while del(6q), del(11q), and del(17p) char-acterize a group with a poorer prognosis. Trisomy 12 carries a high risk of diseaseprogression, but unlike patients with del(17p) and del(11q), patients with trisomy 12respond to therapy with better survival [35]. Follow-up FISH studies are also clini-cally useful, as demonstrated by one study which showed that 27% of CLL patientsacquired new chromosomal aberrations during the course of their disease, and inone-third of these patients, the newly detected abnormalities changed their diseasestatus from low risk to high risk [38].

Plasma Cell Myeloma

Identification of recurrent chromosomal abnormalities by conventional analysis inplasma cell myeloma (PCM) has been hindered by patchy bone marrow infiltration,low mitotic index of malignant plasma cells in vitro, the poor quality of metaphasechromosomes, and the cryptic nature of some IgH gene rearrangements observedin this disease. An abnormal karyotype is found in 30–40% of cases, more often inadvanced stages than in newly diagnosed patients. Highly complex karyotypes arealso common, mostly in later stage disease [39–42].

Three distinct cytogenetic groups are recognized: (1) a hyperdiploid group with47 or more chromosomes observed in 30–50% of cases with a lower frequency

68 M.A. Micale

of IgH/14q32 rearrangement and monosomy 13/del(13q), (2) a hypodiploid groupaccounting for 20–35% of cases with a higher frequency of IgH/14q32 rearrange-ment and monosomy 13/del(13q), and (3) a pseudodiploid group in 20–35% of casescharacterized by IgH/14q32 rearrangement [41]. Clinically, this cytogenetic classi-fication is valuable, since hyperdiploid PCM patients seem to have a better outcomethan do non-hyperdiploid patients.

FISH analysis has demonstrated that chromosomal aberrations can be found inthe majority of PCM cases, despite the relatively high incidence of a normal kary-otype identified by conventional banding studies. In a Mayo Clinic study, Dewaldet al. [42] identified one or more abnormalities by FISH in 86% of newly diagnosedPCM cases using a panel of FISH probes to detect t(4;14), t(11;14), t(14;16), 17p13deletion, and monosomy 13/del(13q).

Identification of monosomy 13/del(13q) by FISH is the most common abnormal-ity detected in PCM. FISH studies have revealed that monosomy 13/del(13q) occursin all stages of plasma cell neoplasms including monoclonal gammopathy of unde-termined significance (MGUS), PCM, and plasma cell leukemia (PCL); however,the net effect of monosomy 13/del(13q) on prognosis is stronger when monosomy13/del(13q) is detected by karyotype than when it is observed by FISH [42]. Thisis because the observation of abnormal metaphases indicates a larger tumor bur-den with a highly proliferative malignant plasma cell component. While monosomy13/del(13q) has been associated with shorter survival and lower response rates totreatment, some recent studies have suggested that it may not, as the sole abnor-mality, be as important a prognostic marker. Instead, its close association with otherpoor prognostic markers such as t(4;14), t(14;16), and p53 gene deletion may con-tribute to the perception that monosomy 13/del(13q) predicts an adverse outcome.At present, the prognostic significance of chromosome 13 abnormalities in PCM isnot completely clear [41].

The IgH/14q32 translocation is detected in more than 50% of PCM cases andis strongly associated with the non-hyperdiploid group [43]. This rearrangementis mediated mostly by errors in immunoglobulin class switch recombination [44]and is believed to be an early, possibly pathogenic event in many cases [45].Three major specific IgH translocations t(11;14)(q13;q32), t(4;14)(p16.3;q32), andt(14;16)(q32;q23) are identified in PCM. The t(4;14) and t(14;16) are cryptictranslocations found in less than 15 and 5% of patients, respectively. They can onlybe detected accurately utilizing the FGFR3/IgH and MAF/IgH FISH dual-fusionFISH probes. Both t(4;14) and t(14;16) are associated with hypodiploidy, an adversedisease outcome with shorter survival, and aggressive clinical features. Less com-mon translocations involving IgH have also been described involving partner genessuch as IRF4, IRTA1/IRTA2, and C-MYC [41]. Secondary IgH translocations thatdysregulate the C-MYCproto-oncogene are found in 5% of PCM cases [40]. Thosetypes of translocations are considered late progression events and are likely to havea negative impact on overall prognosis.

The t(11;14) is the most common translocation observed in PCM, being seen in15–20% of cases. While it can be detected easily by G banding, the IgH/CCND1FISH probe is useful to examine metaphase cells with a complex karyotype or poor


morphology, and to asses interphase nuclei in cases that yield normal metaphases.The t(11;14) results in upregulation of CCND1and is associated with a favorableprognosis. Other cyclin genes such as CCND2 and CCND3 have also been found tobe upregulated in both MGUS and PCM. It appears that almost all cases of MGUSand PCM tumors upregulate at least one cyclin gene, sometimes as a consequence ofan IgH translocation [45]. Some have proposed that a classification system for PCMbased on the type of cyclin gene expressed together with the karyotypic profile maygenerate useful biological and clinical subgroups.

Deletion of 17p13/p53 gene using a locus-specific p53 gene FISH probe isdetected in 9–30% of PCM cases. This abnormality is identified more often in non-hyperdiploid PCM (26%) than in the hyperdiploid group (1%) [40]. Deletion of17p13 is associated with a poor prognosis in PCM.

It is clear that identification of cytogenetic abnormalities by both conventionalkaryotyping and FISH studies provides important diagnostic and prognostic infor-mation in PCM. The FISH panel utilized by most cytogenetics laboratories innewly diagnosed PCM cases includes enumeration probes for chromosomes 3,9, and 15 to screen for ploidy (gain of chromosomes 3, 9, and 15 is found in>90% of hyperdiploid cases) as well as probes to detect monosomy 13/13q dele-tion (RB1/LAMP1), p53 gene deletion, and common IgH translocations (Fig. 2.10).This methodology yields significant prognostic information for risk assessmentand treatment stratification in patients with PCM. In order to increase the sen-sitivity of FISH in PCM, some labs are now employing techniques that enrich

Fig. 2.10 Plasma cell myeloma FISH panel demonstrates the following: (a, b) loss of the D13S319and RB1 loci (one orange signal) with retention of the LAMP1 locus at 13q34 (two green signals)[this differentiates a chromosome 13q deletion from monosomy 13] and (c) loss of the p53 gene(one orange signal indicated by arrows)

70 M.A. Micale

for the plasma cell component in either whole-nuclei preparations or paraffin-embedded tissue sections. Originally described by Ahmann and colleagues [46],simultaneous FISH and cytoplasmic immunoglobulin staining permits analysisof only cells that express a plasma cell phenotype. Other techniques includingMay-Grunwald Giemsa (MGG) staining and FISH (target FISH or T-FISH) [47],FICTION (fluorescence immunophenotyping and interphase cytogenetics as a toolfor the investigation of neoplasms) [48], and SNP microarrays combined with FISH[49] have been used to enrich for the malignant component in FISH analysis ofplasma cell disorders.

Non-Hodgkin Lymphoma

The majority of non-Hodgkin lymphomas (NHLs) demonstrate clonal chromoso-mal abnormalities. The primary aberrations are commonly translocations that causerelocation of oncogenes to the vicinity of highly active promoter/enhancer elementsof immunoglobulin or T-cell receptor genes in B-cell or T-cell lymphoma, respec-tively, resulting in gene deregulation [50]. Unlike most of the translocations in acuteand chronic leukemias that result in a hybrid fusion gene with altered activity, thetranslocations in B-cell lymphoma mostly result in juxtaposition (not fusion) ofthe oncogene to an immunoglobulin gene regulatory sequence. One exception isthe API2–MALT1 fusion gene generated by the t(11;18)(q21;q21) in MALT lym-phoma. In some B-lineage lymphomas such as Burkitt lymphoma [t(8;14)(q24;q32)and its variants] or mantle cell lymphoma [t(11;14)(q13;q32)], one or a few specificIgH rearrangements are identified in the majority of cases and are thus consid-ered pathognomonic for the disease; however, in other B-cell neoplasms such asdiffuse large B-cell lymphoma, IgH rearrangements are detectable in a smaller num-ber of cases and are translocated with a wide variety of partner genes. In general,few translocations identified in B-cell neoplasms are characteristic of a specificlymphoma subtype [41].

Cytogenetic and molecular studies have provided evidence that the process ofoncogenesis in many lymphomas follows a multistep process similar to that origi-nally described for colorectal cancer. It appears that the primary genetic event ofa tumor clone initiates the lymphoid malignancy. These genetic alterations thusserve as diagnostic markers for the malignancy; however, additional changes wouldappear to be necessary for sustained lymphomagenesis. One line of evidence tosupport this notion is the molecular identification in apparently healthy individualsof genetic alterations such as the t(14;18) or the t(11;14). Whether these individ-uals are at higher risk for subsequent development of malignancy is not clear,but it seems that one or more additional genetic alterations are necessary for thedevelopment of frank malignancy. These secondary genetic changes are often iden-tified along with the defined primary change in the diagnostic specimen. These,and further genetic changes, result in increasing complexity of the karyotype andare associated with transformation of an indolent lymphoma to one with moreaggressive biological behavior. Thus, identification of complex karyotypes in the


diagnostic lymphoma specimen, or cytogenetic evolution with increasing karyotypiccomplexity, is associated with a poorer prognosis [41].

Conventional cytogenetic analysis is not always possible in lymphomas due tothe lack of fresh tissue and small biopsy specimens. FISH can be used to estab-lish the diagnosis in viable and fixed tissue and to assess the involvement ofbone marrow by lymphoid tumor. As unfixed tissue may not be available, FISHon paraffin-embedded tissue sections can be an invaluable technique to identifygenetic aberrations in lymphoid malignancies, as can FISH analysis of touch imprintspecimens [51]. Studies have shown that the sensitivity of FISH for detectinglymphoma-associated chromosome translocations is higher and more specific thanPCR owing, in part, to the large genomic region over which some of the transloca-tion breakpoints are spread. This can preclude their detection by molecular methodsin a highly sensitive fashion. In mantle cell lymphoma, for instance, FISH was foundto be superior to PCR with a 95–100% detection rate of IgH/CCND1 gene fusion ascompared with a detection rate of 35–40% by PCR [52].

Follicular lymphoma. The most frequent translocation in B-cell NHL,t(14;18)(q32;q21), juxtaposes the BCL2 proto-oncogene at 18q21 next to the IgHgene locus at 14q32 (Fig. 2.11). This translocation is identified in 80–90% of follic-ular lymphoma (FL) cases and to a lesser extent in diffuse large B-cell lymphoma(20–30%). The translocated BCL2 gene encodes an aberrant protein that inhibitsapoptosis. Only in 10% of cases is the t(14;18) the sole abnormality. A numberof non-random secondary changes are documented, the most common of which isan additional copy of the derivative chromosome 18 originating from the t(14;18)[der(18)t(14;18)(q32;q21)]. Low-grade FL can progress to high-grade FL or trans-form to diffuse large B-cell lymphoma (DLBCL) through acquisition of additionalcytogenetic changes, an event associated with a poorer prognosis [41].

Another recurrent primary abnormality in FL is rearrangement of the BCL6gene at band 3q27. This rearrangement occurs through a variety of chromosomalabnormalities involving various partner genes. In fact, BCL6 rearrangement appearsto be an extremely common event in a variety of B-cell disorders, in particularDLBCL. Rearrangements of BCL6 result in dysregulation through its interaction

Fig. 2.11 (a) A patient with follicular lymphoma demonstrates the t(14;18) by conventional cyto-genetic analysis. (b) FISH analysis utilizing a dual-color, dual-fusion probe reveals one red, onegreen, and two fusion signals (solid arrows) indicating fusion of IGH and BCL2genes. The hatchedarrow indicates a nucleus with a normal signal pattern

72 M.A. Micale

with gene regulatory sequences of the partner gene in the translocation. High-gradeFL with BCL6 rearrangement but without the t(14;18) often progresses to DLBCLas well [41].

Diffuse large B-cell lymphoma. No chromosomal abnormality is specific for dif-fuse large B-cell lymphoma. Many of the chromosomal abnormalities observed inother B-cell lymphomas can be observed in this disease as well. These abnormalitiesinclude BCL6 gene disruption (20–40% of cases); translocations of 14q32 involvingthe IgH locus (20–40%); gain of chromosomes X, 3, 7, 12, and 18; and loss of chro-mosomes Y, 6, 13, 15, and 17. There is at present conflicting evidence regarding theprognostic significance of either BCL6 rearrangement or t(14;18), the most com-monly observed translocation of 14q32 being observed in DLBCL. However, like inother lymphomas, del(17p) involving the p53 gene as well as karyotypic complexityindicates disease progression and a poorer prognosis [41]. Of interest is the findingthat t(14;18) can occur concurrently with chromosome 8q24/MYC gene transloca-tion in a number of B-cell neoplasms, including DLBCL. These neoplasms are ofhigh grade and are associated with a poorer prognosis [53].

Burkitt lymphoma. The Burkitt lymphoma (BL)-associated translocations includet(8;14)(q24;q32), t(2;8)(p12;q24), and t(8;22)(q24;q11). The t(8;14) is observed in75–85% of all BL patients (Fig. 2.12), while the remaining 15–25% of patientspresent one of the variant translocations, with the t(8;22) seen twice as frequently as

Fig. 2.12 This karyotype from a patient with Burkitt lymphoma demonstrates the t(8;14)(q24;q32) (solid arrows). An add (19q) chromosome is also present (open arrow)


the t(2;8). These translocations juxtapose the C-MYC proto-oncogene at 8q24 nextto the promoter for the Ig heavy chain gene at 14q32, Ig kappa locus at 2p12, or Iglambda locus at 22q11. This repositioning of the MYC gene disrupts its regulationand results in its constitutive overexpression leading to malignant transformation.Activation of MYC takes place on the der(14) in the t(8;14) and on the der(8) in thet(2;8) and t(8;22). Molecular analysis of the breakpoints in sporadic, endemic, andimmunodeficiency-associated BL demonstrates different clustering on the der(8)and the der(14), suggesting that different pathogenetic mechanisms may gener-ate the t(8;14) in different disease settings. A characteristic feature of BL is thatone of the three characteristic translocations is generally part of a relatively sim-ple karyotype, with karyotypic complexity indicating disease progression. Amongsecondary chromosomal abnormalities, the most common is structural rearrange-ment of chromosome 1, especially the long arm, as well as trisomy 7 and trisomy12 [41].

Mantle cell lymphoma. The t(11;14)(q13;q32) is present in virtually all cases ofmantle cell lymphoma (MCL). In 20% of cases, it is part of a more complex kary-otype, sometimes associated with loss of the der(11) chromosome. Chromosomenumbers are generally in the diploid or the hyperdiploid range, except in the blas-tic variants where polyploidy is often observed. The t(11;14) involves a breakpointwithin the BCL1 gene locus at 11q13 that results in relocation of the CCND1 gene(which is positioned downstream from BCL1) next to the promoter for the IgHgene. This results in the overexpression of CCND1. Identification of the t(11;14) isimportant as it can differentiate MCL from other low-grade lymphomas, especiallyif immunophenotyping is inconclusive [41].

Splenic marginal zone lymphoma. Up to 40% of splenic marginal zone lym-phomas present a del(7q) chromosome. The t(11;14)(q13;q32) has also beenreported; however, it is unclear whether these cases may have been MCL [41].

Extranodal marginal zone B-cell lymphoma of mucosa-associated lymphoidtissue (MALT type). Three recurrent translocations are observed in MALT lym-phomas. These include t(11;18)(q21;q21), t(14;18)(q32;q21), and t(1;14)(p22;q32).Trisomies 3 and 18 are observed in translocation-negative MALT lymphomas. Likeother lymphomas, as MALT lymphomas progress, they acquire additional secondarychromosomal changes including MYC gene translocations (8q24), del(17p) with lossof p53 and del(9p) with loss of the CDKN2A locus. The presence of the t(11;18)and possibly the t(1;14) is associated with a low probability of cure by antibiotictherapy that targets Helicobacter pylori, the infectious agent responsible for thedevelopment of gastric MALT lymphoma [41].

Anaplastic large-cell lymphoma. The t(2;5)(p23;q35), which fuses the nucle-ophosmin (NPM) gene at 5q35 with the anaplastic lymphoma kinase (ALK) gene at2p23, is the most common translocation observed in anaplastic large-cell lymphoma(ALCL). Tumors with this translocation are generally of high grade and express theCD30 (Ki-1) antigen. The t(2;5) leads to the formation of a chimeric fusion proteinwith constitutive tyrosine kinase activity. Other translocations which fuse ALK toother partner genes have been identified in ALCL as well [41].

74 M.A. Micale

Hodgkin Lymphoma

Chromosome analysis in classical Hodgkin lymphoma (HL) often reveals normalkaryotypes due to the abundance of nonmalignant cells in the lesion; how-ever, cytogenetic studies by classical and FISH methods combined with CD-30immunofluorescence staining have revealed highly complex karyotypes with cyto-genetic instability, triploid/tetraploid metaphases, and multiple aneuploidies in theneoplastic Reed–Sternberg cells. In nodular lymphocyte-predominant Hodgkin lym-phoma (NLPHL), abnormalities involving the BCL6 gene at 3q27 are identified inup to 50% of cases, not surprising given that NLPHL shares many features withDLBCL, and may in fact be a non-Hodgkin lymphoma rather than a HL [41].

Array-Based Genomic Profiling of Hematolymphoid Disorders

The newest generation of hematolymphoid molecular analysis is based on thesimultaneous examination of thousands of small genomic segments utilizing arrayscontaining either oligonucleotides (60-mers) or single-nucleotide polymorphisms(SNPs). SNP analysis appears to be better suited for studying neoplasia as it candetect gene-dosage changes at a higher level of resolution than can oligos andcan also detect copy-number neutral loss of heterozygosity (acquired uniparentaldisomy). From a few hundred thousand to over one million individual loci canbe interrogated in a single assay depending on the type of SNP array used. Twotechnologies currently available involve the spotting of individual SNPs onto genechips (Affymetrix SNP Array) or adsorbed on microbeads (Illumina Infinium HDBeadChip). Some have referred to this technology as “molecular allelokaryotyp-ing” [54, 55]. One significant disadvantage of array-based studies is that presentplatforms cannot detect balanced chromosomal rearrangements, a common featureof many hematolymphoid disorders.

SNP arrays appear to provide concordant results when compared with FISH anal-ysis using disease-specific panels; however, SNP analysis may not be as sensitive asFISH for detecting low-level mosaicism. Sargent et al. [56] studied 100 CLL sam-ples utilizing both a typical CLL FISH panel and a 44 K oligonucleotide array anddemonstrated a high degree of concordance between FISH and array CGH, althoughlow-level mosaicism (<25% of nuclei positive for a chromosomal abnormality) wasoften not detected by array CGH.

Studies utilizing SNP array technology to genomically profile hematolymphoidneoplasms are becoming more numerous in the literature. These studies haverevealed clinically significant information previously unattainable by classical cyto-genetic and FISH analysis. Lehmann et al. performed SNP chip analysis on 56patients with early stage untreated CLL and identified not only abnormalities thatwere detected by simultaneous FISH analysis but also additional abnormalitiesincluding deletions of chromosomes 5q, 6q, and Xp. Whole-chromosome 13 uni-parental disomy (UPD) was also identified and appears to be a common findingin early stage CLL [54]. Kawamata et al. performed SNP chip analysis on 14


ALL samples at diagnosis, remission, and relapse. All cases demonstrated genomicabnormalities at relapse, with 10 samples acquiring additional changes not observedin the diagnostic specimen. These changes included deletion of the INK4A/ARF andNF2 genes. Also at relapse, uniparental disomy of chromosomes originally pre-senting in the diagnostic specimen as trisomy was identified, along with UPD ofchromosome region 16p12.3-pter. Interestingly, this SNP chip study also revealeddisappearance of deletions at relapse, possibly indicating that some of the clonesidentified at relapse were present but not identified at initial diagnosis [55]. SNP chipanalysis of AML/MDS samples reported by Akagi et al. [57] identified genomicabnormalities including uniparental disomy in 49% of samples previously found tohave a normal karyotype. These and other studies clearly demonstrate the power ofSNP array-based genetic analysis; however, for the foreseeable future, a combina-tion of conventional cytogenetics, FISH, and SNP array analysis will likely be thebest approach to studying hematolymphoid disorders.

References

1. Harris NL, Jaffe ES, Stein H, et al. A revised European–American classification oflymphoid neoplasms: a proposal from the international lymphoma study group. Blood.1994;84(5):1361–1392.

2. Harris NL, Jaffe ES, Diebold J, et al. World Health Organization classification ofneoplastic diseases of the haematopoietic and lymphoid tissues: report of the ClinicalAdvisory Committee Meeting, Airlie House, Virginia, November 1997. J Clin Oncol.1999;17(12):3835–3849.

3. Swerdlow SH, Campo E, Harris NL, et al. (eds.). WHO Classification of Tumors ofHaematopoietic and Lymphoid Tissues. Lyon: IARC; 2008.

4. McKusick VA. The growth and development of human genetics as a clinical discipline. Am JHum Genet. 1975;27:261–273.

5. Tjio JH, Levan A. The chromosome number of man. Hereditas. 1956; 42:1–16.6. Nowell PC, Hungerford DA. A minute chromosome in human chronic myelocytic leukemia

(CML). Science. 1960;132:1497.7. Rowley JD. A new consistent chromosomal abnormality in chronic myelogenous leukemia

identified by quinacrine fluorescent and giemsa staining. Nature. 1973;234:290–293.8. Zech L, Haglund U, Nilsson K. Characteristic chromosomal abnormalities in biopsies and

lymphoid-cell lines from patients with Burkitt and non-Burkitt lymphomas. Int J Cancer.1976;17:47–56.

9. Kallioniemi OP, Kallioniemi A, Piper J, et al. Optimizing comparative genomic hybridizationfor analysis of DNA sequence copy number changes in solid tumors. Genes ChromosomesCancer. 1994;10(4):231–243.

10. Schrock E, du Manoir S, Veldman T, et al. Multicolor spectral karyotyping of humanchromosomes. Science. 1996; 273:494–497.

11. Kraan J, von Bergh AR, Kleiverda K, et al. Multicolor fiber FISH. Methods Mol Biol.2002;204:143–153.

12. Shaffer LG, Bejjani BA, Torchia B, et al. The identification of microdeletion syndromes andother chromosome abnormalities: cytogenetic methods of the past, new technologies for thefuture. Am J Med Genet Part C Semin Med Genet. 2007;145C:335–345.

13. Sagoo GS, Butterworth AS, Sanderson S, et al. Array CGH in patients with learning disability(mental retardation) and congenital anomalies: updated systematic review and meta-analysisof 19 studies and 13, 926 subjects. Genet Med. 2009;11(3):139–146.

76 M.A. Micale

14. Gondek LP, Tiu R, O’Keefe CL, et al. Chromosomal lesions and uniparental disomydetected by SNP arrays in MDS, MDS/MPD, and MDS-derived AML. Blood. 2008;111(3):1534–1542.

15. Lehmann S, Ogawa S, Raynaud SD, et al. Molecular allelokaryotyping of early-stageuntreated chronic lymphocytic leukemia. Cancer. 2008;112(6):1296–1305.

16. Gaidzik V, Dohner K. Prognostic implications of gene mutations in acute myeloid leukemiawith normal cytogenetics. Sem Oncol. 2008;35(4):346–355.

17. Shaffer LG, Slovak ML, Campbell LJ, eds. ISCN (2009): An International System for HumanCytogenetic Nomenclature; 2009. S. Karger, Basel.

18. Lupski JR, Stankiewicz P. Genomic disorders: molecular mechanisms for rearrangements andconveyed phenotypes. PLoS Genet. 2005;1(6):e49.

19. Aplan PD. Causes of oncogenic chromosomal translocations. Trends Genet. 2006;22(1):46–55.

20. Gollin SM. Mechanisms leading to nonrandom, nonhomologous chromosomal translocationsin leukemia. Semin Cancer Biol. 2007;17(1):74–79.

21. Janz S, Potter M, Rabkin CS. Lymphoma- and leukemia-associated chromosomal transloca-tions in healthy individuals. Genes Chromosomes Cancer. 2003;36:211–223.

22. Vandenberghe P, Michaux L, Hagemijer A. Chronic myeloproliferative neoplasms. In:Heim S, Mitelman F, eds. Cancer Cytogenetics. 3rd ed. Hoboken, NJ: Wiley; 2009.pp. 209–232.

23. Landstrom AP, Tefferi A. Fluorescent in situ hybridization in the diagnosis, prognosis,and treatment monitoring of chronic myeloid leukemia. Leuk Lymphoma. 2006;47(3):397–402.

24. Dingli D, Schwager SM, Mesa RA, et al. Presence of unfavorable cytogenetic abnormali-ties is the strongest predictor of poor survival in secondary myelofibrosis. Cancer. 2006;106:1985–1989.

25. Cools J, DeAngelo DJ, Gotlib J, et al. A tyrosine kinase created by fusion of the PDGFRA andFIP1L1 genes as a therapeutic target of imatinib in idiopathic hypereosinophilic syndrome.NEJM. 2003;348(13):1201–1214.

26. La Starza R, Specchia G, Cuneo A, et al. The hypereosinophilic syndrome: fluorescence insitu hybridization detects the del(4)(q12)-FIP1L1/PDGFRA but not genomic rearrangementsof other tyrosine kinases. Haematologica. 2005;90(5):596–601.

27. Schiffer CA. World Health Organization and International Prognostic Scoring System:the limitations of current classification systems in assessing prognosis and determiningappropriate therapy in myelodysplastic syndromes. Semin Hematol. 2008;45(1):3–7.

28. Sreekantaiah C. FISH panels for hematologic malignancies. Cytogenet Genome Res.2007;118:284–296.

29. Rigolin GM, Bigoni R, Milani R, et al. Clinical importance of interphase cytogeneticsdetecting occult chromosome lesions in myelodysplastic syndromes with normal karyotype.Leukemia. 2001;15:1841–1847.

30. Atlas of Genetics and Cytogenetics in Oncology and Haematology. URL:http://AtlasGeneticsOncology.org

31. Borthakur G, Estey EE. Therapy-related acute myelogenous leukemia and myelodysplasticsyndrome. Curr Oncol Rep. 2007;9(5):373–377.

32. Harrison CJ, Johansson B. Acute lymphoblastic leukemia. In: Heim S, Mitelman F, eds.Cancer Cytogenetics. 3rd ed. Hoboken, NJ: Wiley; 2009. pp. 233–296.

33. Sutcliffe MJ, Shuster JJ, Sather HN, et al. High concordance from independent studies bythe Children’s Cancer Group (CCG) and Pediatric Oncology Group (POG) associating favor-able prognosis with combined trisomies 4, 10, and 17 in children with NCI standard-riskB-precursor acute lymphoblastic leukemia: a Children’s Oncology Group (COG) initiative.Leukemia. 2005;19(5):734–740.

34. Harrison CJ, Moorman AV, Barber KE, et al. Interphase molecular cytogenetic screeningfor chromosomal abnormalities of prognostic significance in childhood acute lymphoblastic


leukemia: a UK Cancer Cytogenetics Group Study. Br J Haematol. 2005;129(4):520–530.

35. Dohner H, Stilgenbauer S, Benner A, et al. Genomic aberrations and survival in chroniclymphocytic leukemia. N Engl J Med. 2000;343:1910–1916.

36. Reddy KS. Chronic lymphocytic leukaemia profiled for prognosis using a fluorescence in situhybridisation panel. Br J Haematol. 2006;132:705–722.

37. Nelson BP, Gupta R, Dewald GW, et al Chronic lymphocytic leukemia FISH panel: impact ondiagnosis. Am J Clin Pathol. 2007;128:323–332.

38. Dewald GW, Brockman SR, Paternoster SF, et al. Chromosome anomalies detected by inter-phase fluorescence in situ hybridization: correlation with significant biological features ofB-cell chronic lymphocytic leukaemia. Br J Haematol. 2003;121:287– 295.

39. Fonseca R, Barlogie B, Bataille R, et al. Genetics and cytogenetics of multiple myeloma: aworkshop report. Cancer Res. 2004;64:1546–1558.

40. Mohamed AN, Bentley G, Bonnett M, et al. Chromosomes aberrations in a series of 120multiple myeloma cases with abnormal karyotypes. Am J Hematol. 2007;82(12):1080–1087.

41. Siebert R. Mature B- and T-cell neoplasms and Hodgkin lymphoma. In: Heim S, Mitelman F,eds. Cancer Cytogenetics. 3rd ed. Hoboken, NJ: Wiley; 2009. pp. 297–374.

42. Dewald GW, Therneau T, Larson D, et al. Relationship of patient survival and chromosomeanomalies detected in metaphase and/or interphase cells at diagnosis of myeloma. Blood.2005;106(10):3553–3558.

43. Fonseca R, Debes-Marun CS, Picken EB, et al. The recurrent IgH translocations are highlyassociated with nonhyperdiploid variant multiple myeloma. Blood. 2003;102(7):2562–2567.

44. Bergsagel PL, Kuehl WM. Chromosome translocations in multiple myeloma. Oncogene.2001;20:5611–5622.

45. Bergsagel PL, Kuehl WM, Zhan F, et al. Cyclin D dysregulation: an early and unifyingpathogenic event in multiple myeloma. Blood. 2005;106:296–303.

46. Ahmann GJ, Jalal SM, Juneau AL, et al. A novel three-color, clone specific fluorescencein situ hybridization procedure for monoclonal gammopathies. Cancer Genet Cytogenet.1998;101:7–11.

47. Slovak ML, Bedell V, Pagel K, et al. Targeting plasma cells improves detection of cytoge-netic aberrations in multiple myeloma: phenotype/genotype fluorescence in situ hybridization.Cancer Genet Cytogenet. 2005;158:99–109.

48. Saez B, Martin-Subero JI, Odero MD, et al. Multicolor interphase cytogenetics for the studyof plasma cell dyscrasias. Oncol Rep. 2007;18(5):1099–1106.

49. Agnelli L, Mosca L, Fabris S, et al. A SNP microarray and FISH-based procedure to detectallelic imbalances in multiple myeloma: an integrated genomics approach reveals a wide genedosage effect. Genes Chromosomes Cancer. 2009;48(7):603–614.

50. Al-Katib A, Mohamed AN. Non-Hodgkin’s lymphoma and chronic lymphocytic leukemia. In:Runge M, Patterson C, eds. Principles of Molecular Medicine. 2nd ed. Totowa, NJ: HumanaPress; 2009. pp. 794–803.

51. Ventura R, Martin-Subero J, Jones M, et al. FISH analysis for the detection of lymphoma-associated chromosomal abnormalities in routine paraffin-embedded tissue. J Mol Diagn.2006;8:141–151.

52. Belaud-Rotureau MA, Parrens M, Dubus P, et al. A comparative analysis of FISH, RT-PCR, PCR, and immunochemistry for the diagnosis of mantle cell lymphomas. Mod Pathol.2002;15:517–525.

53. Kanungo, A, Medeiros LJ, Abruzzo LV, et al. Lymphoid neoplasms associated with concur-rent t(14;18) and 8q24/c-MYC translocation generally have a poor prognosis. Mod Pathol.2006;19:25–33.

54. Lehmann S. Ogawa S, Raynaud SD, et al. Molecular allelokaryotyping of early-stage,untreated chronic lymphocytic leukemia. Cancer. 2008;112:1296–1305.

55. Kawamata N, Ogawa S, Seeger K, et al. Molecular allelokaryotyping of relapsed pediatricacute lymphoblastic leukemia. Int J Oncol. 2009;34:1603–1612.

78 M.A. Micale

56. Sargent R, Jones D, Abruzzo LV, et al. Customized oligonucleotide array-based compara-tive genomic hybridization as a clinical assay for genomic profiling of chronic lymphocyticleukemia. J Mol Diagn. 2009;11(1):25–34.

57. Akagi T, Ogawa S, Dugas M, et al. Frequent genomic abnormalities in acutemyeloid leukemia/myelodysplastic syndrome with normal karyotype. Haematologica.2008;94(2):213–223.

58. Micale MA. Molecular cytogenetic (FISH) analysis of hematolymphoid disorders. In:Marchant K, ed. Laboratory Hematology Practice. 1st ed. ISLH; 2010, in Press.

Chapter 3Using Cytogenetic and Molecular Testsin Diagnostic Workups with the WHOClassification – 2008

Clarence C. Whitcomb

Keywords WHO classification of hematopoietic neoplasms · Cytogenetictesting · Molecular tests · Immunohistochemical staining · Flow cytome-try · Karyotype · DNA and DNA testing, Chronic myeloproliferative neo-plasms · DNA sequencing · Comparative genomic hybridization · Geneexpression profiles · Chronic myelogenous leukemia · BCR–ABL1 · Polycythemiavera · Essential thrombocythemia · Primary myelofibrosis · Chronic neutrophilicleukemia · Myeloid and lymphoid neoplasms with eosinophilia · PDGFRA· PDGFRB · FGFR1 · Chronic eosinophilic leukemia · Chronic myelomono-cytic leukemia · Atypical chronic myeloid leukemia · Juvenile myelomonocyticleukemia · Myelodysplastic/myeloproliferative neoplasm · Myelodysplastic syn-drome · Refractory cytopenia with unilineage dysplasia · Refractory anemia withexcess blasts · Refractory anemia with ring sideroblasts · Refractory cytopenia withmultilineage dysplasia · Childhood myelodysplastic syndrome · Acute promyelo-cytic leukemia · B-cell ALL · AML with minimal differentiation · AML withoutmaturation · AML with maturation · Acute myelomonocytic leukemia · Acutemonoblastic leukemia · Acute erythroid leukemia · Acute megakaryoblasticleukemia · Acute basophilic leukemia · Acute panmyelosis with myelofibro-sis · Malignant lymphomas · Chronic lymphocytic leukemia · Follicular lym-phoma · Mantle cell lymphoma · Marginal zone B-cell lymphoma · Diffuse largeB-cell lymphoma · Burkitt lymphoma · T-cell lymphoma · Anaplastic large-celllymphoma · NK lymphoma · Hodgkin lymphoma · Mast cell neoplasms

A critical review of the neoplastic disorders of hematopoietic cells has been con-ducted by panels of expert clinicians and pathologists under the sponsorship ofthe World Health Organization (WHO), and from this effort a comprehensive clas-sification for these disorders has been developed. Clinical features, phenotypiccharacteristics of the cellular proliferations, and, in many instances, genotypic fea-tures of the abnormal cells are all used in characterizing these disorders. Detailed

C.C. Whitcomb (B)Department of Pathology, Miller School of Medicine, University of Miami, Miami,FL 33136, USAe-mail: [email protected]


80 C.C. Whitcomb

descriptions and discussions of diagnostically important features for the entitiesdefined within this classification have been summarized and published in a mono-graph – WHO Classification of Tumors of Haematopoietic and Lymphoid Tissues[1] – which represents the consensus of the many experts who have contributed tothe WHO project.

Findings from cytogenetic and molecular test procedures are included as impor-tant diagnostic criteria for many of the entities. Because this classification is usedalmost universally today, cytogenetic and molecular testing of blood, bone marrow,or tissue has become an integral component in the diagnostic workups for patientsin whom these disorders are suspected. The following is a summary of the use ofsuch tests in diagnostic workups, following the guidelines of the WHO classifica-tion. Particular emphasis is focused on entities and situations for which cytogeneticand molecular testing is required or can be very effectively used today.

Detailed discussions of clinical, morphological, phenotypic, and genetic featuresof the hematopoietic neoplasms as well as specific criteria for diagnosing thesedisorders are presented in the WHO monograph cited above [1]. The discussionhere is more general in focus, and this chapter is intended only as a guide to theWHO monograph, which should serve in practice today as the “working manual”for hematopathologists. The disorders are discussed in the sequence in which theyare presented in the WHO monograph and the nomenclature used also follows thestyle of that work. An extensive list of references pertinent to the specific disor-ders is included in the WHO monograph. Selected references of a more generalnature or which relate to practical issues arising during actual diagnostic workupsare included here.

Diagnostic Workups

Cytogenetic and molecular tests are technically specialized procedures that are oftenperformed in regional facilities. Multiple individuals, at the “primary” site of care aswell as in separate reference laboratories, may be involved in the generation of datafrom which a final diagnostic interpretation is formulated. Coordination of the manyand varied activities involved in receiving, preparing, and distributing materials fortesting, correlating test results with observations from traditional morphologicalstudies and integration of the data from all of the procedures into a comprehensivesummary are responsibilities of the hematopathologist. Careful case managementduring the workup is essential, and this is a non-trivial task.

A diagnostic workup is a multi-step process. Procedures are used in a sequentialmanner to refine an evolving diagnostic impression. Evaluation of stained smearsand tissue sections serves as the first step in a workup, for traditional morphologicalobservations still provide a firm basis upon which at least a working differen-tial diagnosis can be formulated [2]. An initial diagnostic impression is refinedby more detailed phenotypic characterization of cells from the lesional tissue.Flow cytometry analysis of cells from the tissue is widely used for this purpose,and this technique is very effective in workups of the leukemic disorders [3].

3 Cytogenetic and Molecular Tests in Diagnostic Workups 81

Immunohistochemical staining procedures are widely used in the workups of tis-sue tumors. The information generated by these “first-stage” activities may sufficefor a working diagnosis, but often important issues will not have been completelyaddressed by the initial procedures. Cytogenetic and molecular tests are then usedas “second-stage” studies to provide additional critical information. The specializedmolecular and cytogenetic procedures can also be very helpful in the first stagesof a workup, however – particularly in workups of tissue tumors – where they canhelp resolve “basic” diagnostic question such as whether a cellular proliferation isneoplastic at all or whether it represents a “benign” reactive process [4–6].

Immunohistochemical staining is a basic tool in the practice of pathology today[7]. The technique is rapid and inexpensive, but its application is limited to fixedcells and tissues, and the antibody reagents used must be reactive with these materi-als. A wider repertoire of antibodies can be used in flow cytometry analysis of cellsin suspensions and that technique has greater analytic sensitivity. However, flowcytometry analysis requires fresh tissue, and, for useful information to be derived,sufficient viable cells must be isolated from the specimen. It also does not affordeasy correlation of analytic results with histomorphology.

Cytogenetic tests provide information related to the cell genome. Conventionalkaryotyping can detect deletions and amplifications of individual chromosomes aswell as a wide range of structural alterations, such as translocations of segmentswithin and among the chromosomes. A karyotype provides a comprehensive viewof the genome at a relatively “gross” level, but it also requires fresh cells and is time-consuming and expensive. In addition, not all diagnostically important structuralabnormalities will be detected by this technique. Fluorescent in situ hybridizationanalysis (FISH) can target specific chromosomal segments or genes, and FISHprocedures are rapid and can be performed with non-dividing cells in smears orhistologic sections [8, 9]. In addition, FISH procedures can detect some struc-tural alterations not discernable by routine karyotype analysis. When performedin situ, these cytogenetic procedures also facilitate correlation of cytogenetic andmorphologic findings.

Analyses of DNA or RNA provide information at a molecular level. When usedto evaluate the antigen receptor genes (ARG) of B cells or T cells, these analysesare very useful for demonstrating the presence of a clonal lymphoid cell popula-tion within a background of “normal” cells. A clonal population is detected byvirtue of the component cells having a similar ARG rearrangement, which is demon-strated in the analysis as a distinct “monoclonal” signal that is distinguished from“polyclonal” readings produced by the background cells. Tests for clonal ARGrearrangements are frequently used to assess whether a cellular proliferation is neo-plastic or reactive – a “positive” finding of a monoclonal T- or B-cell populationbeing considered as evidence supporting an interpretation of neoplasia [10].

Molecular tests do not provide information of a “gross” nature such as gainsor losses of whole chromosomes or chromosomal segments, but they are excellenttools for detecting specific structural abnormalities. Molecular techniques can beused as an alternative to FISH procedures for detecting chromosome translocationsin many instances. Molecular tests provide the greatest analytic sensitivity, and they

82 C.C. Whitcomb

are very useful in evaluating specimens for residual disease detection [11]. However,due to variations in the structural rearrangements that can alter the chromosomes,some important translocations may not be detected. In these cases FISH analysiscan be diagnostically more sensitive and should be used if molecular testing yieldsa “negative” finding.

Molecular tests are performed using DNA and RNA extracted from cells. Carefultissue handling is required to obtain high-quality material for molecular testing.Expeditious processing is particularly required to obtain material suitable for RNAanalysis. In the most direct approach a portion of fresh tissue can be submit-ted for molecular analysis in the first stage of the workup. If molecular testingwill be deferred to a later stage, the tissue aliquot may be preserved by freezing,preferably at –70◦C. Alternatively, for DNA analyses using polymerase chain reac-tion techniques, material for analysis can be extracted from fixed tissue that hasbeen embedded in paraffin blocks. The latter approach is very useful when theamount of diagnostic tissue is limited, and it facilitates correlation of molecularand morphological findings.

Highly technical cytogenetic and molecular procedures such as comparativegenomic hybridization (CGH) [12], sequencing of DNA, and generation of geneexpression profiles (GEP) by array analyses [13] – among other techniques –have been used widely in investigational studies. These studies have contributedimmensely to the current understanding of pathogenetic mechanisms involved inthe neoplastic hematopoietic disorders, and findings from some studies have beentranslated into new approaches in diagnosis and clinical management. These moretechnically demanding procedures are not used in the routine diagnostic settingat this time, and they will not be discussed further here, except in the context ofpossible future applications.

Chronic Myeloproliferative Disorders

The following sections summarize information that is presented in great detail in theWHO monograph cited above. Most of the factual assertions and recommendationsmade in the following comments are taken from that monograph [1]. A very usefulsummary of the cytogenetic and molecular findings used in defining the myeloiddisorders has been also recently published [14].

The myeloproliferative neoplasms (MPN) include chronic myelogenousleukemia, BCR–ABL1 positive, polycythemia vera, essential thrombocythemia, andprimary myelofibrosis, as well as entities designated chronic neutrophilic leukemia,myeloid and lymphoid neoplasms with eosinophilia and abnormalities of PDGFRA,PDGFRB, or FGFR1, chronic eosinophilic leukemia, NOS, and myeloproliferativeneoplasm, unclassifiable. Each of these entities is characterized by the presence of aclonal proliferation of myeloid – i.e., granulocytic, erythroid, and megakaryocytic –cells. Cytogenetic and molecular tests are essential in the workups of these disor-ders. Indeed, the use of such tests in the diagnosis and treatment of patients withchronic myelogenous leukemia (CML) is a paradigm for how such testing may


Table 3.1 Molecular and cytogenetic findings in myeloproliferative neoplasms

WHO category Required findings Useful but not required

Chronic myelogenousleukemia, BCR–ABL1positive

BCR–ABL1 POSSee text

Chronic neutrophilicleukemia

BCR–ABL1 NEG JAK2 POS – occasionalcases

Polycythemia vera BCR–ABL1 NEG JAK2 POS – most casesPrimary myelofibrosis BCR–ABL1 NEG JAK2 POS – 50% of casesEssential thrombocythemia BCR–ABL1 NEG JAK2 POS – 40–50%Myeloid and lymphoid

neoplasms witheosinophilia andabnormalities of PDGFRA,PDGFRB, or FGFR1

BCR–ABL1 NEGVariable translocations

involving PDGFA (4q12),PDGFB(5q31-33), orFGFR1(8p11-12) may be found

Chronic eosinophilicleukemia, NOS

BCR–ABL1 NEG; PDGFA,PDGFB, and FGFR1 NEG

JAK2 POS – occasionalcases

Myeloproliferative neoplasm,unclassifiable

BCR–ABL1 NEG; PDGFA,PDGFB, and FGFR1 NEG

JAK2 POS – occasionalcases

be used in the management of hematopoietic neoplasms. Table 3.1 summarizesfindings of relevant cytogenetic and molecular tests for the disorders of the MPNgroup.

Abnormalities of genes encoding proteins involved in intracellular signalingpathways are pathogenetically associated with the MPN disorders. Tests to demon-strate these abnormalities are useful in diagnostic workups. Demonstration of thecharacteristic “Philadelphia chromosome,” which results from a reciprocal translo-cation of material between chromosomes 9 and 22 – designated in cytogeneticterminology as t[9;22](q34;q11.2) – has long been used for the diagnosis of CML.The translocation juxtaposes portions of the ABL1 gene on chromosome 9 (9q34)and the BCR locus on chromosome 22 (22q11.2) to create an abnormal “fusiongene” BCR–ABL1. A “positive” finding of an abnormal BCR–ABL1 gene is a sinequa non for the diagnosis of CML.

A positive test for BCR–ABL1 will be found in more than 90% of patients withCML. This abnormality can be detected by karyotyping, FISH analysis, or molec-ular procedures. Different BCR–ABL1 fusion genes can be formed depending uponthe site(s) of breakage and recombination within the involved chromosomes. Theusual translocation seen in CML results in a fusion gene that encodes a 210 kDprotein. Breaks at other sites can result in fusion genes and gene products of differ-ent molecular sizes (e.g., p190 or p230 kD). A variant translocation that may not bedetected by routine karyotyping can also occur. Therefore, other analytic techniquesshould be used when a “negative” result is found by karyotyping in a clinically sus-picious case. Quantitative molecular tests for BCR–ABL1 have been developed, andthese are now used to monitor the effects of therapy and to provide guidance formanagement decisions such as the timing of bone marrow transplantation [15].

84 C.C. Whitcomb

All of the other MPN disorders lack the distinctive BCR–ABL1 abnormality, butmutations of the JAK2 gene are associated with several of these entities. The mostcommon mutation of JAK2 is designated as V617F [16]. A mutated JAK2 gene isnot specific for any particular BCR–ABL1-negative MPN, and JAK2 mutations maybe seen in rare cases of CML as well, but a “positive” molecular test finding is veryuseful for distinguishing a BCR–ABL1-negative myeloproliferative neoplasm froma reactive hyperplasia of myeloid cells [17].

The well-recognized BCR–ABL1-negative myeloproliferative neoplasms includethe following:

1. Chronic neutrophilic leukemia: This rare disorder can resemble CML mor-phologically, but it is characterized by a predominance of mature neutrophils.Distinction from reactive neutrophilia is very difficult. A positive test for a JAK2mutation will help resolve this problem. Rare cases of CML with a neutrophiliasimilar to that of chronic neutrophilic leukemia do occur, but the presence of aBCR–ABL1 fusion gene should distinguish these cases. The fusion gene in suchcases of CML is often the result of a translocation producing a p230 kD protein.

2. Polycythemia vera: The JAK2 V617F mutation can be demonstrated in almostall cases. A positive finding for a JAK2 mutation is now a major criterion for thediagnosis of this disorder.

3. Primary myelofibrosis: Characteristic morphological features in bone marrowcore biopsies are the primary features used for diagnosis. A JAK2 mutation canbe found in about half of cases, however, and a positive finding is very useful inmorphologically equivocal cases.

4. Essential thrombocythemia: JAK2 mutation can be demonstrated in about half ofthe cases. A positive test finding is helpful in differentiating this disorder from areactive thrombocytosis.

Other BCR–ABL1-negative disorders in the MPN group include the following:

5. Myeloid and lymphoid neoplasms with eosinophilia and abnormalities ofPDGFRA, PDGFRB, or FGFR1: This diagnosis may be used if a translo-cation involving one of the genes PDGFRA(4q12), PDGFRB(5q31-33), orFGFR1(8p11-12) can be demonstrated. Abnormalities of these genes have beenfound in several neoplastic disorders in which eosinophilia is a conspicuous fea-ture, and, because of these associations, this new entity has been added to theWHO classification in its most recent edition of 2008. Rare cases of CML witheosinophilia can be seen, but the presence of a BCR–ABL1 fusion gene shoulddistinguish these cases. Therefore, tests for BCR–ABL1 should be used in con-junction with tests for these translocations when eosinophilia is a conspicuousfeature.

6. Chronic eosinophilic leukemia, NOS: This diagnosis is appropriate only whenan absence of eosinophilia-associated translocations has been demonstrated.

7. Myeloproliferative neoplasm, unclassifiable: This diagnostic category is appro-priate only for cases which are BCR–ABL1 negative, have no demonstra-ble abnormalities of PDGFRA, PDGFRB, or FGFR1, and which have no


morphological features that would support a diagnosis of one of the other BCR–ABL1-negative categories. Some of the “unclassifiable” cases may be JAK2positive.

Myelodysplastic Disorders

Two groups of entities designated as myelodysplastic/myeloproliferative neoplasms(MDS/MPN) and myelodysplastic syndromes (MDS) are included in the WHO clas-sification. These groups include a complex assortment of disorders characterizedby dysplastic morphological features involving mature and immature granulocytes,erythroid precursors, and/or megakaryocytes.

The MDS/MPN group includes the following:

Chronic myelomonocytic leukemiaAtypical chronic myeloid leukemia, BCR–ABL1 negativeJuvenile myelomonocytic leukemiaMyelodysplastic/myeloproliferative neoplasm, unclassifiableRefractory anemia with ring sideroblasts (RARS) associated with marked

thrombocytosis (provisional entity)

The MDS/MPN disorders typically present with clinical and morphological fea-tures that are more like those of the myeloproliferative neoplasms, i.e., they aremore “proliferative” in nature. Dysplastic features are often present, but may berather minimal in these disorders.

The MDS group includes the following entities:

Refractory cytopenia with unilineage dysplasiaRefractory anemia with ring sideroblastsRefractory cytopenia with multilineage dysplasiaRefractory anemia with excess blastsMyelodysplastic syndrome with isolated del(5q)Myelodysplastic syndrome, unclassifiableChildhood myelodysplastic syndromeRefractory cytopenia of childhood (provisional entity)

The MDS disorders usually present clinically with “cytopenic” features sug-gesting ineffective hematopoiesis. Morphological findings of dyserythropoiesis,dysgranulopoiesis, and/or dysmegakaryopoiesis are a hallmark of these disorders,and an accurate blast cell count is needed for diagnosis of these entities. Carefulevaluation of a peripheral blood smear and adequately cellular bone marrow aspi-rate smears by microscopy is essential, particularly for determination of the “blastcount.” Flow cytometry should not be substituted for morphological analysis for thispurpose.

86 C.C. Whitcomb

The BCR–ABL1 abnormality that characterizes CML is not found in any of theMDS/MPN or MDS disorders, but a variety of non-specific cytogenetic abnormali-ties are common. Deletions of segments from chromosome 7 or 5 or deletions of theentire chromosomes – del(5) or del(7) – are frequently found, and multiple abnor-malities are often present. A list of cytogenetic abnormalities that are typically seenin these disorders is included in the WHO monograph [1]. Although these cyto-genetic abnormalities are not completely specific for the MDS disorders, they areused as “signature” features, and a diagnosis of a disorder from the MDS groupcan be made even in the absence of conspicuous morphological dysplasia if suchcytogenetic abnormalities are present. Because no specific abnormality can be pre-dicted from the initial morphological findings, however, a full karyotype study isrecommended for diagnosis of these disorders.

The genetic abnormalities found in the MDS/MPN and MDS disorders are notassociated directly with the same intracellular signaling pathways involved in theMPN disorders, e.g., the JAK–STAT pathway. In some of the MDS/MPN entities,e.g., Juvenile myelomonocytic leukemia, mutations in the PTPN11, NRAS, KRAS, orNF1 genes that encode components of the RAS pathway are found. A JAK2 mutationmay be seen in some cases of myelodysplasia, but this finding should not supersedethe other genetic abnormalities associated with MDS as a diagnostic criterion.

A finding of a 5q deletion as the only cytogenetic abnormality characterizes theentity designated myelodysplastic syndrome with isolated del(5q). A few patientswith this rare disorder may also have a JAK2 mutation. Patients with the entitydesignated refractory anemia with ring sideroblasts (RARS) may have a markedthrombocytosis, and for these patients tests for JAK2 V617F should be obtained.If tests for JAK2 mutations are negative, studies of the MPL gene mutationsW515K/L are recommended as well. Finally, a diagnosis of any of the entities in theMDS/MPN and MDS groups should not be made without demonstrating “negative”findings in tests for BCR–ABL1 and also for translocations involving PDGFRA,PDGFRB, or FGFR1 if eosinophilia is a prominent feature.

Acute Leukemias

A diagnosis of acute leukemia is made when blasts constitute more than 20% ofthe nucleated cells in a cellular bone marrow aspirate smear. The initial diagnosisis usually made easily by morphological examination of peripheral blood and bonemarrow materials. Microscopic examination of an adequately cellular bone marrowaspirate smear – including a “differential count” of at least 500 nucleated cells – isstill the recommended technique for establishing a diagnosis of acute leukemia. Thecritical blast cell “count” should be determined by microscopy; quantitation by flowcytometry is not recommended.

The blast cells may be of either myeloid or lymphoid lineage. Characterization ofthe blasts to distinguish acute myeloid leukemia (AML) from acute lymphoblasticleukemia (ALL) and, more specifically, to identify B-lineage acute lymphoblas-tic leukemia (B-ALL) or T-lineage acute lymphoblastic leukemia (T-ALL) is a


necessary “first-stage” activity in the workup of an acute leukemia. Flow cytom-etry analysis is the preferred tool for determining the lineage of the blast cells, andthis is usually readily accomplished when specimens containing sufficient blast cellsare available. If sufficient cells cannot be obtained for analysis, morphologic tech-niques using immunostaining can be used. In some cases features of both myeloidand lymphoid lineages may be demonstrable.

Despite the fact that a clonal rearrangement of a lymphoid antigen receptorgene – IGH for B cells or one of the T-cell receptor (TCR) genes for T cells –can be demonstrated in most cases of ALL, and that these same genes are usuallyin germline configuration in most cases of AML, tests for lymphoid ARG rear-rangements are usually not needed in a workup of an acute leukemia. Becauserearrangements of both an IGH and a TCR gene may be found in either B-ALLor T-ALL, these tests are not helpful for lineage determination. Rare cases of AMLhave been reported in which a clonal lymphoid ARG rearrangement was present aswell. However, tests for a clonal T-lineage ARG rearrangement are useful for dis-tinguishing NK cells, which do not have rearranged TCR genes, from true T cells,and tests for a clonal IGH rearrangement can help differentiate leukemic blasts ofB-ALL from immature but polyclonal B cells (hematogones) that are often seen inregenerating bone marrow specimens post chemotherapy.

Subtypes of AML and B-ALL are distinguished by the presence of specific cyto-genetic translocations. These are designated as individual entities within the currentWHO classification. The defining cytogenetic abnormality must be demonstratedfor diagnosis of an individual case as one of these specific subtypes. No subclassi-fication of T-ALL based on cytogenetic abnormalities is part of the current WHOschema. Other non-specific cytogenetic abnormalities as well as mutations in FLTor KIT genes may be found in any of the specific subtypes of acute leukemia, butonly the designated abnormality is used as the “defining” criterion for classification.

The cytogenetically defined subtypes of acute leukemia identify patients forwhich predictable differences in the clinical course have been well demonstrated[18]. Recognition of these specific subtypes of acute leukemia is helpful for clin-ical assessment of risk of relapse, and specific abnormalities can also serve astarget “markers” in follow-up testing for residual or recurrent disease [19]. Withthe exception of acute promyelocytic leukemia (discussed below) complete cytoge-netic characterization for most cases can be completed as “second-stage” proceduresin the workup, but materials for these studies must have been taken and distributedappropriately from the materials initially received. The genetically defined subtypesof acute leukemia are listed in Table 3.2. The complete WHO terminology for thesesubtypes includes a formal designation of the defining translocation and the namesof the genes involved, but for clarity in Table 3.2 the entities are listed with onlythe gene names. The associated cytogenetic abnormality – the required finding – islisted in the following column.

The entity “acute promyelocytic leukemia with PML-RARA” is characterized bya specific cytogenetic translocation t(15;17) involving the retinoic acid receptor gene(RARA). This particular subtype of AML responds well to therapy that is designedto overcome the aberrant intracellular signaling that results from the aberration of

88 C.C. Whitcomb

Table 3.2 Cytogenetic findings in AML and ALL with recurrent genetic abnormalities

WHO category Required findings

AML with RUNX1-RUNX1T1 t(8;21)(q22;q22)AML with CBFB-MYH11 inv(16)(p13.1;q22) or t(16;16)(p13.1;q22)Acute promyelocytic leukemia with

PML-RARAt(15;17)(q22;q12)

AML with MLLT3-MLL t(9;11)(p22;q23)AML with DEK-NUP214 t(6;9)(p23;q34)AML with RPN1-EVI1 inv(3)(q21;26.2) or t(3;3)(q21;q26.2)AML with RBM15-MKL1 t(1;22)(p13;q13)AML with mutated NPM1AML with mutated CEBPAB-ALL with BCR–ABL1 t(9;22)(q34;q11.2)B-ALL with MLL rearranged t(v;11q23)B-ALL with TEL-AML1 (ETV6-RUNX1) t(12;21)(p13;q22)B-ALL with hyperdiploidyB-ALL with hypodiploidyB-ALL with IL3-IGH t(5;14)(q31;q32)B-lymphoblastic leukemia/lymphoma with

E2A-PBX1 (TCF3-PBX1)t(1;19)(q23;p13.3)

Mixed phenotype acute leukemia withBCR–ABL1

t(9;22)q34;q11.2)

Mixed phenotype acute leukemia with MLLrearranged

A rearrangement involving MLL (11q23) mustbe present; t(4;11) is common

the RARA gene. Recognition of this subtype of AML with its unique pathophysiol-ogy is essential for optimal patient care. Some cases have distinctive morphologicalfeatures (e.g., multiple Auer rods), and a finding of decreased or absent expres-sion of HLADR by flow cytometry analysis is also very suggestive for this subtype.Definitive diagnosis requires demonstration of a translocation involving the RARAgene, however. Cytogenetic confirmation of such can be accomplished rapidly withFISH or molecular studies.

The entity “B-ALL with [9;22](q34;q11.2); BCR–ABL1” is characterized by thepresence of a BCR–ABL1 fusion gene. Tests for this signature abnormality shouldbe capable of detecting the common variant translocations of BCR–ABL1. The p190variant is found in most cases of ALL in childhood, while only about half of adultpatients will have this form, the others having the p210 variant which is typicallyseen in CML. Other “leukemia-associated” cytogenetic abnormalities may occur inassociation with the BCR–ABL1 in these cases, but the BCR–ABL1 abnormality isthe defining abnormality for classification.

A variety of partner genes can participate in translocations with the MLL gene(11q23) in acute leukemia. Demonstration of a translocation involving the MLLgene is a requirement for diagnosis of a case as “B-ALL with MLL rearranged.”Karyotypic analysis may be required to detect these variable chromosomal abnor-malities. Hyperploidy can be detected by karyotyping, by FISH analysis, or by aDNA index calculation from flow cytometric DNA ploidy analysis. When present,


trisomies of chromosome 4, 10, and 17 are particularly significant, and a “triple”combination of all three trisomies is a particularly favorable cytogenetic finding.Eosinophilia may be seen in some cases of B-ALL, and this may be associatedparticularly with a t(5;14)(q31;q32) translocation.

Two additional subcategories of acute leukemia designated “AML with mutatedNPM1” and “AML with mutated CEBPA” are included in the WHO classificationas “provisional” diagnostic categories. These are defined by the presence of muta-tions. The karyotype is usually unremarkable in these cases. The special tests forthe defining mutations may not be widely available.

Acute Leukemias with No Specific Cytogenetic Findings

A substantial number of cases of AML will not have any of the specific geneticabnormalities. Cases of AML with no defining abnormalities are designated “acutemyeloid leukemia, NOS”. In such cases, the following diagnostic categories –defined entirely by morphological criteria – may be used.

AML with minimal differentiationAML without maturationAML with maturationAcute myelomonocytic leukemiaAcute monoblastic and monocytic leukemiaAcute erythroid leukemiaAcute megakaryoblastic leukemiaAcute basophilic leukemiaAcute panmyelosis with myelofibrosis

Testing to establish the absence of any of the specific cytogenetic abnormal-ities is essential when these diagnostic categories are used. They should not beused for cases in which no testing has been performed. Studies for mutations ofNPM1, CEBPA, and FLT3 may be recommended in cases of “cytogenetically nor-mal” AML, but these special tests are not required for classification. For cases inwhich the morphological features suggest the rare entity acute basophilic leukemia,cytogenetic tests to establish the absence of a BCR–ABL1 fusion gene, and also theabsence of t(6;9)(p23;q34), should be undertaken.

Acute myeloid leukemia may arise in patients who have had a well-documentedpreexisting myelodysplastic disorder or in whom one or more of the “MDS-associated” cytogenetic abnormalities is demonstrable. Such cases may be diag-nosed as AML with myelodysplasia-related changes. However, if any of thecytogenetic abnormalities specific for one of the genetically defined subtypes ispresent, the case should be classified within that specific subgroup.

A separate group of entities designated acute leukemias of ambiguous lineageincludes several diagnostically challenging disorders. Phenotypic studies using flow

90 C.C. Whitcomb

cytometry and immunohistochemical staining will demonstrate features of multiplelineages for the blasts in these cases, and designation of any single lineage for classi-fication is equivocal. Various cytogenetic abnormalities may be found in these cases.If abnormalities involving the BCR–ABL1 or the MLL genes are present, the casesmay be classified as either mixed phenotype acute leukemia with t(9;22)q34;q11.2);BCR–ABL1 or mixed phenotype acute leukemia with t(v;11q23); MLL rearranged.

Malignant Lymphomas

Neoplasms of mature lymphoid cells presenting as tissue tumors constitute themalignant lymphomas (non-Hodgkin lymphomas) and Hodgkin’s disease (nowcalled Hodgkin lymphoma). The WHO classification separates the mature B-celland T-cell neoplasms into two large groups, each containing numerous specific enti-ties that are differentiated by their morphological, phenotypic, and clinical features.Cytogenetic studies are not used for subclassification of the tissue tumors as theyare for acute leukemia. However, such studies can be useful in resolving some dif-ferential diagnostic issues. Table 3.3 lists malignant lymphomas in which specificcytogenetic abnormalities can provide useful information. Brief discussions of someof the specific types of lymphoma are given below.

Chronic lymphocytic leukemia (CLL), hairy cell leukemia, and some other“leukemic” disorders are grouped with the tumorous disorders in the WHO classifi-cation. The “leukemic” cells in these disorders are mature B or T cells, just as are the

Table 3.3 Cytogenetic abnormalities in non-Hodgkin lymphomas

WHO category Potentially significant or useful findings

Follicular lymphoma t(14;18)(q32;q21) and BCL2Mantle cell lymphoma t(11;14)(q13;q32)

Rare variants may be negativeExtranodal marginal zone lymphoma of

mucosa-associated lymphoid tissue (MALTlymphoma)

t(11;18)(q21;q21), t(14;18)(q32;q21), ort(3;14)(p14.1;q32)

Nodal marginal zone lymphoma Absence of t(11;18), t(14;18), or t(3;14)Splenic marginal zone lymphoma Absence of t(11;18)Primary cutaneous follicle center lymphoma t(14;18)(q32;q21) and BCL2Diffuse large B-cell lymphoma (DLBCL), NOS Rearrangements of MYC may be seen in 10%Burkitt lymphoma A translocation involving MYC is strongly

recommendedB-cell lymphoma, unclassifiable, with features

intermediate between DLBCL and Burkittlymphoma

A translocation involving MYC should bedemonstrated, but an additionaltranslocation involving BCL2 may bepresent as well

Primary cutaneous DLBCL, leg type Absence of t(14;18)Primary DLBCL of the CNS t(14;18) and t(8;14) are seen only rarelyALK-positive large B-cell lymphoma t(2;5)(p23;q35) or t(2;17)(p23;q23) or

variants


cells of the disorders that present as primary tissue tumors. A distinction between“leukemia” and “lymphoma” can be arbitrary in some instances. Patients with CLLfrequently have lymphadenopathy and splenomegaly, while patients presenting withtissue tumors can have a demonstrable “leukemic” component. Phenotyping of theleukemic cells is usually easily accomplished by flow cytometry for the “leukemic”disorders. No specific cytogenetic studies are required for the diagnosis of these“chronic” leukemias, but a general characterization of ploidy or karyotype may beuseful, particularly in the management of patients with CLL.

Molecular studies are often helpful in the first stages of a diagnostic workup oflymphoid proliferations in tissue. Molecular testing for clonal ARG rearrangementscan be very helpful in distinguishing a neoplastic lymphoid cell proliferation from areactive cellular infiltrate. A clonal IGH rearrangement can be found in virtually allof the B-cell lymphomas, and a clonal rearrangement of a TCR gene (usually TRB,TRG) can be demonstrated in most cases of T-cell lymphoma. The rare NK-celllymphomas do not exhibit rearrangements of the T- and B-lineage antigen receptorgenes. Because of the morphologic similarity between T cells and NK cells, molec-ular testing for TCR rearrangements is essential for distinguishing the neoplasms ofthese two lineages. Once a diagnosis of neoplasia has been established, further clas-sification is based on the phenotype. In some cases, molecular techniques targetingspecific genetic abnormalities may be helpful for resolving issues of classification.

Some Diagnostic Problems

When small lymphocytic cells predominate or are admixed in significant num-bers with larger lymphoid cells, the distinction of a neoplastic proliferation froma reactive / inflammatory lymphoid cell infiltrate can be very difficult. Restrictedexpression of immunoglobulin light chains by B cells can be used as an indicatorof a clonal B-cell population, but demonstration of light chain expression is oftendifficult in sections of fixed and paraffin-embedded tissue. Flow cytometry analysiscan demonstrate immunoglobulin light chain restriction if an adequate aliquot ofunfixed cells is available, but this is frequently not possible with small endoscopicbiopsies or needle core biopsies. In addition, neither morphologic nor flow cytomet-ric techniques can demonstrate the presence of a clonal T-cell population. Becauseof these limitations molecular tests to demonstrate a clonal IGH or TCR rearrange-ment are often useful. Material from paraffin-embedded biopsy tissue can be usedfor these molecular studies if PCR-DNA techniques are used. However, clonalitydoes not by itself imply neoplasia.

A “positive” finding of a clonal ARG rearrangement in a morphologically sus-picious lesion provides strong support for an interpretation of a B-cell lymphoidneoplasm. Contrariwise, a “negative” finding should foster caution in proceedingwith such an interpretation. However, the occurrence of clonal B-cell populationsin lymphoid tissue containing hyperplastic follicles has been well-documented[20, 21], and there are also numerous reports of findings of “monoclonal” T-cell orB-cell populations in biopsies interpreted with due consideration as morphologically

92 C.C. Whitcomb

benign [22–24]. Therefore, a positive molecular test finding should not be inter-preted as prima facie evidence for neoplasia. Correlation of the molecular testfindings with the histomorphologic features in the tissue from which the analyticmaterial was derived is essential.

“False-positive” molecular test findings may reflect physiologic expansions ofnon-neoplastic lymphoid clones, but technical aberrations may also produce find-ings that can be discordant with morphology [25–30]. Interpretation of the analyticdata from PCR procedures can be challenging, particularly in specimens in whichonly a small number of cells are available. In such cases the findings may bereported as indicating “oligoclonal” or “pseudoclonal” populations [31]. Moleculartest reports should include some discussion of the limitations of the analytic datawhen this occurs [32].

Follicular Lymphoma (FL)

Follicular lymphoma is characterized by a t(14;18)(q32;q21) chromosome translo-cation that juxtaposes the BCL2 gene (18q21) and the IGH gene (14q32). Rare casesof FL can have a t(8;14) translocation in addition to t(14;18). The t(14;18) translo-cation is not specific for follicular lymphoma, and it can be seen in other types ofmature B-cell lymphoma. In addition, a “positive” finding for a BCL2 translocationalone cannot be taken as an absolute indicator of clinical neoplasia, for abnormal-ities involving BCL2 have been reported in “normal” individuals [33, 34]. Despitethese limitations, cytogenetic tests for a t(14;18) translocation can be useful in sep-arating follicular lymphoma from atypical follicular hyperplasia, and they can serveas an adjunct to tests for ARG rearrangements in ambiguous cases. Due to variationsin the sites of the chromosomal breakages and recombination, molecular tests maynot be as diagnostically sensitive as FISH analysis.

Mantle Cell Lymphoma (MCL)

This mature B-cell lymphoma is characterized by a t(11;14)(q13;q32) translocationthat juxtaposes the CCND1 gene (11q13) and the IGH (14q32). The transloca-tion results in deregulation of the BCL1 gene and consequent over-expression ofcyclin-D1. Demonstration of cyclin-D1 by immunohistochemical staining is usu-ally employed as the primary criterion for a diagnosis of MCL. When expressionof cyclin-D1 cannot be demonstrated, however, cytogenetic tests for a t(11;14)translocation can be useful [35]. The t(11;14) translocation is specific for man-tle cell lymphoma. In rare cases in which neither cyclin-D1 nor t(11;14) can bedemonstrated – and in which no abnormalities characteristic of other subtypes oflymphoma are present – tests for other cyclins (e.g., cyclin-D2 or cyclin-D3) arerecommended to support a diagnosis of MCL. Tests for t(11;14) may be useful formonitoring patients for residual or recurrent disease as well.


Marginal Zone B-Cell Lymphoma (MZL)

The lymphomas of marginal zone B-cell type occur in extranodal sites as wellas lymph nodes [36]. Extranodal marginal zone lymphoma of mucosa-associatedlymphoid tissue (MALT lymphoma) is a relatively common “low-grade” lymphoidneoplasm of the gastrointestinal tract, and morphologically similar lymphomasoccur in other “mucosa-associated” sites, such as the bronchial tree and ocu-lar adnexa. Translocations involving the BCL2 gene may be seen in MZL, butgenes other than IGH are often associated as “partners” in these chromosomalabnormalities. Different translocations may be seen preferentially in lesions fromdifferent sites. For example, a t(11;18)(q21;q21) translocation is most frequentlyfound in gastric or pulmonary tumors, while a t(14;18)(q32;q21) translocation ismore likely to be found in neoplasms involving ocular adnexa or salivary gland. At(3;14)(p14.1;q32) translocation may occur in tumors of the thyroid, ocular adnexa,or skin. Distinguishing MZL from MCL (mantle cell lymphoma) can be challeng-ing, and demonstration of a BCL2 translocation – in the absence of a translocationinvolving CCND1 – can be helpful in resolving this differential diagnostic problem.

Biopsy specimens from extranodal lesions are frequently very small, and diag-nosis by morphological examination can be very difficult. Tests for clonal ARGrearrangements are often needed to address the initial diagnostic question as towhether the lymphoid cell population is a neoplastic proliferation or simply areactive infiltrate. As discussed previously, a “positive” molecular finding in asmall biopsy specimen must be considered carefully. Correlation with morpho-logical – and imaging findings – is essential to avoid “over-interpretation” [37].Demonstration of a cytogenetic abnormality in these situations may help resolveuncertainty in the significance of a “positive” ARG rearrangement. In addition, cyto-genetic testing for the t(11;18)(q21;q21) translocation can be helpful in the workupof gastric biopsies in patients with Helicobacter pylori infection. The presence ofa t(11;18) gene rearrangement may be predictive of a lack of response to antibiotictherapy for H. pylori infection.

Diffuse Large B-Cell Lymphoma (DLBCL)

Diagnosis of a large-cell lymphoma of B-cell lineage, in any of several morpho-logical variants, is often relatively easy from the histomorphological features andthe cellular phenotype as demonstrated by immunohistochemical staining. Tests forclonal IGH rearrangements will be “positive” in virtually all cases, and in manycases molecular testing is not really needed. Cytogenetic studies, if performed, willfrequently demonstrate translocations involving the BCL6 or the BCL2 genes, butcytogenetic findings are not used as primary criteria for diagnosis or classificationof these neoplasms.

Investigations using array-based techniques to evaluate the expression of multi-ple genes within the tissue cell populations (gene expression profiling – GEP) havedemonstrated that the large B-cell lymphomas can be subdivided into at least two

94 C.C. Whitcomb

groups – one in which the cells exhibit an expression profile similar to that of follicleB cells (germinal center-like B-cell lymphomas) and the other in which the profile ismore like that of activated “non-follicle” B cells (activated B-cell lymphoma) [38].Clinical studies have demonstrated significant differences in the behavior of tumorswith these different gene expression profiles. The technically demanding GEP pro-cedures are not suitable for routine diagnostic use at this time, but results from thesestudies have led to refinements in immunohistochemical profiling of the DLBCL.An algorithm incorporating the results of immunohistochemical stains for CD10,BCL6, and MUM1 has been used to sub-classify cases of DLBCL. An expandeddiscriminator incorporating findings for BCL2, BCL6, LMO2, FN1, CCND2, andSCYA3 has been proposed more recently [39].

Burkitt Lymphoma (BL)

A diagnosis of “Burkitt lymphoma” is usually suggested by the histologic appear-ance of the tumor, but in many cases the morphology is “atypical” and distinctionof BL from DLBCL can be problematic [40]. Cytogenetic tests may provide someguidance in resolving this diagnostic problem. Translocations involving the MYCgene on chromosome 8 with various “partner” genes are characteristic of trueBL. Translocations with IGH, IGK, or IGL as partners – e.g., t(8;14), t(8;22),or t(2;8) respectively – are common variants. Tests to detect these transloca-tions are recommended for confirmation of a diagnosis of BL in both classic andproblematic cases.

The translocations involving MYC are not specific for BL, however, and they maybe found in some cases of DLBCL as well. A finding of a MYC translocation as theonly abnormality provides strong support for classification of a case as BL. In con-trast, findings of other abnormalities (e.g., translocations involving BCL6 or BCL2)along with a MYC translocation would favor a diagnosis of DLBCL. The diagnosticproblem in separating BL from DLBCL is recognized in the WHO classification byinclusion of a separate category designated “ML with features intermediate betweenBL and DLBCL,” which can be used for cases in which the difficulties cannot beresolved.

Cytogenetic tests may be helpful in the diagnosis of some other large-cell B-celllymphomas as well. The t(14;18) translocation is typically not seen in “Primarycutaneous DLBCL, leg type” and it is also quite rare in “Primary DLBCL of theCNS.” Therefore, demonstrating an absence of this abnormality may help indirectlyin properly classifying these particular lymphomas. In addition, the presence of aBCL2 translocation is used as an indicator of poor prognosis in DLBCL, and testingfor this abnormality during and after treatment may assist in monitoring therapeuticresponse and detection of minimal residual disease or recurrent lymphoma.

Plasma Cell Neoplasms (PCN)

Neoplastic proliferations of plasma cells present in several clinicopathologicforms. These neoplasms are distinguished by biochemical characteristics of the


immunoglobulins produced by the neoplastic plasma cells as well as by clinicalfeatures. In most cases a plasma cell neoplasm is easily diagnosed by morphologicstudies. Evidence of clonality is provided by demonstrating restricted expression ofimmunoglobulin light chains in the cytoplasm of the plasma cells. Either immuno-histochemical stains or flow cytometry can be used for this purpose. Molecular testsare seldom needed for demonstration of a clonal population. Chromosomal abnor-malities are commonly found in the PCN disorders, but cytogenetic tests are notneeded for primary diagnosis. Recent investigational studies using gene expressionprofiling have led to a proposal for a subclassification of plasma cell myeloma basedon gene expression patterns in association with the expression of cyclins-D1, -D2,and -D3, but this refinement is not included in the current WHO classification [41].

T-Cell Lymphomas

The neoplasms of mature T cells form a large and diverse group of entities. Asdiscussed previously, tests for clonal TCR rearrangements are very useful in theprimary stages of a workup to establish the neoplastic nature of a T-cell proliferation.The subtypes of the T-cell neoplasms are distinguished primarily on the basis ofmorphological, phenotypic, and clinical features. Molecular and cytogenetic testfindings are not used as primary diagnostic criteria for the recognized subtypes.

The antigen receptor expressed at the cell surface in T cells may be composedof either alpha/beta or gamma/delta chains. Designation of a T-cell tumor as agamma/delta T-cell neoplasm or an alpha/beta T-cell neoplasm is based on theantigen receptor expressed at the cell membrane. Tests for rearrangements of theunderlying TCR genes usually target either the TCG (gamma chain) or TCB (betachain) gene. The gene rearrangement(s) demonstrable by molecular testing maynot necessarily correlate with the antigen receptor actually expressed at the cellmembrane, however, and in some cases molecular testing will demonstrate rear-rangements of both genes. Therefore immunohistochemical staining of the surfacereceptor molecule is the preferred technique for definitive lineage assignment withinthe T-cell neoplasms.

The entity designated “T-cell lymphoma AILD like” is a noteworthy subtypeof the T-cell lymphomas [42–44]. Clonal populations of B cells may develop inthese tumors in addition to the dominant clonal T-cell population. Molecular testingmay demonstrate both T-cell and B-cell clonality in such instances. Clonal T-cellpopulations, presumably arising as a “secondary” phenomenon, have been describedin patients with plasma cell myeloma also [45, 46].

Anaplastic Large-Cell Lymphomas (ALCL)

This morphologically distinctive lymphoid neoplasm is characterized by expres-sion of CD30. The cells in many tumors express protein markers of T-cell lineage,and molecular tests for ARG rearrangements will demonstrate a clonal TCR rear-rangement in many cases. The cells may also express a cytoplasmic protein ALK

96 C.C. Whitcomb

(anaplastic lymphoma kinase), which can serve as an additional useful diagnosticfeature.

Cases of ALCL are phenotypically and genotypically heterogeneous, however,and can present diagnostic difficulties. Some cases may not express markers ofT-cell lineage, but may have a “positive” finding for a clonal TCR gene rearrange-ment. Other cases may have “negative” tests for ARG rearrangements, and notall cases express ALK protein. Expression of ALK in ALCL is often associatedwith a t(2;5)(p23;q35) translocation involving the ALK gene of chromosome 2.Cytogenetic tests demonstrating a translocation involving ALK may be helpful inproblematic cases. Molecular tests for the ALK–NPM fusion gene created by thet(2;5) translocation are available, but PCR-based tests may occasionally be negativein cases of variant translocation.

NK-Cell Lymphomas

An initial impression of a possible T-cell neoplasm often arises in the workup ofa disorder of NK cells because of “positive” immunostaining for CD3. Clinicalfeatures can be very helpful in suggesting the NK-cell nature of the proliferation,however. The CD3 “positivity” in these cases reflects the presence of CD3-epislonwithin the cytoplasm of the NK cells rather than a complete CD3 molecule at thecell membrane – a finding that typifies true T cells. A discordant “negative” resultfor CD3 will be obtained from concurrent flow cytometric analysis that assessesmembrane-associated CD3. Application of a comprehensive staining panel for anti-gens associated with NK cells, including CD56 and cytotoxic enzymes TIA1,granzyme, and perforin, should be used to document the NK-cell lineage of theabnormal cells in these cases. Demonstration of EB viral material within tumorcells is also useful in diagnosing some of the NK-cell entities. Molecular tests todemonstrate an absence of a clonal TCR rearrangement will be confirmatory andshould be performed.

Neoplasms of mature B and T lymphoid cells for which no specific cytogeneticabnormalities or molecular test findings have been found to date are included in theWHO classification. Diagnosis of these entities requires consideration of multiplefeatures including the following:

Histopathologic features:

Plasmablastic lymphomaT-cell/histiocyte-rich large B-cell lymphomaIntravascular large B-cell lymphomaPeripheral T-cell lymphoma, NOS (including angioimmunoblastic T-cell lym-

phoma)Anaplastic large-cell lymphoma (C-ALCL)

Clinical presentation:

Splenic marginal zone lymphomaPrimary mediastinal large B-cell lymphoma


Primary effusion lymphomaEnteropathy-associated T-cell lymphomaHepatosplenic T-cell lymphomaCutaneous T-cell lymphomas (multiple subtypes)Subcutaneous panniculitis-like T-cell lymphoma)

Presence of specific pathogenetic viruses:

Adult T-cell leukemia/lymphoma (HTLV-1 associated)Large B-cell lymphoma arising in HHV8-associated multicentric Castleman

diseaseEBV-positive T-cell lymphoproliferative diseases of childhood

Tests for Viruses Are Useful in Diagnosis of HematolymphoidNeoplasms

Tests for EB virus (EBV) are useful in the diagnosis of several entities withinthe WHO classification [47]. Burkitt lymphoma, DLBCL involving the CNS, pri-mary effusion lymphoma, plasmablastic lymphoma, and some of the NK-cellneoplasms are associated with EB virus. Demonstration of the presence of viralmaterials in the cells of these proliferations can be useful for diagnosis of theseentities. Demonstration of viral material within abnormal cells themselves is thediagnostically most important finding, and procedures that afford correlation withhistopathology are recommended. Immunohistochemical staining for viral proteinsand in situ hybridization procedures are generally used to demonstrate EB virus.Molecular tests targeting episomal EB viral DNA can be used to demonstrate clonalpopulations of infected cells as well, but such highly technical techniques are notreadily available.

Molecular tests (e.g., PCR) for EB viral material within CSF are useful in theworkups of patients with CNS lesions that are suspicious for lymphoma when diag-nostic tissue cannot be easily obtained. It is also reasonable to consider using testsfor EBV in any situation in which immunocompromise is likely. Expression of EBVin Hodgkin lymphoma is also of current interest. Other viruses such as HIV, HTLV1,HHV6, HHV8, and HCV have also been implicated as cofactors in the pathogenesisof some lymphomas, and tests to document the presence of these agents may be use-ful in selected cases as well. Therapy designed to eliminate the viral agent may be arational adjunctive therapeutic approach in those lesions associated with infection.

Hodgkin Lymphoma

The tumors of Hodgkin lymphoma contain only a minor population of abnormalcells within a background of “normal” – presumably reactive – lymphoid cells, his-tiocytes, and granulocytic cells. The abnormal cells (Hodgkin and Reed–Sternbergcells) are indeed clonal lymphoid cells (usually of B-cell lineage), but, because

98 C.C. Whitcomb

of their scarcity, testing of whole (aggregate) tumor tissue for clonal ARG rear-rangements often yields “negative” results. The clonal nature of the abnormal cellshas been demonstrated by micro-dissecting them from tumor tissue and analyzingthe minute quantities of DNA extracted therefrom. Such techniques are impracti-cal for routine use at the present time. A diagnosis of HL is based entirely on thehistomorphological features of the tumors.

Cells with morphological features similar to Hodgkin or Reed–Sternberg cellscan be found in some cases of mature T- or B-cell lymphoma. A “positive” findingfor a clonal B-cell population by molecular testing will not distinguish HL from aB-cell lymphoma with “Hodgkin-like” features, but demonstration of a clonal T-cellpopulation by molecular techniques can be helpful in differentiating HL from a T-cell lymphoma. Although cases of HL of T-cell lineage have been reported, a findingof a clonal T-cell population in conjunction with an “atypical” (i.e., “not classicfor HL”) immunophenotype in the abnormal cells, would strongly suggest that thetumor is a T-cell lymphoma rather than HL. Diagnostic difficulties in separating HLfrom DLBCL are recognized in the WHO classification by inclusion of a categorydesignated “ML with features intermediate between HL and DLBCL” [48–50]. Thiscategory can be used for cases in which the diagnostic difficulties are intractablewith current testing modalities.

Other Hematolymphoid Neoplasms

The WHO monograph includes detailed discussions of clinical and pathologicalfeatures for tumors of mast cells, immature granulocytic cells (myeloid sarcomas),dendritic cells, and histiocytes. Proliferations associated with immunodeficiency arealso discussed. With the exception of the neoplastic proliferations of mast cells, inwhich KIT mutations may be demonstrable, no specific cytogenetic or molecularfindings are associated with these other proliferations, and diagnosis is based onmorphological findings. Neoplasms of mast cells have been grouped with the clonal“myeloid” disorders of the MPN group in the 2008 WHO classification. Moleculartesting for KIT mutation may be helpful for diagnosis in cases where morphologicaland phenotypic findings are equivocal and is important for prognosis since it resultsin relative resistance to tyrosine kinase inhibitor (imatinib) therapy.

Summary

The goal of a diagnostic workup should be a concise and clinically useful reportthat summarizes the information generated from the various procedures used andstates a final diagnostic interpretation using current terminology [51]. The reportshould particularly emphasize those attributes that support the diagnosis. When crit-ical information is unavailable or when discordant findings arise during the workup,concise explanatory statements and recommendations for additional procedures willbe helpful to the clinical user. Such a report will provide clear documentation forcurrent and future management.


The WHO classification includes some diagnostic categories intended for casesthat are “unclassifiable” or which can only be classified in a general manner, e.g.,as “not otherwise specified.” A discussion of the use of these diagnostic categoriescan be found in Chapter 2 of the WHO monograph. Although the particular com-ments in that chapter are directed to the use of the “unclassified” descriptor inconnection with myeloproliferative neoplasms, the discussion can be generalizedfor application to other situations in which an adequate workup fails to provide suf-ficient diagnostic information. To quote from that discussion: “In such cases it isoften preferable to describe the morphological findings, and to suggest additionalclinical and laboratory procedures that are needed to further classify the process.The report should summarize the reason for the difficulty in reaching a more spe-cific diagnosis, and, if possible, specify which [entities] can be excluded fromconsideration” [1].

Diagnostic difficulties do remain for a number of the entities in the current WHOclassification, but it should be anticipated that ongoing studies will generate find-ings that may suggest new approaches to these difficulties [52–54]. In addition,continuing studies, particularly of the cytogenetically “normal” acute leukemiasand of both “large-cell” and “small-cell” lymphomas, may very well lead to therecognition of new entities with clinically distinctive features. Since its original pub-lication in 2001, the WHO classification has undergone continual review by panelsof expert editors and consultants. In its most recent fourth edition, published in2008, significantly increased emphasis was placed on cytogenetic and moleculartesting. As members of the expert groups continue to review the classification, itshould be anticipated that further revisions to diagnostic criteria, almost certainlyincorporating additional findings from cytogenetic and molecular testing, will beforthcoming.

References

1. Swerdlow SH, Campo E, Harris NL, Jatte ES, Pileri SA, Stein H, Thiele J, Vardiman JW, eds.WHO Tumours of Haematopoietic and Lymphoid Tissues. International Agency for researchon Cancer. Lyon, France: IARC; 2008. pp. 439.

2. Jaffe ES, Harris NL, Stein H, Isaacson PG. Classification of lymphoid neoplasms: themicroscope as a tool for disease discovery. Blood. 2008;112:4384–4399.

3. Dunphy CH. Applications of flow cytometry and immunohistochemistry to diagnostichematopathology. Arch Pathol Lab Med. 2004;128:1004–1022.

4. Arber DA. Review: molecular diagnostic approach to non-Hodgkin’s lymphoma. J Mol Diagn.2000;2:178–190.

5. Bench AJ, Erber WN, Scott MA. Molecular genetic analysis of haematological malignancies:I. Acute leukaemias and myeloproliferative disorders. Clin Lab Haem. 2005;27:148–171.

6. Bench AJ, Erber WN, Follows GA, Scott MA. Molecular genetic analysis of haematologicalmalignancies II: mature lymphoid neoplasms. Int J Lab Hem. 2007;29:229–260.

7. Garcia F MD, Swerdlow SH. Best practices in contemporary diagnostic immunohistochem-istry panel approach to hematolymphoid proliferations. Arch Pathol Lab Med. 2009;133:756–765.

8. Wolff DJ, Bagg A, Cooley LD, et al. Guidance for fluorescence in situ hybridization testingin hematologic disorders. J Mol Diagn. 2007;9:134–143.

100 C.C. Whitcomb

9. Ventura RA, Martin-Subero JI, Jones M, et al. Review: FISH analysis for the detection oflymphoma-associated chromosomal abnormalities in routine paraffin-embedded tissue. J MolDiagn. 2006;8:141–151.

10. van Krieken JH, Langerak AW, Macintyre EA, et al. Improved reliability of lymphomadiagnostics via PCR-based clonality testing. Leukemia. 2007;21:201–206.

11. van der Velden VHJ, Hochhaus A, Cazzaniga G, et al. Detection of minimal residual dis-ease in hematologic malignancies by real-time quantitative PCR: principles, approaches, andlaboratory aspects. Leukemia. 2003;17:1013–1034.

12. Bejjani BA, Shaffer LG. Review: application of array-based comparative genomic hybridiza-tion to clinical diagnostics. J Mol Diagn. 2006;8:528–533.

13. Dunphy CH. Gene expression profiling data in lymphoma and leukemia: review of the lit-erature and extrapolation of pertinent clinical applications. Arch Pathol Lab Med. 2006;130:483–520.

14. Vardiman JW, Thiele J, Arber DA, et al. The 2008 revision of the World Health Organization(WHO) classification of myeloid neoplasms and acute leukemia: rationale and importantchanges. Blood. 2009;114:937–951.

15. Hughes T, Deininger M, Hochhaus A, et al. Monitoring CML patients responding to treatmentwith tyrosine kinase inhibitors: review and recommendations for harmonizing current method-ology for detecting BCR-ABL transcripts and kinase domain mutations and for expressingresults. Blood. 2006;108:28–37.

16. Tefferi A, Gilliland G. The JAK2V617F tyrosine kinase mutation in myeloproliferative disor-ders: status report and immediate implications for disease classification and diagnosis. MayoClin Proc. 2005;80:947–958.

17. Lay M, Mariappan R, Gotlib J, et al. Detection of the JAK2 V617F mutation by LightCyclerPCR and Probe Dissociation Analysis. J Mol Diagn. 2006;8:330–334.

18. Arber DA, Stein AS, Carter NH, et al. Prognostic impact of acute myeloid leukemia clas-sification: importance of detection of recurring cytogenetic abnormalities and multilineagedysplasia on survival. Am J Clin Pathol. 2003;119:672–680.

19. van der Reijden BA, Simons A, Luiten E, et al. Minimal residual disease quantificationin patients with acute myeloid leukaemia and inv(16)/CBFB-MYH11 gene fusion. Acutemyeloid leukaemia. Br J Haematol. 2002;118:411–418.

20. Iijima T, Inadome Y, Noguch Mi. Clonal proliferation of b lymphocytes in the germinal cen-ters of human reactive lymph nodes: possibility of overdiagnosis of b cell clonal proliferation.Diagn Mol Pathol. 2000;9:132–136.

21. Nam-Cha SH, San-Millan B, Mollejo M, et al. Light-chain-restricted germinal centres inreactive lymphadenitis: report of eight cases. Histopathology. 2008;52:436–444.

22. Dippel E, Klemke CD, Hummel M, et al. T-cell clonality of undetermined significance. Blood.2001;98:247–248.

23. Rawstron AC, Green MJ, Kuzmicki A, et al. Monoclonal B lymphocytes with the character-istics of ‘indolent’ chronic lymphocytic leukemia are present in 3.5% of adults with normalblood counts. Blood. 2002;100:635–639.

24. Dong L, Masaki Y, Takegami T, et al. Clonality analysis of lymphoproliferative disorders inpatients with Sjogren’s syndrome. Clin Exp Immunol. 2007;150:279–284.

25. Zhou XG, Sandvej K, Gregersen N, et al. Detection of clonal B cells in microdissected reactivelymphoproliferations: possible diagnostic pitfalls in PCR analysis of immunoglobulin heavychain gene rearrangement. J Clin Pathol. 1999;52:104–110.

26. Elenitoba-Johnson KS, Bohling SD, Mitchell RS, et al. PCR analysis of the immunoglobulinheavy chain gene in polyclonal processes can yield pseudoclonal bands as an artifact of lowB cell number. J Mol Diagn. 2000;2:92–96.

27. Langerak AW, Molina TJ, Lavender FL, et al. Polymerase chain reaction-based clonality test-ing in tissue samples with reactive lymphoproliferations: usefulness and pitfalls. A report ofthe BIOMED-2 Concerted Action BMH4-CT98-3936. Leukemia. 2007;21:222–229.


28. Murphy KM, Berg KD, Geiger T, et al. Consultations in molecular diagnostics: capillary elec-trophoresis artifact due to eosin: implications for the interpretation of molecular diagnosticassays. J Mol Diagn. 2005;7:143–148.

29. Kojo S, Elenitoba-Johnson J, Bohling SD, et al. PCR analysis of the immunoglobulin heavychain gene in polyclonal processes can yield pseudoclonal bands as an artifact of low B cellnumber. J Mol Diagn. 2000;2:92–96.

30. Ahrens K, Braylan R, Almasri N, et al. IgH PCR of zinc formalin-fixed, paraffin-embeddednon-lymphomatous gastric samples produces artifactual “Clonal” bands not observed inpaired tissues unexposed to zinc formalin. J Mol Diagn. 2002;4:159–163.

31. Bagg A. Commentary: immunoglobulin and t-cell receptor gene rearrangements: mindingyour B’s and T’s in assessing lineage and clonality in neoplastic lymphoproliferative disorders.J Mol Diagn. 2006;8:426–429.

32. Gulley ML, Braziel RM, Halling KC, et al. Clinical laboratory reports in molecular pathology.Arch Pathol Lab Med. 2007;131:852–863.

33. Limpens J, Stad R, Vos C, et al. Lymphoma-associated translocation t(14;18) in blood B cellsof normal individuals. Blood. 1995;85:2528–2536.

34. Summers KE, Goff LK, Wilson AG, et al. Frequency of the Bcl-2/IGH rearrangement innormal individuals: implications for the monitoring of disease in patients with follicularlymphoma. J Clin Oncol. 2001;19:420–424.

35. Wohlschlaeger C, Lange K, Merz H, et al. Aberrant immunophenotypes of mantle celllymphomas. Leuk Lymphoma. 2003;44:269–273.

36. Rao DS, Said JW. Small lymphoid proliferations in extranodal locations. Arch Pathol LabMed. 2007;131:383–396.

37. Hummel M, Oeschger S, Barth TF, et al. Wotherspoon criteria combined with B cell clonal-ity analysis by advanced polymerase chain reaction technology discriminates covert gastricmarginal zone lymphoma from chronic gastritis. Gut. 2006;55:782–787.

38. Alizadeh AA, Eisen MB, Davis RE, et al. Distinct types of diffuse large B-cell lymphomaidentified by gene expression profiling. Nature. 2000;403:503–511.

39. Lossos IS, Czerwinski DK, Alizadeh AA, et al. Prediction of survival in diffuse large-B-celllymphoma based on the expression of six genes. N Engl J Med. 2004;350:1828–1837.

40. Haralambieva E, Boerma E-J, van Imhoff GW, et al. Clinical, immunophenotypic, and geneticanalysis of adult lymphomas with morphologic features of Burkitt lymphoma. Am J SurgPathol. 2005;29:1086–1094.

41. Kuehl WM, Bergsagel PL. Early genetic events provide the basis for a clinical classificationof multiple myeloma. Am Soc Hematology. 2005;346–352.

42. Zettl A, Lee SS, Rudiger T, et al. Epstein-Barr virus-associated B-cell lymphoproliferative dis-orders in angioimmunoblastic T-cell lymphoma and peripheral T-cell lymphoma, unspecified.Am J Clin Pathol. 2002;117:368–379.

43. Luzzatto F, Pruneri G, Benini E, et al. Angioimmunoblastic T-cell lymphoma with hyperplas-tic germinal centres and a high content of EBV-infected large B-cells carrying IgH chain genemonoclonal rearrangement. Histopathology. 2005;46:464–466.

44. Tan BT, Warnke RA, Arber DA. The frequency of B and T cell gene rearrangements andEBV in T-cell lymphomas: a comparison between angioimmunoblastic T-cell lymphoma andperipheral T-cell lymphoma, unspecified, with and without associated B cell proliferations. JMol Diagn. 2006;8:466–475.

45. Lim SH, Badros A, Lue C, et al. Distinct T-cell clonal expansion in the vicinity of tumor cellsin plasmacytoma. Cancer. 2001;91:900–8.

46. Sze DM. Clonality detection of expanded T-cell populations in patients with multiplemyeloma. Methods Mol Med. 2005;113:257–267.

47. Gulley ML, Tang W. Review: laboratory assays for Epstein-Barr virus-related disease. J MolDiagn. 2008;10:279–292.

48. Poppema S, Kluiver JL, Atayar C, et al. Report: workshop on mediastinal grey zonelymphoma. Eur J Haematol. 2005;75(Suppl 66):45–52.

102 C.C. Whitcomb

49. Stein H, Jöhrens K, Anagnostopoulos I. Non-mediastinal grey zone lymphomas and reportfrom the workshop. Eur J Haematol. 2005;75(Suppl 66):42–44.

50. Traverse-Glehen A, Pittaluga S, Gaulard P, et al. Mediastinal gray zone lymphoma: the miss-ing link between classic Hodgkin’s lymphoma and mediastinal large b-cell lymphoma. Am JSurg Pathol. 2005;29:1411–1421.

51. Ogino S, Gulley ML, den Dunnen JT, et al. Standard mutation nomenclature in moleculardiagnostics practical and educational challenges. J Mol Diagn. 2007;9:1–6.

52. Savage KJ, Monti S, Kutok JL, et al. The molecular signature of mediastinal large B-celllymphoma differs from that of other diffuse large B-cell lymphomas and shares features withclassical Hodgkin lymphoma. Blood. 2003;102:3871–3879.

53. Dave SS, Fu K, Wright GW, et al. Molecular diagnosis of Burkitt’s lymphoma. N Engl J Med.2006;354:2431–2442.

54. Hummel M, Bentink S, Berger H, et al. A Biologic definition of Burkitt’s lymphoma fromtranscriptional and genomic profiling. N Engl J Med. 2006;354:2419–2430.

Chapter 4Update on the Molecular Pathologyof Precursor Lymphoid Leukemias

Robert B. Lorsbach

Keywords Acute lymphoblastic leukemia · Lymphoblastic lymphoma · PrecursorB-lymphoblastic leukemia · Precursor T-lymphoblastic leukemia · Precursor lym-phoid malignancy · World Health Organization · BCR–ABL1 · Philadelphiachromosome · IKAROS · IKZF1 · Micro-RNA · miR-230 · Chronic myel-ogenous leukemia · Down syndrome · GATA1 · Acute megakaryoblasticleukemia · JAK2 · JAK2 R683 mutation · Janus kinase · CRLF2 · IL-7 alphareceptor · Leukemia predisposition · ARID5B · Cancer stem cell · Leukemiastem cell · NOD/SCID mouse · NOTCH1 · Delta-like ligand · Gamma-secretase · Gamma-secretase inhibitor · FBXW7 · PTEN (phosphatase and tensinanalog) · PI3K (phosphatidylinositol 3-kinase) · HES1 (hairy and enhancer-of-splitanalog 1) · AKT · mTOR · Corticosteroids · Flow cytometry · Early T-cell precursor

Introduction

The precursor lymphoid malignancies are a group of neoplasms derived fromeither hematopoietic stem cells (HSCs) or committed lymphoid progenitor cells,depending upon the underlying genetic lesion. Because the malignant cells inthese tumors share many of the genetic (e.g., ongoing rearrangement of antigenreceptor genes), and immunophenotypic (e.g., expression of progenitor markerssuch as CD34 and terminal deoxynucleotidyl transferase) properties of normallymphoid progenitors, they are categorized under the World Health Organization(WHO) classification as “precursor lymphoid malignancies” [1]. These neoplasmsinclude precursor B-lymphoblastic leukemia (B-ALL), precursor B-lymphoblasticlymphoma (B-LBL), precursor T-lymphoblastic leukemia (T-ALL), and precursorT-lymphoblastic lymphoma (T-LBL). B-ALL is significantly more common thanT-lineage disease, with the latter comprising only 10–15% and 25% of ALL in chil-dren and adults, respectively. Although they may develop at any age, the precursor

R.B. Lorsbach (B)Department of Pathology, University of Arkansas for Medical Sciences, Little Rock,AR 72205, USAe-mail: [email protected]


104 R.B. Lorsbach

lymphoid malignancies arise most frequently in children, in whom ALL is the mostcommon malignancy. The lineage of the malignant lymphoblasts, most commonlydetermined by flow cytometry-based immunophenotyping, greatly influences thelikelihood of lymphomatous versus leukemic presentation [2]. With B-lineage dis-ease, a leukemic presentation is much more common, whereas T-lymphoblasticmalignancies manifest more frequently as a lymphoma most often an anteriormediastinal thymic mass.

Historically, the precursor lymphoid malignancies have been classified accord-ing to their morphologic, cytochemical, or immunophenotypic properties. However,a wide array of cytogenetic and molecular lesions have been detected in these malig-nancies, which in many instances have been shown to have a profound impact ontheir biologic behavior and ultimately on clinical outcome. Thus, these underlyinggenetic lesions are now acknowledged as critical determinants in the classifica-tion and prognostication of the precursor lymphoid neoplasms, particularly forB-ALL/LBL. This is reflected in the recently published WHO classification, whereseveral cytogenetically or molecularly defined subtypes of precursor ALL/LBL arenow recognized as diagnostic entities. In this chapter, we will briefly review theclassification of the precursor lymphoid malignancies and discuss briefly some gen-eral aspects of their pathogenesis. However, the main focus of discussion will be onrecent advances in our understanding of the molecular pathogenesis of the precursorlymphoid malignancies.

WHO Classification of Precursor Lymphoid Malignancies

The classification of the precursor lymphoid malignancies has been revised andincluded as part of the 2008 WHO classification of hematopoietic and lymphoidmalignancies (Table 4.1) [1]. An important feature of this revised classification is itsdelineation as diagnostic entities of several cytogenetically or molecularly definedsubtypes of ALL/LBL. In particular, the WHO classification includes several sub-types of B-ALL/LBL, each containing a critical genetic lesion (either a translocationor a numerical chromosomal abnormality) that has been extensively characterized

Table 4.1 2008 World health organization classification of precursor lymphoid neoplasms

B-lymphoblastic leukemia/lymphoma, not otherwise specifiedB-lymphoblastic leukemia/lymphoma with recurrent genetic abnormalities

B-lymphoblastic leukemia/lymphoma with t(9;22)(q34;q11.2); BCR–ABL1B-lymphoblastic leukemia/lymphoma with t(v;11q23); MLL rearrangedB-lymphoblastic leukemia/lymphoma with t(12;21)(p13;q22); ETV6-RUNX1B-lymphoblastic leukemia/lymphoma with hyperdiploidyB-lymphoblastic leukemia/lymphoma with hypodiploidyB-lymphoblastic leukemia/lymphoma with t(5;14)(q31;q32); IL3-IGHB-lymphoblastic leukemia/lymphoma with t(1;19)(q23;p13.3); TCF3-PBX1

T-lymphoblastic leukemia/lymphoma

4 Update on the Molecular Pathology of Precursor Lymphoid Leukemias 105

and shown to delineate a distinct tumor type. Particularly in the pediatric population,these subtypes of B-ALL/LBL are often associated with distinctive clinical behav-ior and are of prognostic importance. No subtypes of T-ALL/LBL are recognizedfor diagnostic purposes under WHO criteria. This is due, in part, to the fact thatalthough several genetic lesions have been identified in T-ALL/LBL, their prog-nostic importance is not as well defined as the genetic abnormalities in B-lineagetumors. However, the recent identification in T-ALL/LBL of genetic abnormalitieswhich represent potential therapeutic targets will likely prompt the recognition ofthese abnormalities in future WHO classification schemes. A practical and obviousimplication of the 2008 WHO classification for the routine practice of diagnostichematopathology is the requirement for cytogenetic or molecular characterizationfor final diagnosis of a precursor lymphoid neoplasm. The ongoing application ofgenome-wide, high-throughput genetic techniques to the analysis of the precursorALL/LBL has revealed several novel genetic abnormalities in these malignancies,a subset of which appears to significantly impact on their clinical behavior. Thus,the classification of these malignancies will undoubtedly be further refined to betterreflect advances in our understanding of their underlying molecular pathogenesis aswell as to better define more homogenous leukemia subtypes that might be amenableto molecularly targeted therapeutics.

Overview of Cytogenetic and Molecular Lesions in PrecursorLymphoblastic Malignancies

Numerous genetic lesions in ALL/LBL have been identified and extensively char-acterized, a comprehensive discussion of which is beyond the scope of thischapter. Therefore, in this section only general aspects of leukemogenesis will beaddressed before proceeding to detailed discussion of several select topics on ALLpathogenesis. For a more general discussion of the molecular genetics of the pre-cursor lymphoid leukemias/lymphomas, the reader is referred to several excellentreviews [3–6].

The precursor lymphoid malignancies include two major types in which thereis immunophenotypic and genetic evidence of either B- or T-lineage lymphoid dif-ferentiation. The major genetic lesions found in B-ALL versus T-ALL are almostmutually exclusive. Rarely, one of these genetic abnormalities may be detectedin both B-ALL and T-ALL, for example, the exceptional case of T-ALL express-ing BCR–ABL1, a fusion almost invariably associated with B-ALL. This tightassociation between genetic lesion and disease type reflects the lineage-restrictedexpression and/or function of many of the gene products targeted by these geneticabnormalities. As alluded to above, the major cytogenetic/molecular subtypes ofB-ALL have been incorporated into the current WHO classification (Table 4.1);these prominently include several chromosomal translocations and also subtypes inwhich there are numerical chromosomal abnormalities. The principal genetic lesionsdetected in T-ALL are indicated in Table 4.2; similar to B-ALL, these include not

106 R.B. Lorsbach

Table 4.2 Genetic lesions in T-ALL

Genetic lesion Frequency Target gene(s)

Translocations targeting TCRgenes at 7q34 (TCRB &TCRG) and 14q11 (TCRA& TCRD)

35% TAL1, TAL2, TLX1 (HOX11),TLX3 (HOX11L2), LYL1,LMO1, LMO2, LCK,NOTCH1, OLIG2, CCND2

Aberrant expression due to1p32 rearrangements

20% STIL-TAL1

Fusion oncogene formation 10% PICALM-MLLT10(CALM-AF10)

∼5% MLL fusions∼5% ABL1 fusionsRare NUP98 fusions

Deletion of 9p21 and 6q 65%, del(9p) P15, P1620–30%, del(6q) Unidentified loci

Activating gene mutations 50–60% NOTCH110–15% pediatric T-ALLs FLT3∼10% NRAS, KRAS

only translocations but also other chromosomal rearrangements, as well as pointmutations.

The majority of these lesions target genes encoding transcription factors that arecritically involved in normal B- and T-cell lymphopoiesis. In some instances, a chro-mosomal translocation targeting a transcription factor gene results in formation ofa fusion oncoprotein that aberrantly activates or represses the expression of down-stream transcriptional targets (e.g., ETV6-RUNX1), whereas other chromosomalrearrangements result in overexpression of a transcription factor with perturbationof downstream gene expression (e.g., rearrangements targeting the HOX genes inT-ALL). In contrast to AML and other myeloid disorders, few genetic lesions inALL directly target tyrosine kinases, most notably BCR–ABL1 a constitutively acti-vated tyrosine kinase resulting from the t(9;22). Finally, recent investigation hasidentified several genetic lesions important in the pathogenesis of ALL which maybe detected in several molecular subtypes (Table 4.2). The most notable of these areactivating mutations in NOTCH1 which are detected in more than half of T-ALLcases. Given this, some investigators have proposed defining as a “type A mutation”those genetic lesions which define specific ALL subtypes (e.g., the TCF3–PBX1fusion oncogene resulting from the t(1;19) detected in a subset of B-ALL) andusing “type B mutation” to denote those abnormalities which may be detected intwo or more ALL subtypes (e.g., the aforementioned NOTCH1 mutations) [4]. Ingeneral, these genetic aberrations disrupt the normal ontogenetic program of thelymphoid progenitors in which they occur. This is achieved through the perturbationof several key cell signaling pathways in lymphoid progenitors, inducing variabledegrees of dysregulated proliferation and cell cycle regulation, increased resis-tance to apoptosis, blocked differentiation, and ultimately the development of acuteleukemia.


B-lymphoblastic Leukemia with t(9;22)(q34;q11.2)

The t(9;22)(q34;q11.2), or Philadelphia (Ph) chromosome, was the first describedrecurrent genetic lesion in a human malignancy and is by definition present inchronic myelogenous leukemia (CML) [7, 8]. This translocation targets the BCRand ABL1 genes located on chromosomes 22 and 9, respectively, and results inthe expression of a BCR–ABL1 fusion mRNA transcript and ultimately the BCR–ABL1 fusion oncoprotein [9, 10]. While the chromosomal breakpoint at 9q34 occurswithin a relatively restricted region of the ABL1 gene, within the intron precedingexon 2 (a2), there is much greater variability in the BCR breakpoints. In virtuallyall cases of CML and a minority of Ph+ adult and pediatric B-ALLs, the breakpointin BCR occurs within the so-called major breakpoint cluster region (M-bcr) aftereither exon 13 (e13 or b2) or exon 14 (e14 or b3), resulting in the expression of ap210 fusion protein. However, in most cases of B-ALL, the minor breakpoint clus-ter region (m-bcr), located between exons 1 and 2 of BCR, is utilized. Because itcontains significantly less BCR coding sequence, utilization of the m-bcr results inexpression of the smaller p190 isoform of BCR–ABL1. Although not expressed inB-ALL, a p230 isoform of BCR–ABL1 is expressed in rare cases of CML in whichthe BCR breakpoint occurs within intron 19 [11–13]. The molecular genetics of theBCR–ABL1 rearrangements are detailed in several recent reviews [14, 15] and inChapter 7.

The leukemogenic effects of BCR–ABL1 are largely attributable to dysregu-lated protein tyrosine kinase activity, recently reviewed by Goldman and Melo [14].Homodimerization of BCR–ABL1 is effected through motifs derived from BCR.This results in constitutive protein tyrosine kinase activation due to phosphorylationof critical tyrosine residues within the activation loop of the ABL-derived kinasedomain. The dysregulated signaling induced by BCR–ABL1 is effected throughseveral critical signaling pathways, including RAS, signal transducers and acti-vators of transcription (STATs), phosphatidylinositol 3-kinase, mitogen-activatedprotein (MAP) kinases, and MYC. Thus, BCR–ABL1 signaling within a leukemiccell results in growth factor-independent proliferation, resistance to pro-apoptoticstimuli, and reduced cell adhesion.

In addition to its pathogenetic role in CML, the t(9;22) with expression ofBCR–ABL1 defines a subset of de novo B-ALL, so-called Ph+ B-ALL, account-ing for 25% and approximately 3% of B-ALL in adults and children, respectively.Clinically, Ph+ B-ALL is an aggressive disease, and even in children it is associ-ated with lower rates of complete remission and significantly lower rates of overallsurvival than most other molecular/cytogenetic subtypes of ALL [16, 17], althoughclinical outcome may be improved by addition of imatinib [18]. The mechanismsresponsible for the poor outcome of Ph+ B-ALL are poorly understood; however,a better understanding of the genetic lesions that cooperate with BCR–ABL1 maypermit the development of more effective, targeted therapies which would have asignificant impact on clinical outcome, particularly in adult B-ALL.

A major advance in our understanding of the pathogenesis of Ph+ B-ALL camewith the recent demonstration that most Ph+ ALLs harbor deletions in the IKZF1

108 R.B. Lorsbach

Fig. 4.1 Genomic organization of the IKZF1 gene. IKZF1 encodes the B-cell transcription factorIKAROS. Exons are indicated by blue boxes; the first exon is noncoding. Those segments encodingzinc finger motifs within the N-terminal portion and the C-terminus are indicated by gray bars.As denoted, the N-terminal zinc fingers mediate DNA binding, whereas those located within theC-terminus are responsible for IKAROS dimerization

gene, which encodes the highly conserved transcription factor IKAROS [19–21].IKAROS is expressed in hematopoietic progenitors and mature lymphoid cells.Mice deficient in Ikaros have major defects in lymphoid development as mani-fested by an absence of mature B cells and their progenitors, a marked deficiencyof mature T cells and their progenitors, and an absence of NK cells [22]. Like mosttranscription factors, IKAROS is a modular protein (Fig. 4.1). Multiple zinc fin-ger domains are located in the amino one-half of IKAROS and are responsible forDNA-binding and nuclear localization. Near the carboxy terminus, additional zincfinger motifs are present which mediate protein–protein interaction, including thehomodimerization of IKAROS. In addition, this C-terminal domain mediates theinteraction of IKAROS with other proteins, such as the histone deacetylases, thatmediate chromatin remodeling.

Using high-density single-nucleotide polymorphism (SNP) array analysis withwhole-genome coverage, 60–80% of Ph+ B-ALLs from adult and pediatric patientsIKZF1 were found to harbor IKZF1 mutations, nearly all of which were heterozy-gous [20, 21, 23]. The most common of these were deletions encompassing exons3–6 (�3–6). The resulting truncated transcript is predicted to encode an IKAROSisoform lacking the DNA-binding domain; this �3–6 mutant manifests aberrantsubcellular localization, being localized in the cytoplasm [20]. Furthermore, the�3–6 mutant forms non-functional heterodimers with wild-type IKAROS and thusfunctions in a dominant-negative manner [24]. Less commonly, deletions of exons1–6 were identified, deleting all but the last coding exon; other less commondeletions of IKZF1 were also detected. Interestingly, earlier studies had identifiedmultiple different IKZF1 transcripts in normal B cells as well as Ph+ B-ALL andin both instances were believed to be generated solely through alternative splicing[25, 26]. These more recent findings indicate that the variant transcripts in Ph+ B-ALL are due, in part, to intragenic deletions in IKZF1. Importantly, these mutationsin IKZF1 are not observed in chronic-phase CML but are detected commensuratewith transformation of CML to lymphoid blast crisis but not myeloid blast crisis,suggesting that the leukemogenic potential of these IKZF1 mutations is contextspecific for B-lymphoblastic malignancy [20, 23].

Recent studies provide insight into the basis for this association betweenBCR–ABL1 signaling and altered IKAROS function. Following ligation of the pre-B-cell receptor complex, signaling mediated by phosphorylated STAT5 and the SRCfamily of tyrosine kinases induces cell proliferation through the activation of NF-κB


and ultimately upregulation of MYC and CCND2 [27]. This pro-proliferative sig-naling of the pre-B-cell receptor complex is counteracted by signaling through thelinker proteins SYK and SLP65 which inhibits STAT5 and upregulates expressionof IKAROS, subsequently inducing cell cycle arrest through IKAROS-dependentexpression of p27 [28–30]. Elegant studies from Muschen and colleagues indicatethat in the presence of IKAROS, BCR–ABL1 tyrosine kinase activity is divertedaway from SRC signaling and toward that of SYK and SLP65, which wouldresult in net growth inhibition [31]. Indeed, expression of wild-type IKAROS inBCR–ABL1+ B lymphoblasts induces cell cycle arrest, even though these cells donot express a functional pre-B-cell receptor complex. By contrast, in the absence ofIKAROS, BCR–ABL1 downstream signaling occurs mainly through SRC-mediatedpathways resulting in dysregulated proliferation. Thus, IKAROS appears to func-tion as a tumor suppressor in the specific context of B-cell progenitors expressingBCR–ABL. The mechanism by which IKAROS “redirects” the downstream sig-naling effected by BCR–ABL1 remains to be determined. Nevertheless, theseobservations account for the high incidence of IKAROS mutations in BCR–ABL1+

B-ALL.Although not detected in AML or in most subtypes of ALL, IKZF1 muta-

tions have also been detected in a subset of clinically aggressive ALL lackingBCR–ABL1 expression [32, 33]. Interestingly, these cases have gene expressionprofiles that are quite similar to that of Ph+ B-ALL. These observations suggestthat BCR–ABL– B-ALLs with IKZF1 mutations may harbor occult mutations in anas yet unidentified tyrosine kinase. Indeed, activating mutations in the Janus fam-ily kinases have been described in a subset of high-risk B-ALLs, 70% of whichalso harbor IKZF1 mutations [34]. Finally, these findings indicate that the tumorsuppressor function of IKAROS is not strictly limited to Ph+ ALL.

Micro-RNAs (miRNAs) are short, noncoding RNAs that silence target geneexpression through direct degradation of mRNA transcripts or indirectly throughrepression of translation. miRNAs regulate several important biologic processes,including cell proliferation, differentiation, and apoptosis [35]. Recent studiesdemonstrate that expression of BCR–ABL1, as well as that of native ABL1, is subjectto miRNA-mediated epigenetic regulation [36]. Comparative genomic hybridizationanalysis of mouse leukemias expressing BCR–ABL1 identified loss of heterozy-gosity on chromosome 12, a region of the mouse genome which is particularlyenriched in genes encoding miRNAs. Further analysis of this region revealed thatthe promoter of one gene, miR-230, had undergone extensive CpG hypermethyla-tion. The syntenic region of the human genome containing miR-230 is located onchromosome 14q32, a region that is deleted with progression to blast phase in CML[37]. Importantly, such epigenetic silencing of miR-230 was also detected in pri-mary human Ph+ leukemic cells, and it was specific for leukemic cells expressingBCR–ABL1, as it was not observed in BCR–ABL– cells. Reexpression of miR-230 markedly inhibited proliferation of BCR–ABL1+ cell lines. These results haveseveral important implications. First, they indicate that miR-230 is an importanttumor suppressor in BCR–ABL1+ leukemias. Perhaps more importantly, they sug-gest that restoration of miR-230 expression in BCR–ABL+ leukemic cells may be

110 R.B. Lorsbach

therapeutically useful, an approach that could be achieved through the adminis-tration of demethylating drugs such as decitabine. While certainly preliminary,manipulation of miR-230 expression may represent an important new avenue fortherapeutic intervention in BCR–ABL1+ leukemias.

Down Syndrome-Associated ALL

Children with Down syndrome (DS) have a markedly increased risk of developingacute leukemia, including both AML and ALL, with a 20-fold higher risk for ALLthan the general pediatric population [38]. A significant advance in our understand-ing of DS-associated acute leukemia resulted from the demonstration that mutationsin GATA1, which encodes a megakaryocyte- and erythroid-specific transcription fac-tor, are present in nearly all cases of transient myeloproliferative disorder and acutemegakaryoblastic leukemia (AMKL) occurring in the setting of DS (Fig. 4.2). Thesemutations result in the expression of a truncated form of GATA1 which is believedto result in dysregulated proliferation of megakaryocytic progenitors, which may inthe case of AMKL progress to overt acute leukemia [39].

Cytogenetic analyses have demonstrated distinct differences between DS-associated ALL (DS-ALL) and ALL occurring in children without DS [40]. Forexample, ALL with the t(9;22) or 11q23 rearrangements are distinctly uncom-mon in DS-ALL, compared to non-DS-ALL. By contrast, trisomy of chromosomeX, del(9p), and the t(8;14)(q11;q32) are significantly more common in DS-ALL.While these cytogenetic findings suggest differences in the pathogenesis of DS-ALL, the pathogenesis of ALL in children with DS has remained poorly understooduntil recently.

The Janus kinases are protein tyrosine kinases and include three family members,JAK1, JAK2, and JAK3 [41]. The demonstration that mutations in JAK2, primarilythe V617F mutation, are detected at a high frequency in several myeloproliferativeneoplasms, including polycythemia vera, essential thrombocythemia, and primarymyelofibrosis, has spawned considerable investigation into the role of altered JAKsignaling in leukemogenesis. The Janus kinases physically associate with the cyto-plasmic tails of several cytokine receptors, including those for erythropoietin, IL-3,and GM-CSF, and effect signal transduction upon engagement of the receptor byits cognate ligand through the phosphorylation of critical downstream signalingmolecules, including the signal transducer and activator of transcription (STAT)proteins, mitogen-activated protein kinases (MAPK), and phosphatidylinositol-3-kinase (PI3K) signaling pathways. In contrast to wild-type JAK2, which is activatedonly upon ligand binding, the V617F mutation endows JAK2 with constitutivesignaling activity [42].

Although there were scattered isolated reports describing mutations or translo-cations targeting JAK2, the first definitive evidence for a broader pathogenetic roleof perturbed JAK signaling in DS-ALL came from the identification of JAK2 muta-tions in approximately 20% of analyzed cases of DS-ALL, nearly all of which were


Fig. 4.2 Pathogenesis of Down syndrome-associated acute leukemia. DS is associated with sig-nificantly increased risk for the development of acute leukemia including both AML, particularlyAMKL, and ALL. A critical early genetic lesion in the evolution of DS-associated megakary-ocytic disorders is acquisition of mutations in GATA1 (top portion of figure), which is thought toresult in dysregulated proliferation in megakaryocyte lineage-committed hematopoietic progeni-tors. GATA1 mutations are detected in nearly all TMDs; progression to overt AMKL presumablyreflects the acquisition of additional cooperating mutations (indicated by an asterisk), the iden-tity of which are largely unknown. Recent studies implicate aberrant overexpression of CRLF2in the pathogenesis of DS-associated ALL (bottom portion of figure). Together with IL7-Rα,CRLF2 forms a high-affinity receptor for the cytokine thymic stromal lymphopoietin (TSLP). Themechanism by which CRLF2 overexpression contributes to leukemogenesis is presently unknown.However, acquisition of mutations in tyrosine kinases, such as the JAK R683 mutant which pos-sesses constitutive kinase activity, appears to cooperate with CRLF2 to induce ALL. In bothDS-associated ALL and AMKL, the identity of those genes on chromosome 21 which cooperatewith these mutations is currently unknown

heterozygous [43]. In contrast to the V617F mutation characteristic of myelopro-liferative neoplasms, the JAK2 mutations identified in DS-ALL primarily targeteda conserved arginine residue at position 683 (R683). Subsequent analyses by otherinvestigators have similarly detected JAK2 R683 mutations in 18–28% of DS-ALLcases [44, 45]. Similar to the V617F, transduction of JAK2 R683 mutants intocytokine-dependent cell lines conferred cytokine-independent growth and activateddownstream signaling pathways [43, 44]. Like V617, the R683 residue is locatedwithin the pseudokinase domain of JAK2. Molecular modeling predicts that R683 islocated within the very highly conserved binding pocket of the JAK2 pseudokinase

112 R.B. Lorsbach

domain [43]; thus, this mutation may alter the substrate binding of JAK2. The factthat the mutations targeting V617 and R683 are thought to be on different surfaces ofthe pseudokinase domain may account for the distinct disease associations observedwith each of these JAK2 mutations. The JAK2 R683 mutation does not appear tohave any significant prognostic impact [43, 45].

Deregulated expression of cytokine receptor-like factor 2, or CRLF2, has veryrecently been implicated in the pathogenesis of DS-ALL [46, 47]. CRLF2 isexpressed by early B- and T-cell progenitors, mast cells, and dendritic cells [48].Together with the IL-7 receptor α chain, CRLF2 forms a high-affinity, heterodimericreceptor for thymic stromal-derived lymphopoietin (TSLP), a cytokine that shareshomology with IL-7 [49, 50]. As suggested by its name, TSLP is primarilyexpressed by epithelial cells, keratinocytes, and stromal cells [51, 52]. It potentlyactivates dendritic cells and plays a critical role in the regulation of inflammatoryand allergic responses, although it appears to be dispensable for B-cell development[53–58].

The CRLF2 gene resides within a pseudoautosomal region, PAR1, of the X and Ysex chromosomes [59]. Dysregulation of CRLF2 expression has been demonstratedin both pediatric and adult B-ALL. This may occur secondary to a chromosomaltranslocation involving 14q32, either the t(X;14)(p22;q32) or the t(Y;14)(p11;q32),bringing CRLF2 into close juxtaposition with the immunoglobulin heavy chainenhancer, resulting in its overexpression [46]. Alternatively, an interstitial dele-tion within PAR1, either del(X)(p22.33p22.33) or del(Y)(p11.32p11.32), juxtaposesthe P2RY8 and CRLF2 genes, yielding an in-frame fusion of a noncoding exon ofP2RY8 to the entire coding region of CRLF2, resulting in dysregulated expression ofCRLF2 driven by P2RY8 promoter elements [47]. The PAR1 deletion appears to bemore common in pediatric ALL, whereas both genomic rearrangements are detectedwith comparable frequency in adult ALL. Notably, the deletion is frequently presentin 42–53% of DS with ALL, in contrast to typical pediatric ALL in which a PAR1deletion is detected in only 2–3% of cases [47]. Furthermore, mutations in JAK2are present in approximately 45% of cases with alterations affecting CRLF2. Theseinclude primarily the R683 mutations discussed above, but also JAK2 kinase domainmutations and rarely JAK1 pseudokinase domain mutations. In vitro analyses con-firmed constitutive JAK2 signaling in cells harboring these mutations. Importantly,cell lines overexpressing CRLF2 together with a constitutively active JAK2 R683mutant manifest cytokine-independent growth in vitro. While the underlying leuke-mogenic mechanism is as yet unknown, it is likely that CRLF2 overexpression playsan important role in DS-ALL pathogenesis, given that these cases generally lackthe genetic lesions characteristic of typical pediatric ALL, including the t(12;21),t(1;19), t(9;22), and translocations targeting MLL as well as the typical numericalchromosomal abnormalities, e.g., high hyperdiploidy, hypodiploidy [47].

Although not restricted to DS-ALL, the detection of genomic alterations result-ing in dysregulated CRLF2 expression in nearly half of DS-ALL cases suggeststhat the oncogenic potential of CRLF2 is enhanced or influenced by a coexistingtrisomy 21. Interestingly, in both DS-associated AMKL and ALL, the identifiedrecurrent genetic lesions target genes (GATA1 and CRLF2, respectively) do notreside on chromosome 21. This suggests that in the context of DS, leukemogenesis


requires cooperation between the aberrant signaling induced by these mutations andthe altered expression of an as yet unidentified gene residing in the DS critical regionwhich has been mapped to chromosome 21q22 [60, 61].

Genetic Factors Predisposing to Precursor LymphoblasticLeukemia/Lymphoma

While several genetic disorders are well known to predispose to lymphoblasticleukemia, including Down syndrome, Bloom syndrome, neurofibromatosis, andataxia telangiectasia [62–66], patients with these disorders collectively accountfor fewer than 5% of cases of ALL. Consequently, the existence and identity ofpredisposing genetic factors in the preponderance of patients with ALL remainslargely undefined. The analysis of twins to gain insight into whether there existsa genetic propensity for the development of ALL is complicated by the fact thata significant subset of pediatric ALLs develops in utero with the attendant riskfor transplacental leukemic “metastasis” [67]. Recent whole-genome associationanalyses have provided some initial insights [68, 69]. Using high-density single-nucleotide polymorphism (SNP)-based analysis, several germline polymorphismshave been identified which appear to confer an increased risk for the developmentof ALL. In two independent studies, SNPs located on chromosome 7p12.2 and map-ping to or near the IKZF1 locus were the most strongly associated with risk for ALL.Both groups also identified SNPs mapping to the ARID5B gene (AT-rich interactivedomain 5B). Interestingly, both studies found associations between certain SNPsand specific ALL subtypes. For example, ARID5B polymorphisms were specificallyassociated with hyperdiploid B-ALL, whereas another SNP mapping to ORC2C3was associated with ETV6-RUNX1+ B-ALLs [69].

In addition, population studies indicate increased risk for ALL development inchildren harboring polymorphisms in gene products involved in DNA mismatchrepair, P-glycoprotein-mediated drug efflux, and folate metabolism [70–72]. Aswith most SNP disease association studies of this type, the underlying mecha-nisms by which the identified alleles actually predispose to ALL leukemogenesisare incompletely understood. Collectively, however, these data indicate that the inutero and childhood development of ALL is likely multifactorial and in a subsetof patients may be attributable to inherited abnormalities that directly predispose toleukemia due to dysregulated lymphoid progenitor development and proliferation aswell as genetic lesions that indirectly contribute to leukemogenesis through alteredxenobiotic and micronutrient metabolism.

Cancer Stem Cells in Precursor B-ALL: Definitionsand Controversies

It is now well accepted that pluripotent stem cells exist in virtually all normal tis-sues. Such stem cells possess two essential biologic properties, namely a capacity

114 R.B. Lorsbach

for self-renewal and the ability to generate differentiated progeny. With recognitionof the existence of stem cells in normal tissues, considerable effort in recent yearshas focused on determination of whether a similar hierarchy of differentiation existswithin tumors and specifically whether a pluripotent cell analogous to the afore-mentioned normal tissue stem cells, the so-called cancer stem cell (CSC), likewiseexists.

Our current concept of a CSC derives from the pioneering work of John Dickand colleagues who first showed that the capacity to generate disease upon trans-fer to immunocompromised NOD/SCID murine recipients was largely restricted tocells present in the CD34+CD38– fraction of human AMLs [73, 74]. Subsequentanalysis of a wide array of tumors typically coupled with sophisticated cell-sortingapproaches has revealed that within a given cancer, individual tumor cells vary con-siderably in their capacity to subsequently give rise to tumors when, for example,transplanted to immunocompromised murine recipients [75]. Thus, CSCs appear tobe present in many, if not all, types of malignancies.

Like its normal counterpart, a CSC is defined by its possession of self-renewalproperties as well as the capacity to generate more differentiated cellular progeny.In addition, CSCs are quiescent and are thought to express high levels of severaldrug transporters, which would render them relatively resistant to chemotherapy,similar to their benign counterparts [76]. Given this, it is presently thought that ourfailure to eradicate many human malignancies with currently available chemother-apeutics is attributable, at least in part, to this relative intrinsic resistance of CSCsto chemotherapy when compared to their more differentiated tumor cell progeny.Thus, it is hoped that enhanced understanding of cancer stem cell biology willfacilitate the development of more effective therapeutic approaches that specificallytarget CSCs.

Before turning to leukemia stem cells (LSCs), a few salient comments regard-ing the analysis of stem cells are in order. Given the intrinsic properties of stemcell biology, analysis of any candidate CSC population requires ultimately a bio-logic readout, most stringently one in which the properties of the putative CSCpopulation are assessed in vivo. For example, transplantation of tumor cell pop-ulations into immunodeficient murine recipients is a commonly used techniqueto assay for CSCs. However, several factors can impact quantitatively and qual-itatively on the outcome of in vivo analyses of CSCs, including differences inhow the recipient mice are manipulated prior to transplantation (e.g., irradia-tion, other “conditioning” regimens), the route by which experimental cells areactually administered to recipient animals (e.g., subcutaneous injection versusintravenous injection versus intramedullary injection in the case of LSCs), andwhether experimental cells are administered alone or together with normal “car-rier” cells. Thus, it is important to be cognizant of not only the inherent variabilityof any biologic assay but also the fact that subtle technical differences amongthese techniques have the potential to impact significantly on the experimentalreadout.

In addition to these variables related to experimental technique, it is helpful toremember that while they share some biologic properties with normal stem cells


by definition, CSCs are neoplastic, and their biologic behavior may differ signif-icantly in other regards from that of their benign counterparts. For example, aleukemia-initiating genetic lesion (e.g., a chromosomal translocation resulting inexpression of a fusion oncoprotein) may in fact occur in a more differentiated orlineage-committed cell type, and it could impart “stemness” to this otherwise differ-entiated cell by altered gene expression or epigenetic events resulting directly fromthe initial leukemogenic mutation. In addition, the degree to which a CSC niche isrecapitulated in recipient animals used in transplant assays has significant potentialto influence findings. If such niches are not present due to deficient cytokine sig-naling or lack of stromal interactions (including both inappropriate expression ofcritical molecules and attenuated or absent molecular interaction across species),then the “stemness” of a candidate cell population would not be detected. Finally,xenotransplantation may select for a small subset of cancer cells that can surmountthe biologic obstacles posed by these assays, thus artificially imparting a “stemness”to a cell population that does not actually exist in a human host, an interpretationfor which supporting experimental data exist [77]. The nuances and variables ofassessing the biologic properties of putative CSCs have recently been addressed[78, 79].

Let us now turn to LSCs. Since the characterization of CD34+/CD38− LSCs inAML by Dick and colleagues [73], considerable work has been undertaken to iden-tify and characterize LSCs in ALL [80]. In these studies, the isolation of variousleukemic blast populations has exploited the differential and temporally regulatedexpression of several cell surface antigens during the various stages of normalB-cell ontogeny (Fig. 4.3). It should also be noted that leukemic blasts coexpressCD34 and CD38 in more than 95% of cases of B-ALL [81].

While several studies clearly indicate that acquisition of the primary leuke-mogenic mutation occurs within a progenitor cell, studies published to date haveyielded conflicting findings. For example, some FISH analyses of sorted cell popu-lations from patients with t(12;21) B-ALL have failed to identify this translocationin the CD34+CD19− cell fraction, which presumably contains normal totipotentHSCs [82, 83]. However, other investigators have shown that only cells within eitherthe CD34+CD19− or the CD34+CD10− fraction, and not CD34+ cells coexpressingCD19 or CD10, could generate leukemia in NOD/SCID recipient animals, suggest-ing that the t(12;21) is acquired in a cell more immature than a committed B-cellprogenitor.

The picture is also somewhat muddled with regard to Ph+ ALL. Recent analy-ses of Ph+ ALL suggest that the t(9;22) occurs in a committed B-cell progenitor[82, 84]. However, the exact phenotype of this progenitor is uncertain as contra-dictory findings have been published. Hotfilder et al. demonstrated that the t(9;22)was detectable by FISH in approximately 50% of flow-sorted CD34+CD10− cellsversus over 90% of CD34+CD19+ cells [84]. By contrast, Castor et al. found thatthe t(9;22) was present in CD34+CD38−CD19− cells only in cases expressing thep210 BCR−ABL1 isoform but not those expressing p190 [82]. Despite these dif-ferences, progenitors containing the t(9;22) manifested essentially no capacity formyeloid/erythroid differentiation when assayed either in vitro or in vivo. These

116 R.B. Lorsbach

Fig. 4.3 Temporal acquisition of cell surface protein expression during B-cell ontogeny. Thedevelopment and maturation of human B cells is characterized by the tightly regulated expres-sion of several cell surface markers. As indicated, the expression of certain markers is evanescent,e.g., CD34, whereas other cell surface proteins such as CD19 are expressed throughout much ofB-cell ontogeny at a relatively uniform level. This temporally regulated marker expression has beenexploited to define the indicated stages of B-cell maturation. It should be pointed out that leukemicB lymphoblasts not infrequently manifest patterns or levels of antigen expression that deviate fromthose of normal B-cell progenitors

results appear to contradict the findings of older studies of BCR–ABL+ ALLs withp190 expression where the t(9;22) was detected in normal hematopoietic lineagesin addition to the leukemic blasts, suggesting its presence in a multipotential HSCcompartment [81, 85].

A recent study from Vormoor et al. further confounds the picture [86]. Theseinvestigators fractionated leukemic bone marrow samples into CD34+CD19–,CD34+CD19+, and CD34−CD19+ populations and surprisingly found signifi-cant leukemic engraftment in NOD/SCID recipients for all three cell fractions.Furthermore, they showed that transplant of CD19+ cells coexpressing CD20, amarker expressed relatively late in normal B-cell ontogeny (Fig. 4.3), could likewiseyield leukemic engraftment in recipient mice. The basis for these seemingly contra-dictory results is uncertain at present, but may be attributable to subtle differencesbetween the experimental approaches used by these investigators. For example,Vormoor et al. [86] used intrafemoral injection of their stem cell preparations,whereas the other studies employed intravenous injection. Technical differencesin the cell-sorting strategies employed as well as the utilization of different cellsurface marker combinations may also be confounding these analyses. With fewexceptions, most of the studies indicate that there are significant differences in thecapacity of various sorted ALL to generate leukemia in recipient immunodeficientmice, indicative of the existence of LSCs in B-ALL. However, the apparently con-tradictory findings of some studies highlight the need for a standardized approachto the isolation and analysis of LSCs.


Altered NOTCH Signaling in T-ALL

NOTCH1 encodes a heterodimeric receptor that critically regulates cell fate deci-sions by multipotent progenitors during development. As such, NOTCH1 plays animportant role in regulating proliferation, apoptosis, and differentiation in thesecells. Within the hematopoietic system, NOTCH1 critically influences the gener-ation of T-cells at multiple steps in their ontogeny [87, 88]. The commitment ofhematopoietic progenitors to the T-cell lineage is dependent on NOTCH1 signaling.Loss of NOTCH1 within hematopoietic cells results in a block in T-cell developmentat an early CD4/CD8 double-negative stage and the aberrant development of B-cellswithin the thymus. NOTCH1 signaling also regulates subsequent stages of T-celldevelopment including transition of thymocytes from CD4/CD8 double-negative todouble-positive cells, as well as influencing the choice between αβ and γδ T-celldifferentiation.

NOTCH signaling is rather complex, due in part to the existence of multipleligands [including Delta-like (DLL) 1, 3, and 4 as well as Jagged 1 and 2] aswell as multiple NOTCH receptors (NOTCH1, 2, 3, and 4) [87]; for the sake ofbrevity and relevance, our discussion will be limited to NOTCH1. Both NOTCH1and DLLs are type I transmembrane proteins. The mature NOTCH1 receptor isgenerated by cleavage of the nascent NOTCH1 protein by a furin-like protein,forming a heterodimer receptor complex comprised of an extracellular componentand a transmembrane/intracellular subunit. Upon binding ligand, the latter under-goes proteolytic cleavage mediated in part by γ-secretase with release of ICN, theintracellular portion of NOTCH1. ICN then translocates to the nucleus where itforms a multimeric transcription complex with several other cofactors, including theDNA-binding protein CSL, to induce the expression of NOTCH1 target genes [89].

The first indication of a role for NOTCH1 in T-ALL leukemogenesis came fromthe observation that NOTCH1 was the chromosomal target of the t(7;9)(q34;q34.3)[90]. Given the rarity of this translocation, which is detected in less than 1% ofcases, the broader pathogenetic relevance of altered NOTCH1 signaling in T-ALLwas not appreciated until recently when mutations in NOTCH1 were demonstratedin approximately 60% of cases [91–93]. These mutations target the extracellularheterodimerization domain (HD) and the C-terminal inhibitory PEST domain ofNOTCH1. The HD mutations induce NOTCH1 cleavage independent of its bindingto extracellular ligand, whereas mutations in the PEST domain inhibit the proteaso-mal degradation of ICN1. Thus, both types of mutations result in the amplification oraberrant activation of NOTCH1 signaling and expression of critical downstream tar-gets, such as HES1 and MYC, resulting in leukemic development (Fig. 4.4) [94–97].The leukemogenic effect of perturbed NOTCH1 signaling appears to be restrictedto T-cell progenitors, as NOTCH1 mutations have not been reported in B-ALL.

With the identification of NOTCH1 mutations in the majority of T-ALLs, therehas been considerable interest in exploiting the resulting aberrant NOTCH1 sig-naling for therapeutic purposes. An obvious candidate is γ-secretase, an enzymerequired for NOTCH1 proteolysis. γ-secretase inhibitors (GSIs) were readilyavailable, as they had been developed and used therapeutically for Alzheimer

118 R.B. Lorsbach

Fig. 4.4 Role of aberrant NOTCH1 and PI3K/AKT signaling in T-cell ALL. Upon ligand binding,NOTCH1 undergoes two proteolytic cleavages, the latter catalyzed by γ-secretase. This releasesthe intracellular portion (ICN) which then translocates to the nucleus where it forms multimericcomplexes with CSL and other transcriptional coactivators on the promoters of NOTCH-responsivegenes resulting in their transcriptional activation (upper left). Mutations in NOTCH1 are among thecommonest in T-ALL and result in constitutive NOTCH1 signaling which directly and indirectlythrough PI3K/AKT enhances cell growth and survival. The γ-secretase inhibitors (GSIs) block theproteolytic processing of NOTCH1, thereby inhibiting NOTCH signaling (upper right). However,resistance to GSIs is present in a significant subset of T-ALL cases at diagnosis and relapse (lowerleft). Recent studies indicate that this resistance is due to deletion of PTEN, an indirect NOTCH1transcriptional target and central regulator of PI3K/AKT signaling. This loss of PTEN expressionin leukemic T lymphoblasts results in a shift from their dependency on aberrant NOTCH1 signalingto dysregulated PI3K/AKT activity, resulting in continued leukemic proliferation even in the faceof GSI therapy (lower right). Modified from [114]

disease, where they function to block synthesis of amyloidogenic β-amyloid pep-tides. Unfortunately, despite their apparent efficacy in inhibiting the growth of celllines in vitro, clinical trials with GSIs have been disappointing due to a modest anti-leukemic effect and significant toxicity in vivo. Given their potent lymphotoxicity,corticosteroids are a key component of most multiagent chemotherapeutic regimensfor ALL. Interestingly, recent studies indicate that GSIs and corticosteroids may actsynergistically to exert a more significant anti-leukemic effect by restoring corticos-teroid sensitivity in resistant cell lines [98]. In addition, glucocorticoids may blockthe goblet cell metaplasia and intestinal toxicity resulting from the GSI-inducedinhibition of NOTCH1 signaling. While these findings clearly need to be confirmed


in primary leukemic samples and in preclinical models, they suggest that GSIs mayexert clinically meaningful anti-leukemic effect when administered as a componentof combination chemotherapy.

More recently, mutations have also been detected in FBXW7, an E3 ligasethat ubiquitinates NOTCH1 as well as MYC, targeting them for proteasomaldegradation, and thereby acting as a negative regulator of NOTCH1 signaling[28]. Mutations in FBXW7 have been detected in 11–24% of adult and pediatricT-ALL/LBL [28, 91, 99, 100]. The prognostic significance of mutations that impacton NOTCH1-mediated signaling, including FBXW7 mutations, is currently unre-solved as some studies have shown that such mutations confer a good prognosis[91, 101, 102], whereas others have shown that NOTCH signaling pathway muta-tions lack prognostic significance or even negatively impact on prognosis [100, 103].The basis for these discrepant findings is currently uncertain, but is presumablyattributable to differences in therapy.

Role of PTEN and PI3K in Pathogenesis of T-ALL

PTEN (phosphatase and tensin analog) negatively regulates the phosphatidyli-nositol 3-kinase (PI3K)-AKT signaling pathway. Upon activation of PI3K byextracellular stimuli, phosphatidylinositol-3, 4, 5 trisphosphate (PIP3) is gener-ated which then recruits AKT, a serine–threonine kinase, to the plasma membranewhere it subsequently undergoes phosphorylation-dependent activation, effectedby 3-phosphoinositide-dependent kinase-1. AKT then phosphorylates several keydownstream molecules, including TSC2, MDM2, and FOXO, which enhance cellgrowth, survival, and proliferation [104]. Through its dephosphorylation of PIP3 tophosphatidylinositol-4,5 bisphosphate (PIP2), PTEN functions as a central, criticalnegative regulator of PI3K signaling.

NOTCH1 represses PTEN expression in both normal thymocytes and T-ALLcells [105]. This effect is mediated indirectly through the induction of HES1 (hairyand enhancer-of-split analog-1). Thus, the abrogation of NOTCH1 signaling byGSIs blocks HES1 expression, resulting in the upregulation of PTEN expressionand consequently inhibition of PI3K-AKT signaling (Fig. 4.4). As alluded to above,GSIs have only a modest anti-leukemic effect in vivo, presumably due at least in partto acquisition of resistance. In support of this notion, GSI-resistant T-ALL cell linesbut not GSI-sensitive ones harbor homozygous deletions or biallelic mutations inPTEN, resulting in complete loss or marked down regulation of PTEN expression[105]. PTEN mutations have likewise been detected in primary T-ALL leukemicsamples at initial diagnosis in 6–9% of cases [105–107]. Loss of PTEN in GSI-resistant cells is accompanied by markedly increased levels of phosphorylated AKT,indicating constitutive activation of the PI3K-AKT signaling pathway [105].

Hyperactivation of PI3K-AKT signaling appears to play a critical oncogenicrole in GSI-resistant T-ALL cells, since an AKT inhibitor markedly inhibited theirgrowth and viability, but not that of GSI-sensitive cells. Thus, acquisition of PTENmutations obviates the need for ongoing NOTCH1 signaling, inducing instead a

120 R.B. Lorsbach

critical need, or “oncogene addiction” for constitutive PI3K-AKT signaling. Amore recent analysis using array-based comparative genomic hybridization simi-larly showed that DNA gains or deletions encompassing genes encoding activators[e.g., MAP kinase interacting serine/threonine kinase 2, insulin-like growth factor2 (IGF2), IGF2 receptor), and inhibitors (PTEN, FOXO3A) of PI3K-AKT signal-ing are present in approximately 30% of pediatric T-ALL cases, indicating thatperturbations of this signaling pathway have an important pathogenetic role inT-ALL [107].

An obvious and important implication of these studies is that effective ther-apy for T-ALL likely requires inhibition of both the NOTCH1 and the PI3K-AKTsignaling pathways. Inhibitors targeting various components of PI3K-AKT signal-ing are currently under development and preclinical evaluation [108, 109]. mTOR(mammalian target of rapamycin) is an important downstream target of PI3K-AKT.Several recent studies indicate that inhibition of multiple downstream signaling tar-gets of NOTCH1 and PI3K-AKT, including mTOR, may work synergistically toinduce cell cycle arrest and apoptosis in T-cell lymphoblasts in in vitro and in vivomurine models [110–112].

Integration of Whole-Genome Analyses with Other AnalyticApproaches to Identify Novel ALL Subtypes

The integration of high-throughput methodologies, detailed knowledge of the biol-ogy of lymphoid precursors, and traditional immunophenotypic analytical methodsholds promise for defining novel, clinically relevant subtypes of precursor lym-phoid malignancies. A recent analysis of T-ALL has demonstrated the power of suchan approach. Early T-cell precursors (ETPs) represent a small minority of thymo-cytes which have recently emigrated from the bone marrow to the thymus and havebeen well characterized immunophenotypically. Through the analysis of T-ALLs todetect those cases which have a gene expression profile most closely resemblingthat of normal ETPs, Campana et al. were able to identify a distinctive subset ofT-ALLs [113]. In addition to having a cell surface immunophenotype, similar tothat of normal ETPs, this subset of T-ALLs manifests increased genomic instabil-ity. Clinically, children with ETP-like T-ALL tend to be younger, respond morepoorly to induction chemotherapy with frequently detectable minimal residual dis-ease, and most importantly have a higher rate of hematologic relapse than typicalT-ALL patients. This study demonstrates the power of integrating different analyt-ical approaches to define novel ALL subsets which may possibly share a commonmolecular pathogenesis. Importantly, such approaches may permit the prospectiveidentification of patients who would benefit from more intensive chemotherapy ormolecularly targeted therapeutics.

Conclusion

During the past four decades, tremendous progress has been made in elucidating thepathogenetic mechanisms underlying precursor ALL/LBL and in the development


of improved therapeutics for these malignancies, which has resulted in markedimprovement in clinical outcome and survival, particularly in the pediatric popu-lation. However, the prognosis of adults with ALL/LBL remains poor, and whileeffective in children, the currently available therapies are highly toxic with manysignificant and long-lasting side effects. Improvements in the therapy for the pre-cursor lymphoblastic malignancies and in the survival of patients afflicted by themwill undoubtedly be predicated on continued advances in our understanding of theirmolecular pathogenesis, a goal that should be attainable with the ongoing applica-tion of whole-genome analytic techniques to the investigation of these malignanciesand the development of molecularly targeted therapeutics.

References

1. Swerdlow SH, Campo E, Harris NL, et al. WHO classification of tumours of haematopoieticand lymphoid tissues. Lyon: IARC Press; 2008.

2. Onciu M. Acute lymphoblastic leukemia. Hematol Oncol Clin North Am. 2009;23:655–674.3. Pui CH, Relling MV, Downing JR. Acute lymphoblastic leukemia. N Engl J Med.

2004;350:1535–1548.4. Van Vlierberghe P, Pieters R, Beverloo HB, Meijerink JP. Molecular-genetic insights in

paediatric T-cell acute lymphoblastic leukaemia. Br J Haematol. 2008;143:153–168.5. Aifantis I, Raetz E, Buonamici S. Molecular pathogenesis of T-cell leukaemia and lym-

phoma. Nat Rev Immunol. 2008;8:380–390.6. Teitell MA, Pandolfi PP. Molecular genetics of acute lymphoblastic leukemia. Annu Rev

Pathol. 2009;4:175–198.7. Nowell PC, Hungerford DA. A minute chromosome in human chronic granulocytic

leukemia. Science. 1960;132:1497.8. Rowley JD. A new consistent chromosomal abnormality in chronic myelogenous leukaemia

identified by quinacrine fluorescence and Giemsa staining. Nature. 1973;243:290–293.9. Grosveld G, Verwoerd T, van AT, et al. The chronic myelocytic cell line K562 contains

a breakpoint in bcr and produces a chimeric bcr/c-abl transcript. Mol Cell Biol. 1986;6:607–616.

10. Shtivelman E, Lifshitz B, Gale RP, Canaani E. Fused transcript of abl and bcr genes inchronic myelogenous leukaemia. Nature. 1985;315:550–554.

11. Mittre H, Leymarie P, Macro M, Leporrier M. A new case of chronic myeloid leukemia withc3/a2 BCR/ABL junction. Is it really a distinct disease? Blood. 1997;89:4239–4241.

12. Pane F, Frigeri F, Sindona M, et al. Neutrophilic-chronic myeloid leukemia: a distinct dis-ease with a specific molecular marker (BCR/ABL with C3/A2 junction). Blood. 1996;88:2410–2414.

13. Wilson G, Frost L, Goodeve A, et al. BCR-ABL transcript with an e19a2 (c3a2) junction inclassical chronic myeloid leukemia. Blood. 1997;89:3064.

14. Goldman JM, Melo JV. Chronic myeloid leukemia–advances in biology and new approachesto treatment. N Engl J Med. 2003;349:1451–1464.

15. Pane F, Intrieri M, Quintarelli C, et al. BCR/ABL genes and leukemic phenotype: frommolecular mechanisms to clinical correlations 1. Oncogene. 2002;21:8652–8667.

16. Gaynon PS, Trigg ME, Heerema NA, et al. Children’s Cancer Group trials in childhoodacute lymphoblastic leukemia: 1983–1995. Leukemia. 2000;14:2223–2233.

17. Maloney KW, Shuster JJ, Murphy S, Pullen J, Camitta BA. Long-term results of treatmentstudies for childhood acute lymphoblastic leukemia: Pediatric Oncology Group studies from1986–1994. Leukemia. 2000;14:2276–2285.

18. Schultz KR, Bowman WP, Aledo A, et al. Improved early event-free survival with imatinibin Philadelphia chromosome-positive acute lymphoblastic leukemia: a children’s oncologygroup study. J Clin Oncol. 2009;27:5175–5181.

122 R.B. Lorsbach

19. Mullighan CG, Goorha S, Radtke I, et al. Genome-wide analysis of genetic alterations inacute lymphoblastic leukaemia. Nature. 2007;446:758–764.

20. Iacobucci I, Storlazzi CT, Cilloni D, et al. Identification and molecular characterizationof recurrent genomic deletions on 7p12 in the IKZF1 gene in a large cohort of BCR-ABL1-positive acute lymphoblastic leukemia patients: on behalf of Gruppo Italiano MalattieEmatologiche dell’Adulto Acute Leukemia Working Party (GIMEMA AL WP). Blood.2009;114:2159–2167.

21. Martinelli G, Iacobucci I, Storlazzi CT, et al. IKZF1 (Ikaros) Deletions in BCR-ABL1-positive acute lymphoblastic leukemia are associated with short disease-free survival andhigh rate of cumulative incidence of relapse: A GIMEMA AL WP Report. J Clin Oncol.2009;27:5202–5207.

22. Georgopoulos K. Haematopoietic cell-fate decisions, chromatin regulation and ikaros. NatRev Immunol. 2002;2:162–174.

23. Mullighan CG, Miller CB, Radtke I, et al. BCR-ABL1 lymphoblastic leukaemia is charac-terized by the deletion of Ikaros. Nature. 2008;453:110–114.

24. Reynaud D, Demarco IA, Reddy KL, et al. Regulation of B cell fate commitmentand immunoglobulin heavy-chain gene rearrangements by Ikaros. Nat Immunol. 2008;9:927–936.

25. Klug CA, Morrison SJ, Masek M, et al. Hematopoietic stem cells and lymphoid progenitorsexpress different Ikaros isoforms, and Ikaros is localized to heterochromatin in immaturelymphocytes. Proc Natl Acad Sci U S A. 1998;95:657–662.

26. Klein F, Feldhahn N, Herzog S, et al. BCR-ABL1 induces aberrant splicing of IKAROS andlineage infidelity in pre-B lymphoblastic leukemia cells. Oncogene. 2006;25:1118–1124.

27. Saijo K, Schmedt C, Su IH, et al. Essential role of Src-family protein tyrosine kinases inNF-kappaB activation during B cell development. Nat Immunol. 2003;4:274–279.

28. Thompson EC, Cobb BS, Sabbattini P, et al. Ikaros DNA-binding proteins as inte-gral components of B cell developmental-stage-specific regulatory circuits. Immunity.2007;26:335–344.

29. Ma S, Pathak S, Trinh L, Lu R. Interferon regulatory factors 4 and 8 induce the expression ofIkaros and Aiolos to down-regulate pre-B-cell receptor and promote cell-cycle withdrawalin pre-B-cell development. Blood. 2008;111:1396–1403.

30. Nakayama J, Yamamoto M, Hayashi K, et al. BLNK suppresses pre-B-cell leukemogenesisthrough inhibition of JAK3. Blood. 2009;113:1483–1492.

31. Trageser D, Iacobucci I, Nahar R, et al. Pre-B cell receptor-mediated cell cycle arrestin Philadelphia chromosome-positive acute lymphoblastic leukemia requires IKAROSfunction. J Exp Med. 2009;206:1739–1753.

32. Mullighan CG, Su X, Zhang J, et al. Deletion of IKZF1 and prognosis in acute lymphoblasticleukemia. N Engl J Med. 2009;360:470–480.

33. den Boer ML, van SM, De Menezes RX, et al. A subtype of childhood acute lymphoblasticleukaemia with poor treatment outcome: a genome-wide classification study. Lancet Oncol.2009;10:125–134.

34. Mullighan CG, Zhang J, Harvey RC, et al. JAK mutations in high-risk childhood acutelymphoblastic leukemia. Proc Natl Acad Sci U S A. 2009;106:9414–9418.

35. Ambros V. The functions of animal microRNAs. Nature. 2004;431:350–355.36. Bueno MJ, Perez dCI, Gomez de CM, et al. Genetic and epigenetic silencing of microRNA-

203 enhances ABL1 and BCR-ABL1 oncogene expression. Cancer Cell. 2008;13:496–506.37. Sercan HO, Sercan ZY, Kizildag S, et al. Consistent loss of heterozygosity at 14Q32

in lymphoid blast crisis of chronic myeloid leukemia. Leuk Lymphoma. 2000;39:385–390.

38. Hasle H, Clemmensen IH, Mikkelsen M. Risks of leukemia and solid tumors in individualswith Down’s syndrome. Lancet 2000;355:165–169.

39. Malinge S, Izraeli S, Crispino JD. Insights into the manifestations, outcomes, and mecha-nisms of leukemogenesis in Down syndrome. Blood. 2009;113:2619–2628.


40. Forestier E, Izraeli S, Beverloo B, et al. Cytogenetic features of acute lymphoblastic andmyeloid leukemias in pediatric patients with Down syndrome: an iBFM-SG study. Blood.2008;111:1575–1583.

41. Khwaja A. The role of Janus kinases in haemopoiesis and haematological malignancy. Br JHaematol. 2006;134:366–384.

42. Levine RL, Pardanani A, Tefferi A, Gilliland DG. Role of JAK2 in the pathogenesis andtherapy of myeloproliferative disorders. Nat Rev Cancer. 2007;7:673–683.

43. Bercovich D, Ganmore I, Scott LM, et al. Mutations of JAK2 in acute lymphoblasticleukaemias associated with Down’s syndrome. Lancet. 2008;372:1484–1492.

44. Kearney L, Gonzalez De CD, Yeung J, et al. Specific JAK2 mutation (JAK2R683)and multiple gene deletions in Down syndrome acute lymphoblastic leukemia. Blood.2009;113:646–648.

45. Gaikwad A, Rye CL, Devidas M, et al. Prevalence and clinical correlates of JAK2 mutationsin Down syndrome acute lymphoblastic leukaemia. Br J Haematol. 2009;144:930–932.

46. Russell LJ, Capasso M, Vater I, et al. Deregulated expression of cytokine receptor gene,CRLF2, is involved in lymphoid transformation in B cell precursor acute lymphoblasticleukemia. Blood. 2009;114:2688–2698.

47. Mullighan CG, Collins-Underwood JR, Phillips LA, et al. Rearrangement of CRLF2 inB-progenitor- and Down syndrome-associated acute lymphoblastic leukemia. Nat Genet.2009;41:1243–1246.

48. Wang YH, Liu YJ. Thymic stromal lymphopoietin, OX40-ligand, and interleukin-25 inallergic responses. Clin Exp Allergy. 2009;39:798–806.

49. Pandey A, Ozaki K, Baumann H, et al. Cloning of a receptor subunit required for signalingby thymic stromal lymphopoietin. Nat Immunol. 2000;1:59–64.

50. Park LS, Martin U, Garka K, et al. Cloning of the murine thymic stromal lymphopoi-etin (TSLP) receptor: formation of a functional heteromeric complex requires interleukin7 receptor. J Exp Med. 2000;192:659–670.

51. Saenz SA, Taylor BC, Artis D. Welcome to the neighborhood: epithelial cell-derivedcytokines license innate and adaptive immune responses at mucosal sites. Immunol Rev.2008;226:172–190.

52. Rochman Y, Leonard WJ. Thymic stromal lymphopoietin: a new cytokine in asthma. CurrOpin Pharmacol. 2008;8:249–254.

53. Soumelis V, Reche PA, Kanzler H, et al. Human epithelial cells trigger dendritic cellmediated allergic inflammation by producing TSLP. Nat Immunol. 2002;3:673–680.

54. Carpino N, Thierfelder WE, Chang MS, et al. Absence of an essential role for thymicstromal lymphopoietin receptor in murine B-cell development. Mol Cell Biol. 2004;24:2584–2592.

55. Ying S, O’Connor B, Ratoff J, et al. Thymic stromal lymphopoietin expression is increasedin asthmatic airways and correlates with expression of Th2-attracting chemokines anddisease severity. J Immunol. 2005;174:8183–8190.

56. Yoo J, Omori M, Gyarmati D, et al. Spontaneous atopic dermatitis in mice expressingan inducible thymic stromal lymphopoietin transgene specifically in the skin. J Exp Med.2005;202:541–549.

57. Zhang Z, Hener P, Frossard N, et al. Thymic stromal lymphopoietin overproduced by ker-atinocytes in mouse skin aggravates experimental asthma. Proc Natl Acad Sci U S A.2009;106:1536–1541.

58. He R, Oyoshi MK, Garibyan L, et al. TSLP acts on infiltrating effector T cells to driveallergic skin inflammation. Proc Natl Acad Sci U S A. 2008;105:11875–11880.

59. Helena Mangs A, Morris BJ. The human pseudoautosomal region (PAR): origin, functionand future. Curr Genomics. 2007;8:129–136.

60. Korenberg JR, Kawashima H, Pulst SM, et al. Molecular definition of a region of chro-mosome 21 that causes features of the Down syndrome phenotype. Am J Hum Genet.1990;47:236–246.

124 R.B. Lorsbach

61. McCormick MK, Schinzel A, Petersen MB, et al. Molecular genetic approach to the charac-terization of the “Down syndrome region” of chromosome 21. Genomics. 1989;5:325–331.

62. Stiller CA, Chessells JM, Fitchett M. Neurofibromatosis and childhood leukaemia/ lym-phoma: a population-based UKCCSG study. Br J Cancer. 1994;70:969–972.

63. Passarge E. Bloom’s syndrome: the German experience. Ann Genet. 1991;34:179–197.64. Taylor AM, Metcalfe JA, Thick J, Mak YF. Leukemia and lymphoma in ataxia telangiectasia.

Blood. 1996;87:423–438.65. Ziino O, Rondelli R, Micalizzi C, et al. Acute lymphoblastic leukemia in children

with associated genetic conditions other than Down’s syndrome. The AIEOP experience.Haematologica. 2006;91:139–140.

66. Whitlock JA. Down syndrome and acute lymphoblastic leukaemia. Br J Haematol.2006;135:595–602.

67. Greaves MF, Wiemels J. Origins of chromosome translocations in childhood leukaemia. NatRev Cancer. 2003;3:639–649.

68. Papaemmanuil E, Hosking FJ, Vijayakrishnan J, et al. Loci on 7p12.2, 10q21.2 and14q11.2 are associated with risk of childhood acute lymphoblastic leukemia. Nat Genet.2009;41:1006–1010.

69. Trevino LR, Yang W, French D, et al. Germline genomic variants associated with childhoodacute lymphoblastic leukemia. Nat Genet. 2009;41:1001–1005.

70. Wiemels JL, Smith RN, Taylor GM, et al. Methylenetetrahydrofolate reductase (MTHFR)polymorphisms and risk of molecularly defined subtypes of childhood acute leukemia. ProcNatl Acad Sci U S A. 2001;98:4004–4009.

71. Mathonnet G, Krajinovic M, Labuda D, Sinnett D. Role of DNA mismatch repair geneticpolymorphisms in the risk of childhood acute lymphoblastic leukaemia. Br J Haematol.2003;123:45–48.

72. Urayama KY, Wiencke JK, Buffler PA, et al. MDR1 gene variants, indoor insecticide expo-sure, and the risk of childhood acute lymphoblastic leukemia. Cancer Epidemiol BiomarkersPrev. 2007;16:1172–1177.

73. Bonnet D, Dick JE. Human acute myeloid leukemia is organized as a hierarchy thatoriginates from a primitive hematopoietic cell. Nat Med. 1997;3:730–737.

74. Lapidot T, Sirard C, Vormoor J, et al. A cell initiating human acute myeloid leukaemia aftertransplantation into SCID mice. Nature. 1994;367:645–648.

75. Schulenburg A, Ulrich-Pur H, Thurnher D, et al. Neoplastic stem cells: a novel therapeutictarget in clinical oncology. Cancer. 2006;107:2512–2520.

76. Dean M, Fojo T, Bates S. Tumour stem cells and drug resistance. Nat Rev Cancer.2005;5:275–284.

77. Kelly PN, Dakic A, Adams JM, Nutt SL, Strasser A. Tumor growth need not be driven byrare cancer stem cells. Science. 2007;317:337.

78. Clarke MF, Dick JE, Dirks PB, et al. Cancer stem cells–perspectives on current sta-tus and future directions: AACR Workshop on cancer stem cells. Cancer Res. 2006;66:9339–9344.

79. Sanchez-Garcia I, Vicente-Duenas C, Cobaleda C. The theoretical basis of cancer-stem-cell-based therapeutics of cancer: can it be put into practice? Bioessays. 2007;29:1269–1280.

80. Bernt KM, Armstrong SA. Leukemia stem cells and human acute lymphoblastic leukemia.Semin Hematol. 2009;46:33–38.

81. Lamkin T, Brooks J, Annett G, Roberts W, Weinberg K. Immunophenotypic differ-ences between putative hematopoietic stem cells and childhood B-cell precursor acutelymphoblastic leukemia cells. Leukemia. 1994;8:1871–1878.

82. Castor A, Nilsson L, strand-Grundstrom I, et al. Distinct patterns of hematopoietic stem cellinvolvement in acute lymphoblastic leukemia. Nat Med. 2005;11:630–637.

83. Hotfilder M, Rottgers S, Rosemann A, et al. Immature CD34+CD19- progenitor/stem cells inTEL/AML1-positive acute lymphoblastic leukemia are genetically and functionally normal.Blood. 2002;100:640–646.


84. Hotfilder M, Rottgers S, Rosemann A, et al. Leukemic stem cells in childhood high-riskALL/t(9;22) and t(4;11) are present in primitive lymphoid-restricted CD34+CD19- cells.Cancer Res. 2005;65:1442–1449.

85. Haferlach T, Winkemann M, Ramm-Petersen L, et al. New insights into the biology ofPhiladelphia-chromosome-positive acute lymphoblastic leukaemia using a combination ofMay-Grunwald-Giemsa staining and fluorescence in situ hybridization techniques at thesingle cell level. Br J Haematol. 1997;99:452–459.

86. Le Viseur C., Hotfilder M, Bomken S, et al. In childhood acute lymphoblastic leukemia,blasts at different stages of immunophenotypic maturation have stem cell properties. CancerCell. 2008;14:47–58.

87. Grabher C, von BH, Look AT. Notch 1 activation in the molecular pathogenesis of T-cellacute lymphoblastic leukaemia. Nat Rev Cancer. 2006;6:347–359.

88. Laky K, Fowlkes BJ. Notch signaling in CD4 and CD8 T cell development. Curr OpinImmunol. 2008;20:197–202.

89. Shih I, Wang TL. Notch signaling, gamma-secretase inhibitors, and cancer therapy. CancerRes. 2007;67:1879–1882.

90. Ellisen LW, Bird J, West DC, et al. TAN-1, the human homolog of the Drosophila notch gene,is broken by chromosomal translocations in T lymphoblastic neoplasms. Cell. 1991;66:649–661.

91. Asnafi V, Buzyn A, Le NS, et al. NOTCH1/FBXW7 mutation identifies a large sub-group with favorable outcome in adult T-cell acute lymphoblastic leukemia (T-ALL): aGroup for Research on Adult Acute Lymphoblastic Leukemia (GRAALL) study. Blood.2009;113:3918–3924.

92. van Grotel M, Meijerink JP, van Wering ER, et al. Prognostic significance of molecular-cytogenetic abnormalities in pediatric T-ALL is not explained by immunophenotypicdifferences. Leukemia. 2008;22:124–131.

93. Weng AP, Ferrando AA, Lee W, et al. Activating mutations of NOTCH1 in human T cellacute lymphoblastic leukemia. Science. 2004;306:269–271.

94. Jarriault S, Brou C, Logeat F, et al. Signalling downstream of activated mammalian Notch.Nature. 1995;377:355–358.

95. Weng AP, Millholland JM, Yashiro-Ohtani Y, et al. c-Myc is an important direct tar-get of Notch1 in T-cell acute lymphoblastic leukemia/lymphoma. Genes Dev. 2006;20:2096–2109.

96. Palomero T, Lim WK, Odom DT, et al. NOTCH1 directly regulates c-MYC and activates afeed-forward-loop transcriptional network promoting leukemic cell growth. Proc Natl AcadSci U S A. 2006;103:18261–18266.

97. Palomero T, Ferrando A. Oncogenic NOTCH1 control of MYC and PI3K: challengesand opportunities for anti-NOTCH1 therapy in T-cell acute lymphoblastic leukemias andlymphomas. Clin Cancer Res. 2008;14:5314–5317.

98. Real PJ, Tosello V, Palomero T, et al. Gamma-secretase inhibitors reverse glucocorticoidresistance in T cell acute lymphoblastic leukemia. Nat Med. 2009;15:50–58.

99. Larson GA, Chen Q, Kugel DS, et al. The impact of NOTCH1, FBW7 and PTEN mutationson prognosis and downstream signaling in pediatric T-cell acute lymphoblastic leukemia: areport from the Children’s Oncology Group. Leukemia. 2009;23:1417–1425.

100. Mansour MR, Sulis ML, Duke V, et al. Prognostic implications of NOTCH1 andFBXW7 mutations in adults with T-cell acute lymphoblastic leukemia treated on the MRCUKALLXII/ECOG E2993 protocol. J Clin Oncol. 2009;27:4352–4356.

101. Breit S, Stanulla M, Flohr T, et al. Activating NOTCH1 mutations predict favorable earlytreatment response and long-term outcome in childhood precursor T-cell lymphoblasticleukemia. Blood. 2006;108:1151–1157.

102. Park MJ, Taki T, Oda M, et al. FBXW7 and NOTCH1 mutations in childhood T cell acutelymphoblastic leukaemia and T cell non-Hodgkin lymphoma. Br J Haematol. 2009;145:198–206.

126 R.B. Lorsbach

103. Zhu YM, Zhao WL, Fu JF, et al. NOTCH1 mutations in T-cell acute lymphoblastic leukemia:prognostic significance and implication in multifactorial leukemogenesis. Clin Cancer Res.2006;12:3043–3049.

104. Chalhoub N, Baker SJ. PTEN and the PI3-kinase pathway in cancer. Annu Rev Pathol.2009;4:127–150.

105. Palomero T, Sulis ML, Cortina M, et al. Mutational loss of PTEN induces resistance toNOTCH1 inhibition in T-cell leukemia. Nat Med. 2007;13:1203–1210.

106. Gutierrez A, Sanda T, Grebliunaite R, et al. High frequency of PTEN, PI3K, and AKTabnormalities in T-cell acute lymphoblastic leukemia. Blood. 2009;114:647–650.

107. Remke M, Pfister S, Kox C, et al. High-resolution genomic profiling of childhood T-ALLreveals frequent copy-number alterations affecting the TGF-beta and PI3K-AKT pathwaysand deletions at 6q15–16.1 as a genomic marker for unfavorable early treatment response.Blood. 2009;114:1053–1062.

108. Hennessy BT, Smith DL, Ram PT, Lu Y, Mills GB. Exploiting the PI3K/AKT pathway forcancer drug discovery. Nat Rev Drug Discov. 2005;4:988–1004.

109. Garcia-Echeverria C, Sellers WR. Drug discovery approaches targeting the PI3K/Aktpathway in cancer. Oncogene. 2008;27:5511–5526.

110. Chan SM, Weng AP, Tibshirani R, Aster JC, Utz PJ. Notch signals positively regulate activityof the mTOR pathway in T-cell acute lymphoblastic leukemia. Blood. 2007;110:278–286.

111. Chiarini F, Fala F, Tazzari PL, et al. Dual inhibition of class IA phosphatidylinositol3-kinase and mammalian target of rapamycin as a new therapeutic option for T-cell acutelymphoblastic leukemia. Cancer Res. 2009;69:3520–3528.

112. Cullion K, Draheim KM, Hermance N, et al. Targeting the Notch1 and mTOR pathways ina mouse T-ALL model. Blood. 2009;113:6172–6181.

113. Coustan-Smith E, Mullighan CG, Onciu M, et al. Early T-cell precursor leukaemia: a subtypeof very high-risk acute lymphoblastic leukaemia. Lancet Oncol. 2009;10:147–156.

114. Gutierrez A, Look AT. NOTCH and PI3K-AKT pathways intertwined. Cancer Cell.2007;12:411–413.

Chapter 5Molecular Pathology of Acute MyeloidLeukemias

Karen P. Mann and Debra F. Saxe

Keywords AML · Acute myeloid leukemia · APL · Acute promyelocyticleukemia · Myeloid · RQ-PCR · RT-PCR · PCR · Capillary electrophoresis ·Electropherogram · Sequencing · DNA · RNA · Translocation · Point mutation ·Insertions · Deletions · Gene expression · Prognosis · FLT3 ·NPM1 · PML · RARA · RUNX1 · RUNXT1 · MYH11 · CBFB ·MLL · Double fusion probe · Breakapart probe · Chromosome enumerationprobe · FISH · WT1 · CEBPA · ERG · BAALC · Myelodysplasia · Myelodysplasticsyndrome · Myeloproliferative neoplasm · BCR-ABL1 · JAK2 · MPL ·Exon · Intron · Breakpoints · Primers · MRD · Minimal residual disease · WHOclassification · Karyotyping · Allele-specific PCR · Melt-curve · AML withrecurrent genetic abnormalities · AML with balanced translocations · AML withgene mutations · inv(16) · t(16;16) · t(8;21) · t(15;17) · t(5;17) · t(11;17) · PML–RARA · RUNXI-RUNXITI · CBFB-MYH11 · MLL-X · MLL-PTD · Partialtandem duplication · NRMI · NUMA · PLZF · PLZF-RARA · NPM1-RARA ·NUMA-RARA · ATRA · Core binding factor AML · CBF-AML · KIT ·RUNX2 · RUNX1 · class I · class II · Fusion protein · ETO ·MLLT3-MLL1 · t(9;11) · Pediatric AML · Prognosis · AF9 ·AF10 · ELL · AF6 · ENL · AF17 · SEPT6 · AML-MDS · AMLwith myelodysplasia-related changes · Therapy-related AML · t(6;9) · DEK-NUP214I · inv(3) · t(3;3) · RPN1-EVI1 · t(1;22) · RBM15-MKL1 · AML withmutated NPM1 · AML with mutated CEBPA · NK-AML · Normal karyotypeAML · FLT3-LM · FLT3-TKD · D835 · CCAAT/enhancer-binding proteinalpha · Chromosome 5 · Chromosome 7 · 5q- · 7q- · Monosomy · CTNNAI ·RPS14 · DIAPH1 · Egr1/Krox20 · Alpha-catenin · MLL5 · APS · CUTL1 ·7q22 · D7S486 · D7S498 · D7S505 · 7q31-4 · t(5;11) · t(2;11) · t(11;16) · Tyrosinekinase · Allelic ratio · Wilms tumor 1 · Homozygous · Heterozygous

K.P. Mann (B)Department of Pathology and Laboratory Medicine, Emory University, F143c Emory UniversityHospital, 1364 Clifton Rd NE, Atlanta, GA 30322, USAe-mail: [email protected]


128 K.P. Mann and D.F. Saxe

Introduction

Acute myeloid leukemia (AML) is a family of hematopoietic neoplasms character-ized by proliferation of myeloid-lineage blast cells in the bone marrow, peripheralblood, and/or extramedullary sites. Appropriate classification and subclassificationof these neoplasms is necessary to ensure appropriate therapy. This classification isincreasingly based upon the underlying genetic abnormalities of these diseases, andevaluation for these abnormalities has become standard of care in the diagnosis andtreatment of acute leukemias. The current WHO classification incorporates a com-bination of clinical features, morphologic features, immunophenotype, karyotype,and molecular testing to precisely subclassify these diseases [1]. This is considereda work in progress, and, as more abnormalities are identified, they will be incor-porated into novel diagnostic and prognostic subgroups. In addition, at least someof these mutations are being incorporated into minimal residual disease testing andsome are targeted for directed therapy.

A variety of types of mutations have been described requiring a variety of diag-nostic modalities. These include balanced translocations [e.g., t(15;17)(q22;q21)],insertions (e.g., NPM1), deletions [e.g., del(5q)], point mutations (e.g., FLT 3-D835), and duplications (e.g., MLL-PTD). Some of these mutations define specificsubtypes of disease, whereas others provide prognostic information and/or guidetherapy. A variety of detection techniques can be used, depending on the specificabnormalities. This may include FISH and conventional karyotyping, sequenc-ing, PCR, RT-PCR, and RQ-PCR, each of which has unique advantages anddisadvantages.

Conventional karyotyping provides broad overview of all of the patient’s abnor-malities. It will not, however, detect small abnormalities (<3–5 Mb) or maskedtranslocations and has low analytic sensitivity as typically only 20 cells are eval-uated. FISH increases the sensitivity by approximately 10-fold and is excellent atdetecting masked translocations that are not picked up by chromosome analysis.Since the probes cover large areas of DNA (100 kb–1.5 Mb), polymorphisms ormutations do not interfere with their ability to bind. In addition, FISH excels in sit-uations where a given gene has multiple translocation partners (e.g., MLL). Usingproper probe design, one can demonstrate that a given gene has been involved in atranslocation without needing probes for every possible gene partner. Many labora-tories have devised specific FISH probes panels in addition to standard karyotypingto identify recurring prognostic cytogenetic abnormalities in AML. These panelsmay vary among laboratories but most often include combinations of the follow-ing probes: 5q–/–5, 7q–/–7, t(8;21), inv(16), MLL, t(15;17), and t(9;22). Althoughmost of these abnormalities can be detected readily by chromosome analysis, theaddition of FISH can resolve complex karyotypes involving rearrangement of themany chromosomes included in the panels and are very informative in preparationswith few dividing cells. Once specific abnormalities have been identified by FISH,the patient may then be followed by FISH alone for the appropriate abnormalities.When disease progression is suspected, the entire panel and standard cytogeneticsare warranted to look for additional abnormalities.

5 Molecular Pathology of Acute Myeloid Leukemias 129

Amplification-based techniques (PCR, RT-PCR, and RQ-PCR) have the highestanalytic sensitivity and can be used both at diagnosis and for minimal residual dis-ease (MRD) testing. Point mutations can be detected by sequencing (considered thegold standard), but can also be identified by allele-specific PCR, PCR with melt-curve analysis, and in specific cases PCR followed by restriction enzyme digestionand fragment analysis.

In this chapter, we will review the molecular abnormalities in AML with a focusupon those that are considered standard of care in diagnosis, treatment monitoring,and MRD assessment. In addition, we will describe the testing modalities used todemonstrate the abnormalities, discuss the appropriate utilization of these tests, anddescribe the pros and cons of specific types of testing. A brief review of futurediagnostic possibilities will also be discussed.

Acute Myeloid Leukemia

AML is divided into subcategories based in large part upon underlying geneticabnormalities (Table 5.1) [1]. Many of these abnormalities are balanced transloca-tions which were originally identified by conventional karyotyping. Subsequently,FISH and PCR techniques have been developed to detect the more common ofthese. Detection of these abnormalities has become standard of care in treatmentof patients with AML. In addition, mutations with significant prognostic implica-tions have been and are being identified (Table 5.1). Their usage is increasing asdiagnostic tests and targeted therapies are developed. In specific settings, minimalresidual disease testing is becoming increasingly used as well to monitor responseto therapy and detect early relapse.

Table 5.1 AML-associated mutations

AML with recurrentgeneticabnormalities

Balancedtranslocations

t(8;21)(q22;q22); RUNX1-RUNX1T1inv(16)(p13.1q22) or t(16;16)(p13;q22);

CBFB-MYH11t(15;17)(q22;q12); (PML–RARA)t(9;11)(q22;q23); MLLT3-MLLt(6;9)(p23;q34); DEK-NUP214inv(3)(q21q26.2); RPN1-EVI1t(1;22)(p13;q13); RBM15-MKL1

Gene mutations Mutated NPM1 (provisional entity)Mutated CEBPA (provisional entity)

Other significant gene mutations in AML FLT3-ITDFLT3-D835KIT exon 8 or 17 mutationsWT1MLL translocations [other than t(9;11)]MLL partial tandem duplications


AML with Recurrent Genetic Abnormalities

AML with Balanced Translocations/Inversions

Acute Promyelocytic Leukemia with t(15;17)(q22;q21); PML–RARA (APL)

Demonstration of the t(15;17)(q22;q21) PML–RARA, or more rarely one of sev-eral alternative translocations (X-RARA), is required for diagnosis of APL [1, 2].The translocation results in creation of a fusion protein involving the two genes:PML on chromosome 15q22 and RARA on chromosome 17q21 [3]. The PML pro-tein product is believed to be involved in a number of cellular functions includingapoptosis, tumor suppression, and senescence [4]. The retinoic acid receptor alphais a ligand-dependent nuclear receptor which controls expression of genes involvedin hematopoietic differentiation and growth [3]. The PML–RARA fusion gene joinsthe nuclear localization signal and dimerization domains of the PML gene to theDNA-binding and ligand-binding domains of RARA [3, 4].

Detection of the PML–RARA translocation can be performed using conven-tional karyotyping, FISH, RT-PCR, or RQ-PCR each of which has advantages anddisadvantages as described above. Since patients with APL have a high risk of dis-seminated intravascular coagulopathy, rapid diagnosis followed by institution ofspecific therapy [all-trans retinoic acid (ATRA) along with cytotoxic chemother-apy] is essential. In order to assess response to therapy and likelihood of relapse,RT-PCR testing is routinely performed [2, 5, 6].

FISH probes are commercially available to detect this translocation. The twomost common strategies are dual color, dual fusion probes to detect PML/RARA,and breakapart probe directed at RARA (Fig. 5.1). In the dual fusion probe designeach gene is labeled with a different fluorochrome, typically spectrum orange andspectrum green, each of which spans the breakpoint of the respective genes. Whenthe t(15;17) occurs, the distal portion of each gene is translocated to the other chro-mosome (Fig. 5.1b). The two translocated chromosomes then each contain a fusionsignal (yellow) consisting of half a red and half a green signal. The normal homologsappear as single red and green signals (2F1R1G). Normal cells contain two redand two green signals (2R2G) (Fig. 5.1a). This strategy will also identify t(15;17)variants such as complex 3-way translocations. Alternatively, the RARA breakapartprobe set spans the gene with the distal portion or 3’ end in one color and the prox-imal region or 5’ end in another color (fusion/yellow, F). If there is a translocationinvolving RARA, the break occurs between the two colors, separating the red andgreen signals (Fig. 5.1d). A translocation in RARA appears as one fusion signal,one red signal, and one green signal (1F1R1G). This probe demonstrates that arearrangement in the RARA gene has occurred, but does not designate the partnerchromosome. RARA rearrangements with PML and rare related genes such as NRMI(5q32), NUMA (11q13), PLZF (11q23), and others all show the same breakapartpattern, 1F1R1G.

The presence of the fusion transcript can also be demonstrated by RT-PCR orRQ-PCR. Although the breakpoint in the RARA gene consistently occurs in intron2, there are three distinct breakpoints in the PML gene, bcr1 (50–60%), bcr2 (5%),


Fig. 5.1 Sample FISH results from patient with APL. Interphase cells hybridized with PML/RARAdual color, dual fusion probe set, with normal cells showing 2R2G (a) and t(15;17) cells showing2F1R1G signal patterns (b). Interphase cells hybridized with RARA breakapart probe set, withnormal cells showing 2F (b) and t(?;17) cells showing 1F1R1G signal patterns (d). Signal patternsshown in (c) and (d) could likewise represent a breakapart probe such as is used for AML withinv(16). In this case, normal pattern would show 2F (d) and the abnormal pattern would show1F1R1G

and bcr3 (35–45%) resulting in the L (long), V (variable), or S (short forms,respectively) shown in Fig. 5.2 [3, 4, 7].

Therefore, multiple primers that bind to appropriate loci in the PML gene arenecessary in order to detect all possible transcripts. Consensus primers and probesfor a RQ-PCR assay have been designed by Gabert et al. [8] and validated in amulti-institutional study supported by a Europe against Cancer Program (Fig. 5.3).

Many other laboratories have also developed assays, reviewed by Grimwade andLo Coco [2]. Demonstration of this translocation by RQ-PCR at diagnosis is helpfulnot only to establish the diagnosis (which may also be established by FISH or cyto-genetics) but also to ensure that the primer sets work for the individual patients forminimal residual disease testing. In addition, it allows determination of the changein disease burden with treatment for minimal residual disease testing.

In addition to the common translocation above, alternative rearrangementshave been described including t(11;17)(q23q21) PLZF-RARA, t(5;17)(q32;q21)


Fig. 5.2 Exon–intron structure of PML (yellow boxes) and RARA (blue boxes). Breakpoints areindicated by arrows. Three alternative fusion products are shown (L, V, and S)

Fig. 5.3 RQ-PCR results from two samples from patients with APL with different levels of fusiontranscript burden (PML–RARA). Primers based upon Gabert et al. (8). Results are compared to thereference gene ABL

NPM1-RARA, t(11;17)(q13q21) NUMA-RARA, and others [1–3, 6]. Although rare,they are clinically significant as APL with t(11;17)(q23;q21) does not respondto ATRA, whereas APL with t(5;17)(q32;q21) or t(11;17)(q13q21) do [6, 9, 10].Detection of these alternative translocations is typically performed by conventionalkaryotyping, although they will be detected by both RARA breakapart probes andsome FISH probes designed primarily to detect the PML–RARA fusion. If appro-priately designed, the double fusion probe can demonstrate that an alternativetranslocation involving RARA occurred even though it will not detect the specificfusion partner. In these cases, instead of showing the typical double fusion patternseen in Fig. 5.1b, an extra RARA signal will be seen (2R3G PML/RARA). RT-PCR


or RQ-PCR directed at PML–RARA will not detect the alternative translocations andthe majority of clinical laboratories have not validated RT-PCR/RQ-PCR to detectthese rare translocations.

Molecular minimal residual disease testing has become standard of care in APL.This is typically performed by RT-PCR or RQ-PCR. The persistence of the pml–rara fusion transcript during induction is a well-recognized phenomenon and doesnot indicate failed therapy [6, 11, 12]. A major therapeutic goal is for the fusiontranscript to be undetectable in a bone marrow aspirate sample post-consolidation.Persistence at this time point is associated with an increased rate of relapse [2, 6,11, 12]. If a patient is in molecular remission post-consolidation and the transcriptrecurs, this is likewise associated with increased risk of relapse [2, 12].

A recent international consensus document has been published on behalf ofEuropean LeukemiaNet [6]. This document reiterates that PCR positivity post-induction should not alter therapy. In addition, they recommend molecular moni-toring for PML–RARA every 3 months following consolidation with bone marrowaspirate material as the preferred sample. This monitoring should be continued for3 years. If transcript returns, they recommend early repeat testing to confirm posi-tivity. In this setting positivity indicates that these patients will relapse if not givenadditional therapy.

Core Binding Factor AML

The core binding factor (CBF) AMLs (CBF-AML) are comprised of acuteleukemias with recurrent translocations that involve a gene of the CBF complex.They comprise 10–25% of AML [1, 13]. These are so-called good prognosis AMLas the majority of patients go into complete remission with standard therapy. Themost common of the CBF-AML are AML with t(8;21)(q22;q22) RUNX1[CBFA2]-RUNX1T1 [AML t(8;21)] and AML with inv(16)(p13.1q22) or t(16;16)(p13.1;q22);CBFB-MYH11 [AML inv(16)]. Of note, these translocations are not sufficient fordevelopment of leukemia. Ongoing studies have identified a variety of comple-mentary mutations all of which involve constitutive activation of protein kinases,reviewed by Mrozek et al. [14]. Mutations in the KIT gene in particular have beendescribed in 12–44% of all CBF-AML [14–19] and at least some groups have shownthat these mutations have prognostic significance (see below).

The normal CBF complex is composed of alpha and beta subunits which het-erodimerize and is involved in transcriptional regulation of a host of genes. RUNX1is involved in regulation of normal hematopoietic establishment and differentiation,whereas other members of the RUNX family (RUNX2, RUNX3) are involved in tran-scriptional regulation of other genes [13]. The translocations seen in CBF-AMLresult in fusion proteins that are believed to act as dominant negative inhibitors ofnormal transcription.

AML with t(8;21)(q22;q22) RUNX1-RUNX1T1

AML t(8;21) is one of the two common CBF-AML. This translocation occurs in4–12% of adult AML and 12–30% of pediatric AML [1, 20]. It represents 5–15%


of all AML [1, 13, 21]. This abnormality is considered to be a good prognosiswith the majority of patients going into CR with standard therapy. Although AMLt(8;21) has typical morphologic and immunophenotypic features, diagnosis requiresdemonstration of this unique translocation by conventional cytogenetics, FISH, orRQ-PCR. The use of quantitative RQ-PCR for minimal disease testing is becom-ing standard of care and is described below. A typical FISH strategy is a dualcolor, dual fusion probe using a strategy similar to that described above for APL(Fig. 5.1a, b). Primers and probes have been developed by a number of groupsincluding the Europe Against Cancer Consortium [8, 22–25].

RUNX1 (also known as AML1 and CBFA2) is a member of the core bindingfactor complex which includes RUNX1 and RUNX1T1 [20, 21, 26]. RUNX1 containsa runt homology DNA-binding domain that binds to control elements upstream ofgenes involved in hematopoiesis. Binding is enhanced by heterodimerization withCBFB. Its fusion partner RUNX1T1 [ETO1] has a zinc finger domain as well asNervy homology regions [20, 21, 26]. It acts to inhibit CEBPB. The fusion genecontains the runt homology (i.e., the DNA binding) domain of RUNX1, but lacks itstransactivation domain [26, 27]. This is replaced by almost the entire coding regionof RUNX1T1 (Fig. 5.4). The fusion protein acts in a dominant negative fashionsuppressing function of the normal CBF complex.

Although, AML t(8;21) is generally considered a good prognosis acute leukemia,many patients (30–40%) relapse after completing therapy, and there has been anongoing attempt to identify additional genetic abnormalities that predict relapse[21, 28]. As mentioned above, a subset of patients with CBF-AML harbor KIT genemutations [15–18]. KIT is a type 3 tyrosine kinase receptor. The most commonmutations have been identified in exons 8 and 17 and include the D816 and othermutations in exon 17 [15–20, 29], exon 8 insertions and deletions, and less com-monly internal tandem duplications in exon 11 [18]. KIT mutations are associatedwith early relapse and worse overall survival [15, 16, 18, 19, 29, 30]. Identificationof these abnormalities is typically performed at diagnosis in these patients bysequencing of the appropriate exons of the KIT gene prior to initiating therapy. Othermethodologies including allele-specific PCR and PCR followed by high-resolutionmelt-curve analysis can also be used.

Fig. 5.4 Exon–intron structure of RUNX1 (pink boxes) and RUNX1T1 (blue boxes). Breakpointregions are indicated by arrow or bracket. Fusion product shown below


AML with inv(16)(p13.1q22) or t(16;16)(p13.1;q22); CBFB-MYH11

AML with inv(16)(p13.1q22) or t(16;16)(p13.1;q22); CBFB-MYH11 [AMLinv(16)] is another “good prognosis” CBF-AML. This mutation is identified in 8–12% of patients with AML [13, 20, 31]. It is typically associated with AML withatypical eosinophils with mixed basophilic and eosinophilic granules as well asat least partial monocytic differentiation. The genes involved are the CBFB geneinvolved in the CBF complex as well as the smooth muscle myosin heavy chaingene MYH11 [32–35]. Identification of this translocation can be done by conven-tional karyotyping, FISH, or RT-PCR. Of note, both the inv(16) and the t(16;16)can be difficult to identify by conventional karyotyping and can be easily missed byinexperienced readers. FISH probes are typically designed as a breakapart probe set,which yields the same pattern of 1F1R1G for both the inversion and the translocated16 (see Fig. 5.1c, d). Because of variations in the breakpoints in MYH11 (see below),careful design of primers and probes is essential. RQ-PCR for minimal residualdisease testing is discussed below.

CBFB contains a single domain, the heterodimerization domain, which binds toRUNX1 and stabilizes binding of the CBF complex to DNA [32, 34]. CBFB doesnot directly bind DNA. Breakpoints for the fusion transcript are almost always scat-tered in exon 5, a 15 kb exon of this gene [13, 32, 34, 35]. A single report of anexon 6 breakpoint in CBFB was recently made [36]. MYH11 is a complex gene with42 coding exons which encode a protein of close to 2000 amino acids [37]. Thereare multiple MYH11 breakpoints that occur in AML inv(16), and at least 11 fusiontranscripts have been described to date [8, 31, 33, 36]. Breakpoints found in mul-tiple different exons and in the cDNA span greater than 1000 base pair length ofsequence. Therefore detection requires the use of multiple primer pairs. The mostcommon fusion transcripts are A (88%), D (5%), and E (5%) (Fig. 5.5).

The other alternative transcripts are rare and each one is detected in fewerthan 1% of cases [8, 33, 36, 38, 39]. The fusion in all cases, however, results inthe heterodimerization domain of the CBFB being joined to the multimerizationC-terminal domain of MYH11 [34]. The resulting fusion transcript has a domi-nant negative effect on transcription [33]. Of note, in addition to the variation offusion transcript size determined by MYH11 breakpoint, RUNX1 can have alter-native splicing which results in a fusion protein with only 133 amino acids fromCBFB [34].

KIT mutations are likewise seen in a subset of these patients [15–18]. However,the prognostic significance is less clear. Paschka et al. [16] and Care et al. [15]demonstrated increased relapse rates in patients with AML inv(16) with KIT muta-tions, and Paschka et al. [16] additionally demonstrated decreased overall survivalin these patients. Boissel et al. [17] and Cairoli et al. [18] did not find anydifferences in overall or relapse-free survival based upon KIT mutation status.Additionally, Paschka et al. [16] only found significant changes in patients withexon 17 mutation, whereas Care et al. [15] found changes in patients with exon8 mutations.


Fig. 5.5 Exon–intron structure of CBFB (green boxes) and MYH11 (purple boxes). Breakpointregions indicated by arrows. Most frequent fusion products are shown (a, d, and e)

MRD by RQ-PCR in CBF-AML

The significance of molecular monitoring to evaluate MRD in CBF-AML has beeninvestigated by a number of investigators and reviewed by Marcucci et al. [28]and Mrozek et al. [40]. Initial studies using qualitative RT-PCR and competitiveRT-PCR gave variable results as to the persistence of detectable fusion transcriptsand the significance of their presence or absence during therapy. More recently,investigators have used RQ-PCR to look at fusion transcript levels at diagnosis, afterinduction therapy, after consolidation therapy, and at continued follow-up [41–49].Although each study was designed differently and had somewhat different findings,all demonstrated that monitoring of molecularly determined MRD is of value inCBF-AML.

Multiple investigators demonstrated a wide variation of transcript burdenbetween patients at diagnosis (1–3 log variation) which could not be explainedpurely by blast count or number of positive interphase cells by FISH [43–45, 47].Patients with higher transcript levels were found to have worse overall and event-free survival [44] and increased rate of relapse [45]. This type of variation is unlikewhat has been observed in detection of BCR-ABL1 in chronic-phase CML and isworth noting as many laboratories compare BCR-ABL1 levels to a normalized levelof an average patient in order to determine log decrease or percentage level of tran-script [50]. This would not work in CBF-AML where levels at diagnosis vary sowidely.

Two investigators found a high concordance between levels of MRD in periph-eral blood and bone marrow and proposed that peripheral blood could be used formonitoring in both CBF-AML subtypes [46] or specifically in AML t(8;21) [45].


Other findings include the following: Leroy et al. [45] determined that a >3 logdecrease post-induction or a level >10−5 post-consolidation was associated withdecreased risk of relapse. Weisser et al. [48] used a median level of transcript afterinduction and post-consolidation. At both time points patients with a level greaterthan median showed increased cumulative incidence of relapse and decreased over-all survival and event-free survival. They found a cutoff level of 0.003 (the medianin their study) or a reduction to 0.01% (4 log) could be similarly used. Other pro-posed cutoff levels include greater than 10 copies as compared to a standard curveat end of treatment [47], less than 2 log decrease post-induction [46], greater than1 log increase after complete remission [49], and greater than 2 log reduction [43].Schnittger et al. [44] proposed a score which looked at transcript level at two timepoints, diagnosis and during the first 3–4 months of therapy. He found that patientswho had a level at diagnosis greater than the 75 percentile (as compared to otherpatients) and a level at the 3–4 month time point greater than the median (whichcorresponded to <3 log decrease) are at high risk for treatment failure.

Although these results are promising and indicate the value of MRD detectionin CBF-AML, more work needs to be done, including standardization of RQ-PCRtechniques, standardization of methods of reporting, and development of an interna-tional standard for quantification, before specific cutoffs can guide therapy.

Other Recurrent Translocations

Additional recurrent translocations are included in the WHO classification [1] andare shown in Table 5.1. All of these are rare and vary from representing <1–2% ofadult AML depending on the specific translocation.

AML with t(9;11)(p22q23); MLLT3-MLL and Other MLL Abnormalities

Although AML with t(9;11)(p22q23); MLLT3-MLL [AML t(9;11)] is distinct fromthe other MLL abnormalities in the WHO classification, all MLL abnormalities willbe summarized in this section.

At least 104 distinct MLL rearrangements have been identified. Of these 64 fusionpartners have been characterized at the molecular level [51]. In addition both par-tial tandem duplication (AML-PTD) and MLL amplifications have been described[52–57]. Translocations involving the MLL gene occur in 3–4% of adult AML,14–20% cases of pediatric AML, and 65% of infant AML [58–60]). These translo-cations are frequent in treatment-related AML, specifically those treated withtopoisomerase II inhibitors. MLL-PTD is relatively frequent in adults (5–10%), butoccurs in fewer than 1% of pediatric AML patients [51, 56, 61].

Although the majority of MLL translocations are associated with a poor progno-sis, at least some studies have demonstrated that AML t(9;11) is somewhat uniqueand is associated with an intermediate prognosis in AML [62–66]. Other studieshave not replicated this finding [57, 60]. In addition, studies have demonstratedpoor prognosis [67] for t(6;11)(q27;q23) and good prognosis for t(9;11) in child-hood AML [64]. The MLL gene contains 37 exons and covers over 100 kb [59,68–72]. The protein consists of 3969 amino acids and contains several significant


Fig. 5.6 Exon–intron structure of MLL (yellow boxes). Gene X (green boxes) indicates any knownfusion partners. Fusion transcript and MLL-PTD shown with protein domains: AT (AT hooks),SNL (nuclear localization signal), DMT (dimerization domain), PHD (PHD finger domain), TAD(transactivation domain), and SET (domain involved in histone methylation)

domains including AT hooks, a nuclear localization domain, repressor domains, andPHD domains [68–72]. The breakpoint cluster region (Fig. 5.6) covers over 8 kband breakpoints can occur in introns 5–11 [59, 73]. MLL is involved in a multi-protein complex that is present in the nucleus and involved in remodeling of thenucleosome. It has histone methyltransferase activity and is involved in methyla-tion, acetylation, and nucleosome remodeling [72]. Its role in HOX gene regulationis essential for normal hematopoiesis. Although there are numerous translocationpartners, the majority of MLL translocations in AML involve the following genes:AF9, AF10, ELL, AF6, ENL, AF17, and SEPT [51]. Translocations fall into twotypes: class I mutations involving fusion to a nuclear protein and class II mutationswhich result in cytoplasmic localization [56, 72].

It is important to note that patients with acute leukemia with translocationsinvolving MLL may fall into a number of different categories including AML withrecurrent translocations, AML with myelodysplasia-related changes (AML-MDS),and therapy-related AML. Appropriate classification of these cases requires a com-bination of clinical history as well as identification of specific translocations (seebelow). In cases with MLL translocations that do not have a translocation that fallsunder AML-MDS the specific abnormality is listed in the diagnosis [1].

Although it is theoretically possible to develop RQ-PCR testing to detect all ofthese translocations, it is impractical for the clinical laboratory. Therefore, MLLtranslocations are typically detected by a combination of karyotyping and FISHusing a breakapart probe (Fig. 5.1c, d). FISH using a breakapart probe has the advan-tage of detecting cryptic translocations as well as MLL-PTD and amplifications. Itdoes not, however, demonstrate the translocation partner.

MRD in AML with MLL Abnormalities

Although MRD detection by RQ-PCR is difficult due to the multiple translocationpartners, some studies have demonstrated clinical significance to MRD detection


[51, 55, 74]. Therefore, laboratories should at least consider whether RT-PCRanalysis detecting specific translocations is appropriate in their clinical setting.Conventional karyotyping will, however, detect the majority of these. Of note, MLL-PTD is detectable in up to 100% of healthy adults [56] and in 93% of cord bloodsamples when high-sensitivity techniques are used. The level is typically 4 log lowerthan is seen in AML with MLL-PTD.

Rare Subtypes of AML

In addition to the commonly seen subtypes of AML described above, rare yet recur-rent translocations are also incorporated into the new WHO classification, specif-ically: AML with t(6;9)(p23;q34); DEK-NUP214, AML with inv(3)(q21q26.2) ort(3;3)(q21;q26.2); RPN1-EVI1, and AML with t(1;22)(p13;q13); RBM15-MKL1.Taken together they comprise approximately 2–4% of AML [1]. These mutationsare typically detected by conventional karyotyping instead of RQ-PCR due to theirrarity.

AML with t(6;9) occurs in adults and children. Patients have a poor prognosiswith short survival. There is often multilineage dysplasia (especially in the erythroidand granulocytic series) as well as an absolute basophilia. Like the other AML withrecurrent translocations, a diagnosis of AML with t(6;9) can be made even if blastscomprise fewer than 20% of cells. The translocation is amenable to detection byboth Southern blot and RQ-PCR [75–77].

AML with inv(3) or t(3;3) is also a poor prognosis AML. It occurs bothas a therapy-related AML and de novo. There is often multilineage dysplasiawith megakaryocytic dysplasia often with micromegakaryocytes with unilobated orbilobed nuclei. A subset of cases show marked thrombocytosis at diagnosis [1].Fusion transcripts can be evaluated by RT-PCR [78]; however, these studies are notroutinely performed in the majority of clinical laboratories.

AML with t(1;22) is very rare (<1% of AML) and is typically seen in infants[1, 79, 80]. A single case report exists describing an adult patient [81].

AML with Gene Mutations

In addition to the balanced translocations described above, numerous gene muta-tions have been described and continue to be described associated with AML. Thesecan be divided into class I mutations (involved in proliferation and apoptosis) andclass II mutations (associated with differentiation). Many of these are felt to becooperating mutations and are found in a number of different subtypes of AML,whereas others may represent novel AML subtypes defined by gene mutations. Twoof these are identified as provisional entities in the WHO classification [1].

AML with Mutated Nucleophosmin 1 (NPM1) (Provisional Entity)

Mutation of NPMI is the most common mutation in AML and occurs in approxi-mately 30% of all AML and 50–60% of normal karyotype AML (NK-AML) [1, 61,


82]. It occurs in both pediatric and adult AML, but is significantly rarer in children,4% of all pediatric AML [83]. In NK-AML, the presence of the NPM1 mutation inthe absence of an accompanying FLT3-LM is associated with a good prognosis [61,84–88]. NPM1 encodes a shuttling protein that is typically located in the nucleo-lus of the cell. Mutations typically involve 4–5 base pair insertions in exon 12 butrare alternative translocations have been identified in exons 9 [89] and 11 [90, 91](Fig. 5.7). At least 40 unique mutations have been identified to date. The most com-mon mutations are A, B, and D with A occurring in up to 80% of cases [92, 93]. Allresult in abnormal localization from the nucleus into the cytoplasm due to creationof a nuclear export signal.

Detection of NPM1 can be performed by PCR using primers flanking the siteof insertion (Fig. 5.8), followed by fragment analysis [94]. This should be per-formed at the time of diagnosis in NK-AML and should be paired with evaluationfor FLT3-LM (see below) for appropriate assessment of prognosis. The mutation

Fig. 5.7 Exon–intron structure of NPM1. Major mutation site is indicated by arrow. Raremutations sites are indicated by double-headed arrow

Fig. 5.8 Electropherogramdemonstrating the presence ofan NPM1 exon 12 insertionmutation by PCR withfragment analysis by capillaryelectrophoresis. Unmutatedand mutated genes indicatedby arrows


occurs in a single allele and therefore both the mutated and the unmutated geneswill be detected. This approach also has the advantage that it should detect anymutation within the coding sequence flanked by the primer set. More recently avariety of groups have developed RQ-PCR assays using either specific primer setstargeting common mutations [95] or multiple primer sets that detect multiple dif-ferent mutations [92, 93, 96]. A typical strategy is to utilize a consensus forward orreverse primer paired with mutant-specific reverse or forward primers, respectively.Consensus probes are used [92, 93, 96].

NPM1 is an attractive target of MRD testing as it is believed to occur earlyin oncogenesis and the majority of authors find it stable at relapse [92, 93, 96].There are reports of clonal evolution at relapse [87, 95], but these appear to bethe exception and not the rule. Using a RQ-PCR assay, both Gorello et al. [92]and Chou et al. [96] were able to monitor disease level in these patients and insome cases predict relapse. More recently Schnittger et al. [93] described RQ-PCRassays which were able to define 17 different NPM1 mutations in 252 patients.They additionally found that monitoring of mutant transcript level during treat-ment could predict likelihood of complete clinical response as well as likelihood ofrelapse.

AML with Mutated CCAAT/Enhancer-Binding Protein Alpha (CEBPA)(Provisional Entity)

CEBPA is a member of the CCAAT/enhancer-binding protein family with a rolein proliferation and differentiation in myelopoiesis [97, 98]. Down-regulation ofCEBPA is associated with CBF-AML and APL [98]. Mutations in CEBPA are seenin 10–15% of AML [97] and in approximately 15% of NK-AML [61]. They areseen in 7% of pediatric AML [83]. Patients with mutated CEBPA have a good prog-nosis similar to that seen in the CBF-AML and APL [97, 99]. CEBPA is encodedon chromosome 19q13.1. It encodes multiple domains including several transacti-vation domains as well as a leucine zipper, DNA-binding and dimerization domain[97, 98]. Mutations cluster in several regions and biallelic mutations are common[100]. Mutations in the N-terminal region are typically out-of-frame insertions ordeletions. They result in increase in an alternative isoform which acts in a dominantnegative fashion to inactivate the unmutated gene. Alternatively mutations occur inthe C-terminal DNA-binding domain and interfere with normal transcription acti-vation of myeloid genes involved in differentiation. Although studies are limited,the mutations are not detected at the time of CR; however, if patients relapse, theinitial mutations are once more detectable [101]. To the best of our knowledge,development of novel CEBPA mutations at relapse has not been described. Giventhe variation in mutations seen in this gene (i.e., mutations occur over the entirecoding sequence of the gene), routine testing in the clinical laboratory has not beenadopted; however, multiplex or multiple PCR reactions in conjunction with frag-ment analysis can be used as a screening technique to identify these mutations[102, 103].


AML with Myelodysplasia-Related Changes

Diagnosis of this entity can be made by using a combination of clinical history(prior MDS and absence of prior cytotoxic therapy), morphologic/cytologic findingsof dyspoiesis, absence of recurring genetic abnormality not associated with MDS,and the presence of specific abnormalities associated with MDS [1]. Diagnosis,therefore, depends in part upon demonstrating the presence of MDS-associatedabnormalities by FISH and/or conventional cytogenetics. Specific alterations havebeen described that will fulfill the criteria for AML with MDS-related changes,summarized in Table 5.2 [1]. Unbalanced abnormalities of chromosome 5 and 7 arethe most common abnormalities in this group. These abnormalities result in lossof all of chromosomes 5 and 7 (monosomy) or loss of portions of the long arm(Fig 5.9). Although the breakpoints vary with both 5q and 7q rearrangements, thecommonly deleted regions (CDR) are 5q31-32, 7q22, and 7q32-34 [104, 105]. TheCDR on chromosome 5 contains over 40 genes, several of which have been impli-cated in the pathogenesis of myeloid malignancies. These candidate genes includeCTNNAI [106], RPS14 [107], DIAPH1, Egr1/Krox20, and alpha-catenin [108–111].Such genes as MLL5,APS, and CUTL1 at 7q22 and regions D7S486, D7S498, andD7S505 at 7q31-4 have been suggested to be involved in myeloid disease within thetwo commonly deleted regions on 7q [112–116].

Table 5.2 AML-MDS-related cytogenetic abnormalities

Complex (>3 abnormalities)–5 or del(5q)–7 or del(7q)i(17q) or t(17p)–13 or del(13q)del(11q)del(12p) or t(12p)del(9q)idic(X)q13t(1;3)(p36.3;q21.1)t(11;16)(q23;p13.3)t(2;11)(p21;q23)t(3;21)(q26.2;q22.1)t(3;5)(q25;q34)t(5;10)(q33;q21)t(5;12)(q33;p12)t(5;17)(q33;p13)t(5;7)(q33;q11.2)

Although the majority of MLL translocations are categorized under AML withrecurrent translocations (see above), two specific translocations fall in this categoryt(2;11)(p21;q23) and t(11;16)(q23;p13.3) if the patient does not have prior historyof chemotherapy.


Fig. 5.9 FISH images using enumeration probe set for chromosome 5 with the Cri-du-Chat regionat 5p15.2 labeled in green and the EGR1 region at 5q31 in red. (a) Metaphase and interphase FISHdemonstrate normal pattern of this probe (2G2R). (b) FISH pattern showing deletion/monosomyof chromosome 5 (1G1R). (f ) FISH pattern showing interstitial deletion 5q (2G1R)

Other Gene Mutations in AML with Prognostic Significance

FLT3-LM and FLT3-TKD

Activating mutations in the type 3 receptor tyrosine kinase FLT3 gene were firstidentified by Nakao et al. [117] and subsequently have been described in 25–30%of AML patients and in 40% in NK-AML [61, 118, 119]. These mutations arerarer in pediatric AML and are seen in approximately 10–15% of patients [83,119]. When activated, mutated FLT3 acts on downstream pathways including thePI3 kinase/AKT pathway, RAS/MAP kinase pathway, and STAT5 pathway [120].The gene is, therefore, involved in a number of normal cellular processes includingapoptosis, proliferation, and differentiation.

FLT3 mutations in AML constitutively activate the receptor by two distincttypes of mutations: in-frame length mutations (LM) in a regulatory juxtamembranedomain and point mutations (either at D835 or at I836) in the tyrosine kinase domain(TKD) as shown in Fig. 5.10 [1, 119]. The former has been shown to predict poor


Fig. 5.10 Structure of FLT3 gene with protein domains indicated. IGG (immunoglobulin likedomains), JM (juxtamembrane), TK1, and TK2 (tyrosine kinase domains). Sites of LM (lengthmutations) and TKD (tyrosine kinase domain point mutations) are indicated

prognosis in NK-AML [61, 119]. The significance of the point mutation, however, isless well established. Some studies have shown that the TKD mutations also confera poorer prognosis [118, 121], whereas others either do not show prognostic signif-icance or must be analyzed in conjunction with other mutations [122]. Specifically,they found that TKD mutation in NK-AML does not demonstrate prognostic signif-icance; TKD mutations along with FLT3-LM, PML–RARA, or MLL-PTD mutationsconfer a poor prognosis, and TKD mutations along with NPM1 or CEBPA mutationsshow improved prognosis [122]. FLT3 mutations are common in APL and are seenin approximately 35% of patients [118, 123]. They do not seem to impart the samepoor prognosis in this group. Of note, FLT3 mutations can be lost at time of relapse,may arise at relapse, or the size of the FLT3-LM may change during the diseasecourse [74, 124–127].

Testing for FLT3-LM can be performed by PCR followed by fragment analysisas shown in Fig. 5.11 [94, 128]. The size of the LM varies widely ranging from3 to > 400 base pair and in some patients multiple different size insertions can befound [123, 124]. In the majority of patients, this is a heterozygous mutation, i.e.,

Fig. 5.11 Electropherogram demonstrating detection of the FLT3-LM as performed by PCR fol-lowed by capillary electrophoresis with fragment analysis. Both the unmutated allele (FLT3) andthe mutated allele (FLT3-LM) are indicated


the normal gene is still identified, but in a subset only mutated alleles are detected. Inaddition, some patients have both the LM and the TKD. D835 mutation evaluationcan be performed by PCR followed by restriction enzyme digestion, allele-specificPCR, or sequencing.

The allelic ratio of the LM to the normal allele has been shown to be of prognos-tic significance. In general, higher ratios correlate with worsening prognosis [129].Low levels of allele burden have not been shown to impart the negative prognosisof this mutation. Specific levels of allele burden have been identified in the litera-ture below which the prognosis is no longer poor, < 0.4 in children [130] or < 0.78in adults [118]. It is difficult to know, however, how to apply this in the clinicallaboratory as independent standards are not available to confirm that allelic ratiosare consistent from laboratory to laboratory. Of note, allelic ratio comparisons havebeen performed between laboratories participating in specific Children’s OncologyGroup (COG) trials. Given that this testing is typically performed using end pointPCR and the size of the LM can vary from 3 to > 400 base pair, it is naïve to assumethat the amplification efficiency of the mutated allele will be identical to the unmu-tated gene regardless of insert size. Therefore the establishment of a strict cutoff forallelic burden may not be appropriate unless testing is performed in a more quanti-tative manner. As mentioned above, FLT3 mutations have been known to disappearduring the disease course [126, 127]. Therefore detection of this mutation may notbe a reliable target for MRD testing. FLT3 inhibitors are in early use and show somepromise.

Wilms Tumor 1 (WT1)

WT1 is mutated in 10–15% of AML [131–134] and in approximately 6% cases ofpediatric AML [83]. Two large series specifically analyzed NK-AML and foundmutations in approximately 10% of NK-AML [133, 134]. Mutations tend to clusterin exons 7 and 9 [131–134], but other exons are rarely involved [131]. The role ofthe WT1 protein in hematopoiesis is not fully understood at this time; however, itis known to contain transcription regulatory domains as well as DNA-binding zincfinger domains. Mutations include small insertions and deletions primarily in exon7 and point mutations primarily in exon 9 [131–134]. In the majority of patientsit is a heterozygous mutation, but in some cases only the mutant allele is identi-fied most likely caused by copy number neutral loss of heterozygosity (acquireduniparental disomy). Mutations result in either loss of all or a portion of the DNA-binding domain or missense mutations. Multiple WT1 mutations are also found ina subset of patients [131–134]. Of note, WT1 mutations are associated with poorprognosis and, therefore, may help stratify patients with NK-AML [83, 132–134].Numerous papers have been written describing the use of WT1 expression levels byRT-PCR in monitoring AML.

Gene Expression and Prognosis in NK-AML

Attempts to further stratify prognosis in NK-AML are underway looking at lev-els of gene expression and their effects on prognosis. High expression of the


following genes has been shown to help further stratify these patients: brain andacute leukemia, cytoplasmic (BAALC), meningioma 1 (MN1), and ETS-related gene(ERG).

BAALC is a highly conserved gene normally expressed in mesodermal cells andin normal hematopoietic precursors [135–137]. Its expression is not seen in normalperipheral blood leukocytes and is only seen at low levels in normal bone marrowcells due to the relative paucity of the CD34+ stem cell compartment. Multiple pro-tein isoforms are expressed secondary to alternative splicing [135]. Overexpressionof BAALC has been shown to be associated with poor prognosis in NK-AML bymultiple groups [135, 138–140]. In addition, when evaluated in concert with FLT3-LM mutation status, and allele burden for FLT-LM (when positive), it allowedfurther stratification of both mutation-positive and mutation-negative patients [139,140]. High levels of expression of this gene are associated with other poor prognos-tic factors, including the presence of FLT3-LM, unmutated NPM1, mutated CEBPA,MLL-PTD, and high ERG expression [140].

The MN1 gene was first identified in a patient with meningioma [141]. It wassubsequently found to be involved in rare translocations seen in myeloid neoplasms[142]. High levels of expression of the MN1 gene have been shown to be associ-ated with unmutated NPM1 and to be an independent indicator of poor prognosisin NK-AML [143, 144]. High expression of ERG has likewise been shown to be anindependent indicator of poor prognosis in NK-AML [145, 146]. The authors pro-pose that level of expression of ERG can be used in addition to mutation status forFLT3-LM and NPM1 in patients with NK-AML.

Although expression of these three genes has been shown to have prognosticsignificance in the studies referenced above, it is premature to incorporate theminto routine clinical testing. In order to use gene expression levels to guide ther-apy outside of clinical trials a number of hurdles will need to be overcome. Testingwill have to be performed and reported in a uniform manner from laboratory tolaboratory, and an international standard will need to be available to ensure thatdifferent laboratories are performing the test accurately and that the results are com-parative from lab to lab. In addition, large, prospective, multicenter studies willneed to be performed to define appropriate cutoffs to predict good versus poorprognosis.

Mutations and Translocations Associated with Other MyeloidNeoplasms

Mutations and translocations primarily associated with myeloproliferative neo-plasms or myelodysplastic disorders can also be seen in de novo AML. This includesBCR/ABL1, JAK2, and MPL. Some of these cases may represent blast transfor-mation of previously undiagnosed myeloid neoplasms; however, others appear torepresent de novo disease [1, 147–152]. Molecular detection of these abnormalitiesis described in the chapter on myeloproliferative neoplasms.


Future Directions

As our knowledge and understanding of the underlying pathobiology of AML con-tinues to improve, so will the need for new molecular testing to identify and monitorthese abnormalities. Many of these tests will begin as laboratory-developed tests,whereas others may evolve as companion diagnostics as new drugs are developedfor targeted therapies. Although some suggested diagnostic algorithms have beenproposed [14, 153], substantial work needs to be done in developing guidelinesfor appropriate sensitivity, guidelines for therapy based upon a complex mixture ofprognostic markers, and uniformity in reporting and quantitation. Widely available,well-characterized positive controls and quantitation standards are needed to ensurequality and consistency from laboratory to laboratory. These are non-trivial prob-lems and attempts to solve them are ongoing. In addition, the technology continuesto evolve and in many cases become substantially cheaper. Techniques includ-ing gene expression arrays, whole genome sequencing, and studies involving SNParrays, epigenetic analysis, microRNAs, and proteomics are ongoing.

References

1. Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, Thiele J, Vardiman JW,eds. WHO classification of tumors of haematopoietic and lymphoid tissues. 4th ed. Lyon:IARC; 2008. World Health Organization Classification of Tumours.

2. Grimwade D, Lo Coco F. Acute promyelocytic leukemia: a model for the role of moleculardiagnosis and residual disease monitoring in directing treatment approach in acute myeloidleukemia. Leukemia. 2002;16(10):1959–1973.

3. Huret JL, Chomienne C. t(15;17)(q22;q21). Atlas Genet Cytogenet Oncol Haematol. April1998. URL: http://AtlasGeneticsOncology.org/Anomalies/t1517ID1035.html

4. Brown NJM, Ramalho M, Pedersen EW, Moravcsik E, Solomon E, Grimwade D PMLnuclear bodies in the pathogenesis of acute promyelocytic leukemia: active players orinnocent bystanders? Front Biosci. 2009;14:684–1707.

5. Lo Coco F, Ammatuna E, Sanz MA. Current treatment of acute promyelocytic leukemia.Haematologica. 2007;92(3):289–291.

6. Sanz MA, Grimwade D, Tallman MS, et al. Management of acute promyelocytic leukemia:recommendations from an expert panel on behalf of the European LeukemiaNet. Blood.2009;113(9):1875–1891.

7. Reiter A, Sauele S, Grimwade D, et al. Genomic anatomy of the specific reciprocaltranslocation t(15;17) in acute promyelocytic leukemia. Genes Chromosomes Cancer.2003;36(2):175–188.

8. Gabert J, Beillard E, van der Velden VHJ, et al.Standardization and quality control studiesof ‘real-time’ quantitative reverse transcriptase polymerase chain reaction of fusion genetranscripts for residual disease detection in leukemia –A Europe Against Cancer Program.Leukemia. 2003;17(12):2318–2357.

9. Guidez F, Ivins S, Zhu J, Söderström M, Waxman S, Zelent A. Reduced retinoicacid-sensitivities of nuclear receptor corepressor binding to PML- and PLZF-RARalphaunderlie molecular pathogenesis and treatment of acute promyelocytic leukemia. Blood.1998;91(8):2634–2642.

10. He LZ, Guidez F, Tribioli C, et al. Distinct interactions of PML-RARalpha and PLZF-RARalpha with co-repressors determine differential responses to RA in APL. Nat Genet.1998;18(2):126–135.


11. Lee S, Kim Y-J, Eom K-S, et al. The significance of minimal residual disease kineticsin adults with newly diagnosed PML-RARalpha-positive acute promyelocytic leukemia:results of a prospective trial Haematologica. 2006;91(5):671–674.

12. Santamaría C, Chillón MC, Fernández C, et al. Using quantification of the PML-RARtranscript to stratify the risk of relapse in patients with acute promyelocytic leukemia.Haematologica. 2007;92(3):315–322.

13. Speck NA, Gilliland DG. Core-binding factors in haematopoiesis and leukaemia. Nat. RevCancer. 2002;2(7):502–512.

14. Mrózek K, Marcucci G, Paschka P, Whitman SP, Bloomfield CD. Clinical relevance ofmutations and gene-expression changes in adult acute myeloid leukemia with normal cyto-genetics: are we ready for a prognostically prioritized molecular classification? Blood.2007;109(2):431–448.

15. Care RS, Valk PJM, Goodeve AC, et al. Incidence and prognosis of c-KIT and FLT3mutations in core binding factor (CBF) acute myeloid leukaemias. Br J Haematol.2003;121(5):775–777.

16. Paschka P, Marcucci G, Ruppert AS, et al. Adverse prognostic significance of KIT mutationsin adult acute myeloid leukemia with inv(16) and t(8;21): a Cancer and Leukemia GroupB Study J Clin Oncol. 2006;24(24):3904–3911.

17. Boissel N, Leroy H, Brethon B, et al. Incidence and prognostic impact of c-Kit, FLT3, andRas gene mutations in core binding factor acute myeloid leukemia (CBF-AML). Leukemia.2006;20(6):965–970.

18. Cairoli R, Beghini A, Grillo G, et al. Prognostic impact of c-KIT mutations in core bindingfactor leukemias: an Italian retrospective study. Blood. 2006;107(9):3463–3468.

19. Schnittger S, Kohl TM, Haferlach T, et al. KIT-D816 mutations in AML1-ETO-positive AML are associated with impaired event-free and overall survival. Blood.2006;107(5):1791–1799.

20. Müller AMS, Duque J, Shizuru JA, Lübbert M. Complementing mutations in core bind-ing factor leukemias: from mouse models to clinical applications. Oncogene. 2008;27(44):5759–5773.

21. Downing JR. The AML1-ETO chimaeric transcription factor in acute myeloid leukaemia:biology and clinical significance. Br J Haematol. 1999;106(2):296–308.

22. Kondo M, Kudo K, Kimura H, et al. Real-time quantitative reverse transcription-polymerasechain reaction for the detection of AML1-MTG8 fusion transcripts in t(8;21)-positive acutemyelogenous leukemia. Leuk Res. 2000;24(11):951–956.

23. Fujimaki S, Funato T, Harigae H, et al. A quantitative reverse transcriptase polymerase chainreaction method for the detection of leukaemic cells with t(8;21) in peripheral blood. EurJ Haematol. 2000;64(4):252–258.

24. Viehmann S, Teigler-Schlegel A, Bruch J, Langebrake C, Reinhardt D, Harbott J. Monitoringof minimal residual disease (MRD) by real-time quantitative reverse transcription PCR (RQ-RT-PCR) in childhood acute myeloid leukemia with AML1/ETO rearrangement. Leukemia.2003;17(6):1130–1136.

25. Takenokuchi M, Yasuda C, Takeuchi K, et al. Quantitative nested reverse transcriptasePCR vs. real-time PCR for measuring AML1/ETO (MTG8) transcripts. Clin Lab Haematol.2004;26(2):107–114.

26. Huret J-L. t(8;21)(q22;q22). Atlas Genet Cytogenet Oncol Haematol. 1997. http://AtlasGeneticsOncology.org/Anomalies/t0821.html. Accessed 2009.

27. Xiao Z, Greaves MF, Buffler P, et al. Molecular characterization of genomic AML1-ETOfusions in childhood leukemia. Leukemia. 2001;15(12):1906–1913.

28. Marcucci G, Caligiuri MA, Bloomfield CD. Core binding factor (CBF) acute myeloidleukemia: is molecular monitoring by RT-PCR useful clinically? Eur J Haematol.2003;71(3):143–154.

29. Shimada A, Taki T, Tabuchi K, et al. KIT mutations, and not FLT3 internal tandem dupli-cation, are strongly associated with a poor prognosis in pediatric acute myeloid leukemia


with t(8;21): a study of the Japanese Childhood AML Cooperative Study Group. Blood.2006;107(5):1806–1809.

30. Akagi T, Shih L-Y, Ogawa S, et al. Single nucleotide polymorphism genomic arrays analysisof t(8;21) acute myeloid leukemia cells Haematologica. 2009;94(9):1301–1306.

31. van der Reijden BA, Dauwerse HG, Giles RH, et al. Genomic acute myeloid leukemia-associated inv(16)(p13q22) breakpoints are tightly clustered. Oncogene. 1999;18(2):543–550.

32. Hajra A, Collins FS. Structure of the leukemia-associated human CBFB gene. Genomics.1995;26(3):571–579.

33. Liu PP, Hajra A, Wijmenga C, Collins FS. Molecular pathogenesis of the chromosome 16inversion in the M4Eo subtype of acute myeloid leukemia. Blood. 1995;85(9):2289–2302.

34. Huret J-L. inv(16)(p13q22); t(16;16)(p13;q22); del(16)(q22). Atlas Genet Cytogenet OncolHaematol. 1997. http://AtlasGeneticsOncology.org/Anomalies/inv16.html. Accessed 2009.

35. Huret J-L. inv(16)(p13;q22); t(16;16)(p13;q22); del(16)(q22). Atlas Genet Cytogenet OncolHaematol. 1999. http://AtlasGeneticsOncology.org/Anomalies/inv16.html.

36. Park TS, Lee ST, Song J, et al. Detection of a novel CBFB/MYH11 variant fusion transcript(K-type) showing partial insertion of exon 6 of CBFB gene using two commercially availablemultiplex RT-PCR kits. Cancer Genet Cytogenet. 2009;189(2):87–92.

37. Huret J-L. MYH11 (myosin heavy chain) (incomplete). Atlas Genet Cytogenet OncolHaematol. 1999. http://AtlasGeneticsOncology.org/Genes/MYH11ID43.html. Accessed2009.

38. Van der Reijden BA, de Wit L, van der Poel S, et al. Identification of a novel CBFβ-MYH11transcript: implications for RT-PCR diagnosis. Hematol J. 2001;2(3):206–209.

39. Kadkol SS, Bruno A, Dodge C, Lindgren V, Ravandi F. Comprehensive analysis of CBFB-MYH11 fusion transcripts in acute myeloid leukemia by RT-PCR analysis. J Mol Diagn.2004;6(1):22–27.

40. Mrozek K, Marcucci G, Paschka P, Bloomfield CD. Advances in molecular geneticsand treatment of core-binding factor acute myeloid leukemia. Curr Opin Oncol. 2008;20:711–718.

41. Marcucci G, Caligiuri MA, Döhner H, et al. Quantification of CBF/MYH11 fusion tran-script by real time RT-PCR in patients with INV(16) acute myeloid leukemia. Leukemia.2001;15(7):1072–1080.

42. Buonamici S, Ottaviani E, Testoni N, et al. Real-time quantitation of minimal residual dis-ease in inv(16)-positive acute myeloid leukemia may indicate risk for clinical relapse andmay identify patients in a curable state. Blood. 2002;99(2):443–449.

43. Krauter J, Görlich K, Ottmann O, et al. Prognostic value of minimal residual disease quan-tification by real-time reverse transcriptase polymerase chain reaction in patients with corebinding factor leukemias. J Clin Oncol. 2003;21(23):4413–4422.

44. Schnittger S, Weisser M, Schoch C, Hiddemann W, Haferlach T, Kern W. New score pre-dicting for prognosis in PML-RARA+, AML1-ETO+, or CBFBMYH11+ acute myeloidleukemia based on quantification of fusion transcripts. Blood. 2003;102(8):2746–2755.

45. Leroy H, de Botton S, Grardel-Duflos N, et al. Prognostic value of real-time quantitativePCR (RQ-PCR) in AML with t(8;21). Leukemia. 2005;19(3):367–372.

46. Stentoft J, Hokland P, Østergaard M, Hasle H, Nyvold CG. Minimal residual core bindingfactor AMLs by real time quantitative PCR—Initial response to chemotherapy predicts eventfree survival and close monitoring of peripheral blood unravels the kinetics of relapse. LeukRes. 2006;30(4):389–395.

47. Perea G, Lasa A, Aventín A, et al. Prognostic value of minimal residual disease (MRD) inacute myeloid leukemia (AML) with favorable cytogenetics [t(8;21) and inv(16)]. Leukemia.2006;20(1):87–94.

48. Weisser M, Haferlach C, Hiddemann W, Schnittger S. The quality of molecular responseto chemotherapy is predictive for the outcome of AML1-ETO-positive AML and isindependent of pretreatment risk factors. Leukemia. 2007;21(6):1177–1182.


49. Lane S, Saal R, Mollee P, et al. A ≥1 log rise in RQ-PCR transcript levels defines molecularrelapse in core binding factor acute myeloid leukemia and predicts subsequent morphologicrelapse. Leuk Lymphoma. 2008;49(3):517–523.

50. Hughes T, Deininger M, Hochhaus A, et al. Monitoring CML patients responding to treat-ment with tyrosine kinase inhibitors: review and recommendations for harmonizing currentmethodology for detecting BCR-ABL transcripts and kinase domain mutations and forexpressing results. Blood. 2006;108(1):28–37.

51. Meyer C, Kowarz E, Hofmann J, et al. Oncogenes, fusion genes and tumor suppressor genesnew insights to the MLL recombinome of acute leukemias. Leukemia. 2009;23(8):1490–1499.

52. Schnittger S, Kinkelin U, Schoch C, et al. Screening for MLL tandem duplication in 387 uns-elected patients with AML identify a prognostically unfavorable subset of AML. Leukemia.2000;14(5):796–804.

53. Poppe B, Vandesompele J, Schoch C, et al. Expression analyses identify MLL as a prominenttarget of 11q23 amplification and support an etiologic role for MLL gain of function inmyeloid malignancies. Blood. 2004;103(1):229–235.

54. Arnaud B, Douet-Guilbert N, Morel F, et al. Screening by fluorescence in situ hybridizationfor MLL status at diagnosis in 239 unselected patients with acute myeloblastic leukemia.Cancer Genet Cytogenet. 2005;161(2):110–115.

55. Weisser M, Kern W, Schoch C, Hiddemann W, Haferlach T, Schnittger S. Risk assessmentby monitoring expression levels of partial tandem duplications in the MLL gene in acutemyeloid leukemia during therapy. ©2005 Ferrata Storti Foundation. 2005;90(7):881–889.

56. Basecke J, Whelan JT, Griesinger F, Bertrand FE. The MLL partial tandem duplication inacute myeloid leukaemia. Br J Haematol. 2006;135(4):438–449.

57. Shih L-Y, Liang D-C, Fu J-F, et al. Characterization of fusion partner genes in 114 patientswith de novo acute myeloid leukemia and MLL rearrangement. Leukemia. 2006;20(2):218–223.

58. Schoch C, Schnittger S, Klaus M, Kern W, Hiddemann W, Haferlach T. AML with11q23/MLL abnormalities as defined by the WHO classification: incidence, partner chro-mosomes, FAB subtype, age distribution, and prognostic impact in an unselected series of1897 cytogenetically analyzed AML cases. Blood. 2003;102(7):2395–2402.

59. Chowdhury T, Brady HJM. Insights from clinical studies into the role of the MLL gene ininfant and childhood leukemia. Blood Cell Mol Dis. 2008;40(2):192–199.

60. Balgobind BV, Raimondi SC, Harbott J, et al. Novel prognostic subgroups in childhood11q23/MLL-rearranged acute myeloid leukemia: results of an international retrospectivestudy. Blood. 2009;114(12):2489–2496.

61. Schlenk RF, Döhner K, Krauter J, et al. Mutations and treatment outcome in cytogeneticallynormal acute myeloid leukemia. N Engl J Med. 2008;358(18):1909–1918.

62. Mrozek K, Heinonen K, Lawrence D, et al. Adult patients with de novo acute myeloidleukemia and t(9;11)(p22;q23) have a superior outcome to patients with other transloca-tions involving band 11q23: a cancer and leukemia group B study. Blood. 1997;90(11):4532–4538.

63. Byrd JC, Mrozek K, Dodge RK, et al. Pretreatment cytogenetic abnormalities are predictiveof induction success, cumulative incidence of relapse, and overall survival in adult patientswith de novo acute myeloid leukemia: results from Cancer and Leukemia Group B (CALGB8461). Blood. 2002;100(13):4325–4336.

64. Rubnitz JE, Raimondi SC, Tong X, et al. Favorable impact of the t(9;11) in childhood acutemyeloid leukemia. J Clin Oncology. 2002;20(9):2302–2309.

65. Palle J, Frost BM, Forestier E, et al. Cellular drug sensitivity in MLL-rearranged child-hood acute leukaemia is correlated to partner genes and cell lineage. Br J Haematol.2005;129(2):189–198.

66. Krauter J, Wagner K, Schäfer I, et al. Prognostic factors in adult patients up to 60 yearsold with acute myeloid leukemia and translocations of chromosome band 11q23: individual


patient data–based meta-analysis of the german acute myeloid leukemia intergroup. J ClinOncol. 2009;27(18):3000–3006.

67. Welborn JL, Jenks HM, Hagemeijer A. Unique clinical features and prognostic significanceof the translocation (6;11) in acute leukemia. Cancer Genet Cytogenet. 1993;65(2):125–129.

68. Huret J-L. MLL (myeloid/lymphoid or mixed lineage leukemia). Atlas Genet CytogenetOncol Haematol. 1997;2009. http://AtlasGeneticsOncology.org/Genes/MLL.html. Accessed2009.

69. Hess JLaH JL. MLL (myeloid/lymphoid or mixed lineage leukemia). Atlas GenetCytogenet Oncol Haematol. 2000. http://AtlasGeneticsOncology.org/Genes/MLL.html.Accessed 2009.

70. Marschalek R. MLL (myeloid/lymphoid or mixed lineage leukemia). Atlas Genet CytogenetOncol Haematol. 2002:11.

71. Huret J-L. MLL (myeloid/lymphoid or mixed lineage leukemia). Atlas Genet CytogenetOncol Haematol. 2005. http://AtlasGeneticsOncology.org/Genes/MLL.html. Accessed2009.

72. Slany RK. The molecular biology of mixed lineage leukemia. Haematologica.2009;94(7):984–993.

73. Harper DP, Aplan PD. Chromosomal rearrangements leading to mll gene fusions: clinicaland biological aspects. Cancer Res. 2008;68(24):10024–10027.

74. Scholl C, Schlenk RF, Eiwen K, Döhner H, Fröhling S, Döhner K. The prognostic valueof MLL-AF9 detection in patients with t(9;11)(p22;q23)-positive acute myeloid leukemia.©2005 Ferrata Storti Foundation. 2005;90(12):1626–1634.

75. von Lindern M, Fornerod M, Soekarman N, et al. Translocation t(6;9) in acute non-lymphocytic leukaemia results in the formation of a DEK-CAN fusion gene. Baillieres ClinHaematol. 1992;5(4):857–879.

76. Soekarman D, von Lindern M, Daenen S, et al. The translocation (6;9) (p23;q34) shows con-sistent rearrangement of two genes and defines a myeloproliferative disorder with specificclinical features. Blood. 1992;79(11):2990–2997.

77. Tobal K, Frost L, Liu Yin JA. Quantification of DEK-CAN fusion transcript by real-timereverse transcription polymerase reaction in patients with t(6;9) acute myeloid leukemia.Haematologica. 2004;89(10):1267–1269.

78. Martinelli G, Ottaviani E, Buonamici S, et al. Association of 3q21q26 syndrome with dif-ferent RPN1/EVI1 fusion transcripts. Haematologica/ J Hematol. 2003;88(11):1221–1228.

79. Carroll A, Civin C, Schneider N, et al. The t(1;22) (p13;q13) is nonrandom and restrictedto infants with acute megakaryoblastic leukemia: a Pediatric Oncology Group Study. Blood.1991;78(3):748–752.

80. Lion T, Haas OA. Acute megakaryocytic leukemia with the t(1;22)(p13;q13). LeukLymphoma. 1993;11(1):15–20.

81. Hsiao H-H, Yang M-Y, Liu Y-C, et al. RBM15-MKL1 (OTT-MAL) fusion transcript in anadult acute myeloid leukemia patient. Am J Hematol. 2005;79(1):43–45.

82. Schneider F, Hoster E, Unterhalt M, et al. NPM1 but not FLT3-ITD mutations predict earlyblast cell clearance and CR rate in patients with normal karyotype AML (NK-AML) orhigh-risk myelodysplastic syndrome (MDS). Blood. 2009;113(21):5250–5253.

83. Lapillonne H, Llopis L, Auvrignon A, et al. Extensive mutational status of genes and clinicaloutcome in pediatric acute myeloid leukemia. Leukemia. 2009;24:205–209.

84. Schnittger S, Schoch C, Kern W, et al. Nucleophosmin gene mutations are predictorsof favorable prognosis in acute myelogenous leukemia with a normal karyotype. Blood.2005;106(12):3733–3739.

85. Dohner K, Schlenk RF, Habdank M, et al. Mutant nucleophosmin (NPM1) predicts favor-able prognosis in younger adults with acute myeloid leukemia and normal cytogenetics:interaction with other gene mutations. Blood. 2005;106(12):3740–3746.

86. Verhaak RGW, Goudswaard CS, van Putten W, et al. Mutations in nucleophosmin (NPM1) inacute myeloid leukemia (AML): association with other gene abnormalities and previously


established gene expression signatures and their favorable prognostic significance. Blood.2005;106(12):3747–3754.

87. Suzuki T, Kiyoi H, Ozeki K, et al. Clinical characteristics and prognostic implications ofNPM1 mutations in acute myeloid leukemia. Blood. 2005;106(8):2854–2861.

88. Boissel N, Renneville A, Biggio V, et al. Prevalence, clinical profile, and prognosis of NPMmutations in AML with normal karyotype. Blood. 2005;106(10):3618–3620.

89. Mariano AR, Colombo E, Luzi L, et al. Cytoplasmic localization of NPM in myeloidleukemias is dictated by gain-of-function mutations that create a functional nuclear exportsignal. Oncogene. 2006;25(31):4376–4380.

90. Albiero E, Madeo D, Bolli N, et al. Identification and functional characterization of a cyto-plasmic nucleophosmin leukaemic mutant generated by a novel exon-11 NPM1 mutation.Leukemia. 2007;21(5):1099–1103.

91. Pitiot AS, Santamaría I, García-Suárez O, et al. A new type of NPM1 gene mutation in AMLleading to a C-terminal truncated protein. Leukemia. 2007;21(7):1564–1566.

92. Gorello P, Cazzaniga G, Alberti F, et al. Quantitative assessment of minimal residual dis-ease in acute myeloid leukemia carrying nucleophosmin (NPM1) gene mutations. Leukemia.2006;20(6):1103–1108.

93. Schnittger S, Kern W, Tschulik C, et al. Minimal residual disease levels assessed by NPM1mutation-specific RQ-PCR provide important prognostic information in AML. Blood.2009;114(11):2220–2231.

94. Noguera NI, Ammatuna E, Zangrilli D, et al. Simultaneous detection of NPM1 andFLT3-ITD mutations by capillary electrophoresis in acute myeloid leukemia. Leukemia.2005;19(8):1479–1482.

95. Papadaki C, Dufour A, Seibl M, et al. Monitoring minimal residual disease in acute myeloidleukaemia with NPM1 mutations by quantitative PCR: clonal evolution is a limiting factorBr J Haematol. 2009;144(4):517–523.

96. Chou WC, Tang JL, Wu SJ, et al. Clinical implications of minimal residual disease moni-toring by quantitative polymerase chain reaction in acute myeloid leukemia patients bearingnucleophosmin (NPM1) mutations. Leukemia. 2007;21 (5):998–1004.

97. Leroy H, Roumier C, Huyghe P, Biggio V, Fenaux P, Preudhomme C. CEBPA pointmutations in hematological malignancies. Leukemia. 2005;19:329–334.

98. Mueller BU, Pabst T. C/EBP[alpha] and the pathophysiology of acute myeloid leukemia.Curr Opin Hematol. 2006;13(1):7–14.

99. Ho PA, Alonzo TA, Gerbing RB, et al. Prevalence and prognostic implications of CEBPAmutations in pediatric acute myeloid leukemia (AML): a report from the Children’sOncology Group. Blood. 2009;113(26):6558–6566.

100. Lin L-I, Chen C-Y, Lin D-T, et al. Characterization of CEBPA mutations in acute myeloidleukemia: most patients with CEBPA mutations have biallelic mutations and show a distinctimmunophenotype of the leukemic cells. Clin Cancer Res. 2005;11(4):1372–1379.

101. Shih LY, Liang DC, Huang CF, et al. AML patients with CEBP[alpha] mutations mostlyretain identical mutant patterns but frequently change in allelic distribution at relapse:a comparative analysis on paired diagnosis and relapse samples. Leukemia. 2006;20(4):604–609.

102. Lin L-I, Lin T-C, Chou W-C, Tang J-L, Lin D-T, Tien H-F. A novel fluorescence-based mul-tiplex PCR assay for rapid simultaneous detection of CEBPA mutations and NPM mutationsin patients with acute myeloid leukemias. Leukemia. doi:10.1038/sj.leu.2404331; publishedonline 27 July 2006; 2006;20:1899–1903.

103. Benthaus T, Schneider F, Mellert G, et al. Rapid and sensitive screening for CEBPAmutations in acute myeloid leukaemia. Br J Haematol. 2008;143(2):230–239.

104. Charrin C. del(5q) in myeloid malignancies. Atlas Genet Cytogenet Oncol Haematol. 1998.http://AtlasGeneticsOncology.org/Genes/del5qID1092.html. Accessed 2009.

105. Desangles F. -7/del(7q) in adults. Atlas Genet Cytogenet Oncol Haematol. 1999.http://AtlasGeneticsOncology.org/Anomalies/del7qID1093.html. Accessed 2009.


106. Liu TX, Becker MW, Jelinek J, et al. Chromosome 5q deletion and epigenetic suppres-sion of the gene encoding -catenin (CTNNA1) in myeloid cell transformation Nat Med.2007;13(1):78–83.

107. Ebert BL, Pretz J, Bosco J, et al. Identification of RPS14 as a 5q- syndrome gene by RNAinterference screen. Nature. 2008;451:335–339.

108. Eisenmann KM, Dykema KJ, Matheson SF, et al. 5q- myelodysplastic syndromes: chro-mosome 5q genes direct a tumor-suppression network sensing actin dynamics. Oncogene.2009;28(39):3429–3441.

109. Nishino HT, Chang C-C. Myelodysplastic syndromes: clinicopathologic features,pathobiology, and molecular pathogenesis. Arch Pathol Lab Med. 2005;129(10):299–1310.

110. Van den Berghe H, Michaux L. 5q-, twenty-five years later: a synopsis. Cancer GenetCytogenet. 1997;94(1):1–7.

111. Horrigan SK, Arbieva ZH, Xie HY, et al. Delineation of a minimal interval and identificationof 9 candidates for a tumor suppressor gene in malignant myeloid disorders on 5q31. Blood.2000;95(7):2372–2377.

112. Emerling BM, Bonifas J, Kratz CP, et al. MLL5, a homolog of Drosophila trithorax locatedwithin a segment of chromosome band 7q22 implicated in myeloid leukemia. Oncogene.2002;21(31):4849–4854.

113. Tosi S, Scherer SW, Giudici G, Czepulkowski B, Biondi A, Kearney L. Delineationof multiple deleted regions in 7q in myeloid disorders. Genes Chromosomes Cancer.1999;25(4):384–392.

114. Yokouchi M, Suzuki R, Masuhara M, Komiya S, Inoue A, Yoshimura A. Cloning and char-acterization of APS, an adaptor molecule containing PH and SH2 domains that is tyrosinephosphorylated upon B-cell receptor stimulation. Oncogene. 1997;15(1):7–15.

115. Koike M, Tasaka T, Spira S, Tsuruoka N, Koeffler HP. Allelotyping of acute myelogenousleukemia: loss of heterozygosity at 7q31.1 (D7S486) and q33-34 (D7S498, D7S505). LeukRes. 1999;23(3):307–310.

116. Brezinová J, Zemanová Z, Ransdorfová Š, et al. Structural aberrations of chromosome 7revealed by a combination of molecular cytogenetic techniques in myeloid malignancies.Cancer Genet Cytogenet. 2007;173(1):10–16.

117. Nakao M, Yokota S, Iwai T, et al. Internal tandem duplication of the FLT3 gene found inacute myeloid leukemia. Leukemia. 1996;10(12):1911–1918.

118. Thiede C, Steudel C, Mohr B, et al. Analysis of FLT3-activating mutations in 979 patientswith acute myelogenous leukemia: association with FAB subtypes and identification ofsubgroups with poor prognosis. Blood. 2002;99(12):4326–4335.

119. Gilliland DG, Griffin JD. The roles of FLT3 in hematopoiesis and leukemia. Blood.2002;100(5):1532–1542.

120. Small D. Targeting FLT3 for the treatment of leukemia. Semin Hematol. 2008;45(3 Suppl 2):S17–S21.

121. Yamamoto Y, Kiyoi H, Nakano Y, et al. Activating mutation of D835 within theactivation loop of FLT3 in human hematologic malignancies. Blood. 2001;97(8):2434–2439.

122. Bacher U, Haferlach C, Kern W, Haferlach T, Schnittger S. Prognostic relevance of FLT3-TKD mutations in AML: the combination matters an analysis of 3082 patients. Blood.2008;111(5):2527–2537.

123. Akagi T, Shih L-Y, Kato M, et al. Hidden abnormalities and novel classification oft(15;17) acute promyelocytic leukemia (APL) based on genomic alterations. Blood.2009;113(8):1741–1748.

124. Schnittger S, Schoch C, Dugas M, et al. Analysis of FLT3 length mutations in 1003 patientswith acute myeloid leukemia: correlation to cytogenetics, FAB subtype, and prognosis inthe AMLCG study and usefulness as a marker for the detection of minimal residual disease.Blood. 2002;100(1):59–66.


125. Schnittger S, Schoch C, Kern W, Hiddemann W, Haferlach T. FLT3 length mutations asmarker for follow-up studies in acute myeloid leukaemia. Acta Haematologia. 2004;112(1–2):68–78.

126. Cloos J, Goemans BF, Hess CJ, et al. Stability and prognostic influence of FLT3 mutationsin paired initial and relapsed AML samples. Leukemia. 2006;20(7):1217–1220.

127. Kottaridis PD, Gale RE, Langabeer SE, Frew ME, Bowen DT, Linch DC. Studies of FLT3mutations in paired presentation and relapse samples from patients with acute myeloidleukemia: implications for the role of FLT3 mutations in leukemogenesis, minimal resid-ual disease detection, and possible therapy with FLT3 inhibitors. Blood. 2002;100(7):2393–2398.

128. Murphy KM, Levis M, Hafez MJ, et al. Detection of FLT3 internal tandem duplicationand D835 mutations by a multiplex polymerase chain reaction and capillary electrophoresisassay. J Mol Diagn. 2003;5(2):96–102.

129. Gale RE, Green C, Allen C, et al. The impact of FLT3 internal tandem duplication mutantlevel, number, size, and interaction with NPM1 mutations in a large cohort of young adultpatients with acute myeloid leukemia. Blood. 2008;111(5):2776–2784.

130. Meshinchi S, Stirewalt DL, Alonzo TA, et al. Structural and numerical variation ofFLT3/ITD in pediatric AML. Blood. 2008;111(10):4930–4933.

131. King-Underwood L, Pritchard-Jones K. Wilms’ Tumor (WT1) Gene mutations occurmainly in acute myeloid leukemia and may confer drug resistance. Blood. 1998;91(8):2961–2968.

132. Summers K, Stevens J, Kakkas I, et al. Wilms’ tumour 1 mutations are associated with FLT3-ITD and failure of standard induction chemotherapy in patients with normal karyotype AML.Leukemia. 2007;21(3):550–551.

133. Virappane P, Gale R, Hills R, et al. Mutation of the Wilms’ tumor 1 gene is a poor prog-nostic factor associated with chemotherapy resistance in normal karyotype acute myeloidleukemia: The United Kingdom Medical Research Council Adult Leukaemia Working Party.J Clin Oncol. 2008;26(33):5429–5435.

134. Paschka P, Marcucci G, Ruppert AS, et al. Wilms’ tumor 1 gene mutations indepen-dently predict poor outcome in adults with cytogenetically normal acute myeloid leukemia:A Cancer and Leukemia Group B Study. J Clin Oncol. 2008;26(28):4595–4602.

135. Tanner SM, Austin JL, Leone G, et al. BAALC, the human member of a novel mammalianneuroectoderm gene lineage, is implicated in hematopoiesis and acute leukemia. Proc NatlAcad of Sci U S A. 2001;98(24):13901–13906.

136. Baldus CD, Tanner SM, Kusewitt DF, et al. BAALC, a novel marker of human hematopoieticprogenitor cells. Exp. Hematol. 2003; 31(1):1051–1056.

137. Satoskar AA, Tanner SM, Weinstein M, Qualman SJ, de la Chapelle A. Baalc, a marker ofmesoderm and muscle. Gene Expr Patterns. 2005;5(4):463–473.

138. Baldus CD, Tanner SM, Ruppert AS, et al. BAALC expression predicts clinical outcome ofde novo acute myeloid leukemia patients with normal cytogenetics: a Cancer and LeukemiaGroup B Study. Blood. 2003;102(5):1613–1618.

139. Baldus CD, Thiede C, Soucek S, Bloomfield CD, Thiel E, Ehninger G. BAALC expressionand FLT3 internal tandem duplication mutations in acute myeloid leukemia patients withnormal cytogenetics: prognostic implications. J Clin Oncol. 2006;24(5):790–797.

140. Langer C, Radmacher MD, Ruppert AS, et al. High BAALC expression associates withother molecular prognostic markers, poor outcome, and a distinct gene-expression signaturein cytogenetically normal patients younger than 60 years with acute myeloid leukemia: aCancer and Leukemia Group B (CALGB) study. Blood. 2008;111(11):5371–5379.

141. Lekanne Deprez RH, Riegman PH, Groen NA, et al. Cloning and characterization of MN1,a gene from chromosome 22q11, which is disrupted by a balanced translocation in ameningioma. Oncogene. 1995;10(8):1521–1528.

142. Grosveld GC. MN1, a novel player in human AML. Blood Cell Mol Dis. 2007;39(3):336–339.


143. Heuser M, Beutel G, Krauter J, et al. High meningioma 1 (MN1) expression as a pre-dictor for poor outcome in acute myeloid leukemia with normal cytogenetics. Blood.2006;108(12):3898–3905.

144. Langer C, Marcucci G, Holland KB, et al. Prognostic importance of MN1 transcript levels,and biologic insights from mn1-associated gene and MicroRNA expression signatures incytogenetically normal acute myeloid leukemia: a Cancer and Leukemia Group B Study.J Clin Oncol. 2009;27(19):3198–3204.

145. Marcucci G, Baldus CD, Ruppert AS, et al. Overexpression of the ETS-related gene, erg,predicts a worse outcome in acute myeloid leukemia with normal karyotype: a Cancer andLeukemia Group B Study. J Clin Oncol. 2005;23(36):9234–9242.

146. Marcucci G, Maharry K, Whitman SP, et al. High expression levels of the ETS-related gene,erg, predict adverse outcome and improve molecular risk-based classification of cytogeneti-cally normal acute myeloid leukemia: a Cancer and Leukemia Group B Study. J Clin Oncol.2007;25(22):3337–3343.

147. Berger R, Chen SJ, Chen Z. Philadelphia-positive acute leukemia, cytogenetic and molecularaspects. Cancer Genet Cytogenet. 1990;44:143–152.

148. Pardanani AD, Levine RL, Lasho T, et al. MPL515 mutations in myeloproliferative and othermyeloid disorders: a study of 1182 patients. Blood. 2006;108(10):3472–3476.

149. Illmer T, Schaich M, Ehninger G, Thiede C. Tyrosine kinase mutations of JAK2 are rareevents in AML but influence prognosis of patients with CBF-leukemias. Haematologica.2007;92(1):137–138.

150. Schnittger S, Bacher U, Kern W, Haferlach T, Haferlach C. JAK2V617F as progressionmarker in CMPD and as cooperative mutation in AML with trisomy 8 and t(8;21): acomparative study on 1103 CMPD and 269 AML cases. Leukemia. 2007;21(8):1843–1845.

151. Nishii K, Nanbu R, Lorenzo VF, et al. Expression of the JAK2 V617F mutation is not foundin de novo AML and MDS but is detected in MDS-derived leukemia of megakaryoblasticnature. Leukemia. 2007;21(6):1337–1338.

152. Hussein K, Bock O, Theophile K, et al. MPLW515L mutation in acute megakaryoblasticleukaemia. Leukemia. 2009;23(5):852–855.

153. Scholl S, Fricke H-J, Sayer HG, Höffken K. Clinical implications of molecular geneticaberrations in acute myeloid leukemia. J Cancer Res Clin Oncol. 2009;135(4):491–505.

Chapter 6Molecular Pathology of Mature B-Celland T-Cell Lymphomas

Sophia L. Yohe, David W. Bahler, and Marsha C. Kinney

Keywords B-cell maturation · Immunoglobulin heavy chain (IgH) generearrangement · Kappa immunoglobulin light chain gene rearrangement · Lambdaimmunoglobulin light chain gene rearrangement · Class (isotype) switch · Somatichypermutation · Clonality testing · B-cell · Follicular lymphoma · B-celllymphoma/CLL 2 (BCL2) translocation –t(14;18) · Mantle cell lymphoma · CyclinD1 translocation –t(11;14) · Diffuse large B-cell lymphoma (DLBCL) · Germinalcenter type · Activated B-cell type · B-cell lymphoma/CLL 6 (BCL6) · Marginalzone lymphoma · Mucosal associated · t(11;18) · Nodal marginal zonelymphoma · Splenic marginal zone lymphoma · Burkitt lymphoma · MYCbreakpoints · t(8:14) · Lymphomas intermediate between Burkitt andDLBCL · Lymphoplasmacytic lymphoma · T-cell maturation · T-cell receptor(TCR) alpha gene rearrangement · T-cell receptor (TCR) beta gene rear-rangement · T-cell receptor (TCR) delta gene rearrangement · T-cell receptor(TCR) gamma gene rearrangement · Clonality testing · T-cell · Clonal peakheight · Flow cytometry · Vbeta · Anaplastic large cell lymphoma · ALK-NPM translocation –t(2;5) · ALK signaling pathway · AngioimmunoblasticT-cell lymphoma · Mycosis fungoides · Sezary syndrome · Hepatosplenic T-celllymphoma · Enteropathy associated T-cell lymphoma · Extranodal NK/T-celllymphoma · Peripheral T-cell lymphoma NOS · Nodular lymphocyte predomi-nant Hodgkin lymphoma · Aberrant somatic hypermutation · Classical Hodgkinlymphoma · Southern blot · Polymerase chain reaction (PCR) – Fluorescentin-situ hybridization (FISH) · Immunohistochemical stains · Comparative genomichybridization · Gene expression profiling

S.L. Yohe (B)Department of Laboratory Medicine and Pathology, University of Minnesota Medical Center,Fairview, Mayo Room D219-7, 420 Delaware St. SE, Minneapolis, MN 55455, USAe-mail: [email protected]


158 S.L. Yohe et al.

Introduction

Over the last two decades, molecular genetic testing has assumed a prominent rolein the diagnosis, classification, and clinical management of lymphoma [1]. In somelymphoid and more so in myeloid tumors, a specific genetic abnormality may be thedefining feature. More often genetic abnormalities are characteristic of one type ofdisease but may be present in other tumors and contribute to their pathogenesis. Inaddition to the standard assays for receptor gene rearrangements and specific geneticlesions (particularly translocations) used in clinical practice, sophisticated assayssuch as expression microarray, array comparative genomic hybridization (CGH),microRNA analysis, and epigenetic testing are more readily available; data derivedfrom these complex analyses are continuously being translated into clinically perti-nent information regarding pathogenesis, diagnosis, prognosis, and targeted therapyfor lymphoma. This chapter will focus on the molecular pathogenesis of lymphomaand discuss current clinical molecular diagnostic testing and its limitations andemphasize interpretation of molecular results in the context of clinical features,morphology, or other studies.

Molecular Testing for B-Cell Non-HodgkinLymphoma (B-NHL)

B-Cell Biology and Maturation

Events that occur during B-cell development play an important role in moleculartesting for B-cell lymphomas. Furthermore, many B-NHLs correspond to differentstages of B-cell development; therefore an understanding of B-cell development iscritical in the diagnosis of B-NHL (Table 6.1).

B cells are part of the adaptive immune system, producing antibodies againstvarious antigens. Given the large number of antigens and unpredictable exposure indifferent individuals, mechanisms must be in place to create a wide range of antibod-ies (an antibody repertoire) from a limited number of genes. The first step in creatingthis diversity is rearrangement of the heavy- and light-chain immunoglobulin genes,which occurs in B lymphoblasts residing in the bone marrow. Rearrangement ofthe immunoglobulin heavy-chain gene (IGH or H) occurs first followed by rear-rangement of the kappa (κ) and lambda (λ) immunoglobulin light-chain genes (L).Additional diversity is created by the addition and subtraction of nucleotides atthe sites of rearrangement, pairing of different heavy and light chains, and somatichypermutation. These differences are responsible for the polyclonal population of Bcells in normal individuals.

The IGH gene on chromosome 14q32 is composed of 40–52 functional variable(V), 25 diversity (D), six joining (J), and five constant (C) segments [2, 3] as shownin Fig. 6.1.

6 Molecular Pathology of Mature B-Cell and T-Cell Lymphomas 159

Tabl

e6.

1O

verv

iew

ofB

-cel

ldev

elop

men

tsho

win

gth

etim

ing

ofim

mun

oglo

bulin

gene

rear

rang

emen

tand

som

atic

hype

rmut

atio

n(S

HM

)

Loc

atio

nC

ell

IgH

SHM

κ/λ

Imm

unop

heno

type

Neo

plas

ticco

unte

rpar

t

Bon

em

arro

wE

arly

pro-

Bce

llD

H–J

HG

erm

line

Ger

mlin

eT

dT,C

D34

,CD

10,

CD

19,C

D38

AL

L

Lat

epr

o-B

cell

VH

–DH

J HG

erm

line

Ger

mlin

eT

dT,C

D10

,CD

19,C

D38

AL

LPr

e-B

cell

Rea

rran

ged

intr

acel

lula

rμ

Ger

mlin

eV

L–J

LC

D19

,CD

20,C

D38

,Les

sT

dT,v

aria

ble

CD

20A

LL

Imm

atur

eB

cell

Rea

rran

ged

surf

ace

IgM

Ger

mlin

eR

earr

ange

dC

D19

,CD

20

Lym

phoi

dtis

sue

Man

tlezo

neN

aïve

Bce

llR

earr

ange

dsI

gM/s

IgD

Ger

mlin

e(6

0–85

%)

orlo

w

Rea

rran

ged

CD

19,C

D20

,BC

L2

MC

LC

LL

/SL

L

Ger

min

alce

nter

Cen

troc

yte

Rea

rran

ged

surf

ace

Ong

oing

Rea

rran

ged

surf

ace

CD

10,B

CL

6,C

D19

,C

D20

FL

GC

Bce

llR

earr

ange

dsu

rfac

eH

igh

Rea

rran

ged

surf

ace

CD

30,C

D15

,C

HL

Cen

trob

last

Rea

rran

ged

surf

ace

Ong

oing

Rea

rran

ged

surf

ace

CD

19,C

D20

,BC

L6

NL

PHL

GC

Bce

llR

earr

ange

dsu

rfac

eH

igh

and

aber

rant

Rea

rran

ged

surf

ace

CD

10,B

CL

6,C

D19

,C

D20

DL

BC

L(G

Cty

pe)

GC

orpo

st-G

CB

cell

Rea

rran

ged

surf

ace

Hig

hR

earr

ange

dsu

rfac

eC

D10

,hig

hK

i67

BL

Post

-ge

rmin

alce

nter

Act

ivat

edpe

riph

eral

Bce

llR

earr

ange

dsu

rfac

eH

igh

Rea

rran

ged

surf

ace

MU

M1

No

CD

10D

LB

CL

(AB

Cty

pe)

Mem

ory

Bce

llR

earr

ange

dsu

rfac

eH

igh

Rea

rran

ged

surf

ace

No

CD

5or

CD

10M

ZL

Post

-fol

licul

arB

cell

Rea

rran

ged

surf

ace

and

cyto

plas

mic

Hig

hR

earr

ange

dsu

rfac

ean

dcy

topl

asm

icN

oC

D5

orC

D10

LPL

Plas

ma

cell

Rea

rran

ged

cyto

plas

mic

Cla

sssw

itch

Hig

hR

earr

ange

dcy

topl

asm

icC

D13

8,C

D38

,MU

M1

PCN

The

cell

ofor

igin

and

itsne

opla

stic

coun

terp

art

are

nota

ted.

Seve

ral

lym

phom

asm

ayar

ise

from

mor

eth

anon

ece

llof

orig

in.A

LL

,acu

tely

mph

obla

stic

lym

phom

a;M

CL

,man

tlece

llly

mph

oma;

FL:f

ollic

ular

lym

phom

a;C

HL

,cla

ssic

alH

odgk

inly

mph

oma;

NL

PHL

,nod

ular

lym

phoc

yte-

pred

omin

antH

odgk

inly

mph

oma;

DL

BC

L,

diff

use

larg

eB

-cel

llym

phom

a;B

L,B

urki

ttly

mph

oma;

MZ

L,m

argi

nal-

zone

lym

phom

a;L

PL,l

ymph

opla

smac

ytic

lym

phom

a;PC

N,p

lasm

ace

llne

opla

sm


Fig. 6.1 Immunoglobulin heavy-chain gene rearrangement. DH–JH rearrangement occurs first fol-lowed by VH–DHJH rearrangement. There are 40–52 functional VH regions. Non-template (N) andpalindromic (P) nucleotides are added at the joins by terminal deoxynucleotidyl transferase (TdT)and recombination-activating gene (RAG) proteins, respectively. After rearrangement a single V,D, and J region is present. PCR primers are directed toward the framework regions (FRs) whichare more conserved between V regions than the complementarity-determining regions (CDRs).Forward primers are directed toward FR I, FR II, and/or FR III in the VH segment, while thereverse primer is directed toward FR IV in the JH segment. L refers to a leader sequence

Ultimately a single V, D, J, and C regions are joined to each other with the inter-vening gene sequences removed. Not all recombinations create a functional proteinas a stop codon may be created or rearrangement may occur with a pseudogene(especially in the variable region). Rearrangement starts on the IGH gene in earlypro-B cells with a DH to JH segment joining [2]. This rearrangement occurs onboth alleles and is usually successful (rarely produces a stop codon) due to themakeup of the DH segment [2]. The next step is VH to DHJH rearrangement inlate pro-B cells (Fig. 6.1). Recombination at this step is less often successful andoccurs in one allele at a time [2]. The second allele will rearrange only if recom-bination is unsuccessful on the first and both alleles can join different remainingVH segments. Despite the option of multiple rearrangements on one allele and sub-sequently on the other, at least 45% of pro-B cells do not successfully completethis step and are lost [2]. Pro-B cells that complete this step produce mu (μ) heavychains and become pre-B cells. Successful rearrangement is signaled by a receptorcomposed of the newly formed mu heavy chain, CD79A, CD79B, and surrogatelight chains encoded by non-rearranged genes [2]. Theoretically VH–DH–JH rear-rangements can lead to ∼6,000 different combinations in the heavy-chain gene [2].Additional diversity is created by the addition and subtraction of nucleotides at thesites of recombination by the action of terminal deoxynucleotidyl transferase (TdT)and various other enzymes. TdT, which is highly expressed in pro-B cells, adds upto 20 random or non-template nucleotides (N nucleotides) at the DH–JH and VH-DH


joins [2]. Palindromic sequences of nucleotides (P nucleotides) are added by therecombination-activating gene (RAG) proteins [2]. Various DNA repair enzymes areresponsible for the removal of nucleotides. The IGH gene now has a VDJ composedof four framework regions (FRs) and three complementarity-determining regions(CDRs) as shown in Fig. 6.1. FR I, FR II, and FR III are located in the VH region,while FR IV is in the JH region. FRs are similar in all VH segments and undergoless somatic hypermutation than do CDRs. CDRs are the antigen-binding areas, arevariable from one VH segment to another, and are prone to somatic hypermutation.

Pre-B cells proliferate prior to recombining their light-chain genes, thereforeseveral cells with the same IGH rearrangement will pair with different kappa orlambda light chains. The light-chain genes lack a diversity segment and have Nnucleotides added only about 25% of the time due to lower expression of TdTin pre-B cells [2] as shown in Fig. 6.2a. The kappa light-chain gene (IGK) onchromosome 2p11 rearranges first >90–95% of the time (Fig. 6.2a). One of thevariable segments recombines with one of the joining segments. If this rearrange-ment is unsuccessful, remaining V and J segments can recombine and if one alleleis completely unsuccessful, the second allele will undergo recombination. If bothalleles are unsuccessful, then the second light chain, lambda (IGL) on chromo-some 22q11, will undergo rearrangement in the same manner (Fig. 6.2b). The kappa

a

b

Fig. 6.2 Kappa and lambda light-chain gene rearrangement. (a) Kappa light-chain gene rearrange-ment. Vκ–Jκ rearrangement occurs if unsuccessful subsequent Vκ–Jκ rearrangements can occurwith any of the remaining Vκ and Jκ segments until there are no remaining segments. Non-template(N) nucleotides are only added to the join approximately 25% of the time. A kappa-deletingelement (Kde) segment can rearrange with a variable segment or the intron between Jκ and Cκ

(intron-RSS) leading to inactivation of the allele. (b) Lambda light-chain gene rearrangement.Rearrangement is similar to kappa light-chain gene rearrangement, except that there is no analogto Kde. N nucleotides are only added to the join approximately 25% of the time


immunoglobulin light-chain gene has a unique area called the kappa-deleting ele-ment (Kde) (Fig. 6.2a). The Kde area can recombine with a variable segment oran area in the intron between Jκ and Cκ (intron-RSS) (Fig. 6.2a). Both possibleKde recombinations delete the constant region and recombination with the variableregion also deletes the junctional region leading to inactivation and inhibition ofsomatic hypermutation of that allele. Rearrangement of the light-chain genes theo-retically results in 320 different combinations and 1.92 × 106 different combinationsare theoretically possible with a heavy chain paired to a light chain [2]. Diversity isincreased to about 5 × 1013 by the addition of N and P nucleotides and subtractionof nucleotides in the areas of rearrangement [2]. Successful light-chain rearrange-ment results in the expression of surface IgM and the cell becomes an immatureB cell.

Immature B cells leave the bone marrow (unless they react to self-antigens) andtravel to the peripheral lymphoid tissues such as spleen and lymph node where theyare now considered mature B cells. Mature B cells enter in the interfollicular areasand move through the germinal center where they are exposed to foreign antigens tobecome plasma cells and memory B cells. Antigen-naïve B cells in the mantle zonehave not switched the CH region and express IgM or IgD using alternative transcrip-tion or splicing [4]. These cells compete for entry into the germinal center whereproliferation, class (isotype) switching, and somatic hypermutation take place.

Class switch to IgG, IgA, or IgE occurs upon presentation to antigen and isdependent on signals from the microenvironment, including helper T cells and den-dritic cells [4, 5]. Activation-induced cytidine deaminase (AID) is important forboth class switching and somatic hypermutation by converting cytosine to uracil[4]. Uracil is removed by base excision repair leading to a single-stranded (ss) DNAbreak [4]. The resulting gap is usually repaired by DNA polymerase which replacesthe missing nucleotide based on reading the opposing DNA strand. However, forclass switch to occur, DNA polymerase fails to correct the problem. Instead, themismatch repair (MMR) system excises the nucleotides near an ssDNA break onone allele and an ssDNA break on the other allele leading to a double-strandedDNA break [4]. In some cases, two uracils on opposing DNA strands will be closeenough to form a double-stranded break without the action of MMR. Double-stranded breaks formed by either of these methods are cleaned up to form bluntends. A blunt end in a donor switch (S) region is then recombined with a blunt endin an acceptor S region via non-homologous end joining [5] as shown in Fig. 6.3.

Somatic hypermutation starts in the dark zone of the germinal center in cen-troblasts and continues as the cell traverses the germinal center and becomes acentrocyte. Studies have shown mutation-active periods and mutation-silent peri-ods during which antigen selection takes place [6]. A cell that has a good fit with apresented antigen goes on to the next round of somatic hypermutation, while a cellthat does not fit is targeted for apoptosis. Somatic hypermutation functions in anti-gen selection by producing point mutations of the V segment of the heavy- and/orlight-chain genes which may lead to a single amino acid change during cell division.Over several cell divisions, multiple mutations accumulate. Still these mutationsusually lead to only subtle changes in affinity to antigen compared to the original


Fig. 6.3 Class switch of the immunoglobulin heavy-chain gene. Activation-induced cytosinedeaminase (AID) changes cytosine to uracil in the switch regions ultimately leading to double-stranded DNA breaks and recombination between two different switch regions. In this example,there is class switching to IgG2

parent-naïve B cell. AID converts cytosine to uracil which is repaired, sometimesincorrectly resulting in a point mutation. Studies have shown that AID is active innon-immunoglobulin regions of the genome and may play a role in immunoglobulintranslocations with other genes [7]. The effects of AID outside of the immunoglob-ulin loci are repaired at a high rate, whereas repair in the immunoglobulin regionsis error prone, leading to mutations in 1 per 1,000 nucleotides per cell division [8].This mechanism seems to involve the V region promoter, since somatic hypermu-tation occurs over a limited area downstream from the promoter and the rate ofmutation is proportional to the distance from the promoter [8].

Errors in these normal physiologic B-cell processes can lead to genetic alterationsthat are lymphomagenic or leukemogenic. The double-stranded breaks that occurduring VDJ recombination and class switching can lead to translocations involvingthe immunoglobulin genes such as t(14;18), t(11;14), and t(8;14). Aberrant somatichypermutation can occur in areas of the genome other than the immunoglobulingenes leading to multiple point mutations or translocations. Testing for aberrantsomatic hypermutation is not generally available; however, testing for pathologicrearrangements and physiologic VDJ rearrangements is widely used and help-ful for determining clonality, supporting a diagnosis of malignancy, lymphomasubclassification, and minimal residual disease evaluation.

B-Cell Clonality Testing

Clonality testing takes advantage of the fact that once a B cell has rearrangedits immunoglobulin genes, every daughter cell will have the same rearrange-ment. Indications for gene rearrangement clonality testing include evaluation ofB-cell proliferations when morphology and immunophenotype are inconclusivefor malignancy, lymphoproliferations in immunosuppressed individuals (includingpost-transplant patients), evaluation of minimal residual disease, and comparison of


two lymphoid malignancies for a clonal relationship. Gene rearrangement testingis not lineage specific and should not be used to establish lineage. IGH rearrange-ment has been detected in 4–18% of AML, 8–16% of mature T-cell lymphomas, andapproximately 22% of T-lymphoblastic leukemia (T-ALL), especially CD3-negativeor γδ (gamma delta) T-ALL, although lineage infidelity with IGH is less frequentthan with T-cell receptor (TCR) gene rearrangements [3, 9–11].

Southern blot, once the gold standard for detecting clonal populations, is lesscommonly performed than PCR because it is time consuming, labor intensive, tech-nically demanding, and requires large amounts of high-quality DNA (fresh tissue).Additionally it has a lower sensitivity than PCR, requiring the presence of 5–10%tumor cells for detection. Southern blot relies on the detection of non-germline (i.e.,rearranged) DNA fragments after restriction enzyme digest. Testing of IGH and IGKis more common because these genes not only have a simple gene structure allowingthe use of one to two probes but also have a large repertoire. False positives are rareand related to incomplete enzyme digestion or polymorphisms of restriction enzymesites. False negatives are also rare as long as the DNA quality is good and adequate(5–10%) tumor cells are present.

PCR has several advantages over Southern blot: it is quicker, less labor inten-sive, requires smaller amounts of DNA, can tolerate lower quality DNA (such asDNA obtained from fixed tissue), and has a better sensitivity. However, PCR ismore prone to false positives and false negatives (especially in germinal centeror post-germinal center-derived malignancies). PCR testing takes advantage of thelength differences created during VDJ rearrangement, especially from the additionof N and P nucleotides and the subtraction of nucleotides. In normal or reactiveB-cell populations, PCR product sizes will show a Gaussian distribution, oftenreferred to as a polyclonal background (Fig. 6.4). A clonal B-cell population hasthe same immunoglobulin rearrangement in all cells and that PCR product size willpredominate with or without a polyclonal background (Fig. 6.4).

PCR testing usually targets the VH–JH region with consensus or family primersto at least two of FR I, II, and III and consensus primers to FR IV (Fig. 6.1);

Fig. 6.4 Example of PCR with capillary electrophoresis gene scanning using primers to FR III.The top panel shows a clonal peak. The bottom panel demonstrates a polyclonal population havinga bell-shaped curve distribution of sizes in the expected range for the primer set used (71–150 bp)


however, testing of other segments can increase diagnostic sensitivity. Testingwith a single FR yields sensitivities from 50–75% depending on the primer used[12–14]. Testing with multiple FR primers increases sensitivity to approximately80% [12–14]. Addition of primers to evaluate kappa immunoglobulin light-chaingene rearrangement increases sensitivity further to approximately 90% [12, 13, 15].The BIOMED-2 primers target FR I, II, and III, as well as incomplete DH–JH rear-rangement, IGK, and IGL. Testing of DH–JH rearrangement targets the incompletelyrearranged, non-coding allele and increases the yield in germinal center/post-germinal center malignancies as an incompletely rearranged allele does not undergosomatic hypermutation [3]. BIOMED-2 primers for IGK include evaluation ofVκ–Jκ as well as rearrangements with Kde. Many studies have shown good sen-sitivity and specificity for detecting clonality in B-cell neoplasms versus reactiveconditions when using a combination of BIOMED2 primer sets (tubes). Generallyusing both κ (kappa) tubes (Vκ–Jκ and Vκ/intron-RSS-Kde) with some combinationof the FR I, II, or III tubes gives a sensitivity from 91 to 98%, with good DNA qual-ity (>300–400 base pairs) [12, 13]. The DH–JH primer set may be useful in clonalitydetection in plasma cell neoplasms which are positive in about 60% of cases andin improving detection of marginal-zone lymphoma to >90% [12, 13]. Detection ofclonality in lymphoplasmacytic lymphoma is also increased slightly with the use ofthe DH–JH primer set [16]. Use of a lambda primer set does not provide significantadditional information [12, 13].

Because of its high sensitivity, false positives are often seen with PCR testing.These pseudoclonal or often oligoclonal populations result from the amplification ofa limited benign population or repertoire. Amplification of a few B cells (for exam-ple, hypocellular samples after chemotherapy or with aplastic anemia) can lead toan apparent clonal gene rearrangement. Oligoclonal/pseudoclonal populations dueto limited numbers of cells will amplify different peaks on different runs. Therefore,running a sample in duplicate (done in most laboratories) or retesting an apparentlyclonal population avoids this complication [12, 13]. A limited repertoire can be seenin immunosuppressed patients (including transplant patients) and in reactive lymphnodes because of the presence of antigen-selected clones.

False negatives most often occur due to poor annealing of primers; there-fore, primer design is very important. However, even with good primer design,somatic hypermutation can affect the primer-binding sites and lead to false nega-tives. As expected, false negatives are more common in B-cell malignancies thathave undergone somatic hypermutation (germinal center or post-germinal centermalignancies), although targeting the kappa gene rearrangements mentioned abovegreatly improves detection [12, 13]. Degraded poor-quality DNA and the presenceof PCR inhibitors can lead to falsely negative results. Both of these occur moreoften when using fixed and paraffin-embedded tissue [3]. Both the type of fixativeand the age of the specimen can affect PCR success. Degradation of DNA can beassessed by measuring the size of DNA fragments. Fragments less than 200 basepairs lead to extremely poor detection rates by PCR (16%), while fragments atleast 300 base pairs in size lead to reasonable rates of amplification (>76–96%)[3, 12].


Somatic Hypermutation Testing

In the research setting, somatic hypermutation status of the IGH variable region(VH) may be used to help distinguish the cell of origin of B-cell neoplasms.However, in the clinical setting, testing is usually performed to determine the VHmutational status of chronic lymphocytic leukemia/small lymphocytic lymphoma(CLL/SLL) and splenic marginal-zone lymphoma. Testing for somatic hypermu-tation involves sequencing the VH segment of a clonal population and comparingthe resulting sequence to a database of germline sequences of the different VH seg-ments. Significant differences from the most homologous VH segment sequenceindicate that the population has undergone somatic hypermutation. A difference of≥2% is considered mutated, whereas >98% homology is considered unmutated. InCLL/SLL, unmutated VH is associated with a worse prognosis. Ongoing somatichypermutation is indicated by detecting several different sequences from the clonalpopulation as opposed to a single sequence. The presence of a single sequenceindicates that a cell has been exposed to antigen within the germinal center andcompleted somatic hypermutation prior to clonal expansion and therefore is eithera late germinal center or a post-germinal center B cell. The presence of ongoingsomatic hypermutation suggests a cell of origin of a germinal center B cell at thestage of antigen presentation, although a more mature B-cell stage that has failed toshut off or has re-initiated somatic hypermutation is another possibility.

Follicular Lymphoma

Follicular lymphoma is derived from germinal center B cells (centrocytes and cen-troblasts) which have rearranged heavy- and light-chain genes and have ongoingsomatic hypermutation. Morphology and immunophenotype typically reflect thisorigin with a follicular growth pattern and expression of germinal center markersCD10 and BCL6. Rearrangement of immunoglobulin genes can be detected by PCR,although somatic hypermutation decreases the detection rate to approximately 60%by VH–JH testing. Targeting the incomplete DH–JH rearrangement and kappa lightchain can increase the detection rate to > 90% [12, 13]. Follicular lymphoma is char-acterized by t(14;18) or variant B-cell CLL/lymphoma 2 (BCL2) rearrangementssuch as t(2;18) and t(18;22) which are found in up to 90% of follicular lymphomas.

Follicular Lymphoma and BCL2 t(14;18)

Normal BCL2 protein plays a role in mitochondrial permeability and has an anti-apoptotic effect. The BCL2 breakpoints are in the untranslated region and thereforethe translated portion of BCL2 is fused to the 3′ immunoglobulin gene which bringsBCL2 under the influence of the immunoglobulin promoter leading to overexpres-sion of a functional protein and decreased apoptosis. Approximately 65–70% of


the rearrangements occur in the major breakpoint region (MBR) which is locatedin the untranslated region of the last exon (exon 3) and another 10% occur in theminor cluster region (mcr) which is about 30 kb outside of the BCL2 gene (Fig. 6.5).Approximately 20% occur either at the 3′-end of the BCL2 gene or 5′ to the mcr andin some patients the breakpoint is unknown [17, 18]. The IGH breakpoint consis-tently occurs in the junctional region. Although some studies have noted differentclinical characteristics associated with the location of the BCL2 breakpoint, otherstudies have failed to confirm those findings [17, 18].

Testing for the BCL2 rearrangement is indicated if there is a suspicion of fol-licular lymphoma, but clonality cannot be demonstrated by immunoglobulin generearrangement studies and when follicular lymphoma is in the differential of anothersmall B-cell lymphoma displaying a nodular pattern, such as marginal-zone lym-phoma with colonization of the germinal center. Testing can be done by conventionalcytogenetics, FISH, or PCR. Immunohistochemical staining of BCL2 does not indi-cate the presence of the BCL2 translocation as other mechanisms can cause BCL2expression and staining is positive in most small B-cell lymphomas and somenormal B cells, such as mantle zone cells, and normal T cells.

Conventional cytogenetics detects most cases of t(14;18) and its variants as wellas other abnormalities; however, fresh tissue is not always available and there arecryptic rearrangements that require additional testing to detect. Most PCR assayshave primers to only the MBR and mcr, and the detection rate of t(14;18) is about60% [3]. Some assays have been designed with additional primers to the 5′ mcrand 3′ MBR breakpoint regions, but the detection rate still reaches only 60–88%[3, 19, 20]. FISH detects >90% of t(14;18) and, depending on the probe strategy,

Fig. 6.5 BCL2/IGH rearrangement at the major breakpoint region (MBR). The IGH breakpointis consistently in the joining region. The BCL2 breakpoints are variable and include the following:the major breakpoint region (MBR) within the 3′ non-coding portion of exon 3; the minor clusterregion (mcr), located 20–30 kb 3′ to the MBR; and additional breakpoints/clusters between theMBR and the mcr [3′ BCL2, 3′ MBR, intermediate cluster region (icr), and 5′ mcr] in most of theremaining cases


will usually detect variant cases; therefore, FISH is generally preferred over PCRfor diagnosis [20, 21].

Positivity of BCL2 by PCR requires correlation with other information, as thet(14;18) can occur in other lymphomas, particularly diffuse large B-cell lymphomas,and can occur in the peripheral blood of healthy blood donors and in hyperplasticlymphoid tissue [22, 23]. False-negative PCR results occur with alternate break-points or mutations of the primer-binding sites. About 10% of nodal follicularlymphomas are t(14;18) negative by current testing methods; however, absence of aBCL2 translocation is the rule for pediatric follicular lymphoma and primary cuta-neous follicle center cell lymphoma. A small subset of t(14;18)-negative follicularlymphomas have an alternate mechanism of increasing BCL2 expression, such as+18q [24]. Translocations of BCL2 are less common in grade 3B follicular lym-phoma with <15% of cases positive compared to >80% positivity in grades 1 and 2and >70% positivity in grade 3A [19]. Grade 3B follicular lymphomas frequentlyhave alterations in BCL6 on 3q27 and other cytogenetic abnormalities similar toDLBCL [19].

Other Genetic Abnormalities in Follicular Lymphoma

Since t(14;18) can be found in the blood and hyperplastic lymphoid tissues ofhealthy individuals, additional abnormalities are likely required for follicular lym-phoma to develop [22, 23, 25]. The translocation is thought to be an early eventin B-cell development leading to a prolonged life span of the B cell, giving ampletime to develop other genetic defects. In fact, follicular lymphoma usually has atleast one additional abnormality by routine cytogenetics such as gains of 1q, 2p, 7,8q, 12q, 18q, and X and losses of 1p, 6q, 10q, 13q, and 17p [25]. Findings asso-ciated with a bad prognosis are thought to be late events and include gains of 1q,12, and X and losses of 1p, 17p, and 17q [25]. Some of these same abnormalitiesare associated with large-cell transformation which occurs in 10–60% of follicu-lar lymphomas (Table 6.2). Transformation to a higher grade lymphoma, usuallydiffuse large B-cell lymphoma, occurs in 25–35% of patients with follicular lym-phoma [1]. Transformation has been associated with gains of 7, 12q, and X; lossesof 4q, 13q, and 17p; inactivation of TP53 and CDKN2A; and MYC deregulation[25]. BCL6 rearrangements are common in grade 3B follicular lymphomas with adiffuse large B-cell component but are rare in lower grade follicular lymphomasand grade 3B follicular lymphomas with a pure follicular pattern [26]. Follicularlymphoma rarely has a t(8;14) which is associated with a particularly aggressivecourse [27]. CGH has shown chromosomal gains similar to conventional cytoge-netics but also found gains in 18p and 12p in over 10% of the cases, although thesignificant gene(s) affected are not known [28]. Deletions by CGH include thoseseen by conventional cytogenetics, as well as deletions of 9p, 3q, and 11q [28].The deletion of 9q often involves the CDKN2A and CDKN2B loci and is associ-ated with worse overall survival [28]. The deletion of 3q involves the LIM domaincontaining preferred translocation partner in lipoma (LPP) gene approximately half


Table 6.2 Genetic changes associated with higher grade (grade 3B) follicular lymphoma (FL),a poor prognosis, and transformation

Grade 3B FL Poor prognosis/transformation

Less frequent t(14;18) del 6q23-263q27 abnormalities del 17p

including BCL6 TP53 mutations/inactivationrearrangementsa –1p, +12, +18p, +Xp

MYC rearrangementsInactivation p16INK4A

In general these are all associated with more complex karyotypes andan increased number of abnormalitiesaWhen diffuse growth is present

of the time [28]. Deletions of 5p and 6q were also associated with a worse overallsurvival [28].

Prognosis in follicular lymphoma has also been linked to gene expression pro-filing (GEP) of the background cells, which shows two distinct signatures termedimmune response 1 (IR1) and immune response 2 (IR2). IR1 displays increasedexpression of T-cell genes and the macrophage genes TNFSF13B and ACTN1 andhas a favorable prognosis. IR2 shows increased expression of follicular dendriticcell genes and other macrophage genes and has an unfavorable prognosis [25].

Mantle Cell Lymphoma

Mantle cell lymphomas (MCLs) originate from the small CD5-positive B cellsthat reside in the mantle zone areas surrounding germinal centers [29]. Mostare thought to be derived from naïve pre-germinal center cells, while the minor-ity with mutated immunoglobulin genes may arise from post-germinal centermemory B cells [29, 30]. Virtually all MCLs have a characteristic translocation,the t(11;14)(q13;q32), that brings the CCND1 gene at 11q13 encoding cyclin D1under control of the immunoglobulin heavy-chain locus at 14q32 [1, 29]. The netresult is constitutive expression of cyclin D1 which is normally not expressed byB cells. In addition, levels of cyclin D1 are often further increased in MCL bydeletions and point mutations that result in the removal of destabilization elementsin the cyclin D1 mRNA leading to truncated transcripts with an increased half-life [31]. Increased levels of cyclin D1 cause deregulation of cell cycle control atthe G1–S phase checkpoint by mitigating the suppressor effects of the retinoblas-toma protein and the cyclin-dependent kinase inhibitors p27kip1 and p21 [29, 32].The t(11;14)(q13;q32) is thought to occur as a mistake during the immunoglobulingene rearrangement process in early B-cell development and is necessary but notsufficient for the subsequent development of MCL [29].


Detection of Cyclin D1 Dysregulation

The diagnosis of MCL usually requires direct or indirect demonstration of ant(11;14)(q13;q32) [1]. Only very rare cases of MCL lack t(11;14)(q13;q32) and asa result are difficult to diagnose with certainty using standard techniques [1, 33].The few cases of MCL that do not have t(11;14)(q13;q32) typically show abnor-mal expression of cyclin D2 or cyclin D3 [1, 33]. Testing for t(11;14)(q13;q32)should generally be performed on all CD5+ small B-cell neoplasms that do nothave the characteristic phenotypic and morphologic features of CLL/SLL. In addi-tion, t(11;14)(q13;q32) testing is also appropriate for B-cell neoplasms that do notfit into other diagnostic categories since MCL may occasionally lack CD5 and/orhave other unusual features [1]. In most cases, indirect testing for t(11;14)(q13;q32)by immunohistochemical staining paraffin-fixed tissue section for cyclin D1 thathas histologic evidence of lymphoma is sufficient for diagnosis [1]. Although othersmall B-cell neoplasms should be negative, cases of hairy cell leukemia as well asthe proliferation centers of CLL/SLL may sometimes show increased reactivity forcyclin D1, creating a potential diagnostic pitfall [34, 35]. In addition, histiocytes andendothelial cells which can be admixed with lymphoma cells may normally expresscyclin D1. Also, approximately 20% of plasma cell myeloma cases can express ele-vated levels of cyclin D1 secondary to having a t(11;14)(q13;q32) and often have amore lymphoid cytologic appearance than do typical plasma cells [36]. In suspectedcases of MCL that have equivocal cyclin D1 staining results and/or other unusualfeatures, testing for t(11;14)(q13;q32) by FISH is the method of choice. FISH testingfor t(11;14)(q13;q32) can be reliably performed using standard paraffin-embeddedfixed tissue biopsies and has a sensitivity of over 95% for MCL [37, 38]. Becauseof the variability in t(11;14)(q13;q32) breakpoints, molecular PCR-based testing isinformative in only approximately 40–50% of cases [38, 39]. Classical cytogeneticsand Southern blot analysis can also be used for detection of t(11;14)(q13;q32), butboth methods require fresh non-fixed tissue or cells, are relatively expensive andtime consuming, and are only about 70% sensitive [40].

Other Genetic Abnormalities in MCL

Most MCLs also have large numbers of other detectable genetic abnormalitiesincluding gains, losses, and high copy amplification of particular chromosomalregions [29, 32, 41]. In particular, gains of 3q25-qter have been reported in 30–50%of cases, 7p21–22 gains and 8q21-qter gains in 15–35%, and gains of 18q11–23in 16% of cases [1, 29]. Frequent areas of chromosome loss include 1p13–31 in30–50% of cases, 6q23–27 in 20–40%, 9p21–22 in 20–30%, 9q21-qter in 20–30%,11q22–23 in 20–60%, 13q11–13 in 25–55%, 13q14–34 in 40–50%, and 17p13-pterin 20–45% of cases [1, 29]. In addition to the above listed structural abnormalities,inactivating mutations of the ATM gene at 11q22–23 appear to be present in 40–75%of MCL cases [42]. Abnormalities of the 8q24 locus that contains the MYC gene arenot common but have been reported in very aggressive MCL [43].


Clinical Implications

Most MCLs are aggressive with patients typically experiencing only short responsesto current treatments and having median survivals of only 3–4 years [1, 29].However, small numbers of MCL patients have been identified with very indo-lent clinical courses, even without receiving treatment, indicating that the clinicalbehavior of MCL can be highly variable [29, 44, 45]. Recent studies have suggestedthat MCL in patients who experience indolent disease may have relatively few ifany identifiable genetic abnormalities other than the t(11;14)(q13;q32), unlike themajority of MCL [29, 46]. Gene expression array profiling of large numbers of MCLcases has identified a proliferation signature using 20 genes that can divide patientsinto four prognostic groups with median survivals ranging between 10 months (mostaggressive group) and 6.7 years (least aggressive group) [47]. As might be expected,given these findings, more routinely used measures of cell proliferation such as themitotic index or percentage of Ki-67 positively stained cells have also been nega-tively correlated with survival [48]. The negative survival impact of 3q region gains,or 9p, 9q, and 17p region losses which are often present in clinically aggressivecases, appears to be independent of the array-based proliferation score [29], indi-cating that multiple measures of prognosis may need to be considered to optimizetreatment decisions for MCL patients. Although clinical tests for many of the cyto-genetic abnormalities frequently seen in MCL are not readily available, quantitativePCR assays for copy number and other MCL alterations have been reported that maybecome suitable for more routine prognostic stratification [49]. Unlike CLL/SLL,the mutational status of the expressed immunoglobulin heavy-chain variable genesegments (VH) does not appear to have prognostic value [45, 50]. However, MCLsappear to show preferential use of certain VH gene segments relative to normalCD5+ B cells, indicating that direct antigen receptor stimulation may be playing apositive role in lymphoma cell development and growth [50, 51]. Moreover, the useof particular VH gene segments such as V3–21 has been associated with increasedsurvival and fewer genetic alterations [46, 50, 51].

Diffuse Large B-Cell Lymphoma

Diffuse large B-cell lymphoma (DLBCL) is a heterogeneous entity, the minorityof which falls into defined categories (Tables 6.3 and 6.4) with specific morpho-logic, immunophenotypic, and molecular features, and the majority are referred toas DLBCL, not otherwise specified.

These are a heterogeneous group of tumors that can be subtyped by differentgene expression profiling signatures. Subtyping of DLBCL by comparing signa-tures to normal B cells has identified two distinct signatures: germinal centerB-cell (GC) type, which has a profile similar to germinal center B cells, and thepost-germinal center or activated B-cell (ABC) type, which has a profile like acti-vated peripheral B cells, although some cases do not display either signature [52].Alternative clustering methods have been used to divide DLBCL based on potential


Table 6.3 Genetic alterations in specific diffuse large B-cell subtypes

Entity Recurrent abnormalitiesImmunoglobulingene SHM

Geneexpressionprofiling Cell of origin

T-cell-rich,histiocyte-richDLBCL

4q and 19p anomaliesonly seen by CGH

RearrangedOngoing SHM

Host immune-responseprofile

GC B cell

Primary DLBCLof CNS

BCL6 rearrangementsBiased use of VH4/34PIM-1, MYC,

RhoH/TTFn,and PAX5

6q del and gains of 12qand 22q

Gains of 18q21 and copynumber increase inBCL2 and MALT-1

Deletions at 9p21 affectCDKN2A/p16INK4A

array CGH 6p21.3 (HLAregion)

No t(14;18)


(< 27%)

Activated(late) GC Bcell

PrimarycutaneousDLBCL, legtype

Similar to DLBCB, NOSBCL-6, MYC, and IgH

rearrangementsHigh-level amplifications

of 18q21.31–q21.33including BCL-2 andMALT1

del of 9p21.3 (CDKN2Aand CDKN2B)

No t(14;18)


(some cases)

ABC type Peripheralpost-GC Bcells

EBV+ DLBCLof the elderly

None reported Rearranged Unknown Mature B celltransformedby EBV

DLBCL, diffuse large B-cell lymphoma; SHM, somatic hypermutation; GC, germinal cen-ter; CNS, central nervous system; NOS, not otherwise specified; ABC, activated B cell; EBV,Epstein–Barr virus

pathogenetic mechanisms into three groups: oxidative phosphorylation, BCR(B-cell receptor)/proliferation, and host response [53]. Although the prognosis in thesegroups is similar, identification of such groups provides insight into pathogenesisand may help guide research for targeted therapy [53].

BCL6 Alterations in DLBCL

The B-cell CLL/lymphoma 6 (BCL6) gene, located on chromosome 3q27, encodes azinc finger protein that is a transcriptional repressor normally expressed in germinal


Table 6.4 Genetic alterations in other lymphomas of large B cells

EntityRecurrentabnormalities

Gene expressionprofiling Other

Primary mediastinal (thymic)large B-cell lymphoma

+9p24, +2p15,+Xp11.4–21,+Xq24–26

Similar to classicalHodgkin lymphoma

Negative

Intravascular large B-celllymphoma

Unknown Unknown Negative

DLBCL associated withchronic inflammation

Distinct profile EBV positive

Lymphomatoidgranulomatosis

None Unknown EBV positive

ALK-positive DLBCL t(2;17) (CLTC-ALK)t(2;5) lesscommon

Unknown Negative

Plasmablastic lymphoma None Unknown EBV positive60–75%

Large B-cell lymphomaarising inHHV-8-associatedmulticentric Castlemandisease

Unknown Unknown EBV negativeHHV8 positive

Primary effusion lymphoma None Distinct profile EBV positiveHHV8 positive

DLBCL, diffuse large B-cell lymphoma; EBV, Epstein–Barr virus; HHV, human herpes virus

center B cells but downregulated with maturation into plasma cells. Expression ofBCL6 appears to block differentiation to memory B cells and inhibit apoptosis.BCL6 translocations are the most common translocation found in DLBCL but otherabnormalities, such as gains/amplifications or mutations, involving BCL6 also occur[54]. Approximately 40% of DLBCLs show some alteration of BCL6 which is morecommon in the ABC type than the GC type [54, 55]. These genetic changes do nothave a uniform effect on BCL6. Although constitutive activation of BCL6 is a fre-quent outcome, not all cases display protein overexpression [54, 56]. Differentialbinding of BCL6 protein to its normal target genes has been described and mayexplain the variability in protein expression [56]. Increased BCL6 protein expres-sion has been associated with a better prognosis [54, 57]. Changes in BCL6 are notlimited to DLBCL and are seen in other lymphomas, as well as non-hematologicmalignancies.

BCL6 translocations occur more commonly with ABC-type DLBCL and part-ner with IGH about half the time [54]. The translocations generally bring BCL6under the effect of a new promoter [25]. BCL6 translocations have also been foundin primary mediastinal DLBCL [54], cutaneous diffuse large B-cell lymphoma,leg type [58], follicular lymphoma, and nodular lymphocyte-predominant Hodgkinlymphomas [59, 60]. The breakpoint on BCL6 is most commonly in the major break-point region (MBR) which spans the non-coding exon 1 and 5′ region of the first


intron, but a few cases occur at an alternative breakpoint region [61]. As would beexpected with alternate breakpoints and multiple translocation partners, PCR test-ing for BCL6 translocations has a poor detection rate and FISH with a break-apartprobe for BCL6 is the preferred method of testing for BCL6 translocations. A BCL6translocation is not specific for DLBCL, and there is no clear association with prog-nosis; therefore, routine clinical testing for BCL6 translocation is not recommendedcurrently.

Mutations of BCL6 are common in DLBCL and involve exon 1 and the 5′region of intron 1. Multiple mutations are often present suggesting aberrant somatichypermutation. The study by Iqbal et al. [54] detected higher messenger RNA(mRNA) levels and a trend toward higher protein levels in DLBCL cases with BCL6mutations.

Other Genetic Alterations in DLBCL

Many other genetic alterations have been found in DLBCL, including other translo-cations, amplifications, aberrant somatic hypermutation, deletions, and inactivation.Genetic alterations vary with the type of DLBCL. GC-type DLBCL is more fre-quently associated with t(14;18), amplification of REL on chromosome 2p, and gainof 12q [25]. BCL2 translocations, such as the t(14;18) seen in follicular lymphoma,occur in 20–30% of the GC-type DLBCL [25]. REL is a transcription factor inthe nuclear factor kappa B (NF-κB) family. The NF-κB pathway is also alteredin the ABC-type DLBCL, with constitutive activation of NF-κB being described.Other changes seen primarily in the ABC type are gains of 3q and 18q, loss of 6q(including PRDM1), and inactivating mutations of PRDM1 [25]. MYC rearrange-ments occur in up to 10% of DLBCLs and are associated with a worse prognosiseven in rituximab-treated patients and often, but not always, a high proliferative rateat >90% by Ki67 [1, 25, 62, 63]. MYC-positive DLBCLs are often indistinguishablefrom DLBCLs without MYC translocations [62]. The partner of the rearranged MYCis an immunoglobulin gene in 60% of cases [1]. Aberrant somatic hypermutationsin genes, such as PIM1 (a protooncogene), MYC, RHOH (a RAS family GTPase),and PAX5, occur in more than 50% of DLBCLs [25].

Immunohistochemical stains can be used to categorize DLBCL into germinalcenter like or ABC type, although the correlation with gene expression profiling isnot perfect. Cases that are CD10 positive or BCL6 positive and MUM1 negativeare germinal center like and all other staining patterns are the ABC type [64, 65].Different chromosomal abnormalities are found in these two entities as well withBCL2 rearrangements and gains of 12q12 occurring in the germinal center-typeDLBCL, while 3q27 abnormalities, gains of 18q21–q22, and losses of 6q21–q22occur more often in the ABC-type DLBCL. As a group, DLBCL shows rear-rangement of the immunoglobulin genes and somatic hypermutation in the variableregions. As expected, only the GC type of DLBCL has ongoing somatic hypermu-tation [1]. Early studies showed the GC subtype to have a better prognosis with a


50–60% 5-year survival, compared to 15–30% 5-year survival for the ABC sub-type. More recent studies, with rituximab as a part of standard therapy, have hadconflicting results. A gene expression profiling study by Lenz et al. confirmed abetter overall and progression-free survival in the GC type of DLBCL; however,other studies showed no survival difference between the GC and ABC subtypes asidentified by immunohistochemistry [66–68]. Gene expression profiling of the non-malignant background cells has shown two distinct signatures that are associatedwith outcome. One signature, referred to as stromal 1, has expression of genes thatencode extracellular matrix components and remodeling proteins and macrophagegenes and is associated with a better prognosis [66]. A second signature, referred toas stromal 2, has expression of endothelial genes and genes associated with angio-genesis and is associated with a poorer prognosis [66]. Both the stromal 1 andstromal 2 gene expression profiles were seen in GC and ABC types of DLBCL[66].

Expression of BCL2 protein by immunohistochemical staining has been asso-ciated with prognosis. In general, BCL2 positivity is associated with a worseprognosis, although studies vary as to whether the prognostic difference occurs inthe GC type or non-GC type of DLBCL [65, 69].

Prognosis is also affected by the presence of histologic and molecular bone mar-row involvement. Patients with histologic marrow involvement have the poorest5-year survival of 12% [70]. However, patients without histologic involvement but apositive gene rearrangement study have a 5-year survival of 37% compared to 66%for patients who lack both [70].

Marginal-Zone Lymphomas

Extranodal marginal-zone lymphomas of mucosa-associated lymphoid tissue(MALT lymphomas) represent the majority of lymphomas that arise outside ofprimary hematopoietic tissues (lymph node, spleen, bone marrow) [1]. The cellof origin is thought to be a post-germinal center memory B cell which normallyresides in the marginal zones that surround follicular mantles. As would be expectedbased on the cell of origin, the vast majority of MALT lymphomas lack expressionof CD5 and CD10 [1]. MALT lymphomas develop at sites that do not normallyhave lymphoid tissue but where lymphoid tissue has been acquired in response to achronic infection or an autoimmune disease [71]. The most common site of MALTlymphoma development is the stomach, where Helicobacter pylori is the infec-tious agent causing acquisition of the reactive precursor MALT [72, 73]. MALTlymphomas that develop in the salivary gland or thyroid are preceded by reactiveinfiltrates related to the autoimmune diseases Sjögren’s syndrome and Hashimoto’sthyroiditis, respectively [1]. The highly restricted use of certain VH gene seg-ments by salivary gland lymphomas implicates direct antigen stimulation mediatedthrough the immunoglobulin receptor complex as playing an important role in lym-phoma development [74], while in the stomach the importance of H. pylori-specific


T cells to lymphoma cell growth and survival highlights the complementary roleof indirect antigen stimulation [75]. With continued antigenic stimulation, clonalB-cell populations sometimes develop in the acquired infiltrates with molecular andgenetic abnormalities that give rise to lymphomas [76, 77].

Genetic Abnormalities in MALT Lymphomas

Four recurrent balanced translocations have been reported in MALT lymphomas,t(11;18)(q21;q21), t(14;18)(q32;q21), t(1;14)(p22;q32), and t(3;14)(p14.1;q32) [76,77]. The t(11;18)(q21;q21) fuses the amino end of API2 at 11q21 to the carboxylterminal of MALT1 at 18q21, generating a chimeric fusion protein that activatesNF-κB. The t(14;18)(q32;q21) and the t(1;14)(p22;q32) result in deregulation ofMALT1 and BCL10 at 1p22 by bringing them under the control of the immunoglob-ulin heavy-chain (IgH) locus at 14q32, which also result in the activation of theNF-κB pathway. The more recently identified t(3;14)(p14.1;q32) results in thederegulation of FOXP1 (forkhead box protein P1). The incidence of these fourtranslocations varies greatly depending on the lymphoma site (Table 6.5), witht(11;18)(q21;q21) being more common in gastric, intestinal, and lung MALT lym-phomas, while the t(14;18)(q32;q21) and t(3;14)(p14.1;q32) more often identified insalivary gland, and ocular MALT lymphomas [78–80]. These observations suggestthat MALT lymphomagenesis has location-specific features that may in turn dependon the type of antigenic stimulation triggering the precursor-reactive infiltrates. Theconcept of site-dependent factors affecting MALT lymphoma development is fur-ther supported by these translocations being mutually exclusive [78]. In addition,the frequency of MALT lymphoma-associated translocations may also vary withgeographical region, in that European-based studies have found higher incidencesof t(11;18)(q21;q21) in gastric MALT lymphomas relative to North American-basedstudies, while t(3;14)(p14.1;q32), identified in 10% of European MALT lymphoma

Table 6.5 Frequencies (%) of MALT lymphoma translocations and trisomies

Site

t(11;18)a

API2–MALT1

t(14;18)a

IgH–MALT1

t(1;14)a

IgH–BCL10

t(3;14)a

IgH–FOXP1 Trisomy 3b Trisomy 18b

Stomach 5, 24 5, 1 0, 0 0, 0 11 6Intestine 42, 12 0, 0 0, 12 0, 0 75 25Lung 31, 53 10, 7 2, 7 0, 0 20 7Salivary

gland0, 1 0, 12 0, 2 0, 0 55 19

Ocularadnexa

0, 1 0, 24 0, 0 0, 20 38 14

Thyroid 0, 17 0, 0 0, 0 0, 50 17 0

aLeft most data is from North American study of Remstein et al. [78] and right most data is fromEuropean studies of Streubel et al. [79, 80]bData from Streubel et al. [79]


cases, was not detected in a large North American study (Table 6.5) [78–80]. Inaddition to the translocations noted above, trisomies of chromosomes 3 and 18 arealso common, being present in 25–40% of cases (Table 6.5) [78, 79]. The pres-ence of trisomies often occurs in cases that do not show evidence of translocations,indicating that testing for both types of genetic abnormalities is advisable [78].

Detection of MALT Lymphoma Translocations

A definitive diagnosis of MALT lymphoma can often be difficult to render becauseof the presence of the precursor-reactive infiltrate in many biopsy specimens. IGHrearrangement studies can be used to detect clonal B-cell populations. Detectionof one of the MALT lymphoma-associated translocations or trisomies would havea more significant diagnostic impact, especially with equivocal histologic and/orimmunophenotypic findings, and strongly favor a diagnosis of MALT lymphoma.Clinical testing for t(11;18)(q21;q21) is often done by RT-PCR, which workswell for small tissue specimens typically obtained from endoscopic procedures.Moreover, it is also well suited for fixed paraffin-embedded specimens typicallyused for histology, being able to detect 96% of t(11;18)(q21;q21) with only threeprimer sets. FISH is also used by many labs for detection of t(11;18)(q21;q21),although perhaps being slightly less sensitive than PCR in some cases. Clinical test-ing for detection of t(14;18)(q32;q21) as well as for trisomies 18 and 3 is alsowidely available and usually done by FISH. Tests for detection of the less fre-quently encountered t(1;14)(p22;q32), and t(3;14)(p14.1;q32) are presently limitedto research laboratories.

Clinical Implications

Besides having diagnostic importance, the presence of t(11;18)(q21;q21) alsohas prognostic and treatment-related significance. MALT lymphomas witht(11;18)(q21;q21) are much more likely to be present in regional lymph nodesand other distal sites at diagnosis, and are less likely to undergo transformationto large-cell lymphomas [71–81]. Moreover, gastric MALT lymphomas that harbort(11;18)(q21;q21) generally do not respond to antibiotic treatments that eliminateH. pylori, while complete responses can be obtained with this approach in 70% ormore of other gastric MALT lymphomas [76, 82].

Other Marginal-Zone Lymphomas

Nodal marginal-zone lymphomas (MZLs) resemble other types of marginal-zonelymphomas but differ by originating from lymph node-based marginal-zone B cellsand do not show evidence of extranodal or splenic involvement [1]. The four


translocations associated with MALT lymphomas described above are not foundin nodal MZLs [83].

Splenic MZLs originate from spleen-based marginal-zone B cells that phenotyp-ically differ from other marginal-zone B cells in frequently expressing IgD alongwith IgM [1]. MALT-associated translocations are also not seen, but approximately40% of cases show allelic loss of 7q31–32 which is also associated with moreaggressive disease [84]. Similar to CLL/SLL, about half of splenic MZL cases havemutated VH genes and about half have unmutated VH genes, and those cases withunmutated VH genes have a more aggressive clinical course [85]. However, theVH gene mutational status may not be independent of the effect of 7q31–32 loss,which occurs primarily in those cases with unmutated VH genes. More recently,gene expression profiling has suggested that survival of splenic MZL patients isnegatively associated with the expression of CD38 and genes associated with theNF-κB pathway [86]. Both mutated and unmutated splenic MZLs also show biaseduse of the VH1–2 segment, which may be expressed by approximately 40% ofcases [85, 87], further supporting a role for direct antigen receptor stimulation insplenic MZL development. Splenic MZL has a unique transcriptional profile com-pared to other small B-cell lymphomas with expression of genes involved in theAKT1 signaling pathway [88].

Burkitt Lymphoma

Burkitt lymphoma (BL) is an aggressive, rapidly proliferating tumor characterizedmorphologically by medium-sized monotonous cells with many mitoses and apop-totic cells. Macrophages with apoptotic debris are scattered throughout the tumorgiving the classic “starry sky” pattern. The cells have an immunophenotype of ger-minal center B cells and express B-cell markers, CD10, and BCL6 but are negativefor BCL2. The high rate of proliferation is demonstrated by staining with Ki67which is nearly 100% positive in the tumor cells. Translocations of MYC are presentin about 95% of cases, with t(8;14) being present 80% of the time and t(2;14) ort(14;22) comprising the remainder of cases [1]. Rearrangement is more commonwith the kappa light-chain gene than with the lambda light-chain gene. There arethree main types of BL which differ in their epidemiology, clinical characteristics,EBV involvement, and MYC and immunoglobulin breakpoints (Table 6.6).

Burkitt Lymphoma and MYC

MYC [v-myc myelocytomatosis viral oncogene homolog (avian), also known asC-MYC] on chromosome 8q24 encodes a nuclear phosphoprotein that functionsas a transcription factor and plays a role in cell cycle progression, apoptosis, andcellular transformation. Two isoforms exist that result from alternate translationalstart signals. Synthesis of the longer isoform is suppressed in BL. Endemic andEBV-positive BLs have the MYC breakpoint far 5′ from the MYC gene (class III


Table 6.6 Differences between endemic, sporadic, and immunodeficient Burkitt lymphomas(BLs)

Endemic BL Sporadic BL Immunodeficient BL

Geographicdistribution

Equatorial Africa,Papua, New Guinea

Worldwide Worldwide

Anatomicpredilection

Jaws, facial bones(50%)

Abdominal Nodal

EBV+ ∼100% ∼30% 25–40%IGH breakpoint VDJ Cμ switch region VDJMYC breakpoint Class III Classes I and II Class IIIOther HIV+

breakpoint) and the IGH breakpoint in the VDJ region suggesting occurrence dur-ing somatic hypermutation and show a high rate of somatic hypermutation withoutongoing mutation, suggesting a post-germinal center B-cell or memory B-cell ori-gin [6, 89, 90]. Sporadic and EBV-negative BLs have an IGH breakpoint in the classswitch region and the MYC breakpoint occurs in exon 1 or intron 1 (class I break-points) or close 5′ from the MYC gene (class II breakpoints) as shown in Fig. 6.6,and have a low rate of somatic hypermutation, suggesting an early germinal centerB-cell origin such as a centroblast [6, 89, 90].

Fig. 6.6 MYC/IGH rearrangement in endemic (eBL) and sporadic (sBL) Burkitt lymphoma. IneBL, the IGH breakpoint is in the DHJH region and the MYC breakpoint is far 5′ from exon 1 ofMYC (class III). In sBL, the IGH breakpoint is in the switch region of CH and the MYC breakpointis either in exon or intron 1 (class I) or toward the 5′-end but close to exon 1 (class II)


Testing for MYC translocation is done to confirm a diagnosis of classic BL andmay be helpful in cases that are atypical for BL; however, some cases of DLBCL andcases intermediate between BL and DLBCL have MYC translocations [62, 63, 91].A final diagnosis in these cases must take into account the genetic changes with mor-phology and immunophenotype. Conventional cytogenetics can detect many casesof t(8;14) and its variants as well as other abnormalities; however, there are crypticrearrangements that require additional testing to detect. The breakpoints occur overtoo large a region to be amplified by standard PCR; however, long-range PCR tech-niques have reported detection rates up to 87% in cases of sporadic BL with t(8;14)[92, 93]. The breakpoint in endemic BL yields too large a product to be detectedeven with long-range PCR and this method does not detect translocations with kappaor lambda immunoglobulin light-chain genes, making the overall detection of MYCtranslocations lower. Long-range PCR assays generally use the MYC/04 primer toexon 2 of the MYC gene and primers to JH, Cμ, Cγ, and Cα and yield PCR productsranging in size from 1.5 to 12 kb [93, 94]. The sensitivity for detection is 1 in 1,000to 1 in 10,000 cells; therefore it can be used to monitor minimal residual disease[93, 94]. FISH detects about 90% of t(8;14) and, depending on the probe strategy,will usually detect variant translocations; therefore, FISH is generally preferred overPCR for diagnosis [95]. The presence of t(8;14), t(2;8), or t(8;22) is not synonymouswith a diagnosis of Burkitt lymphoma as it can be present in DLBCL and rarely intransformed follicular lymphoma [1].

Other Genetic Abnormalities in Burkitt Lymphoma

Additional genetic events can occur in BL and commonly involve the CDKN2A(p14ARF and p16INK4a)–MDM2–p53 pathway and the BCL2 family of proteinsthrough BIM which binds and inactivates BCL2. Mutations of TP53 occur in about30% of BL, whereas inactivation of CDKN2A (p14ARF) and overexpression ofMDM2 are less common but represent alternative mechanisms to inhibit p53 [96].CDKN2A produces several proteins, including p16INK4a and p14ARF, throughalternate transcription. MDM2 blocks the transactivation domain and exports p53 tothe cytoplasm for degradation, while p14ARF inhibits MDM2. Therefore both over-expression of MDM2 and inactivation of p14ARF lead to decreased expression ofp53 [96]. Inactivation of CDKN2A (p16INK4a) by promoter methylation occurs in asubstantial number of cases and affects the retinoblastoma (Rb) protein by prevent-ing phosphorylation. Other mechanisms of inactivating CDKN2A, such as deletionsand point mutations, occur but less commonly [96]. Chromosome abnormalitiesin addition to MYC translocations are common occurring in 92% of pediatriccases; gains of 1q and 7q have been seen and are associated with a poor outcome[25, 97, 98]. However, most cases have a relatively simple karyotype with only a fewadditional abnormalities [25]. Presence of t(14;18) is associated with an aggressiveclinical course and a poor prognosis [91, 99].

Lymphomas with features of both BL and DLBCL tend to have a more com-plex karyotype than does BL, and although 35–50% have MYC translocations, the


partner is often not an immunoglobulin gene [1, 100]. It is more common to haveBCL2 or BCL6 translocations with a MYC translocation (a double-hit lymphoma) inthese indeterminant lymphomas rather than in either BL or DLBCL and a complexkaryotype with MYC rearrangement by CGH with > 6 abnormalities) [1, 100–102](Table 6.7). Gene expression profiling shows a signature for BL that is distinct fromDLBCL and can categorize some of these cases; however, even with this technique,intermediate cases are found [100, 103].

Table 6.7 Genetic differences between Burkitt lymphoma (BL), diffuse large B-cell lymphoma(DLBCL), and lymphomas intermediate between BL and DLBCL

Genetic features BLIntermediateBL/DLBCL DLBCL

MYC rearrangement Yes (95%) Common RareIG-MYCa Yes ∼30% RareNon-IG-MYCa No ∼20% Rare

BCL2 without MYC rearrangement No Rare 20–30%BCL6 without MYC rearrangement No Rare 30–40%Double hitb No ∼50% RareMYC-simple karyotypec Yes Rare RareMYC-complex karyotypec Rare Common Rare

aIG-MYC includes translocation with immunoglobulin heavy- or light-chain genesbDouble-hit lymphomas contain either a BCL2 or a BCL6 rearrangement in addition to an MYCrearrangementcComplex karyotype by array CGH has more than six abnormalities, while a simple karyotypeshows no or only a few cytogenetic or CGH abnormalities in addition to the MYC rearrangement

Lymphoplasmacytic Lymphoma

Lymphoplasmacytic lymphoma (LPL) is thought to originate from post-follicularcenter B cells that show some capacity to differentiate into plasma cells. Most casesof LPL express IgM and those showing bone marrow involvement and an IgMparaprotein of any concentration meet criteria for the entity termed Waldenström’smacroglobulinemia [1]. The diagnosis of LPL is one of exclusion and can oftenbe difficult to differentiate from marginal-zone lymphoma using only histologyand immunohistochemical staining [1]. However, LPL has no specific recurrentchromosomal or molecular abnormalities; so detecting molecular changes associ-ated with marginal-zone or other lymphomas can help rule out LPL. Contrary toearlier reports, the t(9;14) translocation, which brings PAX5 on chromosome 9under regulation of the IgH gene on chromosome 14, is only occasionally foundin well-characterized LPL [104]. Deletion of 6q may be present in up to 50%of bone marrow-based LPL but is not specific for LPL and is rarely reported intissue-based LPL [105, 106]. However, those cases with 6q deletions have beenreported to have more aggressive disease and worse prognoses [107]. Trisomies of


chromosome 4 have been reported in approximately 20% of cases of Waldenström’smacroglobulinemia [108].

Molecular Testing for T-Cell Non-Hodgkin lymphoma (T-NHL)

T-Cell Biology and Maturation

As with B cells, knowledge of the developing T-cell and T-cell receptor (TCR)rearrangement is important for understanding molecular clonality testing. Most αβ

(alpha beta) T cells are part of the adaptive immune system, responding to a varietyof antigens. T-cell receptors recognize antigens that are presented by major histo-compatibility complex (MHC) class I and II proteins on other cells. Given the largenumber and unpredictable makeup of antigens, mechanisms must be in place to cre-ate a wide range of T-cell receptors (repertoire) with a limited number of genes. Thefirst step in creating this diversity is rearrangement of the individual T-cell recep-tor genes, which occurs in T cells in the thymus. Additional diversity is created byaddition and subtraction of nucleotides at the sites of rearrangement and pairing ofdifferent chains. These differences are responsible for the polyclonal population ofT cells in normal individuals.

Rearrangement of the TCR genes is similar to rearrangement of the immunoglob-ulin genes. TCR beta (TRB) and TCR delta (TRD) are similar to IGH and havevariable, diversity, and joining regions. TCR alpha (TRA) and TCR gamma (TRG)lack the diversity region similar to the immunoglobulin light-chain genes. T-cellreceptors differ from immunoglobulins in that TRB only pairs with TRA and TRDonly pairs with TRG defining two subsets of T cells: alpha beta (αβ) T cells andgamma delta (γδ) T cells. Furthermore, TRD is unique because it is entirely locatedwithin the TRA gene locus and is therefore deleted when TRA rearrangement takesplace (Fig. 6.7).

Lymphoid progenitors migrate from the bone marrow to the thymus, where theyare triggered to become T cells and undergo TCR rearrangement. The earliest T cells

Fig. 6.7 The TRA locus contains 70–80 variable (V) regions and 61 joining (J) regions. TRD haseight V regions interspersed with TRA but can also rearrange with TRA V regions. The three D andfour J regions of TRD are located between the shared variable region and the remainder of TRAand are deleted with rearrangement of TRA


are CD3-negative and CD4- and CD8-negative (double-negative) thymocytes. TRDon chromosome 14q12 rearranges first followed by TRG on chromosome 7p15. If nofurther rearrangement takes place, the cells become γδ (gamma delta) T cells whichremain negative for CD4 and CD8. However, cells that become αβ (alpha beta) Tcells will rearrange TRB on chromosome 7q32. Successful rearrangement of TRBleads to expression of CD4 and CD8 (double-positive thymocytes), proliferation,and subsequent rearrangement of TRA on chromosome 14q12. Several different Tcells with the same TRB will have different TRA rearrangements increasing diver-sity, which approximates 5.8 × 106 for TRB and TRA combinations. Addition andsubtraction of nucleotides adds about 2 × 1011 possibilities for a total of about 1018

αβ (alpha beta) T-cell repertoire [2]. Mature αβ (alpha beta) T cells express eitherCD4 or CD8.

The majority of T cells are αβ (alpha beta) T cells and both TRA and TRB havemuch greater diversity than do TRG and TRD. TRB is a highly complex gene with52 V regions of which 39–47 are functional and two different sets of D and J regions[2, 3] (Fig. 6.8).

Fig. 6.8 The TRB locus contains 52 variable (V) regions (39–47 are functional), a diversity (D)region with six associated joining (J) regions and a constant region (Cβ1), and a second D regionwith seven associated J regions and a constant region (Cβ2). Because of the two separate D and Jregions, it is possible to have two Dβ–Jβ rearrangements on a single allele

Because of the latter there may be two different DB–JB rearrangements present onone allele. The first DB region has six corresponding JB regions, while the secondDB region has seven JB regions. TRA with 70–80 functional V regions and 61 Jregions is also a complex gene [2, 3] (Fig. 6.7). The TRD V regions are interspersedwith the TRA V regions, while the TRD D, J, and C regions are located between theshared V region and the TRA J region (Fig. 6.7). Therefore, when TRA undergoesrearrangement of the VA and JA segments, the TRD regions are lost.

The TRD and TRG loci have less diversity and a more limited repertoire (Figs. 6.7and 6.9). Although TRG rearrangement is present in both αβ (alpha beta) and γδ

(gamma delta) T cells, the TRG gene has 14 variable regions of which only 10 arefunctional, 5 J segments, 2 C segments, and a limited number of N and P nucleotides;therefore, not only is there less diversity but the length differences between differentrearrangements are only 20–30 base pairs versus about 60 base pairs for IGH [2].TRD has three D and four J regions located between the shared V segments andthe TRA J region. TRD has eight V regions, but it can also use some of the TRAV regions [2]. Both the TRG and the TRD loci show preferential rearrangement of


Fig. 6.9 The TRG locus contains 14 variable (V) regions of which only 10 are functional andundergo rearrangement, 5 joining (J) regions, and 2 constant regions (C). Primers are generallydesigned against V1–V8, V9, V10, and V11 (forward arrows) with multiple primers against the Jregions (reverse arrows). JP1, T-cell receptor gamma joining P1; JP, T-cell receptor gamma joiningP; J1, T-cell receptor gamma joining 1; JP2, T-cell receptor gamma joining P2; J2, T-cell receptorgamma joining 2

the variable regions at particular anatomic sites after birth. For example, most γδ

(gamma delta) T cells in the intestines and the spleen express region Vδ1, Vδ2 ispreferred in the skin, and Vγ9/Vδ2 predominates in peripheral blood γδ (gammadelta) T cells in adults [3].

Unlike the immunoglobulin genes, the TCR genes are not commonly involved intranslocations in T-cell leukemias or lymphomas. The exception is T-cell prolym-phocytic leukemia (T-PLL), where about 90% of cases have activation of TCL1Aand TCL1B on 14q32 by inv(14q11–q32) or t(14;14)(q11;q32) and placement nextto TRA [1].

T-Cell Clonality Testing

Clonality testing takes advantage of the fact that once a T cell has rearrangedits TCR genes, every daughter cell will have the same rearrangement. Indicationsfor gene rearrangement clonality testing include evaluation of suspicious T-cellproliferations, lymphoproliferations in immunosuppressed individuals (includingpost-transplant patients), evaluation for minimal residual disease, and comparisonof two lymphoid malignancies for a clonal relationship. Gene rearrangement testingis not lineage specific and should not be used to establish lineage. TCR rearrange-ment occurs in 40–70% of B-ALL, 4–14% of AML, and 2–10% of mature B-cellmalignancies [3, 9, 109, 110].

Southern blot testing usually targets TCRβ because this gene not only allowsthe use of one to two probes but also has a large repertoire of possible rearrange-ments. Southern blot depends on the detection of non-germline (i.e., rearranged)DNA fragments after restriction enzyme digest. TRB gene rearrangement is presentin virtually all αβ (alpha beta) T-cell lymphomas, 95% of CD3-positive T-ALL, and80% of CD3-negative T-ALL [3]. Limitations and the occurrence of false positivesand false negatives are the same as for testing IGH (see B-cell Clonality Testingsection).


PCR for TCR gene rearrangement generally targets TRG because primer designis simpler, (Fig. 6.9) although BIOMED-2 primers also target TRB and TRD. TheTRA gene is so complex that analysis of it is generally not attempted. TRG gene rear-rangements are detected overall in approximately 80%-90% of T-cell lymphomaswith the incidence varying between subtypes and techniques [8]. TRG detectionrates are better (>90%) for T-lymphoblastic leukemia (T-ALL), T-LGL, and T-PLL[3]. Improved overall detection rates of approximately 95% have been seen with theBIOMED-2 primer sets for both TRG and TRB [9, 111, 112]. Testing with the fullBIOMED-2 set of primers against TRG, TRB, and TRD has a sensitivity of 98% andspecificity of 93% for T-cell neoplasms, as long as good quality DNA (>300–400base pairs) is used [9].

There are many different primer designs for detecting T-cell receptor gammagene rearrangements. Primers are generally designed against VG1–8, VG9, VG10,and VG11 regions with multiple primers against the JG regions (Fig. 6.9).Approximately 60%-70% of clones occur in the VG1–8 primer set [11]. Usingmultiple primers against the variable and joining regions increases detection ratesto over 80% [8, 113]. Some PCR methods use PCR primers only against T-cellreceptor gamma joining regions 1 and 2, however, better results are obtained ifprimers against the other joining regions are also employed [3, 8, 113]. The join-ing region primers may be labeled with different fluorescent dyes and multiplexedto decrease the number of tubes and to assist in comparing a peak to the polyclonalbackground [114].

The limited TRG repertoire and more limited size of rearrangements means thata small clonal population that has a common rearrangement may blend in with thebackground and not be detectable or may look clonal. Furthermore, canonical rear-rangements such as those that occur in γδ (gamma delta) T cells may look clonalbecause these cells preferentially rearrange the same V-J segments and limited Nnucleotides are added. Most TRG clones have one or two rearrangements present;however, approximately 10%-15% have more than two rearrangements suggesting asecond minor clone or oligoclones, genomic instability, or aneuploidy [112]. Thesepeaks may be present in a polyclonal background and it may be difficult to determineif the peaks represent a clonal population or a pseudoclonal population.

Clonal peaks should be significantly higher than the polyclonal background;however, what the exact cutoff should be to call a clonal peak is debated. Severalmethods have been proposed that compare a predominant peak to the polyclonalbackground. A commonly used method is to compare the relative peak heights. Onestudy suggested that if the predominant peak is at least three times the height of thepolyclonal background, it should be considered a clonal peak, whereas peaks 1.5–3times the height of the background may be clonal and require further evaluation[115]. However, reactive T-cell lesions may display clonal peaks even using strictcriteria.

The TRD locus has more diversity than its partner; however, it is deleted withrearrangement of TRA on the same allele, and therefore TRD rearrangement ispresent only in about 35% of αβ (alpha beta) T cells. Furthermore, TRD has a morerestricted repertoire and can lead to false-positive clones and should be interpreted


in the context of concomitant TRG and TRB analyses. In the BIOMED-2 studies,TRD did not significantly add to detection, although it could be considered for usein known γδ (gamma delta) T-cell neoplasms, such as hepatosplenic lymphoma orcutaneous γδ (gamma delta) T-cell lymphoma [112].

False-positive and false-negative results occur for the same reasons that theyoccur with immunoglobulin gene rearrangement (see B-cell Clonality Testing sec-tion). Additionally false-positive clones are more common with PCR of TRGbecause of its more limited repertoire and preferential rearrangement [3]. As men-tioned previously, running samples in duplicate or retesting apparently clonal popu-lations will help to differentiate oligoclonal/pseudoclonal populations by amplifyingdifferent peaks on different runs [3, 12]. Clonal T-cell populations can occur inapproximately 5–10% of ostensibly reactive T-cell proliferations and should promptcareful review of histopathology, close follow-up, and additional testing to includere-biopsy of another site particularly in cutaneous lesions [111]. Finding identicalpeaks at two different sites or on two separate runs is strong evidence for a trueclonal process.

Another method to assess clonality of a T-cell population is to perform flow cyto-metric immunophenotyping with antibodies to the variable region of TRB (Vβ (beta)).Commercial antibodies against class-specific sequences of Vβ (beta) are availableand cover approximately 70% of the Vβ (beta) repertoire [116]. The normal distribu-tion of T-cell expression of these Vβ (beta) classes is well defined and substantiallyincreased numbers of T cells expressing a single Vβ (beta) class suggest a clonalT-cell population [117]. This method can be used to assess a subpopulation of Tcells identified by other surface markers or an entire T-cell population (such asCD8-positive T cells). Sensitivities of >90% and a specificity of 80% are achievedwhen evaluating some disorders such as T-cell large granular lymphocyte leukemiaand Vβ (beta) analysis may help when evaluating pseudoclonal TRG PCR results[116, 118]. However, Vβ (beta) analysis by flow cytometry is more prone to falsenegatives when small numbers of neoplastic cells are present [118].

Anaplastic Large-Cell Lymphoma (ALCL)

Anaplastic large-cell lymphoma (ALCL) is defined by cohesive clusters and sheetsof large dysplastic CD4+ T cells that invade the paracortex and sinuses of lymphnodes and strongly express the activation antigen CD30 in a membrane and Golgipattern in virtually every cell [119]. Approximately 30% of systemic ALCL involveextranodal sites such as skin, bone, soft tissue, liver, and lung. Systemic ALCL isdivided into two categories based on the expression of anaplastic lymphoma kinase(ALK), a type II transmembrane receptor tyrosine kinase belonging to the insulinreceptor superfamily [119–121]. ALK– ALCL is morphologically indistinguishablefrom ALK+ ALCL and is separated from peripheral T-cell lymphoma, not otherwisespecified (NOS) based on its strong expression of CD30 in virtually every cell. Atthe molecular level, ALK– ALCL appears to have a distinct genetic profile, although


with some overlapping features with ALK+ ALCL and PTCL-NOS [121]. A thirdentity limited to the skin, primary cutaneous ALCL, is ALK negative (with very rareexceptions), clinically indolent, and will not be discussed in detail [122].

ALK (on chromosome 2p23) is highly conserved across species, is expressed inembryonic neural tissue, and is involved in midgut and neural tissue development[123]. Midkine and pleiotrophin are the putative normal ALK ligands. ALK is notnormally expressed in lymphoid tissue, but as a result of translocation with othergenes, a fusion protein is created that forms homodimers or heterodimers (as isthe case with NPM–ALK) resulting in transphosphorylation and activation of theALK signaling pathway. NPM (5q35), a gene involved in ribosome biogenesis andshuttling between the cytoplasm and the nucleolus, is the most frequent ALK partner(70–85% of cases), but approximately eight other partner genes have been describedin ALCL (Table 6.8).

The genomic breakpoints in ALK are almost invariably located in the intronflanked by exons 16 and 17 with exons 17–26 encoding the intracytoplasmicdomain. The fusion gene is composed of the 5′-end partner fused to the ALK tyrosinekinase domain at the 3′-end (Fig. 6.10).

The subcellular compartmentalization of the fusion protein and particular sig-naling pathway activated are fusion gene dependent (Table 6.8). The interaction ofNPM–ALK with wild-type NPM in the centrosome protein complex may explainthe frequent numerical chromosome aberrations in ALCL through deregulatedphosphorylation of cell-division regulators [124].

Activation of the ALK signaling pathway leads to proliferation, prolongedtumor cell survival, and cytoskeletal rearrangement and cell migration, reviewed

Table 6.8 ALK translocations seen in anaplastic large-cell lymphoma

Cytogeneticabnormality

ALK partnergene ALK staining pattern

Approximatepercentage of cases

t(2;5)(p23;q35) NPM Nuclear, diffusecytoplasmic

80

t(1;2)(q25;p23) TPM3 Diffuse cytoplasmic withmembrane accentuation

10–15

Inv(2)(p23q35) ATIC Diffuse cytoplasmic <5t(2;3)(p23;q21) TFGa Diffuse cytoplasmic <5t(2;17)(p23;q23) CLTC Granular cytoplasmic <5t(2;17)(p23;q35) ALO17 Diffuse cytoplasmic <1t(2;19)(p23;p13.1) TPM4 Diffuse cytoplasmic <1t(2;22)(p23;q11.2) MYH9 Diffuse cytoplasmic <1t(2;X)(p23;q11–12) MSN Membrane staining <1

aThree variants based on different fusion protein lengthsNPM, nucleophosmin gene; TPM3 and TPM4, non-muscular tropomyosin gene; ATIC, amino-terminus of 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase/IMP cyclohy-drolase gene; TFG, TRK-fused gene; CLTC, clathrin heavy polypeptide gene; ALO17, ALKlymphoma oligomerization partner on chromosome 17; MYH9, myosin heavy-chain 9 gene; MSN,moesin gene


Fig. 6.10 Simplified schematic of the molecular structure of ALK, NPM, and ALK fusion genes.With the exception of MSN (moesin)–ALK and MYH9 (myosin heavy chain)–ALK, all ALKfusion proteins contain the entire intracytoplasmic portion (amino acids 1,058–1,620) of ALK.Homodimerization of the fusion protein or heterodimerization with NPM mimics ligand-mediatedaggregation of full-length wild-type ALK, resulting in constitutive activation of the ALK signalingpathway. X, other variant fusion partners (Table 6.8); arrows indicate breakpoint; numbers indicatethe amino acid

by Chiarle et al. [125]. Proliferative effects are primarily the result of activation ofcyclins and enhanced expression of genes such as FOS, JUN, and MYC. NPM–ALKacts as a docking molecule for downstream adaptors [IRS1 (insulin receptor sub-strate), SRC (sarcoma), SHC (SH2 domain-containing transforming protein), andPLCγ (phospholipase C gamma)] that activate the RAS–ERK (extracellular signal-related kinase) pathway. One of the key steps in lymphomagenesis appears to beinteraction of the SHP2/GRB2 (growth factor receptor-bound protein 2) complexwith ALK through SHC to enhance phosphorylation of ERK1/2 through SRC andSOS (son of sevenless) [126]. Recent evidence suggests that IGF-1R (type 1 insulin-like growth factor receptor) tyrosine kinase interacts with NPM–ALK to potentiateits effects and may function as an extracellular domain to maintain the phosphoryla-tion status of NPM–ALK [127]. The effect of ALK on survival is mediated throughthe JAK–STAT pathway (particularly STAT3) that regulates downstream moleculessuch as BCL-2, BCLXL, C/EBPβ (CCAAT/enhancer binding protein beta), survivin,


MCL1 (myeloid cell leukemia sequence 1) and through the PI3K (phosphatidylinos-itol 3-kinase)–AKT pathway with inhibition of FOXO3A (forkhead box O3A) andBAD (BCL-2 agonist of cell death). Interestingly ALK’s effect on the ERK1/2 path-way (increased JUNB expression) and STAT3 also enhance transcription of CD30.Changes in actin filament depolymerization and loss of cell-matrix adhesion resultfrom ALK activation of the Rho family GTPases through phosphorylation of mem-bers of the VAV family of guanine nucleotide exchange factors (GEFs) and maycontribute to the unusual sinus growth pattern of this lymphoma [128].

Deregulated expression of full-length or ALK fusion protein has been describedin non-lymphoid tumor cell lines and tumors from multiple tissue types includ-ing tumors of neural origin (retinoblastoma, neuroblastoma, glioblastoma) or solidtumors such as melanoma, breast carcinoma, and rhabdomyosarcoma [125–131].Approximately 50% of inflammatory myofibroblastic tumors express ALK fusionswith tropomyosin genes TPM3 and TPM4 and less frequently RANBP2 (RAN-binding protein 2), CARS (cysteinyl-tRNA synthetase), and SEC31A (SEC-like 1)[125, 132].

Recent evaluation of genes located near the t(2;5) breakpoints such as the onco-genic AP-1 transcription factor FRA2 (on 2p23) and the HLH protein inhibitor ofdifferentiation ID2 (2p25) and the tyrosine kinase CSF-1 receptor (5q33.1) hasshown upregulation of these genes in both ALK+ and ALK– ALCL [133], sug-gesting that aberrant expression of these genes through unknown mechanisms maycontribute to lymphomagenesis, precede the t(2;5), and create conditions favorablefor translocations to occur.

Detection of ALK Dysregulation

The t(2;5)(p23;q35) and other variant translocations can be detected on routinecytogenetics, but fresh tissue is often not available. RT-PCR methods to detectNPM–ALK fusion transcripts are only rarely used in routine diagnosis due theinability to detect other partners and the more time-consuming technique. Inaddition, NPM–ALK translocations have been detected in normal individuals inperipheral blood and lymph nodes using sensitive RT-PCR techniques [134–136].Real-time quantitative (RQ)-RT-RCR can potentially be used to detect bone mar-row involvement, minimal residual disease, and early relapse in ALCL but has notbeen evaluated in a large number of patients [136, 137].

Immunohistochemical detection using the ALK-1 antibody is the routine methodof detecting ALK protein (CD256). The pattern of staining reflects the nature ofthe fusion protein (and cytogenetic lesion), with diffuse staining in the nucleus(due to heterodimerization of wild-type NPM with oncogenic NPM–ALK fusionprotein) and cytoplasm with NPM–ALK and diffuse or granular cytoplasmic ormembrane staining with other partners (Table 6.8). Caution should be taken ininterpreting immunostaining as other tumors (inflammatory myofibroblastic tumors,rhabdomyosarcoma, and others) may express ALK.


FISH break-apart rearrangement probes targeting sequences on either side of theALK gene breakpoint are routinely used to detect the t(2;5) and variant translo-cations and numerical abnormalities. Additional copies of 2p23 or 2p have beenreported in ALK– ALCL suggesting other mechanisms of ALK dysregulation ina subset of ALCL [138–140]. A rare t(11; 22) type 2 (EWS exon 7/FLI exon5) typical for Ewing sarcoma has been detected by PCR in a ALK+ ALCL withexpression of CD99 [141]. Rare three-way translocations of ALK and NPM withchromosome 3p21 and 13q14 have been described particularly in the small-cellvariant [142].

Other Genetic Abnormalities in ALCL

Secondary chromosomal imbalances have been detected by conventional CGHanalysis in approximately 60–90% of ALK+ and ALK– ALCL, with the specificchromosomal abnormalities being somewhat variable in the limited number of casesinvestigated [139, 140, 143]. Loss of 11q or 13q has been reported in ALK+ andALK– ALCL in approximately 15–30% of cases. Losses of chromosomes 4 [140]and 9p and 10p [139] in ALK+ ALCL and of 6q and 16p in ALK– ALCL [139, 140,143] are reported. Gains of 7p and 6p [139] and 17p and 17q [140] in ALK+ ALCLand of 1q, 3p, 6p, or 8q in approximately 20–50% of ALK– ALCL [140, 143] havebeen reported. Although loss of 13q and gain of 17q are commonly present in otherT-cell lymphomas, other numerical abnormalities such as loss of 9p21-pter, 5q21,or 12q21–22 seen in approximately 30% of PTCL-NOS were detected in less than5% of ALK– or ALK+ ALCL. Sub-megabase resolution tiling (SMRT) array CGHperformed on ALCL cell lines DEL and SR-786 revealed gains of 5p15.32–p14.3,20p12.3–q13.11, and 20q13.2–q13.32 and losses of 18q21.32–18q23 [144].

A limited number of gene expression microarray profiling (GEP) studies havebeen performed. Thompson et al. [145] found that ALK+ ALCL overexpressesgenes encoding signal transduction molecules such as SYK, LYN, CDC37 andunderexpresses transcription factors such as HOXC6 and HOXA3, and both ALK+and ALK– ALCL highly express kinase genes (LCK, protein kinase C, VAV2,and NKIAMRE). GEP of nodal peripheral T-cell lymphoma distinguished ALCLfrom other peripheral T-cell lymphomas [146]. Analysis of 25 ALK+ ALCL and 7ALK– ALCL found that BCL-6, PTPN12 (tyrosine phosphatase), C/EBPβ, andSERPINA1 (alpha-1 antitrypsin) genes were differentially expressed in ALK+ALCL and CCR7, CNTFR, IL22, and IL21 in ALK– ALCL [147]. Further geneontology (GO) analysis (association of gene products with regard to their bio-logic processes, cellular components, and molecular functions) has shown thatthe ALK+ ALCL profile was related to immune response, the IkB kinase/NF-κB cascade, and transendothelial migration pathways. The ALK+ tumors showeddifferent expression profiles based on variant morphology (common versus smallcell), in particular ALCL with variant morphology overexpressed genes involvedin the cell cycle regulation and proliferation and genes encoding proteins involved


in adhesion and migration. The variant histology group also had upregulation ofbiochemical pathways reflecting a hyperactive metabolic state. GEP performed onmicrodissected tumor cells and compared to normal T-cell and NK-cell subsetshas confirmed the upregulation of NF-κB target genes and an activated T-cell phe-notype with loss of T-cell-specific signaling and differentiation molecules [148].Interestingly, the number of genes differentially expressed between ALK+ andALK– ALCL were few and seem to be related to genes activated by ALK signaling,suggesting that these two tumors are closely related. This study also showed thatfew genes were differentially expressed between systemic and primary cutaneousALCL, suggesting that the different clinical behavior may be influenced by themicroenvironment, particularly increased numbers of T-regulatory cells in systemicALCL.

Angioimmunoblastic T-Cell Lymphoma (AITL)

Angioimmunoblastic T-cell lymphoma (AITL) is a tumor of CD4+ T cells with afollicular helper phenotype (CD10+, BCL-6+, CXCL13+, PD-1+) [149]. The tumoris heterogeneous with admixed plasma cells, eosinophils, and histiocytes and anEBV+ large B-cell population being almost invariably present. Other characteristicfeatures include vascular proliferation and expanded follicular dendritic cell mesh-work. Due to the polymorphous nature of the infiltrate resembling a reactive process,clonality studies are important in confirming the presence of a neoplastic popu-lation. Clonal TCR rearrangement is detected in 75–90% of cases; 25–30% haveclonal IGH as well [150–152]. In a small subset, an EBV+ DLBCL may develop asa composite lymphoma or at relapse [153].

Genetic Abnormalities in AITL

The most frequent recurrent cytogenetic abnormalities include +3, +5, and +X[154, 155]. Recurrent breakpoints have been reported at 1p31–32; 3p24–25; 4p13;9q21–22; 12q13; 14q11; 14q32. Compared to PTCL-NOS and ALCL, AITL hasfewer structural abnormalities and lacks polyploidy or gains of chromosome 7 seenin approximately 20% of PTCL-NOS and rarely in ALCL [156].

Comparative genomic hybridization studies in AITL are limited. In one series,complete or partial chromosome gains (the majority being the result of trisomicevents, particularly 5q31q35 in 55% or 21 in 41%) or losses (most commonly 6qin 23%) have been identified in 91 and 36% of AITL, respectively [143]. The +21was consistently associated with a gain of 5. 5q31–32 appears to be distinctive, andgrowth factor and growth factor receptor genes (such as IL-3, PDGFRB) localize tothis area. In another study, Thorns et al. failed to detect significant gains of chromo-somes 3 or 5 or 21 but showed recurrent gains of 22q, 19, and 11q13 and losses of13q [157].


Gene expression profiling has confirmed the CD4+ follicular helper T-cell originof AITL with overexpression of such genes as CXCL13 [chemokine (C–X–C motif)ligand 13], BCL6, PDCD1 (programmed cell death 1), CD40L (CD40 ligand),NFATC1 (nuclear factor of activated T-cell cytoplasmic, calcineurin-dependent 1)[158]. In addition, there appears to be a strong microenvironmental component withoverexpression of B-cell and follicular dendritic cell-related genes, chemokines, andgenes related to the extracellular matrix and vascular biology. A set of eight geneshas been identified that may accurately separate AITL from PTCL-NOS [158]. Inone study, approximately 75% of AITL have elevated levels of c-MAF (cellularhomologue of the transforming gene of the avian retrovirus AS42), a transcriptionfactor on 16q23 belonging to the AP1 superfamily detected by RT-PCR [159].

Mycosis Fungoides (MF)

Mycosis fungoides is a tumor composed of skin homing CD4+ T cells with dysplas-tic cerebriform nuclei that infiltrate the epidermis [160]. The clinical appearancevaries as the tumor progresses from patches to infiltrative plaques and tumors andcorrelates with the number of cells present and the depth of invasion into the der-mis. Several clinical/morphologic variants have been described but as of yet do nothave characteristic recurrent cytogenetic or molecular genetic abnormalities. Theimmunophenotype often shows little loss of pan T-cell antigens except CD7, whichcan be seen in reactive lymphoid infiltrates.

Early lesions of MF often resemble benign dermatoses, and the diagnosis isproblematic relying on the detection of a T-cell clone in the appropriate clini-cal and morphologic context [161, 162]. Rearrangements of TRG and TRB aredetected by PCR in approximately 70–90% of MF using the BIOMED-2 primersand other PCR techniques [163, 164] and are highest in patients with tumors versuspatches/plaques. PCR analysis for TCR gamma rearrangement with denaturing gra-dient gel electrophoresis (DGGE) (or related techniques such as PCR temperaturegradient gel electrophoresis or PCR/single-stranded conformational polymorphismanalysis) with thresholds of detection of 1% increases the sensitivity of clonalitydetection up to approximately 90% [165, 166]. PCR amplification followed byribonuclease protection analysis has a detection rate of 1/100,000 cells but is toosensitive for routine diagnosis as a clone may be detected in microscopically nor-mal blood, bone marrow, on lymph nodes in patients with patch stage MF whonever develop involvement of these sites and have a normal life expectancy [166].Using these more sensitive techniques, clonal T-cell populations have been detectedin reactive lymphoproliferations including skin infiltrates (approximately 2–11% ofcases) [163–165, 167, 168]. These clonal dermatoses may in fact represent a “false-positive” precursor lesion with approximately 20% having a risk of later developingMF [166] or a cutaneous T-cell lymphoid dyscrasia [169]. The presence of a match-ing clone from a different site improves the specificity of the clone; however, thiscan occur in clonal dermatitis as well [164, 170].


Molecular Staging of Mycosis Fungoides (MF)

The detection of a T-cell clone in the peripheral blood should not be consideredperipheral blood involvement by MF unless the clone is identical to that seen inthe skin or other involved sites such as a lymph node [166, 171, 172]. The clonesdetected may represent benign oligoclonal or clonal cytotoxic T-cell large granularlymphocyte (LGL) expansions that are relatively common in the elderly [173, 174]or that represent a restricted LGL response to tumor cell antigens or a mono-clonal T-cell dyscrasia of undetermined significance with or without erythroderma[175, 176]. PCR/DGGE is particularly useful in demonstrating clonal relatednessin that the separation of PCR products relies on nucleotide sequences and impliesan identical product more than PCR alone. In addition, another consideration isthat comparison of DNA from paraffin-embedded tissue versus DNA isolated fromfresh lymphocytes in the blood may not be directly comparable, and there may befalse-negative or false-positive results.

Detection of a T-cell clone in histologically uninvolved lymph nodes frompatients with MF is associated with a reduced probability of survival and appearsto be useful in distinguishing tumor involvement from benign dermatopathic lym-phadenopathy [177, 178]. However, the molecular detection of a clone by PCR inhistologically uninvolved lymph nodes has not been demonstrated to be statisti-cally significant in multivariate analysis [179]. Currently the International Societyfor Cutaneous Lymphomas (ISCL) and the European Organization of Researchand Treatment of Cancer (EORTC) recommend that nodal rating still be basedprincipally on histopathology with the recommendation that lymph nodes lack-ing architectural effacement (NCI/VA LN3 or Dutch grade 2) be divided into twogroups, N2A (clone negative) and N2B (clone positive), based on TCR results todetermine the prognostic significance between patients with N2B versus N3 (partialeffacement of node architecture with many atypical cerebriform lymphocytes, NC1-VA LN4) [180]. Detection of a T-cell clone in the bone marrow does not appear toprovide additional prognostic value over examination of the peripheral blood [181].

Genetic Abnormalities in Mycosis Fungoides

Genomic instability with numerous random and non-random structural chromosomeabnormalities has been reported in up to 70% of MF by both conventional cytoge-netics and more sophisticated studies. These structural abnormalities have primarilybeen detected in late-stage disease as it is more difficult to culture or isolate cells inearly patch/plaque MF. Most frequent abnormalities involve chromosomes 1, 8, 9,10, 11, 12, and 17, and include loss of chromosome material at 1p22 and 1p36, and9p21, 10q, and 17p [182–184]. The involvement of regions containing the T-cellreceptor subunits is observed rarely. Reciprocal translocation is very rare in MF;a t(3;9) has been reported in one case of the granulomatous slack skin variant ofMF [185].


Gene expression profiling comparing MF and inflammatory dermatosis hasrevealed upregulation of 10 genes involved in anti-apoptotic signaling with inhi-bition of proapoptotic pathways through tumor necrosis factor receptor 1 (TNFR1).Interleukin-2 signaling may also be critical in MF pathogenesis through activationof JAK2 and STAT4 and induction of oncogenes such as MYC, LYN, and HCK [186].Unsupervised hierarchical clustering has revealed two major subclasses, one beingmore aggressive including tumoral-stage MF and the other less agressive (plaquestage). Six genes were identified that could be used as a prediction model to distin-guish MF and inflammatory conditions in 97% of cases. Signatures associated withabnormal immunophenotypes and with tumoral-stage disease were also detected.Transcriptional profiling of MF at various clinical stages has revealed three clus-ters with distinct clinical and biologic characteristics: patients in cluster 1 had theworst prognosis/tumor stage disease and upregulation of genes involved in lym-phocyte activation and the TNF pathway; patients in cluster 2 had less aggressivedisease and upregulation of genes associated with epidermal development; and clus-ter 3 patients had more extensive disease, decreased event-free survival, and poorresponse to therapy and increased expression of genes involved in inflammation andbenign epidermal hyperproliferation [187].

Conventional comparative genomic hybridization studies have shown chromo-some imbalances in 56–94% of MF cases and corroborate some of the abnormalitiesdetected on cytogenetic studies [188–190]. Chromosome imbalances are stagerelated with the highest incidence with large-cell transformation. DNA losses havebeen reported at 1p (38%), 10q (15%), 13q (<10%), 17p (21%), and 19 (15%) [189].DNA gains involve 4q (18%), 18 (15%), and 17q (12%). 1p33–36 and 10q26 mayrepresent regions of minimal recurrent deletion; on chromosome 1p, two regions ofminimal common deletion at 1p36 (D1S228 marker) and 1p22 (D1S2766 marker)have been defined by allelotyping [189]. Loss of heterozygosity (LOH) studieshas also identified abnormalities in 1p22 (putative tumor suppressor gene), 9p21(p15/CDKN2B and p16/CDKN2A), 10q23 (PTEN), and 17p13 (TP53), the mostcommon being in 10q23, particularly in tumor stage MF [191]. Alterations in p15and p16/CDKN2A (allelic loss or promoter hypermethylation) have been identifiedin all stages of MF, being somewhat more frequent in tumor stage [192, 193]. Intransformed MF, loss of 17p and gain of 17q can result from the formation of i(17q)with subsequent loss of TP53 and overexpression of genes on 17q such as STAT3[188]. Evidence of widespread epigenetic instability, with promoter hypermethyla-tion of multiple tumor suppressor genes, has been reported, again particularly inadvanced stage disease [194]. Amplification of JUNB (19p13.2), a member of theAP-1 transcription factor complex, has been reported in MF, Sézary syndrome (SS),and ALCL [195].

A recent study of array CGH performed on tumor stage MF with evidence oflarge-cell transformation (>25% large cells) revealed gains of 7q36 and 7q21–7q22and losses of 5q13 and 9p21 [196]. Integration of array CGH data with expres-sion array from the same patients revealed that the most frequent copy numberabnormality, gain of 7q36, is associated with increased expression of FASTK, a ser-ine/threonine protein kinase that attenuates apoptosis. Loss of 9p21 is associated


with diminished expression of p16/CDKN2A and 13q14 with decreased RB1 andhypermethylation of DLEU1; this region also contains miRNA15A and miRNA16-1 which also have tumor suppressor properties. Gain of 1q21–22 is associated withhigher expression of MCL-1 (myeloid cell leukemia sequence 1) which modulatesglucocorticoid resistance; glucocorticoid sensitivity can be restored by the mTORinhibitor rapamycin suggesting another therapeutic modality for tumors with thisabnormality. Loss of 9p21, gain of 8q24.3, or gain of 1q21–1q22 was associatedwith lower survival rates in this study. In a previous publication, the same authorsfound that SS is characterized by gain of 17q22–25 and 8q22–24 and loss of 17p13and 10q25, with amplification of the MYC gene seen in 75% of SS patients [197]supporting the hypothesis that MF and SS are distinct diseases.

Sézary Syndrome (SS)

Sézary syndrome is a rare disease defined by the presence of erythroderma (oftenassociated with pruritus), generalized lymphadenopathy, and the presence of anidentical T-cell clone in the involved tissues and blood. An absolute Sézary cellcount of 1,000 or more cells per cubic millimeter, or a CD4/CD8 ratio of 10 ormore, or loss of one or more pan T-cell antigens other than CD7 (CD2, CD3, CD4,and/or CD5) must be identified in the peripheral blood T cells. A monotonous cere-briform cell infiltrate in the skin is present, but epidermotropism may be absent orminimal. Bone marrow involvement may be sparse and predominantly interstitial.Patients with MF may rarely be erythrodermic but usually are distinguished from SSby slow onset with an antecedent history of patch or plaque-type lesions and fewerabnormal cells in the peripheral blood. Although SS and MF are considered distinctentities in the current WHO classification, there is no distinct pathogenetic markerin either, and there is some overlap in clinical and pathologic features as well asgenetic abnormalities.

Genetic Abnormalities in Sézary Syndrome

Similar to MF, complex karyotypes are seen with many numerical and structuralabnormalities particularly involving deletions of portions of chromosomes 1p, 6q,10q [particularly 10(q22.3–q26.13)], and 13q and additions of 17p and 19 [189,198–200]. Unbalanced translocations between chromosomes 8 and 17 have beenreported in a small number of patients [183, 198].

Gene expression profiling has not yielded entirely uniform and reproduciblemolecular signatures in SS but has confirmed some previous observations andidentified novel molecular abnormalities. Sézary cells have a Th2 pattern of dif-ferentiation with suppression of Th1 differentiation through decreased STAT4 andincreased GATA3 [201, 202]. Pro-survival and anti-apoptotic signaling is important.CGH has revealed aberrant high expression of the tyrosine kinase receptor EpHA4


(phosphorylation of STAT3) and the transcription factor TWIST that prevents MYC-induced apoptosis by antagonizing TP53 [203]. Upregulation of molecules such asCD01 (cysteine dioxygenase, the rate-limiting step in the synthesis of taurine, whichprotects T cells from apoptosis through CD95 and is regulated by MYC) and DNM3(regulated by TWIST1) appears to distinguish skin lesions of SS from MF [204].RT-PCR assay for three upregulated genes (PLS3, DNM3, CD01) and one downreg-ulated gene (STAT4) may potentially prove to be highly sensitive and specific forthe diagnosis of SS [204, 205].

Array CGH and correlative quantitative PCR studies interpreted in the con-text of previous genetic studies suggest that at least three molecular mechanismsare involved in the pathogenesis of SS [197]. Dysregulation of MYC due to gainof MYC, loss of MYC antagonists MX11 (10q25) and MNT (17p13), or distur-bances in MYC induced apoptosis with loss of BIM (2q12) or genetic lesions inTP53, TWIST, or CDKN2A, which have been previously described. In addition,FAS (10q24), which is a key regulator in mature T cells, can be lost. Potentiationof IL-2, IL-7, and IL-15 signaling resulting in STAT3/STAT5 phosphorylation anddeletions of inhibitors of IL-2 (DUSP5 and TCF8) is seen in the majority of patientsand leads to uncontrolled proliferation. The loss of chromosomal regions con-taining TP53 and genome maintenance genes (RPA1 impairs double-strand breakrepair and HICI leads to de-acetylation of TP53) likely contributes to the chro-mosomal instability seen in SS patients. Recurrent duplication of 17q11.2∼q12 (aregion containing STAT5 and ERBB2, alias HER2/neu) has been identified in earlystage MF and SS and suggested to be an early event in the pathogenesis of theseneoplasms [206].

Hepatosplenic T-Cell lymphoma (HSTL)

Hepatosplenic T-cell lymphoma is a rare, aggressive extranodal, cytotoxic, T-celllymphoma with a predominantly CD4–, CD8–/+ gamma delta T-cell phenotype[207]. Patients are typically young males with hepatosplenomegaly, little or noadenopathy, and frequent bone marrow involvement. Approximately 20% of thetumors arise in chronically immunosuppressed patients.

At the molecular level, HSTL shows biallelic rearrangement of the TRG genes.In the small number of HSTL with an alpha beta T-cell phenotype, the TRB genesare rearranged. Non-productive TRB gene rearrangements are seen in some gammadelta HSTL. It should be remembered that a T cell is designated as a gamma deltaT cell based on the expression of the gamma delta T-cell receptor proteins, not onmolecular receptor rearrangements.

Isochromosome 7q is present in most cases but is not entirely pathognomonic asit can be seen in other lymphomas [156, 208, 209]. Multiple copies of isochromo-some 7q can be seen with progression [210]. Other mechanisms of 7q amplificationincluding ring chromosomes have been reported [211]. Trisomy 8 and loss of a sexchromosome may also be present.


Gene expression profiling in a small number of cases has shown a signaturedifferent from AITL and gamma delta T-cell lymphomas arising at other sites[212]. Overexpressed genes included those related to natural killer cell-associatedmolecules such as killer cell immunoglobulin-like molecules (KIRs), CD16 (genesFCGR3B and FCGR3A), and KLRC4 (gene for protein NKG2F, a killer cell lectin).GO analysis revealed enriched molecules for cellular defense response, signaltransduction, receptor activity, and IgG binding.

Enteropathy-Associated T-Cell Lymphoma (EATL)

Enteropathy-associated T-cell lymphoma is uncommon in most parts of the worldexcept in northern Europe, where celiac disease is more prevalent [213]. The tumorpresents as mucosal masses, most commonly in the small intestine (jejunum). Thetumor is composed of intraepithelial CD5–, predominantly alpha beta T cells andcan be divided into two subtypes based on cell morphology and immunopheno-type (classical EATL with variable sized, CD4–, CD8–/+, CD56– T cells in patientswith celiac disease versus type II (monomorphic) EATL with small to mediummonomorphic, CD4–, CD8+/–, CD56+ T cells). There is a strong association withthe HLA-DQ2/DQ8 phenotype in classical EATL.

Cytogenetic studies reveal that most tumors (classical and type II) have com-plex segmental amplifications of the 9q31.3-qter region or deletions of 16q12.1[214–216]. The classical EATL has more frequent gains of 1q and 5q, while type IIEATL has more 8q24 (MYC) amplifications. Patients with refractory celiac diseasewith gains of 1q and monoclonal TRG rearrangement and decreased expression ofCD8 in T cells in the surrounding enteropathic mucosa likely have intraepithelialT-cell lymphoma or in situ EATL [217].

Extranodal Natural Killer-/T-Cell Lymphoma

True natural killer (NK) cell malignancies are rare in the Western world and moreprevalent in the Far East, Mexico, and South America [218]. Due to their infrequentnature and overlap in morphology and immunophenotype with cytotoxic T-cellneoplasms, their diagnosis is difficult. The prototypic NK-cell malignancy is theextranodal NK-/T-cell lymphoma, nasal type. As the name implies, the cell of ori-gin is an activated NK cell or a cytotoxic T cell. Both NK cells and cytotoxic Tcells express cytotoxic granule protein TIA-1, granzyme B, and perforin. There arecurrently no lineage-specific NK-cell markers; NK-cell lineage determination relieson the absence of T-cell receptor gene rearrangements and the lack of surface CD3,CD5, and T-cell receptor proteins alpha beta or gamma delta that would be expressedin cytotoxic T cells. Extranodal NK-/T-cell lymphoma, nasal type most commonlypresents in the nose or paranasal sinuses but can involve the skin, soft tissue, gas-trointestinal tract, and testis. Characteristic pathologic features include invasion and


destruction of blood vessels by small, medium, or large tumor cells with areas ofzonal necrosis. Tumor cells are CD4–, CD8–/+, CD16+, CD56+, express cytotoxicgranule markers, and most are EBV+.

Extranodal NK-/T-cell lymphoma (and other NK-cell malignancies) does nothave specific recurrent translocations. Many cytogenetic abnormalities have beenreported [219–221], the most frequent being del(6)(q21q25) or i(6)p10. As TCRrearrangements are not present in NK-cell malignancies, cytogenetic or moleculargenetic abnormalities are the only tests available to demonstrate clonality in NK-cellmalignancies and may be useful in distinguishing fulminant reactive proliferationsof cytotoxic T cells or NK cells as seen in hemophagocytic lymphohistiocytosis.

CGH and LOH studies of NK-cell malignancies in general have shown manyabnormalities, some of which are recurrent, and include gain of 1q31.3-qterand other regions in 1q, 2q13–q14 and 2q31.1–q32.2, 7q11.2 and 7q31.1–q31.2,17q21.1, and 20pter-qter and loss of 1p36.23–p36.33, several regions in 6q, 4q12,5q34–q35.3, 7q21.3–q22.1, 9p21.3–p22.1, 11q22.3–q23.3, 13q14.11, 15q11.2–q14,and 17p13–p13.1 [222–224]. Nakashima et al. [223] have found that abnor-mal regions preferentially detected in extranodal NK-/T-cell lymphoma, nasaltype include gains of 2q11.2–q37 and losses of 6q16.1–q27, 11q22.3–q23.3, and4q31.3–q32.1 and in aggressive NK-cell leukemia gains of 1q23.1–q24.2 and1q31.3–q44 and loss of 17p13.1. Several regions in 6q are deleted, the most com-mon being 6q21. GEP has shown three genes in the 6q21 deleted region thathave decreased expression; these include PRDM1 (PR domain zinc finger protein1), Blimp-1 (B-lymphocyte induced maturation protein, a transcriptional repres-sor and master regulator of B-cell differentiation), ATG5 (autophagy 5, involvedin IFN-γ-induced autophagic cell death), and AIM1 (absent in melanoma, tumorsuppressor gene implicated in melanoma); highly methylated CpG islands 5′ ofPRDM1 and AIM1 correlate with their decreased expression [224]. Loss in the9p22.1–p22.3 region includes tumor suppressor genes CDKN2B (p15INK4B) andCDKN2A (p16INK4A). Mutations in the 17p region may result in the loss of tumorsuppressor genes other than TP53. Recent Northern blot and QT-PCR studies haveshown that miRNA21 (located on 17q23) and miRNA155 (located on 21q21) areoverexpressed in NK-cell lymphoma/leukemia and dysregulate AKT signaling viarepression of PTEN (phosphatase and tensin homologue) and SHIP1 (SH2 domaincontaining inositol-5-phosphatase), respectively, as a possible mechanism for lym-phomagenesis [225]. In other focused studies, partial deletion of FAS or mutation inTP53, beta-catenin, K-RAS or C-KIT, changes of unknown significance, have beenidentified in extranodal NK-/T-cell lymphoma, nasal type [226–228].

Peripheral T-Cell Lymphoma, Not Otherwise Specified(PTCL-NOS)

Peripheral T-cell lymphoma not otherwise specified is a heterogeneous group ofT-cell lymphomas that do not fit into well-defined categories. PTCL-NOS predom-inantly arises from a CD4+ T cell involving lymph nodes, but CD8+ tumors and


extranodal disease are described. More than half of the cases have complex kary-otypes without specific cytogenetic lesions. Recurrent losses of 4q, 5q, 6q, 13q, and9p, 10p, 10q, and 12q and 13q and recurrent gains of 1q, 3p, 5p, 7q22-qter, 8q,17q, and 22q are described with variable detection between series [139, 143, 157].It should be noted that abnormalities in 1p36, 6q21, 7p15, 11q13, 14q11.2, 14q32,16q22, 16q24, 17p13-q25, 19q13, 22q11.2, and 22q13 may be seen in PTCL-NOS,AITL, or ALCL.

A recent study by Nelson et al. found that the most frequent abnormality is gainof 7q (minimal overlapping region [MR] 7q22q31) (33%) followed by losses on 6q(MR 6q22q24) (26%) and 10p (MR1013pter) (26%) [143]. Gains of 7q are presentin several benign and malignant neoplasms and in apparently normal tissue andmay be associated with disease progression. Loss of 6q is a common finding innon-Hodgkin lymphoma, including PTCL where the area clusters at 6q21. In PTCL-NOS, the minimal region extends distally from 6q22–6q24 [143]. Translocationsinvolving 14q11.2, the site of the TCR alpha/TCR delta locus, and 11q23 (MLLregion) rearrangements are also found.

GEP has shown deregulation of genes involved in proliferation, apoptosis, matrixremodeling, cell adhesion, and transcriptional regulation [146, 158, 229, 230].Although PTCL-NOS in most cases is distinct from ALCL and AITL, there is over-lap. It appears that a subset of CD30– PTCL-NOS may include some lymphomasderived from AITL or lymphomas with a follicular helper T-cell origin that have apathogenesis distinct from AILT [158]. Molecular subgroups within PTCL-NOS donot correlate with a CD4+ helper or CD8+ suppressor phenotype [229]. A prolifer-ation signature with expression of genes associated with the cell cycle [e.g., CCNA(cyclin A2), CCNB (cyclin B1), TOP2A topoisomerase DNA II alpha] and PCNA(proliferating cell nuclear antigen) correlates with a shorter survival and is inverselyrelated to inflammatory response genes and genes regulating a T-cell-specific pro-gram [230]. Based on GEP and using a multi-class predictor, PTCL-NOS can bedivided into three molecular subgroups: one with a poor outcome based on theexpression of genes such as CCND2; another with overexpression of genes involvedin T-cell activation and apoptosis, including NFKB1 and BCL-2; and lastly a groupwith overexpression of genes in the IFN/JAK/STAT pathway.

Hodgkin Lymphoma

Hodgkin lymphomas typically present in young adults as lymphadenopathy, espe-cially in the cervical region. Morphologically, the tumor cells are generally sparseand scattered within a background of non-neoplastic inflammatory cells. Molecularabnormalities are often masked by the prominent inflammatory background unlessthe neoplastic cells are isolated for testing (such as by microdissection). Two maindisease entities are recognized: classical Hodgkin lymphoma (CHL) and nodu-lar lymphocyte-predominant Hodgkin lymphoma (NLPHL), which differ in theirclinical, morphologic, and molecular features.


Nodular Lymphocyte-Predominant Hodgkin Lymphoma (NLPHL)

A diagnosis of NLPHL requires identification of the neoplastic lymphocyte-predominant (LP) cells within an inflammatory background and with an at leastpartially nodular architecture. The folded or multilobate nuclei of LP cells, whichgive rise to the moniker “popcorn” cells, generally contain multiple nucleoli thatare smaller than CHL Reed–Sternberg (RS) cells; however, occasionally LP cellscan be morphologically indistinguishable from classic RS cells. The immunophe-notype differs greatly as LP cells express CD45 and normal B-cell markers such asCD20 and CD79a and lack CD15 and CD30. The background shows a folliculardendritic meshwork with small B cells and CD3+, CD57+ T cells form rosettesaround the LP cells. The T-cell immunophenotype resembles germinal center Tcells with BCL6 and MUM1 expression. The neoplastic LP cells are thought tobe derived from centroblastic germinal center B cells as they express BCL6, haverearranged immunoglobulin genes, and show evidence of ongoing somatic hyper-mutation. Aberrant somatic hypermutation is found in approximately 80% of cases,most commonly in PAX5 and to a lesser extent in PIM1, Rho/TTF, and MYC [231].BCL6 rearrangements are present in about one-half of cases and have a variety ofpartners including immunoglobulin genes, IKAROS, and ABR [59, 60].

Classical Hodgkin lymphoma (CHL)

Four subtypes of CHL are recognized which differ in morphology (especially thecomposition of the background cells), clinical characteristics, and frequency of EBVassociation; however, the neoplastic cells share the same immunophenotype andgenetics. The neoplastic cells are composed of classic RS cells and mononuclearvariants termed Hodgkin (H) cells. Despite the fact that the HRS cells are B cellsin >98% of cases, they lack or have weak staining for many B-cell markers and arenegative for CD45. CD79a is usually absent; CD20 shows variable staining of theneoplastic cells in 30–40% of cases. PAX5 almost always stains the neoplastic cells,but staining is often weaker than in normal B cells. The cells are classically positivefor CD30 and CD15 in a membranous and Golgi pattern and also express MUM1.Immunoglobulin gene rearrangement is present in the HRS cells in >98% of casesand the VH region contains a high number of somatic mutations without evidenceof ongoing mutation; therefore the cell of origin is thought to be a mature germi-nal center B cell in these cases [24]. Rare cases of CHL express T-cell markers andapproximately 85% of these demonstrate clonal immunoglobulin gene rearrange-ment in the RS cells and are thought to be of B-cell origin with aberrant expressionof T-cell markers [232, 233]. The remaining cases lack immunoglobulin gene rear-rangement and instead show TRG rearrangement in the HRS cells supporting a rareT-cell origin [232, 233].

Although no recurrent cytogenetic abnormalities are described for CHL, ane-uploid and hypertetraploid karyotypes are frequently found. Smaller gains andamplifications have been discovered on 2p, 9p, 12q, 4p16, and 4q23–q24 by CGH


analysis of a few cases [234]. Using gene sequencing, aberrant somatic hypermu-tation is found in approximately 55% of cases, most commonly in MYC and to alesser extent in PIM1, Rho/TTF, and PAX5 [231].

Gene expression profiling (GEP) of the background cells can differentiate CHLfrom T-cell-rich, histiocyte-rich DLBCL and shows a different signature for EBV-positive and EBV-negative cases of CHL [235]. EBV-positive cases express anti-viral-related genes associated with activated T cells, especially Th1-positive cells,and macrophages [235]. Outcome has also been correlated with certain gene expres-sion profile signatures. Expression of apoptotic genes and B-cell-related genes,such as BCL11A and CCL21, is correlated with a good outcome [235]. Conversely,expression of extracellular matrix and stromal remodeling genes, such as collagens,is associated with an unfavorable outcome [235]. Immunohistochemistry can beused as a surrogate for these expression profiles. Increased background cells stain-ing with CD20 and BCL11A (indicating increased reactive B cells) and increasedFOXP3-positive T-regulatory cells are associated with a good outcome, whileincreased TIA1-positive or topoisomerase IIa reactive background T cells are asso-ciated with an unfavorable outcome [235]. GEP of the HRS cells shows a distinctsignature that is most similar to EBV-transformed B cells and ABC-type DLBCLand has upregulation of several genes including fascin. There is downregulation ofseveral B-cell genes, such as CD19, CD20, CD52, and TNFRSF17 (BCMA), leadingto the weak expression of classic B-cell markers [236]. Downregulation of CD19,CD79a, and IGH has been shown to be due to inhibition of transcription factor TCF3(E2A) by Id2 and MSC (ABF-1) [237]. Furthermore, inhibition of TCF3 leads toupregulation of several T-cell and macrophage genes [237]. The overlap with ABC-type DLBCL included several genes (cyclinD2, IRF-4/MUM1, CCR7, IκBα, andcFLIP) that promote proliferation and inhibit apoptosis and are regulated by nuclearfactor kappa B (NF-κB) and HRS cells show constitutive activation of NF-κB [237].

References

1. Swerdlow SH, Campo E, Harris NL, et al. eds., WHO Classification of Tumours ofHaematopoietic and Lymphoid Tissues. 4th ed. Lyon: IARC Press; 2008.

2. Murphy KM, Travers P, Walport M. Janeway’s Immunobiology. 7th ed. New York: GarlandScience; 2008.

3. van Dongen JJ, Langerak AW, Bruggemann M, et al. Design and standardization of PCRprimers and protocols for detection of clonal immunoglobulin and T-cell receptor generecombinations in suspect lymphoproliferations: report of the BIOMED-2 Concerted ActionBMH4-CT98-3936. Leukemia. 2003;17:2257–2317.

4. Stavnezer J, Guikema JE, Schrader CE. Mechanism and regulation of class switch recombi-nation. Annu Rev Immunol. 2008;26:261–292.

5. Kotnis A, Du L, Liu C, Popov SW, Pan-Hammarstrom Q. Non-homologous end joining inclass switch recombination: the beginning of the end. Philos Trans R Soc Lond B Biol Sci.2009;364:653–665.

6. Bellan C, Lazzi S, Hummel M, et al. Immunoglobulin gene analysis reveals 2 distinctcells of origin for EBV-positive and EBV-negative Burkitt lymphomas. Blood. 2005;106:1031–1036.


7. Liu M, Schatz DG. Balancing AID and DNA repair during somatic hypermutation. TrendsImmunol. 2009;30:173–181.

8. Teng G, Papavasiliou FN. Immunoglobulin somatic hypermutation. Annu Rev Genet.2007;41:107–120.

9. Boeckx N, Willemse MJ, Szczepanski T, et al. Fusion gene transcripts and Ig/TCR gene rear-rangements are complementary but infrequent targets for PCR-based detection of minimalresidual disease in acute myeloid leukemia. Leukemia. 2002;16:368–375.

10. Szczepanski T, Pongers-Willemse MJ, Langerak AW, et al. Ig heavy chain gene rearrange-ments in T-cell acute lymphoblastic leukemia exhibit predominant DH6–19 and DH7–27gene usage, can result in complete V-D-J rearrangements, and are rare in T-cell receptoralpha beta lineage. Blood. 1999;93:4079–4085.

11. Tan BT, Seo K, Warnke RA, Arber DA. The frequency of immunoglobulin heavy chain geneand T-cell receptor gamma-chain gene rearrangements and Epstein–Barr virus in ALK+ andALK– anaplastic large cell lymphoma and other peripheral T-cell lymphomas. J Mol Diagn.2008;10:502–512.

12. Liu H, Bench AJ, Bacon CM, et al. A practical strategy for the routine use of BIOMED-2 PCR assays for detection of B- and T-cell clonality in diagnostic haematopathology. BrJ Haematol. 2007;138:31–43.

13. Evans PA, Pott C, Groenen PJ, et al. Significantly improved PCR-based clonality test-ing in B-cell malignancies by use of multiple immunoglobulin gene targets. Report of theBIOMED-2 Concerted Action BHM4-CT98–3936. Leukemia. 2007;21:207–214.

14. Nikiforova MN, Hsi ED, Braziel RM, et al. Detection of clonal IGH gene rearrangements:summary of molecular oncology surveys of the College of American Pathologists. ArchPathol Lab Med. 2007;131:185–189.

15. Pai RK, Chakerian AE, Binder JM, Amin M, Viswanatha DS. B-cell clonality determinationusing an immunoglobulin kappa light chain polymerase chain reaction method. J Mol Diagn.2005;7:300–307.

16. Martin-Jimenez P, Garcia-Sanz R, Balanzategui A, et al. Molecular characterization of heavychain immunoglobulin gene rearrangements in Waldenstrom’s macroglobulinemia and IgMmonoclonal gammopathy of undetermined significance. Haematologica. 2007;92:635–642.

17. Lopez-Guillermo A, Cabanillas F, McDonnell TI, et al. Correlation of bcl-2 rearrange-ment with clinical characteristics and outcome in indolent follicular lymphoma. Blood.1999;93:3081–3087.

18. Buchonnet G, Jardin F, Jean N, et al. Distribution of BCL2 breakpoints in follicular lym-phoma and correlation with clinical features: specific subtypes or same disease? Leukemia.2002;16:1852–1856.

19. Ott G, Katzenberger T, Lohr A, et al. Cytomorphologic, immunohistochemical, and cyto-genetic profiles of follicular lymphoma: 2 types of follicular lymphoma grade 3. Blood.2002;99:3806–3812.

20. Gu K, Chan WC, Hawley RC. Practical detection of t(14;18)(IgH/BCL2) in follicularlymphoma. Arch Pathol Lab Med. 2008;132:1355–1361.

21. Aster JC, Longtine JA. Detection of BCL2 rearrangements in follicular lymphoma. AmJ Pathol. 2002;160:759–763.

22. Summers KE, Goff LK, Wilson AG, Gupta RK, Lister TA, Fitzgibbon J. Frequency of theBcl-2/IgH rearrangement in normal individuals: implications for the monitoring of diseasein patients with follicular lymphoma. J Clin Oncol. 2001;19:420–424.

23. Schmitt C, Balogh B, Grundt A, et al. The bcl-2/IgH rearrangement in a population of204 healthy individuals: occurrence, age and gender distribution, breakpoints, and detectionmethod validity. Leuk Res. 2006;30:745–750.

24. Leich E, Salaverria I, Bea S, et al. Follicular lymphomas with and without translocationt(14;18) differ in gene expression profiles and genetic alterations. Blood. 2009;114:826–834.

25. Hartmann EM, Ott G, Rosenwald A. Molecular biology and genetics of lymphomas.Hematol Oncol Clin North Am. 2008;22:807–823, vii.


26. Katzenberger T, Ott G, Klein T, Kalla J, Muller-Hermelink HK, Ott MM. Cytogenetic alter-ations affecting BCL6 are predominantly found in follicular lymphomas grade 3B with adiffuse large B-cell component. Am J Pathol. 2004;165:481–490.

27. Voorhees PM, Carder KA, Smith SV, Ayscue LH, Rao KW, Dunphy CH. Follicular lym-phoma with a Burkitt translocationpredictor of an aggressive clinical course: a case reportand review of the literature. Arch Pathol Lab Med. 2004;128:210–213.

28. Schwaenen C, Viardot A, Berger H, et al. Microarray-based genomic profiling reveals novelgenomic aberrations in follicular lymphoma which associate with patient survival and geneexpression status. Genes Chromosomes Cancer. 2009;48:39–54.

29. Jares P, Colomer D, Campo E. Genetic and molecular pathogenesis of mantle cell lym-phoma: perspectives for new targeted therapeutics. Nat Rev Cancer. 2007;7:750–762.

30. Kienle D, Krober A, Katzenberger T, et al. VH mutation status and VDJ rearrangement struc-ture in mantle cell lymphoma: correlation with genomic aberrations, clinical characteristics,and outcome. Blood. 2003;102:3003–3009.

31. Wiestner A, Tehrani M, Chiorazzi M, et al. Point mutations and genomic deletions inCCND1 create stable truncated cyclin D1 mRNAs that are associated with increasedproliferation rate and shorter survival. Blood. 2007;109:4599–4606.

32. Fernandez V, Hartmann E, Ott G, Campo E, Rosenwald A. Pathogenesis of mantle-cell lym-phoma: all oncogenic roads lead to dysregulation of cell cycle and DNA damage responsepathways. J Clin Oncol. 2005;23:6364–6369.

33. Fu K, Weisenburger DD, Greiner TC, et al. Cyclin D1-negative mantle cell lym-phoma: a clinicopathologic study based on gene expression profiling. Blood. 2005;106:4315–4321.

34. Cheuk W, Wong KO, Wong CS, Chan JK. Consistent immunostaining for cyclin D1 can beachieved on a routine basis using a newly available rabbit monoclonal antibody. Am J SurgPathol. 2004;28:801–807.

35. de Leon ED, Alkan S, Huang JC, Hsi ED. Usefulness of an immunohistochemical panelin paraffin-embedded tissues for the differentiation of B-cell non-Hodgkin’s lymphomas ofsmall lymphocytes. Mod Pathol. 1998;11:1046–1051.

36. Robillard N, Avet-Loiseau H, Garand R, et al. CD20 is associated with a smallmature plasma cell morphology and t(11;14) in multiple myeloma. Blood. 2003;102:1070–1071.

37. Li JY, Gaillard F, Moreau A, et al. Detection of translocation t(11;14)(q13;q32) inmantle cell lymphoma by fluorescence in situ hybridization. Am J Pathol. 1999;154:1449–1452.

38. Remstein ED, Kurtin PJ, Buno I, et al. Diagnostic utility of fluorescence in situ hybridizationin mantle-cell lymphoma. Br J Haematol. 2000;110:856–862.

39. Belaud-Rotureau MA, Parrens M, Dubus P, Garroste JC, de Mascarel A, Merlio JP. A com-parative analysis of FISH, RT-PCR, PCR, and immunohistochemistry for the diagnosis ofmantle cell lymphomas. Mod Pathol. 2002;15:517–525.

40. Katz RL, Caraway NP, Gu J, et al. Detection of chromosome 11q13 breakpoints by inter-phase fluorescence in situ hybridization. A useful ancillary method for the diagnosis ofmantle cell lymphoma. Am J Clin Pathol. 2000;114:248–257.

41. Katzenberger T, Kienle D, Stilgenbauer S, et al. Delineation of distinct tumour profilesin mantle cell lymphoma by detailed cytogenetic, interphase genetic and morphologicalanalysis. Br J Haematol. 2008;142:538–550.

42. Camacho E, Hernandez L, Hernandez S, et al. ATM gene inactivation in mantle celllymphoma mainly occurs by truncating mutations and missense mutations involving thephosphatidylinositol-3 kinase domain and is associated with increasing numbers of chromo-somal imbalances. Blood. 2002;99:238–244.

43. Vaishampayan UN, Mohamed AN, Dugan MC, Bloom RE, Palutke M. Blastic mantle celllymphoma associated with Burkitt-type translocation and hypodiploidy. Br J Haematol.2001;115:66–68.


44. Nodit L, Bahler DW, Jacobs SA, Locker J, Swerdlow SH. Indolent mantle cell lymphomawith nodal involvement and mutated immunoglobulin heavy chain genes. Hum Pathol.2003;34:1030–1034.

45. Orchard J, Garand R, Davis Z, et al. A subset of t(11;14) lymphoma with mantle cell featuresdisplays mutated IgVH genes and includes patients with good prognosis, nonnodal disease.Blood. 2003;101:4975–4981.

46. Thelander EF, Walsh SH, Thorselius M, et al. Mantle cell lymphomas with clonalimmunoglobulin V(H)3–21 gene rearrangements exhibit fewer genomic imbalancesthan mantle cell lymphomas utilizing other immunoglobulin V(H) genes. Mod Pathol.2005;18:331–339.

47. Rosenwald A, Wright G, Wiestner A, et al. The proliferation gene expression signature is aquantitative integrator of oncogenic events that predicts survival in mantle cell lymphoma.Cancer Cell. 2003;3:185–197.

48. Tiemann M, Schrader C, Klapper W, et al. Histopathology, cell proliferation indices andclinical outcome in 304 patients with mantle cell lymphoma (MCL): a clinicopathologicalstudy from the European MCL Network. Br J Haematol. 2005;131:29–38.

49. Jardin F, Picquenot JM, Parmentier F, et al. Detection of gene copy number aberrations inmantle cell lymphoma by a single quantitative multiplex PCR assay: clinicopathologicalrelevance and prognosis value. Br J Haematol. 2009;146:607–618.

50. Walsh SH, Thorselius M, Johnson A, et al. Mutated VH genes and preferential VH3–21 usedefine new subsets of mantle cell lymphoma. Blood. 2003;101:4047–4054.

51. Camacho FI, Algara P, Rodriguez A, et al. Molecular heterogeneity in MCL defined bythe use of specific VH genes and the frequency of somatic mutations. Blood. 2003;101:4042–4046.

52. Alizadeh AA, Eisen MB, Davis RE, et al. Distinct types of diffuse large B-cell lymphomaidentified by gene expression profiling. Nature. 2000;403:503–511.

53. Monti S, Savage KJ, Kutok JL, et al. Molecular profiling of diffuse large B-cell lymphomaidentifies robust subtypes including one characterized by host inflammatory response. Blood.2005;105:1851–1861.

54. Iqbal J, Greiner TC, Patel K, et al. Distinctive patterns of BCL6 molecular alterations andtheir functional consequences in different subgroups of diffuse large B-cell lymphoma.Leukemia. 2007;21:2332–2343.

55. Baron BW, Zeleznik-Le N, Baron MJ, et al. Repression of the PDCD2 gene by BCL6 andthe implications for the pathogenesis of human B and T cell lymphomas. Proc Natl Acad SciUSA. 2007;104:7449–7454.

56. Ci W, Polo JM, Cerchietti L, et al. The BCL6 transcriptional program features repressionof multiple oncogenes in primary B cells and is deregulated in DLBCL. Blood. 2009;113:5536–5548.

57. Uccella S, Placidi C, Marchet S, et al. Bcl-6 protein expression, and not the germinal centreimmunophenotype, predicts favourable prognosis in a series of primary nodal diffuse largeB-cell lymphomas: a single centre experience. Leuk Lymphoma. 2008;49:1321–1328.

58. Hallermann C, Kaune KM, Gesk S, et al. Molecular cytogenetic analysis of chromosomalbreakpoints in the IGH, MYC, BCL6, and MALT1 gene loci in primary cutaneous B-celllymphomas. J Invest Dermatol. 2004;123:213–219.

59. Wlodarska I, Stul M, De Wolf-Peeters C, Hagemeijer A. Heterogeneity of BCL6 rear-rangements in nodular lymphocyte predominant Hodgkin’s lymphoma. Haematologica.2004;89:965–972.

60. Wlodarska I, Nooyen P, Maes B, et al. Frequent occurrence of BCL6 rearrangementsin nodular lymphocyte predominance Hodgkin lymphoma but not in classical Hodgkinlymphoma. Blood. 2003;101:706–710.

61. Bastard C, Deweindt C, Kerckaert JP, et al. LAZ3 rearrangements in non-Hodgkin’s lym-phoma: correlation with histology, immunophenotype, karyotype, and clinical outcome in217 patients. Blood. 1994;83:2423–2427.


62. Obermann EC, Csato M, Dirnhofer S, Tzankov A. Aberrations of the MYC gene in unse-lected cases of diffuse large B-cell lymphoma are rare and unpredictable by morphologicalor immunohistochemical assessment. J Clin Pathol. 2009;62:754–756.

63. Savage KJ, Johnson NA, Ben-Neriah S, et al. MYC gene rearrangements are associatedwith a poor prognosis in diffuse large B-cell lymphoma patients treated with R-CHOPchemotherapy. Blood. 2009;114:3533–3537.

64. Colomo L, Lopez-Guillermo A, Perales M, et al. Clinical impact of the differentiation profileassessed by immunophenotyping in patients with diffuse large B-cell lymphoma. Blood.2003;101:78–84.

65. Hans CP, Weisenburger DD, Greiner TC, et al. Confirmation of the molecular classifica-tion of diffuse large B-cell lymphoma by immunohistochemistry using a tissue microarray.Blood. 2004;103:275–282.

66. Lenz G, Wright G, Dave SS, et al. Stromal gene signatures in large-B-cell lymphomas. NEngl J Med. 2008;359:2313–2323.

67. Copie-Bergman C, Gaulard P, Leroy K, et al. Immuno-fluorescence in situ hybridizationindex predicts survival in patients with diffuse large B-cell lymphoma treated with R-CHOP:a GELA study. J Clin Oncol. 2009;27:5573–5579.

68. Seki R, Ohshima K, Fujisaki T, et al. Prognostic impact of immunohistochemical biomarkersin diffuse large B-cell lymphoma in the rituximab era. Cancer Sci. 2009;100:1842–1847.

69. Song MK, Chung JS, Shin DH, et al. Prognostic significance of the Bcl-2 negative ger-minal centre in patients with diffuse large B cell lymphoma treated with R-CHOP. LeukLymphoma. 2009;50:54–61.

70. Mitterbauer-Hohendanner G, Mannhalter C, Winkler K, et al. Prognostic significance ofmolecular staging by PCR-amplification of immunoglobulin gene rearrangements in diffuselarge B-cell lymphoma (DLBCL). Leukemia. 2004;18:1102–1107.

71. Isaacson PG, Du MQ. MALT lymphoma: from morphology to molecules. Nat Rev Cancer.2004;4:644–653.

72. Isaacson PG, Spencer J. The biology of low grade MALT lymphoma. J Clin Pathol.1995;48:395–397.

73. Parsonnet J, Hansen S, Rodriguez L, et al. Helicobacter pylori infection and gastriclymphoma. N Engl J Med. 1994;330:1267–1271.

74. Miklos JA, Swerdlow SH, Bahler DW. Salivary gland mucosa-associated lymphoid tissuelymphoma immunoglobulin V(H) genes show frequent use of V1–69 with distinctive CDR3features. Blood. 2000;95:3878–3884.

75. Hussell T, Isaacson PG, Crabtree JE, Spencer J. The response of cells from low-grade B-cellgastric lymphomas of mucosa-associated lymphoid tissue to Helicobacter pylori. Lancet.1993;342:571–574.

76. Farinha P, Gascoyne RD. Molecular pathogenesis of mucosa-associated lymphoid tissuelymphoma. J Clin Oncol. 2005;23:6370–6378.

77. Sagaert X, De Wolf-Peeters C, Noels H, Baens M. The pathogenesis of MALT lymphomas:where do we stand? Leukemia. 2007;21:389–396.

78. Remstein ED, Dogan A, Einerson RR, et al. The incidence and anatomic site specificityof chromosomal translocations in primary extranodal marginal zone B-cell lymphoma ofmucosa-associated lymphoid tissue (MALT lymphoma) in North America. Am J SurgPathol. 2006;30:1546–1553.

79. Streubel B, Simonitsch-Klupp I, Mullauer L, et al. Variable frequencies of MALTlymphoma-associated genetic aberrations in MALT lymphomas of different sites. Leukemia.2004;18:1722–1726.

80. Streubel B, Vinatzer U, Lamprecht A, Raderer M, Chott A. T(3;14)(p14.1;q32) involv-ing IGH and FOXP1 is a novel recurrent chromosomal aberration in MALT lymphoma.Leukemia. 2005;19:652–658.

81. Starostik P, Patzner J, Greiner A, et al. Gastric marginal zone B-cell lymphomas of MALTtype develop along 2 distinct pathogenetic pathways. Blood. 2002;99:3–9.


82. Liu H, Ruskon-Fourmestraux A, Lavergne-Slove A, et al. Resistance of t(11;18) positivegastric mucosa-associated lymphoid tissue lymphoma to Helicobacter pylori eradicationtherapy. Lancet. 2001;357:39–40.

83. Traverse-Glehen A, Felman P, Callet-Bauchu E, et al. A clinicopathological study ofnodal marginal zone B-cell lymphoma. A report on 21 cases. Histopathology. 2006;48:162–173.

84. Mateo M, Mollejo M, Villuendas R, et al. 7q31–32 allelic loss is a frequent finding in splenicmarginal zone lymphoma. Am J Pathol. 1999;154:1583–1589.

85. Algara P, Mateo MS, Sanchez-Beato M, et al. Analysis of the IgV(H) somatic mutationsin splenic marginal zone lymphoma defines a group of unmutated cases with frequent 7qdeletion and adverse clinical course. Blood. 2002;99:1299–1304.

86. Ruiz-Ballesteros E, Mollejo M, Rodriguez A, et al. Splenic marginal zone lymphoma: pro-posal of new diagnostic and prognostic markers identified after tissue and cDNA microarrayanalysis. Blood. 2005;106:1831–1838.

87. Bahler DW, Pindzola JA, Swerdlow SH. Splenic marginal zone lymphomas appear tooriginate from different B cell types. Am J Pathol. 2002;161:81–88.

88. Thieblemont C, Nasser V, Felman P, et al. Small lymphocytic lymphoma, marginal zoneB-cell lymphoma, and mantle cell lymphoma exhibit distinct gene-expression profilesallowing molecular diagnosis. Blood. 2004;103:2727–2737.

89. Shiramizu B, Barriga F, Neequaye J, et al. Patterns of chromosomal breakpoint locationsin Burkitt’s lymphoma: relevance to geography and Epstein–Barr virus association. Blood.1991;77:1516–1526.

90. Neri A, Barriga F, Knowles DM, Magrath IT, Dalla-Favera R. Different regions of theimmunoglobulin heavy-chain locus are involved in chromosomal translocations in distinctpathogenetic forms of Burkitt lymphoma. Proc Natl Acad Sci USA. 1988;85:2748–2752.

91. Haralambieva E, Boerma EJ, van Imhoff GW, et al. Clinical, immunophenotypic, and geneticanalysis of adult lymphomas with morphologic features of Burkitt lymphoma. Am J SurgPathol. 2005;29:1086–1094.

92. Joos S, Falk MH, Lichter P, et al. Variable breakpoints in Burkitt lymphoma cells with chro-mosomal t(8;14) translocation separate c-myc and the IgH locus up to several hundred kb.Hum Mol Genet. 1992;1:625–632.

93. Basso K, Frascella E, Zanesco L, Rosolen A. Improved long-distance polymerase chainreaction for the detection of t(8;14)(q24;q32) in Burkitt’s lymphomas. Am J Pathol.1999;155:1479–1485.

94. Akasaka T, Muramatsu M, Ohno H, et al. Application of long-distance polymerase chainreaction to detection of junctional sequences created by chromosomal translocation inmature B-cell neoplasms. Blood. 1996;88:985–994.

95. Haralambieva E, Schuuring E, Rosati S, et al. Interphase fluorescence in situ hybridizationfor detection of 8q24/MYC breakpoints on routine histologic sections: validation in Burkittlymphomas from three geographic regions. Genes Chromosomes Cancer. 2004;40:10–18.

96. Lindstrom MS, Wiman KG. Role of genetic and epigenetic changes in Burkitt lymphoma.Semin Cancer Biol. 2002;12:381–387.

97. Lones MA, Sanger WG, Le Beau MM, et al. Chromosome abnormalities may correlatewith prognosis in Burkitt/Burkitt-like lymphomas of children and adolescents: a report fromChildren’s Cancer Group Study CCG-E08. J Pediatr Hematol Oncol. 2004;26:169–178.

98. Macpherson N, Lesack D, Klasa R, et al. Small noncleaved, non-Burkitt’s (Burkitt-Like)lymphoma: cytogenetics predict outcome and reflect clinical presentation. J Clin Oncol.1999;17:1558–1567.

99. Kanungo A, Medeiros LJ, Abruzzo LV, Lin P. Lymphoid neoplasms associated with concur-rent t(14;18) and 8q24/c-MYC translocation generally have a poor prognosis. Mod Pathol.2006;19:25–33.

100. Hummel M, Bentink S, Berger H, et al. A biologic definition of Burkitt’s lymphoma fromtranscriptional and genomic profiling. N Engl J Med. 2006;354:2419–2430.


101. Zhao XF, Hassan A, Perry A, Ning Y, Stass SA, Dehner LP. C-MYC rearrangements arefrequent in aggressive mature B-cell lymphoma with atypical morphology. Int J Clin ExpPathol. 2008;1:65–74.

102. Salaverria I, Zettl A, Bea S, et al. Chromosomal alterations detected by comparative genomichybridization in subgroups of gene expression-defined Burkitt’s lymphoma. Haematologica.2008;93:1327–1334.

103. Dave SS, Fu K, Wright GW, et al. Molecular diagnosis of Burkitt’s lymphoma. N Engl JMed. 2006;354:2431–2442.

104. Cook JR, Aguilera NI, Reshmi-Skarja S, et al. Lack of PAX5 rearrangements in lympho-plasmacytic lymphomas: reassessing the reported association with t(9;14). Hum Pathol.2004;35:447–454.

105. Cook JR, Aguilera NI, Reshmi S, et al. Deletion 6q is not a characteristic marker of nodallymphoplasmacytic lymphoma. Cancer Genet Cytogenet. 2005;162:85–88.

106. Sargent RL, Cook JR, Aguilera NI, et al. Fluorescence immunophenotypic and interphasecytogenetic characterization of nodal lymphoplasmacytic lymphoma. Am J Surg Pathol.2008;32:1643–1653.

107. Ocio EM, Schop RF, Gonzalez B, et al. 6q deletion in Waldenstrom macroglobulinemia isassociated with features of adverse prognosis. Br J Haematol. 2007;136:80–86.

108. Terre C, Nguyen-Khac F, Barin C, et al. Trisomy 4, a new chromosomal abnormality inWaldenstrom’s macroglobulinemia: a study of 39 cases. Leukemia. 2006;20:1634–1636.

109. Kneba M, Bergholz M, Bolz I, et al. Heterogeneity of immunoglobulin gene rearrangementsin B-cell lymphomas. Int J Cancer. 1990;45:609–613.

110. Brumpt C, Delabesse E, Beldjord K, et al. The incidence of clonal T-cell receptor rear-rangements in B-cell precursor acute lymphoblastic leukemia varies with age and genotype.Blood. 2000;96:2254–2261.

111. van Krieken JH, Langerak AW, Macintyre EA, et al. Improved reliability of lymphomadiagnostics via PCR-based clonality testing: report of the BIOMED-2 Concerted ActionBHM4-CT98-3936. Leukemia. 2007;21:201–206.

112. Bruggemann M, White H, Gaulard P, et al. Powerful strategy for polymerase chain reaction-based clonality assessment in T-cell malignancies Report of the BIOMED-2 ConcertedAction BHM4 CT98-3936. Leukemia. 2007;21:215–221.

113. Lawnicki LC, Rubocki RJ, Chan WC, Lytle DM, Greiner TC. The distribution of gene seg-ments in T-cell receptor gamma gene rearrangements demonstrates the need for multipleprimer sets. J Mol Diagn. 2003;5:82–87.

114. Greiner TC, Rubocki RJ. Effectiveness of capillary electrophoresis using fluorescent-labeledprimers in detecting T-cell receptor gamma gene rearrangements. J Mol Diagn. 2002;4:137–143.

115. Lee SC, Berg KD, Racke FK, Griffin CA, Eshleman JR. Pseudo-spikes are common inhistologically benign lymphoid tissues. J Mol Diagn. 2000;2:145–152.

116. Carey JL, McCoy JP, Keren DF, eds. Flow Cytometry in Clinical Diagnosis. 4th ed. Chicago:ASCP Press; 2007.

117. van den Beemd R, Boor PP, van Lochem EG, et al. Flow cytometric analysis of the Vbetarepertoire in healthy controls. Cytometry. 2000;40:336–345.

118. Morice WG, Kimlinger T, Katzmann JA, et al. Flow cytometric assessment of TCR-Vbetaexpression in the evaluation of peripheral blood involvement by T-cell lymphoproliferativedisorders: a comparison with conventional T-cell immunophenotyping and molecular genetictechniques. Am J Clin Pathol. 2004;121:373–383.

119. Delsol G, Falini B, Muller-Hermelink HK, et al. Anaplastic large cell lymphoma (ALCL),ALK-positive. In: Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, ThieleJ, Vardiman JW, eds. WHO Classification of Tumours of Haematopoietic and LymphoidTissues. 4th ed. Lyon: IARC Press; 2008. pp. 312–316.

120. Morris SW, Kirstein MN, Valentine MB, et al. Fusion of a kinase gene, ALK, to a nucleolarprotein gene, NPM, in non-Hodgkin’s lymphoma. Science. 1994;263:1281–1284.


121. Mason DY, Harris NL, Delsol G, et al. Anaplastic large cell lymphoma, ALK-negative.In: Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, Thiele J, VardimanJW, eds. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed.Lyon: IARC Press; 2008. pp. 317–319.

122. Ralfkiaer E, Willemze R, Paulli M, Kadin ME. Primary cutaneous CD30-positive T-celllymphoproliferative disorders. In: Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA,Stein H, Thiele J, Vardiman JW, eds. WHO Classification of Tumours of Haematopoieticand Lymphoid Tissues. 4th ed. Lyon: IARC Press; 2008. pp. 300–301.

123. Palmer RH, Vernersson E, Grabbe C, Hallberg B. Anaplastic lymphoma kinase: signallingin development and disease. Biochem J. 2009;420:345–361.

124. Ventura RA, Martin-Subero JI, Knippschild U, et al. Centrosome abnormalities in ALK-positive anaplastic large-cell lymphoma. Leukemia. 2004;18:1910–1911.

125. Chiarle R, Voena C, Ambrogio C, Piva R, Inghirami G. The anaplastic lymphoma kinase inthe pathogenesis of cancer. Nat Rev Cancer. 2008;8:11–23.

126. Voena C, Conte C, Ambrogio C, et al. The tyrosine phosphatase Shp2 interacts withNPM–ALK and regulates anaplastic lymphoma cell growth and migration. Cancer Res.2007;67:4278–4286.

127. Shi P, Lai R, Lin Q, et al. IGF-IR tyrosine kinase interacts with NPM–ALK oncogene toinduce survival of T-cell ALK+ anaplastic large-cell lymphoma cells. Blood. 2009;114:360–370.

128. Colomba A, Courilleau D, Ramel D, et al. Activation of Rac1 and the exchange factorVav3 are involved in NPM–ALK signaling in anaplastic large cell lymphomas. Oncogene.2008;27:2728–2736.

129. Dirks WG, Fahnrich S, Lis Y, Becker E, MacLeod RA, Drexler HG. Expression and func-tional analysis of the anaplastic lymphoma kinase (ALK) gene in tumor cell lines. Int JCancer. 2002;100:49–56.

130. Soda M, Choi YL, Enomoto M, et al. Identification of the transforming EML4-ALK fusiongene in non-small-cell lung cancer. Nature. 2007;448:561–566.

131. Perez-Pinera P, Chang Y, Astudillo A, Mortimer J, Deuel TF. Anaplastic lymphoma kinaseis expressed in different subtypes of human breast cancer. Biochem Biophys Res Commun.2007;358:399–403.

132. Griffin CA, Hawkins AL, Dvorak C, Henkle C, Ellingham T, Perlman EJ. Recurrent involve-ment of 2p23 in inflammatory myofibroblastic tumors. Cancer Res. 1999;59:2776–2780.

133. Mathas S, Kreher S, Meaburn KJ, et al. Gene deregulation and spatial genome reorganizationnear breakpoints prior to formation of translocations in anaplastic large cell lymphoma. ProcNatl Acad Sci USA. 2009;106:5831–5836.

134. Trumper L, Pfreundschuh M, Bonin FV, Daus H. Detection of the t(2;5)-associatedNPM/ALK fusion cDNA in peripheral blood cells of healthy individuals. Br J Haematol.1998;103:1138–1144.

135. Maes B, Vanhentenrijk V, Wlodarska I, et al. The NPM–ALK and the ATIC–ALK fusiongenes can be detected in non-neoplastic cells. Am J Pathol. 2001;158:2185–2193.

136. Kalinova M, Krskova L, Brizova H, Kabickova E, Kepak T, Kodet R. Quantitative PCRdetection of NPM/ALK fusion gene and CD30 gene expression in patients with anaplasticlarge cell lymphomaresidual disease monitoring and a correlation with the disease status.Leuk Res. 2008;32:25–32.

137. Mussolin L, Pillon M, d’Amore ES, et al. Prevalence and clinical implications of bonemarrow involvement in pediatric anaplastic large cell lymphoma. Leukemia. 2005;19:1643–1647.

138. Kansal R, Sait SN, Block AW, et al. Extra copies of chromosome 2 are a recurring aberrationin ALK-negative lymphomas with anaplastic morphology. Mod Pathol. 2005;18:235–243.

139. Zettl A, Rudiger T, Konrad MA, et al. Genomic profiling of peripheral T-cell lymphoma,unspecified, and anaplastic large T-cell lymphoma delineates novel recurrent chromosomalalterations. Am J Pathol. 2004;164:1837–1848.


140. Salaverria I, Bea S, Lopez-Guillermo A, et al. Genomic profiling reveals different geneticaberrations in systemic ALK-positive and ALK-negative anaplastic large cell lymphomas.Br J Haematol. 2008;140:516–526.

141. Schlaak M, Renner R, Treudler R, et al. CD30+ anaplastic lymphoma kinase-positiveanaplastic large cell lymphoma with an unusual translocation t(11;22). Br J Dermatol.2008;159:240–242.

142. Sano F, Tasaka T, Nishimura H, et al. Small cell variant of anaplastic large cell lymphomadiagnosed by a novel chromosomal abnormality t(2;5;3)(p23;q35;p21) of bone marrow cells.Pathol Int. 2008;58:494–497.

143. Nelson M, Horsman DE, Weisenburger DD, et al. Cytogenetic abnormalities and clinicalcorrelations in peripheral T-cell lymphoma. Br J Haematol. 2008;141:461–469.

144. Fadlelmola FM, Zhou M, de Leeuw RJ, et al. Sub-megabase resolution tiling (SMRT)array-based comparative genomic hybridization profiling reveals novel gains and losses ofchromosomal regions in Hodgkin lymphoma and anaplastic large cell lymphoma cell lines.Mol Cancer. 2008;7:2.

145. Thompson MA, Stumph J, Henrickson SE, et al. Differential gene expression in anaplasticlymphoma kinase-positive and anaplastic lymphoma kinase-negative anaplastic large celllymphomas. Hum Pathol. 2005;36:494–504.

146. Ballester B, Ramuz O, Gisselbrecht C, et al. Gene expression profiling identifiesmolecular subgroups among nodal peripheral T-cell lymphomas. Oncogene. 2006;25:1560–1570.

147. Lamant L, de Reynies A, Duplantier MM, et al. Gene-expression profiling of systemicanaplastic large-cell lymphoma reveals differences based on ALK status and two distinctmorphologic ALK+ subtypes. Blood. 2007;109:2156–2164.

148. Eckerle S, Brune V, Doring C, et al. Gene expression profiling of isolated tumour cells fromanaplastic large cell lymphomas: insights into its cellular origin, pathogenesis and relationto Hodgkin lymphoma. Leukemia. 2009;23:2129–2138.

149. Dogan A, Gaulard P, Jaffe ES, Ralfkiaer E, Muller-Hermelink HK. AngioimmunoblasticT-cell lymphoma. In: Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, SteinH, Thiele J, Vardiman JW, eds. WHO Classification of Tumours of Haematopoietic andLymphoid Tissues. 4th ed. Lyon: IARC Press; 2008. pp. 309–311.

150. Weiss LM, Strickler JG, Dorfman RF, Horning SJ, Warnke RA, Sklar J. Clonal T-cell popula-tions in angioimmunoblastic lymphadenopathy and angioimmunoblastic lymphadenopathy-like lymphoma. Am J Pathol. 1986;122:392–397.

151. Tan BT, Warnke RA, Arber DA. The frequency of B- and T-cell gene rearrangements andEpstein–Barr virus in T-cell lymphomas: a comparison between angioimmunoblastic T-celllymphoma and peripheral T-cell lymphoma, unspecified with and without associated B-cellproliferations. J Mol Diagn. 2006;8:466–475; quiz 527.

152. Attygalle AD, Chuang SS, Diss TC, Du MQ, Isaacson PG, Dogan A. Distinguishingangioimmunoblastic T-cell lymphoma from peripheral T-cell lymphoma, unspecified, usingmorphology, immunophenotype and molecular genetics. Histopathology. 2007;50:498–508.

153. Attygalle AD, Kyriakou C, Dupuis J, et al. Histologic evolution of angioimmunoblasticT-cell lymphoma in consecutive biopsies: clinical correlation and insights into natural historyand disease progression. Am J Surg Pathol. 2007;31:1077–1088.

154. Dogan A, Attygalle AD, Kyriakou C. Angioimmunoblastic T-cell lymphoma. Br J Haematol.2003;121:681–691.

155. Schlegelberger B, Zhang Y, Weber-Matthiesen K, Grote W. Detection of aberrant clonesin nearly all cases of angioimmunoblastic lymphadenopathy with dysproteinemia-typeT-cell lymphoma by combined interphase and metaphase cytogenetics. Blood. 1994;84:2640–2648.

156. Leich E, Haralambieva E, Zettl A, et al. Tissue microarray-based screening for chromosomalbreakpoints affecting the T-cell receptor gene loci in mature T-cell lymphomas. J Pathol.2007;213:99–105.


157. Thorns C, Bastian B, Pinkel D, et al. Chromosomal aberrations in angioimmunoblastic T-celllymphoma and peripheral T-cell lymphoma unspecified: a matrix-based CGH approach.Genes Chromosomes Cancer. 2007;46:37–44.

158. de Leval L, Rickman DS, Thielen C, et al. The gene expression profile of nodal periph-eral T-cell lymphoma demonstrates a molecular link between angioimmunoblastic T-celllymphoma (AITL) and follicular helper T (TFH) cells. Blood. 2007;109:4952–4963.

159. Murakami YI, Yatabe Y, Sakaguchi T, et al. c-Maf expression in angioimmunoblastic T-celllymphoma. Am J Surg Pathol. 2007;31:1695–1702.

160. Ralfkiaer E, Cerroni L, Sander CA, Smoller BR, Willemze R. Mycosis fungoides. In:Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, Thiele J, Vardiman JW,eds. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed.Lyon: IARC Press; 2008. pp. 296–298.

161. Pimpinelli N, Olsen EA, Santucci M, et al. Defining early mycosis fungoides. J Am AcadDermatol. 2005;53:1053–1063.

162. Willemze R, Jaffe ES, Burg G, et al. WHO-EORTC classification for cutaneous lymphomas.Blood. 2005;105:3768–3785.

163. Sandberg Y, Heule F, Lam K, et al. Molecular immunoglobulin/T- cell receptor clonalityanalysis in cutaneous lymphoproliferations. Experience with the BIOMED-2 standardizedpolymerase chain reaction protocol. Haematologica. 2003;88:659–670.

164. Thurber SE, Zhang B, Kim YH, Schrijver I, Zehnder J, Kohler S. T-cell clonality analysisin biopsy specimens from two different skin sites shows high specificity in the diagnosis ofpatients with suggested mycosis fungoides. J Am Acad Dermatol. 2007;57:782–790.

165. Wood GS, Tung RM, Haeffner AC, et al. Detection of clonal T-cell receptor gamma generearrangements in early mycosis fungoides/Sezary syndrome by polymerase chain reac-tion and denaturing gradient gel electrophoresis (PCR/DGGE). J Invest Dermatol. 1994;103:34–41.

166. Smoller BR, Santucci M, Wood GS, Whittaker SJ. Histopathology and genetics of cutaneousT-cell lymphoma. Hematol Oncol Clin North Am. 2003;17:1277–1311.

167. Langerak AW, Molina TJ, Lavender FL, et al. Polymerase chain reaction-based clonal-ity testing in tissue samples with reactive lymphoproliferations: usefulness and pitfalls.A report of the BIOMED-2 Concerted Action BMH4-CT98-3936. Leukemia. 2007;21:222–229.

168. Ponti R, Quaglino P, Novelli M, et al. T-cell receptor gamma gene rearrangement bymultiplex polymerase chain reaction/heteroduplex analysis in patients with cutaneousT-cell lymphoma (mycosis fungoides/Sezary syndrome) and benign inflammatory disease:correlation with clinical, histological and immunophenotypical findings. Br J Dermatol.2005;153:565–573.

169. Guitart J, Magro C. Cutaneous T-cell lymphoid dyscrasia: a unifying term for idiopathicchronic dermatoses with persistent T-cell clones. Arch Dermatol. 2007;143:921–932.

170. Vega F, Luthra R, Medeiros LJ, et al. Clonal heterogeneity in mycosis fungoides and itsrelationship to clinical course. Blood. 2002;100:3369–3373.

171. Delfau-Larue MH, Laroche L, Wechsler J, et al. Diagnostic value of dominant T-cell clonesin peripheral blood in 363 patients presenting consecutively with a clinical suspicion ofcutaneous lymphoma. Blood. 2000;96:2987–2992.

172. Muche JM, Sterry W, Gellrich S, Rzany B, Audring H, Lukowsky A. Peripheral blood T-cellclonality in mycosis fungoides and nonlymphoma controls. Diagn Mol Pathol. 2003;12:142–150.

173. Khan N, Shariff N, Cobbold M, et al. Cytomegalovirus seropositivity drives the CD8T cell repertoire toward greater clonality in healthy elderly individuals. J Immunol.2002;169:1984–1992.

174. Posnett DN, Sinha R, Kabak S, Russo C. Clonal populations of T cells in normalelderly humans: the T cell equivalent to “benign monoclonal gammapathy”. J Exp Med.1994;179:609–618.


175. Gniadecki R, Lukowsky A. Monoclonal T-cell dyscrasia of undetermined significanceassociated with recalcitrant erythroderma. Arch Dermatol. 2005;141:361–367.

176. Klemke CD, Poenitz N, Dippel E, Hummel M, Stein H, Goerdt S. T-cell clonality ofundetermined significance. Arch Dermatol. 2006;142:393–394.

177. Assaf C, Hummel M, Steinhoff M, et al. Early TCR-beta and TCR-gamma PCR detec-tion of T-cell clonality indicates minimal tumor disease in lymph nodes of cutaneous T-celllymphoma: diagnostic and prognostic implications. Blood. 2005;105:503–510.

178. Fraser-Andrews EA, Mitchell T, Ferreira S, et al. Molecular staging of lymph nodes from60 patients with mycosis fungoides and Sezary syndrome: correlation with histopathologyand outcome suggests prognostic relevance in mycosis fungoides. Br J Dermatol. 2006;155:756–762.

179. Juarez T, Isenhath SN, Polissar NL, et al. Analysis of T-cell receptor gene rearrangementfor predicting clinical outcome in patients with cutaneous T-cell lymphoma: a compari-son of Southern blot and polymerase chain reaction methods. Arch Dermatol. 2005;141:1107–1113.

180. Olsen E, Vonderheid E, Pimpinelli N, et al. Revisions to the staging and classificationof mycosis fungoides and Sezary syndrome: a proposal of the International Society forCutaneous Lymphomas (ISCL) and the cutaneous lymphoma task force of the EuropeanOrganization of Research and Treatment of Cancer (EORTC). Blood. 2007;110:1713–1722.

181. Sibaud V, Beylot-Barry M, Thiebaut R, et al. Bone marrow histopathologic and molecularstaging in epidermotropic T-cell lymphomas. Am J Clin Pathol. 2003;119:414–423.

182. Scarisbrick JJ, Woolford AJ, Russell-Jones R, Whittaker SJ. Allelotyping in mycosis fun-goides and Sezary syndrome: common regions of allelic loss identified on 9p, 10q, and 17p.J Invest Dermatol. 2001;117:663–670.

183. Thangavelu M, Finn WG, Yelavarthi KK, et al. Recurring structural chromosome abnormal-ities in peripheral blood lymphocytes of patients with mycosis fungoides/Sezary syndrome.Blood. 1997;89:3371–3377.

184. Karenko L, Sarna S, Kahkonen M, Ranki A. Chromosomal abnormalities in relation toclinical disease in patients with cutaneous T-cell lymphoma: a 5-year follow-up study. BrJ Dermatol. 2003;148:55-64.

185. Ikonomou IM, Aamot HV, Heim S, Fossa A, Delabie J. Granulomatous slack skin with atranslocation t(3;9)(q12;p24). Am J Surg Pathol. 2007;31:803-806.

186. Tracey L, Villuendas R, Dotor AM, et al. Mycosis fungoides shows concurrent deregulationof multiple genes involved in the TNF signaling pathway: an expression profile study. Blood.2003;102:1042–1050.

187. Shin J, Monti S, Aires DJ, et al. Lesional gene expression profiling in cutaneous T-celllymphoma reveals natural clusters associated with disease outcome. Blood. 2007;110:3015–3027.

188. Prochazkova M, Chevret E, Mainhaguiet G, et al. Common chromosomal abnormalities inmycosis fungoides transformation. Genes Chromosomes Cancer. 2007;46:828–838.

189. Mao X, Lillington D, Scarisbrick JJ, et al. Molecular cytogenetic analysis of cutaneous T-celllymphomas: identification of common genetic alterations in Sezary syndrome and mycosisfungoides. Br J Dermatol. 2002;147:464–475.

190. Fischer TC, Gellrich S, Muche JM, et al. Genomic aberrations and survival in cutaneous Tcell lymphomas. J Invest Dermatol. 2004;122:579–586.

191. Katona TM, O’Malley DP, Cheng L, et al. Loss of heterozygosity analysis identifies geneticabnormalities in mycosis fungoides and specific loci associated with disease progression.Am J Surg Pathol. 2007;31:1552–1556.

192. Navas IC, Ortiz-Romero PL, Villuendas R, et al. p16(INK4a) gene alterations are frequentin lesions of mycosis fungoides. Am J Pathol. 2000;156:1565–1572.

193. Scarisbrick JJ, Woolford AJ, Calonje E, et al. Frequent abnormalities of the p15 and p16genes in mycosis fungoides and sezary syndrome. J Invest Dermatol. 2002;118:493–499.


194. van Doorn R, Zoutman WH, Dijkman R, et al. Epigenetic profiling of cutaneous T-cell lym-phoma: promoter hypermethylation of multiple tumor suppressor genes including BCL7a,PTPRG, and p73. J Clin Oncol. 2005;23:3886–3896.

195. Mao X, Orchard G, Lillington DM, Russell-Jones R, Young BD, Whittaker SJ. Amplificationand overexpression of JUNB is associated with primary cutaneous T-cell lymphomas. Blood.2003;101:1513–1519.

196. van Doorn R, van Kester MS, Dijkman R, et al. Oncogenomic analysis of mycosis fungoidesreveals major differences with Sezary syndrome. Blood. 2009;113:127–136.

197. Vermeer MH, van Doorn R, Dijkman R, et al. Novel and highly recurrent chromosomalalterations in Sezary syndrome. Cancer Res. 2008;68:2689–2698.

198. Batista DA, Vonderheid EC, Hawkins A, et al. Multicolor fluorescence in situ hybridiza-tion (SKY) in mycosis fungoides and Sezary syndrome: search for recurrent chromosomeabnormalities. Genes Chromosomes Cancer. 2006;45:383–391.

199. Mao X, Lillington DM, Czepulkowski B, Russell-Jones R, Young BD, Whittaker S.Molecular cytogenetic characterization of Sezary syndrome. Genes Chromosomes Cancer.2003;36:250–260.

200. Karenko L, Kahkonen M, Hyytinen ER, Lindlof M, Ranki A. Notable losses at specificregions of chromosomes 10q and 13q in the Sezary syndrome detected by comparativegenomic hybridization. J Invest Dermatol. 1999;112:392–395.

201. Kari L, Loboda A, Nebozhyn M, et al. Classification and prediction of survival inpatients with the leukemic phase of cutaneous T cell lymphoma. J Exp Med. 2003;197:1477–1488.

202. Hahtola S, Tuomela S, Elo L, et al. Th1 response and cytotoxicity genes are down-regulatedin cutaneous T-cell lymphoma. Clin Cancer Res. 2006;12:4812–4821.

203. van Doorn R, Dijkman R, Vermeer MH, et al. Aberrant expression of the tyrosine kinasereceptor EphA4 and the transcription factor twist in Sezary syndrome identified by geneexpression analysis. Cancer Res. 2004;64:5578–5586.

204. Booken N, Gratchev A, Utikal J, et al. Sezary syndrome is a unique cutaneous T-cell lym-phoma as identified by an expanded gene signature including diagnostic marker moleculesCDO1 and DNM3. Leukemia. 2008;22:393–399.

205. Nebozhyn M, Loboda A, Kari L, et al. Quantitative PCR on 5 genes reliably identifiesCTCL patients with 5% to 99% circulating tumor cells with 90% accuracy. Blood. 2006;107:3189–3196.

206. Barba G, Matteucci C, Girolomoni G, et al. Comparative genomic hybridization identifies17q11.2 approximately q12 duplication as an early event in cutaneous T-cell lymphomas.Cancer Genet Cytogenet. 2008;184:48–51.

207. Gaulard P, Jaffe ES, Krenacs L, Macon WR. Hepatosplenic T-cell lymphoma. In: SwerdlowSH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, Thiele J, Vardiman JW, eds. WHOClassification of Tumours of Haematopoietic and Lymphoid Tissue. 4th ed. Lyon: IARCPress; 2008. pp. 292–293.

208. Alonsozana EL, Stamberg J, Kumar D, et al. Isochromosome 7q: the primary cytoge-netic abnormality in hepatosplenic gammadelta T cell lymphoma. Leukemia. 1997;11:1367–1372.

209. Feldman AL, Law M, Grogg KL, et al. Incidence of TCR and TCL1 gene translocations andisochromosome 7q in peripheral T-cell lymphomas using fluorescence in situ hybridization.Am J Clin Pathol. 2008;130:178–185.

210. Wlodarska I, Martin-Garcia N, Achten R, et al. Fluorescence in situ hybridization studyof chromosome 7 aberrations in hepatosplenic T-cell lymphoma: isochromosome 7q as acommon abnormality accumulating in forms with features of cytologic progression. GenesChromosomes Cancer. 2002;33:243–251.

211. Tamaska J, Adam E, Kozma A, et al. Hepatosplenic gammadelta T-cell lymphoma with ringchromosome 7, an isochromosome 7q equivalent clonal chromosomal aberration. VirchowsArch. 2006;449:479–483.


212. Miyazaki K, Yamaguchi M, Imai H, et al. Gene expression profiling of peripheral T-celllymphoma including gammadelta T-cell lymphoma. Blood. 2009;113:1071–1074.

213. Isaacson PG, Chott A, Ott G, Stein H. Enteropathy-associated T-cell lymphoma. In:Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, Thiele J, Vardiman JW,eds. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed.Lyon: IARC Press; 2008. pp. 289–291.

214. Zettl A, Ott G, Makulik A, et al. Chromosomal gains at 9q characterize enteropathy-typeT-cell lymphoma. Am J Pathol. 2002;161:1635–1645.

215. Deleeuw RJ, Zettl A, Klinker E, et al. Whole-genome analysis and HLA genotyping ofenteropathy-type T-cell lymphoma reveals 2 distinct lymphoma subtypes. Gastroenterology.2007;132:1902–1911.

216. Zettl A, Rudiger T, Muller-Hermelink HK. [Enteropathy type T-cell lymphomas: pathologyand pathogenesis]. Pathologe. 2007;28:59–64.

217. Verkarre V, Romana SP, Cellier C, et al. Recurrent partial trisomy 1q22–q44 in clonalintraepithelial lymphocytes in refractory celiac sprue. Gastroenterology. 2003;125:40–46.

218. Chan JKC, Quintanilla-Martinez L, Ferry JA, Peh S-C. Extranodal NK/T-cell lymphoma,nasal type. In: Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, ThieleJ, Vardiman JW, eds. WHO Classification of Tumours of Haematopoietic and LymphoidTissues. 4th ed. Lyon: IARC Press; 2008. pp. 285–288.

219. Siu LL, Chan V, Chan JK, Wong KF, Liang R, Kwong YL. Consistent patterns of allelic lossin natural killer cell lymphoma. Am J Pathol. 2000;157:1803–1809.

220. Tien HF, Su IJ, Tang JL, et al. Clonal chromosomal abnormalities as direct evidence forclonality in nasal T/natural killer cell lymphomas. Br J Haematol. 1997;97:621–625.

221. Wong KF, Zhang YM, Chan JK. Cytogenetic abnormalities in natural killer cell lym-phoma/leukaemiais there a consistent pattern? Leuk Lymphoma. 1999;34:241–250.

222. Siu LL, Wong KF, Chan JK, Kwong YL. Comparative genomic hybridization analysis of nat-ural killer cell lymphoma/leukemia. Recognition of consistent patterns of genetic alterations.Am J Pathol. 1999;155:1419–1425.

223. Nakashima Y, Tagawa H, Suzuki R, et al. Genome-wide array-based comparative genomichybridization of natural killer cell lymphoma/leukemia: different genomic alteration pat-terns of aggressive NK-cell leukemia and extranodal Nk/T-cell lymphoma, nasal type. GenesChromosomes Cancer. 2005;44:247–255.

224. Iqbal J, Kucuk C, Deleeuw RJ, et al. Genomic analyses reveal global functional alter-ations that promote tumor growth and novel tumor suppressor genes in natural killer-cellmalignancies. Leukemia. 2009;23:1139–1151.

225. Yamanaka Y, Tagawa H, Takahashi N, et al. Aberrant overexpression of microRNAsactivate AKT signaling via down-regulation of tumor suppressors in natural killer-celllymphoma/leukemia. Blood. 2009;114:3265–3275.

226. Hongyo T, Hoshida Y, Nakatsuka S, et al. p53, K-ras, c-kit and beta-catenin gene mutationsin sinonasal NK/T-cell lymphoma in Korea and Japan. Oncol Rep. 2005;13:265–271.

227. Quintanilla-Martinez L, Kremer M, Keller G, et al. p53 Mutations in nasal natural killer/T-cell lymphoma from Mexico: association with large cell morphology and advanced disease.Am J Pathol. 2001;159:2095–2105.

228. Shen L, Liang AC, Lu L, et al. Frequent deletion of Fas gene sequences encodingdeath and transmembrane domains in nasal natural killer/T-cell lymphoma. Am J Pathol.2002;161:2123–2131.

229. Piccaluga PP, Agostinelli C, Califano A, et al. Gene expression analysis of peripheral T celllymphoma, unspecified, reveals distinct profiles and new potential therapeutic targets. J ClinInvest. 2007;117:823–834.

230. Cuadros M, Dave SS, Jaffe ES, et al. Identification of a proliferation signature related tosurvival in nodal peripheral T-cell lymphomas. J Clin Oncol. 2007;25:3321–3329.

231. Liso A, Capello D, Marafioti T, et al. Aberrant somatic hypermutation in tumor cellsof nodular-lymphocyte-predominant and classic Hodgkin lymphoma. Blood. 2006;108:1013–1020.


232. Seitz V, Hummel M, Marafioti T, Anagnostopoulos I, Assaf C, Stein H. Detection ofclonal T-cell receptor gamma-chain gene rearrangements in Reed–Sternberg cells of classicHodgkin disease. Blood. 2000;95:3020–3024.

233. Muschen M, Rajewsky K, Brauninger A, et al. Rare occurrence of classical Hodgkin’sdisease as a T cell lymphoma. J Exp Med. 2000;191:387–394.

234. Joos S, Kupper M, Ohl S, et al. Genomic imbalances including amplification of the tyrosinekinase gene JAK2 in CD30+ Hodgkin cells. Cancer Res. 2000;60:549–552.

235. Chetaille B, Bertucci F, Finetti P, et al. Molecular profiling of classical Hodgkin lym-phoma tissues uncovers variations in the tumor microenvironment and correlations withEBV infection and outcome. Blood. 2009;113:2765–3775.

236. Kuppers R, Klein U, Schwering I, et al. Identification of Hodgkin and Reed–Sternberg cell-specific genes by gene expression profiling. J Clin Invest. 2003;111:529–537.

237. Mathas S, Janz M, Hummel F, et al. Intrinsic inhibition of transcription factor E2A byHLH proteins ABF-1 and Id2 mediates reprogramming of neoplastic B cells in Hodgkinlymphoma. Nat Immunol. 2006;7:207–215.

Chapter 7Molecular Pathology of MyeloproliferativeNeoplasms

David S. Bosler

Keywords Introduction · Classification · Roles of molecular diagnostics · Chronicmyelogenous leukemia (CML) · Philadelphia chromosome · t(9;22) · BCR–ABL1 discovery · Imatinib · CML · Clinical findings · Laboratoryfindings · Histology · CML blast phase · BCR–ABL1 structure · BCR–ABL1 breakpoints · M-BCR · m-BCR · p210 · p190 · p230 · BCR–ABL1 aschimeric protein · ABL1 tyrosine kinase · BCR–ABL1 pathogenesis · BCR–ABL1domains · BCR–ABL1 diagnostic testing · cytogenetics · RT-PCR · BCR–ABL1fluorescence in situ hybridization (FISH) · Multiplex PCR · CML diseasemonitoring · Hematologic response · Cytogenetic response · Imatinib · Majormolecular response · CML disease monitoring · BCR–ABL1 transcript mea-surement · Quantitative RT-PCR · Normalized copy number · BCR–ABL1control gene · BCR–ABL1 quantitative RT-PCR platforms · BCR–ABL1reporting · Normalized copy number · BCR–ABL1 measurement standardiza-tion · BCR–ABL1 measurement optimization · Imatinib mechanism · Imatinibresistance · ABL1 kinase domain mutations · P-loop · T315I · ABL1 mutationtesting methods · DNA sequencing · ABL1 mutation testing standardization · CMLmolecular diagnostic testing · BCR–ABL1-negative classic myeloproliferativeneoplasms (MPNs) · JAK2 · MPL · BCR–ABL1-negative MPNs: epidemi-ology · Clinical findings · JAK2 V617F · Secondary erythrocytosis · JAK2V617F as marker of clonality · JAK2 V617F pathogenesis · JAK2 exon 12 muta-tions · polycythemia vera JAK2 mutations · JAK2 allelic burden · JAK2 detectionmethods · JAK2 detection methods · JAK2 V617F detection · JAK2 detectionmethods · JAK2 exon 12 mutations · MPL · MPL W515 mutations · PDGFR ·FGFR1 · Eosinophilia · Chronic eosinophilic leukemia · Hypereosinophilic syn-drome · PDGFRA · CHIC2 deletion · FIP1L1–PDGFRA fusion · FIP1L1–PDGFRAfusion imatinib response · Eosinophilia · FIP1L1–PDGFRA detection · PDGFRB ·ETV6–PDGFRB fusion · Chronic myelomonocytic leukemia · PDGFRB

D.S. Bosler (B)Department of Clinical Pathology, Cleveland Clinic, 9500 Euclid Ave, Cleveland,OH 44195, USAe-mail: [email protected]


216 D.S. Bosler

abnormality detection · FGFR1 · 8p11 myeloproliferative syndrome · FGFR1disease spectrum · FGFR1 fusion partners · FGFR1 imatinib resistance · FGFR1abnormality detection · Mast cell disease · KIT D816V · KIT D816V tyrosine kinaseinhibitor therapy · KIT D816V detection · KIT D816V detection methods · Mastcell disease differential diagnosis

Introduction

As a group, myeloproliferative neoplasms affect an estimated 6–10 per 100,000 peo-ple [1]. Although many of the individual myeloproliferative neoplasms had beenpreviously recognized clinical entities, it was Dameshek in 1950 that proposedthe currently recognized concept of the myeloproliferative neoplasms as a groupof similar entities united by poorly controlled proliferation of various hematopoi-etic elements [2]. With the exception of chronic myelogenous leukemia (CML),few additional advancements had been made, and other members of this group hadlargely continued to be defined by a combination of clinical and pathologic crite-ria until very recently. Continued advances in molecular diagnostic techniques andresearch into these entities have recently yielded additional insights into the patho-biology of the myeloproliferative neoplasms. Although much of the story remains tobe told, these insights are now reflected in the way myeloproliferative neoplasms arediagnosed and classified. The 2008 WHO Classification [1] groups the myelopro-liferative neoplasms largely according to their associated molecular abnormalities,many of which involve tyrosine kinases (see Table 7.1). CML is defined by thepresence of the BCR–ABL1 fusion. The diagnosis of the other “classic” myelopro-liferative neoplasms – polycythemia vera, essential thrombocythemia, and primarymyelofibrosis – is greatly aided by the presence of JAK2 mutations or MPL muta-tions. Myeloproliferative neoplasms involving PDGFRA, PDGFRB, and FGFR1have been defined as a group containing somewhat heterogeneous but also unify-ing clinicopathologic features. Diagnosis of mast cell disease is also sometimesaided by the presence of a common mutation in KIT. This classification systemplaces molecular diagnostics in a central role in the diagnosis and classification of

Table 7.1 Tyrosine kinases involved in myeloproliferative neoplasms

Tyrosine kinase Disease

Abl kinase Chronic myelogenous leukemiaJanus kinase 2 (JAK2) Polycythemia vera, essential thrombocythemia, primary

myelofibrosisPlatelet-derived growth factor

receptors (PDGFRA, PDGFRB)Myeloid and lymphoid neoplasms with eosinophilia and

abnormalities of PDGFRA, PDGFRB, or FGFR1Fibroblast growth factor receptor 1

(FGFR1)KIT Mast cell disease (also acute myeloid leukemia,

gastrointestinal stromal tumor, melanoma)

7 Molecular Pathology of Myeloproliferative Neoplasms 217

myeloproliferative neoplasms. Many of the recent changes in the classification sys-tem have largely been driven by the advent and widespread use of tyrosine kinaseinhibitors such as imatinib mesylate, with part of the goal of classification being reli-able prediction of which neoplasms will respond to this type of therapy. Additionalroles for molecular diagnostics have also emerged, such as prognosis and diseasemonitoring. Although these potential roles are yet to be fully realized in many dis-eases, CML serves as an illustration of the various ways that molecular diagnosticscan contribute to management of malignancy. This chapter is organized by diseasegroups within the myeloproliferative neoplasms as recognized by the WHO clas-sification. This chapter will explain the basics of the abnormalities encountered ineach group and show how they are identified. It will also illustrate additional rolesof molecular diagnostics where applicable, and it will discuss some of the salientissues that challenge molecular diagnosticians today. A large portion of the chap-ter is devoted to chronic myelogenous leukemia, reflecting its importance from anhistorical perspective as well as the range of test applications and the complex-ity of issues confronting the contemporary molecular diagnostic lab related to thisdisease.

Chronic Myelogenous Leukemia

Historical Perspective and Current Relevance

The story of CML merits special attention in any comprehensive discussion ofmolecular diagnostics. It has particular relevance from an historical perspective, andit serves as an illustrative microcosm of the potential spectrum of molecular diagnos-tics’ applications in clinical oncology. From the early cytogenetics-based discoveryof the Philadelphia chromosome and the discovery of BCR–ABL1, encoding one ofthe earliest known fusion proteins in neoplasia, to the more recent development oftherapy targeted specifically to inhibit that fusion protein, CML has been the sub-ject of many ground-breaking discoveries that have led the advance of the broaderscience and practice of hematology and oncology (Fig. 7.1).

CML was first described as early as 1845 [3]. Relatively soon after Dameshek’sdescription of myeloproliferative neoplasms in 1950, a series of rapid advances incytogenetics techniques resulted in numerous discoveries related to human chro-mosomes. One of these landmark discoveries was the Philadelphia chromosomein CML, the first described structural chromosome abnormality, by Nowell andHungerford in 1960 [4]. Looking at karyotypes from normal individuals, acuteleukemias, and chronic granulocytic leukemias, these researchers identified anabnormal minute chromosome present only in those with CML. Later dubbed the“Philadelphia chromosome,” there was understandably initially no idea how it wasderived. Thirteen years later in 1973, Janet Rowley demonstrated using bandingtechniques that the Philadelphia chromosome resulted from a reciprocal translo-cation between chromosomes 9 and 22 by recognizing that, in cases with the

218 D.S. Bosler

1960 1970 1980 1990 2000

1960: Philadelphia chromosome

1973: t(9;22)

1982: ABL localizedto chromosome 9

1985: BCR-ABL fusion

1990: BCR-ABL ascause of CML

2001: ImatinibFDA Approval

1845

1845: Firstdescription of

CML

Fig. 7.1 Timeline of landmark developments from the first descriptions of CML to the use oftargeted therapy

Philadelphia chromosome, chromosome 9 contained additional material similar inlength and staining pattern to that lost from chromosome 22 [5]. Also a pioneer-ing discovery, this was the first identified translocation in human chromosomes.Numerous studies and researchers in the 1980s contributed to the ultimate demon-stration of the BCR–ABL1 fusion [6], including the localization of ABL1 to the longarm of chromosome 9 [7], mapping of the breakpoint regions on both chromosomes9 and 22 showing proximity and/or involvement of ABL1 and BCR, respectively [8,9], and demonstration that ABL1 is involved in the t(9;22) and other variant translo-cations of CML, but is not translocated in Philadelphia chromosome-negative cases[10]. The presence of the BCR–ABL1 fusion transcript in CML cells was ultimatelydemonstrated in 1985 by Shtivelman et al. [11], and introduction of BCR–ABL1was shown to cause a CML-like disease in mice in 1990 [12]. A growing body ofresearch furthered the notion that CML containing the t(9;22) or similar variantswas a distinct entity from similar appearing neoplasms without such abnormalities.Based on evidence implicating the ABL1 tyrosine kinase activity within BCR–ABL1 in the leukemogenesis of CML, STI571 (later called imatinib) was developedas a specific inhibitor of BCR–ABL1’s tyrosine kinase activity. After demonstrationof its clinical utility, imatinib was approved for use in 2001 for the treatment ofCML and was the first molecular-targeted therapy approved for use in human cancer[13, 14].

The impact of molecular diagnostics on care for patients with CML has onlygrown with time, with integral facets of contemporary management includingdisease-defining diagnostic tests, providing prognostic information, guiding choiceof therapy, monitoring of response to therapy through minimal residual diseasetesting, and testing for development of resistance to therapy.


Epidemiology, Clinical, and Laboratory Features

Chronic myelogenous leukemia accounts for 7–20% of leukemias, has an esti-mated incidence of one to two per 100,000 worldwide, and will be diagnosed inan estimated 5050 people in the United States in 2009 [15, 16]. The median age atdiagnosis is in the fifth and sixth decades [1, 17]. At diagnosis, patients most oftencomplain of fatigue, bleeding tendencies, weight loss, sweats, and symptoms relatedto splenomegaly such as left upper quadrant pain, sensation of a mass, abdominalfullness, and/or swelling [18, 19]. Less common manifestations include bone painand infection. Twenty to 40% are asymptomatic, presenting incidentally with lab-oratory abnormalities [16, 18]. Seventy-five percent of patients have splenomegaly,and purpura is also a common finding [18]. The most common laboratory findingsat presentation are granulocytic leukocytosis and anemia, with or without throm-bocytosis. Average values range 174–225 × 109/L for white blood cell count,9.7–10.3 g/dL for hemoglobin, and 430–485 × 109/L for platelets [18, 20]. Thereis generally a loose inverse relationship between the leukocyte count and the sever-ity of anemia. The granulocytes are left-shifted in the peripheral blood, with fullmaturation to neutrophils, a prominence of myelocytes, and relatively few blastscompared to acute leukemias. Absolute basophilia is a virtually invariant feature,and absolute eosinophilia and monocytosis are present in the majority of cases.Although leukocytosis is the rule, exceptional cases may have normal white bloodcell counts and present instead with markedly elevated platelet counts, myelofibro-sis, or anemia [18, 20]. The bone marrow is markedly hypercellular with a markedpredominance of granulocytes and thick paratrabecular cuffs of immature granulo-cytic precursors [19]. The granulocytes are left-shifted and show full maturation.Blasts account for less than 10% of cellularity in chronic phase and have morphol-ogy that cannot be distinguished from normal myeloid blasts [19, 21, 22]. Smallhypolobated megakaryocytes are a frequent finding, as are bone marrow basophiliaand eosinophilia [19, 22].

The natural history of CML is biphasic or triphasic. Most patients presentin chronic phase, with untreated cases almost invariably progressing to blastphase, and about two-thirds of cases passing through a transitional acceleratedphase [17]. While the overall survival at 3 years in imatinib-treated patients inchronic phase is 95%, progression to blast phase portends a much more aggres-sive course, with 10% survival at 3 years [21]. Although there has been debateregarding the criteria for diagnosis of accelerated phase and blast phase, the currentWHO criteria for accelerated phase include persistent or worsening WBC count,thrombocytosis or splenomegaly unresponsive to therapy, thrombocytopenia, theappearance of additional clonal abnormalities, basophilia >20%, or blast count 10–19%, while blast phase is defined as >20% blasts in the peripheral blood/marrowor extramedullary blast proliferation [1]. Chromosomal abnormalities in additionto the Philadelphia chromosome frequently accompany the onset of accelerated orblast phase, described in more detail below. Blastic transformation may be myeloid,lymphoid, or undifferentiated [17, 23].

220 D.S. Bosler

The Structure and Pathogenesis of BCR–ABL1

The Philadelphia chromosome is created by a reciprocal translocation of chro-mosomes 9 and 22, with the portions of these chromosomes telomeric to theirrespective breakpoints at 9q34 and 22q11 essentially trading places with each other[5] (Fig. 7.2). The “Philadelphia chromosome” is the abnormal chromosome 22 thatis derived from this reciprocal translocation and is easily recognized by its smallsize [4]. The BCR–ABL1 fusion is on this newly derived chromosome 22, created bytranslocation of the 3′ end of the ABL1 gene from chromosome 9 to chromosome 22,where it is juxtaposed to the 5′ part of BCR. The t(9;22) is the most common mech-anism of creating a BCR–ABL1 fusion and is present in over 90% of CML cases[10]. In the remaining cases, the BCR–ABL1 fusion is either created by a complextranslocation involving chromosomes 9, 22, and one or two other chromosomes oris present but cannot be detected by cytogenetics [10, 24, 25].

Although the breakpoints in ABL1 vary over a relatively large 300 kb range ofchromosome 9, these varied breakpoints most often ultimately result in the sameportion of ABL1 being incorporated into the fusion [6, 24]. This homogeneity occursbecause breakpoints in ABL1 span the regions of the gene including exons Ib and Ia,which are both ultimately spliced out of the mature BCR–ABL1 fusion transcriptregardless of the breakpoint [25]. The end result is that exons 2 through 11 are the

Chr 9

Chr 22

ABL1 9q34

BCR 22q11

(der)Chr 9

Ph Chr

BCR-ABL1

BCR5’ 3’e1

e13 e14 e15 e19

m-BCR M-BCR µ-BCR

ABL15’ 3’a2 a31b 1a a11

e13a2

e14a2

p210

p190

p230

e1a2

e19a2

Fig. 7.2 The Philadelphia chromosome results from a reciprocal translocation involving the longarms of chromosomes 9 and 22 – t(9;22)(q34;q11). The Philadelphia chromosome is the alteredchromosome 22 that is derived from this translocation, and contains a fusion of the 5′ end ofBCR and the 3′ end of ABL1. Varying BCR breakpoints result in transcripts and fusion proteins ofvarious lengths, including the most common p210 fusion protein (containing breakpoints of e13a2or e14a2), as well as the p190 (e1a2) and p230 (e19a2)


most frequent portion of ABL1 that is incorporated into the fusion. Rarely, exon 2of ABL1 is also excluded from the fusion transcript, resulting in incorporation ofexons 3 through 11 of ABL1. These cases produce a similar fusion protein and havea clinical presentation without distinguishable differences [6, 24].

The breakpoints within BCR have a more limited and clustered distribution. Mostcommonly, BCR is disrupted in a 5.8 kb region known as M-BCR (major breakpointcluster region) spanning exons 12–16 of the BCR gene [8, 24]. Breakpoints withinthe M-BCR yield a 210 kDa protein called the p210 protein that is present in mostCML cases, with the most frequent fusions being b3a2 (exon 14 of BCR to exon 2of ABL1) and b2a2 (exon 13 of BCR to exon 2 of ABL1). Some studies suggest thatcases involving breakpoints within the more 3′ end of this region are more likelyto be associated with thrombocytosis or present as essential thrombocythemia like,but these findings are not well established [24]. At least two other breakpoint clus-ter regions are present on BCR, and although they occur rarely in CML, they areassociated with unique phenotypes. BCR–ABL1 fusions involving m-BCR (minorbreakpoint cluster region) are most often seen in acute lymphoblastic leukemia andare also rarely seen in CML. This region spans 55 kb of BCR between exons 1 and 2,so rearrangements involving this region include only the most 5′ end of BCR (e1a2– exon 1 of BCR to exon 2 of ABL1) [6, 24]. As might be expected, this fusionresults in a smaller protein, the p190 fusion protein. Although rare in CML, thisfusion has been associated with an increase in monocytes and phenotypic featuresoverlapping with chronic myelomonocytic leukemia (CMML) and has been associ-ated with inferior response to imatinib [24, 26]. Interestingly, very small amountsof the p190 fusion protein are frequently co-expressed in cases with M-BCR break-points and p210 as the major fusion protein. Rare cases of CML also involve theμ-BCR (micro breakpoint cluster region), which creates an e19a2 (exon 19 of BCRto exon 2 of ABL1) rearrangement that includes more of BCR, and results in thelonger p230 fusion protein. These cases have been associated with a greater propor-tion of mature neutrophils within the neoplastic proliferation and show clinical andhematologic overlap with chronic neutrophilic leukemia [6, 24]. BCR–ABL1 rear-rangements involving areas of BCR outside these three clustered regions have alsobeen rarely reported, including intron 6, intron 8, and intron 10 [24].

The BCR–ABL1 fusion results in a chimeric protein that contains part BCR andpart ABL1. This chimeric fusion is a different result from that of most translo-cations in lymphoma, which bring an intact oncogene under the regulation ofa constitutively active gene, such as the immunoglobulin heavy chain gene inB-cell lymphomas. Despite its chimeric nature, the BCR–ABL1 fusion proteinretains the non-receptor tyrosine kinase activity that is attributed to ABL1. Normalintact ABL1 is a member of a family of tyrosine kinases, which use a phosphatefrom adenosine triphosphate (ATP) to phosphorylate tyrosine amino acid residueswithin proteins [25]. This tyrosine phosphorylation serves as the mechanism ofsignal transduction that normal ABL1 uses to help regulate cytoskeleton structureand the cell cycle [6, 27]. The BCR–ABL1 fusion has deregulated and increasedABL1 kinase activity. Studies in mouse models have shown that the ABL1 tyro-sine kinase activity of the BCR–ABL1 fusion protein is necessary for induction of

222 D.S. Bosler

CML-like disease and is most effective at inducing CML-like disease when intro-duced into hematopoietic stem cells [12, 27]. Although increased ABL1 kinaseactivity has some transformative properties, murine studies have also shown thatABL1 kinase activation alone is insufficient to reliably induce a CML-like dis-ease. Other functional regions within the BCR–ABL1 fusion are also necessaryfor transformation, such as the BCR coiled-coil multimerization domain, the BCRGRB-2 binding site, the ABL1 SH2 domain, and various other tyrosine phospho-rylation sites [27]. The coiled-coil multimerization domain acts to enhance thekinase activity of ABL1, while the SH2 domain facilitates activation of the RASpathway, and the GRB-2 binding site participates in cell proliferation and survivalsignals through activation of RAS, SHP2 and PI3K-AKT pathways [27]. The down-stream effects of BCR–ABL1 include induction of hematopoietic growth factorssuch as interleukin-3 (IL-3), granulocyte colony-stimulating factor (G-CSF), andgranulocyte–macrophage colony-stimulating factor (GM-CSF). Altered apoptoticsignaling pathways such as upregulation of BCL-2 also play a role. The result isuninhibited growth and proliferation that, combined with reduced apoptosis, pro-duce the morphologic and clinical findings in CML. Taken together, these findingshighlight the central role of BCR–ABL1 in the pathogenesis of CML.

Induction of blast phase is associated with additional compounding genetic alter-ations. One study found additional abnormalities in 65 and 82% of blast-phase andaccelerated-phase cases, respectively, with the most frequent additional abnormali-ties being a second Philadelphia chromosome, trisomy 8, and isochromosome 17q[23]. Other abnormalities associated with disease progression include mutationsin p53, RB1, P16, c-MYC, and RAS; increased BCR–ABL1 transcript expression;and methylation abnormalities of ABL1 [28]. Silencing of various tumor suppressorgenes may occur through a variety of mechanisms, including deletion, inactivatingmutations, and promoter methylation.

Diagnostic Testing

As CML is a disease defined by the presence of the BCR–ABL1 fusion, a reliableand sensitive method of detecting the fusion is critical to the diagnostic processwhenever a diagnosis of CML is considered. Although cytogenetic karyotyping is areliable method for detection of t(9;22) and complex translocations, a small percent-age of cases are not detectable by this method and would be missed if cytogeneticswere used as a sole method of molecular testing [24]. Additionally, the process ofcytogenetic karyotyping takes a minimum of several days to complete, resulting ina delay in diagnosis that is unnecessary if other methods can provide an answersooner. RT-PCR designed to detect the BCR–ABL1 transcript for the p210 fusion isa rapid and very analytically sensitive technique that is useful in diagnosing CML ifpositive, but a negative result does not completely exclude CML since a small per-centage of cases have alternate fusion transcripts that would be missed by a primerset designed to detect a single fusion protein [29]. Additionally, rare deletions occur-ring in or near the primers may cause false-negative results by PCR even when the


correct breakpoints are targeted [30]. Given the limitations of these methods, othertechniques have been employed when necessary as adjuncts to provide the optimalclinical and analytical sensitivity.

One method commonly used to detect the BCR–ABL1 fusion gene at diagno-sis is fluorescence in situ hybridization (FISH). One of the most common formsof this method is the locus-specific identifier (LSI) BCR/ABL Dual Color, DualFusion Translocation Probe by Vysis (Downers Grove, IL). This probe set is a com-bination of LSI BCR probe (22q11.2) labeled with SpectrumGreen and LSI ABL1probe (9q34) labeled with SpectrumOrange. Since the probe sets span and flankthe breakpoints, the reciprocal translocation creates two fusion signals, one on thePhiladelphia chromosome (derived chromosome 22) and the other on the alteredchromosome 9. Since FISH can be performed on interphase cells, results are notdependent on cell culture, resulting in the ability to perform testing on a widerrange of samples as well as a more rapid turn around time than can be achievedby the culture-dependent cytogenetic karyotyping [31]. Additionally, BCR–ABL1fusions can be detected by FISH in rare Philadelphia chromosome-negative cases ofCML [24, 31, 32].

Comprehensive multiplex PCR strategies have also been employed as a reliableand sensitive means of detecting BCR–ABL1 fusions. These methods often use RT-PCR, since starting with messenger RNA means that the introns have been splicedout, allowing amplification of products with a variety of breakpoints using fewerprimers. One published multiplex RT-PCR method using multiple labeled primerscombined with capillary electrophoresis fragment size analysis detects transcriptswith a wide array of breakpoints, including e1, e13, e14, e19 on BCR, as well asa2 and a3 on ABL1 [33]. Others have employed bead-array-based technology in thedetection phase to distinguish products with various breakpoints. One advantage thatthese methods have over FISH is that they provide more detailed information aboutthe breakpoints present in the fusion, which may have clinical implications and isimportant in identifying a target for detection of minimal residual disease. Othertechnologies such as protein-based flow cytometric immunobead assays have shownpromising results in some studies and may ultimately provide another powerful andflexible option for detection [34].

Disease Monitoring/Response to Therapy

As therapeutic options for CML have improved, methods for detecting smalleramounts of residual disease have become increasingly relevant (summarized inTable 7.2).

Hematologic remission, or normalization of peripheral blood counts and spleensize, was an important indicator of control of disease when hydroxyurea was amainstay of CML therapy [35, 36]. Cytogenetic response became important inthe era of interferon-α (alpha) therapy, since it could induce complete cytogeneticresponse (loss of detection of the Philadelphia chromosome by cytogenetic meth-ods), in 10–20% of patients, and detection of the level of cytogenetic response

224 D.S. Bosler

Table 7.2 Definition of various response types and equivalent estimated tumor burden

Response type Hematologic Cytogenetic Molecular

Definition Complete:normalization ofperipheral bloodcounts and spleensize

Partial (major):<35% Ph+ cellsdetected

Complete: no Ph+cells detected

Major: 3 log reductionfrom standardizedlevel at diagnosis

Complete: notranscript detected(poorly reproducedlevel)

Approximateequivalent tumorburden (number ofcells)

1010–1012 Complete:109–1010

Major: <109

Complete: ? 106

Reproduced and adapted with permission from Baccarani et al. [42].

was important because it was associated with both a longer duration of chronicphase and an improved survival compared with those that did not achieve thislevel of response [35, 36]. Patients could be stratified into prognostically significantgroups based on whether they had a complete cytogenetic response, major cyto-genetic response (<35% of cells with the Philadelphia chromosome), or no majorcytogenetic response (>35%). The advent of imatinib mesylate therapy broughtabout major changes in the course and survival of CML patients as a group andrequired yet another level of detection sensitivity to provide adequate stratificationof response to therapy.

The superiority of imatinib was established in the landmark IRIS (InternationalRandomized Study of Interferon versus STI571) clinical trial, in which 1106 sub-jects with newly diagnosed CML were randomized to receive either imatinib(STI571) or interferon-α plus cytarabine [37]. The rates of complete cytogeneticremission at 12 months were 69% for the imatinib group and 7% for the interferon-αgroup. The imatinib response rates were not only durable but also actually improvedover time, with 87% of imatinib-treated subjects achieving complete cytogeneticresponse by 60 months [38]. Importantly, the cytogenetic response was highlypredictive of a favorable outcome – of those subjects who achieved complete cyto-genetic response at 12 months, 97% were progression free at 60 months, comparedwith 81% for those that had not achieved a major cytogenetic response [38]. Amongsubjects with a complete cytogenetic response, quantitative measurement of BCR–ABL1 transcript levels provided additional useful prognostic information. Subjectsreceiving imatinib achieved major molecular response (>3 log reduction from stan-dardized baseline) faster and at a higher rate than those receiving interferon-α(37).Fifty-seven percent of imatinib-treated subjects with a complete cytogenetic remis-sion achieved a major molecular response by 12 months (39% of all imatinib-treatedsubjects), and these subjects were 100% free from progression to accelerated-phaseor blast-phase crisis at 60 months of follow-up [37, 38]. Failure to achieve at leasta 2 log reduction of BCR–ABL1 transcripts by the time of complete cytogenetic


response and failure to reach a 3 log reduction at any time have subsequently beenassociated with poorer progression-free survival [39]. Monitoring of BCR–ABL1transcript levels early in treatment (4 weeks and 3 months) has also been shownto be predictive of eventual major cytogenetic response and progression-free sur-vival [40]. Imatinib therapy has so changed the course of disease in CML that manyprognostic factors recognized during previous eras of therapy are now supersededby response to imatinib [17, 38, 41].

As management of CML has evolved, it has become necessary to integratethe various methods of disease monitoring in the optimal way for measuringboth initial response and maintenance of response to therapy. Recommendationsfrom the European LeukemiaNet for CML disease monitoring are summarized inTable 7.3 [42].

The prioritized goals of treatment according to these recommendations are(in order) complete hematologic response, complete cytogenetic response, majormolecular response, and “complete” molecular response. If these goals are achieved,they most often happen in order as imatinib therapy continues over time, and asTable 7.4 shows, the various goals are built into the evaluation of response to therapy

Table 7.3 Samples and recommended frequencies for various types of response

Hematologic response Cytogenetic response Molecular response

Sample Peripheral blood (CBC) Bone marrow aspirate Peripheral blood or marrowRecommended

frequencyEvery 2 weeks until

complete response;then every 3 months

Every 6 months untilcomplete response;then every year

Every 3 months; mutationalanalysis as indicated

Reproduced and adapted with permission from Baccarani et al. [42].

Table 7.4 Criteria for lack of response

Time afterdiagno-sis Failure Suboptimal response

3 months No response Less than complete hematologicresponse

6 months 1. Less than complete hematologic response2. No cytogenetic response (Ph >95%)

Less than partial cytogeneticresponse (Ph >35%)

12 months Less than partial cytogenetic response(Ph >35%)

Less than complete cytogeneticresponse

18 months Less than complete cytogenetic response Less than major molecular responseAnytime 1. Loss of complete hematologic response

2. Loss of complete cytogenetic response3. Mutationa

1. Additional cytogeneticabnormalities

2. Loss of major molecular response3. Mutationa

aMutation should be one with known high level of insensitivity to imatinib in order to fulfill thiscriterion.Reproduced and adapted with permission from Baccarani et al. [42].

226 D.S. Bosler

in a time-dependent manner. If the goals are not achieved or are lost, evaluation forresistance to therapy is often warranted as described in more detail below.

Since evaluations of hematologic response and molecular response can be per-formed on peripheral blood, only cytogenetic evaluation requires periodic bonemarrow sampling. Although FISH testing may allow evaluation of cytogeneticresponse in peripheral blood due to its increased sensitivity, the panel did notrecommend FISH for this purpose because outcome data had been based on cytoge-netics, and because cytogenetics has the important advantage of detecting additionalchromosomal abnormalities [42].

Quantitative RT-PCR in Disease Monitoring – Optimization,Control Genes, and Reporting

As we have seen, the realization of widespread imatinib use necessitated widespreadimplementation of a sensitive and reliable test to monitor response to therapy. Sincepivotal imatinib trials had used quantitative RT-PCR, this method was a logical ini-tial choice [37, 38]. Modern quantitative PCR methods have many advantages thatare important to BCR–ABL1 fusion monitoring, including high sensitivity, a closedsystem that reduces the possibility for contamination, comparatively good repro-ducibility and a wide dynamic range of 5–6 logs [29]. The basic steps involved inperforming this assay include RNA extraction, reverse transcription, amplificationwith simultaneous detection of target cDNA using technology such as Taqman orFRET probes, and comparison to a calibration curve that converts the cycle numberat detection (Ct) to a quantitative value (Fig. 7.3). The optimization and standardiza-tion of each of these steps has been the focus of extensive study and discussion [43,

Fig. 7.3 Quantitative RT-PCR is the established method used in molecular level disease moni-toring of BCR–ABL1 transcripts in CML treated with tyrosine kinase inhibitors. Curves showingdetection of both the BCR–ABL1 fusion at the M-BCR breakpoint and the control gene (in thiscase ABL1) are shown. Reporting the quantity of BCR–ABL1 relative to the quantity of ABL1 inthe same sample (referred to as the normalized copy number) can help control for variability dueto RNA degradation as the sample ages


44]. The Europe Against Cancer (EAC) Program established a standardized protocolfor minimal residual disease testing using quantitative RT-PCR with TaqMan detec-tion technology in 26 laboratories across 10 countries, and their design and protocolfor BCR–ABL1 transcript measurement have provided a reference standard againstwhich other methods can be compared [44].

Quantitative RT-PCR has the advantages and disadvantages associated with mes-senger RNA as a starting material. As discussed earlier, one advantage is that intronshave been spliced out, overcoming much of the variability of breakpoints in ABL1and allowing the use of fewer primer sets. One disadvantage is that RNA is unsta-ble and degrades over time, leading to quantitative variability depending on thetime from draw to analysis unless appropriate compensatory mechanisms are inplace. For this reason, a control gene must be incorporated into quantitative RT-PCR assays. Reporting the BCR–ABL1 expression level (in copy number) as a ratioto the transcript copy number of an appropriate control gene, a relative value knownas the normalized copy number (NCN), can help compensate for RNA degradationthat occurs with aging of the specimen, as well as variability in the efficiency ofthe extraction and reverse transcription steps [43–45]. The ideal control gene wouldhave a predictable, stable expression level and rate of degradation that is comparableto that of BCR–ABL1, yet its measurement would be unaffected by BCR–ABL1 lev-els [45, 46]. Beillard et al. extensively studied the appropriateness of 14 candidatecontrol genes as part of the EAC effort and, based on stability, expression levelsin various samples, and lack of pseudogene amplification, concluded that ABL1,beta-glucuronidase (GUSB), and beta-2-microglobulin (B2M) had the appropriatecharacteristics to potentially serve as control genes for quantitative minimal residualdisease testing [45]. Ultimately, the authors selected ABL1 as the most appropri-ate based on comparatively consistent expression levels in different sample typesand in both normal and leukemic samples (BCR was not evaluated). In practice,many different control genes are used, with BCR, ABL1, and GUSB being the mostprevalent. BCR and ABL1 have both been established by use in high-profile stud-ies, both have the advantage of convenience from an assay design standpoint, andABL1 particularly has demonstrated similar stability to BCR–ABL1 [37, 43–46]. Apotential limitation of ABL1 as a control gene is that, since its primers may alsoamplify BCR–ABL1 fusion transcripts, the control genes may be over-representedwhen BCR–ABL1 levels are high, resulting in underestimation of BCR–ABL1 whenexpressed as NCN [45, 47]. Since precise quantitation is most important when BCR–ABL1 levels are low rather than at high levels where the potential for skewed resultsis greatest however [48], this bias is of undetermined clinical relevance, and ABL1remains in widespread use as a BCR–ABL1 assay control gene, along with BCR,GUSB, and others.

Many acceptable platforms are available for BCR–ABL1 quantitative monitor-ing, with variable capacity for throughput available depending on anticipated testingvolume. Silvy et al. compared seven different platforms using serial dilutions froma BCR–ABL1-positive cell line, plasmid standard curves, and results expressed asnormalized copy number according to the EAC protocol and found the methodsgenerally equivalent and within acceptable performance standards [49].

228 D.S. Bosler

Another issue confronting clinical laboratories performing quantitative BCR–ABL1 testing is how best to report the results. As described above, reportingBCR–ABL1 levels as copy number per microgram does not account for variability inage of the specimen or efficiency of the assay. While using normalized copy numbercorrects many of these issues, it does not account for variability between labora-tories, and raw NCNs do not have well-established clinical relevance. For thesereasons, most laboratories report BCR–ABL1 levels using the method first describedby Hughes et al. for use in evaluating subjects on the IRIS trial [37]. In order tostandardize results between laboratories, this group converted their normalized copynumber results into a log reduction from a standardized median created from mea-surement of 30 newly diagnosed (baseline) patients at each laboratory. Using thismethod to report results has a few distinct advantages. First, since each lab createsa standard baseline using its own method, reporting the results relative to each lab’sbaseline may correct some of the variability between laboratories. Second, sincethis reporting method was used in the pivotal IRIS trial, it has established clinicalvalidity – a 3 log reduction from the standardized median baseline was the cutoffvalue shown to have clinical relevance, as described in more detail earlier [37]. Inpractice, many labs report both the NCN and the log reduction.

Quantitative RT-PCR in Disease Monitoring – InterlaboratoryStandardization Efforts

Despite the use of methods with high internal reproducibility and standardization byreporting log reduction from a standardized median, considerable variation betweenlaboratories remains an issue. One sample sharing study of 38 laboratories foundthat results varied considerably from lab to lab, with log reduction ranges betweenlaboratories spanning 2.5–3.0 logs for the same dilution [47]. These results empha-size the importance not only of careful optimization of methods but also the needfor standardization between laboratories. Additional efforts at standardization thatare currently underway include the application of an internationally standardizedconversion factor analogous to the International Normalized Ratio (INR) used inmonitoring anticoagulation with warfarin, and preparation and dissemination ofstable standard reference and calibration materials. Proposals for an internationalstandardization include use of standardized calibrators to produce conversion fac-tors for laboratories that would allow normalization to two anchor points, with themedian standard baseline equal to 100% and the 3 log reduction equal to 0.1%[48, 50]. In an international study comparing laboratory values, Branford et al.showed that application of such a laboratory-specific conversion factor can resultin higher rates of concordance between laboratories [51]. Meanwhile, efforts atproducing stable, standardized reference material are progressing. A lyophilizedpreparation of K562 cell line has been shown to have potential as a stable andreliable quality control reagent that could serve as an international standard [52].Availability of such standards appears imminent and will significantly advance theefforts at inter-laboratory standardization.


Resistance to Tyrosine Kinase Inhibitor Therapy

A brief review of imatinib’s mechanism of action is warranted as preparationfor a discussion regarding imatinib resistance. Originally known as signal trans-duction inhibitor 571 (STI 571), imatinib inhibits the tyrosine kinase activity ofBCR–ABL1, platelet-derived growth factor receptors (PDGFRA and PDGFRB),and c-KIT, by competitively blocking the ATP binding site (Fig. 7.4). Imatinibachieves high affinity and specificity for BCR–ABL1 through specific binding toBCR–ABL1’s inactive form, which locks the tyrosine kinase in an inactive state[14, 53]. With the kinase in the inactive form and access to ATP blocked, substratephosphorylation cannot take place, preventing the downstream cascade of eventsthat drives transformation in CML (as described above in more detail).

Indications for resistance testing include either signs indicating initial treatmentfailure or suboptimal response (see Table 7.4), or indications that response has beenlost, such as accelerated- or blast-phase, clonal evolution, loss of cytogenetic orhematologic response, or increasing BCR–ABL1 transcript level [29, 42, 54]. Aconfirmed increase in BCR–ABL1 transcript levels of 5–10-fold or greater has beenproposed as a threshold that would prompt evaluation for resistance [43, 54].

A.

B.

PPP

BCR-ABL1Substrate

Tyr Substrate

TyrP

BCR-ABL1Substrate

Tyrimatinib

PPP

Substrate

Tyr

Fig. 7.4 (a) BCR–ABL1 acts as a constitutively active tyrosine kinase, using ATP as a phosphatesource to phosphorylate substrate proteins at tyrosine residues, resulting in activation of down-stream pathways. (b) Imatinib blocks the ATP binding site and binds BCR–ABL1 in its inactiveconformation, preventing tyrosine phophorylation of substrates

230 D.S. Bosler

Resistance to imatinib therapy may be primary or secondary, and either BCR–ABL1 dependent or independent [29]. Initial therapy failure (primary resistance)is generally related to pharmacokinetic and/or pharmacodynamic properties of ima-tinib and the treated host including abnormal cellular drug transport [55]. Secondaryresistance is more common, occurring at an annual rate of about 4%, and mostcommonly results from mutations in the ABL1 kinase domain that block or over-come imatinib’s activity [29, 55–57]. Approximately 30–50% of chronic-phaseCML patients with signs of resistance have a detectable ABL1 kinase domainmutation, with higher rates in accelerated and blast phase [54, 58]. Other mech-anisms of resistance include BCR–ABL1 amplification and/or overexpression andBCR–ABL1-independent mechanisms such as clonal evolution [42, 55]. Some stud-ies have demonstrated LYN-mediated upregulation of BCL-2 as a mechanism ofBCR–ABL1-independent resistance [59, 60].

ABL1 kinase domain mutations tend to occur in one of four functional regions:the P-loop, the imatinib binding site, the catalytic domain, and the activation loop[14, 54]. Although over 70 ABL1 kinase domain mutations have been reported,mutations within just eight codons (M244V, G250E, Y253F/H, E255K/V, T315I,M351T, F359V, H396R) represent up to 85% of resistance [54, 58]. Not all muta-tions are functionally equivalent, and the implications of detection may vary. Forexample, imatinib dose escalation may be a reasonable approach with mutationssuch as M351T that show mild resistance, while other mutations may prompt aswitch to second-generation tyrosine kinase inhibitors, and yet others such as T315Imay indicate resistance not only to imatinib but also to newer tyrosine kinaseinhibitors [54]. Mutations within the P-loop (codons 248–256) and the T315I muta-tion typically confer the greatest level of resistance [54, 58]. Although newer andmore potent tyrosine kinase inhibitors such as dasatinib and nilotinib serve aseffective second-line therapies in patients who have failed imatinib [57], contin-ued experience with these drugs has demonstrated that specific mutations are alsoassociated with their use [54, 61]. In one study of subjects treated with dasatinibor nilotinib after imatinib failure, new mutations developed in 26% of cases [61].Although new mutations during dasatinib or nilotinib therapy tended to occur atcertain codons (e.g., 253, 317, 359), only the T315I mutation was associated with ahigher level of resistance. The high level of resistance to available therapies foundin T315I-mutated cases has prompted the development of newer potential therapiessuch as the aurora kinase inhibitors, many of which have shown promising clinicaland/or laboratory activity against T315I, and are in early phase clinical trials [62].

Several questions regarding ABL1 kinase domain mutations and resistanceremain unanswered. The number of ABL1 kinase domain mutations reported contin-ues to grow, and the functional significance of many remains undetermined. It is notclear whether mutations occur after initiation of therapy or if they become apparentas a dominant mutated clone emerges under selective therapeutic pressure, althoughdetection of low levels of mutation in treatment naïve subjects suggest the latter[63]. Additionally, detection of these low-level mutations was not associated withdevelopment of resistance or shortened survival [63], prompting questions about atwhat level a mutated clone becomes clinically relevant.


Several methods have been implemented in detection of ABL1 kinase domainmutations, including denaturing HPLC, direct chain termination-based (Sanger)sequencing, pyrosequencing, allele-specific oligonucleotide (ASO) PCR, and liquidbead array [43, 54]. The currently recommended technique for mutation screeningis Sanger sequencing [43, 54]. Although this method is less sensitive than others,it offers the advantages of comprehensive evaluation of the targeted area as wellas bidirectional confirmation and is routinely available in most labs. Additionally,while methods such as ASO-PCR offer much greater sensitivity of detection, low-level detection of mutations does not appear to have clinical relevance at this time,and ASO-PCR assays are limited to detection of the specific mutations for whichthey are designed. One potential challenge with sequencing is detection of mutationswith uncertain clinical significance. Although there is generally good correlationbetween in vitro resistance to tyrosine kinase inhibitors and clinical resistance[54, 61], new or recently reported mutations may not have been evaluated.Additionally, correlation of BCR–ABL1 transcript levels with levels of the ABL1-mutated clone has shown that the presence of a mutation does not invariably accountfor clinical resistance [64].

To address some of the challenges associated with testing for ABL1 kinasedomain mutations, the Association for Molecular Pathology convened an ABLMutation Working Group. The resulting publication of laboratory practice guide-lines [54] provides guidance on testing methods and recommendations for reportingand makes proposals for standardization, reference materials, proficiency test-ing, and development of a publicly available BCR–ABL1 kinase domain mutationdatabase for use in interpreting the clinical significance of results.

Summary: Integration of Molecular DiagnosticTesting into CML Management

Molecular diagnostic testing is an important component of many aspects of diag-nosis and management of CML, including diagnosis itself, disease monitoringand prognosis, and response to therapy and resistance. At diagnosis, reliable andsensitive methods for detection of BCR–ABL1 fusion include fluorescence in situhybridization and some comprehensive multiplex RT-PCR assays. Cytogenetickaryotyping and PCR for the p210 fusion also detect the vast majority of CMLcases, but they miss cases that are cytogenetically cryptic or that involve rare vari-ant fusions, respectively. Nonetheless, it remains important to perform these tests atdiagnosis in order to get baseline information that will become important in diseasemonitoring. Assuming that a p210 BCR–ABL1 fusion is present at diagnosis, quan-titative RT-PCR designed to detect the p210 fusion is the most sensitive methodof monitoring residual disease, and the increased sensitivity of this method hasbeen shown to be clinically relevant. Obtaining a baseline result for the RT-PCRp210 assay is helpful in order to ensure that cases that are subsequently nega-tive for the BCR–ABL1 fusion during disease monitoring are negative because theyhave achieved complete molecular response rather than because they have a variant

232 D.S. Bosler

rearrangement that is not detected by the assay. Additionally, cytogenetic karyotyp-ing remains an important method of detecting additional cytogenetic abnormalities,such as a second Philadelphia chromosome, trisomy 8, or isochromosome 17q,which may be harbingers of disease progression and are not detected by RT-PCR orFISH. When molecular disease monitoring, hematologic findings, or clinical coursesuggests loss of response to therapy, analysis of the ABL1 kinase for mutations thatconfer resistance to imatinib and/or other tyrosine kinase inhibitors helps to guidechoice of subsequent therapy.

BCR–ABL1-Negative “Classic” Myeloproliferative Neoplasms

Like CML, the BCR–ABL1-negative classic myeloproliferative neoplasms (MPNs)are long-established recognized clinical entities. Although some were known bydifferent names, polycythemia vera (PV), essential thrombocythemia (ET), andprimary myelofibrosis (PMF) were all included along with CML in Dameshek’sfirst conceptual grouping of myeloproliferative neoplasms in 1951 [2]. Unlikethe early advancements in pathogenesis and diagnostic markers seen in CMLhowever, substantive discoveries regarding the BCR–ABL1-negative MPNs laggedand their diagnosis relied almost exclusively on clinicopathologic findings untilvery recently. The discovery of JAK2 mutations and, to a lesser extent, MPLmutations in the BCR–ABL1-negative classic MPNs have both advanced theknowledge of pathogenesis and provided much more robust markers of clon-ality that have greatly simplified their diagnosis, with much remaining to beexplained.

Epidemiology, Clinical, and Laboratory Features

As a group, PV, ET, and PMF occur with about three times the frequency of CML[65]. Estimated incidence rates of each vary by study, but it appears that ET maybe slightly more common than the other two, with all having a peak incidencein the sixth and seventh decades [1, 65, 66]. These three closely related enti-ties are united by the presence of unregulated proliferation of myeloid precursorcells. This proliferation involves varying combinations of erythroid, granulocytic,and/or megakaryocytic cell lines. Although there can be some overlap betweenthem, PV generally shows predominant erythroid proliferation, while ET showspredominant megakaryocytic proliferation, and PMF shows predominant myelofi-brosis [1, 67]. The clinicopathologic overlap intimates the underlying clonal stemcell abnormality that is present in each of these disorders. Patients present withsplenomegaly and variable abnormalities of peripheral blood cell counts, includingelevated red blood cell mass (PV) or platelets (ET), or combinations of elevatedcounts and cytopenias with a leukoerythroblastic blood picture indicating marrowreplacement by fibrosis (PMF). Leukocytosis may also be present. Risk of throm-boembolic events is significantly increased and is the leading etiology of morbidity


and mortality in MPN patients [1, 68]. Bone marrow morphology can be normo-cellular to mildly hypercellular (ET), markedly hypercellular (PV), or have markedfibrosis (PMF). Marrow histology overlaps, with hyperplasia of various myeloidcomponents, morphologically abnormal and clustered megakaryocytes, and retic-ulin fibrosis found to varying degrees in each entity. Dysplasia is not a prominentfeature. The clinical course is indolent and variable, with risk of marrow failure dueto progressive myelofibrosis, and progression to blast crisis in a minority. ET is theleast likely to progress, with those affected most often showing an indolent course[1, 69, 70]. Although distinction between ET and PMF is of prognostic significance,the distinction based on morphologic grounds can be unreliable [71]. Despite theoverlap of clinical and pathologic features, the heterogeneity indicates that thereare likely varied pathophysiologic mechanisms at work, many of which remainunexplained.

JAK2 V617F’s Contribution to Diagnosis of MPNs

Prior to the discovery of the JAK2 mutation, only about 20% of the BCR–ABL1-negative classic MPNs had recognizable clonal abnormalities [1]. The combinationof the lack of a marker of clonality in a majority of cases and significant over-lap with non-neoplastic conditions resulted in a complicated diagnostic processthat most often included an exhaustive exclusion of alternative explanations forsigns and symptoms before a diagnosis of a myeloproliferative neoplasm couldbe made. For example, the differential diagnosis of erythrocytosis includes a widevariety of etiologies aside from polycythemia vera [72]. Congenital causes such asmutations in the VHL gene, activating mutations of EPO, or high oxygen affinityhemoglobin variants result in erythrocytosis. Additionally, a number of poten-tially more common causes of acquired erythrocytosis must be excluded, includingthose resulting from hypoxia such as tobacco smoking, chronic lung disease orhigh altitude, and paraneoplastic erythropoietin production from a variety of non-hematolymphoid neoplasms (e.g., renal, hepatocellular, uterine leiomyomas). Lowerythropoietin (EPO) levels are seen in PV and can help to rule out many of themore common causes of secondary erythrocytosis. In addition to ruling out alter-native causes, the evaluation for a myeloproliferative neoplasm most often alsoincluded the relatively cumbersome measurement of red cell mass in order to ruleout apparent (relative) polycythemia [72]. Similarly, thrombocytosis has a wide vari-ety of potential etiologies other than ET, such as iron deficiency anemia, infection,inflammation, other underlying malignancies (hematologic or other), or response tomedications [73].

When present, the JAK2 V617F mutation provides a marker of clonality that dra-matically simplifies the diagnostic process for BCR–ABL1-negative MPNs. Locatedon chromosome 9p24, the JAK2 gene encodes a non-receptor protein tyrosine kinasethat associates with homodimeric growth factor receptors. The JAK2 V617F muta-tion’s presence in a majority of BCR–ABL1-negative MPNs was reported essentiallysimultaneously by four research groups in 2005 [74–77]. The mutation is found

234 D.S. Bosler

Table 7.5 JAK2 and MPL in BCR–ABL-negative MPNs

Mutation

JAK2 V617F JAK2 Exon 12 MPL W515

PV 95% ∼ 5% –ET 50% – 1–4%PMF 50% – 5–11%

in about 95% of PV, roughly half of ET and PMF (Table 7.5), and in small per-centages of other myeloid neoplasms including chronic myelomonocytic leukemia,atypical CML, acute myeloid leukemias, and refractory anemia with ring sider-oblasts and thrombocytosis (RARS-T) [78–82]. The mutation consists of a singlenucleotide substitution (G1849T) in exon 14 of JAK2 that results in a single aminoacid change from valine to phenylalanine at position 617. This mutation within theJH2-negative regulatory domain results in phosphorylation and constitutive acti-vation of JAK2, which facilitates ligand-independent activation at homodimericgrowth factor receptors (such as the EPO receptor, thrombopoietin receptor, andGCSFR) and activation of STAT signaling proteins [83, 84]. Ultimately, thereis unregulated (ligand independent) downstream activation of transcription fac-tors involved in proliferation and survival via activation of the MAPK and PI3Kpathways (Fig. 7.5) [81]. As this mutation is not detected at significant levels innormal subjects, it serves as a marker of clonality, precluding the need for exhaus-tive exclusion of potential alternative etiologies of signs and symptoms before adiagnosis can be made. It is important to remember that the JAK2 V617F muta-tion is not specific for any MPN, nor does its absence exclude the possibility ofan MPN.

Other JAK2 Mutations in PV

Investigation of some of the 5% of PV cases that are JAK2 V617F negative hasrevealed that most of these cases contain alternate JAK2 abnormalities. The first suchinvestigation showed four novel mutations within exon 12 of JAK2, with detectedabnormalities including the amino acid substitutions K539L and H538Q withK539L, as well as a small deletion, N542-E543del, and a deletion/insertion, F537-K539delinsL [85]. Subsequent studies have confirmed these results and uncoveredadditional mutations within exon 12, most often occurring within the region con-taining amino acid residues 530–550 [86–88]. Overall, the N542-E543del is themost frequent abnormality [82]. These mutations appear to occur only in PV, andphenotype–genotype correlations suggest that these cases more often present withan isolated erythrocytosis. In contrast to the V617F mutation, most of the exon 12mutated cases are heterozygous [82, 85–87]. One recent study screening 20,000samples from patients with suspected MPNs suggests that alternative JAK2 muta-tions may occur over a much broader range of the gene than initially suggested [89].


Erythropoietin receptor

JAK2

STAT

JAK2P P

PI3K MAPK

Proliferation, Survival

Erythropoietin receptor

STAT

P P

JAK2 V617FJAK2 V617F


A. B.

C.

JAK2JAK2

STAT

P P

MPL W515L/K


Thrombopoietin receptor (MPL)

Fig. 7.5 (a) Physiologic activation of the erythropoietin receptor and downstream pathwaysrequires binding by erythropoietin, which results in JAK2 activation and phosphorylation, withrecruitment and activation of STAT signaling proteins and downstream induction of proliferationand survival genes via the PI3K and MAPK pathways. (b) Mutated JAK2 protein results in consti-tutive activation of JAK2 and downstream pathways that is independent of erythropoietin binding.(c) Mutations in MPL (thrombopoietin receptor) result in activation of the same pathway due toligand-independent constitutive activation of the receptor itself

Ma et al. confirmed the presence of mutations within the previously reported exon12 hot spots but also reported mutations throughout exons 13–15 [89]. Although thefunctional significance of these new mutations and their association with diseaseare not established, the study suggests that a broader net be cast when testing foralternative mutations.

Effect of JAK2 Allelic Burden

A growing body of literature is examining the relevance of the quantity of JAK2V617F present in BCR–ABL1-negative classic MPNs. A subset of PV cases(25–30%) has a homozygous V617F mutation most often resulting from mitoticrecombination (acquired uniparental disomy) rather than deletional loss of heterozy-gosity [76, 77, 90]. Homozygosity is more common in PV, occurring relativelyrarely in ET and PMF [76, 77, 90]. Homozygous PV cases are associated withincreased hematocrit and leukocyte counts, lower platelet counts, a greater degree

236 D.S. Bosler

of splenomegaly, and pruritus [90]. More recently, studies have focused on quan-titative JAK2 V617F burden rather than zygosity, with mixed results. As might beexpected, increasing allelic burden in PV correlates with the findings in homozy-gous cases, and ET and PMF cases with increased JAK2 V617F allelic burdentend to show similar findings such as increased hemoglobin and leukocyte count[91–94]. Although many studies have attempted to stratify prognosis and throm-botic risk according to allelic burden, results have been conflicting and likely reflecta more complex underlying pathobiology that remains to be more completely sortedout. A few studies have shown that low JAK2 V617F allele burden in PMF correlateswith poor prognosis [93, 95].

JAK2 Mutation Detection Methods

Numerous methods can be employed to detect the JAK2 V617F mutation. Onerecent study performed a head-to-head comparison of four methods, including com-mercial assays using PCR with enzyme digestion and fragment size analysis, PCRwith differentially fluorescent allele-specific Taqman probes, and allele-specificoligonucleotide (ASO) PCR and a laboratory-developed test using PCR followedby melt curve analysis [96]. The quantitative ASO-PCR showed the greatest sensi-tivity (0.0098%), followed by the two semi-quantitative methods (PCR with Taqmanprobes and PCR with enzyme digestion and capillary electrophoresis), which bothhad a sensitivity of 1.25%. The PCR with melt curve analysis showed the least sensi-tivity, with a slight curve abnormality at 10%, and clearly visible peaks only presentat the 20% and undiluted (100%) levels. Evaluation of 33 patients’ bone marrowand blood samples including a variety of MPNs as well as non-MPN malignanciesand a variety of reactive conditions showed concordance between all four methodsin 30 of 33 cases. Two of the three discordant cases were negative by the less sensi-tive melt curve analysis and positive by the other three methods [96]. Although theauthors felt that the commercial assays had all performed well, the PCR followed byenzyme digestion and electrophoresis required more hands-on time and posed someinterpretation difficulties with low-level positives (Fig. 7.6). The ASO-PCR tech-nique has the added benefit that it can be used as a quantitative assay. Although theclinical relevance of JAK2 V617F quantitation is currently unclear, it may becomemore relevant in the future either as a prognostic tool if issues surrounding alleleburden become clearer or as a disease monitoring tool if JAK2 inhibitors proveeffective. One potential challenge encountered when using an assay with such highsensitivity is what to do with low-level positive results. Since low-level positivesare seen in subjects without any evidence of MPNs (presumably normal), exten-sive study of normal populations is recommended to establish a normal range priorto implementation. Also, although one might be concerned about quantitative dif-ferences between blood and bone marrow samples, one relatively limited recentstudy using parallel ASO-PCR reactions for wild-type and V617F mutant JAK2 toevaluate peripheral blood and bone marrow samples in 11 untreated MPN subjects


Fig. 7.6 Detection of JAK2 by PCR followed by melting curve analysis is accomplished by usingflanking PCR primers that amplify both wild-type and mutated JAK2, with application of a FRETprobe. In the following melting curve analysis, rising temperatures cause the probes to “melt” offof the amplified product to which they’ve hybridized, resulting in loss of fluorescence. The probein this case is 100% homologous to the wild-type sequence, requiring a higher melting temperatureto denature, while the partially mismatched mutated sequence denatures at a slightly lower temper-ature. Detection of wild-type and or mutated JAK2 is achieved by comparing the melting curves ofthe samples to controls. Although this method works well, it only detects sequence changes thatoccur under the probe, and may have limited sensitivity in detecting low levels of JAK2 V617F

showed good correlation between peripheral blood and bone marrow levels [97].Besides the sample size, one limitation of this study was that no low-level sampleswere tested.

Since the V617F mutation accounts for 95% of PV, analysis for other JAK2 muta-tions may remain largely limited to high-volume laboratories. A direct sequencingassay might be adequate to detect most of these mutations as long as they arepresent in high enough amounts (25–30%) to be detected by direct sequencing (e.g.,at diagnosis). Strategies to increase sensitivity such as use of wild-type blockingoligonucleotides in targeted areas may be of benefit. Alternatively, mutation screen-ing methods can be employed. One recent study reports successful screening ofthe entire coding region of JAK2 exons 11–15, incorporating all of the reportedabnormalities, using a non-isotopic RNA cleavage assay [98]. Others have shownhigh sensitivity and specificity in detection of exon 12 and 14 mutations withhigh-resolution melting analysis [99].

JAK2 – Unresolved Issues

Many issues surrounding JAK2 and MPNs remain incompletely resolved and arethe subject of investigation, including whether JAK2 is the initiating event in MPNs,the role that heredity may play in susceptibility to acquiring JAK2 mutations, themechanisms driving pathogenesis in JAK2- and MPL-negative ET and PMF cases,and the role of JAK2 inhibitor therapy [100]. Between 10 and 20 JAK2 inhibitors

238 D.S. Bosler

are in the development pipeline, with five in clinical trials. Although early resultsshow improvement in signs and symptoms such as blood counts, splenomegaly, andconstitutional symptoms, whether any will induce remission or alter disease courseis yet to be determined [82, 100].

MPL

Mutations in MPL have recently been discovered in a small percentage of JAK2-negative ET and PMF cases. Located at chromosome 1p34, MPL encodes the throm-bopoietin receptor, which stimulates global hematopoiesis as well as megakaryocytegrowth and differentiation. When bound to its ligand, MPL acts in a fashion similarto the erythropoietin receptor, with activation of the JAK/STAT pathway and down-stream stimulation of proliferation and survival signals via the MAPK and PI3Kpathways [81]. MPL mutations in MPNs were first reported in 2006, when a groupscreening JAK2 V617F-negative MPNs for mutations reported a single nucleotidesubstitution (G1544T) resulting in a single amino acid change from tryptophan toleucine (W515L) in 4 of 45 evaluated PMF cases [101]. In vitro models showed thatthe mutation conferred thrombopoietin-independent growth and activation of down-stream pathways (Fig. 7.3C), and mice transduced with the mutation-developedlethal MPNs with a short latency [101]. A subsequent larger study evaluated 1182myeloid neoplasms for MPL mutations and found W515L or W515K (lysine substi-tuted for tryptophan) mutations in 4 of 318 ET, 13 of 290 PMF, and 3 of 126 acutemyeloid leukemia cases (the three positives were diagnosed as blast transformationof a pre-existing myeloproliferative neoplasm) and did not detect mutations in anyof 242 PV cases, 206 other myeloid neoplasms, or 64 controls [102]. Six subjectshad W515L and JAK V617F concurrently. More recent studies have found addi-tional somatic mutations, including W515S and S505N, and report incidence ratesof 1–4% of ET and 5–11% of PMF [82]. Compared with MPL wild-type MPN sub-jects, mutated cases are older, more anemic, and more likely to require transfusionsupport [103].

The primary clinical utility of MPL testing at this time is as a marker of clon-ality in cases of suspected MPN when JAK2 analysis and cytogenetics reveal noabnormalities. Since this utility impacts a relatively small number of cases, testingfor MPL mutations is currently relegated to only a few laboratories, with the mostcommon method used being reverse transcription, PCR, and direct sequencing oftargeted areas of MPL.

PDGFR and FGFR1 Abnormalities

This rare group of neoplasms was segregated into its own chapter in the 2008 WHOclassification, with each of the three members in the group defined and classifiedaccording to its associated molecular abnormality. This change in classificationhas largely been driven by the recognition that some of these disorders can be


successfully treated with tyrosine kinase inhibitors such as imatinib, underscor-ing the importance of recognizing them when present despite their rare incidence.Overlapping clinicopathologic features both within this group and with othermyeloid and lymphoid neoplasms combine with varying responses to tyrosinekinase inhibitors to make molecular diagnostic techniques an indispensible part ofthe diagnostic process when the possibility of one of these entities is considered.Table 7.6 summarizes some of the salient features of neoplasms with PDGFRA,PDGFRB, and FGFR1 abnormalities (Table 7.6).

Table 7.6 Key points for PDGFRA, PDGFRB, and FGFR1 abnormalities

Gene PDGFRA PDGFRB FGFR1

Locus 4q12 5q31∼32 8p11Detection FISH Karyotypea KaryotypeCommon

presentationCEL CMML CEL, AML, T-LBL

TKI therapy Sensitive Sensitive Not effective

aMolecular confirmation recommended.

Spectrum of Eosinophilia-Related Disorders

Before further exploring the PDGFR- and FGFR1-associated neoplasms, a briefdiscussion is warranted regarding the broader context of eosinophilia, one oftheir most common manifestations. Distinction of neoplastic processes involvingeosinophils from reactive eosinophilias has historically been challenging [104].Eosinophilia is most often a T-cell-mediated, cytokine-driven process resultingfrom a variety of non-neoplastic conditions including parasitic or other infections,allergy, connective tissue disease, certain pulmonary diseases, and drug reactions[104, 105]. Additionally, reactive eosinophilias can be induced by a variety ofmalignancies, including Hodgkin lymphoma, T-cell lymphomas, lymphoblasticleukemia/lymphomas, and mastocytosis. Less than 1% of eosinophilias are partof a clonal neoplastic process, and these more often occur as part of a variety ofmyeloid neoplasms other than chronic eosinophilic leukemia, including CML orother myeloproliferative neoplasms, myelodysplastic syndromes, and acute myeloidleukemias [20, 105, 106]. Clonal eosinophils can also be seen with lymphoblas-tic lymphoma and systemic mastocytosis. Chronic eosinophilic leukemia (CEL)is a comparatively rare eosinophil-predominant myeloid neoplasm with increasedperipheral blood and bone marrow eosinophils and either evidence of clonality orincreased blasts (<20%). Due to their potential responsiveness to tyrosine kinaseinhibitors, PDGFR- and FGFR1-associated neoplasms presenting with the mor-phologic findings of CEL are specifically excluded from a diagnosis of CEL, nototherwise specified, and are instead classified according to their specific molecu-lar abnormalities [1] as described in more detail below. Finally, hypereosinophilicsyndrome (HES) is a diagnosis of exclusion in the setting of persistent unexplained

240 D.S. Bosler

eosinophilia and evidence of organ involvement. It is a process of incompletelyexplained etiology that, while often morphologically indistinguishable from CEL,has no demonstrable evidence of clonality, and potential underlying etiologiesof reactive eosinophilia have been excluded [1, 106, 107]. Although clonal T-cell receptor gene rearrangements have been demonstrated in some subjects withan HES-like presentation [108], these cases are excluded from the diagnosis ofHES, and generally regarded as aberrant T-cell populations with cytokine-inducedeosinophilia rather than a myeloid neoplasm. Because of the potential for tissuedamage (including potentially life-threatening cardiac pathology) resulting fromeosinophil tissue infiltration and degranulation regardless of etiology, potential cor-rectable etiologies for eosinophilia should be exhaustively explored and excludedbefore a diagnosis of hypereosinophilic syndrome is considered [105].

PDGFRA

Platelet-derived growth factor receptor (PDGFR) alpha is a member of the typeIII family of receptor tyrosine kinases that also includes PDGFR beta and KIT[82]. Its gene, PDGFRA, resides at 4q12. Abnormalities involving this locus resultin constitutive activation of the PDGFRA tyrosine kinase activity and have beendemonstrated in neoplasms with a range of clinical and morphologic features.Although sporadic reports of abnormalities involving the PDGFRA locus had pre-viously been reported in eosinophilic neoplasms, the most common abnormality ofthis locus was first reported by Cools et al. in 2003 [109]. Their strategy was tolook for abnormalities in patients with HES that had responded to imatinib ther-apy, with attention focused on PDGFRA, PDGFRB, and KIT based on previousstudies showing that their tyrosine kinase products are targets of imatinib. Part oftheir analysis included using FISH probes to more closely evaluate a translocationinvolving 4q12 in one subject, revealing a deletion at this locus involving the CHIC2gene. Additional studies showed the presence of a fusion gene, FIP1L1–PDGFRA,which resulted not from the translocation but from an 800 kb interstitial deletionon 4q12. Many subjects without any cytogenetically evident abnormalities also hadthe deletion and fusion, with 9 of 16 evaluated HES subjects ultimately testing pos-itive [109]. Further characterization of this entity has shown male predominanceand a median age of onset in the fifth decade [110, 111]. Splenomegaly and cardiacinvolvement are frequent findings, as are elevated vitamin B12 and serum tryptaselevels [110–112].

The FIP1L1–PDGFRA fusion was also subsequently demonstrated by Pardananiet al. in three of five evaluated subjects with systemic mastocytosis with eosinophiliaand, in marked contrast to the morphologically similar systemic mastocytosis casesthat had the KIT D816V mutation, those with the FIP1L1–PDGFRA fusion achievedsustained complete response with imatinib [113]. A subsequent study by the samegroup screened 89 subjects with eosinophilia for the FIP1L1–PDGFRA fusion andfound that after exclusion of clearly reactive cases (all of which were negative forthe fusion), 14% of primary eosinophilias harbored the fusion [112]. Although 10


of 11 positive cases were categorized by this group as systemic mastocytosis witheosinophilia, 7 of these had originally been diagnosed clinically as HES and werereclassified after elevated serum tryptase levels were seen in some, and carefulbone marrow histologic and immunophenotypic evaluation showed increased num-bers of immunophenotypically aberrant mast cells. The FIP1L1–PDGFRA fusionhas also been described in association with T-lymphoblastic lymphomas and acutemyeloid leukemias with eosinophilia that responded to imatinib [114]. The positiveresponses to imatinib in this clinically and morphologically diverse group of entitieswith features that overlap significantly with diseases that do not respond to imatinibunderscores the relevance of molecular diagnostic testing in these cases.

When present, the FIP1L1–PDGFRA most often occurs as a sole abnormality[110]. Unlike the BCR–ABL1 fusion, the FIP1L1–PDGFRA fusion does not dependon its 5′ component, FIP1L1, for its function [115]. The breakpoints in FIP1L1 sig-nificantly vary compared with the relatively stable breakpoint region in PDGFRA inexon 12 involving the autoinhibitory juxtamembrane region [82, 108]. Disruption ofthis autoinhibitory domain has been shown to be required for constitutive activationand transforming potential [115].

An important point for hematopathologists and molecular diagnosticians is thatthis 800 kb deletion is too small to be detected at the level of conventional cytoge-netic karyotyping. As such, one must remember to order a specific diagnostic testwhen this entity is considered in the differential diagnosis. Although some havesuccessfully used RT-PCR to confirm the presence of the fusion [108, 110], the var-ied breakpoints in FIP1L1 make FISH a practical strategy of detection for clinicaluse. Perhaps the most common FISH probe strategy uses three probes, includingprobes 5′ to the FIP1L1 breakpoints, 3′ to the PDGFRA breakpoints, and within theexpected deleted region [104, 110]. With this strategy, the presence of three signalsindicates lack of deletion, while absence of the middle signal (within the deletedregion) indicates presence of the deletion (Fig. 7.7).

PDGFRB

Myeloid neoplasms associated with PDGFRB abnormalities represent another rareset of neoplasms with a range of clinicopathologic findings unified by the aforemen-tioned molecular abnormality as well as their general responsiveness to tyrosinekinase inhibitors. PDGFR beta, also a receptor tyrosine kinase, is encoded by thePDGFRB gene located at chromosome 5q31∼32. The most common PDGFRB-associated abnormality is the t(5;12) (q31∼33;p12∼13) translocation, which createsa ETV6–PDGFRB fusion that was first described by Golub et al. in 1994 insubjects with chronic myelomonocytic leukemia (CMML) [111, 116]. A hostof other rare translocations involving alternative fusion partners have also beendescribed, all of which have been detectable by conventional cytogenetic karyotyp-ing [111]. Alternative fusion partners with known or likely response to imatinibdescribed to date include WDR48, GPIAP1, TPM3, PDE4DIP, PRKG2, GOLGA4,HIP1, CCDC6, GIT2, NIN, KIAA1509, TP53BP1, NDE1, RABEP1, and SPECC1

242 D.S. Bosler

FIP1L1 PDGFRA

CHIC2

KIT

~800 bp deletionA.

C.B.

Fig. 7.7 (a) The FIP1L1–PDGFRA fusion occurs through deletion of an interstitial DNA segmentof approximately 800 kb. FISH probes placed in regions flanking the genes of interest and withinthe deleted area can be used to detect the fusion via loss of the signal in the deleted region (orangein this case). (b) The image on the left shows a normal signal, with two copies of an intact segmentof the locus of interest on chromosome 4q12 resulting in two signals containing all three probes.The image on the right is from a cell line that contains two copies of the deletion and one normalallele, resulting in two signals containing only aqua and green (with orange absent), and one normalsignal with all three probes

[1, 111, 117]. Regardless of the fusion partner, the fusion with PDGFRB results inconstitutive activation of the tyrosine kinase via ligand-independent dimerization,which has demonstrated transformative properties [118]. In addition to CMML,PDGFRB abnormalities have also been described in cases with the clinicopatho-logic features of atypical CML, juvenile myelomonocytic leukemia, and chroniceosinophilic leukemia [117, 119–121]. As is seen with PDGFRA-associated neo-plasms, mast cells may be increased and have morphologic and immunophenotypicaberrancies [117]. Overall, responses to imatinib in these disorders have beenrapid, complete and relatively durable [116]. One study following 12 subjectswith PDGFRB-associated chronic myeloproliferative disorders treated with ima-tinib reported that all 12 had rapid response and 11 had complete normalizationof peripheral blood counts [122]. Median overall survival since diagnosis was 65months, with 10 of 12 subjects still alive at last follow-up.


Although the PDGFRB abnormalities are detected by conventional cytogenetics,the 2008 WHO Classification recommends that for definitive diagnosis these abnor-malities should be confirmed at the molecular level, because t(5;12) translocationsthat do not contain that ETV6–PDGFRB fusion can be cytogenetically indistin-guishable from those that do [1]. Since there are numerous potential partners, abreakapart FISH probe strategy using probes surrounding PDGFRB would be apotentially practical approach for this confirmation. Multiplex RT-PCR strategiesthat detect most of the common ETV6–PDGFRB fusion have also been demon-strated [123]. Additionally, the dramatic responses to imatinib in PDGFRA- andPDGFRB-associated neoplasms raise the question of whether methods of diseasemonitoring analogous to those used in CML should be implemented. The compar-ative rarity of these neoplasms has prevented a thorough analysis of this questionto date, however, and more detailed studies are needed before a consensus can bereached.

FGFR1

FGFR1-associated hematolymphoid neoplasms are quite rare, with only about50 cases reported [1, 111]. The first cases reported in the 1990s were character-ized as presenting with T-lymphoblastic lymphoma with eosinophilia, subsequentlydeveloping myeloid malignancies such as AML or myeloid sarcoma and showing at(8;13) translocation most often involving 8p11 [124, 125]. Also known as the 8p11myeloproliferative syndrome and 8p11 stem cell leukemia/lymphoma syndrome, theinvolvement of the FGFR1 gene on 8p11 in these neoplasms was first characterizedin 1998 by Xiao et al., who localized the defect to the disruption of FGFR1 intron8 and described a ZNF198-FGFR1 fusion in four patients with t(8;13) (p11;q12)translocation-associated malignancies [126]. This most common fusion involves the5′ end of ZNF198 on 13q12 and the 3′ end of FGFR1 (including exon 9) on 8p11.Although the 3′ end of the FGFR1 gene is invariantly involved, numerous alterna-tive 5′ fusion partners have been described, including CEP110, FGFR1OP1, BCR,TRIM24, MYO18A, HERVK, FGFR1OP2, and CPSF6 [1, 82, 111, 127]. Similar toPDGFRB fusions, the FGFR1 fusions result in constitutive activation of FGFR1’styrosine kinase activity via ligand-independent dimerization and have been shownto induce hematolymphoid neoplasms in murine models that are analogous to thatseen in humans [82, 128–130].

The heterogeneous spectrum of clinicopathologic presentations seen in theFGFR1-associated hematolymphoid neoplasms reflects the presence of an under-lying defect in pluripotent stem cells with the ability to differentiate into myeloidor lymphoid cells [111]. The neoplasms can present as a myeloproliferative neo-plasm such as chronic eosinophilic leukemia or CML-like disease, as acute myeloidleukemia, or as lymphoblastic leukemia/lymphoma (T-cell more often than B-cell)[131]. Acute leukemias can also be of mixed myeloid and lymphoid phenotype.Despite this heterogeneous range of presentations, about 90% have peripheral bloodand/or bone marrow eosinophilia at presentation. In those that present with chronic

244 D.S. Bosler

disease, transformation to one of the acute manifestations usually occurs rapidlywithin 1–2 years of diagnosis [131]. Presentation is often relatively early in life, themedian age at diagnosis is 32 years, and there is a slight male predominance [131].The disease most often behaves aggressively and is refractory to chemotherapeu-tic regimens, with most patients dying of persistent or relapsed disease within 1.5years [125, 131]. Allogeneic stem cell transplant is currently the only chance forcure. Unlike the other members of this group, FGFR1-associated neoplasms are notsensitive to imatinib. A newer tyrosine kinase inhibitor, PKC412, has been shownto inhibit tyrosine kinase activity in transformed cell lines and prolong survival inmurine models [130]. Although these researchers also reported that one human sub-ject with a FGFR1-associated myeloproliferative neoplasm showed improvementin leukocytosis and reduction of lymphadenopathy and splenomegaly when treatedwith PKC412, much more work is needed to more definitively characterize thedrug’s effectiveness in this disease.

The lack of established therapeutic regimens that might require precise quantita-tion for disease monitoring and the lack of known genotype–phenotype correlationswith prognostic relevance make the current role of the molecular diagnosticianrelatively limited for the FGFR1-associated hematolymphoid neoplasms. The lab-oratory’s main role is ensuring that this rare entity is recognized when present.Since all reported abnormalities to date have been detectable by cytogenetic kary-otyping, this method should be adequate means of detection in most cases. In therare instance where clinical suspicion is very high despite negative cytogenetics,or where material for cytogenetic analysis is not available, an alternative strategysuch as breakapart FISH at the 8p11 locus may be of benefit if demonstration of theabnormality is required.

Mast Cell Disease

Mastocytosis is a rare and clinically heterogeneous disease characterized by a neo-plastic proliferation of abnormal, clonal mast cells. Manifestations can be eitherprimarily cutaneous or systemic, with frequent sites of systemic involvement includ-ing the gastrointestinal tract, liver, spleen, bone marrow, and lymph nodes [1].Affected patients have an increased risk of associated myeloid neoplasms. Thediagnosis and classification process of mastocytosis contains complexities that arebeyond the scope of this text, and interested readers are referred to the 2008 WHOClassification and/or one of many published reviews on the topic [1, 132–134]. Insummary, the diagnosis relies on some combination of the clinical findings, the pres-ence of morphologically and immunophenotypically abnormal clusters of mast cellsat affected sites, elevated serum tryptase levels, and the presence of the KIT D816Vmutation.

Like the other myeloproliferative neoplasms, the central molecular abnormal-ity in mastocytosis causes constitutive activation of a tyrosine kinase. Mastocytosisinvolves KIT, a receptor tyrosine kinase in the same family as PDGFRA andPDGFRB that binds stem cell factor as its ligand [135]. The gene for KIT resides in


close proximity to PDGFRA at 4q12 and contains 21 exons. Like the other tyro-sine kinases discussed in this chapter, the downstream effects of KIT activationinclude cell proliferation, maturation, differentiation, and prolonged survival andare mediated through several familiar pathways including PI3K, MAPK, JAK/STAT,and protein kinase C. KIT is normally expressed in a variety of cell types, includ-ing mast cells, hematopoietic stem cells, germ cells, melanocytes, and interstitialcells of Cajal present within the gastrointestinal stroma. Interestingly, this distribu-tion reflects the types of tumors in which KIT mutations arise, including not onlymastocytosis but also acute myeloid leukemias and lymphomas, germ cell tumors,melanomas, and gastrointestinal stromal tumors [135].

A single nucleotide substitution in exon 17 of KIT (A7176T) resulting in a singleamino acid change from aspartic acid to valine (D816V) is present in over 90% ofadult systemic mastocytosis cases [136]. This most common mutation occurs withinthe activation loop of the TK2 domain and results in destabilization of the inactiveconformation of KIT and ligand-independent activation [135, 137]. The mutation ispresent in mast cells, with variable expression in other hematopoietic cells [136]. Asignificant subset of D816V-negative cases has other KIT mutations (Fig. 7.8), whichtend to cluster within the TK2 activation loop near D816 (amino acids 815–839) andwithin the extracellular cleavage, transmembrane, and juxtamembrane regulatorydomains (amino acids 419 and 509–560) [135]. The importance of distinguishingbetween these mutation sites is that, while the D816V and other activation loopmutations are by-and-large resistant to imatinib therapy, some mutations withinthe transmembrane and juxtamembrane domains show sensitivity to imatinib [82].Some newer generation tyrosine kinase inhibitors such as dasatinib have shown lim-ited activity against systemic mastocytosis with D816V mutations, and PKC412 has

Extracellular

Tyrosine kinase

Transmembrane

Juxtamembrane

D816V

PI3K MAPK

TK1

TK2

KIT Domains:

PKC

MastocytosisHotspots

Imatinib

sensitive

resistant

Fig. 7.8 This schematic representation of the KIT receptor highlights the various function domainsand hot spots for mutations within KIT that are associated with mast cell disease. The most commonmutation in mast cell disease, KIT D816V, occurs in the TK2 domain. As is true with D816V,mutations in this region tend to be resistant to imatinib, while rarer mutations in the transmembraneand juxtamembrane domains may show some imatinib sensitivity

246 D.S. Bosler

also shown activity against D816V cases in very limited clinical experience [82,138, 139].

KIT testing in mastocytosis has potential clinical relevance for both diagnosisand choice of therapy. Detection of D816V is one of the diagnostic criteria formastocytosis and can help to confirm the diagnosis in otherwise challenging cases.Additionally, knowledge of the D816V mutation’s presence in a given case willlikely dissuade the use of imatinib given the established lack of benefit. By contrast,detecting one of the rare mutations with potential sensitivity to imatinib may be ofsubstantial benefit in individual cases, although the rarity of these cases likely makesroutine evaluation for these mutations impractical except in the highest volumereference laboratories.

Although detection of a single nucleotide substitution such as the KIT D816Vmutation should be easily achieved by a variety of methods, the relative paucity ofcells harboring the mutation, particularly in peripheral blood, can pose a challengefor detecting D816V in mastocytosis. Peripheral blood samples may preferentiallybe positive for mastocytosis cases in which the underlying defect originates in amultipotent stem cell and is therefore present in a variety of hematopoietic cellsrather than restricted solely to mast cells [137]. Higher detection rates are there-fore typical when testing bone marrow or lesional tissue compared with peripheralblood. Enrichment for mast cells, such as with flow cytometric sorting based onCD25 expression, often has a higher diagnostic yield than testing of unsorted spec-imens. Additionally, analysis of mRNA via RT-PCR helps to compensate for thepaucity of cells carrying the mutation, since these cells express KIT at a high level[137]. Numerous methods have been employed in the detection of D816V for clin-ical purposes, including RT-PCR followed by RFLP analysis or direct sequencing,allele-specific PCR, and peptide nucleic acid (PNA)-mediated polymerase chainreaction clamping of the wild-type allele combined with oligonucleotide hybridiza-tion probes [137, 140]. Allele-specific PCR offers the advantage of sensitivity thatcan overcome the small percentage of tumor cells. For example, a sensitivity of 1%has been reported using DNA from formalin-fixed, paraffin-embedded tissue usingallele-specific PCR combined with a wild-type blocking oligonucleotide [141]. Thedisadvantage of allele-specific PCR is that, since it detects only the mutations forwhich it has been designed, alternative methods would be required to detect the rarenon-D816V mutations.

One final relevant point for mast cell disease is that, since abnormal mast cellsand increased eosinophils can occur in both of the PDGFR-associated myeloid neo-plasms and mastocytosis, evaluation for all of these entities should be consideredwhen this overlapping presentation is encountered. Although rarely encountered,this distinction has particular clinical relevance due to striking differences inresponse to imatinib therapy.

References

1. Swerdlow S, Campo E, Harris N, et al. WHO Classification of Tumours of Haematopoieticand Lymphoid Tissues. 4th ed. Lyon: IARC; 2008.


2. Dameshek W. Some speculations on the myeloproliferative syndromes. Blood. 1951;6:372–375.

3. Bennett J. Case of hypertrophy of the spleen and liver, in which death took place fromsuppuration of the blood. Edinburgh Med Surg J. 1845;64:413–423.

4. Nowell PC, Hungerford DA. Chromosome studies on normal and leukemic human leuko-cytes. J Natl Cancer Inst. 1960;25:85–109.

5. Rowley JD. Letter: a new consistent chromosomal abnormality in chronic myeloge-nous leukaemia identified by quinacrine fluorescence and Giemsa staining. Nature.1973;243:290–293.

6. Pane F, Intrieri M, Quintarelli C, Izzo B, Muccioli GC, Salvatore F. BCR/ABL genesand leukemic phenotype: from molecular mechanisms to clinical correlations. Oncogene.2002;21:8652–8667.

7. Heisterkamp N, Groffen J, Stephenson JR, et al. Chromosomal localization of humancellular homologues of two viral oncogenes. Nature. 1982;299:747–749.

8. Groffen J, Stephenson JR, Heisterkamp N, de KA, Bartram CR, Grosveld G. Philadelphiachromosomal breakpoints are clustered within a limited region, bcr, on chromosome 22.Cell. 1984;36:93–99.

9. Heisterkamp N, Stephenson JR, Groffen J, et al. Localization of the c-ab1 oncogene adjacentto a translocation break point in chronic myelocytic leukaemia. Nature. 1983;306:239–242.

10. Bartram CR, de KA, Hagemeijer A, et al. Translocation of c-ab1 oncogene correlateswith the presence of a Philadelphia chromosome in chronic myelocytic leukaemia. Nature.1983;306:277–280.

11. Shtivelman E, Lifshitz B, Gale RP, Canaani E. Fused transcript of abl and bcr genes inchronic myelogenous leukaemia. Nature. 1985;315:550–554.

12. Daley GQ, Van Etten RA, Baltimore D. Induction of chronic myelogenous leukemia in miceby the P210bcr/abl gene of the Philadelphia chromosome. Science. 1990;247:824–830.

13. Druker BJ, Lydon NB. Lessons learned from the development of an abl tyrosine kinaseinhibitor for chronic myelogenous leukemia. J Clin Invest. 2000;105:3–7.

14. Piccaluga PP, Rondoni M, Paolini S, Rosti G, Martinelli G, Baccarani M. Imatinib mesylatein the treatment of hematologic malignancies. Expert Opin Biol Ther. 2007;7:1597–1611.

15. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Thun MJ. Cancer statistics, 2009. CA Cancer JClin. 2009;59:225–249.

16. Redaelli A, Bell C, Casagrande J, et al. Clinical and epidemiologic burden of chronicmyelogenous leukemia. Expert Rev Anticancer Ther. 2004;4:85–96.

17. Cortes J. Natural history and staging of chronic myelogenous leukemia. Hematol Oncol ClinNorth Am. 2004;18:569–84, viii.

18. Savage DG, Szydlo RM, Goldman JM. Clinical features at diagnosis in 430 patients withchronic myeloid leukaemia seen at a referral centre over a 16-year period. Br J Haematol.1997;96:111–116.

19. Cotta CV, Bueso-Ramos CE. New insights into the pathobiology and treatment of chronicmyelogenous leukemia. Ann Diagn Pathol. 2007;11:68–78.

20. Spiers AS, Bain BJ, Turner JE. The peripheral blood in chronic granulocytic leukaemia.Study of 50 untreated Philadelphia-positive cases. Scand J Haematol. 1977;18:25–38.

21. Cortes JE, Talpaz M, O’Brien S, et al. Staging of chronic myeloid leukemia in the imatinibera: an evaluation of the World Health Organization proposal. Cancer. 2006;106:1306–1315.

22. Lorand-Metze I, Vassallo J, Souza CA. Histological and cytological heterogeneity ofbone marrow in Philadelphia-positive chronic myelogenous leukaemia at diagnosis. Br JHaematol. 1987;67:45–49.

23. Muehleck SD, McKenna RW, Arthur DC, Parkin JL, Brunning RD. Transformation ofchronic myelogenous leukemia: clinical, morphologic, and cytogenetic features. Am J ClinPathol. 1984;82:1–14.

24. Melo JV. The diversity of BCR-ABL fusion proteins and their relationship to leukemiaphenotype. Blood. 1996;88:2375–2384.

248 D.S. Bosler

25. Kurzrock R, Gutterman JU, Talpaz M. The molecular genetics of Philadelphia chromosome-positive leukemias. N Engl J Med. 1988;319:990–998.

26. Verma D, Kantarjian HM, Jones D, et al. Chronic myeloid leukemia (CML) with P190BCR-ABL: analysis of characteristics, outcomes, and prognostic significance. Blood.2009;114:2232–2235.

27. Ren R. Mechanisms of BCR-ABL in the pathogenesis of chronic myelogenous leukaemia.Nat Rev Cancer. 2005;5:172–183.

28. Shteper PJ, Ben-Yehuda D. Molecular evolution of chronic myeloid leukaemia. SeminCancer Biol. 2001;11:313–323.

29. Ou J, Vergilio JA, Bagg A. Molecular diagnosis and monitoring in the clinical managementof patients with chronic myelogenous leukemia treated with tyrosine kinase inhibitors. AmJ Hematol. 2008;83:296–302.

30. Dessars B, El HH, Lambert F, Kentos A, Heimann P. Rational use of the EAC real-timequantitative PCR protocol in chronic myelogenous leukemia: report of three false-negativecases at diagnosis. Leukemia. 2006;20:886–888.

31. Landstrom AP, Ketterling RP, Knudson RA, Tefferi A. Utility of peripheral blood dualcolor, double fusion fluorescent in situ hybridization for BCR/ABL fusion to assesscytogenetic remission status in chronic myeloid leukemia. Leuk Lymphoma. 2006;47:2055–2061.

32. Costa D, Espinet B, Queralt R, et al. Chimeric BCR/ABL gene detected by fluorescencein situ hybridization in three new cases of Philadelphia chromosome-negative chronicmyelocytic leukemia. Cancer Genet Cytogenet. 2003;141:114–119.

33. Chasseriau J, Rivet J, Bilan F, et al. Characterization of the different BCR-ABL transcriptswith a single multiplex RT-PCR. J Mol Diagn. 2004;6:343–347.

34. Weerkamp F, Dekking E, Ng YY, et al. Flow cytometric immunobead assay for the detectionof BCR-ABL fusion proteins in leukemia patients. Leukemia. 2009;23:1106–1117.

35. Salesse S, Verfaillie CM. BCR/ABL: from molecular mechanisms of leukemia induction totreatment of chronic myelogenous leukemia. Oncogene. 2002;21:8547–8559.

36. Mark HF, Sokolic RA, Mark Y. Conventional cytogenetics and FISH in the detection ofBCR/ABL fusion in chronic myeloid leukemia (CML). Exp Mol Pathol. 2006;81:1–7.

37. Hughes TP, Kaeda J, Branford S, et al. Frequency of major molecular responses to imatinibor interferon alfa plus cytarabine in newly diagnosed chronic myeloid leukemia. N Engl JMed. 2003;349:1423–1432.

38. Druker BJ, Guilhot F, O’Brien SG, et al. Five-year follow-up of patients receiving imatinibfor chronic myeloid leukemia. N Engl J Med. 2006;355:2408–2417.

39. Press RD, Love Z, Tronnes AA, et al. BCR-ABL mRNA levels at and after the time of acomplete cytogenetic response (CCR) predict the duration of CCR in imatinib mesylate-treated patients with CML. Blood. 2006;107:4250–4256.

40. Wang L, Pearson K, Ferguson JE, Clark RE. The early molecular response to imatinibpredicts cytogenetic and clinical outcome in chronic myeloid leukaemia. Br J Haematol.2003;120:990–999.

41. Marin D, Kaeda J, Szydlo R, et al. Monitoring patients in complete cytogenetic remissionafter treatment of CML in chronic phase with imatinib: patterns of residual leukaemia andprognostic factors for cytogenetic relapse. Leukemia. 2005;19:507–512.

42. Baccarani M, Saglio G, Goldman J, et al. Evolving concepts in the management of chronicmyeloid leukemia: recommendations from an expert panel on behalf of the EuropeanLeukemiaNet. Blood. 2006;108:1809–1820.

43. Hughes T, Deininger M, Hochhaus A, et al. Monitoring CML patients responding to treat-ment with tyrosine kinase inhibitors: review and recommendations for harmonizing currentmethodology for detecting BCR-ABL transcripts and kinase domain mutations and forexpressing results. Blood. 2006;108:28–37.

44. Gabert J, Beillard E, van dV, V, et al. Standardization and quality control studies of ‘real-time’ quantitative reverse transcriptase polymerase chain reaction of fusion gene transcripts


for residual disease detection in leukemia – a Europe Against Cancer Program. Leukemia.2003;17:2318–2357.

45. Beillard E, Pallisgaard N, van dV V, et al. Evaluation of candidate control genes for diagnosisand residual disease detection in leukemic patients using ‘real-time’ quantitative reverse-transcriptase polymerase chain reaction (RQ-PCR) – a Europe against cancer program.Leukemia. 2003;17:2474–2486.

46. van dV V, Boeckx N, Gonzalez M, et al. Differential stability of control gene and fusiongene transcripts over time may hamper accurate quantification of minimal residual disease –a study within the Europe Against Cancer Program. Leukemia. 2004;18:884–886.

47. Zhang T, Grenier S, Nwachukwu B, Wei C, Lipton JH, Kamel-Reid S. Inter-laboratory com-parison of chronic myeloid leukemia minimal residual disease monitoring: summary andrecommendations. J Mol Diagn. 2007;9:421–430.

48. Cross NC, Hughes TP, Hochhaus A, Goldman JM. International standardisation of quantita-tive real-time RT-PCR for BCR-ABL. Leuk Res. 2008;32:505–506.

49. Silvy M, Mancini J, Thirion X, Sigaux F, Gabert J. Evaluation of real-time quantitativePCR machines for the monitoring of fusion gene transcripts using the Europe against cancerprotocol. Leukemia. 2005;19:305–307.

50. Iacobucci I, Galletti L, Amabile M, Soverini S, Baccarani M, Martinelli G. MonitoringBCR-ABL transcript levels by real-time quantitative polymerase chain reaction: a linearregression equation to convert from BCR-ABL/B2M ratio to estimated BCR-ABL/ABLratio. Haematologica. 2007;92:429–430.

51. Branford S, Fletcher L, Cross NC, et al. Desirable performance characteristics for BCR-ABL measurement on an international reporting scale to allow consistent interpretation ofindividual patient response and comparison of response rates between clinical trials. Blood.2008;112:3330–3338.

52. Saldanha J, Silvy M, Beaufils N, et al. Characterization of a reference material for BCR-ABL(M-BCR) mRNA quantitation by real-time amplification assays: towards new standards forgene expression measurements. Leukemia. 2007;21:1481–1487.

53. Schindler T, Bornmann W, Pellicena P, Miller WT, Clarkson B, Kuriyan J. Structural mech-anism for STI-571 inhibition of abelson tyrosine kinase. Science. 2000;289:1938–1942.

54. Jones D, Kamel-Reid S, Bahler D, et al. Laboratory practice guidelines for detecting andreporting BCR-ABL drug resistance mutations in chronic myelogenous leukemia and acutelymphoblastic leukemia: a report of the Association for Molecular Pathology. J Mol Diagn.2009;11:4–11.

55. Melo JV, Chuah C. Resistance to imatinib mesylate in chronic myeloid leukaemia. CancerLett. 2007;249:121–132.

56. Deininger MW. Optimizing therapy of chronic myeloid leukemia. Exp Hematol.2007;35:144–154.

57. Jabbour E, Cortes J, Kantarjian H. Novel tyrosine kinase inhibitors in chronic myelogenousleukemia. Curr Opin Oncol. 2006;18:578–583.

58. Soverini S, Colarossi S, Gnani A, et al. Contribution of ABL kinase domain muta-tions to imatinib resistance in different subsets of Philadelphia-positive patients: by theGIMEMA Working Party on Chronic Myeloid Leukemia. Clin Cancer Res. 2006;12:7374–7379.

59. Dai Y, Rahmani M, Corey SJ, Dent P, Grant S. A Bcr/Abl-independent, Lyn-dependent formof imatinib mesylate (STI-571) resistance is associated with altered expression of Bcl-2.J Biol Chem. 2004;279:34227–34239.

60. O’Hare T, Eide CA, Deininger MW. Persistent LYN signaling in imatinib-resistant,BCR-ABL-independent chronic myelogenous leukemia. J Natl Cancer Inst. 2008;100:908–909.

61. Cortes J, Jabbour E, Kantarjian H, et al. Dynamics of BCR-ABL kinase domain mutations inchronic myeloid leukemia after sequential treatment with multiple tyrosine kinase inhibitors.Blood. 2007;110:4005–4011.

250 D.S. Bosler

62. Bixby D, Talpaz M. Mechanisms of resistance to tyrosine kinase inhibitors in chronicmyeloid leukemia and recent therapeutic strategies to overcome resistance. Hematology.2009;461–476.

63. Willis SG, Lange T, Demehri S, et al. High-sensitivity detection of BCR-ABL kinase domainmutations in imatinib-naive patients: correlation with clonal cytogenetic evolution but notresponse to therapy. Blood. 2005;106:2128–2137.

64. Khorashad JS, Anand M, Marin D, et al. The presence of a BCR-ABL mutant allele in CMLdoes not always explain clinical resistance to imatinib. Leukemia. 2006;20:658–663.

65. Delhommeau F, Pisani DF, James C, Casadevall N, Constantinescu S, Vainchenker W.Oncogenic mechanisms in myeloproliferative disorders. Cell Mol Life Sci. 2006;63:2939–2953.

66. Girodon F, Bonicelli G, Schaeffer C, et al. Significant increase in the apparent incidenceof essential thrombocythemia related to new WHO diagnostic criteria: a population-basedstudy. Haematologica. 2009;94:865–869.

67. Anastasi J. The myeloproliferative neoplasms: insights into molecular pathogenesis andchanges in WHO classification and criteria for diagnosis. Hematol Oncol Clin North Am.2009;23:693–708.

68. Austin SK, Lambert JR. The JAK2 V617F mutation and thrombosis. Br J Haematol.2008;143:307–320.

69. Kreft A, Buche G, Ghalibafian M, Buhr T, Fischer T, Kirkpatrick CJ. The incidenceof myelofibrosis in essential thrombocythaemia, polycythaemia vera and chronic idio-pathic myelofibrosis: a retrospective evaluation of sequential bone marrow biopsies. ActaHaematol. 2005;113:137–143.

70. Thiele J, Kvasnicka HM. Myelofibrotic transformation in essential thrombocythemia.Haematologica. 2009;94:431–433.

71. Wilkins BS, Erber WN, Bareford D, et al. Bone marrow pathology in essential throm-bocythemia: interobserver reliability and utility for identifying disease subtypes. Blood.2008;111:60–70.

72. Tefferi A. The diagnosis of polycythemia vera: new tests and old dictums. Best Pract ResClin Haematol. 2006;19:455–469.

73. Sanchez S, Ewton A. Essential thrombocythemia: a review of diagnostic and pathologicfeatures. Arch Pathol Lab Med. 2006;130:1144–1150.

74. Baxter EJ, Scott LM, Campbell PJ, et al. Acquired mutation of the tyrosine kinase JAK2 inhuman myeloproliferative disorders. Lancet. 2005;365:1054–1061.

75. James C, Ugo V, Le Couedic JP, et al. A unique clonal JAK2 mutation leading to constitutivesignalling causes polycythaemia vera. Nature. 2005;434:1144–1148.

76. Levine RL, Wadleigh M, Cools J, et al. Activating mutation in the tyrosine kinase JAK2 inpolycythemia vera, essential thrombocythemia, and myeloid metaplasia with myelofibrosis.Cancer Cell. 2005;7:387–397.

77. Kralovics R, Passamonti F, Buser AS, et al. A gain-of-function mutation of JAK2 inmyeloproliferative disorders. N Engl J Med. 2005;352:1779–1790.

78. Steensma DP, Dewald GW, Lasho TL, et al. The JAK2 V617F activating tyrosinekinase mutation is an infrequent event in both “atypical” myeloproliferative disorders andmyelodysplastic syndromes. Blood. 2005;106:1207–1209.

79. Ceesay MM, Lea NC, Ingram W, et al. The JAK2 V617F mutation is rare in RARS butcommon in RARS-T. Leukemia. 2006;20:2060–2061.

80. Scott LM, Campbell PJ, Baxter EJ, et al. The V617F JAK2 mutation is uncommon in can-cers and in myeloid malignancies other than the classic myeloproliferative disorders. Blood.2005;106:2920–2921.

81. Levine RL, Pardanani A, Tefferi A, Gilliland DG. Role of JAK2 in the pathogenesis andtherapy of myeloproliferative disorders. Nat Rev Cancer. 2007;7:673–683.

82. Tefferi A. Molecular drug targets in myeloproliferative neoplasms: mutant ABL1, JAK2,MPL, KIT, PDGFRA, PDGFRB and FGFR1. J Cell Mol Med. 2009;13:215–237.


83. Zhao R, Xing S, Li Z, et al. Identification of an acquired JAK2 mutation in polycythemiavera. J Biol Chem. 2005;280:22788–22792.

84. Lu X, Levine R, Tong W, et al. Expression of a homodimeric type I cytokine recep-tor is required for JAK2V617F-mediated transformation. Proc Natl Acad Sci U S A.2005;102:18962–18967.

85. Scott LM, Tong W, Levine RL, et al. JAK2 exon 12 mutations in polycythemia vera andidiopathic erythrocytosis. N Engl J Med. 2007;356:459–468.

86. Pietra D, Li S, Brisci A, et al. Somatic mutations of JAK2 exon 12 in patients with JAK2(V617F)-negative myeloproliferative disorders. Blood. 2008;111:1686–1689.

87. Schnittger S, Bacher U, Haferlach C, et al. Detection of JAK2 exon 12 mutationsin 15 patients with JAK2V617F negative polycythemia vera. Haematologica. 2009;94:414–418.

88. Pardanani A, Lasho TL, Finke C, Hanson CA, Tefferi A. Prevalence and clinicopatho-logic correlates of JAK2 exon 12 mutations in JAK2V617F-negative polycythemia vera.Leukemia. 2007;21:1960–1963.

89. Ma W, Kantarjian H, Zhang X, et al. Mutation profile of JAK2 transcripts in patients withchronic myeloproliferative neoplasias. J Mol Diagn. 2009;11:49–53.

90. Vannucchi AM, Antonioli E, Guglielmelli P, et al. Clinical profile of homozygous JAK2617 V>F mutation in patients with polycythemia vera or essential thrombocythemia. Blood.2007;110:840–846.

91. Kittur J, Knudson RA, Lasho TL, et al. Clinical correlates of JAK2V617F allele burden inessential thrombocythemia. Cancer. 2007;109:2279–2284.

92. Tefferi A, Strand JJ, Lasho TL, et al. Bone marrow JAK2V617F allele burden and clinicalcorrelates in polycythemia vera. Leukemia. 2007;21:2074–2075.

93. Tefferi A, Lasho TL, Huang J, et al. Low JAK2V617F allele burden in primary myelofibro-sis, compared to either a higher allele burden or unmutated status, is associated with inferioroverall and leukemia-free survival. Leukemia. 2008;22:756–761.

94. Vannucchi AM, Antonioli E, Guglielmelli P, Pardanani A, Tefferi A. Clinical corre-lates of JAK2V617F presence or allele burden in myeloproliferative neoplasms: a criticalreappraisal. Leukemia. 2008;22:1299–1307.

95. Guglielmelli P, Barosi G, Specchia G, et al. Identification of patients with poorer sur-vival in primary myelofibrosis based on the burden of JAK2V617F mutated allele. Blood.2009;114:1477–1483.

96. Cankovic M, Whiteley L, Hawley RC, Zarbo RJ, Chitale D. Clinical performance ofJAK2 V617F mutation detection assays in a molecular diagnostics laboratory: evaluationof screening and quantitation methods. Am J Clin Pathol. 2009;132:713–721.

97. Larsen TS, Pallisgaard N, Moller MB, Hasselbalch HC. Quantitative assessment of the JAK2V617F allele burden: equivalent levels in peripheral blood and bone marrow. Leukemia.2008;22:194–195.

98. Kambas K, Mitroulis I, Kourtzelis I, Chrysanthopoulou A, Speletas M, Ritis K. Fast andreliable mutation detection of the complete exon 11–15 JAK2 coding region using non-isotopic RNase cleavage assay (NIRCA). Eur J Haematol. 2009;83:215–219.

99. Rapado I, Grande S, Albizua E, et al. High resolution melting analysis for JAK2 Exon 14and Exon 12 mutations: a diagnostic tool for myeloproliferative neoplasms. J Mol Diagn.2009;11:155–161.

100. Goldman JM, Green AR, Holyoake T, et al. Chronic myeloproliferative diseases with andwithout the Ph chromosome: some unresolved issues. Leukemia. 2009;23:1708–1715.

101. Pikman Y, Lee BH, Mercher T, et al. MPLW515L is a novel somatic activating mutation inmyelofibrosis with myeloid metaplasia. PLoS Med. 2006;3:e270.

102. Pardanani AD, Levine RL, Lasho T, et al. MPL515 mutations in myeloproliferative and othermyeloid disorders: a study of 1182 patients. Blood. 2006;108:3472–3476.

103. Guglielmelli P, Pancrazzi A, Bergamaschi G, et al. Anaemia characterises patients withmyelofibrosis harbouring Mpl mutation. Br J Haematol. 2007;137:244–247.

252 D.S. Bosler

104. Gotlib J, Cross NC, Gilliland DG. Eosinophilic disorders: molecular pathogenesis,new classification, and modern therapy. Best Pract Res Clin Haematol. 2006;19:535–569.

105. Brito-Babapulle F. The eosinophilias, including the idiopathic hypereosinophilic syndrome.Br J Haematol. 2003;121:203–223.

106. Bain BJ. Cytogenetic and molecular genetic aspects of eosinophilic leukaemias. Br JHaematol. 2003;122:173–179.

107. Chusid MJ, Dale DC, West BC, Wolff SM. The hypereosinophilic syndrome: analysis offourteen cases with review of the literature. Medicine (Baltimore). 1975;54:1–27.

108. Roche-Lestienne C, Lepers S, Soenen-Cornu V, et al. Molecular characterization of the idio-pathic hypereosinophilic syndrome (HES) in 35 French patients with normal conventionalcytogenetics. Leukemia. 2005;19:792–798.

109. Cools J, DeAngelo DJ, Gotlib J, et al. A tyrosine kinase created by fusion of the PDGFRAand FIP1L1 genes as a therapeutic target of imatinib in idiopathic hypereosinophilicsyndrome. N Engl J Med. 2003;348:1201–1214.

110. Vandenberghe P, Wlodarska I, Michaux L, et al. Clinical and molecular features of FIP1L1-PDFGRA (+) chronic eosinophilic leukemias. Leukemia. 2004;18:734–742.

111. Bain BJ, Fletcher SH. Chronic eosinophilic leukemias and the myeloproliferative vari-ant of the hypereosinophilic syndrome. Immunol Allergy Clin North Am. 2007;27:377–388.

112. Pardanani A, Brockman SR, Paternoster SF, et al. FIP1L1-PDGFRA fusion: prevalence andclinicopathologic correlates in 89 consecutive patients with moderate to severe eosinophilia.Blood. 2004;104:3038–3045.

113. Pardanani A, Ketterling RP, Brockman SR, et al. CHIC2 deletion, a surrogate for FIP1L1-PDGFRA fusion, occurs in systemic mastocytosis associated with eosinophilia and predictsresponse to imatinib mesylate therapy. Blood. 2003;102:3093–3096.

114. Metzgeroth G, Walz C, Score J, et al. Recurrent finding of the FIP1L1-PDGFRA fusion genein eosinophilia-associated acute myeloid leukemia and lymphoblastic T-cell lymphoma.Leukemia. 2007;21:1183–1188.

115. Stover EH, Chen J, Folens C, et al. Activation of FIP1L1-PDGFRalpha requires disruptionof the juxtamembrane domain of PDGFRalpha and is FIP1L1-independent. Proc Natl AcadSci U S A. 2006;103:8078–8083.

116. Golub TR, Barker GF, Lovett M, Gilliland DG. Fusion of PDGF receptor beta to a novel ets-like gene, tel, in chronic myelomonocytic leukemia with t(5;12) chromosomal translocation.Cell. 1994;77:307–316.

117. Walz C, Metzgeroth G, Haferlach C, et al. Characterization of three new imatinib-responsivefusion genes in chronic myeloproliferative disorders generated by disruption of the platelet-derived growth factor receptor beta gene. Haematologica. 2007;92:163–169.

118. Carroll M, Tomasson MH, Barker GF, Golub TR, Gilliland DG. The TEL/platelet-derivedgrowth factor beta receptor (PDGF beta R) fusion in chronic myelomonocytic leukemiais a transforming protein that self-associates and activates PDGF beta R kinase-dependentsignaling pathways. Proc Natl Acad Sci U S A. 1996;93:14845–14850.

119. Wittman B, Horan J, Baxter J, et al. A 2-year-old with atypical CML with a t(5;12)(q33;p13)treated successfully with imatinib mesylate. Leuk Res. 2004;28 (Suppl 1):S65–S69.

120. Morerio C, Acquila M, Rosanda C, et al. HCMOGT-1 is a novel fusion partner to PDGFRBin juvenile myelomonocytic leukemia with t(5;17)(q33;p11.2). Cancer Res. 2004;64:2649–2651.

121. Steer EJ, Cross NC. Myeloproliferative disorders with translocations of chromosome 5q31-35: role of the platelet-derived growth factor receptor Beta. Acta Haematol. 2002;107:113–122.

122. David M, Cross NC, Burgstaller S, et al. Durable responses to imatinib in patientswith PDGFRB fusion gene-positive and BCR-ABL-negative chronic myeloproliferativedisorders. Blood. 2007;109:61–64.


123. Curtis CE, Grand FH, Waghorn K, Sahoo TP, George J, Cross NC. A novel ETV6-PDGFRBfusion transcript missed by standard screening in a patient with an imatinib responsivechronic myeloproliferative disease. Leukemia. 2007;21:1839–1841.

124. Abruzzo LV, Jaffe ES, Cotelingam JD, Whang-Peng J, Del DV, Jr., Medeiros LJ. T-celllymphoblastic lymphoma with eosinophilia associated with subsequent myeloid malignancy.Am J Surg Pathol. 1992;16:236–245.

125. Inhorn RC, Aster JC, Roach SA, et al. A syndrome of lymphoblastic lymphoma,eosinophilia, and myeloid hyperplasia/malignancy associated with t(8;13)(p11;q11):description of a distinctive clinicopathologic entity. Blood. 1995;85:1881–1887.

126. Xiao S, Nalabolu SR, Aster JC, et al. FGFR1 is fused with a novel zinc-finger gene, ZNF198,in the t(8;13) leukaemia/lymphoma syndrome. Nat Genet. 1998;18:84–87.

127. Hidalgo-Curtis C, Chase A, Drachenberg M, et al. The t(1;9)(p34;q34) and t(8;12)(p11;q15)fuse pre-mRNA processing proteins SFPQ (PSF) and CPSF6 to ABL and FGFR1. GenesChromosomes Cancer. 2008;47:379–385.

128. Roumiantsev S, Krause DS, Neumann CA, et al. Distinct stem cell myeloproliferative/T lym-phoma syndromes induced by ZNF198-FGFR1 and BCR-FGFR1 fusion genes from 8p11translocations. Cancer Cell. 2004;5:287–298.

129. Guasch G, Delaval B, Arnoulet C, et al. FOP-FGFR1 tyrosine kinase, the product of a t(6;8)translocation, induces a fatal myeloproliferative disease in mice. Blood. 2004;103:309–312.

130. Chen J, DeAngelo DJ, Kutok JL, et al. PKC412 inhibits the zinc finger 198-fibroblastgrowth factor receptor 1 fusion tyrosine kinase and is active in treatment of stem cellmyeloproliferative disorder. Proc Natl Acad Sci U S A. 2004;101:14479–14484.

131. Macdonald D, Reiter A, Cross NC. The 8p11 myeloproliferative syndrome: a distinct clinicalentity caused by constitutive activation of FGFR1. Acta Haematol. 2002;107:101–107.

132. Horny HP, Sotlar K, Valent P. Mastocytosis: state of the art. Pathobiology. 2007;74:121–132.133. Pardanani A, Akin C, Valent P. Pathogenesis, clinical features, and treatment advances in

mastocytosis. Best Pract Res Clin Haematol. 2006;19:595–615.134. Hungness SI, Akin C. Mastocytosis: advances in diagnosis and treatment. Curr Allergy

Asthma Rep. 2007;7:248–254.135. Orfao A, Garcia-Montero AC, Sanchez L, Escribano L. Recent advances in the understand-

ing of mastocytosis: the role of KIT mutations. Br J Haematol. 2007;138:12–30.136. Garcia-Montero AC, Jara-Acevedo M, Teodosio C, et al. KIT mutation in mast cells and

other bone marrow hematopoietic cell lineages in systemic mast cell disorders: a prospectivestudy of the Spanish Network on Mastocytosis (REMA) in a series of 113 patients. Blood.2006;108:2366–2372.

137. Akin C. Molecular diagnosis of mast cell disorders: a paper from the 2005 WilliamBeaumont Hospital Symposium on Molecular Pathology. J Mol Diagn. 2006;8:412–419.

138. Gleixner KV, Mayerhofer M, Sonneck K, et al. Synergistic growth-inhibitory effects of twotyrosine kinase inhibitors, dasatinib and PKC412, on neoplastic mast cells expressing theD816V-mutated oncogenic variant of KIT. Haematologica. 2007;92:1451–1459.

139. Gotlib J, Berube C, Growney JD, et al. Activity of the tyrosine kinase inhibitor PKC412 in apatient with mast cell leukemia with the D816V KIT mutation. Blood. 2005;106:2865–2870.

140. Sotlar K, Escribano L, Landt O, et al. One-step detection of c-kit point mutations usingpeptide nucleic acid-mediated polymerase chain reaction clamping and hybridization probes.Am J Pathol. 2003;162:737–746.

141. Corless CL, Harrell P, Lacouture M, et al. Allele-specific polymerase chain reaction for theimatinib-resistant KIT D816V and D816F mutations in mastocytosis and acute myelogenousleukemia. J Mol Diagn. 2006;8:604–612.

Chapter 8Molecular Pathology of Chronic LymphocyticLeukemia

Daniela Hoehn, L. Jeffrey Medeiros, and Sergej Konoplev

Keywords Chronic lymphocytic leukemia (CLL) · Small lymphocyticlymphoma (SLL) · Genetic predisposition · Conventional cytogenetics · Fluorescencein situ hybridization (FISH) · 13q14 deletion · Retinoblastoma gene (RB1) ·D13S25 · D13S319 · miR-15a · miR-16-1 · 11q22-q23 deletion · Ataxia telang-iectasia mutated (ATM) · MLL · RDX · NPAT · CUL5 · PPP2R1B · Trisomy12 · CDK2 · CDK4 · STAT6 · APAF-1 · MDM-2 · 17p13 dele-tion · miR-34a · p53 mutation · 6q deletion · 3q27 trisomy · BCL6 · 8q24gain · MYC · t(14;19)(q32;q13) · BCL3 · BCL2 · t(2;14)(p16;q32) · DNAmethylation · Denaturing high-performance liquid chromatography(DHPLC) · Comparative genomic hybridization (CGH) · Single nucleotidepolymorphism arrays (SNP arrays) · Multiplex ligation-dependent probe ampli-fication (MLPA) · Gene expression profiling · microRNA · B-cell receptor(BCR) · IgH variable region (IgHV) · ZAP-70 · CD38 · Activation-inducedcytidine deaminase (AID) · Somatic hypermutation (SHM) · Class-switchrecombination (CSR) · Stereotyped BCR · Pharmacological inhibitionof BCR signaling · LYN · SYK · AKT · ERK · MCL-1 · BCL-XL · Dasatinib · Apoptosis · Caspases · CD95 · TOSO · CD40 · Tumornecrosis factor-related apoptosis-inducing ligand (TRAIL) · TCL-1 · Survivin ·IL-4 · Stromal cells · CCL22/MDC · CCL-17/TARC

Introduction

Chronic lymphocytic leukemia (CLL) and small lymphocytic lymphoma (SLL)are mature B-cell lymphoid neoplasms with an identical immunophenotype thatare currently thought to be very closely related if not identical. As a result,the current World Health Organization classification of hematolymphoid diseases

S. Konoplev (B)Department of Hematopathology, M.D. Anderson Cancer Center, Box 72, 1515 HolcombeBoulevard, Houston, TX 7030, USAe-mail: [email protected]


256 D. Hoehn et al.

has designated this disease as chronic lymphocytic leukemia/small lymphocyticlymphoma. Patients who present as CLL have lymphocytosis and bone marrowinvolvement and usually also have tissue sites of disease (lymph nodes, spleen,other organs). In contrast, patients who present as SLL have lymphadenopathy andcommonly have involvement of bone marrow and other tissue sites of disease,but lack lymphocytosis. This chapter will discuss the cytogenetic and molecu-lar genetic findings in CLL/SLL, and most of the data are derived from patientswith CLL. For simplicity, from this point forward CLL/SLL will be referred toas CLL.

Chronic lymphocytic leukemia is the most common hematological neoplasmof adults in western countries with an incidence of approximately 2–6 cases per100,000 per year, increasing to 12.8 cases per 100,000 per year in the seventhdecade. CLL represents approximately 7% of all non-Hodgkin lymphomas. Agenetic predisposition to CLL is thought likely. Familial clustering of CLL casescan be documented in 5–10% of patients (Swedish cancer database). In addition,CLL has marked geographic variation. Chronic lymphocytic leukemia is rare inAsian countries, and this low incidence is maintained in populations that migrate towestern countries, further supporting a genetic predisposition.

Cytogenetic Abnormalities

The identification and frequency of chromosomal abnormalities in CLL are highlydependent on the method employed. The two methods typically used are conven-tional cytogenetics and fluorescence in situ hybridization (FISH). Traditionally,FISH has been far more sensitive than conventional cytogenetic analysis. Theobserved false-negative rate of conventional cytogenetic analysis has been attributedto the low proliferation rate of CLL cells. Using conventional cytogenetic tech-niques, chromosomal abnormalities are identified in approximately 50% of CLL,whereas genetic abnormalities can be detected in up to 80% of CLL casesusing FISH probes. Recently, stimulation of CLL cells with either CD40 ligand-expressing cells and IL-4 or a combination of CpG-oligodeoxynucleotides and IL-2has led to an increased frequency of metaphase spreads for detailed chromosomeanalysis in CLL [1, 2]. In the initial report [1], this approach revealed transloca-tions in 33 of 96 (34%) of CLL patients. The presence of these translocations alsodefined a new prognostic subgroup with significantly inferior overall survival andshorter treatment-free survival (Fig. 8.1).

Later studies showed that most of these translocations are not balanced and areaccompanied by DNA losses at the breakpoints. In another study, after stimulationwith CpG-oligodeoxynucleotides and IL-2, the rate of detection of abnormalitiesin CLL was comparable to the rate of detection by parallel interphase FISH [2].Conventional cytogenetics using stimulation with CpG-oligodeoxynucleotides andIL-2 also frequently detected balanced and unbalanced translocations. The stud-ies also suggested that there is a higher frequency of complex aberrations (morethan three aberrations) in CLL than was appreciated using traditional methods. Ina recent, large update on the results of metaphase cytogenetics after stimulation

8 Molecular Pathology of Chronic Lymphocytic Leukemia 257

Fig. 8.1 The median survival times for the groups with 17p deletion, 11q deletion, 12q trisomy,normal karyotype, and 13q deletion as the sole abnormality were 32, 79, 114, 111, and 133 months,respectively. (Reproduced with permission from [4])

in more than 506 patients with CLL [3], 500 (98.8%) cases yielded metaphases.Aberrations were detected in 415 (83.0%) cases by conventional banding and in392 (78.4%) cases by FISH. Conventional karyotyping detected a total of 832abnormalities compared with 502 by FISH. A subgroup of CLL cases with a com-plex karyotype (16.4%) was identified. In addition, deletion 13q, the most commonabnormality in CLL, could be further subdivided on the basis of the presence oftranslocations or complex aberrations [3].

Due to economical restraints there is a tendency to combine several probes spe-cific to most common cytogenetic abnormalities in a panel, which makes FISHtesting more cost-effective. Examples of application of this technique are illustratedin Fig. 8.2.

Deletion 13q14

Deletion of band 13q14 is the most common chromosomal aberration in CLL, in40–60% of cases assessed by FISH. This deletion is associated with a long inter-val between diagnosis and the need for treatment [4]. The 13q14 locus is thought

258 D. Hoehn et al.

Fig. 8.2 An example of a commercially available fluorescence in situ hybridization panel for CLL.(a) So-called R tube containing the spectrum orange probe specific for p53 gene at 17p13.1 andthe spectrum green probe specific for ATM gene at 11q22.3. Two cells labeled 2G2R (one being ininterphase and one being in mitotic spread) demonstrate two preserved copies of both genes. Cellslabeled 1G2R demonstrate loss of one allele of the ATM gene; a cell labeled 2G1R demonstratesloss of one allele of p53 gene; a cell labeled 1G1R shows loss of both p53 and ATM genes. (b)So-called B tube containing the spectrum orange probe specific for the 13q13.4 locus, the spectrumgreen probe specific for the centromeric region of chromosome 12, and the spectrum cyanic probespecific for the 13q34 locus. Two cells labeled 2R2A2G (one being in interphase and one beingin mitotic spread) demonstrate a normal pattern. A cell labeled 1R2A2G demonstrates deletion of13q13.4 locus; a cell labeled 2R1A2G demonstrates loss of the 13q34 locus; a cell labeled 2R2A3Gshows gain of chromosome 12 (trisomy 12). (Microphotographs were kindly provided by LynneV. Abruzzo, M.D., Ph.D)

to be the site of a tumor suppressor gene. Since the retinoblastoma gene (RB1)is located at 13q14, and FISH has shown monoallelic loss of the RB1 in 30% ofCLL patients, RB1 deletion has been implicated in pathogenesis. RB1 encodes fora nuclear phosphoprotein that is involved in cell cycle regulation and transcrip-tion control. Disruption of both RB1 alleles by deletion or mutation, however, isextremely rare. These findings raise questions about the hypothesized pathogeneticrole of RB1 in CLL. Other studies have shown that the minimally deleted region at13q14 is 1.6 cM telomeric to RB1. Additional loci, D13S25 and D13S319, locateddistal to band 13q14 and more frequently deleted in CLL have been identified sug-gesting a novel tumor suppressor gene located in these regions. High-resolutiongenomic maps of these critical regions revealed several new genes from this area.BCMS and BCMSUN are more commonly deleted and require further research aspotential tumor suppressor genes.

Deletion or downregulation of two micro-RNA genes, miR15 and miR16, alsohas been implicated in CLL pathogenesis [5]. miR-15a and miR-16-1 are locatedin the minimally deleted region of 13q13.4. A germline mutation of miR-16-1 andmiR-15a was able to cause lower levels of micro-RNA expression and was associ-ated with deletion of normal alleles. Deletion or downregulation of these miRNAgenes has been shown in 70% of CLL cases (but not in CD5+ B lymphocytesobtained from healthy donors) [6]. Moreover, a mutation that reduces the expressionof miRNA15a and miRNA16-1 has been associated with loss of the other allele inCLL cells. This would support the hypothesis that this miRNA cluster functions as a


tumor suppressor gene [5]. The roles of miR-15a and miR-16-1 as tumor suppressorgenes in CLL have been verified in multiple studies that elucidated an inverse cor-relation between miRNA expression and BCL2 gene expression. Furthermore thiscluster directly targets and represses expression of BCL2 in leukemic cells [7]. Cellgrowth and cell cycle progression have been shown to be negatively regulated bymiR-16-1 [8]. miRNA expression also has prognostic implications as CLL patientswith monoallelic 13q14 deletion have slower lymphocyte growth kinetics than doCLL patients with biallelic 13q14 deletion [9].

Deletion 11q22-q23

Structural aberrations of chromosome 11 are reported in 12–25% of CLL cases.11q deletions are clinically associated with younger patient age, advanced clin-ical stage, and a worse prognosis. FISH studies have identified a critical regionaround the neural cell adhesion molecule (NCAM) at 11q23. This region is deletedin many hematological neoplasms [10]. A minimal consensus-deleted region of 2–3Mb in size in bands 11q22.3-q23.1 has been shown [11]. This area is rich in tumorsuppressor and other genes including FDX, ataxia telangiectasia mutated (ATM),MLL, and RDX. The ATM gene is involved in DNA repair. The frequent observationof lymphoma development in ATM knockout mice supports the function of ATMas a tumor suppressor gene [12, 13]. The absence of ATM protein expression inCLL cases further supports ATM being involved in CLL pathogenesis. Inactivationof ATM by deletion and/or mutation has been shown in multiple studies. Morethan two-thirds of CLL cases have monoallelic loss of ATM, while the other non-deleted allele often bears point mutations, most frequently located in the PI-3 kinasedomain.

Chronic lymphocytic leukemia cells carrying 11q22.3 deletions often showupregulation of genes that control cell cycle progression and signaling path-ways. ATM-mutant CLL cases exhibit a deficient ATM-dependent p21 responseto gamma irradiation, fail to upregulate tumor necrosis factor-related apoptosis-inducing ligand receptor 2 (TRAIL R2), and therefore have an inability torepair chromosomal breaks. ATM mutations are associated with genetic insta-bility, clonal evolution, and disease progression. ATM-mutant CLL cases uni-formly lack immunoglobulin variable region (IGHV) somatic hypermutation.This association indicates their crucial key role for ATM at the pregerminalcenter stage of B-cell maturation, subsequently contributing to developmentof CLL.

ATM gene mutations have been shown in CLL patients with and without del 11q[14, 15]. In a study of 155 CLL patients [15], ATM mutations were shown in 12%of cases and were associated with poorer prognosis. Most CLL patients with ATMmutations were refractory to DNA damaging chemotherapeutic drugs, and it hasbeen suggested that these patients might benefit from agents bypassing the classicalDNA damage pathway [15]. Interestingly, ATM mutations are only found in about

260 D. Hoehn et al.

one-third of the cases with 11q deletion suggesting that more complex mechanismsunderlie the poorer prognosis of patients with deletion 11q [16].

A hypothesized regulator of ATM expression is the miR34b/miR-34c cluster. Thiscluster is located at the deleted region 11q23, centromeric to ATM. Other candidategenes in the 11q22-23 region include NPAT (cell cycle regulation), CUL5 (ubiquitin-dependent apoptosis regulation), and PPP2R1B (component of the cell cycle andapoptosis regulating PP2A) which are all down-regulated in CLL cases with 11qdeletion.

Trisomy 12

Trisomy 12 was one of the first recurrent aberrations described in CLL, in 10–25%of cases. Detailed assessment by FISH studies has shown a minimally gainedregion limited to bands 12q13-q15. The gene (genes) in this region are presum-ably amplified and hypothesized to be oncogenes. Genes of oncogenic potential thatare localized in this region include CDK2, CDK4, STAT6, APAF-1, and MDM-2.Although these genes play critical roles in the regulation of oncogenesis, cell cyclecontrol, and apoptosis, their role in the CLL pathogenesis is yet to be elucidated.DNA microchip technology employing matrix comparative genomic hybridization(CGH) has shown the smallest replicated genomic regions in bands 12q13-q21 inCLL [17, 18].

Deletion 17p13

Chromosomal deletion of 17p13 occurs in 10–15% of CLL cases [19]. 17p deletionis associated with more aggressive clinical behavior and worse prognosis. Many ofthese cases show numerous copy number changes, many recurrent, and a highlyunstable genome [20]. Chromosome 17p13 is the location of the p53 gene. Variousstudies have shown a prevalence of p53 mutations in the range of 10% of all CLLcases and higher in patients with 17p13 deletions. p53 mutations have been foundin 4–17% patients with early stage CLL and have been associated with poor prog-nosis in a number of studies [21–23]. However, the exact prognostic relevance ofTP53 mutations has not been conclusively documented within prospective trials.There is growing evidence that functional testing of p53 may be used to predictprognosis [24].

Deletion of 17p/p53 mutation has been associated with downregulation of miR-34a. MiR-34a is transcriptionally induced by p53 and directly targets CDK6,CCND1, CDK4, CCNE2, and MET [25] expression in patients with CLL [26].Furthermore, there appears to be a common ancestry for the miR-15a/miR-16-1and miR-34b/miR-34c clusters [27]. Investigators have analyzed the transcriptomeinduced by overexpression of miR-34 in CLL. The transcriptome is highly enrichedfor genes that regulate cell cycle progression, DNA repair, angiogenesis, and apopto-sis [28]. Multiple studies have shown miR-34a to be proficient in inducing cell cycle


arrest and subsequent caspase-dependent apoptosis via BCL2 and E2F3 repression[29, 30].

Deletion 6q

Deletions of the long arm of chromosome 6 have been shown by conventional cytyo-genetics in 4–6%, and by FISH in up to 9%, of CLL cases. 6q deletions are notspecific for CLL. At least two independently deleted regions have been shown invarious types of non-Hodgkin lymphoma: 6q21-23 and 6q25-q27 [31, 32]. A mini-mally deleted region in band 6q21 spanning 4–5 Mb has been identified [33], but theidentity of candidate gene(s) in this areas is unknown. Deletions of 6q are often asso-ciated with high leukocyte counts, extensive lymphadenopathy, and splenomegaly.Cases of CLL with del(6q) commonly have atypical morphologic features, includ-ing cleaved lymphocytes and increased larger cells, either immunoblast-like orprolymphocytoid cells.

Trisomy 3q27

Trisomy 3q is a rare, recurrent abnormality in CLL. The distal arm of 3q contains aminimally duplicated region and this region includes 3q27 which carries the BCL6gene. BCL6 is commonly involved in chromosomal translocations in several typesof B-cell lymphoma. The BCL6 gene also can undergo point mutations, somatichypermutations, and microdeletions.

Translocations involving BCL6 occur in 3% of CLL cases. The partnergenes in BCL6 translocations are most commonly the Ig heavy or light chaingenes; however, other BCL6 partners include Ras homolog gene family memberH (RHOH) in the t(3;4), histone H1F1 in the t(3;6), Oct binding factor 1(OBF1) in the t(3;11), and lymphocyte cytosolic protein1(LCP1) in the t(3;13).In most translocations the first noncoding exon of the partner gene fuses withthe second exon of BCL6, resulting in deregulated expression of normal BCL6protein.

BCL6 encodes for a 706 amino acid zinc finger transcription factor and con-tains an N-terminal POZ domain. The 5′ portion of BCL6 encodes for the BTB/POZdomain (broad-complex/tramtrack/bric-a-brac/pox virus/zinc finger) and the 3′ endencodes for six DNA binding zinc fingers. BCL6 protein acts as a sequence-specificrepressor of transcription. BCL6 can bind to DNA in a sequence-specific fashion andrepress transcription and can recruit a variety of other POZ-containing proteins thatwill function as transcription corepressors/protein repressors. The carboxy termi-nus of BCL6 is responsible for sequence-specific DNA binding through its six zincfingers, mediated through the consensus sequence TTCCT(A/C)GAA. The protein–protein interactions on the other hand are mediated through the BTB/POZ domain.BCL6 protein is required for normal germinal center formation and antibody affinitymaturation and is highly expressed in germinal center B cells.

262 D. Hoehn et al.

8q24 gain

Gains in 8q occur in a small subset of CLL cases [34]. The 8q24 locus is thesite of MYC (previously known as c-MYC) which belongs to the MYC family oftranscription factors. MYC family transcription factors contain a basic helix-loop-helix leucine zipper domain. MYC encodes a transcription factor that is involved inregulating the expression of a significant number of genes in major cellular path-ways. MYC abnormalities arise via translocation, mutation, or amplification. Theseabnormalities lead to upregulation of MYC expression and therefore unregulatedexpression of many other genes.

The MYC gene has three exons. Two promoters, P1 and P2, control transcriptionand the choice of promoter depends on the MYC protein level. The P2 promoteris considered as most active, generating a 2.25 kb transcript, whereas the P1 pro-moter generates a 2.4 kb transcript. Expression of MYC, which is among the earliestevents following stimulation of the protein kinase signal transduction pathway, canbe successfully induced in CLL cells [35].

MYC overexpression in CLL is a marker of resistance to apoptosis [36]. Althoughrare in CLL, MYC translocations are associated with a poor prognosis. Affectedpatients have shorter survival, commonly present with increased prolymphocytes inblood and bone marrow, and additional cytogenetic abnormalities are common [37].A typical finding of a case with MYC rearrangement is illustrated in Fig. 8.3.

A mouse model attests to the role MYC plays in a subset of CLL. MYC over-expression cooperates with the B-cell receptor to induce lymphomas. In mice withMYC overexpression, with constitutive B-cell receptor signaling the tumors resem-bled CLL. In contrast, with antigenic stimulation of the B-cell receptor the tumorsresembled Burkitt lymphoma [38].

t(14;19)(q32;q13)

t(14;19)(q32;q13) is rare in CLL accounting for approximately 1% of cases. Thistranslocation involves the BCL3 gene at chromosome 19q13 and the IgH gene atchromosome 14q32 [39]. BCL3 and IgH are joined in a head-to-head configurationresulting in overexpression of BCL3.

The BCL3 gene encodes a protein of the IκB family and is involved in regulatingthe NF-kappa B family of transcription factor proteins. Unlike other IκB pro-teins, BCL3 functions as a coactivator of transcription and it inhibits NF-kappa B.Originally identified by its involvement in the t(14:19), BCL3 expression has beenreported in 12% of non-Hodgkin lymphomas, including a subset of CLL cases, and41% of Hodgkin lymphomas.

Patents with CLL associated with t(14;19) tend to be of younger age and oftenhave rapid disease progression and poorer prognosis. These neoplasms often haveatypical morphologic features manifested by a mixture of small and larger lympho-cytes, – and cleaved/indented nuclei. Occasional prolymphocytes are seen in somecases, but never exceed 10% Trisomy 12 and other abnormalities of chromosome


Fig. 8.3 CLL with MYC rearrangement. (a) The bone marrow aspirate smear shows increasedprolymphocytes (Wright-Giemsa ×1000); (b) In lymph node a pseudofollicle contains larger cellswith prominent nucleoli (H&E ×400); (c) Conventional cytogenetic analysis performed on aG-banded metaphase demonstrates t(8;14)(q24·1;q32); (d) FISH study performed on metaphaseand interphase cells using probes to the MYC (red signal) and IGH (green signal) loci demonstratesthat the neoplastic cells contain a reciprocal translocation between the two loci (yellow signals).(Reproduced with permission from [37])

19 are common in these tumors [40, 41]. Typical findings in a case of CLL witht(14;19) are illustrated in Fig. 8.4.

Chromosomal Translocations Involving BCL2

The BCL2 gene at chromosome 18q21 is rearranged in rare CLL cases.Rearrangement usually occurs as part of translocations with Ig genes, witht(2;18)(p11;q21)/Igkappa-BCL2 and t(18;22)(q21;q11)/Iglambda-BCL2 more com-mon than t(14;18)(q32;q21)/IgH-BCL2.

Although translocations are rare, BCL2 is upregulated in all cases of CLL,and by suppressing apoptosis contributes to the prolonged lifecycle of CLL cells.Upregulation of BCL2 is most commonly related to deletions of two genes,miRNA15a and miRNA16-1, located at 13q14. In 70% of CLL cases deletions ofmiRNA15a and miRNA16-1 lead to overexpression of BCL2.

264 D. Hoehn et al.

Fig. 8.4 CLL with t(14;19)(q32;q13). (a) Bone marrow aspirate smear showing atypical mor-phology of neoplastic cells with high nuclear:cytoplasmic ratio and nuclear indentations; Wright-Giemsa, ×1000. (b) Many of the neoplastic cells show strong nuclear staining for bcl3 (× 400).(c) A representative karyogram demonstrates the t(14;19)(q32;q13) and +12. Trisomy 12 is oftenseen in cases of CLL with t(14;19)(q32;q13). (Reproduced with permission from [41])

The miRNA genes belong to a highly conserved noncoding gene family, whosetranscripts inhibit gene expression by causing degradation of mRNA or by blockingtranscription of mRNA. MiR-15a and miR-16-1 are major direct negative regula-tors of the BCL2 antiapoptotic protein and, in addition are indirect activators of theintrinsic apoptotic program. This in turn activates the apoptotic peptidase activatingfactor:1/caspase-9/poly(ADP-ribose) polymerase pathway. The downmodulation ofthese two miRNAs in CLL has been correlated with altered expression of other anti-apoptotic proteins and miRNAs. For example miR-29b expression is known to bedownregulated in patients with CLL having poor prognosis. This miRNA suppressesMCL1, an antiapoptotic member of the BCL2 family.


t(2;14)(p13;q32)

The t(2;14)(p16;q32) is a rare, recurrent abnormality in CLL. The t(2;14) juxta-poses the BCL11A gene at 2p13 with IgH at 14q32. The BCL11A gene is the humanhomolog of the mouse EVI9 gene [42, 43]. The t(2;14) presumably occurs in acti-vated B cells in the course of T-cell-independent immune responses outside thegerminal center. This can be supported by the fact that these translocations aremainly targeted to the switch regions of the IgG2 gene, without T-cell-dependentimmune responses [43].

The t(2;14) in CLL has been associated with younger patient age (includingchildren) and atypical morphologic features, including plasmacytoid differentiation,irregular nuclei, and increased prolymphocytes. These neoplasms commonly haveunmutated IGVH genes and express ZAP70 and CD38 [44]. Typical findings in acase of CLL with t(2;14) are illustrated in Fig. 8.5.

Fig. 8.5 CLL with t(2;14). (a) The aspirate smear demonstrates that many of the cells arecytologically atypical, with irregular nuclear contours and increased prolymphocytes (Wright-Giemsa, ×1,000). (b) Conventional cytogenetic analysis demonstrating the t(2;14)(p16;q32);(c) Fluorescence in situ hybridization analysis was performed on interphase cells using a dual-color, dual-fusion probe to BCL11A at 2p16 and IgH at 14q32. Reciprocal translocation involvingthe BCL11A and IgH genes generates yellow fusion signals on both derivative chromosomes. Thenormal chromosome 14 shows a green signal, and the normal chromosome 2 shows a red signal.A cell that contains the translocation is on the left; a cell without the translocation is on the right.(Reproduced with permission from [44])

266 D. Hoehn et al.

Translocations Involving Chromosome 14q32

When all translocations involving chromosome 14q32/IgH are tabulated, they occurin a small percentage of cases of CLL. These translocations are currently bestdetected by FISH. In general, patients with CLL that carry translocations involvingchromosome 14q32/IgH, regardless of the partner chromosome, more commonlyhave advanced stage disease, unmutated IgH variable region genes, express CD38and ZAP70, and have a poorer prognosis [45].

Epigenetic Changes

Gene methylation is another mechanism of gene inactivation, and the field focusedon the study of this phenomenon is known as epigenetics. With the advent of newmethods for assessing DNA methylation, the importance of epigenetic changes inhuman cancers is well recognized [46].

In CLL, global DNA hypomethylation was initially detected 16 years ago [47].The genome-wide patterns of DNA methylation in CLL are rather specific for CLLcells and can be reproduced in different patients [48, 49] suggesting that DNAmethylation plays a specific role in the pathogenesis of CLL. Aberrant methyla-tion has been described for specific genes in CLL, for example, hypomethylationof BCL2 and TCL1 [50, 51]. Unbiased genome-wide screens of CLL have shownspecific methylation of the promoter region of TWIST2 [52] and of DAPK1 kinase[53]. When DNA methylation analyses were performed on the promoter regionof ZAP-70, a specific CpG was methylated in 51 of 53 ZAP70-negative CLLcases. By contrast, no methylation was found in 30 of 32 ZAP70-positive CLLcases [54].

Apart from their role in pathogenesis and their potential utility in diagnosis,epigenetic modifications are also an interesting target for therapy as epigenetic mod-ifications are reversible (in contrast with genetic aberrations). Several compoundshave been tested on cells in vitro and in clinical studies and shown to be biologicallyand clinically active. An inhibitor of histone deacetylases (depsipeptide FR901228)was shown to selectively target CLL cells in vitro as compared with normal bloodB cells and BM-derived progenitors [55].

High-Throughput Molecular Methods to Assess CLL

Sequence and Mutation Analysis

The availability of high-throughput techniques for mutation screening, e.g., denatur-ing high-performance liquid chromatography (DHPLC) will allow using mutationdetection in larger cohorts of patients and in a timely fashion. The technique allowsthe detection of point mutations and small deletions on the basis of heteroduplex for-mation and the different mobility of the homoduplex (no mutation) and heteroduplexstrand (mutation).


The systematic analysis of mutations of these and other tumor suppressors withinclinical trials will help to further delineate the prognostic value and clinical use ofmutation assessment for clinical decision making in the treatment of patients withCLL. Currently there is no consensus on which method is best used for mutationanalysis but the choice will often depend on local expertise and equipment. In addi-tion to DHPLC, mutation analysis can be performed by direct sequencing, DGGE,SSCP, and array-based mutation analysis.

Array Comparative Genomic Hybridization

Comparative genomic hybridization (including array-CGH) is a microarrayplatform-based test, in which genomic hybridization is compared to an arrayof defined DNA fragments. General principles of array-CGH are illustratedin Fig. 8.6.

Fig. 8.6 General principles of comparative genomic hybridization array (CGH-A) and singlenucleotide polymorphism array (SNP-A) arrays. (a) In array-CGH, control DNA (oligo or bac-terial artificial chromosome probes) is used as reference for the test DNA (putative tumor DNA).Decreased copy number in the tumor DNA results in decreased intensity of the signal for the testand increased signal for reference DNA. (b) In SNP array, hybridization of amplified and labeledDNA with probes corresponding to alleles for each locus results in a genotyping pattern allowingfor determination of the heterozygosity or homozygosity for each allele. At the same time, inten-sity of the hybridization signals allows for determination of copy number changes. (Reproducedwith permission from [192].)

268 D. Hoehn et al.

Fig. 8.7 An example of array-CGH technique applied to CLL cases and revealing frequent sub-microscopic 22q11 deletions. (a) Monoallelic loss of 22q11 was observed in a total of 24 casesand (b) biallelic loss was observed in a total of four cases. One case with (c) monoallelic lossand one case with (d) biallelic loss also exhibited concomitant gains of whole chromosome 22.(e) Oligonucleotide array-CGH analysis of the twenty-eight 22q11 deletion cases identified byBAC array-CGH revealed a minimally deleted 340,279 bp region that contained the ZNF280A(SUHW1), ZNF280B (SUHW2), PRAME, and GGTLC2 (GGTL4) genes. One CLL case with amonoallelic 22q11 deletion is shown. The green and red dots represent the fluorescence ratios ofindividual oligonucleotide probes on the microarray. Red dots represent probes with positive flu-orescence ratios and green dots represent probes with negative fluorescence ratios. The cluster ofgreen probes significantly shifted to the left of zero at 22q11 represents a small deletion. The ver-tical green bar at the left of the figure and the green-shaded rectangle represent the deleted regionat 22q11. (Reproduced with permission from [56])

Array-CGH has been proven to be a sensitive tool in detecting new recurrentcytogenetic abnormalities [18]. Comparing array-CGH to chromosomal CGH in alarge series of CLL cases, Schwaenen and colleagues demonstrated that the spatialresolution of array-CGH was much better and allowed the detection of small-sizedimbalanced regions [18]. In this study, array-CGH allowed detection of previouslyunrecognized recurrent genomic imbalances as a copy number gain of chromo-some 19 and the MYCN oncogene on 2p24 [18]. Recently, Gunn and colleaguesused BAC array-based CGH to detect genomic imbalances in 187 CLL cases anddemonstrated submicroscopic deletions of chromosome 22q11 in 28 cases (15%)with the frequency of these deletions being second only to loss of the 13q14 region[56] (Fig. 8.7). Oligonucleotide-based array-CGH analysis showed that the 22q11deletions ranged in size from 0.34 Mb up to 1 Mb. The minimally deleted regionincluded the ZNF280A, ZNF280B, GGTLC2, and PRAME genes. Quantitative real-time PCR revealed that ZNF280A, ZNF280B, and PRAME mRNA expression wassignificantly lower in the 22q11 deletion cases compared with nondeleted cases(Fig. 8.7) [56].

Single Nucleotide Polymorphisms Arrays

Single nucleotide polymorphisms arrays (SNP-arrays) have recently been usedfor genome-wide detection of copy number changes in CLL [57]. In addition,


SNP-arrays allow for the detection of uniparental disomy, a potential cause of loss ofheterozygosity (LOH) without copy number changes. Pfeifer and colleagues appliedSNP arrays to a large series of CLL patients and detected chromosomal imbalancesin up to 82% of cases [57]. This study observed that aberrations of prognostic impor-tance were identified at expected frequency [57]. In addition, 24 large (>10 Mb)copy number neutral regions with LOH were identified in 14 cases. These abnor-malities are not detectable by other methods and may harbor relevant genes orloss-of-function alleles which may be important for the pathogenesis of CLL [57].

Multiplex Ligation-Dependent Probe Amplification

Multiplex ligation-dependent probe amplification (MLPA) is a novel PCR-basedtechnique to detect genomic alterations, which allows the analysis of more than40 different small (50 pb) DNA sequences in a single reaction [58]. This techniquerelies on the comparative quantitation of specifically bound probes that are amplifiedby polymerase chain reaction (PCR) with universal primers and allows simultaneousprocessing of multiple samples. Coll-Mulet and colleagues performed MLPA analy-sis in CLL patients with the simultaneous identification of 55 genomic CLL-specifictargets, and compared the results of the analysis with FISH data [58]. Resultsshowed a good correlation between MLPA and FISH, as most of the alterationswere detected by both techniques [58].

Gene Expression Profiling

Array-based gene expression studies in CLL have shown remarkable results. Theinitial studies were able to detect a small group of genes that could differentiatemutated from unmutated CLL cases after supervised clustering [59, 60] (Fig. 8.8).A number of these genes have been confirmed to distinguish prognostic subgroupsin CLL including ZAP70. Further studies on the gene expression profile in CLLdemonstrated several important findings such as a gene dosage effect of the chro-mosomal aberrations [61], a p53-dependent signature in response to fludarabinetreatment [62], and a highly sophisticated temporal program following the stimu-lation of CLL cells by B-cell receptor cross-linking [63]. Recently, a gene profilingstudy demonstrated the relationship of clinical variability and the expression of twogene clusters, associated with B-cell receptor signaling and mitogen-activated pro-tein kinase activation [64]. The expression of these clusters dramatically separatedpatients into three groups with treatment-free survival probabilities at 5 years of 83,50, and 17% [64].

MicroRNA

This approach is one of more recently developed technologies applied to the analysisof CLL and other lymphomas and leukemias. As this is specifically discussed inanother chapter in this book, we have mentioned relevant aspects of microRNAbiology in this chapter but refer the reader to the complete discussion elsewhere.

270 D. Hoehn et al.

Fig. 8.8 Relative gene expression levels in CLL. (a) Hierarchical clustering of gene expres-sion data for 205 array elements representing approximately 175 genes that were differentiallyexpressed between Ig-mutated and Ig-unmutated CLL samples (P < 0.001). (b) Hierarchical clus-tering of genes that most strongly discriminated between the CLL subtypes. Also shown for eachgene is the ratio of mean expression of the gene in Ig-unmutated CLL samples versus mean expres-sion in Ig-mutated (high) CLL samples, together with the P values (Student’s t test) that quantitatethe significance of the difference in mean expression between the two CLL subtypes. (c) RT-PCRanalysis of ZAP-70 expression. Shown are data from two Ig-unmutated and two Ig-mutated CLLcases, a T-cell line (Jurkat), various B-cell lines found by microarray analysis to express ZAP-70(LILA, LK6, OCI-Ly2), and a B-cell line not expressing ZAP-70. The control lane represents areaction in which the reverse transcriptase was omitted. (Reproduced with permission from [59])

B-Cell Receptor

Most cases of CLL express surface monotypic Ig light chain, IgM, and IgD. SurfaceIgs are usually expressed at low density/dim intensity for, in large part, still unknown


reasons [65]. The Ig genes are composed of variable and constant regions, andthe variable regions can undergo mutations in the complementarity determiningregions (CDR), thought to be induced by exposure to antigen. Physiologically,these mutations permit the development of more specific antigen-antibody binding[66–71].

The complex of Ig light and heavy chain genes and other molecules expressed onthe cell surface of all B cells, and specifically CLL, comprises the B-cell receptor(BCR). The BCR is a key molecule for understanding the molecular mechanismsof CLL. For the remainder of this chapter, we attempt to place molecular geneticfindings in the context of BCR signaling.

Somatic Mutations of Ig Variable Region Genes

A milestone in understanding of CLL was the observation that somatic mutations ofthe Ig variable region genes occur in a subset of CLL cases and that the absence orpresence of these mutations correlates with clinical course [72, 73] (Fig. 8.9).

Somatic mutations occur in both the IgH and Ig light chain genes, but IgH vari-able region (IGHV) analysis is easier and adequate for prognostic purposes. Anarbitrary cutoff of 2% or more mutations divides CLL into two prognostic groups.Patients with CLL in which the IGHV genes show < 2% somatic mutations (unmu-tated CLL) have a poor clinical outcome. It appears that CLL cells in this subsetreceive continuous antiapoptotic and/or proliferating microenvironmental stimulivia the BCR leading to more aggressive disease. In contrast, CLL patients in which

Fig. 8.9 (a) Somatic hypermutation of the V region of the IgH/IgL loci. Mutations that are tar-geted to the V region can result in either silent or replacement mutations at the amino acid level.(b) Kaplan–Meier plot comparing survival based on the absence (unmutated) or presence (mutated)of significant numbers (<2%) of V gene mutations in 47 B-CLL cases. Median survival of unmu-tated group: 9 years; median survival of mutated group not reached; P = 0.0001; log-rank test).(a) Reproduced with permission from [193]. (b) Reproduced with permission from [72]

272 D. Hoehn et al.

the IGHV genes show ≥ 2% mutations (mutated CLL) have a clinically indolentcourse [72, 73]. A difference in outcome was also demonstrated in CLL patientsreceiving an autologous stem cell transplant (ASCT); all patients with unmutatedIGHV genes relapsed and progressed after a 4-year follow-up, whereas most patientswith mutated IGHV genes remained in molecular remission [74].

Initially it was thought that CLL cases with unmutated IGHV genes (U-CLL)were derived from naive B cells, and that CLL cases with mutated IGHV genesCLL (M-CLL) derived from antigen-experienced B cells. It is now known that allCLL cells bear the surface membrane immunophenotype of being antigen experi-enced (CD27) and activated (expression of CD23, CD25, CD69, and CD71). Geneexpression profiling has shown that CLL cells have a profile similar to memoryB cells irrespective of IGHV mutational status [59, 60]. These findings suggest thatall cases of CLL have a common cellular origin and/or common mechanism oftransformation [59] and a continued requirement for antigen after the transforma-tional event [75]. Comparative analysis of U-CLL and M-CLL, however, suggeststhat the neoplastic cells, by differing in their IGHV genes, have different antigenencounter histories [75]. It also has been suggested that an antigen-driven pro-cess might be critical for determining clinical features and for modulating diseaseoutcome, irrespective of mutation status in B-CLL [76].

Signaling downstream of the BCR in CLL is dominated by the kinases LYNand SYK, which transduce survival and antiapoptotic signals after antigen triggersthe BCR [77]. Antiapoptotic BCR signaling has been associated with prolongedactivation of the MEK/ERK and PI3K/AKT pathways and with AKT-induced ele-vated expression of antiapoptotic MCL-1, which leads to increased survival of CLLcells [78]. Signal transduction via LYN is regulated and amplified via CD19 andthese signals are responsible for the establishment of baseline signaling thresholdsin B cells before antigen-receptor ligation, in addition to augmenting tonic signal-ing following BCR engagement [79]. LYN was identified as a major contributor toantigen-independent BCR signaling, as it is strongly overexpressed, constitutivelyactive and aberrantly present in the cytosol [77]. The recruitment and subsequentactivation of SYK to immunoreceptor tyrosine-based activation motifs within thecytoplasmic tails of Igα and Igβ seems to be disturbed in CLL, as alternative tran-scripts of Ig have been described [80]. CLL clones with a proliferative responseto BCR ligation have considerably higher Syk levels than nonresponsive ‘anergic’CLL cells [81], and the tyrosine kinase ZAP-70 can partially restore BCR signalingwhen Syk is not expressed [82].

ZAP-70, which is involved in T-cell receptor signaling, is aberrantly expressedin some patients with CLL and shows partial, but not complete, overlap with thepresence of unmutated IGHV genes [83, 84]. Currently, the detection of ZAP-70 inneoplastic cells can be performed both by flow cytometry immunophenotypic stud-ies and by immunohistochemistry. An example of immunohistochemical analysis inbone marrow is illustrated in Fig. 8.10.

High ZAP-70 expression in CLL cells is associated with more aggressive disease.While M-CLL cells are considered anergized, U-CLL cells seem to retain somecapacity for competent BCR signaling, with an increased tendency to phosphorylate


Fig. 8.10 Immunohistochemical study for ZAP70 expression using fixed, paraffin-embedded tis-sue of a bone marrow aspirate clot section. Note dim ZAP70 expression by CLL cells and brighterZAP70 expression by intermixed reactive T cells. (a), ×200; (b), ×1000

SYK and to recruit and phosphorylate ZAP-70 [85]. The presence of ZAP-70 canenhance and prolong BCR signaling in CLL independent of its tyrosine kinasefunction, probably by acting as an adaptor protein [82].

CD38 is another potential modifier of BCR signaling that shows substantialoverlap with ZAP70 expression and unmutated IGHV genes. CD38 is a moleculethat affects proliferation and longevity of the neoplastic clone [86]. CD38 ligationby monoclonal antibodies results in proliferation and blastic transformation of asubset of CLL cells, and engagement of CD38 as a receptor is a Lyn-dependentprocess [87]. Dynamic localization to lipid rafts and lateral association of CD38with the BCR, CD19, and CD81 was reported to be a prerequisite for CD38-mediated signaling and both enhancement and refinement of BCR signaling [88].Importantly, ZAP-70 represents a limiting factor in the CD38 signaling pathway,probably serving as a cross-point where BCR signals are enhanced and where migra-tory signals from chemokine receptors intersect with growth signals mediated viaCD38 [89].

Activation-induced cytidine deaminase (AID), an enzyme involved in thesomatic hypermutation (SHM) process and class-switch recombination (CSR) dur-ing normal B-cell differentiation [90], is upregulated in U-CLL cells [91] and isfunctional with generation of isotype-switched transcripts and mutations in the pre-switch μ region [92, 93]. AID upregulation causes mutation in genes associatedwith aggressive disease (e.g., BCL6, PAX5, MYC, RHOH) [94, 95]. In addition, anassociation of AID expression with deletions in 11q- and loss of TP53 has beendemonstrated [96].

A main determinant of BCR-mediated signaling in CLL may be the expressionlevel of surface IgM, which is generally upregulated in U-CLL and downregulatedto anergy in M-CLL [85]. Another possible explanation for anergy in M-CLL maybe chronic exposure to soluble antigens in the absence of co-stimulatory signals [97]and of unresponsiveness to BCR signaling owing to receptor desensitization. It alsohas been shown that BCR translocation to lipid rafts differs in M-CLL comparedwith U-CLL. Constitutive exclusion of the BCR from lipid rafts was observed in

274 D. Hoehn et al.

M-CLL and could be responsible for impaired interactions between the BCR andthe actin cytoskeleton [98].

This molecular signature of anergy in CLL is characterized by constitutive activa-tion of MEK and ERK [99]. As anergy occurs primarily in M-CLL, this may explainwhy ERK phosphorylation defines CLL cases with a favorable prognosis [99].

Stereotypy

Biased use of IGHV genes has been described in CLL. The VH1 gene family ofthe IgH gene and the VIIIb family of the Igkappa gene are used in the forma-tion of Ig gene rearrangements more often than can be explained by chance alone.DNA sequence analysis of the variable (V), diversity (D), and joining (J) seg-ments involved in IgH and Igkappa gene rearrangement has shown that a subsetof CLL have somatic mutations [4, 100]. It has been demonstrated that unrelatedand geographically distant patients with CLL share quasi-identical sequences ofBCR. In one large series, 15 of 1,220 unrelated patients (1.3%) carried virtuallyidentical unmutated IGHV created by rearrangement of the IGHV1-69, IGHD3-16, and IGHJ3 genes [101]. Several groups have reported CLL subsets carryingBCR characterized by non-random pairing of specific IGHV, highly homologous,or identical HCDR3 often associated with a restricted selection of IGVK or IGVLlight chains (the so-called stereotyped BCR) [101–109]. These stereotyped BCRhave been detected in more than 20% of CLL cases [102, 104, 105, 109]; the non-random composition of the expressed BCR on the CLL cells with IG binding haslead to the hypothesis of that similar/identical antigens are involved in pathogenesis[102, 104, 105, 110].

The frequency of stereotyped BCR is higher in U-CLL [107]. Most clusters arecomposed of U-CLL cases [102, 104–107, 109]. In particular, these clusters includeCLL cases that express autoreactive and polyreactive BCR, allegedly derived fromthe B-cell compartment devoted to the production of natural antibodies [108, 110,111]. Among “common” clusters, of particular clinical interest is a cluster composedby U-CLL cases with stereotyped BCR expressing genes from the IGHV1 genefamily other than IGHV1-69 (IGHV1-2, IGHV1-18, IGHV1-3,IGHV1-46, IGHV7-4-1), homologous HCDR3 bearing the QWL amino acid motif, and IGKV1-39 lightchains [102, 104, 105]. The prognosis of CLL patients whose neoplasms expressthis stereotyped BCR is poor, compared with all the other CLL patients or CLLpatients in whom their tumor expresses the same IGHV genes but without the samestereotyped BCR [102, 105].

Among the few clusters of M-CLL cases there are two clusters, both express-ing IgG, composed by cases expressing IGHV4-34 and IGHV4-39, respectively[102, 104, 105, 112, 113]. Specific cluster-biased genomic aberrations have beenfound; 13q has been associated with the IGHV4-34/IGKV2-30 cluster and tri-somy 12 has been associated with the IGHV4-39/IGKV1-39 cluster [113]. Thelatter cluster has been associated with the development of large B-cell lymphoma(Richter syndrome) [114, 115]. Other clusters, mainly composed of M-CLL casesand expressing IGHV3 are less frequent and might be subjected to a geographical


bias. Of particular interest is a group of IGHV3-21 CLL, composed of cases witheither unmutated or mutated IGHV genes, that express a stereotyped BCR char-acterized by an unusually short and highly homologous HCDR3 associated withIGLV3-21 [103, 106, 116–118]. A significantly skewed representation of this par-ticular cluster has been well documented in different European and non-Europeancountries and even in different regions from the same country [103, 116–118].It was clearly demonstrated that patients belonging to IGHV3-21/IGLV3-21 CLLcluster have shorter time to treatment compared with all M-CLL cases and withM-CLL cases that express IGHV3-21, but not included in this stereotyped cluster[103, 105, 116]. The molecular basis for a more aggressive clinical behavior ofCLL belonging to the IGHV3-21/IGLV3-21 CLL cluster is also suggested by geneexpression profiling and immunophenotypic analyses [103]. There are current pro-posals to use IGHV3-21 expression to drive clinical decisions in prospective trials[119, 120].

It also has been observed that cases expressing the IGHV3-23 gene are con-stantly absent from stereotyped BCR clusters [121], despite the fact that IGHV3-23is frequently used, usually in cases of M-CLL [102, 103, 105]. A possible explana-tion justifying the absence of IGHV3-23 genes from clusters of stereotyped BCR isthat IGHV3-23-expressing BCR might be selected through non-CDR-based recog-nition mechanisms, e.g., through interactions with superantigens, a general featureof BCR expressing IGHV3 subgroup genes [121–124]. IGHV3-23 expression hasbeen identified as an independent negative prognosticator in the context of M-CLL[121].

Another example of the clinicobiological implications of BCR stereotypy isprovided by cases that utilize the IGHV4-34 gene [102, 104]. This gene encodesantibodies that are intrinsically autoreactive by virtue of universal and largely lightchain-independent recognition of the N-acetyllactosamine (NAL) antigenic determi-nant of the I/i blood group antigen [125, 126]. The IGHV4-34 gene is used at a highfrequency in normal individuals; however, IGHV4-34 cells are censored at multiplecheckpoints during B-cell development to alleviate their autoreactivity [127]. Thisfinding explains why IGHV4-34 antibodies are virtually undetectable in healthysera, despite the abundance of IGHV4-34 B cells in normal individuals. In contrast,IGHV4-34 antibodies are secreted at high levels in patients with systemic lupuserythematosus (SLE) and closely related to tissue damage and disease activity [127].

The IGHV4-34 gene is also frequently used in CLL and usually employed inM-CLL cases [102, 104, 128], perhaps reflecting the fact that IGHV4-34 sequencesmust undergo somatic hypermutation (SHM) in order to negate their autoreactiv-ity and become sufficiently ‘safe’ to be allowed into the functioning Ig repertoire.Several groups have reported that a major subset of CLL cases expressing stereo-typed IgG-switched, mutated BCRs employ the IGHV4-34 gene in association withthe IGKV2-30 gene [102, 104, 107]. The IGHV4-34/IGKV2-30 stereotype (sub-set 4) is shared by greater than 1% of patients with CLL. As first shown by aMediterranean group, subset 4 is also characterized by striking clinical similarities[102]. In particular, cases belonging to this subset were significantly younger atdiagnosis and had indolent disease.

276 D. Hoehn et al.

The IGHV4-34/IGKV2-30 Ig stereotype of subset 4 cases is characterized bylong, positively charged HCDR3, enriched in aromatic and positively charged aminoacids, similar to pathogenic anti-DNA antibodies [129]. Furthermore, subset 4 VHand VK domains exhibit very distinctive SHM patterns, typified by the frequentintroduction of acidic residues (especially aspartic and glutamic acid residues)[104]. All subset 4 IGHV4-34 sequences were also recently reported to carry intactNAL-binding motifs [104]: thus, in theory, these IGHV4-34-expressing CLL cellscould still be bound (and stimulated for clonal expansion) by NAL-containing epi-topes present in various auto- and exo-antigens. These superantigen-like interactionscould provide the signals promoting survival, expansion, malignant transformation,and potentially clonal evolution.

CLL cells often produce BCRs/mAbs (monoclonal antibodies) that bindautoantigens in a polyreactive manner [130–132], similar to ‘natural antibodies’[133]. In recent studies using recombinantly expressed mAbs from CLL patientswith both unmutated and mutated IGHV, about 80% of U-CLL mAbs and about15% of M-CLL mAbs reacted in enzyme immunoassays with various self andforeign antigens in a polyreactive pattern [111]. Similarly, these mAbs reacted withfixed and permeabilized cells, as in a standard clinical anti-nuclear antibody assaythat is used to define systemic autoimmune disorders. However, the reactivity ofCLL mAbs with intracellular targets was primarily directed against cytoplasmicstructures [111], unlike Abs from patients with autoimmune disorders like SLE thatfrequently, although not exclusively, react with molecules residing in the nucleus. Inrelated studies, mAbs derived from Epstein–Barr virus (EBV)-immortalized CLLB cells and from soluble mAb elaborated by CLL cells stimulated in vitro withdifferentiating agents, reacted with tissue antigens of human tonsil and rat aorticsmooth muscle [110]. As the monoreactivity of several of the M-CLL mAbs couldbe changed to polyreactivity by reverting the Ab amino acid structure to thatof the germline gene, it appeared that most CLL B cells emanate from normalB lymphocytes with poly/autoreactive BCRs [111].

In an attempt to define the cytoplasmic, autoantigenic targets recognized by CLLcells, mAbs from subset 1 patients [102] were used to probe HEp-2 cell lysatesin immunoprecipitation experiments [134]. The BCRs/mAbs of subset 1 patientsare characterized by unmutated rearrangements involving IGHV1-69, IGHD3-16,and IGHJ3, with nearly identical HCDR3 sequences that are paired with unmu-tated IGKV3-20 having equally restricted KCDR3s [101, 102, 107]. These mAbsselectively isolated a molecule of 225 kDa, which upon amino acid sequencingwas identified as nonmuscle myosin heavy chain IIA (MYHIIA) [134], a majormolecular motor of normal cells. By using another member of this mAb subsetin similar immunoprecipitation experiments, the identity of MYHIIA as the tar-get of this group of CLL mAbs was corroborated. In addition, exposure of cells toreagents that alter MYHIIA amounts (specific siRNA) and cytoplasmic localization(blebbistatin) resulted in a corresponding change in binding to these mAbs, therebyconfirming that MYHIIA was their intracellular target [134]. Similar studies werecarried out with a mAb from a patient whose clone belonged to subset 8 that exhibitsa stereotypic rearrangement of IGHV4-39, IGHD6-13, and IGHJ5 associated with


an IGKL1-39 rearrangement containing an Arg at the IGKV–IGKJ junction. Thesestudies revealed vimentin as a target for this subgroup of mAbs [B75]. Furthermore,these studies identified other cytoskeletal components as targets of CLL mAbs, e.g.,filamin B and cofilin-1 [110].

Pharmacological Inhibition of BCR Signaling

Pharmacological inhibition of BCR signaling could provide a twofold effect disrupt-ing both tonic and antigen-ligation-dependent BCR signaling. Potential therapeutictargets for inhibition of BCR signaling in CLL are LYN and SYK [97]. Specificinhibition of LYN and SYK induces apoptosis of CLL cells in vitro, which indicatesthat these kinases primarily transmit prosurvival signals [97]. Inhibition of LYNmight partially mimic an anergic state, as is present in M-CLL. Dasatinib, an ABLand SRC-kinase inhibitor, induces apoptosis in primary CLL cells in vitro, with apreference for the U-CLL or ZAP-70+ CLL subsets, at concentrations clinicallyachievable by oral administration [135]. Preliminary data from a currently ongoingphase II clinical trial of dasatinib in heavily pretreated patients with relapsed CLLdemonstrated several patients with partial responses in lymph nodes and reductionsof lymphocyte counts [136]. Decreased signaling via survival-signaling cascadesinvolving AKT and ERK, reduced expression of antiapoptotic MCL-1 and BCL-XL,and increased p53 levels are also considered to be the basis of the antiproliferativeand proapoptotic capacity of dasatinib [135]. Ex vivo, CLL lymph node samplesdisplay strong ERK activation and high levels of BCL-XL and MCL-1; this hasbeen attributed to CD40-triggered events [137].

Protein kinase C (PKC) and phosphatidylinositol 3-kinase (PI3K)-mediated sig-naling pathways are centrally involved in controlling apoptosis and CLL survival[138]. Approaches that target several isoforms of PKC are under clinical investi-gation. Several PKC isoforms are constitutively active in CLL [139]. Some PKCisoforms are key mediators of BCR signaling (for example, PKCβII, which isoverexpressed in CLL) [140], whereas other isoforms induce AKT activation inde-pendent of BCR ligation and PI3K in CLL cells [141]. PKCδ is permanentlyactivated and downstream of the constitutively activated PI3K in CLL. Specificblockade of PKCδ by rottlerin induces apoptosis and synergizes with vincristinein CLL cells but not in normal B lymphocytes [142]. It is possible that the synergis-tic effect of rottlerin is sufficient to disrupt the balance between antagonizing PKCα

and PKCδ isoforms [142, 143]. The PKC modulator, bryostatin 1, increases CD20expression via MEK1/ERK signaling in a PKC-dependent manner in CLL cells andleads to a twofold increase in apoptosis induction by rituximab [144]. Rituximabachieves inhibition of the RAF/MEK/ERK signaling pathway, resulting in Bcl-xL downregulation and chemosensitization [145]. Farnesyl transferase inhibitors,known to block RAS activity and ERK phosphorylation, have been shown to induceapoptosis in primary CLL cells refractory to standard therapy [146]. Inhibition ofMEK significantly enhances cytotoxicity of purine analogs in a CLL cell line [147].

278 D. Hoehn et al.

Apoptosis

It is of interest that MYHIIA and vimentin, and possibly other nucleic acid-bindingintracellular molecules, transmigrate to the external surface of cells undergoing pro-grammed cell death [148–150]. This finding is consistent with the observation thatmany U-CLL mAbs (about 60%) and some M-CLL mAbs (about 10%) react withthe surfaces of apoptotic cells and not the same cells in the viable state [110, 151,152]. Within the apoptosis-reactive CLL cases, those using IGHV1-69 were themost abundant, and mAbs from patients with mutated IGHV3-21 were less reac-tive, indicating that not all CLL mAbs encoded by IGHV genes associated withpoor clinical outcome necessarily bind apoptotic targets.

CLL blood cells typically undergo apoptosis when cultured in vitro, indicatingthat in vivo accumulation of leukemic lymphocytes is favored by other factors prob-ably originating from the microenvironment. By studying cells at different timesafter the induction of apoptosis, it was found that some of the intracellular antigensbound by CLL cells translocated to the surface membrane and could be identified inapoptotic blebs [151, 152]. Animal studies indicate that some of the target antigensthat appear during apoptosis represent chemical modifications of lipids, lipopro-teins, or proteins[153]. These modifications are often the consequences of oxidation[154]. It was demonstrated that proteins or lipoproteins that were not normally rec-ognized by CLL mAbs in their native state became targets of these mAbs afteroxidative modification [110, 151]. Such modified molecules [e.g., bovine serumalbumin (BSA) derivatized with malondialdehyde] inhibited the binding of certainCLL mAbs to apoptotic cells [110, 151], suggesting that at least some of the epi-topes recognized on apoptotic cells by CLL mAbs were neoantigens created duringthe apoptotic process

Antibodies reactive with neoepitopes created during apoptosis can also recog-nize phosphorylcholine (PC) a molecule hidden in cell membranes of mammalianand microbial cells but exposed during apoptosis [155]. It has been shown that CLLmAbs recognize PC and PC-substituted molecules [110, 151], suggesting that CLLmAbs reactive with autoantigens or neoantigens also bind bacterial cell wall compo-nents. These data confirm the observation that most U-CLL mAbs react with intactbacteria of multiple strains [156].

Apoptotic Pathway

Multiple signals converge in CLL cells to influence cellular fate. Cell death isregulated by a network of cellular signaling cascades that are also influenced byintracellular sensor modules [157]. Upon ligation, these receptors create a platformfor activation of initiator caspases that directly feed into a caspase cascade leadingto cell death [158].

A number of death signaling components have been characterized in CLL.CLL cells express CD95 but are largely resistant to CD95 cross-linking [159].Overexpression of the CD95 regulator TOSO in CLL might contribute to CD95resistance [160]. Although CD40 signaling upregulates CD95, CLL cells commonly


remain resistant to CD95 triggering [161]. IL-15 treatment also is able to sensitizeCD40-stimulated CLL cells to CD95 cross-linking [162]. In addition, CD40L-stimulated CLL cells were rendered CD95 sensitive by XIAP (X-linked inhibitorof apoptosis protein) inhibition using non-SMAC (second mitochondria-derivedactivator of caspases) mimetic, synthetic compounds [163]. Also, CD40-activatedCLL cells are effectively killed by genetically engineered effector cells expressingboth the CD95 ligand (CD178) and TRAIL (tumor necrosis factor-related apoptosis-inducing ligand [164]. The ex vivo phenotype of CLL is one of resistance to TRAIL[165]. However, two publications have shown that histone deacetylase inhibitionsensitizes CLL cells to the effects of TRAIL, and that the TRAIL receptor DR4, butnot DR5, is involved in cell killing [165, 166].

As mentioned above, CD40 has a prominent role in CLL pathogenesis. In termsof cell death regulation, CD40 signaling has been suggested to trigger major repro-gramming along the BCL-2 governed pathway to cell death [167–169]. In parallelto a pattern observed in CLL from lymph node tissue, CD40L triggers upregulationof Mcl-1 and A1, as well as Bcl-xL antiapoptotic proteins [168]. BCR signalinghas been reported to also regulate Mcl-1 [78]. This was suggested to be mediatedvia AKT signaling. AKT has also been shown to target BH3-only proteins BADand BIM for deactivation by phosphorylation [170, 171]. Interestingly the Akt sig-nal seems to be positively modified by TCL-1 overexpression, which is found in arelatively large proportion of CLL patients and has been proposed to be a conse-quence of dysregulation of the microRNAs, miR-29, and miR-181 [170, 172]. Inaddition, miR-15 and miR-16 have been proposed to be the basis of constitutiveBCL-2 overexpression in CLL [7, 170, 173]. Another reported signal that modifiescell death machinery is IL-21, which mediates apoptosis through upregulation ofBIM [174].

Role of the Microenvironment

Chronic lymphocytic leukemia cells maintain their capacity to respond to selectedexternal stimuli that confer a growth advantage and extended survival. In vitro,spontaneous apoptosis of CLL cells can be rescued by the co-culture with ‘nurse-like cells’ or stromal cells [175–179]. Adherent nurse-like cells are able to protectleukemic cells from spontaneous apoptosis [180]. Activated autologous T cellsco-cultured with CLL cells also prevent apoptosis [177, 178]; this action can besomehow replaced by T-cell derived cytokines (i.e. IL-4) and exposure to T-cell-related molecules (i.e., sCD40L) [178, 181]. In addition to rescuing leukemic cellsfrom apoptosis, CD40 stimulation can induce their proliferation [182, 183] and acti-vation as shown by upregulation of cell surface molecules (e.g., CD80, CD95), aswell induction of chemokine production (e.g., CCL-22/MDC, CCL-17/TARC) [184,185] and apoptosis regulators (e.g., survivin) [182].

Normal activated T cells are numerous in tissues involved by CLL where they aremainly located in proliferation centers [184] and intermingled with leukemic cellsthat express activation molecules and have increased proliferative activity [184].

280 D. Hoehn et al.

These findings suggest that T-cell-mediated activation is also occurring in vivo,resembling a sort of immune reaction, and reinforce the concept of a sustainedimmune stimulation in CLL.

The presence of prolymphocytes and paraimmunoblasts, the correlation betweensize and numbers of proliferation centers, and the lymphocyte doubling time [186]strongly indicate that proliferation centers are the reservoir of dividing leukemiccells. Therefore, the dogma that CLL is a static disease, resulting solely froman accumulation of apoptosis-resistant leukemic cells, needs to be reconsidered.Investigation of telomere length and telomerase activity [187, 188] as well as kinet-ics studies [189] indicate that CLL is a dynamic process composed of cells thatproliferate and die, often at appreciable levels. It seems clear that most proliferatingcells are found in tissues where CLL cells can exploit microenvironmental inter-actions in order to avoid apoptosis and acquire a growth advantage. Experimentalfindings suggest that CLL cells in tissues interact with activated T cells that influ-ence leukemic cell proliferation and provide short-term antiapoptotic support, whilestromal cells (and other accessory cells, e.g., nurse-like cells) provide long-termsupport that favors the extended survival and relentless CLL cell accumulation[177, 190]. These and other pieces of experimental evidence indicate that differ-ent microenvironmental components deliver fundamental and specific signals forthe maintenance and expansion of leukemic B cells at different time points in thenatural history of CLL. At the same time, CLL cells are active players in shapingthe microenvironment according to their needs, thanks to the production of selectedchemokines (i.e., CCL22/MDC and CCL-17/TARC) which recruit activated T lym-phocytes that will ensure provision of signals (e.g., IL-4 and CD40 ligation) favoringmalignant cell growth and survival. As CLL cells express CXCR4 [157], they candrift away from T cells toward the surrounding stromal cells which are the mainproducers of CXCL12/SDF-1 [191], the CXCR4-specific ligand, probably avoid-ing a T-cell-mediated limit to their expansion (e.g., via CD95–CD95L interactions)[161], and at the same time, taking advantage of the long-term support by stromalcells. All these data support the importance of the microenvironment in the naturalhistory of CLL.

Conclusions

Molecular characterization of the abnormalities in CLL has been and continues tobe important for a number of reasons. From the scientific view point these studieshelp to identify the genes and the mechanisms involved in hematopoiesis and thepathogenesis of CLL. Further progress may help to identify targets for rational drugdesign or gene therapy. From a more immediate clinical perspective, informationgleaned from these studies has improved the accuracy of diagnosis, helped to predicttherapeutic response, provided criteria for selecting high-risk patient groups whomay benefit from intensive but highly toxic chemotherapy protocols or bone marrowtransplantation, and aids in the detection of minimal residual disease before clinicalrelapse.


References

1. Mayr C, Speicher MR, Kofler DM, et al. Chromosomal translocations are associated withpoor prognosis in chronic lymphocytic leukemia. Blood. 2006;107(2):742–751.

2. Dicker F, Schnittger S, Haferlach T, Kern W, Schoch C. Immunostimulatory oligonucleotide-induced metaphase cytogenetics detect chromosomal aberrations in 80% of CLL patients: astudy of 132 CLL cases with correlation to FISH, IgVH status, and CD38 expression. Blood.2006;108(9):3152–3160.

3. Haferlach C, Dicker F, Schnittger S, Kern W, Haferlach T. Comprehensive genetic char-acterization of CLL: a study on 506 cases analysed with chromosome banding anal-ysis, interphase FISH, IgV(H) status and immunophenotyping. Leukemia. 2007;21(12):2442–2451.

4. Dohner H, Stilgenbauer S, Benner A, et al. Genomic aberrations and survival in chroniclymphocytic leukemia. N Engl J Med. 2000;343(26):1910–1916.

5. Calin GA, Ferracin M, Cimmino A, et al. A MicroRNA signature associated with prognosisand progression in chronic lymphocytic leukemia. N Engl J Med. 2005;353(17):1793–1801.

6. Calin GA, Dumitru CD, Shimizu M, et al. Frequent deletions and down-regulation of micro-RNA genes miR15 and miR16 at 13q14 in chronic lymphocytic leukemia. Proc Natl AcadSci USA. 2002;99(24):15524–15529.

7. Cimmino A, Calin GA, Fabbri M, et al. miR-15 and miR-16 induce apoptosis by targetingBCL2. Proc Natl Acad Sci USA. 2005;102(39):13944–13949.

8. Linsley PS, Schelter J, Burchard J, et al. Transcripts targeted by the microRNA-16 familycooperatively regulate cell cycle progression. Mol Cell Biol. 2007;27(6):2240–2252.

9. Raveche ES, Salerno E, Scaglione BJ, et al. Abnormal microRNA-16 locus with synteny tohuman 13q14 linked to CLL in NZB mice. Blood. 2007;109(12):5079–5086.

10. Kobayashi H, Espinosa R, 3rd, Fernald AA, et al. Analysis of deletions of the long arm ofchromosome 11 in hematologic malignancies with fluorescence in situ hybridization. GenesChromosomes Cancer. 1993;8(4):246–252.

11. James MR, Richard CW, 3rd, Schott JJ, et al. A radiation hybrid map of 506 STS markersspanning human chromosome 11. Nat Genet. 1994;8(1):70–76.

12. Barlow C, Hirotsune S, Paylor R, et al. Atm-deficient mice: a paradigm of ataxia telangiec-tasia. Cell. 1996;86(1):159–171.

13. Rotman G, Shiloh Y. ATM: from gene to function. Hum Mol Genet. 1998;7(10):1555–1563.

14. Schaffner C, Stilgenbauer S, Rappold GA, Dohner H, Lichter P. Somatic ATM muta-tions indicate a pathogenic role of ATM in B-cell chronic lymphocytic leukemia. Blood.1999;94(2):748–753.

15. Austen B, Powell JE, Alvi A, et al. Mutations in the ATM gene lead to impaired overall andtreatment-free survival that is independent of IGVH mutation status in patients with B-CLL.Blood. 2005;106(9):3175–3182.

16. Austen B, Skowronska A, Baker C, et al. Mutation status of the residual ATM allele isan important determinant of the cellular response to chemotherapy and survival in patientswith chronic lymphocytic leukemia containing an 11q deletion. J Clin Oncol. 2007;25(34):5448–5457.

17. Wessendorf S, Lichter P, Schwanen C, et al. Potential of chromosomal and matrix-basedcomparative genomic hybridization for molecular diagnostics in lymphomas. Ann Hematol.2001;80(Suppl 3):B35–B37.

18. Schwaenen C, Nessling M, Wessendorf S, et al. Automated array-based genomic profilingin chronic lymphocytic leukemia: development of a clinical tool and discovery of recurrentgenomic alterations. Proc Natl Acad Sci USA. 2004;101(4):1039–1044.

19. Bigoni R, Cuneo A, Roberti MG, et al. Chromosome aberrations in atypical chroniclymphocytic leukemia: a cytogenetic and interphase cytogenetic study. Leukemia.1997;11(11):1933–1940.

282 D. Hoehn et al.

20. Forconi F, Rinaldi A, Kwee I, et al. Genome-wide DNA analysis identifies recurrentimbalances predicting outcome in chronic lymphocytic leukaemia with 17p deletion. Br JHaematol. 2008;143(4):532–536.

21. Gaidano G, Ballerini P, Gong JZ, et al. p53 mutations in human lymphoid malignancies:association with Burkitt lymphoma and chronic lymphocytic leukemia. Proc Natl Acad SciUSA. 1991;88(12):5413–5417.

22. Trbusek M, Malcikova J, Smardova J, et al. Inactivation of p53 and deletion of ATM inB-CLL patients in relation to IgVH mutation status and previous treatment. Leukemia.2006;20(6):1159–1161.

23. Grever MR, Lucas DM, Dewald GW, et al. Comprehensive assessment of genetic and molec-ular features predicting outcome in patients with chronic lymphocytic leukemia: results fromthe US Intergroup Phase III Trial E2997. J Clin Oncol. 2007;25(7):799–804.

24. Pettitt AR, Sherrington PD, Stewart G, Cawley JC, Taylor AM, Stankovic T. p53 dysfunc-tion in B-cell chronic lymphocytic leukemia: inactivation of ATM as an alternative to TP53mutation. Blood. 2001;98(3):814–822.

25. He L, He X, Lowe SW, Hannon GJ. microRNAs join the p53 network–another piece in thetumour-suppression puzzle. Nat Rev Cancer. 2007;7(11):819–822.

26. Dijkstra MK, van Lom K, Tielemans D, et al. 17p13/TP53 deletion in B-CLL patients isassociated with microRNA-34a downregulation. Leukemia. 2009;23(3):625–627.

27. Auer RL, Riaz S, Cotter FE. The 13q and 11q B-cell chronic lymphocytic leukaemia-associated regions derive from a common ancestral region in the zebrafish. Br J Haematol.2007;137(5):443–453.

28. Bommer GT, Gerin I, Feng Y, et al. p53-mediated activation of miRNA34 candidate tumor-suppressor genes. Curr Biol. 2007;17(15):1298–1307.

29. Wei JS, Song YK, Durinck S, et al. The MYCN oncogene is a direct target of miR-34a.Oncogene. 2008;27(39):5204–5213.

30. Welch C, Chen Y, Stallings RL. MicroRNA-34a functions as a potential tumorsuppressor by inducing apoptosis in neuroblastoma cells. Oncogene. 2007;26(34):5017–5022.

31. Merup M, Moreno TC, Heyman M, et al. 6q deletions in acute lymphoblastic leukemia andnon-Hodgkin’s lymphomas. Blood. 1998;91(9):3397–3400.

32. Stilgenbauer S, Bullinger L, Benner A, et al. Incidence and clinical significance of 6qdeletions in B cell chronic lymphocytic leukemia. Leukemia. 1999;13(9):1331–1334.

33. Zhang Y, Matthiesen P, Harder S, et al. A 3-cM commonly deleted region in 6q21in leukemias and lymphomas delineated by fluorescence in situ hybridization. GenesChromosomes Cancer. 2000;27(1):52–58.

34. Bentz M, Dohner H, Huck K, et al. Comparative genomic hybridization in the investigationof myeloid leukemias. Genes Chromosomes Cancer. 1995;12(3):193–200.

35. Drexler HG, Janssen JW, Brenner MK, Hoffbrand AV, Bartram CR. Rapid expression of pro-tooncogenes c-fos and c-myc in B-chronic lymphocytic leukemia cells during differentiationinduced by phorbol ester and calcium ionophore. Blood. 1989;73(6):1656–1663.

36. Vallat L, Magdelenat H, Merle-Beral H, et al. The resistance of B-CLL cells to DNAdamage-induced apoptosis defined by DNA microarrays. Blood. 2003;101(11):4598–4606.

37. Huh YO, Lin KI, Vega F, et al. MYC translocation in chronic lymphocytic leukaemiais associated with increased prolymphocytes and a poor prognosis. Br J Haematol.2008;142(1):36–44.

38. Refaeli Y, Young RM, Turner BC, Duda J, Field KA, Bishop JM. The B cell antigen receptorand overexpression of MYC can cooperate in the genesis of B cell lymphomas. PLoS Biol.2008;6(6):e152.

39. McKeithan TW, Takimoto GS, Ohno H, et al. BCL3 rearrangements and t(14;19) in chroniclymphocytic leukemia and other B-cell malignancies: a molecular and cytogenetic study.Genes Chromosomes Cancer. 1997;20(1):64–72.


40. Schlette E, Rassidakis GZ, Canoz O, Medeiros LJ. Expression of bcl-3 in chronic lympho-cytic leukemia correlates with trisomy 12 and abnormalities of chromosome 19. Am J ClinPathol. 2005;123(3):465–471.

41. Huh YO, Abruzzo LV, Rassidakis GZ, et al. The t(14;19)(q32;q13)-positive smallB-cell leukaemia: a clinicopathologic and cytogenetic study of seven cases. Br J Haematol.2007;136(2):220–228.

42. Satterwhite E, Sonoki T, Willis TG, et al. The BCL11 gene family: involvement of BCL11Ain lymphoid malignancies. Blood. 2001;98(12):3413–3420.

43. Kuppers R, Sonoki T, Satterwhite E, et al. Lack of somatic hypermutation of IG V(H) genesin lymphoid malignancies with t(2;14)(p13;q32) translocation involving the BCL11A gene.Leukemia. 2002;16(5):937–939.

44. Yin CC, Lin KI, Ketterling RP, et al. Chronic lymphocytic leukemia With t(2;14)(p16;q32)involves the BCL11A and IgH genes and is associated with atypical morphologic featuresand unmutated IgVH genes. Am J Clin Pathol. 2009;131(5):663–670.

45. Cavazzini F, Hernandez JA, Gozzetti A, et al. Chromosome 14q32 translocations involvingthe immunoglobulin heavy chain locus in chronic lymphocytic leukaemia identify a diseasesubset with poor prognosis. Br J Haematol. 2008;142(4):529–537.

46. Smith LT, Otterson GA, Plass C. Unraveling the epigenetic code of cancer for therapy.Trends Genet. 2007;23(9):449–456.

47. Wahlfors J, Hiltunen H, Heinonen K, Hamalainen E, Alhonen L, Janne J. Genomichypomethylation in human chronic lymphocytic leukemia. Blood. 1992;80(8):2074–2080.

48. Smiraglia DJ, Plass C. The study of aberrant methylation in cancer via restriction landmarkgenomic scanning. Oncogene. 2002;21(35):5414–5426.

49. Rush LJ, Raval A, Funchain P, et al. Epigenetic profiling in chronic lymphocytic leukemiareveals novel methylation targets. Cancer Res. 2004;64(7):2424–2433.

50. Hanada M, Delia D, Aiello A, Stadtmauer E, Reed JC. bcl-2 gene hypomethylationand high-level expression in B-cell chronic lymphocytic leukemia. Blood. 1993;82(6):1820–1828.

51. Yuille MR, Condie A, Stone EM, et al. TCL1 is activated by chromosomal rearrangement orby hypomethylation. Genes Chromosomes Cancer. 2001;30(4):336–341.

52. Raval A, Lucas DM, Matkovic JJ, et al. TWIST2 demonstrates differential methylationin immunoglobulin variable heavy chain mutated and unmutated chronic lymphocyticleukemia. J Clin Oncol. 2005;23(17):3877–3885.

53. Raval A, Tanner SM, Byrd JC, et al. Downregulation of death-associated protein kinase 1(DAPK1) in chronic lymphocytic leukemia. Cell. 2007;129(5):879–890.

54. Corcoran M, Parker A, Orchard J, et al. ZAP-70 methylation status is associated withZAP-70 expression status in chronic lymphocytic leukemia. Haematologica. 2005;90(8):1078–1088.

55. Byrd JC, Shinn C, Ravi R, et al. Depsipeptide (FR901228): a novel therapeutic agent withselective, in vitro activity against human B-cell chronic lymphocytic leukemia cells. Blood.1999;94(4):1401–1408.

56. Gunn SR, Bolla AR, Barron LL, et al. Array CGH analysis of chronic lymphocytic leukemiareveals frequent cryptic monoallelic and biallelic deletions of chromosome 22q11 thatinclude the PRAME gene. Leuk Res. 2009;33(9):1276–1281.

57. Pfeifer D, Pantic M, Skatulla I, et al. Genome-wide analysis of DNA copy number changesand LOH in CLL using high-density SNP arrays. Blood. 2007;109(3):1202–1210.

58. Coll-Mulet L, Santidrian AF, Cosialls AM, et al. Multiplex ligation-dependent probeamplification for detection of genomic alterations in chronic lymphocytic leukaemia.Br J Haematol. 2008;142(5):793–801.

59. Rosenwald A, Alizadeh AA, Widhopf G, et al. Relation of gene expression phenotype toimmunoglobulin mutation genotype in B cell chronic lymphocytic leukemia. J Exp Med.2001;194(11):1639–1647.

284 D. Hoehn et al.

60. Klein U, Tu Y, Stolovitzky GA, et al. Gene expression profiling of B cell chronic lympho-cytic leukemia reveals a homogeneous phenotype related to memory B cells. J Exp Med.2001;194(11):1625–1638.

61. Haslinger C, Schweifer N, Stilgenbauer S, et al. Microarray gene expression profiling ofB-cell chronic lymphocytic leukemia subgroups defined by genomic aberrations and VHmutation status. J Clin Oncol. 2004;22(19):3937–3949.

62. Rosenwald A, Chuang EY, Davis RE, et al. Fludarabine treatment of patients withchronic lymphocytic leukemia induces a p53-dependent gene expression response. Blood.2004;104(5):1428–1434.

63. Vallat LD, Park Y, Li C, Gribben JG. Temporal genetic program following B-cell recep-tor cross-linking: altered balance between proliferation and death in healthy and malignantB cells. Blood. 2007;109(9):3989–3997.

64. Rodriguez A, Villuendas R, Yanez L, et al. Molecular heterogeneity in chronic lympho-cytic leukemia is dependent on BCR signaling: clinical correlation. Leukemia. 2007;21(9):1984–1991.

65. Dighiero G, Hamblin TJ. Chronic lymphocytic leukaemia. Lancet. 2008;371(9617):1017–1029.

66. Degan M, Bomben R, Bo MD, et al. Analysis of IgV gene mutations in B cell chronic lym-phocytic leukaemia according to antigen-driven selection identifies subgroups with differentprognosis and usage of the canonical somatic hypermutation machinery. Br J Haematol.2004;126(1):29–42.

67. Chang B, Casali P. The CDR1 sequences of a major proportion of human germlineIg VH genes are inherently susceptible to amino acid replacement. Immunol Today.1994;15(8):367–373.

68. Chang B, Casali P. A sequence analysis of human germline Ig VH and VL genes. The CDR1sof a major proportion of VH, but not VL, genes display a high inherent susceptibility toamino acid replacement. Ann NY Acad Sci. 1995;764:170–179.

69. Lossos IS, Tibshirani R, Narasimhan B, Levy R. The inference of antigen selection on Iggenes. J Immunol. 2000;165(9): 5122–5126.

70. Keating MJ, Chiorazzi N, Messmer B, et al. Biology and treatment of chronic lymphocyticleukemia. Hematology Am Soc Hematol Educ Program. 2003:153–175.

71. Bomben R, Dal Bo M, Zucchetto A, et al. Mutational status of IgV(H) genes in B-cellchronic lymphocytic leukemia and prognosis: percent mutations or antigen-driven selection?Leukemia. 2005;19(8):1490–1492.

72. Damle RN, Wasil T, Fais F, et al. Ig V gene mutation status and CD38 expression as novelprognostic indicators in chronic lymphocytic leukemia. Blood. 1999;94(6):1840–1847.

73. Hamblin TJ, Davis Z, Gardiner A, Oscier DG, Stevenson FK. Unmutated Ig V(H) genesare associated with a more aggressive form of chronic lymphocytic leukemia. Blood.1999;94(6):1848–1854.

74. Dreger P, Stilgenbauer S, Benner A, et al. The prognostic impact of autologous stem celltransplantation in patients with chronic lymphocytic leukemia: a risk-matched analysis basedon the VH gene mutational status. Blood. 2004;103(7):2850–2858.

75. Damle RN, Ghiotto F, Valetto A, et al. B-cell chronic lymphocytic leukemia cells expressa surface membrane phenotype of activated, antigen-experienced B lymphocytes. Blood.2002;99(11):4087–4093.

76. Mauerer K, Zahrieh D, Gorgun G, et al. Immunoglobulin gene segment usage, location andimmunogenicity in mutated and unmutated chronic lymphocytic leukaemia. Br J Haematol.2005;129(4):499–510.

77. Contri A, Brunati AM, Trentin L, et al. Chronic lymphocytic leukemia B cells containanomalous Lyn tyrosine kinase, a putative contribution to defective apoptosis. J Clin Invest.2005;115(2):369–378.

78. Longo PG, Laurenti L, Gobessi S, Sica S, Leone G, Efremov DG. The Akt/Mcl-1 pathway plays a prominent role in mediating antiapoptotic signals downstream of


the B-cell receptor in chronic lymphocytic leukemia B cells. Blood. 2008;111(2):846–855.

79. Fujimoto M, Poe JC, Jansen PJ, Sato S, Tedder TF. CD19 amplifies B lymphocyte sig-nal transduction by regulating Src-family protein tyrosine kinase activation. J Immunol.1999;162(12):7088–7094.

80. Alfarano A, Indraccolo S, Circosta P, et al. An alternatively spliced form of CD79b gene mayaccount for altered B-cell receptor expression in B-chronic lymphocytic leukemia. Blood.1999;93(7):2327–2335.

81. Lankester AC, van Schijndel GM, van der Schoot CE, van Oers MH, van Noesel CJ, van LierRA. Antigen receptor nonresponsiveness in chronic lymphocytic leukemia B cells. Blood.1995;86(3):1090–1097.

82. Gobessi S, Laurenti L, Longo PG, Sica S, Leone G, Efremov DG. ZAP-70 enhancesB-cell-receptor signaling despite absent or inefficient tyrosine kinase activation in chroniclymphocytic leukemia and lymphoma B cells. Blood. 2007;109(5):2032–2039.

83. Wiestner A, Rosenwald A, Barry TS, et al. ZAP-70 expression identifies a chronic lympho-cytic leukemia subtype with unmutated immunoglobulin genes, inferior clinical outcome,and distinct gene expression profile. Blood. 2003;101(12):4944–4951.

84. Durig J, Nuckel H, Cremer M, et al. ZAP-70 expression is a prognostic factor in chroniclymphocytic leukemia. Leukemia. 2003;17(12):2426–2434.

85. Mockridge CI, Potter KN, Wheatley I, Neville LA, Packham G, Stevenson FK. Reversibleanergy of sIgM-mediated signaling in the two subsets of CLL defined by VH-genemutational status. Blood. 2007;109(10):4424–4431.

86. Deaglio S, Vaisitti T, Aydin S, Ferrero E, Malavasi F. In-tandem insight from basic sciencecombined with clinical research: CD38 as both marker and key component of the patho-genetic network underlying chronic lymphocytic leukemia. Blood. 2006;108(4):1135–1144.

87. Cockayne DA, Muchamuel T, Grimaldi JC, et al. Mice deficient for the ecto-nicotinamideadenine dinucleotide glycohydrolase CD38 exhibit altered humoral immune responses.Blood. 1998;92(4):1324–1333.

88. Deaglio S, Vaisitti T, Billington R, et al. CD38/CD19: a lipid raft-dependent signalingcomplex in human B cells. Blood. 2007;109(12):5390–5398.

89. Deaglio S, Vaisitti T, Aydin S, et al. CD38 and ZAP-70 are functionally linked and markCLL cells with high migratory potential. Blood. 2007;110(12):4012–4021.

90. Muramatsu M, Kinoshita K, Fagarasan S, Yamada S, Shinkai Y, Honjo T. Class switchrecombination and hypermutation require activation-induced cytidine deaminase (AID), apotential RNA editing enzyme. Cell. 2000;102(5):553–563.

91. Pasqualucci L, Guglielmino R, Houldsworth J, et al. Expression of the AID protein in normaland neoplastic B cells. Blood. 2004;104(10):3318–3325.

92. Cerutti A, Zan H, Kim EC, et al. Ongoing in vivo immunoglobulin class switch DNA recom-bination in chronic lymphocytic leukemia B cells. J Immunol. 2002;169(11):6594–6603.

93. Oppezzo P, Vuillier F, Vasconcelos Y, et al. Chronic lymphocytic leukemia B cells expressingAID display dissociation between class switch recombination and somatic hypermutation.Blood. 2003;101(10):4029–4032.

94. Sahota SS, Davis Z, Hamblin TJ, Stevenson FK. Somatic mutation of bcl-6 genescan occur in the absence of V(H) mutations in chronic lymphocytic leukemia. Blood.2000;95(11):3534–3540.

95. Reiniger L, Bodor C, Bognar A, et al. Richter’s and prolymphocytic transforma-tion of chronic lymphocytic leukemia are associated with high mRNA expression ofactivation-induced cytidine deaminase and aberrant somatic hypermutation. Leukemia.2006;20(6):1089–1095.

96. Heintel D, Kroemer E, Kienle D, et al. High expression of activation-induced cyti-dine deaminase (AID) mRNA is associated with unmutated IGVH gene status andunfavourable cytogenetic aberrations in patients with chronic lymphocytic leukaemia.Leukemia. 2004;18(4):756–762.

286 D. Hoehn et al.

97. Efremov DG, Gobessi S, Longo PG. Signaling pathways activated by antigen-receptorengagement in chronic lymphocytic leukemia B-cells. Autoimmun Rev. 2007;7(2):102–108.

98. Allsup DJ, Kamiguti AS, Lin K, et al. B-cell receptor translocation to lipid rafts and asso-ciated signaling differ between prognostically important subgroups of chronic lymphocyticleukemia. Cancer Res. 2005;65(16):7328–7337.

99. Muzio M, Apollonio B, Scielzo C, et al. Constitutive activation of distinct BCR-signaling pathways in a subset of CLL patients: a molecular signature of anergy. Blood.2008;112(1):188–195.

100. Juliusson G, Gahrton G. Chromosome aberrations in B-cell chronic lymphocytic leukemia.Pathogenetic and clinical implications. Cancer Genet Cytogenet. 1990;45(2):143–160.

101. Widhopf GF, 2nd, Rassenti LZ, Toy TL, Gribben JG, Wierda WG, Kipps TJ. Chroniclymphocytic leukemia B cells of more than 1% of patients express virtually identicalimmunoglobulins. Blood. 2004;104(8):2499–2504.

102. Stamatopoulos K, Belessi C, Moreno C, et al. Over 20% of patients with chronic lym-phocytic leukemia carry stereotyped receptors: pathogenetic implications and clinicalcorrelations. Blood. 2007;109(1):259–270.

103. Bomben R, Dal Bo M, Capello D, et al. Comprehensive characterization of IGHV3-21-expressing B-cell chronic lymphocytic leukemia: an Italian multicenter study. Blood.2007;109(7):2989–2998.

104. Murray F, Darzentas N, Hadzidimitriou A, et al. Stereotyped patterns of somatic hypermu-tation in subsets of patients with chronic lymphocytic leukemia: implications for the role ofantigen selection in leukemogenesis. Blood. 2008;111(3):1524–1533.

105. Bomben R, Dal Bo M, Capello D, et al. Molecular and clinical features of chronic lympho-cytic leukaemia with stereotyped B cell receptors: results from an Italian multicentre study.Br J Haematol. 2009;144(4):492–506.

106. Tobin G, Thunberg U, Karlsson K, et al. Subsets with restricted immunoglobulin generearrangement features indicate a role for antigen selection in the development of chroniclymphocytic leukemia. Blood. 2004;104(9):2879–2885.

107. Messmer BT, Albesiano E, Efremov DG, et al. Multiple distinct sets of stereotyped antigenreceptors indicate a role for antigen in promoting chronic lymphocytic leukemia. J Exp Med.2004;200(4):519–525.

108. Widhopf GF, 2nd, Goldberg CJ, Toy TL, et al. Nonstochastic pairing of immunoglobulinheavy and light chains expressed by chronic lymphocytic leukemia B cells is predicated onthe heavy chain CDR3. Blood. 2008;111(6):3137–3144.

109. Messmer BT, Raphael BJ, Aerni SJ, et al. Computational identification of CDR3 sequencearchetypes among immunoglobulin sequences in chronic lymphocytic leukemia. Leuk Res.2009;33(3):368–376.

110. Lanemo Myhrinder A, Hellqvist E, Sidorova E, et al. A new perspective: molecular motifson oxidized LDL, apoptotic cells, and bacteria are targets for chronic lymphocytic leukemiaantibodies. Blood. 2008;111(7):3838–3848.

111. Herve M, Xu K, Ng YS, et al. Unmutated and mutated chronic lymphocytic leukemias derivefrom self-reactive B cell precursors despite expressing different antibody reactivity. J ClinInvest. 2005;115(6):1636–1643.

112. Ghiotto F, Fais F, Valetto A, et al. Remarkably similar antigen receptors among asubset of patients with chronic lymphocytic leukemia. J Clin Invest. 2004;113(7):1008–1016.

113. Athanasiadou A, Stamatopoulos K, Gaitatzi M, Stavroyianni N, Fassas A, AnagnostopoulosA. Recurrent cytogenetic findings in subsets of patients with chronic lymphocytic leukemiaexpressing IgG-switched stereotyped immunoglobulins. Haematologica. 2008;93(3):473–474.

114. Rossi D, Cerri M, Capello D, et al. Biological and clinical risk factors of chronic lym-phocytic leukaemia transformation to Richter syndrome. Br J Haematol. 2008;142(2):202–215.


115. Rossi D, Spina V, Cerri M, et al. Stereotyped B-cell receptor is an independent risk fac-tor of chronic lymphocytic leukemia transformation to Richter syndrome. Clin Cancer Res.2009;15(13):4415–4422.

116. Ghia P, Stamatopoulos K, Belessi C, et al. Geographic patterns and pathogenetic implica-tions of IGHV gene usage in chronic lymphocytic leukemia: the lesson of the IGHV3-21gene. Blood. 2005;105(4):1678–1685.

117. Tobin G, Thunberg U, Johnson A, et al. Chronic lymphocytic leukemias utilizing the VH3-21gene display highly restricted Vlambda2-14 gene use and homologous CDR3s: implicatingrecognition of a common antigen epitope. Blood. 2003;101(12):4952–4957.

118. Thorselius M, Krober A, Murray F, et al. Strikingly homologous immunoglobulingene rearrangements and poor outcome in VH3-21-using chronic lymphocytic leukemiapatients independent of geographic origin and mutational status. Blood. 2006;107(7):2889–2894.

119. Krober A, Bloehdorn J, Hafner S, et al. Additional genetic high-risk features such as11q deletion, 17p deletion, and V3-21 usage characterize discordance of ZAP-70 and VHmutation status in chronic lymphocytic leukemia. J Clin Oncol. 2006;24(6):969–975.

120. Zenz T, Dohner H, Stilgenbauer S. Genetics and risk-stratified approach to therapy in chroniclymphocytic leukemia. Best Pract Res Clin Haematol. 2007;20(3):439–453.

121. Bomben R, Dal Bo M, Benedetti D, et al. Expression of mutated IGHV3-23 genes has inde-pendent prognostic relevance in chronic lymphocytic leukemia and identifies a disease subsetwith peculiar molecular and biological features (Abstract). Haematologica. 2009;94(s2):141.

122. Hillson JL, Karr NS, Oppliger IR, Mannik M, Sasso EH. The structural basis ofgermline-encoded VH3 immunoglobulin binding to staphylococcal protein A. J Exp Med.1993;178(1):331–336.

123. Silverman GJ, Goodyear CS. A model B-cell superantigen and the immunobiology ofB lymphocytes. Clin Immunol. 2002;102(2):117–134.

124. Silverman GJ, Goodyear CS. Confounding B-cell defences: lessons from a staphylococcalsuperantigen. Nat Rev Immunol. 2006;6(6):465–475.

125. Potter KN, Hobby P, Klijn S, Stevenson FK, Sutton BJ. Evidence for involvement of ahydrophobic patch in framework region 1 of human V4-34-encoded Igs in recognition ofthe red blood cell I antigen. J Immunol. 2002;169(7):3777–3782.

126. Silberstein LE, George A, Durdik JM, Kipps TJ. The V4-34 encoded anti-i autoantibodiesrecognize a large subset of human and mouse B-cells. Blood Cells Mol Dis. 1996;22(2):126–138.

127. Pugh-Bernard AE, Silverman GJ, Cappione AJ, et al. Regulation of inherently autore-active VH4-34 B cells in the maintenance of human B cell tolerance. J Clin Invest.2001;108(7):1061–1070.

128. Fais F, Ghiotto F, Hashimoto S, et al. Chronic lymphocytic leukemia B cells expressrestricted sets of mutated and unmutated antigen receptors. J Clin Invest. 1998;102(8):1515–1525.

129. Jang YJ, Stollar BD. Anti-DNA antibodies: aspects of structure and pathogenicity. Cell MolLife Sci. 2003;60(2):309–320.

130. Borche L, Lim A, Binet JL, Dighiero G. Evidence that chronic lymphocytic leukemiaB lymphocytes are frequently committed to production of natural autoantibodies. Blood.1990;76(3):562–569.

131. Broker BM, Klajman A, Youinou P, et al. Chronic lymphocytic leukemic (CLL) cells secretemultispecific autoantibodies. J Autoimmun. 1988;1(5):469–481.

132. Sthoeger ZM, Wakai M, Tse DB, et al. Production of autoantibodies by CD5-expressing B lymphocytes from patients with chronic lymphocytic leukemia. J Exp Med.1989;169(1):255–268.

133. Riboldi P, Kasaian M, Mantovani L, Ikematsu H, Casali P. Natural Antibodies. New York:Hardwood Academic Publishers; 1992.

288 D. Hoehn et al.

134. Chu CC, Catera R, Hatzi K, et al. Chronic lymphocytic leukemia antibodies with a com-mon stereotypic rearrangement recognize nonmuscle myosin heavy chain IIA. Blood.2008;112(13):5122–5129.

135. Veldurthy A, Patz M, Hagist S, et al. The kinase inhibitor dasatinib induces apoptosis inchronic lymphocytic leukemia cells in vitro with preference for a subgroup of patients withunmutated IgVH genes. Blood. 2008;112(4):1443–1452.

136. Garg RJ, et al. E. Phase II study of dasatinib in patients with relapsed CLL (Abstract). Blood.2008;112:a4197.

137. Hallaert DY, al. E. c-Abl kinase inhibitors overcome CD40-mediated drug resistance in CLL(Abstract). Blood. 2007;110:905A.

138. Michie AM, Nakagawa R. Elucidating the role of protein kinase C in chronic lymphocyticleukaemia. Hematol Oncol. 2006;24(3):134–138.

139. Alkan S, Huang Q, Ergin M, et al. Survival role of protein kinase C (PKC) in chronic lym-phocytic leukemia and determination of isoform expression pattern and genes altered byPKC inhibition. Am J Hematol. 2005;79(2):97–106.

140. Abrams ST, Lakum T, Lin K, et al. B-cell receptor signaling in chronic lympho-cytic leukemia cells is regulated by overexpressed active protein kinase CbetaII. Blood.2007;109(3):1193–1201.

141. Barragan M, de Frias M, Iglesias-Serret D, et al. Regulation of Akt/PKB by phosphatidyli-nositol 3-kinase-dependent and -independent pathways in B-cell chronic lymphocyticleukemia cells: role of protein kinase C{beta}. J Leukoc Biol. 2006;80(6):1473–1479.

142. Ringshausen I, Oelsner M, Weick K, Bogner C, Peschel C, Decker T. Mechanismsof apoptosis-induction by rottlerin: therapeutic implications for B-CLL. Leukemia.2006;20(3):514–520.

143. Nakagawa R, Soh JW, Michie AM. Subversion of protein kinase C alpha signaling inhematopoietic progenitor cells results in the generation of a B-cell chronic lymphocyticleukemia-like population in vivo. Cancer Res. 2006;66(1):527–534.

144. Wojciechowski W, Li H, Marshall S, Dell’Agnola C, Espinoza-Delgado I. Enhanced expres-sion of CD20 in human tumor B cells is controlled through ERK-dependent mechanisms.J Immunol. 2005;174(12):7859–7868.

145. Jazirehi AR, Vega MI, Chatterjee D, Goodglick L, Bonavida B. Inhibition of the Raf-MEK1/2-ERK1/2 signaling pathway, Bcl-xL down-regulation, and chemosensitization ofnon-Hodgkin’s lymphoma B cells by Rituximab. Cancer Res. 2004;64(19):7117–7126.

146. Marzo I, Perez-Galan P, Giraldo P, et al. Farnesyltransferase inhibitor BMS-214662 inducesapoptosis in B-cell chronic lymphocytic leukemia cells. Leukemia. 2004;18(10):1599–1604.

147. Smal C, Lisart S, Maerevoet M, Ferrant A, Bontemps F, Van Den Neste E. Pharmacologicalinhibition of the MAPK/ERK pathway increases sensitivity to 2-chloro-2’-deoxyadenosine(CdA) in the B-cell leukemia cell line EHEB. Biochem Pharmacol. 2007;73(3):351–358.

148. Kato M, Fukuda H, Nonaka T, Imajoh-Ohmi S. Cleavage of nonmuscle myosin heavy chain-A during apoptosis in human Jurkat T cells. J Biochem. 2005;137(2):157–166.

149. Moisan E, Girard D. Cell surface expression of intermediate filament proteins vimentinand lamin B1 in human neutrophil spontaneous apoptosis. J Leukoc Biol. 2006;79(3):489–498.

150. Zhang M, Alicot EM, Chiu I, et al. Identification of the target self-antigens in reperfusioninjury. J Exp Med. 2006;203(1):141–152.

151. Catera R, Silverman GJ, Hatzi K, et al. Chronic lymphocytic leukemia cells recognizeconserved epitopes associated with apoptosis and oxidation. Mol Med. 2008;14(11–12):665–674.

152. Catera R, Hatzi K, Chu CC, al. E. Polyreactive monoclonal antibodies synthesized by someB-CLL cells recognize specific antigens on viable and apoptotic T cells (Abstract). Blood.2006;108:796a.

153. Chang MK, Bergmark C, Laurila A, et al. Monoclonal antibodies against oxidized low-density lipoprotein bind to apoptotic cells and inhibit their phagocytosis by elicited


macrophages: evidence that oxidation-specific epitopes mediate macrophage recognition.Proc Natl Acad Sci U S A. 1999;96(11):6353–6358.

154. Chou MY, Hartvigsen K, Hansen LF, et al. Oxidation-specific epitopes are important targetsof innate immunity. J Intern Med. 2008;263(5):479–488.

155. Shaw PX, Goodyear CS, Chang MK, Witztum JL, Silverman GJ. The autoreactivity ofanti-phosphorylcholine antibodies for atherosclerosis-associated neo-antigens and apoptoticcells. J Immunol. 2003;170(12):6151–6157.

156. Hatzi K, Catera R, Ferrarini M, al. E. B-cell chronic lymphocytic leukemia (B-CLL) cellsexpress antibodies reactive with antigenic epitopes expressed on the surface of commonbacteria (Abstract). Blood. 2006;108:12a.

157. Burger JA, Burger M, Kipps TJ. Chronic lymphocytic leukemia B cells express functionalCXCR4 chemokine receptors that mediate spontaneous migration beneath bone marrowstromal cells. Blood. 1999;94(11):3658–3667.

158. Carlo-Stella C, Lavazza C, Locatelli A, Vigano L, Gianni AM, Gianni L. Targeting TRAILagonistic receptors for cancer therapy. Clin Cancer Res. 2007;13(8):2313–2317.

159. Tinhofer I, Marschitz I, Kos M, et al. Differential sensitivity of CD4+ and CD8+ T lym-phocytes to the killing efficacy of Fas (Apo-1/CD95) ligand+ tumor cells in B chroniclymphocytic leukemia. Blood. 1998;91(11):4273–4281.

160. Proto-Siqueira R, Panepucci RA, Careta FP, et al. SAGE analysis demonstratesincreased expression of TOSO contributing to Fas-mediated resistance in CLL. Blood.2008;112(2):394–397.

161. Chu P, Deforce D, Pedersen IM, et al. Latent sensitivity to Fas-mediated apoptosis afterCD40 ligation may explain activity of CD154 gene therapy in chronic lymphocytic leukemia.Proc Natl Acad Sci U S A. 2002;99(6):3854–3859.

162. Anether G, Tinhofer I, Senfter M, Greil R. Tetrocarcin-A-induced ER stress medi-ates apoptosis in B-CLL cells via a Bcl-2–independent pathway. Blood. 2003;101(11):4561–4568.

163. Kater AP, Dicker F, Mangiola M, et al. Inhibitors of XIAP sensitize CD40-activatedchronic lymphocytic leukemia cells to CD95-mediated apoptosis. Blood. 2005;106(5):1742–1748.

164. Dicker F, Kater AP, Fukuda T, Kipps TJ. Fas-ligand (CD178) and TRAIL synergisti-cally induce apoptosis of CD40-activated chronic lymphocytic leukemia B cells. Blood.2005;105(8):3193–3198.

165. Inoue S, MacFarlane M, Harper N, Wheat LM, Dyer MJ, Cohen GM. Histone deacetylaseinhibitors potentiate TNF-related apoptosis-inducing ligand (TRAIL)-induced apoptosis inlymphoid malignancies. Cell Death Differ. 2004;11(Suppl 2):S193–206.

166. MacFarlane M, Inoue S, Kohlhaas SL, et al. Chronic lymphocytic leukemic cells exhibitapoptotic signaling via TRAIL-R1. Cell Death Differ. 2005;12(7):773–782.

167. Hallaert DY, Jaspers A, van Noesel CJ, van Oers MH, Kater AP, Eldering E. c-Abl kinaseinhibitors overcome CD40-mediated drug resistance in CLL: implications for therapeutictargeting of chemoresistant niches. Blood. 2008;112(13):5141–5149.

168. Smit LA, Hallaert DY, Spijker R, et al. Differential Noxa/Mcl-1 balance in peripheral versuslymph node chronic lymphocytic leukemia cells correlates with survival capacity. Blood.2007;109(4):1660–1668.

169. Willimott S, Baou M, Naresh K, Wagner SD. CD154 induces a switch in pro-survivalBcl-2 family members in chronic lymphocytic leukaemia. Br J Haematol. 2007;138(6):721–732.

170. Datta SR, Dudek H, Tao X, et al. Akt phosphorylation of BAD couples survival signals tothe cell-intrinsic death machinery. Cell. 1997;91(2):231–241.

171. Qi XJ, Wildey GM, Howe PH. Evidence that Ser87 of BimEL is phosphorylated by Akt andregulates BimEL apoptotic function. J Biol Chem. 2006;281(2):813–823.

172. Pekarsky Y, Santanam U, Cimmino A, et al. Tcl1 expression in chronic lymphocyticleukemia is regulated by miR-29 and miR-181. Cancer Res. 2006;66(24):11590–11593.

290 D. Hoehn et al.

173. Calin GA, Cimmino A, Fabbri M, et al. MiR-15a and miR-16-1 cluster functions in humanleukemia. Proc Natl Acad Sci U S A. 2008;105(13):5166–5171.

174. Gowda A, Roda J, Hussain SR, et al. IL-21 mediates apoptosis through up-regulation ofthe BH3 family member BIM and enhances both direct and antibody-dependent cellularcytotoxicity in primary chronic lymphocytic leukemia cells in vitro. Blood. 2008;111(9):4723–4730.

175. Lagneaux L, Delforge A, Bron D, De Bruyn C, Stryckmans P. Chronic lymphocytic leukemicB cells but not normal B cells are rescued from apoptosis by contact with normal bonemarrow stromal cells. Blood. 1998;91(7):2387–2396.

176. Panayiotidis P, Jones D, Ganeshaguru K, Foroni L, Hoffbrand AV. Human bone marrowstromal cells prevent apoptosis and support the survival of chronic lymphocytic leukaemiacells in vitro. Br J Haematol. 1996;92(1):97–103.

177. Ghia P, Circosta P, Scielzo C, et al. Differential effects on CLL cell survival exertedby different microenvironmental elements. Curr Top Microbiol Immunol. 2005;294:135–145.

178. Ranheim EA, Kipps TJ. Activated T cells induce expression of B7/BB1 on nor-mal or leukemic B cells through a CD40-dependent signal. J Exp Med. 1993;177(4):925–935.

179. Granziero L, Circosta P, Scielzo C, et al. CD100/Plexin-B1 interactions sustain proliferationand survival of normal and leukemic CD5+ B lymphocytes. Blood. 2003;101(5):1962–1969.

180. Burger JA, Tsukada N, Burger M, Zvaifler NJ, Dell’Aquila M, Kipps TJ. Blood-derivednurse-like cells protect chronic lymphocytic leukemia B cells from spontaneous apoptosisthrough stromal cell-derived factor-1. Blood. 2000;96(8):2655–2663.

181. Buske C, Gogowski G, Schreiber K, Rave-Frank M, Hiddemann W, Wormann B.Stimulation of B-chronic lymphocytic leukemia cells by murine fibroblasts, IL-4, anti-CD40antibodies, and the soluble CD40 ligand. Exp Hematol. 1997;25(4):329–337.

182. Granziero L, Ghia P, Circosta P, et al. Survivin is expressed on CD40 stimulation andinterfaces proliferation and apoptosis in B-cell chronic lymphocytic leukemia. Blood.2001;97(9):2777–2783.

183. Patten PE, Buggins AG, Richards J, et al. CD38 expression in chronic lymphocytic leukemiais regulated by the tumor microenvironment. Blood. 2008;111(10):5173–5181.

184. Ghia P, Strola G, Granziero L, et al. Chronic lymphocytic leukemia B cells are endowedwith the capacity to attract CD4+, CD40L+ T cells by producing CCL22. Eur J Immunol.2002;32(5):1403–1413.

185. Yellin MJ, Sinning J, Covey LR, et al. T lymphocyte T cell-B cell-activatingmolecule/CD40-L molecules induce normal B cells or chronic lymphocytic leukemiaB cells to express CD80 (B7/BB-1) and enhance their costimulatory activity. J Immunol.1994;153(2):666–674.

186. Pileri SA, Ascani S, Sabattini E, et al. The pathologist’s view point. Part I–indolentlymphomas. Haematologica. 2000;85(12):1291–1307.

187. Damle RN, Batliwalla FM, Ghiotto F, et al. Telomere length and telomerase activity delin-eate distinctive replicative features of the B-CLL subgroups defined by immunoglobulinV gene mutations. Blood. 2004;103(2):375–382.

188. Terrin L, Trentin L, Degan M, et al. Telomerase expression in B-cell chronic lymphocyticleukemia predicts survival and delineates subgroups of patients with the same igVH mutationstatus and different outcome. Leukemia. 2007;21(5):965–972.

189. Messmer BT, Messmer D, Allen SL, et al. In vivo measurements document the dynamiccellular kinetics of chronic lymphocytic leukemia B cells. J Clin Invest. 2005;115(3):755–764.

190. Nishio M, Endo T, Tsukada N, et al. Nurselike cells express BAFF and APRIL, which canpromote survival of chronic lymphocytic leukemia cells via a paracrine pathway distinctfrom that of SDF-1alpha. Blood. 2005;106(3):1012–1020.


191. Bleul CC, Fuhlbrigge RC, Casasnovas JM, Aiuti A, Springer TA. A highly effica-cious lymphocyte chemoattractant, stromal cell-derived factor 1 (SDF-1). J Exp Med.1996;184(3):1101–1109.

192. Maciejewski JP, Mufti GJ. Whole genome scanning as a cytogenetic tool in hematologicmalignancies. Blood. 2008;112:965–974.

193. Tobin G, Rosenquist R. Prognostic usage of V(H) gene mutation status and its surrogatemarkers and the role of antigen selection in chronic lymphocytic leukemia. Med Oncol.2005;22:217–228.

Chapter 9Targeted Therapy in Hematologic Malignancies

Barbara Zehnbauer and Mona Nasser

Keywords Tumor · Genetics · Therapy · Targeted · Chemotherapy ·Molecular · Cancer · Therapies · Antibodies · Anti-cancer · Immune ·Cytotoxicity · Cytotoxic · Tumor · Monoclonal · Antibodies · Human ·Therapeutic · Chimeric · Resistant · Treatment · Trials · Kinase · Inhibitor · Kinase ·Cancer · JAK2 · Activation · Molecule · Inhibitors · Imatinib · CML · Binding ·Resistance · Mutation · Nilotinib · Dasatinib · ATRA · Retinoic · Repression · APL ·AML · Mutations · Classification · Target · Angiogenesis · Thalidomide · Immuno-modulatory · Apoptotic · Relapsed · Vaccine · Immune · Virus · Membrane ·Targeted · Agents · Inhibitor · Antitumor

Introduction

The utility of tumor-specific molecular genetic signatures in the clinical care ofpatients with cancer has been well documented. Many diagnostic criteria of diseasecategories and subtypes have been revised to include defining genetic signaturessuch as the BCR–ABL1 fusion protein produced as a result of the formation of thePhiladelphia chromosome, a translocation between chromosomes 9 and 22. This isthe hallmark diagnostic finding in more than 95% of patients with chronic myeloge-nous leukemia (CML). Genetic markers have also been recognized as prognosticindicators of likelihood of response to treatment or disease progression. BCR–ABL1is an indicator of poor prognosis in some pediatric patients with ALL because thismarker is associated with poor response to therapy and a poor duration of response.Studies from many clinical centers have also validated the utility of using these sig-natures to stratify patients to the most effective treatment regimens based on historicoutcome-based studies. These therapies have ranged from chemotherapy and stemcell transplantation to immunotherapy and small molecule drugs. In addition, the

B. Zehnbauer (B)Division of Laboratory Systems, Laboratory Practice Evaluation and Genomics Branch, Centersfor Disease Control and Prevention, 1600 Clifton Road NE, Mail Stop G23, Atlanta, GA 30329,USAe-mail: [email protected]


294 B. Zehnbauer and M. Nasser

defining molecular signature may be used to track the efficacy of the treatment bymonitoring levels of residual cancer cells throughout the course of treatment. Nowanti-cancer therapies may specifically target the tumor-specific gene products withinthe cancer cells, bypassing normal cells of the same lineage. Only cancers that bearthe genetic mutations will respond to these targeted therapies, requiring specific,sensitive, and timely molecular diagnostic tools to aid the oncologist in character-izing the cancers of patients who may be candidates for these genetically designeddrugs. In this chapter we will review the categories of targeted therapies that are inuse to treat hematologic malignancies including leukemia, lymphoma, and multiplemyeloma.

Targeted Therapy: Definitions and Classification

The concept of a “magic bullet” therapy to selectively target specific cells to curedisease was originally introduced by Paul Ehrlich for microbial infections in 1891.Nearly a century later, this concept was applied in the anti-cancer therapeutic field.Recent molecular and genetic advances have led to the identification of many ofthe molecules that are overexpressed in cancer cells, or involved in the processof transformation of normal cells into cancer cells (carcinogenesis) [1–3]. As aresult, a new generation of anti-cancer drugs, called “targeted therapy,” has beendeveloped to target and specifically act on these molecular targets. These strategieshave been termed “molecular targeted therapy” and are integral components of cur-rent personalized medicine. These targeted treatments hold the promise of not onlymore effective anti-cancer action but also reduced toxicities and adverse side effectsbecause normal cells are not targeted [4, 5].

Three definitions for targeted cancer therapies have generally been applied [6].The first definition describes targeted therapy as the use of any drug that preciselyfocuses on a distinct molecular target and/or interferes with a specific signalingpathway resulting in the prevention of cancer growth and progression. The sec-ond usage expands the definition of targeted therapy to include antibodies joinedwith cytotoxic agents, cytotoxic radioisotopes, and cellular poisons to selectivelytarget cancer cells. The third, adopted by the Food and Drug Administration (FDA),defines targeted therapy as a drug with an approved label that can be administeredonly after an approved diagnostic test has been performed to establish a patient’seligibility for that drug. In other words, the label of the drug has a reference tothe prerequisite of testing for a valid molecular biomarker. For example, patientswith metastatic colorectal cancer are candidates for cetuximab if overexpressionof EGFR (epithelial growth factor receptor) can be demonstrated with a diagnos-tic test required to qualify for the drug. Both the test and the drug are considereda combination product as they meet the following criterion outlined by the FDA:“Any investigational drug, device or biological product packaged separately thataccording to its proposed labeling is for use only with another individually specified

9 Targeted Therapy in Hematologic Malignancies 295

investigational drug, device or biological product where both are required to achievethe intended use, indication or effect” [7].

Conventional cytotoxic chemotherapeutic agents act mainly on rapidly prolifer-ating cells by inhibiting cell division or inducing damage to the DNA replicated inthese cells. Inadequate specificity is the major drawback to this mechanism becauserapid proliferation is a shared feature of both cancer and some normal cells. Thus,toxicity to actively dividing normal cells is a common side effect to this type oftherapy. In contrast, targeted therapy achieves specificity by focusing on distinctcharacteristics and specific pathways of cancer cells [8, 9].

The main advantage of targeted therapy over traditional cancer therapies, such aschemotherapy and radiation, is the ability to specifically target cancer cells withoutharming or destroying normal cells [4]. The specificity requires a molecule that iscritical to the malignant phenotype (expression) but at the same time is not expressedin vital organs or tissues. The availability of an analytical testing method for quanti-fying the molecular target of the cancer cells is an essential factor for identificationof the population of patients who are eligible for this therapy based on the like-lihood of a favorable response to treatment. Moreover, an ideal molecular targethas a pivotal prognostic role; interfering with that target provides a significant andmeasurable impact on the clinical course of the disease [6]. The ideal target maybe difficult to define; BCR–ABL1 tyrosine kinase has been the best example of atumor-specific signature targeted by a specific small molecule inhibitor, imatinibmesylate. This hybrid protein in patients with chronic myeloid leukemia (CML) iscrucial to the malignant process [10].

There are several methods to categorize targeted therapies in the literature. TheNational Cancer Institute classification is the one most commonly used [5]. Itcategorizes targeted therapeutic agents as follows:

1. Therapeutic monoclonal antibodies2. Small molecule drugs3. Angiogenesis inhibitors4. Apoptosis-inducing drugs5. Cancer vaccines6. Gene therapy

Small molecule drugs and therapeutic monoclonal antibodies are considered tobe the main categories of targeted therapy for hematologic malignancies [9, 11].However, there may be some overlap between different categories because somedrugs may have more than one pharmacodynamic property. Medical care providersmay view the third and fourth categories as subgroups of the first and second. Forexample, bevacizumab is considered to be a monoclonal antibody that has anti-angiogenic properties, while bortezomib may be classified as a small molecule drugthat induces apoptosis [4]. Some agents of targeted therapy have been approved bythe FDA, while many others are still in preclinical testing or clinical trials [5, 8].


Of 20 anti-cancer therapies approved by the FDA during 2000–2008, 15 have beentargeted agents [9, 12].

Therapeutic Monoclonal Antibodies in Targeted Therapy

Monoclonal antibodies (mABs) used in targeted anti-cancer therapy are antibodiesproduced by genetic engineering to be directed against specific antigens that areexpressed primarily or solely by malignant cells [8]. They are water-soluble proteinsof high molecular weight (150,000 Da) which are unable to penetrate cells; thusthey target extracellular components of the cells, such as an extracellular bindingdomain of receptor protein molecules [9, 11]. Monoclonal antibodies are usuallymore efficient in targeting hematologic malignancies than solid tumors due to theease of accessing hematopoietic cells and the low diffusion efficiency through solidtumor tissues to individual cells [1, 8, 11]. These agents are usually administeredby intravenous injections to avoid denaturation of the protein content, which wouldoccur in the gastrointestinal tract following oral intake [9].

The basic antibody molecule is composed of four immunoglobulin polypeptidechains: two identical light chains and two identical heavy chains. The light chainseach have two regions – one constant region domain (CL) and one variable regiondomain (VL) – while the heavy chain has four regions – three constant regiondomains (CH1, CH2, CH3) and one variable region domain (VH). The variableregion domains of the fragment antigen-binding (FAB) region have highly vari-able amino acid sequences and define the complementarity-determining regions(CDRs) which are responsible for antigen-binding specificity [11]. The fragmentcrystallizable region (Fc region) is derived from the constant domains and formsthe tail part of an immunoglobulin protein [8]. The constant region has moreconserved amino acid content and encodes the protein region responsible for theinduction of an immune-mediated response upon binding of the Fc receptors inthe effector cells to proteins of the complement complex [1, 9, 11, 13, 14]. Mosttherapeutic monoclonal antibodies with a human framework are of IgG1 sub-class due to their long half-life and the capability to induce immune responsesmediated through their Fc regions such as antibody-dependent cellular cytotox-icity (ADCC), complement-dependent cell-mediated cytotoxicity (CDCC), andcomplement-dependent cytotoxicity (CDC) [1, 11].

Therapeutic mABs may execute their targeted anti-cancer action via three differ-ent approaches: direct inhibitory effect on the tumor, induction of immune-mediatedmechanisms, and delivery of cytotoxic materials to the tumor cells [1, 9, 11, 15]. Inthe first approach, mABs act directly by binding to certain receptors or ligands,usually on the cell surface of the tumor, to inhibit signal transduction pathwaysthat are crucial to cancer cell survival or proliferation. This direct antigen–antibodybinding mechanism may exert its effect by blocking the binding site of the lig-and, inhibiting receptor heterodimerization, or promoting receptor internalization[16, 17]. Inhibiting the molecular signaling cascade can lead to the arrest of thecell cycle, the reduction of angiogenesis [18], the inhibition of DNA repair [19],


or the induction of apoptosis [11]. Monoclonal antibodies targeting proteins of theepithelial growth factor receptor (EGFR) family are considered the most effectivesignal transduction inhibitors [1].

In the second approach, mABs induce immune-mediated responses in whichmABs tag the target tumor cells which the immune system components subse-quently attack and destroy [8, 14]. Binding of a mAB to its target on the tumorcell surface leads to the interaction between the antibody Fc region and the Fcreceptor on the immune effector cells (macrophages and natural killer cells).Theeffector cells mediate phagocytosis and lysis of the antibody-coated tumor cellsvia the so-called antibody-dependent cellular cytotoxicity (ADCC) mechanism [11].Moreover, activation of the complement cascade can be initiated by binding mAB tothe tumor cell. This cascade reaction either induces complement-dependent cytotox-icity (CDD) during which a membrane attack complex (MAC) leads to tumor celllysis or induces complement-dependent cell-mediated cytotoxicity (CDCC) throughthe generation of C3b, an opsonin which enhances phagocytosis of the tumor cellsby the effector cells [11, 20].

In the third approach, mABs are conjugated with cytotoxic agents such as tox-ins, radioisotopes, drugs, or cytokines to selectively transport these agents to thetargeted cancer cells [1, 11]. Though mABs may show high selectivity toward aspecific target, they may lack sufficient cytotoxicity. Binding of a mAB to a certaintarget may not guarantee a powerful killing activity by itself. On the contrary, thecytotoxic agents frequently have potent cell killing effects but lack adequate selec-tivity. The conjugation process may be perceived as a mechanism to either providethe required cytotoxicity to the mAB or the selectivity of the mAB to the cytotoxicagents. This combined transport mechanism facilitates the use of potent cytotoxicagents by avoiding the potential high toxicity and adverse side effects on normaltissue [21]. Conjugation with the mAB prolongs the serum half-life of the conjugatecytotoxin and “detoxifies” it. A cytotoxic drug remains joined to the mAB duringcirculation, which keeps it biologically inactive, i.e., “non toxic” during transit. Itresumes its cytotoxicity on separation from the mAB within the cell.

Upon attachment to the target cell, internalization of the conjugate takes placeby endocytosis. After internalization, cleavage of the link between the drug andthe mAB occurs in the lysosomes to discharge the cytotoxic drug in its activeform within the cell. Internalization is important for cytotoxic drugs and toxinsto destroy the cells and spread throughout the tumor [21]. This is not a limita-tion of radioisotope conjugates because isotopic decay particles may penetrate 1 cmto destroy cells along the particle path [14]. Though considered to be an attributefor unconjugated antibodies, prolonged circulation half-life of the radioconjugatesis undesirable because prolonged exposure of the bone marrow to radiation mayproduce severe myelosuppression [22].

In 1975, Georges Kohler and Cesar Milstein developed the somatic hybridiza-tion technique, or hybridoma technology [23], by fusing malignant myeloma cellswith antibody-producing B cells to generate specific monoclonal antibodies. Thistechnique produced clonogenic hybrid cells with immortalized replication and pro-duction of specific antibodies to enable mass production of identical antibodies


that target selected antigens. Initially, monoclonal antibodies were derived frommouse cell hybridomas which hindered their use in the therapeutic field due to theirimmunogenicity [24], clearance by human anti-mouse antibodies, short half–life,and reduced ability to induce an immune effector action [9, 11, 25]. Advances inrecombinant DNA technology and genetic engineering facilitated the subsequentproduction of chimeric, humanized, and human types of monoclonal antibodieswith a greater fraction of human immunoglobulin component, consequent lowerimmunogenicity, longer half-life, and increased efficacy [26]. Both chimeric andhumanized types of mABs are derived from two species, human and murine,in different proportions. In the chimeric antibodies, the constant regions of theimmunoglobulin are derived from human origin and represent 75% of the mABprotein content, while the variable regions are of murine origin [24]. The suffix ofthe name of the mAB-drug indicates the type of monoclonal antibody by denotingthe species of origin as follows: mumab (human), -zumab (humanized), -momab(murine), -ximab (chimeric) [9].

In 1997, the FDA approved rituximab (a chimeric antibody) as the first ther-apeutic monoclonal antibody to target cancer, which achieved a radical changein the treatment of non-Hodgkin’s lymphoma (NHL) [11, 14, 27]. Subsequently,several chimeric mABs have been approved by the FDA for therapeutic purposes(Table 9.1).

Although chimeric mABs have notably decreased immunogenicity and thehuman anti-mouse antibody reactions, these adverse side effects have not beenentirely eliminated [28]. There remains a need for further reduction of the immuno-genicity to accomplish tolerance to the increased and multiple dosing required tocombat some cancer targets [29, 30]. Increasing the human component in mABsnot only decreases the xenogenicity and immunogenicity but also enhances theimmune function mediated by the mAB through effective interaction betweenthe human Fc region and the Fc receptor on the immune effector cells [14]. Thehumanized type of mAB has been developed with 95% human component withthe complementarity-determining regions (CDRs) remaining as the only murine-derived fraction [11, 31]. Specific targeting is maintained in both the chimericand the humanized mABs by the murine variable region fragment. Ultimately,development of human monoclonal antibodies with 100% human component hasbeen achieved (Table 9.1) [1, 6, 11, 14, 32].

Specific therapeutic monoclonal antibodies approved by the FDA for treatmentof various hematologic malignancies are summarized in Table 9.1. General featuresof most of these mABs are as follows:

1. Most of these therapies target lymphoid malignancies.2. They target cell surface molecules that are expressed on both normal and

neoplastic cells.3. The target molecule is not expressed on hematopoietic stem cells.4. All are administered intravenously.5. They may be used in conjunction with chemotherapeutic agents without

increased toxicity.


Tabl

e9.

1T

hera

peut

icm

onoc

lona

lant

ibod

ies:

appr

oved

byth

eFD

Afo

rhe

mat

olog

icm

alig

nanc

ies

orin

vest

igat

iona

lclin

ical

tria

ls

Gen

eric

nam

eof

the

agen

tT

rade

nam

eTa

rget

Ant

ibod

yM

echa

nism

ofac

tion

Hem

atol

ogic

mal

igna

ncy

Adv

erse

effe

cts

Mon

itori

ngFD

Aap

prov

al

Ritu

xim

abR

ituxa

nC

D20

Chi

mer

icIg

G1

Imm

une-

med

iate

dcy

toto

xici

ty

Non

-Hod

gkin

’sly

mph

oma

(NH

L)

Rhe

umat

oid

arth

ritis

;ly

mph

ocyt

open

ia;H

BV

reac

tivat

ion;

seve

rem

ucoc

utan

eous

reac

tions

Liv

eva

ccin

essh

ould

beav

oide

d

CB

C;s

igns

ofac

tive

HB

Vin

fect

ion

orhe

patit

isin

patie

nts

who

are

HB

Vca

rrie

rs

1997

Gem

tuzu

mab

ozog

amic

inM

ylot

arg

CD

33H

uman

ized

IgG

4C

onju

gatio

nw

ithto

xin

calic

heam

icin

;cl

eave

sD

NA

;ca

spas

e-m

edia

ted

apop

tosi

s

Acu

tem

yelo

idle

ukem

ia(A

ML

)

Seve

rem

yelo

supp

ress

ion;

hepa

toto

xici

ty

CB

C;e

lect

roly

tele

vels

;liv

erch

emis

trie

s

2000

Ale

mtu

zum

abC

ampa

thC

D52

Hum

aniz

edIg

G1

Imm

une-

med

iate

dcy

toto

xici

tyC

hron

icly

mph

ocyt

icle

ukem

ia(C

LL

)

Hem

atol

ogic

toxi

city

oppo

rtun

istic

infe

ctio

ns;r

ash

CB

C;C

D4

coun

ts;

avoi

dliv

eva

ccin

es;H

erpe

san

dP

neum

ocys

tis

prop

hyla

xis

reco

mm

ende

d

2001

90Y

-Ibr

itum

omab

tiuxe

tan

Zev

alin

CD

20M

urin

eIg

G1

Con

juga

tion

with

yttr

ium

-90

radi

oiso

tope

NH

LSe

vere

,pro

long

edm

yelo

supp

ress

ion;

seve

rem

ucoc

utan

eous

reac

tions

;ris

kof

seco

ndar

ym

alig

nanc

ies

(e.g

.,ac

ute

mye

loid

leuk

emia

);ra

diat

ion

safe

typr

ecau

tions

requ

ired

for

1w

eek

afte

rad

min

istr

atio

n

CB

C;p

retr

eatm

ent

antib

ody

titer

sin

patie

nts

who

have

rece

ived

othe

rm

urin

e-ba

sed

radi

oim

mun

othe

r-ap

yre

gim

ens

2002


Tabl

e9.

1(c

ontin

ued)

Gen

eric

nam

eof

the

agen

tT

rade

nam

eTa

rget

Ant

ibod

yM

echa

nism

ofac

tion

Hem

atol

ogic

mal

igna

ncy

Adv

erse

effe

cts

Mon

itori

ngFD

Aap

prov

al

[131I]

Tosi

tum

omab

Bex

xar

CD

20M

urin

eIg

G2

Con

juga

tion

with

iodi

ne-1

31ra

dioi

soto

pe

NH

LH

ypot

hyro

idis

m;

seve

re,p

rolo

nged

mye

losu

ppre

ssio

n;na

usea

and

vom

iting

;se

cond

ary

mal

igna

ncie

s(e

.g.,

acut

em

yelo

idle

ukem

ia);

radi

atio

nsa

fety

prec

autio

nsre

quir

edfo

r1

wee

kaf

ter

adm

inis

trat

ion

CB

C;t

hyro

idfu

nctio

nte

sts;

pret

reat

men

tan

tibod

ytit

ers

inpa

tient

sw

hoha

vere

ceiv

edot

her

mur

ine-

base

dra

dioi

mm

unot

her-

apy

regi

men

s

2003

Ofa

tum

umab

Arz

erra

CD

20H

uman

ized

IgG

1(κ

,ka

ppa)

Imm

une

effe

ctor

B-c

elll

ysis

(CD

Can

dA

DC

C)

Ref

ract

ory

B-c

ellC

LL

Cyt

open

ia,p

neum

onia

,fe

ver,

coug

h,di

arrh

ea,

fatig

ue,d

yspn

ea,r

ash,

naus

ea

CB

C;m

ultif

ocal

leuk

oenc

epha

lopa

-th

y;ri

skfo

rhe

patit

isB

viru

sre

activ

atio

n

2009

Lum

ilixi

mab

NA

CD

23N

AIm

mun

e-m

edia

ted

cyto

toxi

city

B-c

ellC

LL

Uns

peci

fied

Uns

peci

fied

Phas

eII

clin

ical

tria

lsE

prat

uzum

abN

AC

D22

Hum

aniz

edIg

G1(

κ,

kapp

a)

Imm

une-

med

iate

dcy

toto

xici

tyR

elap

sed

orre

frac

tory

,in

dole

ntB

-cel

lN

HL

Uns

peci

fied

Uns

peci

fied

Clin

ical

tria

ls


Tabl

e9.

1(c

ontin

ued)

Gen

eric

nam

eof

the

agen

tT

rade

nam

eTa

rget

Ant

ibod

yM

echa

nism

ofac

tion

Hem

atol

ogic

mal

igna

ncy

Adv

erse

effe

cts

Mon

itori

ngFD

Aap

prov

al

AM

E-1

33,

GA

101,

veltu

zum

ab

NA

CD

20Fc

γ(g

amm

a)R

IIIa

;ac

tivat

esna

tura

lki

ller

(NK

)ce

lls;

FcR

γ(g

amm

a)II

I,A

DC

C;

casp

ase-

inde

pend

ent

apop

tosi

s

CL

LU

nspe

cifie

dU

nspe

cifie

dC

linic

altr

ials

orpr

eclin

ical

inve

stig

atio

ns

Apo

lizum

abR

emito

gen

HL

A-

DR

ß(b

eta)

-1D

10

Hum

aniz

edIm

mun

eef

fect

orB

-cel

llys

is(C

DC

and

AD

CC

)

Rel

apse

dor

refr

acto

ryC

LL

Uns

peci

fied

Uns

peci

fied

Phas

eII

clin

ical

tria

ls

HC

D12

2N

AC

D40

Hum

anA

DC

CC

LL

Uns

peci

fied

Uns

peci

fied

Phas

e1

clin

ical

tria

lsM

DX

-141

1N

AC

D70

Hum

anA

DC

Cof

tum

orne

cros

isfa

ctor

(TN

F)-e

xpre

ssin

gce

lls

CL

LU

nspe

cifie

dU

nspe

cifie

dPh

ase

1cl

inic

altr

ials

Gal

ixim

abN

AC

D80

Hum

anIg

G1

AD

CC

Rel

apse

dB

-cel

lNH

LU

nspe

cifie

dU

nspe

cifie

dPh

ase

III

clin

ical

tria

lsB

linat

umom

abM

T10

3C

D19

and

CD

3R

ecom

bina

ntbi

spec

ific

CD

CT-

cell

CL

LU

nspe

cifie

dU

nspe

cifie

dPh

ase

1cl

inic

altr

ials

CD

,clu

ster

ofdi

ffer

entia

tion;

CB

C,c

ompl

ete

bloo

dco

unt;

CD

C,c

ompl

emen

t-de

pend

entc

ytot

oxic

ity;A

DC

C,a

ntib

ody-

depe

nden

tcel

lula

rcyt

otox

icity

;NH

L,

non-

Hod

gkin

’sly

mph

oma;

AM

L,a

cute

mye

loid

leuk

emia

;CL

L,c

hron

icly

mph

ocyt

icle

ukem

ia;N

A,n

otav

aila

ble.

Dat

afr

om[1

,4,6

,7,9

,11,

14]


Examples of Investigational Monoclonal Antibodiesin Hematologic Malignancies

The success of rituximab and alemtuzumab in targeting non-Hodgkin’s lymphomaand chronic lymphocytic leukemia has increased the interest in the development ofmore targeted antibody therapeutic approaches. Rituximab is a chimeric antibody;thus there is a need to develop fully humanized antibodies to minimize infusion reac-tions and eliminate the development of human antibodies against the drug. Clinicalevaluation of alternative antibodies based on knowledge of antigen expression onthe surface of lymphoma cells [33] has led to the development of antibodies againstCD22 [unconjugated epratuzumab and calicheamicin conjugate CMC-544 (ino-tuzumab ozogamicin)], CD80 (galiximab), CD52 (alemtuzumab), CD2 [MEDI-507(siplizumab)], CD30 [SGN-30 and MDX-060 (iratumumab)], and CD40 (SGN-40)(Table 9.1). In addition, the VEGF inhibitor bevacizumab, which was first approvedfor the treatment of colon cancer, is currently under investigation for treating NHL.Small molecules that bind directly to receptor molecules (agonists) directed to tumornecrosis factor-related apoptosis-inducing ligand (TRAIL) are also being investi-gated as treatments for both advanced solid tumors and NHL, instead of targetedantibodies [33]. Cancer cells may become resistant to a targeted therapy by acti-vating an alternative pathway to evade apoptosis. This has prompted studies ofcombination treatment regimens of mABs (such as epratuzumab plus rituximab)or mABs plus chemotherapy such as CHOP (cyclophosphamide, doxorubicin, vin-cristine, and prednisone) and other schema [33]. Soluble forms of the antigens(sCD30) represent potential mechanisms of resistance by binding mAB before themAB can bind to the lymphoma cell. Antibody-based therapeutic approaches havealready had a profound impact on the treatment of NHL, and continued refinementwill optimize the clinical benefits [33].

The characteristic B-cell immunophenotype of the lymphocytosis associ-ated with chronic lymphocytic leukemia (CLL) (CD5+/CD19+/CD20+/HLA-DR+/CD23+/surface immunoglobulin dim) [34] is also a target of specific mABagents including lumiliximab (anti-CD23) [35] (summarized in Table 9.1). OthermABs such as AME-133 not only have high specificity for CD20 but also con-tains an Fc region which binds CD16 [Fcγ(gamma)RIIIa] with high affinity, thusimproving its ability to activate natural killer (NK) cells. GA101 is a humanizedanti-CD20 with high affinity for FcRγ (gamma)III, producing enhanced ADCC andstrong caspase-independent apoptosis activity to bind CD20 [36]. Veltuzumab isanother humanized IgG1κ (kappa) monoclonal antibody that, despite targeting thesame CD20 epitope as rituximab [37], has shown clinical activity at lower dosesthan rituximab. Epratuzumab, an anti-CD22 mAB of the IgG1κ (kappa) class, isactive as a single agent [38], and in combination with rituximab, for the treat-ment of patients with relapsed or refractory, indolent B-cell NHL. Apolizumab(Hu1D10; Remitogen) is designed to target the HLA-DRß (beta) polypeptide, on80–90% of cases of CLL, and is currently being tested in patients with relapsedor refractory CLL [39]. HCD122 and MDX-1411 are mABs which target CD40and CD70, respectively; they are being evaluated in phase I trials for patients


with CLL [35]. Several novel immunotherapeutics may also be implemented inthe treatment of CLL including bispecific monoclonal antibodies (anti-CD20 andanti-CD22) [40]; small molecule immunopharmaceuticals targeting CD37, a glyco-protein strongly expressed on the surface of CLL cells; and T-cell antibodies such asblinatumomab (MT103/MEDI-538), which targets CD19 and CD3 simultaneously.These are currently undergoing evaluation in phase I clinical trials [41].

Small Molecule Drugs

Small molecule drugs, also known as kinase inhibitors, signal transductioninhibitors, and small molecule inhibitors [5, 42, 43], are drugs that interfere withthe function of molecules involved in the processes and pathways of carcinogenesisprimarily targeting kinases. FDA-approved small molecule drugs used in treatingpatients with hematologic malignancies are summarized in Table 9.2.

Kinases regulate the phosphorylation of other proteins by the transfer of theterminal phosphate of adenosine triphosphate (ATP) to serine, threonine, or tyro-sine amino acid residues [42, 44]. The human genome encodes nearly 518 kinaseswhich mediate multiple signal transduction pathways by regulating the transfer ofphosphate molecules through cascades that control diverse cell processes includingproliferation, differentiation, contact inhibition, angiogenesis, apoptosis, and cellcycle progression [45]. Genetic mutations, gene overexpression, or chromosomaltranslocations may cause deregulation of phosphorylation patterns and consequentlyaberrant signal transduction deregulation that can lead to tumorigenesis [4, 9,11, 44].

Approximately 100 tyrosine kinases have been identified, which include twogroups of enzymes: cytosolic, non-receptor tyrosine kinases (non-RTKs) and trans-membrane receptor tyrosine kinases (RTKs) [8, 11, 43]. At least 58 receptortyrosine kinases (RTKs) have been identified and are grouped into 20 subfami-lies [45]. Epidermal growth factor (EGF) receptor, platelet-derived growth factor(PDGF) receptor, vascular endothelial growth factor (VEGF) receptor, and theRET proto-oncogene protein are examples of receptor tyrosine kinases [4, 8, 11].Binding of a ligand (e.g., growth factor) to the extracellular domain of the recep-tor tyrosine kinase activates the receptor. This initiates a cascade of signalingvia phosphorylation of intracellular proteins which conveys the extracellular sig-nal to the inside of the cell [45, 46]. The extracellular ligand-binding domain ofthe receptor is frequently targeted by a therapeutic monoclonal antibody, whilethe intracellular catalytic domain may be targeted by a small molecule drug [8,11, 47].

Examples of non- receptor tyrosine kinases (non-RTKs) are the oncogenic pro-tein products of the SRC, ABL1, and JAK2 genes. The SRC oncogenic protein wasthe first non-receptor tyrosine kinase identified [43, 48] and represents a group ofnine cytoplasmic proteins that are important in many cellular processes includingcell growth and differentiation, cell adhesion and motility, carcinogenesis, immunecell function, and even learning and memory [49, 50]. JAK2, a member of the Janus


Tabl

e9.

2Sm

alld

rug

mol

ecul

esap

prov

edby

the

FDA

for

hem

atol

ogic

mal

igna

ncie

s

Gen

eric

nam

eof

the

drug

Tra

dena

me

Targ

etM

echa

nism

ofac

tion

Hem

atol

ogic

mal

igna

ncie

sA

dver

sesi

deef

fect

sM

onito

ring

Adm

inis

trat

ion

The

FDA

appr

oval

Imat

inib

Gle

evec

BC

R–A

BL

1,c-

KIT

,PD

GFR

Kin

ase

inhi

bitio

nA

LL

,CM

L,

hype

reos

inop

hilic

synd

rom

e,sy

stem

icm

asto

cyto

sis,

GIS

Ta

Ras

h;w

eigh

tgai

n;ed

ema;

pleu

rale

ffus

ion;

card

iac

toxi

city

;na

usea

and

vom

iting

;ar

thra

lgia

san

dm

yalg

ias;

mye

losu

ppre

ssio

n

CB

C;l

iver

chem

istr

ies;

wei

ght;

sign

san

dsy

mpt

oms

offlu

idre

tent

ion

Ora

lrou

te20

01

Das

atin

ibSp

ryce

lB

CR

–AB

L1,

Src

fam

ily,

c-K

IT,

PDG

FR

Kin

ase

inhi

bitio

nC

ML

,AL

LR

ash;

diar

rhea

;ple

ural

effu

sion

;flui

dre

tent

ion;

muc

ositi

s;m

yelo

supp

ress

ion;

QT

inte

rval

prol

onga

tion

CB

C;E

CG

;liv

erch

emis

trie

s;w

eigh

t;si

gns

and

sym

ptom

sed

ema

Ora

lrou

te20

06

Nilo

tinib

Tasi

gna

BC

R–A

BL

1K

inas

ein

hibi

tion

CM

LC

ytop

enia

,ras

h,he

adac

he,

naus

ea,i

tchi

ngC

ardi

acir

regu

lari

ties

Ora

lrou

te20

07

All-

tran

s-re

tinoi

cac

id(A

TR

A),

tret

inoi

n

Ves

anoi

dPM

L–

RA

Rα

(alp

ha)

(ret

inoi

cac

idre

cept

orfu

sion

prot

ein)

Rev

oke

the

bloc

kca

used

byPM

L–R

AR

α

fusi

onon

the

diff

eren

tiatio

nof

imm

atur

epr

omye

locy

tes

into

norm

alm

atur

ebl

ood

cells

APL

Hyp

erle

ukoc

ytos

is;w

eigh

tga

in;p

erip

hera

lede

ma;

dysp

nea;

feve

r

CB

C,l

iver

func

tion

Ora

lrou

te19

95


Tabl

e9.

2(c

ontin

ued)

Gen

eric

nam

eof

the

drug

Tra

dena

me

Targ

etM

echa

nism

ofac

tion

Hem

atol

ogic

mal

igna

ncie

sA

dver

sesi

deef

fect

sM

onito

ring

Adm

inis

trat

ion

The

FDA

appr

oval

5-az

acyt

idin

e,de

cita

bine

Vid

aza,

Dac

ogen

DN

Am

ethy

l-tr

ansf

eras

esH

ypom

ethy

latin

gag

ent

Mye

lody

spla

stic

synd

rom

esN

ause

a,vo

miti

ng,

diar

rhea

,cyt

open

iaC

BC

,liv

erfu

nctio

nIn

trav

enou

sro

ute

2004

Suni

tinib

Sute

ntFL

T3

Kin

ase

inhi

bitio

n,an

ti-an

giog

enic

AM

L,r

enal

cell

carc

inom

a,G

IST

aH

eada

che,

feve

r,dr

ysk

in,

naus

ea,v

omiti

ng,r

ash,

flu-l

ike

sym

ptom

s

CB

C,t

hyro

idfu

nctio

nO

ralr

oute

2006

Rom

idep

sin

Isto

dax

His

tone

com

plex

onco

nden

sed

chro

mat

in

His

tone

deac

etyl

ase

(HD

AC

)in

hibi

tion

Cut

aneo

usT-

cell

lym

phom

aN

ause

a,fa

tigue

,bon

em

arro

wsu

ppre

ssio

nC

BC

,EC

G,

Intr

aven

ous

rout

e20

09

Pral

atre

xate

Folo

tyn

RFC

-1Fo

late

path

way

inhi

bitio

n,D

NA

synt

hesi

sin

hibi

tion

T-ce

llly

mph

oma

Muc

ositi

s,th

rom

bocy

tope

nia,

naus

ea,f

atig

ue

CB

C,m

ucos

itis,

rena

lfun

ctio

n,liv

erfu

nctio

n,pr

egna

ncy

Intr

aven

ous

rout

e20

09

The

FD

A-a

ppro

ved

apop

tosi

s-in

duci

ngag

ents

inhe

mat

olog

icm

alig

nanc

ies

Bor

tezo

mib

Vel

cade

26S

prot

easo

me

Indu

ctio

nof

apop

tosi

sby

prot

easo

me

inhi

bitio

n

Mul

tiple

mye

lom

a,m

antle

cell

lym

phom

a

Peri

pher

alne

urop

athy

;m

yelo

supp

ress

ion;

rash

;co

nstip

atio

n;di

arrh

ea;

edem

a;na

usea

and

vom

iting

Sign

san

dsy

mpt

oms

ofpe

riph

eral

neur

opat

hy;

CB

C

Intr

aven

ous

rout

e20

08

Clo

fara

bine

Clo

lar

DN

Asy

nthe

sis

inhi

bitio

nan

din

duct

ion

ofap

opto

sis

Puri

nenu

cleo

side

anal

ogue

Pedi

atri

cle

ukem

ia,

AL

L,T

-cel

lly

mph

oma

Vom

iting

,nau

sea,

diar

rhea

;bon

em

arro

wsu

ppre

ssio

n;he

pato

bilia

ryto

xici

ty

CB

C,E

CG

,he

patic

and

rena

lfun

ctio

n

Intr

aven

ous

rout

e20

04


Tabl

e9.

2(c

ontin

ued)

Gen

eric

nam

eof

the

drug

Tra

dena

me

Targ

etM

echa

nism

ofac

tion

Hem

atol

ogic

mal

igna

ncie

sA

dver

sesi

deef

fect

sM

onito

ring

Adm

inis

trat

ion

The

FDA

appr

oval

The

FD

A-a

ppro

ved

angi

ogen

esis

inhi

bito

rsin

hem

atol

ogic

mal

igna

ncie

s

Tha

lidom

ide

Tha

lom

idA

nti-

angi

ogen

esis

Inhi

bitio

nof

endo

thel

ialc

ells

(ant

i-an

giog

enic

)

Mul

tiple

mye

lom

aSe

datio

n,ne

urop

athy

,co

nstip

atio

n,de

epve

nous

thro

mbo

sis,

mye

losu

ppre

ssio

n,te

rato

geni

c

Preg

nanc

y,C

BC

,O

ral

2006

Len

alid

omid

eR

evlim

idA

nti-

angi

ogen

esis

Inhi

bitio

nof

endo

thel

ialc

ells

(ant

i-an

giog

enic

),im

mun

o-m

odul

ator

y

Mul

tiple

mye

lom

a,m

yelo

dysp

last

icsy

ndro

me

Hem

atol

ogic

toxi

citie

s,de

epve

nous

thro

mbo

sis,

tera

toge

nic

Preg

nanc

y,C

BC

,cl

ottin

gfu

nctio

n,liv

erfu

nctio

n

Ora

l20

06

a FDA

-app

rove

dno

n-he

mat

olog

icin

dica

tions

AL

L,A

cute

lym

phoc

ytic

leuk

emia

;CM

L,c

hron

icm

yelo

idle

ukem

ia;G

IST,

gast

roin

test

inal

stro

mal

tum

or;A

PL,a

cute

prom

yelo

cytic

leuk

emia

;RA

R,r

etin

oic

acid

rece

ptor

;PM

L,p

rom

yelo

cytic

leuk

emia

;AM

L,a

cute

mye

loid

leuk

emia

;CB

C,c

ompl

ete

bloo

dco

unt;

EC

G,e

lect

roca

rdio

gram

.Dat

afr

om[4

,6,7

,9,1

1,12

,69]


family of kinases, was incidentally discovered in 1989 and considered “just anotherkinase” [51]. The name was later modified to Janus kinases (JAKs; JAK1, JAK2,JAK3, and tyrosine kinase 2). JAK gene mutations relevant to human cancershave been described for JAK1 [T-cell acute lymphocytic leukemia (T-ALL), acutemyeloid leukemia (AML), breast cancer, lung cancer], JAK3 (AML cell lines andprimary cells, breast cancer, gastric cancer) [52], and JAK2 [myeloproliferative neo-plasms (MPN) and other myeloid malignancies such as trisomy 21-associated ALL].Only the JAK2 mutations in MPN [53] and JAK1 mutations in T-ALL are observedwith significant frequency [54]. Like JAK2, Abelson 1 (ABL1) is a cytoplasmic tyro-sine kinase. Normally, the ABL1 protein plays a role in non-erythroid myelopoiesis[55], cytoskeletal rearrangement, and inhibition of cell migration. Its oncogeniccounterpart the BCR–ABL1 fusion protein is a diagnostic feature in the pathogene-sis of chronic myelogenous leukemia (CML) and an indicator of poor prognosis forsome patients with ALL or AML [56].

Kinases may be classified into three categories according to their varying roles inthe process of carcinogenesis [42]. The first category of kinases demonstrates trans-forming or oncogenic activity with constant activation of these kinases requiredfor the development and proliferation of cancer cells. An example of this categoryis the V617F (Val617Phe)-activating mutation in the auto-inhibitory pseudokinasedomain of JAK2 which is frequently observed in patients with polycythemia vera,essential thrombocythemia, and idiopathic myelofibrosis [53]. This finding plus therapid proliferation of molecular diagnostic testing for these disorders have stimu-lated the rapid progression of several JAK2 inhibitors into phase I studies [57]. Thesecond category of kinases is not associated with cell transformation but is essentialfor the proliferation or survival of cancer cells. They may also be components of thedownstream signaling pathway of the first, transforming category of kinases [58].An example of this functional category is MTOR, which is part of the PI3K–AKTsignaling pathway [59]. MTOR inhibitors are being tested as therapies for renalcell carcinoma. The third category includes kinases which exhibit action duringdifferent stages of cancer establishment and preservation. For example, the vascu-lar endothelial growth factor receptor (VEGFR) and the fibroblast growth factorreceptor (FGFR) kinases are significant in developing and maintaining blood sup-ply of solid tumors (neovascularization) but are rarely deregulated in hematologicneoplasms [60].

Small molecular weight drug molecules penetrate the plasma membrane ofcells to target the intracellular, cytoplasmic domains of cell surface receptors orintracellular signaling molecules [11]. Most of the current kinase inhibitors arecompetitive binding molecules that target the ATP-binding sites of the kinase acti-vation loop, constitutively locking that domain in either the active or the inactiveconformation regardless of ligand binding [42]. All kinases have a conserved acti-vation loop important to the regulation of kinase activity which can assume a largenumber of conformations. The conformation may be catalytically competent andnormally phosphorylated, or an “inactive” conformer in which the activation loopnormally serves to block a substrate-binding site; phosphorylation alters the loopthree-dimensional structure and reveals the binding site [61, 62]. Therefore, smallmolecule drugs may act directly on the key signaling proteins, in contrast to many


mABs, which indirectly exert their targeted action through the induction of immuneresponses [11].

Unlike monoclonal antibodies, small molecule inhibitors are administered orally,with the exception of bortezomib (an apoptosis-inducing drug), which is admin-istered intravenously [9]. The half-life of the small molecule drugs is a fewhours compared to days or weeks for mABs. This correlates with the daily dos-ing of small molecule inhibitors versus the weekly or monthly administration ofmAB. The majority of small molecule inhibitors may interfere with metabolism bycytochrome P450 enzymes. Thus patients also taking multiple medications such asazole antifungals, anticonvulsants, dexamethasone, macrolide antibiotics, isoniazid,protease inhibitors, rifampin, St. John’s wort, warfarin, and verapamil may havedrug side effects [9, 63]. Other small molecule inhibitors are directly metabolizedby cytochrome P450 drug-metabolizing enzymes [63]. Therapeutic monoclonalantibodies are developed by genetic engineering, which is a more expensive andcomplex process in comparison to small molecule drugs which are chemicallymanufactured [64].

Small molecule inhibitors may be less specific than mABs due to their poten-tial to target several signaling pathways without specificity to a single enzyme [65].Though this multi-targeting approach may carry a risk for increased toxicity, it mayalso be regarded as a therapeutic advantage [66]. Multi-targeting could provide effi-cient elimination of the cancer cells since most cancers, especially solid tumors, aremulti-factorial and result from multiple gene mutations [8, 67]. For example, ima-tinib mesylate (Glivec R© or Gleevec R©), one of the early and most predominantlyused small drug inhibitors, has dramatically changed the treatment of CML [68]and is an example of multi-targeting. This agent targets the intracellular portions ofat least three tyrosine kinases: BCR–ABL1, which is the product of the fusion geneformed by the t(9;22) (q34;q11) chromosomal translocation implicated in the patho-genesis of CML [68]; KIT, which is involved in metastatic gastrointestinal stromaltumors (GISTs), and PDGFRA, which is involved in neoplasms such as glioblas-toma and chronic myeloproliferative syndromes characterized by eosinophilia [8,11, 69]. The downstream signal transduction pathways that receive signals fromthese kinases include RAS/RAF/ mitogen-activated kinase (MAPK), SRC familykinases, JUN NH2-terminal kinase/stressactivated protein kinase, phosphatidylinos-itol 3 kinase (PI3K), STAT5/Janus kinase, nuclear factor-κB, CRC oncogene-likeprotein/focal adhesion kinase, and MYC [70].

Imatinib mesylate [71, 72] and the second-generation tyrosine kinaseinhibitors, dasatinib [73] and nilotinib [69], directly target the fusion proteinBCR–ABL1 and are approved for the treatment of CML. Imatinib mesylate(2-phenylaminopyrimidine; imatinib, Gleevec R©, or Glivec R©) [69, 72], first desig-nated CGP57148B and later signal transduction inhibitor 571(STI571) [71], wasdeveloped as a result of a long process of random screening of different compoundsthat could target the ATP-binding site of the kinase [74, 75]. The tyrosine kinaseactivity is essential to the transforming function of BCR–ABL1. By competing withadenosine triphosphate (ATP) for binding to the tyrosine kinase, Gleevec R© ren-ders the BCR–ABL1 fusion protein unable to activate downstream effector tyrosine


kinase molecules that drive white blood cell proliferation [76, 77]. It does not affectthe activity of the normal ABL1 tyrosine kinase in the same cells. Imatinib is bothhighly effective in treating CML and well tolerated by the patients [69, 78]. Seminalwork by Druker [56] also demonstrated substantial therapeutic utility in the blastcrisis phase of CML and in BCR–ABL1-positive ALL.

Imatinib also inhibits the products of the KIT (CD117) and platelet-derivedgrowth factor receptor (PDGFR) genes [79]. KIT is a receptor tyrosine kinasewhich is commonly mutated in gastrointestinal stromal tumors (GISTs), result-ing in ligand-independent activity, autophosphorylation, stimulation of downstreamsignaling pathways, and uncontrolled cell proliferation [80]. Imatinib is effec-tive at downregulation of these mutant products and is an effective treatment forpatients with GIST. Similar activating mutations of KIT are observed in somepatients with somatic mastocytosis and imatinib has also been an effective tyrosinekinase inhibitor in some of these cases [81]. The PDGFRA gene on chromosome4q12 is frequently involved in an interstitial deletion producing a fusion genewith the FIP1L1 gene in chronic myeloproliferative syndromes characterized byeosinophilia. The fusion gene is a constitutively active tyrosine kinase that is alsoeffectively inhibited by imatinib to control hematopoietic cell transformation [8, 79].

Different mechanisms have been suggested for developing resistance to kinaseinhibitors. Point mutations that produce amino acid substitutions that alter theprotein folding and interfere with drug binding; activation of alternative kinasepathways; amplification of target molecules [82]; and presence of quiescent,hematopoietic stem cells that are resistant to kinase inhibitors are some routeswhich contribute to resistance [83]. There are typically three stages in the clinicalcourse of CML: chronic phase (CP), accelerated phase (AP), and blast crisis(BC) [84]. Imatinib is a highly effective therapy for early, chronic phase CML.However, the imatinib-based therapy has three main problems: the limited responseof CML-blast crisis or Philadelphia chromosome-positive ALL patients to imatinib;the development of resistance in approximately 40% of patients with CML, thatdevelops through the emergence of cell populations with mutations in the BCR–ABL1 kinase domain, which impair the binding of imatinib required for inhibition;and the relative insensitivity of CML stem cells to imatinib constituting a reservoirof malignant cells for leukemic relapse [82, 83, 85]. As the number of Philadelphiachromosomes increases with disease progression, the amount of BCR–ABLprotein expressed in the cell increases and the efficacy of imatinib decreases [86].Numerous second-generation ATP-competitive ABL tyrosine kinase inhibitorssuch as dasatinib, nilotinib, bosutinib, and INNO-406 have been developed tocounter this resistance. These new agents provide some clinical success in targetingmost of the ABL kinase point mutations responsible for imatinib resistance.However, a nucleotide change that creates an amino acid substitution of isoleucinefor threonine at codon position 315 (Thr315Ile or T315I) is the most frequentlyobserved mutation. It is located within the ATP-binding polypeptide loop and is noteffectively targeted by any of these agents [69].

Nilotinib or AMNI07, an aminopyrimidine derivative of imatinib, is asmall molecule drug rationally developed from the crystalline structure of the


imatinib–ABL complex [70, 87]. The N-methylpiperazine ring of imatinib has beenreplaced by a trifluoromethylimidazole-substituted phenyl group, which allows forincreased binding affinity to residues lining the binding site of the kinase. Likeimatinib, nilotinib binds to and stabilizes BCR–ABL in the inactive state [88, 89].Nilotinib has some increased potency (30-fold in vitro) and activity as a result of animproved topologic fit penetrating further into the central region of the kinase andblocking ATP binding [90].

Dasatinib (BMS-354825; Bristol-Myers Squibb) is structurally distinct fromimatinib. Dasatinib is a dual SRC/ABL kinase inhibitor with less stringent con-formational requirements allowing binding to both the active and the inactiveconformations of the ABL kinase domain [73, 91]. Dasatinib does not interact withthe same amino acid residues of the kinase P-loop region; thus it is still an effectiveinhibitor against many imatinib-resistant BCR–ABL1 mutations with the exceptionof T315I [73, 77]. The loss of the hydroxyl side chain and addition of a methylgroup of the substituted isoleucine residue prevents binding of both dasatinib andnilotinib in the Thr315Ile variants.

Dasatinib also inhibits SRC and Src-family kinases including FGR, FYN, HCK,LCK, LYN, and YES55 [92]. Src kinase inhibition might be advantageous inimatinib-resistant disease. Also, dasatinib has some inhibitory action against BCR–ABL1 in CD34+CD38− CML stem cells, at least more effectively than nilotinib orimatinib [91, 93]. Current investigations with dasatinib include potential antitumoreffects in cell lines of head and neck squamous cell carcinoma, non-small cell lungcancer (NSCLC), prostate cancer, breast cancer, and multiple myeloma [94].

All-Trans-Retinoic Acid (ATRA)

All-trans-retinoic acid (ATRA or tretinoin), a metabolic product of vitamin A(retinol), is one of the most important morphogens which induce the differentiationof immature blood cells to mature end-stage cells which then die [95]. In additionto being the first molecularly targeted cancer therapy [96], ATRA is also the firstexample of cancer differentiation induction therapy [97].

ATRA selectively targets the cancer cells of the majority of patients (>95%)with acute promyelocytic leukemia (APL) which harbor a t(15;17) (q22; q21) chro-mosomal translocation between the promyelocytic leukemia (PML) gene and theretinoic acid receptor alpha gene [RARα (alpha)]. This PML–RARα (alpha) con-figuration produces a retinoic acid receptor fusion protein which blocks cells atthe immature promyelocytic stage of myeloid differentiation. This drug releasesthe maturation block caused by this oncoprotein allowing the cells to differenti-ate to normal, mature blood cells; decreasing the proliferation of promyelocytes;and producing cells which will undergo programmed cell death [96–98]. Less than5% of patients with APL may have other variant rearrangements between RARα

and translocation partner genes other than PML [96, 98]. These alternative fusionsmay involve the promyelocytic leukemia zinc finger (PLZF) or the nucleophosmin(NPM), nuclear mitotic apparatus (NUMA), and STAT5b partner genes. The natureof the fusion partner impacts the response of the cells to ATRA therapy [98, 99].


RARα is a ligand-dependent transcription factor and a nuclear hormone receptorprotein. In the presence of its ligand, retinoic acid, RARα regulates the expressionof a large number of target genes, a subset of which is important for normal myeloidcell differentiation [96]. RARα normally dimerizes with a member of the retinoid-Xreceptor (RXR) family leading to the formation of a heterodimer with high DNA-binding affinity. In the absence of retinoic acid, these RARα/RXR heterodimersbind to specific retinoic acid response elements (RAREs) in the promoters of tar-get genes and induce transcriptional repression. This repression is mediated bythe RARα/RXR heterodimer interacting with transcriptional repressors includingthe nuclear receptor–corepressor (N-CoR)/silencing mediator of the retinoid andthyroid (SMRT) receptor, the transcriptional corepressors SIN3A or SIN3B, andhistone deacetylases (HDACs) [96, 100].

ATRA binds with more affinity to the fusion oncoprotein PML–RARα than itdoes to these normal cellular targets [100]. This drug induces a conformationalchange in PML–RARα that dissociates the N-CoR and recruits transcriptional co-activators [96, 100]. Moreover, ATRA also induces cleavage of the PML–RARα

oncoprotein through a proteasome-mediated pathway to reverse the fusion proteintargeting of normal PML function. Together, these activities induce growth arrestand terminal differentiation of the malignant promyelocytes [96]. Leukemic cellswith variant chromosomal translocations producing NPM–RARα and NPM–RARAα

fusion genes are also sensitive to ATRA [100].Although ATRA is highly effective at inducing the terminal differentiation of

promyelocytes, it is not curative as a single agent. Resistance to ATRA can developfrom the acquisition of mutations in the RARα ligand-binding domain of the fusionprotein. ATRA is now used in combination with other chemotherapeutic agents,such as anthracyclines, with responses that are superior to treatment with eitherATRA or chemotherapy alone [96, 101]. Arsenic trioxide (As2O3) has also beenused to treat patients with APL. This compound does not bind to RARα but insteadbinds to the PML portion of the chimeric product, inducing its degradation througha proteasome-dependent mechanism. Arsenic trioxide’s primary effect on promye-locytes is thought to be induction of apoptosis and is an option for patients who haverelapsed APL following ATRA therapy [96].

Examples of Small Molecule Investigational Agentsfor Acute Myeloid Leukemia (AML)

AML is characterized by many different molecular genetic abnormalities, bothamong different subtypes of AML and throughout the clinical progression of sub-types. Additional genetic changes may accrue to increase the genomic instabilityand dysregulation of cellular functions. Unlike CML, in which the principal onco-genic protein is BCR–ABL1, successful results with newly synthesized inhibitorswould represent a compelling case for the power of targeted molecular therapy inAML, a genetically complex cancer [96].


Anti-cancer agents being examined for possible treatment of acute leukemias(Table 9.3) include immunoconjugate drugs; inhibitors of the multidrug resistanceP-glycoprotein, ABCB1; farnesyltransferase inhibitors; histone deacetylase and pro-teasome inhibitors; anti-angiogenic agents, anti-sense oligonucleotides to blockBCL2 gene transcription; inhibitors of MTOR; alkylating agents; purine analogues;inhibitory anti-FLT3 antibodies; and finally, small molecule FLT3 tyrosine kinaseinhibitors [102].

Table 9.3 Novel targeted agents under investigation for treatment of patients with acute myeloidleukemia (AML)

Agents Mechanism of action Target

Zosuquidar Direct drug resistancemodulation

P-glycoprotein and other multidrugresistance proteins

Oblimersen (18-mer BCL2anti-sense oligonucleotide)

Drug resistancemodulators

BCL2 gene expression; overexpressionof BCL2 is common, poorprognostic indicator in AML [119]

• PKC412• CEP701• MLN518

Tyrosine kinaseinhibitors (TKI) ofFLT3

Clinical trials of TKIs of FLT3 ITDmutations in combination withchemotherapy; in both relapsed andnewly diagnosed FLT3-mutantAML patients [103]

Anti-GM-CSF receptor Immunotherapy Antibody conjugated to truncateddiphtheria toxin targets GM-CSFreceptor [119]

Tipifarnib (Zarnestra) Farnesyltransferaseinhibitors

Post-translational modification ofRAS, lamin A, and HJJ-2 with afarnesyl lipid moiety is inhibitedwhich restricts subsequenttranslocation to the cell membranesurface [119]

ITD, internal tandem duplication; GM-CSF, granulocyte macrophage colony-stimulating factor

The FMS-related tyrosine kinase (FLT3) gene encodes an RTK expressed onearly hematopoietic progenitor cells that is activated upon binding with its ligand,FL, to generate a cascade of tyrosine phosphorylation of FLT3 and other down-stream targets. Somatic mutations of FLT3 include internal tandem duplications(ITDs) detected in ∼30% of patients with AML. The ITDs activate the FLT3 kinasefunction in the absence of ligand and is associated with a more aggressive diseasecourse and poor prognosis [103].

Examples of Investigational Agents for the Treatmentof Myeloproliferative Neoplasms (MPN)

The 2008 World Health Organization (WHO) classification system [104] nowdescribes myeloproliferative neoplasms (MPN) as chronic myelogenous leukemia


(CML), polycythemia vera (PV), essential thrombocythemia (ET) and primarymyelofibrosis (PMF), systemic mastocytosis (SM), chronic neutrophilic (CNL) andeosinophilic (CEL) leukemias to replace the former “myeloproliferative disorders”terminology. Therapeutically validated oncoproteins for targeted treatment of MPNinclude BCR–ABL1 and rearranged PDGFR proteins. Just as these genotype–phenotype associations have been effectively exploited in the development of highlyaccurate diagnostic assays and molecular targeted therapy, similar approaches arebeing explored for other MPN with specific genetic alterations: polycythemiavera (JAK2 V617F and other JAK2 mutations), essential thrombocythemia (JAK2V617F and MPL W515L mutations), primary myelofibrosis (JAK2 V617F and MPLW515L mutations), systemic mastocytosis (KIT D816V and other KIT mutations),and stem cell leukemia/lymphoma (ZNF198–FGFR1 and other FGFR1 fusiongenes) [96].

Table 9.4 New agents under development for treatment of chronic myeloid leukemia (CML)[69, 70]

Agent Primary target/inhibition

Effectiveness againstBCR–ABL1 T315Imutant

Bosutinib (SKI-606) Tyrosine kinases including c-Abl, v-Abl,BCR–ABL, Src family kinases

No (second-generationAbl TKIs)

INNO-406 (S-187,CNS-9)

Tyrosine kinases: c-Abl, v-Abl, BCR–ABL,and Lyn kinase

No (second-generationAbl TKIs)

• MK-0457 (VX-680)• PHA-739358• AT-9283

Aurora kinases. These kinases play animportant role in the regulation of mitoticprocess during cell division [70]

Yes

• ON-012380 Abl kinase (substrate competitive) Yes

• Tipifarnib(R115777, Zarnestra)

• Lonafarnib(SCH66336, Sarasar)

• BMS-214662

Farnesyltransferase inhibitors blockpost-translational modification andconstitutive activation of RAS inBCR–ABL1-positive cells

Yes

• Sorafenib (Nexavar) Multiple tyrosine kinases are inhibited Yes

• LAQ824• LBH589• Suberoylanilide

hydroxamic acid

Histone deacetylase inhibitordown-decreases levels of mutantBCR–ABL1 with T3151; inducesapoptosis

Yes

• Geldanamycin• IPI504• PEG-ZnPP,

SMA-Znpp

Binds heat-shock protein (HSP90); inducesBCR–ABL1 protein degradation

Yes

• Homoharringtonine(HHT; cephalotaxinealkylating agent)

Inhibits protein synthesis and inducesapoptosis

Yes


Novel Targeted Agents Under Investigation for the Treatmentof CML

Second-generation ABL1 TKIs, nilotinib and dasatinib, have been developed tooverride imatinib resistance. Third-generation INNO-46 and bosutinib are underdevelopment but the T315I missense mutation remains the biggest obstacle. Severalpromising inhibitors which may target the Aurora kinase, heat-shock proteins,and farnesyltransferase among others are in clinical trials as summarized inTable 9.4 [69].

Angiogenesis-Inhibiting Drugs

Angiogenesis, the formation of new blood vessels from existing vasculature, isessential to maintain sufficient supply of nutrients and oxygen to the tissue nec-essary for growth of tumors and spread of tumor cells to other parts of the body(metastasis). Cancer cells contribute to this process by secreting growth factors, suchas vascular endothelial growth factor (VEGF) and platelet-derived endothelial cellgrowth factor (PDECGF), which stimulate endothelial cell proliferation requiredfor development of capillaries. Most research on tumor growth and angiogenesisfactors has focused on solid tumors and the extent to which vascularization may pre-dict the aggressive course of tumor growth. However, anti-angiogenesis drugs mayalso serve in the treatment of multiple myeloma, myelofibrosis, and myelodysplasticsyndrome.

Since the pharmacological effects of thalidomide extended beyond its neuroseda-tive effects, it was subsequently investigated in a number of dermatologic, rheuma-tologic, and malignant diseases [105]. Thalidomide and its analogue lenalidomide(Table 9.2) have immuno-modulatory, anti-angiogenic, and anti-neoplastic proper-ties. Lenalidomide was designed to enhance immunologic and anti-cancer propertieswhile potentially decreasing the neurotoxic and teratogenic adverse effects ofthe parent compound, thalidomide [106]. Unexpectedly, thalidomide was foundto have anti-myeloma activity when it was thought its anti-angiogenic activitycould slow the disease by inhibiting the formation of new blood vessels in thishighly vascularized cancer. However, the anti-cancer activity of thalidomide andits immunomodulators in multiple myeloma (MM) likely occurs through differentmechanisms and sites in the bone [107]. At least four distinct, but potentially com-plementary, mechanisms have been proposed to account, at least in part, for theantitumor activity of thalidomide and its derivatives: (a) direct anti-proliferative/pro-apoptotic effects against multiple myeloma (MM) cells, including inhibition of thetranscriptional activity of nuclear factor kappa B(NF-κB) and its anti-apoptotic tar-get genes; (b) indirect targeting of MM cells by blocking interactions with bonemarrow stromal cells; (c) immuno-modulatory inhibition of endothelial cell migra-tion, adhesion and capillary-tube formation, and inhibition of key pro-angiogenicgrowth factors such as VEGF [105, 108]; and (d) immuno-modulatory enhancementof natural killer (NK) cell-mediated cytotoxicity against tumor cells.


Apoptosis-Inducing Drugs

Apoptosis or programmed cell death is an important mechanism for elimination ofdamaged, excess, or abnormal cells. Apoptosis is a common feature of aging cellsbut also occurs during embryonic development of select tissues and organs. In can-cer cells, apoptosis is often inhibited and contributes to accumulation of an excesscell population which, in turn, may also be susceptible to accumulation of geneticmutations. Apoptosis-inducing drugs are molecules which inhibit proteasomes, thehistone protein complexes that regulate gene expression [4].

Bortezomib (Velcade) is a modified boronic acid dipeptide with the molecularformula C19H25BN4O4. It is a selective and reversible proteasome inhibitor thatinduces apoptosis through inhibition of the chymotrypsin-like activity of the 20Sproteasome, a subunit of the 26S proteasome [109, 110]. The inhibition of theproteasome system creates an imbalance of various regulatory proteins, triggeringcell cycle arrest at the G1–S and G2–M phases of the cell cycle and activat-ing apoptotic pathways within the cell [111], including the caspase-8-mediatedextrinsic death-receptor pathway, the intrinsic mitochondrial apoptotic pathway,involving caspase-9 activation, and the endoplasmic reticulum stress response path-way, involving caspase-12. In addition, bortezomib also mediates downregulation ofcytokine signaling, cell-adhesion molecules, and angiogenesis factors, via inhibitionof the NF-κB signaling pathway [110].

Bortezomib was originally FDA approved in 2004 for the treatment of relapsedand refractory multiple myeloma [6] and is now FDA approved as a frontlinetreatment for myeloma patients [109] and for relapsed mantle cell lymphoma [112].

While multiple myeloma remains incurable, improving overall survival (OS)is the ultimate goal for new treatment options. Complete response (CR) hasbecome a well-established surrogate for OS. For myeloma patients who maybe candidates for hematopoietic stem cell transplant, frontline therapy must notadversely affect the ability to harvest sufficient stem cells, or their viability.Combinations of bortezomib plus established and novel agents, such as melphalan–prednisone, dexamethasone, doxorubicin, thalidomide–dexamethasone, and, mostrecently, lenalidomide–dexamethasone, may prove more promising than previousstandards of care [109, 110].

Cancer Vaccines

The goal of cancer vaccines is to increase the recognition of cancer cell componentsby the host immune system. Cancer vaccines are developed by obtaining cancer cellsfrom the patient (autologous) or from an established line of cancer cells (allogeneic).These are genetically engineered in vitro or fused with mouse/human hybridomacells to create a fusion that secretes the specific idiotype or cancer cell immunoglob-ulin protein of each patient. This purified fraction is linked to an immune stimulantand then administered to the patient to trigger a specific immune response by theautologous immune system to specifically target the patient’s cancer cell profile.


These vaccines may be designed to either treat or prevent solid tumors. For exam-ple, GARDASIL R© has been approved for the prevention of infection from sometypes of human papillomaviruses which are specifically associated with the major-ity of cases of cervical cancer plus some head and neck, vulvar, vaginal, penile, andanal cancers as well as genital warts [4]. BiovaxID R©, a personalized therapeuticvaccine with the potential for treating follicular B cell non-Hodgkin’s lymphoma,is in FDA- and NIH-approved phase 3 clinical trials. BiovaxID R© includes tumor-specific immunoglobulin protein, idiotype (ID), that is expressed on the surfaceof cancerous B cells to initiate an immune response that targets these cells[113–116].

Gene Therapy

Gene therapy is a general term applied to a therapy that implements geneticmaterial to modify cells. Gene therapy is generally achieved using the processof oncolytic virotherapy or gene transfer with many of these technologies stillunder development [117]. Oncolytic virotherapy involves genetically engineeringviruses to target and kill cancer cells, while sparing healthy tissue. Early trialsidentified unanticipated problems with the use of viral agents and incompletelyunderstood requirements for patient safety. Also, most people have antibodies tothe viruses commonly used as vectors to transfer the genes such as adenovirusor herpes simplex virus type 1 [4]. The virus constructs readily enter the cellsbut are neutralized by host antibodies that were developed due to prior viralinfection.

Gene transfer is the process of introducing a foreign gene into the genome of acancer cell, or the tissue surrounding it, to replace an abnormal or disease-causinggene copy. Virus-derived vectors are required to deliver the gene into the cancercells and maintain the gene copy until it is inserted into the host cell genome [4].Regulatory gene sequences are commonly included to enhance the expression of theintroduced therapeutic gene copy. Non-viral alternatives have also been investigatedfor introducing genes into cancer cells. These range from directly introducing ther-apeutic DNA into the cells (requires large quantities of DNA); encapsulating theDNA into a liposome which fuses with the cell membrane to pass the DNA intothe cell; and chemically linking the DNA to molecules that will bind to a cancercell surface receptor which will then invaginate into the cell membrane and transferthe DNA to the interior. Gene therapy introduces DNA coding for tumor suppres-sor genes, to restrict growth and proliferation, or suicide genes, which expressenzymes that can convert an inactive prodrug into an active anti-neoplastic com-pound. Attempts have also been made to block or replace oncogenes in cancer cells[4]. Many challenges have been encountered, most commonly the lack of efficientand selective vectors to deliver the genes, failure to mobilize the genes to the nucleusfor expression, and insufficient or poorly regulated promotion of the expression ofgene functions [8].


Challenges and Changes in Clinical Practice in the Eraof Targeted Therapy

Targeted therapies have expanded the concept of individually tailored cancer treat-ment because some of these agents may be effective only in patients with cancersthat carry a specific molecular target but lack response in the absence of the target.This distinction may be influenced by patient ethnicity and sex, as well as by tumorhistology [9]. In addition, targeted therapies require new approaches to determineoptimal dosing, to assess patient adherence to therapy, and to evaluate treatmenteffectiveness. The intravenous administration of the traditional chemotherapy inan observed infusion area facilitated monitoring of compliance and managementof toxicities. Most small molecule inhibitors are taken at home on a long-termdaily basis. Thus, assessing patient adherence resembles the challenges encoun-tered with therapies for chronic diseases such as diabetes and hypertension. Limitedstudies indicate that patient adherence to oral cancer treatment regimens can behighly variable and somewhat unpredictable [9]. The cost of these agents, whichcan exceed several thousand dollars per month, may become an important issuein health-care economics [9]. Substituting oral, small molecule inhibitors for tra-ditional chemotherapy eliminates some treatment costs, including those associatedwith vascular access and intravenous infusions. However, targeted therapy is oftenused in addition to, rather than in place of, traditional chemotherapy. If targetedtherapy includes monoclonal antibodies, costs can escalate exponentially [9].

Tumorigenesis can involve dozens of independent genetic mutations in multiplepathways; thus targeting even a few gene products may be overly simplistic and evenineffective. As many as 12 different pathways can be involved in a single cancertype because biological processes have alternate pathways, developed as a result ofevolutionary pressures, giving rise to a redundancy, should one path become blockedor targeted, that is unlikely to be bypassed by a single, highly targeted agent or evenby groups of targeted agents such as TKIs [118].

Selective inhibition of an enzyme target may produce unexpected consequences,such as when the protein or its subtypes being inhibited have multiple roles; it nowseems that this is the case for many enzymes. The range of problems that canresult from high selectivity includes the rapid development of resistance, whichis more likely if a single molecular target is being inhibited with high selectivity[118]. Strategies for developing multiple inhibitors to simultaneously target differ-ent kinase sites and for discovering synergistic inhibitor combinations are urgentlyneeded [42].

The development of these agents will require new skill sets and research tech-nologies. First, the efficacy of targeted agents requires that the subset of tumorsshow dependency on the target for cancer cell growth or survival. Patient selec-tion strategies will likely include molecular genetic diagnostic assays to confirm thepresence of the specific target prior to treatment selection; drug and diagnostic testcombinations will proceed together through the FDA-approved process. Second, itmay be instructive to screen for the presence of the target in the tumors and to seekindications of target association with the study agent. Third, the therapy endpoints


with these agents may require new definitions and assessments of outcome and sur-vival. Some of these agents may not induce (detectable) tumor shrinkage, thus themeasures of response to past chemotherapy may prove insensitive and insufficientin gauging anti-neoplastic effect(s). Fourth, some of these agents will have limitedactivity by themselves yet may have the capacity to markedly enhance the anti-tumor activity of conventional agents like chemotherapy or even other biologicalagents. This latter point is well exemplified by the anti-angiogenesis mAB beva-cizumab, which has no activity as a single agent and yet is clinically active whencombined with chemotherapy. Lastly, there is a risk that novel agents that are testedin a previously treated patient population may not be the ideal population to detectthe antitumor activity of novel agents because their cancers may have resistance toany type of therapy [43].

References

1. Adams GP, Weiner LM. Monoclonal antibody therapy of cancer. Nat Biotechnol.2005;23:1147–1157.

2. Mishra BK, Parikh PM. Targeted therapy in oncology. MJAFI. 2006;62:169–173.3. Armitage JO. Targeted therapy and hematological malignancy. Target Oncol. 2009;4:1–2.4. Kalyn R. Overview of targeted therapies in oncology. J Oncol Pharm Pract. 2007;13:

199–205.5. National Cancer Institute FactSheet. Targeted cancer therapies. http://www.nci.nih.gov/

cancertopics/factsheet/Therapy/targeted. Accessed 15 December 2009.6. Ross JS, Schenken DP, Pietrusko R, Rolfe M, Linette GP, Stec J, et al. Targeted therapies for

cancer 2004. Am J Clin Pathol. 2004;122:598–609.7. US Food and Drug Administration Combination Product Definition. http://www.fda.gov/

combinationproducts/aboutcombinationproducts/ucm118332.htm. Accessed 15 December2009.

8. Segota E, Bukowski RM. The promise of targeted therapy: cancer drugs become morespecific. Cleve Clin J Med. 2004;71:551–560.

9. Gerber DE. Targeted therapies: a new generation of cancer treatments. Am Fam Physician.2008;77:311–319.

10. Summerhayes M. What targeted therapy for cancer is and how it is achieved. Pharm J.2006;277:518–525.

11. Imai K, Takaoka A. Comparing antibody and small-molecule therapies for cancer. Nat RevCancer. 2006;6:714–727.

12. Centerwatch. Approved drugs for oncology. http://www.centerwatch.com/drug-information/fda-approvals/drug-areas.aspx?AreaID=12. Accessed 15 December 2009.

13. Steplewski Z, Lubeck MD, Koprowski H. Human macrophages armed with murineimmunoglobulin G2a antibodies. Science 1983;221:865–867.

14. Sharkey M, Goldenberg DM. Targeted therapy of cancer: new prospects for antibodies andimmunoconjugates. CA Cancer J Clin. 2006;56:226–243.

15. Newsome BW, Ernstoff MS. The clinical pharmacology of therapeutic monoclonal antibod-ies in the treatment of malignancy; have the magic bullets arrived? Br J Clin Pharmacol.2008;66:6–19.

16. Harding J, Burtness B. Cetuximab: an epidermal growth factor receptor chimeric human-murine monoclonal antibody. Drugs Today (Barc.) 2005;41:107–127.

17. Maier LA, Xu FJ, Hester S, Boyer CM, et al. Requirements for the internalization of amurine monoclonal antibody directed against the HER-2/neu gene product c-erbB-2. CancerRes. 1991;51:5361–5369.


18. Izumi Y, Xu L, di Tomaso E, Fukumura D, Jain RK. Tumour biology: herceptin acts as ananti-angiogenic cocktail. Nature. 2002;416:279–280.

19. Pietras RJ, Fendly BM, Chazin VR, et al. Antibody to HER-2/neu receptor blocksDNA repair after cisplatin in human breast and ovarian cancer cells. Oncogene. 1994;9:1829–1838.

20. Gorter A, Meri S. Immune evasion of tumor cells using membrane-bound complementregulatory proteins. Immunol. Today. 1999;20:576–582.

21. Chari RVJ. Targeted cancer therapy: conferring specificity to cytotoxic drugs. Acc ChemRes. 2008;41:98–107.

22. Olafsen T, Kenanova VE, Sundaresan G, et al. Optimizing radiolabeled engineeredantip185HER2 antibody fragments for in vivo imaging. Cancer Res. 2005;65:5907–5916.

23. Kohler G, Milstein C. Continuous cultures of fused cells secreting antibody of predefinedspecificity. Nature. 1975;256:495–497.

24. Carter P. Improving the efficacy of antibody-based cancer therapies. Nat Rev Cancer.2001;1:118–129.

25. Green LL, Hardy MC, Maynard-Currie CE, et al. Antigen-specific human monoclonal anti-bodies from mice engineered with human Ig heavy and light chain YACs. Nat Genet.1994;7:13–21.

26. Moroney S, Plückthun A. Modern antibody technology: the impact on drug development.In: Knäblein J, ed. Modern Biopharmaceuticals. Weinheim, Germany: Wiley-VCH VerlagGmbH & Co KGaA; 2005. pp. 1147–1186.

27. Maloney DG, Grillo-López AJ, White CA, et al. IDEC-C2B8 (Rituximab) anti-CD20 mon-oclonal antibody therapy in patients with relapsed low-grade non-Hodgkin’s lymphoma.Blood. 1997;90:2188–2195.

28. Kuus-Reichel K, Grauer LS, Karavodin LM, et al. Will immunogenicity limit the use,efficacy, and future development of therapeutic monoclonal antibodies? Clin Diagn LabImmunol. 1994;1:365–372.

29. Isaacs JD. From bench to bedside: discovering rules for antibody design, andimproving serotherapy with monoclonal antibodies. Rheumatology (Oxford). 2001;40:724–738.

30. Watkins NA, Ouwehand WH. Introduction to antibody engineering and phage display. VoxSang. 2000;78:72–79.

31. Qu Z, Griffiths GL, Wegener WA, et al. Development of humanized antibodies as cancertherapeutics. Methods. 2005;36:84–95.

32. Mass RD, Press MF, Anderson S, et al. Improved survival benefit from Herceptin(trastuzumab) in patients selected by fluorescence in situ hybridization (FISH). Proc ASCO.2001;21:85A.

33. Fanale MA, Younes A. Monoclonal antibodies in the treatment of non-Hodgkin’s lymphoma.Drugs. 2007;67:333–350.

34. Cheson BD, Bennett JM, Grever M, et al. National Cancer Institute-sponsored WorkingGroup guidelines for chronic lymphocytic leukemia: revised guidelines for diagnosis andtreatment. Blood. 1996;87:4990–4997.

35. Quintás-Cardama A, O’Brien S. Targeted therapy for chronic lymphocytic leukemia. TargetOncol. 2009;4:11–21.

36. Umana P, Ekkehard M, Bruenker P, et al. Novel 3rd generation humanized type II CD20 anti-body with glycoengineered Fc and modified elbow hinge for enhanced ADCC and superiorapoptosis induction. Blood. 2006;11:(abstract 229).

37. Stein R, Qu Z, Chen S, et al. Characterization of a new humanized anti-CD20 monoclonalantibody, IMMU-106, and its use in combination with the humanized anti-CD22 antibody,epratuzumab, for the therapy of non-Hodgkin’s lymphoma. Clin Cancer Res. 2004;10:2868–2878.

38. Leonard JP, Coleman M, Ketas JC, et al. Phase I/II trial of epratuzumab (humanized anti-CD22 antibody) in indolent non-Hodgkin’s lymphoma. J Clin Oncol. 2003;21:3051–3059.


39. Leonard JP, Coleman M, Ketas J, et al. Combination antibody therapy with epratuzumaband rituximab in relapsed or refractory non-Hodgkin’s lymphoma. J Clin Oncol. 2005;23:5044–5051.

40. Qu Z, Goldenberg DM, Cardillo TM, et al. Bispecific anti-CD20/22 antibodies inhibit B-celllymphoma proliferation by a unique mechanism of action. Blood. 2008;111:2211–2219.

41. Bargou R, Leo E, Zugmaier G, et al. Tumor regression in cancer patients by very low dosesof a T cell-engaging antibody. Science. 2008;321:974–977.

42. Zhang J, Yang PL, Gray NS. Targeting cancer with small molecule kinase inhibitors. NatRev Cancer. 2009;9:28–39.

43. Baselga J. Novel agents in the era of targeted therapy: what have we learned and how hasour practice changed? Ann Oncol. 2008;19(Suppl. 7):281–288.

44. Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S. The protein kinase comple-ment of the human genome. Science. 2002;298:1912–1934.

45. Robinson DR, Wu YM, Lin SF. The protein tyrosine kinase family of the human genome.Oncogene. 2000;19:5548–5557.

46. Spencer J. Targeted drugs for cancer treatment. Pharm Pract 2006;22:1–7.47. Held G, Poschel V, Pfreundschuh M. Rituximab for the treatment of diffuse large B-cell

lymphomas. Expert Rev Anticancer Ther. 2006;6:1175–1186.48. Martin GS. The hunting of the Src. Nat Rev Mol Cell Biol. 2001;2:467–475.49. Thomas SM, Brugge JS. Cellular functions regulated by Src family kinases. Annu Rev Cell

Dev Biol. 1997;13:513–609.50. Cohen DM. SRC family kinases in cell volume regulation. Am J Physiol Cell Physiol.

2005;288:C483–C493.51. Wilks AF. The JAK kinases: not just another kinase drug discovery target. Semin Cell Dev

Biol. 2008;19:319–328.52. Jeong EG, Kim MS, Nam HK, et al. Somatic mutations of JAK1 and JAK3 in acute

leukemias and solid cancers. Clin Cancer Res. 2008;14:3716–3721.53. James C, Ugo V, Le Couedic JP, Staerk J, Villeval JL, Constantinescu SN, Casadevall N,

Vainchenker W, et al. A unique clonal JAK2 mutation leading to constitutive signallingcauses polycythaemia vera. Nature. 2005;434:1144–1148.

54. Tefferi A. Molecular drug targets in myeloproliferative neoplasms: mutant ABL1, JAK2,MPL, KIT, PDGFRA, PDGFRB and FGFR1. J Cell Mol Med. 2009;13:215–237.

55. Caracciolo D, Valtieri M, Venturelli D, Peschle C, Gewirtz AM, Calabretta B. Lineage-specific requirement of c-abl function in normal hematopoiesis. Science. 1989;245:1107–1110.

56. Druker BJ, Sawyers CL, Kantarjian H, Resta DJ, Reese SF, Ford JM, et al. Activity of aspecific inhibitor of the BCR-ABL tyrosine kinase in the blast crisis of chronic myeloidleukemia and acute lymphoblastic leukemia with the Philadelphia chromosome. N Engl JMed. 2001;344:1038–1042.

57. Morgan KJ, Gilliland DG. A role for JAK2 mutations in myeloproliferative diseases. AnnRev Med. 2008;59:213–222.

58. Kaelin WG Jr. The concept of synthetic lethality in the context of anticancer therapy. NatRev Cancer. 2005;5:689–698.

59. Faivre S, Kroemer G, Raymond E. Current development of mTOR inhibitors as anticanceragents. Nat Rev Drug Discov. 2006;5:671–688.

60. Kerbel RS. Tumor angiogenesis. N Engl J Med. 2008;358:2039–2049.61. Pearson MA, Fabbro D. Targeting protein kinases in cancer therapy: a success? Expert Rev

Anticancer Ther. 2004;4:1113–1124.62. Liu Y, Gray NS. Rational design of inhibitors that bind to inactive kinase conformations. Nat

Chem Biol. 2006;2:358–364.63. Kantarjian HM, Cortes JE, O’Brien S, et al. Imatinib mesylate therapy in newly diag-

nosed patients with Philadelphia chromosome-positive chronic myelogenous leukemia: highincidence of early complete and major cytogenetic responses. Blood. 2003;101:97–100.


64. Tanner JE. Designing antibodies for oncology. Cancer Metastasis Rev. 2005;24:585–598.65. Huang S, Armstrong EA, Benavente S, et al. Dual-agent molecular targeting of the epider-

mal growth factor receptor (EGFR): combining anti-EGFR antibody with tyrosine kinaseinhibitor. Cancer Res. 2004;64:5355–5362.

66. Xia W, Gerard C M, Liu L, et al. Combining lapatinib (GW572016), a small moleculeinhibitor of ErbB1 and ErbB2 tyrosine kinases, with therapeutic anti-ErbB2 antibod-ies enhances apoptosis of ErbB2-overexpressing breast cancer cells. Oncogene. 2005;24:6213–6221.

67. Hambley TW, Hait WN. Is anticancer drug development heading in the right direction?Cancer Res. 2009;69:1259–1262.

68. Deininger M, Buchdunger E, Druker BJ. The development of imatinib as a therapeutic agentfor chronic myeloid leukemia. Blood. 2005;105:2640–2653.

69. Tanaka R, Kimura S. Abl tyrosine kinase inhibitors for overriding Bcr-Abl/T315I: from thesecond to third generation. Expert Rev Anticancer Ther. 2008;8:1387–1398.

70. Gora-Tybor J, Robak T. Targeted drugs in chronic myeloid leukemia. Curr Med Chem.2008;15:3036–3051.

71. Buchdunger E, O’Reilly T, Wood J. Pharmacology of imatinib (STI571). Eur J Cancer2002;38(Suppl. 5):S28–S36.

72. Yee KW, Keating A. Advances in targeted therapy for chronic myeloid leukemia. Expert RevAnticancer Ther. 2003;3:295–310.

73. Talpaz M, Shah NP, Kantarjian H, et al. Dasatinib in imatinib-resistant Philadelphiachromosome-positive leukemias. N Engl J Med. 2006;354:2531–2541.

74. Nagar B, Bornmann WG, Pellicena P, et al. Crystal structures of the kinase domain of c-Ablin complex with the small molecule inhibitors PD173955 and imatinib (STI-571). CancerRes. 2002;62:4244–4255.

75. Soverini S, Martinelli G, Iacobucci I, Baccarani M. Imatinib mesylate for the treatment ofchronic myeloid leukemia. Expert Rev Anticancer Ther. 2008;8:853–864.

76. Fausel C. Targeted chronic myeloid leukemia therapy: seeking a cure. Am J Health SystPharm. 2007; 64(24 Suppl. 15):S9–S15.

77. Frame D. New strategies in controlling drug resistance. J Manag Care Pharm. 2007;13(8 Suppl. A):13–17.

78. Druker BJ, Guilhot F, O’Brien SG, et al. Five-year follow-up of patients receiving imatinibfor chronic myeloid leukemia. N Engl J Med. 2006;355:2408–2417.

79. Piccaluga PP, Martinelli G, Baccarani M. Advances in the treatment for haematologicalmalignancies. Expert Opin Pharmacother. 2006;7:721–732.

80. Verweij J, van Oosterom A, Blay JY, et al. Imatinib mesylate (STI-571 Glivec, Gleevec) isan active agent for gastrointestinal stromal tumors but does not yield responses in other soft-tissue sarcomas that are unselected for molecular target: results from an EORTC Soft Tissueand Bone Sarcoma Group Phase II study. Eur J Cancer. 2003;39:2006–2011.

81. Droogendijk HJ, Kluin-Nelemans HJ, van Doormaal JJ, Oranje AP, van de Loosdrecht AA,van Daele PL. Imatinib mesylate in the treatment of systemic mastocytosis: a phase II trial.Cancer. 2006;107:345–351.

82. Gorre ME, Sawyers CL. Molecular mechanisms of resistance to STI571 in chronic myeloidleukemia. Curr Opin Hematol. 2002;9:303–307.

83. Graham SM, Jørgensen HG, Allan E, et al. Primitive, quiescent, Philadelphia-positive stemcells from patients with chronic myeloid leukemia are insensitive to STI571 in vitro. Blood.2002;99:319–325.

84. Alvarez RH, Kantarjian H, Cortes JE. The biology of chronic myelogenous leukemia:implications for imatinib therapy. Semin Hematol. 2007;44(1 Suppl. 1):S4–S14.

85. Bellodi C, Lidonnici MR, Hamilton A, Helgason GV, Soliera AR, Ronchetti M,et al. Targeting autophagy potentiates tyrosine kinase inhibitor-induced cell death inPhiladelphia chromosome-positive cells, including primary CML stem cells. J Clin Invest.2009;119:1109–1123.


86. Gorre ME, Mohammed M, Ellwood K, et al. Clinical resistance to STI-571 can-cer therapy caused by BCR-ABL gene mutation or amplification. Science. 2001;293:876–880.

87. Jarkowski A, Sweeney RP. Nilotinib: a new tyrosine kinase inhibitor for the treatment ofchronic myelogenous leukemia. Pharmacotherapy. 2008;28:1374–1382.

88. Kantarjian HM, Giles F, Gattermann N, et al. Nilotinib (formerly AMN107), a highlyselective BCR-ABL tyrosine kinase inhibitor, is effective in patients with Philadelphiachromosome-positive chronic myelogenous leukemia in chronic phase following imatinibresistance and intolerance. Blood. 2007;110:3540–3546.

89. Weisberg E, Manley PW, Breitenstein W, et al. Characterization of AMN107, a selectiveinhibitor of native and mutant Bcr-Abl. Cancer Cell. 2005;7:129–141. (Erratum in CancerCell. 2005;7:399.)

90. Jorgensen HG, Allan EK, Jordanides NE, et al. Nilotinib exerts equipotent antiprolifer-ative effects to imatinib and does not induce apoptosis in CD34+ CML cells. Blood.2007;109:4016–4019.

91. Copland M, Hamilton A, Elrick LJ, et al. Dasatinib (BMS-354825) targets an earlier progen-itor population than imatinib in primary CML but does not eliminate the quiescent fraction.Blood. 2006;107:4532–4539.

92. Das J, Chen P, Norris D, et al. 2-Aminothiazole as a novel kinase inhibitor template.Structure-activity relationship studies toward the discovery of N-(2-chloro-6-methylphenyl)-2-[[6-[4-(2-hydroxyethyl)-1- piperazinyl)]-2-methyl-4-pyrimidinyl]amino)]-1,3-thiazole-5-carboxamide (dasatinib, BMS-354825) as a potent pan-Src kinase inhibitor. J Med Chem.2006;49:6819–6832.

93. Jorgensen HG, Holyoake TL. Characterization of cancer stem cells in chronic myeloidleukaemia. Biochem Soc Trans. 2007;35(Pt 5):1347–1351.

94. Steinberg M. Dasatinib: a tyrosine kinase inhibitor for the treatment of chronic myelogenousleukemia and Philadelphia chromosome-positive acute lymphoblastic leukemia. Clin Ther.2007;29:2289–2308.

95. Lu J, Tan L, Li P, Gao H, Fang B, Ye S, et al. All-trans retinoic acid promotes neurallineage entry by pluripotent embryonic stem cells via multiple pathways. BMC Cell Biol.2009;10:57.

96. Downing JR. Targeted therapy in leukemia. Mod Pathol. 2008;21(Suppl. 2):S2–S7.97. Zhou GB, Zhang J, Wang ZY, Chen SJ, Chen Z. Treatment of acute promyelocytic leukaemia

with all-trans retinoic acid and arsenic trioxide: a paradigm of synergistic moleculartargeting therapy. Philos Trans R Soc Lond B Biol Sci. 2007;362:959–971.

98. Ohnishi K. PML-RARalpha inhibitors (ATRA, tamibarotene, arsenic trioxide) for acutepromyelocytic leukemia. Int J Clin Oncol. 2007;12:313–317.

99. Redner RL. Variations on a theme: the alternate translocations in APL. Leukemia.2002;16:1927–1932.

100. Fang J, Chen SJ, Tong JH, Wang ZG, Chen GQ, Chen Z. Treatment of acute promyelocyticleukemia with ATRA and As2O3: a model of molecular target-based cancer therapy. CancerBiol Ther. 2002;1:614–620.

101. Sanz MA. Treatment of acute promyelocytic leukemia. Hematol Am Soc Hematol EducProgram. 2006:147–155.

102. Schenone S, Brullo C, Botta M. Small molecules ATP-competitive inhibitors of FLT3: achemical overview. Curr Med Chem. 2008;15:3113–3132.

103. Small D. Targeting FLT3 for the treatment of leukemia. Semin Hematol. 2008;45(3 Suppl. 2):S17–S21.

104. Swerdlow SH, Campo E, Harris NL, et al. WHO Classification of Tumours ofHaematopoietic and Lymphoid Tissues; 2008. Lyon, France: IARC.

105. Melchert M, List A. The thalidomide saga. Int J Biochem Cell Biol. 2007;39:1489–1499.106. Chanan-Khan AA, Cheson BD. Lenalidomide for the treatment of B-cell malignancies.

J Clin Oncol. 2008;26:1544–1552.


107. Teo SK, Stirling DI, Zeldis JB. Thalidomide as a novel therapeutic agent: new uses for anold product. Drug Discov Today. 2005;10:107–114.

108. Galustian C, Dalgleish A. Lenalidomide: a novel anticancer drug with multiple modalities.Expert Opin Pharmacother. 2009;10:125–133.

109. Ataergin SA, Kindwall-Keller T, Berger NA, Lazarus HM. New generation pharmacotherapyin elderly multiple myeloma patients. Expert Opin Pharmacother. 2009;10:81–98.

110. Richardson PG, Mitsiades C, Schlossman R, Ghobrial I, Hideshima T, Munshi N, et al.Bortezomib in the front-line treatment of multiple myeloma. Expert Rev Anticancer Ther.2008;8:1053–1072.

111. Gilardini A, Marmiroli P, Cavaletti G. Proteasome inhibition: a promising strategy fortreating cancer, but what about neurotoxicity? Curr Med Chem. 2008;15:3025–3035.

112. Sterz J, von Metzler I, Hahne JC. The potential of proteasome inhibitors in cancertherapy.Expert Opin Investig Drugs. 2008;17:879–895.

113. Reinis M. BiovaxID, a personalized therapeutic vaccine against B-cell lymphomas. CurrOpin Mol Ther. 2008;10:526–534.

114. Koumarianou A, Kountourakis P, Economopoulos T. Vaccine strategies in the treatmentof low-grade non-Hodgkin lymphoma. J Steroid Biochem Mol Biol. 2008 Apr;109(3–5):230–232.

115. Lee ST, Jiang YF, Park KU. BiovaxID: a personalized therapeutic cancer vaccine for non-Hodgkin’s lymphoma. Expert Opin Biol Ther. 2007;7:113–122.

116. Flowers CR. BiovaxID idiotype vaccination: active immunotherapy for follicular lymphoma.Expert Rev Vaccines. 2007;6:307–317.

117. Cross D, Burmester JK. Gene therapy for cancer treatment: past, present and future. ClinMed Res. 2006;4:218–227.

118. Hambley TW, Hait WN. Is anticancer drug development heading in the right direction?Cancer Res. 2009;69:1259–1262.

119. Stone MR. Targeted agents in AML: much more to do. Best Pract Res Clin Hematol.2007;20:39–48.

Chapter 10Micro-RNAs in Hematologic Malignancies

Muller Fabbri and George A. Calin

Keywords Micro-RNAs · Leukemias · Lymphomas · NoncodingRNAs · Polyadenylated precursor pri-miRNA · Precursor pre-miRNA · RISCcomplex · Cancer-associated genomic regions (CAGRs), Tumor suppressorgenes · Fragile sites · Loss of heterozygosity (LOH) · Insulin-like growth factorreceptor (IGFR) · Chemokine receptor 4 (CXCR4) · Promyelocytic leukemia zincfinger (PLZF) · Burkitt lymphoma

Introduction

Micro-RNAs (miRNAs) are noncoding RNAs (ncRNAs) which regulate geneexpression. MiRNAs are involved in a variety of biological processes, spanningfrom development, differentiation, apoptosis, and proliferation to senescence andmetabolism [1–6]. MiRNA biogenesis is initiated by an RNA polymerase II,which initially transcribes the miRNA gene into a long, capped, and polyadeny-lated precursor, called pri-miRNA [7, 8]. By means of a double-stranded RNA-specific ribonuclease called Drosha, in conjunction with its binding partner DGCR8(DiGeorge syndrome critical region gene 8, or Pasha), the pri-miRNA is processedinto a hairpin RNA precursor (pre-miRNA), about 70–100 nucleotides (nt) long [9].The following step is a translocation of pre-miRNA from the nucleus to the cyto-plasm, by means of Exportin 5. Once in the cytoplasm, the precursor is cleaved intoa 18–24 nt duplex by a ribonucleoprotein complex, composed of a ribonuclease III(Dicer), and TRBP (HIV-1 transactivating response RNA binding protein). Finally,

M. Fabbri (B)Department of Molecular Virology, Immunology and Medical Genetics, The Ohio State UniversityComprehensive Cancer Center, Columbus, OH 43210, USAe-mail: [email protected]

G.A. Calin (B)Departments of Experimental Therapeutics and Cancer Genetics, University of Texas,M.D. Anderson Cancer Center, Houston, TX, USAe-mail: [email protected]


326 M. Fabbri and G.A. Calin

the duplex interacts with a large protein complex called RISC (RNA-induced silenc-ing complex), which includes proteins of the Argonaute family (Ago1-4 in humans).One strand of the miRNA duplex remains stably associated with RISC and becomesthe mature miRNA, which guides the RISC complex mainly (but not exclusively)to the 3’-UTR (3’-untranslated region) of the target mRNAs. According to themiRNA:mRNA degree of base-pair complementarity, the target mRNA can becleaved (in case of perfect Watsonian match) or its translation into protein canbe prevented (in case of imperfect Watsonian match). Figure 10.1 summarizes theevents that occur during miRNA biogenesis. Overall, the effect of miRNAs is tosilence the expression of the target mRNAs either by mRNA cleavage or by transla-tional repression. However, researchers have discovered that miRNAs can actuallyalso increase the expression of a target mRNA [10]. Each miRNA can target sev-eral different transcripts. For instance, it has been demonstrated that a cluster of twomiRNAs (namely miR-15a and miR-16) can affect the expression of about 14% ofthe human genome in a leukemic cell line [11]. In addition, the same mRNA can betargeted by several miRNAs [12].

By using high-throughput profiling methods [13, 14], differences in themiRNome (defined as the full complement of miRNAs in a genome) have beendetected in normal versus pathologic tissues or in the same tissues at differentstages of differentiation. The first evidence of a relationship between miRNAs andhuman cancer is derived from the observation that miRNAs are more frequentlylocated in cancer-associated genomic regions (CAGRs), which include fragile sites

RISCPassengerStrand

RISC

Target mRNA

Translational RepressionmRNA cleavage

Ago Dicer

Ago

Rna Pol II

Exp

ortin

5

TRBP + Dicer

miRNA gene

Pri-miRNA

Pre-miRNA

DGCR8 + Drosha

Cell Nucleus

Pre-miRNA

Cell Cytoplasm

miRNA duplex

Fig. 10.1 Biogenesis of miRNAs

10 Micro-RNAs in Hematologic Malignancies 327

(FRA) where tumor suppressor genes (TSGs) are located and regions of frequentloss of heterogeneity (LOH), deletion, amplification, and translocation. After hav-ing mapped 186 miRNAs and compared their locations to those of previouslyreported nonrandom genetic alterations, it was observed that 52.5% of miRNAsare in CAGRs [15]. Overall 19% of miRNAs are located inside or close to fragilesites (FRA), including FRA in which no known tumor suppressor genes map (e.g.,FRA7H and miR-29a and miR-29b-1). About 43% of miRNAs are in LOH regionsor in regions of amplification [15]. Since then, several groups have identified aber-rancies of the miRNome in almost all human tumors [16–18], and specific signaturesof de-regulated miRNAs have been associated with specific tumors, and some-times harbor prognostic implications [19–24]. In hematology, miRNA expressiondiffers during normal hematopoiesis, and miRNA expression aberrations can leadto pathologic phenotypes. This chapter will focus on the role of miRNAs in humanhematological malignancies, after a brief description of the physiological changesof the miRNome during normal hematopoiesis.

Micro-RNAs in Normal Hematopoiesis

Physiologic variations in miRNA expression levels occur during normalhematopoiesis and affect differentiation and commitment of the multipotent hema-tologic progenitor (MPP). The differentiation of MPP cell into either the commonmyeloid progenitor (CMP) or the common lymphoid progenitor (CLP) cell iscontrolled by miR-128a and miR-181a. On one side miR-146 blocks lymphoid dif-ferentiation, whereas miR-155, -24a, and -17 inhibit myeloid differentiation at anearly stage [25]. The expression of miR-223 is low in CD34+ MPPs and CMPs, butincreases steadily in the granulocyte compartment, while it is downregulated in themonocytes lineage [26]. Fazi et al. demonstrated that miR-223 targets NFI-A andC/EBPa, two transcription factors involved in human granulopoiesis, which, in turn,can silence or activate miR-223 expression, respectively [27]. Despite these find-ings, in miR-223 knockout mice models it has been described increased numbersof granulocyte progenitors in the bone marrow and higher levels of mature circu-lating neutrophils [28]. These effects are probably mediated by downregulation ofeither MEF2c, a transcription factor that promotes myeloid progenitor proliferation,or the insulin-like growth factor receptor (IGFR) [28] and establish a role for miR-223 as a negative regulator of maturation but not differentiation of granulocytes.Overall, an important role for miR-223 in human granulopoiesis can be claimed,albeit further investigations are needed to clarify whether high or low expression ofthis miRNA is associated with myeloid differentiation. By targeting the transcrip-tion factor NFI-A, miR-424 induces monocytic/macrophage differentiation in acutemyeloblastic leukemia (AML) cell lines and in CD34+ MPPs, therefore promotingmyeloid hematopoiesis [29].

A systematic analysis of miRNA expression during erythroid commitment oferythrocyte precursors showed three different miRNA expression patterns: a first


group of miRNAs whose expression progressively reduces during erythroid dif-ferentiation (including miR-150, -155, -221, -222), a group of miRNAs whoseexpression increases (namely miR-451, -16 at late stages), and a group of miRNAswith biphasic trend (miR-339, -378) [30]. Previously, another group showed thaterythroid differentiation of MPPs is paralleled by a progressive downregulation ofmiR-221 and miR-222 and upregulation of their direct target: the Kit receptor [31].Moreover, Georgantas et al. showed that miR-155 transduction in normal primaryhuman CD34+ cells inhibits both myeloid and erythroid colony formation [25]. Inaddition, the C57BL6 mouse model transplanted with mice MPPs overexpressingmiR-155 develops a myeloproliferative disorder associated with a decrease in theerythroid/megakaryocytic lineage in the bone marrow [32], supporting other stud-ies indicating an miR-155 block of erythrocytic/megakaryocytic differentiation [25,30, 33]. MiR-451 belongs to an miR-144/451 cluster, whose expression is under thecontrol of the master erythrocyte transcription factor GATA-1 [34]. In a zebra fishembryo model, silencing of miR-451 with anti-miRNA molecules resulted in nor-mal erythroid precursors with strong impairment of their development into maturecirculating red cells. Conversely, no alterations were observed when miR-144 wassilenced, revealing an miR-451-specific function in the late stages of erythropoiesis[34], and supporting that high levels of miR-451 are needed in order for a normalerythropoiesis to occur.

Megakaryocyte differentiation occurs in parallel with the downregulation of apanel of 20 miRNAs, which includes miR-10a and miR-130a [33]. These twomiRNAs target MAFB and HOXA-1, two genes overexpressed during megakary-opoiesis, indicating that miRNAs are responsible for the regulation of their expres-sion level during megakaryocytic commitment [33]. Labbaye et al. have shown thatin megakaryopoietic cultures of CD34+ progenitors, high levels of the promye-locytic leukemia zinc finger (PLZF) protein transactivate miR-146a, which inturn directly silences the chemokine receptor 4 (CXCR4) [35], a key factor formegakaryocytic proliferation, differentiation, and maturation [36].

MiRNAs are differentially expressed also during normal lymphoiddifferentiation.

Ectopic expression of miR-181a in hematopoietic progenitor cells which weresubsequently transplanted into lethally irradiated mice resulted in increased B cellsand a paucity of T lymphocytes [37]. More recently, Neilson et al. have shownthat high expression of miR-181 occurs also in the thymus and in the DP (double-positive CD4+/CD8+) stage of thymocyte development [38]. MiR-181a directlytargets CD69. Since the CD69 signaling pathway affects the egress of lymphocytesfrom the thymus [39], it can be postulated that the ectopic expression of miR-181passing the DP stage, as performed by Chen et al. might have resulted in low CD69-expressing CD4 or CD8 lymphocytes [37]. As a result, those cells have a reducedability to leave the thymus, leading to a general decrease in circulating T cells. Inaddition to CD69, miR-181a directly targets also BCL2 and TCR-α, whose levels ofexpression are known to increase in DP thymocytes following positive selection tothe CD4 or CD8 stage [38]. Reduced levels of TCR-α shift the threshold for positiveand negative selection, while the targeting of BCL2 upon positive selection would


result in cell death. These two mechanisms contribute to the overall reduced levelsof peripheral T cells upon forced expression of miR-181. An miRNA regulating thetransition from pro-B to pre-B-cell stage is miR-150 [40]. Selectively expressed inmature, resting B and T cells, but not in their progenitors [41], miR-150 significantlyreduces the number of mature B cells in spleen, lymph nodes, and peripheral blood,when ectopically expressed in murine hematopoietic stem and progenitor cells [40].These effects are dependent on miR-150 direct target c-MYB, a transcription factorwhich directs multiple steps of lymphocyte development [42]. By silencing c-MYB,miR-150 induces apoptosis of pro-B cells [42]. Deletion of Dicer at an early B-cellstage blocks almost completely the pro-B to pre-B-cell transition, which coincideswith a significant upregulation of the pro-apoptotic protein Bim [43]. At least inpart responsible for this effect is the miR-17-92 cluster, since a targeted deletion ofthe cluster leads to increased levels of the pro-apoptotic protein Bim in mice, andinhibits B-cell development at the pro-B to pre-B transition [44].

Overall, miRNAs are involved in normal hematopoiesis and act as “fine tuners”of their target expression levels, therefore orchestrating commitment and differen-tiation of the pluripotent hematopoietic progenitors. These physiological mecha-nisms are aberrant in cancer and contribute to the pathogenesis of hematologicalmalignancies.

Micro-RNAs in Lymphomas

MiRNAs are involved in human lymphomagenesis (Table 10.1). Tam et al. initiallyobserved that the final part of the B-cell integration cluster (BIC) noncoding RNA(ncRNA), where miR-155 is located [45], accelerates MYC-mediated lymphoma-genesis in a chicken model [46]. Subsequently, high levels of BIC/miR-155 weredescribed in pediatric Burkitt lymphoma (BL) [47], but not in the adult primarycases [48], probably indicating a specific age-dependent role of this miRNA-basedon the age of onset of BL. In the B-cell-specific miR-155 transgenic (TG) mousemodel an acute lymphoblastic leukemia/high-grade lymphoma at approximately9 months of age was described [49]. These malignancies are preceded by apolyclonal pre-B-cell proliferation, have variable clinical presentation, are trans-plantable, and develop oligo/monoclonal expansion [49]. Recently, it was shownthat in these TG mice the B-cell precursors with the highest miR-155 expres-sion were at the origin of the leukemias [50]. Moreover, by directly targeting theSrc homology 2 domain-containing inositol-5-phosphatase (SHIP) and the CCAATenhancer-binding protein beta (C/EBPbeta), two key regulators of the interleukin-6signaling pathway, miR-155 triggers a chain of events that promotes the accumula-tion of large pre-B cells and acute lymphoblastic leukemia/high-grade lymphoma[50]. Two different groups have studied miR-155 knockout (KO) mice modelsand have demonstrated that lack of this miRNA switches cytokine productiontoward TH2 differentiation [51], and also compromises the ability of dendritic cells(DC) to activate T cells, because of a defective antigen presentation or abnormalco-stimulatory functions [52].


Table 10.1 Most frequently de-regulated miRNAs in human leukemias and lymphomas

miRNAChromosomallocation Deregulation Diseases Targets

miR-155 21q21.3 Up BL, DLBCL, HL,CLL, AML(FLT-IDT+)

SHIP, C/EBP-β

miR-17-92 cluster 13q31.3 Up B-cell lymphomas,CML, ALL

PTEN, BIM, E2F1

miR-106a-363cluster

Xq26.2 Up B-cell lymphomas,T-cell leukemias

BIM, TGF-βsignaling

miR-106b-25cluster

7q22.1 Up B-cell lymphomas BIM, TGF-βsignaling

miR-143/145cluster

5q33.1 Down B-cell lym-phomas/leukemias

ERK5

miR-9 1q22 (miR-9-1)5q14.3 (miR-9-2)15q26.1

(miR-9-3)

Up HL PRDM1/BLIMP-1

let-7a 9q22.32(let-7a-1)

11q24.1(let-7a-2)

11q13.31(let-7a-3)

Up HL PRDM1/BLIMP-1

miR-135a 3p21.2(miR-135a-1)

13q23.1(miR-135a-2)

Down HL JAK2

miR-15a/16-1cluster

13q14.2 Down Indolent CLL BCL2, MCL1

miR-29b 7q32.3(miR-29b-1)

1q32.2(miR-29b-2)

Down Aggressive CLL,AML

TCL1, DNMT3A,DNMT3B, SP1,MCL1,CXXC6, CDK6

miR-181b 1q32.1(miR-181b-1)

9q33.3(miR-181b-2)

Down Aggressive CLL,aggressive AML

TCL1, TLR, andIL-6 pathways

miR-203 14q32.33 Down CML ABL1miR-10a 17q21.32 Down CML USF2miR-128b 3p22.3 Up ALL UnknownmiR-124a 8p23.1

(miR-124a-1)8q12.3

(miR-124a-2)20q13.33

(miR-124a-3)

Down ALL CDK6

miR-204 9q21.11 Down AML HOXA10

BL, Burkitt lymphoma; DLBCL, diffuse large B-cell lymphoma; HL, Hodgkin lymphoma; CLL,chronic lymphocytic leukemia; AML, acute myeloid leukemia; CML, chronic myeloid leukemia;ALL, acute lymphoblastic leukemia. For the targets legend, see main text.


High levels of miR-155 have been described also in diffuse large B-cell lym-phoma (DLBCL), the most frequent lymphoma in adults worldwide [53, 54]. Bycomparing miR-155 levels in the activated B-cell phenotype of DLBCL (ABC-DLBCL), versus the germinal center B-cell-like phenotype (GCB-DLBCL), miR-155 was significantly higher in the ABC phenotype [45, 53]. Since ABC-DLBCLand GCB-DLBCL have 5-year survival rates of 30 and 59%, respectively [55],miR-155 expression in DLBCL has a prognostic value. A correlation between miR-155 and NF-kB expression was found in DLBCL cell lines and patients [56]. Inaddition to miR-155, high levels of miR-21 and miR-221 are also associated withABC-DLBCL and severe prognosis [51]. Roehle et al. identified miRNA-specificsignatures for DLBCLs and follicular lymphomas (FLs) [54], and showed that fourmiRNAs (namely miR-330, -17-5p, -106a, and -210) can accurately differentiateDLBCL, FL, and reactive lymph nodes with an overall accuracy of 98% [54].Noteworthly, miR-17-5p and miR-106a belong to two paralogous clusters locatedon chromosome 13 and X, respectively, with a well-established oncogenic role inseveral human malignancies, both solid and hematologic [57].

The miR-17-92 cluster is located at 13q31-32, a region frequently amplified inmalignant B-cell lymphomas [58], and is overexpressed in over 60% of B-cell lym-phoma patients [59]. Overexpression of the cluster in murine pluripotent cells fromMYC transgenic mice accelerates lymphomagenesis [59]. The oncogenic poten-tial of this miRNA cluster is supported also by B-cell miR-17-92 cluster TG micemodels, in which a higher than expected rate of lymphoproliferative disorders andautoimmunity and premature death did occur [60]. The molecular bases of theobserved phenotype reside, at least in part, in the direct targeting of the TSG PTEN,and the pro-apoptotic Bim protein, which controls B-lymphocyte apoptosis [60].Members of the miR-17-92 cluster have homologues in two other clusters: on chro-mosome 7 (the miR-106b-25 cluster) and on chromosome X (the miR-106a-363cluster). The OG c-MYC transactivates both clusters on chromosomes 7 and 13[61] in addition to E2F1, a transcription factor which promotes cell cycle progres-sion [62]. In turn, E2F1 regulates the host genes for the miR-106b-25 and for themiR-17-92 clusters. The miR-106a-363 polycistron is also overexpressed in 46% ofacute and chronic human T-cell leukemias [63], claiming a role in leukemogenesis.Interestingly, two of the three clusters (namely miR-106b-25 and miR-17-92) inter-fere with the transforming growth factor beta (TGF-β) signaling [64], a pathwaywhich is inhibited in several tumors [65]. Moreover, Ventura et al. have shown thatthe miR-17-92 and miR-106b-25 double-knockout mouse model has a more severephenotype than the miR-17-92 single-knockout mouse model [44], suggesting thatboth clusters control apoptosis.

Also miR-143 and miR-145 are frequently downregulated in B-cell lymphomasand leukemias [66]. In non-Hodgkin lymphoma cell lines, restoration of these twomiRNAs induced a dose-dependent growth inhibitory effect which was associatedwith downregulation of Erk5 [66], a recently characterized MAPK, most similar tothe well-studied ERK1/2 subfamily [67].

MiRNA expression in Hodgkin lymphoma (HL) has been object of some studies.Navarro et al. identified a distinctive signature of 25 miRNAs discriminating HL


from reactive lymph nodes and 36 miRNAs differentially expressed in the nodu-lar sclerosis and mixed cellularity subtypes [68]. Three miRNAs (namely miR-96,-128a, and -128b) were selectively downregulated in EBV+ HL [68]. Since onlyone of the miRNAs differentially expressed in EBV+ cases was also included inthe 25-miRNA signature distinguishing HL from reactive lymph nodes, it seemssafe to conclude that EBV is not a primary transforming event in HL. Among theupregulated miRNAs in HL there are miR-9 [68, 69] and let-7a [69], which directlytarget PRDM1/blimp-1 [69], a master regulator in terminal B-cell differentiation[70]. A more recent study has compared miRNA profiles of microdissected Reed–Sternberg cells and Hodgkin cell lines versus CD77+ B cells [71]. In this study aprofile of 12 over and 3 underexpressed miRNAs was identified [71], showing only apartial overlap with Navarro’s profile. This discrepancy might be due to the differentprocedure used to collect the HL cells. MiRNAs have also prognostic implicationsin HL. Low expression of miR-135a has been associated to higher relapse risk andshorter disease-free survival for HL patients [72]. The TSG nature of miR-135a isdetermined by its direct targeting of JAK2, an activator of the antiapoptotic geneBcl-XL [72]. Also in HL higher expression of miR-155 has been reported [69, 71,73], although the function of this upregulation in Hodgkin Reed–Sternberg cells isstill poorly understood.

Micro-RNAs in Leukemias

MiRNAs are also involved in leukemogenesis (Table 10.1). Chronic lymphocyticleukemia (CLL) is the most common leukemia among adults in the Western worldand is characterized by slow accumulation in blood, bone marrow, and lymphatic tis-sue of small, non-proliferating, mature B lymphocytes, which display typical surfacemarkers such as CD19 and CD20 in addition to CD5 [74].

The majority of CLLs are characterized by hemizygous and/or homozygous dele-tion of the genomic region 13q14.3 [75], where a cluster of miRNAs (namely themiR-15a/16-1 cluster) is located [76]. It has been demonstrated that both miR-15a and miR-16-1 are deleted or downregulated in approximately 68% of CLLcases [76], suggesting a role as TSG for this miRNA cluster. Indeed, miR-15aand miR-16 directly target the antiapoptotic BCL2 [77], a protein which is overex-pressed in the majority of CLL malignant B cells [78], and it is believed to mediatethe anti-tumoral effect of these miRNAs. Restoration of miR15a/16-1 expressionin the leukemic MEG-01 cell line (which recapitulates the genetic abnormalitiesof CLL with 13q deletion) leads to apoptosis and inhibition of tumor growth inxenograft mice models, further corroborating a role for miR15a/16-1 as TSGs [77,79]. Interestingly, the pattern of these miRNA cluster-controlled genes includes bothoncogenes (OGs) and TSGs, suggesting that miRNAs cannot be simply described asOGs or TSGs, but dual in nature [80], probably depending on the specific microen-vironment in which they act, which differs among cell types and species. MiRNAsalso harbor prognostic implications in CLL, since a specific miRNA signature can


distinguish between the indolent form of CLL (characterized by low levels of ZAP-70- and IgVH-mutated status), and the aggressive form [81]. Finally, miRNAs areinvolved also in familial CLLs, since a germ-line mutation in pre-miR-16 sequence,which causes low levels of micro-RNA expression both in vitro and in vivo wasassociated with deletion of the normal allele in leukemic cells of two CLL patients,one of which with a family history of CLL and breast cancer [81]. Intriguingly,in the New Zealand Black mouse strain model characterized by spontaneouslyoccurring late-onset CLL [82], Raveche et al. described a point mutation adjacentto the miR-16-1 locus, which is responsible for lower expression of this miRNAin this CLL-prone mouse model [83], further suggesting that reduced levels ofmiR-15a/16-1 contribute to CLL genesis.

High expression of TCL1 (T-cell leukemia/lymphoma 1A) is associated withaggressive CLL. Pekarsky et al. have shown that miR-29b and miR-181b directlytarget TCL1, therefore impacting on the protein kinase AKT (v-akt murine thymomaviral oncogene homolog 1) pathways which affect cell survival, proliferation, anddeath [84]. Another study, conducted in 110 patients, showed a correlation betweenlow levels of miR-29c and poor prognosis CLL [85]. Interestingly, the authors foundthe first evidence of a specific threshold of expression for miR-29c and miR-223,able to predict treatment-free survival (TFS) and overall survival (OS) [85].

Finally, high levels of miR-155 have been described also in CLL versus normalCD19+ B cells [86].

In chronic myeloid leukemia (CML), the miR-17-92 cluster seems to have acentral role. Indeed, the cluster is transactivated both by c-MYC and by BCR–ABL1, the fusion protein which results from the reciprocal translocation t(9;22),hallmark of the disease (Philadelphia chromosome) [87]. The BCR–ABL1–MYCcomplex can transactivate miR-17-92 only in early chronic phase, but not in blastcrisis CML CD34+ cells [87], suggesting a role for miR-17-92 cluster in the earlyphases of CML pathogenesis. MiR-203 directly targets ABL1, and high expres-sion of this miRNA inhibits cancer cell proliferation in an ABL1-dependent manner[88]. In turn, both genetic and epigenetic mechanisms coordinately inactivate thismiRNA, and a high rate of miR-203 promoter hypermethylation has been describedin Ph+ tumors, including B-cell ALLs, primary CMLs, and cultured CML cell lines,whereas no methylation was observed in other hematologic tumors that do not carryABL1 alterations [88]. Overall, miR-203 and miR-17-92 cluster expression seemsto be intertwined and acts as key player in CML pathogenesis. Finally, a role formiR-10a also emerged in CML. In a group of 85 newly diagnosed CML patients,miR-10a was found downregulated in 71% of cases, and an inverse correlationwith the expression of the oncogenic upstream stimulatory factor 2 (USF2) wasdescribed [89].

Acute lymphoblastic leukemia (ALL) is the most common childhood cancer.Zanette et al. compared miRNA expression profile in seven ALL patients versusnormal CD19+ B cells from six healthy individuals and described the miR-17-92cluster as upregulated in ALL samples [90]. Recently, a role for miR-17-92 clusterhas been described also in the less common T-cell subtype of ALL [91]. In anotherstudy, Mi et al. have identified a specific miRNA signature able to discriminate ALL


from AML (acute myeloid leukemia) with high accuracy. In particular four miRNAs(namely miR-128a and -128b upregulated in ALL versus AML and let-7b and miR-223 downregulated in ALL versus AML) can differentially diagnose between theacute leukemias with an accuracy rate of 98% [92]. Moreover, miR-128b was alsoupregulated in ALL versus normal CD19+ cells, suggesting a high specificity forALL [92]. The leukemogenic mechanism of miR-128b is still poorly understood.Epigenetic factors affect the expression of miRNAs in ALL and harbor prognosticimplications. In a recent report conducted on 353 ALL samples, Roman-Gomezet al. observed that 65% of patients had at least one miRNA methylated, and thismethylation status was associated with reduced disease-free and overall survival[93]. In particular miR-124a is frequently downregulated because of its promoterhypermethylated status, and this contributes to the development of the malignantphenotype, since this miRNA directly targets cyclin-dependent kinase 6 (CDK6),an oncogene that promotes cell proliferation by inducing phosphorylation of Rb[94]. Also histone modifications regulate miRNA expression profile in ALL. In thisleukemia miR-22 expression can indeed be restored by treatment with trichostatinA, a well-known histone deacetylase inhibitor [95]. In childhood pre-B ALL patientshigh levels of miR-222, -339, and -142-3p, paralleled by low expression of miR-451and miR-373∗, were also described [96].

In acute myeloid leukemia (AML), high levels of miR-191 and miR-199a seemto have prognostic implications, since they correlate with reduced overall anddisease-free survival [97]. Specific miRNA signatures are also associated with bal-anced 11q23 translocations, isolated trisomy 8, and FLT3-ITD (fms-like tyrosinekinase 3 internal tandem duplications) mutations [97]. Also in AML with normalkaryotype (which represents about 30–40% of all AMLs), mutations of NPM1(nucleophosmin-1) and FLT3-ITD occur [98], and a specific set of miRNAs areable to differentiate these mutation statuses in normal karyotype AMLs [99]. In par-ticular, high levels of miR-10a, -10b, several let-7, and miR-29 family members,as well as downregulation of miR-204, characterize NPM1-mutated versus NPM1-unmutated cases [99]. Given that miR-204 directly targets HOXA10, the high levelsof HOX proteins observed in NPM1-mutated AMLs might derive, at least in part,from low expression of HOX-regulating miRNAs [99]. Despite an overexpressionof miR-155 is associated with FLT3-ITD+ status, there is evidence that this upreg-ulation is actually independent from FLT3 signaling [99]. Therefore, a combinedtherapy with anti-miR-155 molecules and FLT3-ITD pathway inhibitors might rep-resent a rationale approach for this subset of AML patients. In AML with normalcytogenetics but high-risk molecular features (such as FLT3-ITD+, or unmutatedNPM1, or both) low expression of miR-181 family contributes to an aggressiveAML phenotype through mechanisms associated with the activation of pathwayscontrolled by toll-like receptors and interleukin-1b [100]. The t(8;21) transloca-tion, which is the most common chromosomal aberrancy in AML, generates theAML1/ETO fusion oncoprotein. This fusion product causes epigenetic silencing ofmiR-223, by recruiting chromatin remodeling enzymes at an AML1-binding siteon the pre-miR-223 gene [27]. By silencing miR-223 expression, the oncoproteininhibits the differentiation of myeloid precursors, therefore actively contributing


to the pathogenesis of this myeloproliferative disorder. More recently, it has beendemonstrated that miR-29b is a key player in the epigenetics and AML. Both in celllines and in primary samples, miR-29b directly regulates the two “de novo” DNAmethyltransferases (DNMT3A and DNMT3B) [101], as previously observed also inlung cancer [102], and indirectly modulates the levels of the “maintenance” DNMT(DNMT1), by directly targeting DNMT1 activator Sp1 [101]. These effects lead tothe re-expression of epigenetically silenced TSGs, such as ESR1 (estrogen receptoralpha) and p15(INK4b) [101]. Moreover, restoration of miR-29b in AML cell linesand primary samples suppresses the expression of OGs such as MCL1, CXXC6,and CDK6, which are direct targets of miR-29b [103].

Overall, miRNAs play an important role in all kinds of human leukemias, byaffecting the expression levels of important genes which control hematopoiesis.

Concluding Remarks

From the first evidence that aberrancies of the miRNome occur in hematologi-cal malignancies to the progressive understanding of the molecular meaning ofthese aberrations, scientists are progressively reaching the threshold of introducingmiRNA-based therapies into the common clinical management of these malignan-cies. A better understanding of the many targets of the most frequently and widelyde-regulated miRNAs has been replaced by a more pathway-based kind of inquiry,aimed at defining which molecular pathways are mainly affected by the miRNomeabnormality. This approach has proven to be very successful, providing pathogeneticand prognostic information of the utmost importance. Finally, the encouraging (andin some cases even astonishing) results obtained by miRNA-based treatments inxenograft mouse models and in transgenic and knockout mice models have pro-vided the final proof not only that miRNAs can be used to treat cancer but alsoin some cases that they should be used as therapeutics. The next challenge will beto determine how to effectively combine miRNAs and more traditional anticancerdrugs, in order to achieve better efficacy and/or lower incidence of side effects. Anot so far future will answer these questions.

References

1. Ambros V, Lee RC. Identification of microRNAs and other tiny noncoding RNAs by cDNAcloning. Methods Mol Biol. 2004;265:131–158.

2. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cel1.2004;116:281–297.

3. He L, Hannon GJ. MicroRNAs: small RNAs with a big role in gene regulation. Nat RevGenet. 2004;5:522–531.

4. Plasterk RH. Micro RNAs in animal development. Cell. 2006;124:877–881.5. Pasquinelli AE, Hunter S, Bracht J. MicroRNAs: a developing story. Curr Opin Genet Dev.

2005;15:200–205.6. Carleton M, Cleary MA, Linsley PS. MicroRNAs and cell cycle regulation. Cell Cycle.

2007;6:2127–2132.


7. Lee Y, Kim M, Han J, et al. MicroRNA genes are transcribed by RNA polymerase II. EMBOJ. 2004;23:4051–4060.

8. Cai X, Hagedorn CH, Cullen BR. Human microRNAs are processed from capped,polyadenylated transcripts that can also function as mRNAs. RNA. 2004;10:1957–1966.

9. Cullen BR. Transcription and processing of human microRNA precursors. Mol Cell.2004;16:861–865.

10. Vasudevan S, Tong Y, Steitz JA. Switching from repression to activation: microRNAs canup-regulate translation. Science. 2007;318:1931–1934.

11. Calin GA, Cimmino A, Fabbri M, et al. MiR-15a and miR-16-1 cluster functions in humanleukemia. Proc Natl Acad Sci U S A. 2008;105:5166–5171.

12. Vatolin S, Navaratne K, Weil RJ. A novel method to detect functional microRNA targets. JMol Biol. 2006;358:983–996.

13. Liu CG, Calin GA, Meloon B, et al. An oligonucleotide microchip for genome-widemicroRNA profiling in human and mouse tissues. Proc Natl Acad Sci USA. 2004;101:9740–9744.

14. Lu J, Getz G, Miska EA, et al. MicroRNA expression profiles classify human cancers.Nature. 2005;435:834–838.

15. Calin GA, Sevignani C, Dumitru CD, et al. Human microRNA genes are frequentlylocated at fragile sites and genomic regions involved in cancers. Proc Natl Acad Sci USA.2004;101:2999–3004.

16. Fabbri M, Croce CM, and Calin GA. MicroRNAs in the ontogeny of leukemias andlymphomas. Leuk. Lymphoma. 2009;50:160–170.

17. Garzon R, Fabbri M, Cimmino A, et al. MicroRNA expression and function in cancer. TrendsMol Med. 2006;12:580–587.

18. Croce CM. Causes and consequences of microRNA dysregulation in cancer. Nat. Rev.2009;10:704–714.

19. Calin GA, Croce CM. MicroRNA signatures in human cancers. Nat. Rev. 2006;6:857–866.20. Pekarsky Y, Calin GA, Aqeilan R. Chronic lymphocytic leukemia: molecular genetics and

animal models. Curr Top Microbiol Immunol. 2005;294:51–70.21. Barbarotto E, Schmittgen TD, Calin GA. MicroRNAs and cancer: profile, profile, profile.

Int J Cancer. 2008;122:969–977.22. Khoshnaw SM, Green AR, Powe DG, et al. MicroRNA involvement in the pathogenesis and

management of breast cancer. J Clin Pathol. 2009;62:422–428.23. Lowery AJ, Miller N, Devaney A, et al. MicroRNA signatures predict oestrogen receptor,

progesterone receptor and HER2/neu receptor status in breast cancer. Breast Cancer Res.2009;11:R27.

24. Ortholan C, Puissegur MP, Ilie M, et al. MicroRNAs and lung cancer: new oncogenes andtumor suppressors, new prognostic factors and potential therapeutic targets. Curr Med Chem.2009;16:1047–1061.

25. Georgantas RW, Hildreth R, Morisot S, et al. CD34+ hematopoietic stem-progenitor cellmicroRNA expression and function: a circuit diagram of differentiation control. Proc NatlAcad Sci USA. 2007;104:2750–2755.

26. Garzon R, Pichiorri F, Palumbo, et al. MicroRNA gene expression during retinoic acid-induced differentiation of human acute promyelocytic leukemia. Oncogene. 2007;26:4148–4157.

27. Fazi F, Racanicchi S, Zardo G, et al. Epigenetic silencing of the myelopoiesis regulatormicroRNA-223 by the AML1/ETO oncoprotein. Cancer Cell. 2007;12:457–466.

28. Johnnidis JB, Harris MH, Wheeler RT, et al. Regulation of progenitor cell proliferation andgranulocyte function by microRNA-223. Nature. 2008;451:1125–1129.

29. Fontana L, Pelosi E, Greco P, et al. MicroRNAs 17-5p-20a-106a control monocytopoiesisthrough AML1 targeting and M-CSF receptor upregulation. Nat Cell Biol. 2007;9:775–787.


30. Bruchova H, Yoon D, Agarwal AM, et al. Regulated expression of microRNAs in normaland polycythemia vera erythropoiesis. Exp Hematol. 2007;35:1657–1667.

31. Felli N, Fontana L, Pelosi E, et al. MicroRNAs 221 and 222 inhibit normal erythropoiesisand erythroleukemic cell growth via kit receptor down-modulation. Proc Natl Acad Sci USA.2005;102:18081–18086.

32. O’Connell RM, Rao DS, Chaudhuri AA, et al. Sustained expression of microRNA-155in hematopoietic stem cells causes a myeloproliferative disorder. J Exp Med. 2008;205:585–594.

33. Garzon R, Pichiorri F, Palumbo T, et al. MicroRNA fingerprints during human megakary-ocytopoiesis. Proc Natl Acad Sci USA. 2006;103:5078–5083.

34. Dore LC, Amigo JD, Dos Santos CO, et al. A GATA-1-regulated microRNA locus essentialfor erythropoiesis. Proc Natl Acad Sci USA. 2008;105:3333–3338.

35. Labbaye C, Spinello I, Quaranta MT, et al. A three-step pathway comprising PLZF/miR-146a/CXCR4 controls megakaryopoiesis. Nat Cell Biol. 2008;10:788–801.

36. Pang L, Weiss MJ, and Poncz M Megakaryocyte biology and related disorders. J Clin Invest.2005;115:3332–3338.

37. Chen CZ, Li L, Lodish HF, et al. MicroRNAs modulate hematopoietic lineage differentia-tion. Science. 2004;303:83–86.

38. Neilson JR, Zheng GX, Burge CB, et al. Dynamic regulation of miRNA expression inordered stages of cellular development. Genes Dev. 2007;21:578–589.

39. Nakayama T, Kasprowicz DJ, Yamashita M, et al. The generation of mature, single-positivethymocytes in vivo is dysregulated by CD69 blockade or overexpression. J Immunol.2002;168:87–94.

40. Zhou B, Wang S, Mayr C, et al. miR-150, a microRNA expressed in mature B and T cells,blocks early B cell development when expressed prematurely. Proc Natl Acad Sci USA.2007;104:7080–7085.

41. Monticelli S, Ansel KM, Xiao C, et al. MicroRNA profiling of the murine hematopoieticsystem. Genome Biol. 2005;6:R71.

42. Xiao C, Calado DP, Galler G, et al. MiR-150 controls B cell differentiation by targeting thetranscription factor c-Myb. Cell. 2007;131:146–159.

43. Koralov SB, Muljo SA, Galler GR, et al. Dicer ablation affects antibody diversity and cellsurvival in the B lymphocyte lineage. Cell. 2008;132:860–874.

44. Ventura A, Young AG, Winslow MM, et al. Targeted deletion reveals essential and overlap-ping functions of the miR-17 through 92 family of miRNA clusters. Cell. 2008;132:875–886.

45. Eis PS, Tam W, Sun L, et al. Accumulation of miR-155 and BIC RNA in human B celllymphomas. Proc Natl Acad Sci USA. 2005;102:3627–3632.

46. Tam W, Hughes SH, Hayward WS, et al. Avian bic, a gene isolated from a common retroviralsite in avian leukosis virus-induced lymphomas that encodes a noncoding RNA, cooperateswith c-myc in lymphomagenesis and erythroleukemogenesis. J Virol. 2002;76:4275–4286.

47. Metzler M, Wilda M, Busch K, et al. High expression of precursor microRNA-155/BICRNA in children with Burkitt lymphoma. Genes Chromosomes Cancer. 2004;39:167–169.

48. Kluiver J, Haralambieva E, de Jong D, et al. Lack of BIC and microRNA miR-155 expressionin primary cases of Burkitt lymphoma. Genes Chromosomes Cancer. 2006;45:147–153.

49. Costinean S, Zanesi N, Pekarsky Y, et al. Pre-B cell proliferation and lymphoblasticleukemia/high-grade lymphoma in E(mu)-miR155 transgenic mice. Proc Natl Acad SciUSA. 2006;103:7024–7029.

50. Costinean S, Sandhu SK, Pedersen IM, et al. Src homology 2 domain-containing inositol-5-phosphatase and CCAAT enhancer-binding protein beta are targeted by miR-155 in B cellsof Emicro-MiR-155 transgenic mice. Blood. 2009;114:1374–1382.

51. Thai TH, Calado DP, Casola S, et al. Regulation of the germinal center response bymicroRNA-155. Science. 2007;316:604–608.

52. Rodriguez A, Vigorito E, Clare S, et al. Requirement of bic/microRNA-155 for normalimmune function. Science. 2007;316:608–611.


53. Lawrie CH, Soneji S, Marafioti T, et al. Microrna expression distinguishes between germinalcenter B cell-like and activated B cell-like subtypes of diffuse large B cell lymphoma. Int JCancer. 2007;121:1156–1161.

54. Roehle A, Hoefig KP, Repsilber D, et al. MicroRNA signatures characterize diffuse largeB-cell lymphomas and follicular lymphomas. Br J Haematol. 2008;142:732–744.

55. Rosenwald A, Wright G, Leroy K, et al. Molecular diagnosis of primary mediastinal B celllymphoma identifies a clinically favorable subgroup of diffuse large B cell lymphoma relatedto Hodgkin lymphoma. J Exp Med. 2003;198:851–862.

56. Rai D, Karanti S, Jung I, et al. Coordinated expression of microRNA-155 and predictedtarget genes in diffuse large B-cell lymphoma. Cancer Genet Cytogenet. 2008;181:8–15.

57. Mendell JT. miRiad roles for the miR-17-92 cluster in development and disease. Cell.2008;133:217–222.

58. He L, Thomson JM, Hemann MT, et al. A microRNA polycistron as a potential humanoncogene. Nature. 2005;435:828–833.

59. Ota A, Tagawa H, Karnan S, et al. Identification and characterization of a novel gene,C13orf25, as a target for 13q31-q32 amplification in malignant lymphoma. Cancer Res.2004;64:3087–3095.

60. Xiao C, Srinivasan L, Calado DP, et al. Lymphoproliferative disease and autoimmunityin mice with increased miR-17-92 expression in lymphocytes. Nat Immunol. 2008;9:405–414.

61. O’Donnell KA, Wentzel EA, Zeller KI, et al. c-Myc-regulated microRNAs modulate E2F1expression. Nature. 2005;435:839–843.

62. Leone G, DeGregori J, Sears R, et al. Myc and Ras collaborate in inducing accumulation ofactive cyclin E/Cdk2 and E2F. Nature. 1997;387:422–426.

63. Landais S, Landry S, Legault P, et al. Oncogenic potential of the miR-106-363 cluster andits implication in human T-cell leukemia. Cancer Res. 2007;67:5699–5707.

64. Petrocca F, Visone R, Onelli MR, et al. E2F1-regulated microRNAs impair TGFbeta-dependent cell-cycle arrest and apoptosis in gastric cancer. Cancer Cell. 2008;13:272–286.

65. Derynck R, Akhurst RJ, Balmain A. TGF-beta signaling in tumor suppression and cancerprogression. Nat Genet. 2001;29:117–129.

66. Akao Y, Nakagawa Y, Kitade Y, et al. Downregulation of microRNAs-143 and -145 in B-cellmalignancies. Cancer Sci. 2007;98:1914–1920.

67. Nishimoto S, Nishida E MAPK signalling: ERK5 versus ERK1/2. EMBO Rep. 2006;7:782–786.

68. Navarro A, Gaya A, Martinez A, et al. MicroRNA expression profiling in classical Hodgkinlymphoma. Blood. 2007;111:2825–2832.

69. Nie K, Gomez M, Landgraf P, et al. MicroRNA-mediated down-regulation ofPRDM1/Blimp-1 in Hodgkin/Reed-Sternberg cells: a potential pathogenetic lesion inHodgkin lymphomas. Am J Pathol. 2008;173:242–252.

70. Pasqualucci L, Compagno M, Houldsworth J, et al. Inactivation of the PRDM1/BLIMP1gene in diffuse large B cell lymphoma. J Exp Med. 2006;203:311–317.

71. Van Vlierberghe P, De Weer A, Mestdagh P, et al. Comparison of miRNA profiles ofmicrodissected Hodgkin/Reed-Sternberg cells and Hodgkin cell lines versus CD77+ B-cellsreveals a distinct subset of differentially expressed miRNAs. Br J Haematol. 2009;147:686–690.

72. Navarro A, Diaz T, Martinez A, et al. Regulation of JAK2 by miR-135a: prognostic impactin classic Hodgkin lymphoma. Blood. 2009;114:2945–2951.

73. Kluiver J, Poppema S, de Jong D, et al. BIC and miR-155 are highly expressed inHodgkin, primary mediastinal and diffuse large B cell lymphomas. J Pathol. 2005;207:243–249.

74. Cheson BD, Bennett JM, Grever M, et al. National Cancer Institute-sponsored WorkingGroup guidelines for chronic lymphocytic leukemia: revised guidelines for diagnosis andtreatment. Blood. 1996;15:4990–4997.


75. Bullrich F, Fujii H, Calin G, et al. Characterization of the 13q14 tumor suppressor locus inCLL: identification of ALT1, an alternative splice variant of the LEU2 gene. Cancer Res.2001;61:6640–6648.

76. Calin GA, Dumitru CD, Shimizu M, et al. Frequent deletions and down-regulation of micro-RNA genes miR15 and miR16 at 13q14 in chronic lymphocytic leukemia. Proc Natl AcadSci USA. 2002;99:15524–15529.

77. Cimmino A, Calin GA, Fabbri M, et al. miR-15 and miR-16 induce apoptosis by targetingBCL2. Proc Natl Acad Sci USA. 2005;102:13944–13949.

78. Kitada S, Andersen J, Akar S, et al. Expression of apoptosis-regulating proteins in chroniclymphocytic leukemia: correlations with in vitro and in vivo chemoresponses. Blood.1998;91:3379–3389.

79. Bonci D, Coppola V, Musumeci M, et al. The miR-15a-miR-16-1 cluster controls prostatecancer by targeting multiple oncogenic activities. Nat Med. 2008;14:1271–1277.

80. Fabbri M, Ivan M, Cimmino A, et al. Regulatory mechanisms of microRNAs involvementin cancer. Expert Opin Biol Ther. 2007;7:1009–1019.

81. Calin GA, Ferracin M, Cimmino A, et al. A MicroRNA signature associated with prognosisand progression in chronic lymphocytic leukemia. N Engl J Med. 2005;353:1793–1801.

82. Scaglione BJ, Salerno E, Balan M, et al. Murine models of chronic lymphocyticleukaemia: role of microRNA-16 in the New Zealand Black mouse model. Br J Haematol.2007;139:645–657.

83. Raveche ES, Salerno E, Scaglione BJ, et al. Abnormal microRNA-16 locus with synteny tohuman 13q14 linked to CLL in NZB mice. Blood. 2007;109:5079–5086.

84. Pekarsky Y, Santanam U, Cimmino A, et al. Tcl1 expression in CLL is regulated by miR-29and miR-181. Cancer Res. 2006;66:11590–11593.

85. Stamatopoulos B, Meuleman N, Haibe-Kains B, et al. MicroRNA-29c and microRNA-223downregulation has in vivo significance in chronic lymphocytic leukemia and improvesdisease risk stratification. Blood. 2009;113:5237–5245.

86. Marton S, Garcia MR, Robello C, et al. Small RNAs analysis in CLL reveals a deregu-lation of miRNA expression and novel miRNA candidates of putative relevance in CLLpathogenesis. Leukemia. 2008;22:330–338.

87. Venturini L, Battmer K, Castoldi M, et al. Expression of the miR-17-92 polycistron inchronic myeloid leukemia (CML) CD34+ cells. Blood. 2007;109:4399–4405.

88. Bueno MJ, Perez de Castro I, Gomez de Cedron M, et al. Genetic and epigenetic silenc-ing of microRNA-203 enhances ABL1 and BCR-ABL1 oncogene expression. Cancer Cell.2008;13:496–506.

89. Agirre X, Jimenez-Velasco A, San Jose-Eneriz E, et al. Down-regulation of hsa-miR-10a inchronic myeloid leukemia CD34+ cells increases USF2-mediated cell growth. Mol CancerRes. 2008;6:1830–1840.

90. Zanette DL, Rivadavia F, Molfetta GA, et al. miRNA expression profiles in chronic lympho-cytic and acute lymphocytic leukemia. Brazilian J Med Biol Res. 2007;40:1435–1440.

91. Nagel S, Venturini L, Przybylski GK, et al. Activation of miR-17-92 by NK-like home-odomain proteins suppresses apoptosis via reduction of E2F1 in T-cell acute lymphoblasticleukemia. Leuk Lymphoma. 2009;50:101–108.

92. Mi S, Lu J, Sun M, et al. MicroRNA expression signatures accurately discriminateacute lymphoblastic leukemia from acute myeloid leukemia. Proc Natl Acad Sci USA.2007;104:19971–19976.

93. Roman-Gomez J, Agirre X, Jimenez-Velasco, et al. Epigenetic regulation of microRNAs inacute lymphoblastic leukemia. J Clin Oncol. 2009;27:1316–1322.

94. Agirre X, Vilas-Zornoza A, Jimenez-Velasco A, et al. Epigenetic silencing of the tumor sup-pressor microRNA Hsa-miR-124a regulates CDK6 expression and confers a poor prognosisin acute lymphoblastic leukemia. Cancer Res. 2009;69:4443–4453.

95. Li X, Liu J, Zhou R, et al. Gene silencing of MIR22 in acute lymphoblastic leukaemiainvolves histone modifications independent of promoter DNA methylation. Br J Haematol.2010;148:69–79.


96. Ju X, Li D, Shi Q, et al. Differential microRNA expression in childhood B-cell precursoracute lymphoblastic leukemia. Pediatr Hematol Oncol. 2009;26:1–10.

97. Garzon R, Volinia S, Liu CG, et al. MicroRNA signatures associated with cytogenetics andprognosis in acute myeloid leukemia. Blood. 2008;111:3183–3189.

98. Burnett AK. Current controversies: which patients with acute myeloid leukaemiashould receive a bone marrow transplantation?–an adult treater’s view. Br J Haematol.2002;118:357–364.

99. Garzon R, Garofalo M, Martelli MP, et al. Distinctive microRNA signature of acutemyeloid leukemia bearing cytoplasmic mutated nucleophosmin. Proc Natl Acad Sci USA.2008;105:3945–3950.

100. Marcucci G, Radmacher MD, Maharry K, et al. MicroRNA expression in cytogeneticallynormal acute myeloid leukemia. N Engl J Med. 2008;358:1919–1928.

101. Garzon R, Liu S, Fabbri M, et al. MicroRNA-29b induces global DNA hypomethylationand tumor suppressor gene re-expression in acute myeloid leukemia by targeting directlyDNMT3A and 3B and indirectly DNMT1. Blood. 2009;113:6411–6418.

102. Fabbri M, Garzon R, Cimmino A, et al. MicroRNA-29 family reverts aberrant methylationin lung cancer by targeting DNA methyltransferases 3A and 3B. Proc Natl Acad Sci USA.2007;104:15805–15810.

103. Garzon R, Heaphy CE, Havelange V, et al. MicroRNA 29b functions in acute myeloidleukemia. Blood. 2009;114:5331–5341.

Index

A

Note: The letters ‘f’ and ‘t’ following locators refer to figures and tables respectively.

Abelson 1 (ABL1), 20, 50, 55–57, 57f, 65–66,82–86, 83t, 88–90, 88t, 104t,105–110, 106t, 115, 136, 146,216–218, 220–233, 235, 241, 293,295, 303, 304t, 307–313, 333

Absorbance, 9ACD, see Acid–citrate–dextrose (ACD)Acid–citrate–dextrose (ACD), 4Acute leukemia, 43, 58–59, 61, 86–90, 99,

106, 110–111, 116f, 128, 133–134,138, 146, 217, 219, 243, 312, 334

cytogenetic findings, 88twith no specific cytogenetic findings,

89–90subcategories, 89

Acute lymphoblastic leukemia (ALL), 20, 30,44, 51, 66, 86, 221, 329, 333

Acute megakaryoblastic leukemia (AMKL),63, 89, 110–112, 116f

Acute myeloid leukemia (AML), 13, 59–63,127–147

AML with myelodysplasia-relatedchanges, 62

cytogenetic abnormality in, WHOclassification scheme

AML (megakaryoblastic) witht(1;22)(p13;q13) – RBM15-MKL1, 62

AML (promyelocytic) witht(15;17)(q22;q12) – PML/RARα,60–61, 61f

AML with inv(16)(p13q22) ort(16;16)(p13;q22) – CBFβ/MYH11,61f, 62

AML with inv(3)(q21q26.2) ort(3;3)(q21;q26.2) – RPN1/EVI1, 62

AML with t(6;9)(p23;q34) –DEK/NUP214, 62

AML with t(9;11)(p22;q23) –MLLT3/MLL, 61, 61f

AML with t(8;21)(q22;q22) –RUNX1/RUNX1T1, 60

“myeloid cytogenetic markers,” 63rearrangements wrt chemotherapy, 62therapy-related AML (t-AML), 62See also AML, molecular pathology of

ADCC, see Antibody-dependent cellularcytotoxicity (ADCC)

Agarose gels, 11–12AITL, see Angioimmunoblastic T-cell

lymphoma (AITL)ALCL, see Anaplastic large-cell lymphoma

(ALCL)ALL, see Acute lymphoblastic leukemia

(ALL)Allele-specific PCR, 20–21, 129, 134, 145, 246All-trans-retinoic acid (ATRA or tretinoin),

61–62, 310–311AMKL, see Acute megakaryoblastic leukemia

(AMKL)AML, see Acute myeloid leukemia (AML)AML, molecular pathology of

AML-associated mutations, 129tAML with recurrent genetic abnormalities

AML with balanced translocations/inversions, 130–133

AML with gene mutations, 139–142core binding factor AML, 133–137gene expression and prognosis in

NK-AML, 145–146gene mutations with prognostic

significance, 143–145other recurrent translocations, 137–139rare subtypes of AML, 139

mutations and translocations associatedwith other myeloid neoplasms, 146

D. Crisan (ed.), Hematopathology, Molecular and Translational Medicine,DOI 10.1007/978-1-60761-262-9, C© Springer Science+Business Media, LLC 2010

341

342 Index

AML with mutated CCAAT/enhancer-bindingprotein alpha (CEBPA), 141

AML with mutated nucleophosmin 1 (NPM1),139–141

AML with myelodysplasia-related changes,142, 142t, 143f

AML with recurrent genetic abnormalitiesAML with balanced translocations/

inversionsPML–RARA translocation, detection

of, 130–133AML with gene mutations, 139–142

AML with mutated CCAAT/enhancer-binding protein alpha (CEBPA), 141

AML with mutated nucleophosmin 1(NPM1), 139–141

AML with myelodysplasia-relatedchanges, 142, 142t

CBF AMLAML with inv(16)(p13.1q22) or

t(16;16)(p13.1;q22); CBFB-MYH11, 135

AML with t(8;21)(q22;q22)RUNX1-RUNX1T1, 133–134, 134f

good prognosis AML, 133MRD by RQ-PCR in CBF-AML,

136–137gene expression and prognosis in


significance, 143–145FLT3-LM and FLT3-TKD, 143–145WT1, 145

other recurrent translocations, 137–139MLLT3-MLL and other MLL

abnormalities, 137–138, 138fMRD in AML with MLL abnormalities,

138–139rare subtypes of AML, 139

AMNI07, see NilotinibAnaplastic large-cell lymphoma (ALCL), 73,

95–96, 186–189ALK translocations in, 187tdetection of ALK dysregulation, 189–190genetic abnormalities in, 190–191

Aneuploid cell, 48Angiogenesis-inhibiting drugs, 314Angioimmunoblastic T-cell lymphoma (AITL),

96, 191, 196–197, 199genetic abnormalities in, 191–192

An International System of Human CytogeneticNomenclature (2009), 48

Antibody-dependent cellular cytotoxicity(ADCC), 296–297

Anti-cancer action of mABs, approaches,296–297

delivery of cytotoxic materials to tumorcells, 297

direct inhibitory effect on the tumorantigen–antibody binding mechanism,

296EGFR as target, effective signal

transduction inhibitors, 297induction of immune-mediated

mechanisms, 297Anti-cancer therapies, 294, 296Antigen receptor, 26–31, 27t, 81, 87, 91, 95,

103, 171, 178, 272Apolizumab, 301t, 302–303Apoptosis, 278–279Apoptosis-inducing drugs, 295, 315A Proposed Standard System of Nomenclature

of Human Mitotic Chromosomes, 48Arcturus/MDS Analytical Technologies, 3–4ARID5B gene, 113Array comparative genomic hybridization

(array-CGH), 41, 47, 158, 267–268Asymmetric PCR, 25Ataxia telangiectasia, 113, 259Automation, 12, 15, 17

BB-ALL, see B-lymphoblastic leukemia

(B-ALL)“Banding era,” 42Banding patterns of “banding era”

C banding, 42Giemsa banding, 42quinacrine banding, 42

B-cellbiology and maturation, 158–163

class switch of IGH, 162–163, 163fFRs/CDRs in IGH gene, 161IGH rearrangement, 160fimmunoglobulin gene rearrangement/

SHM, 158–163, 159tKappa and lambda light-chain gene

rearrangement, 161, 161flymphomagenic or leukemogenic

genetic alterations, 163pro-B cells turning pre-B cells, 160somatic hypermutation, role in antigen

selection, 165TdT expression in pro-B cells, 160

Index 343

clonality testing, 163–165indications, 163PCR over Southern blot,

advantages, 164PCR testing, false positives/negatives,

164–165PCR with CE gene scanning using

primers to FR3, 164–165, 164fsomatic hypermutation testing, 166

VH mutational status determination forCLL/SLL/MZL, 166

B-cell integration cluster (BIC), 329B-cell non-Hodgkin lymphoma (B-NHL),

158–182B-cell receptor

pharmacological inhibition of BCRsignaling, 277

somatic mutations of Ig variable regiongenes, 271–274

stereotypy, 274–277BCL6 gene, 71–72, 74, 172, 261BCR–ABL1-negative “classic”

myeloproliferative neoplasmseffect of JAK2 allelic burden, 235–236epidemiology, clinical, and laboratory

features, 232–233JAK2 mutation detection methods,

236–237JAK2 mutations in PV, 234–235JAK2 – unresolved issues, 237–238JAK2 V617F’s contribution to diagnosis of

MPNs, 233–234MPL, 238

BCR–ABL1-negative disorders, 84–85BCR–ABL1-negative myeloproliferative

neoplasmschronic neutrophilic leukemia, 84essential thrombocythemia, 84polycythemia vera, 84primary myelofibrosis, 84

Bevacizumab, 295, 302, 318BIC, see B-cell integration cluster (BIC)BigDye R©, 15BIOMED-2 primers, 165, 185, 192BiovaxID R©, 316B-LBL, see Precursor B-lymphoblastic

lymphoma (B-LBL)Bloom syndrome, 113B-lymphoblastic leukemia (B-ALL), 63–66,

88t, 103, 107high hyperdiploidy ALL, 64hypodiploid ALL, 64

pediatric B-cell ALL, trisomy ofchromosomes in, 65f

pediatric B-cell ALL with hyperdiploidkaryotype, 64f

B-NHL, see B-cell non-Hodgkin lymphoma(B-NHL)

Bone marrow basophilia, 219Bone marrow eosinophilia, 219Bone marrow/leukemic blood, cytogenetic

analysis ofculture harvesting/slide preparation/

staining, 46historical perspectives

banding patterns of “banding era,” 42Burkitt lymphoma, 42CISH, 42FISH, diagnostic utility in cytogenetics,

42–43“microarray era,” impact, 43Philadelphia chromosome, CML, 42

karyotype and cytogenetic nomenclaturechromosomal abnormalities, types, 48G-banded bone marrow karyotype, 47fhematolymphoid chromosomal

abnormalities (ISCNnomenclature), 49t

refinement of chromosome morphology,techniques, 48

structural chromosomal rearrangementsin neoplasia, 49t

microscopic analysis of, guidelines for,46–47

specimen collection and storage, 43–44specimen processing and tissue culture

culture conditions used forhematolymphoid disorders, 45t

optimal cell density determination,methods, 44

short-term cultures, advantages, 44–45Bone marrow/leukemic blood, FISH

analysis ofadvantages/disadvantages of FISH, 54–55clinical indications, 51FISH, basic principles of, 50–51FISH probes used, types of, 52f

CEPs, 52commercially available probes for

hematolymphoid disorders, 53t–54tLSI, 52WCP, 52

“Buffy coat,” 3

344 Index

Burkitt lymphoma (BL), 42, 70, 72–73, 72f,90t, 94, 97, 178–180, 179t, 262

endemic, sporadic, and immunodeficientBLs, 179t

and MYC, 179–180MYC/IGH rearrangement in

endemic/sporadic BL, 179fMYC translocation testing, diagnosis of

BL, 180other genetic abnormalities in, 180–181

genetic differences betweenBL/DLBCL, 181t

“starry sky” pattern of macrophages, 178

CCAGRs, see Cancer-associated genomic

regions (CAGRs)Cancer-associated genomic regions (CAGRs),

326Cancer stem cell (CSC), 113–116Cancer vaccines, 295, 315

BiovaxID R©, 316GARDASIL R©, 316

Capillary electrophoresis (CE), 10, 12–13, 28,140f, 144f, 164f, 223, 236

advantages over conventional gelelectrophoresis, 12

applications, 13detection window, 12separation of analytes, 12shadow peak, 12

Carl Zeiss, 4C banding, 42, 46CBFB-MYH11, 133, 135

exon–intron structure of, 136fCCD, see Charge-coupled device (CCD)CDC, see Complement-dependent cytotoxicity

(CDC)CDCC, see Complement-dependent

cell-mediated cytotoxicity (CDCC)CDRs, see Complementarity-determining

regions (CDRs)Cell enrichment, 3–4Centromere enumeration probes (CEPs), 52CEPs, see Centromere enumeration probes

(CEPs)CGH, see Comparative genomic hybridization

(CGH)Chaotropic salt, 7–8Charge-coupled device (CCD), 12Chemokine receptor 4 (CXCR4), 328Chemotherapy, 60, 62, 63f, 87, 114, 119–120,

130, 142, 165, 293, 295, 302, 311,312t, 317–318

Chimeric mABs, 298CHL, see Classical Hodgkin lymphoma (CHL)Chromatin remodeling, 108, 334Chromogenic in situ hybridization (CISH), 42Chromosomal abnormalities

numerical abnormalities, 48structural abnormalities, 48

Chromosomal rearrangements in neoplasia,molecular mechanisms, 48–50, 49t

Chromosomal translocations resulting intoBCL2 rearrangement, 263–264

Chromosome band, 48Chromosome microarray analysis/microarray

CGH, 43Chronic eosinophilic leukemia/idiopathic

hypereosinophilic syndrome, 58Chronic lymphocytic leukemia (CLL), 29, 43,

45t, 51, 67, 90, 166, 255–280, 299t,302, 332

See also CLL, molecular pathology of;Cytogenetic abnormalities in CLL

Chronic myelogenous leukemia (CML),13, 333

diagnostic testingcytogenetic karyotyping, 222FISH, 223PCR strategies, 223RT-PCR technique, 222

disease monitoring/response to therapy,223–226

criteria for lack of response, 225tcytogenetic response, 223–226hematologic response, 223–226imatinib mesylate therapy, 224–225molecular response, 223–226response types and equivalent estimated

tumor burden, 224tsamples/recommended frequencies for

various types of response, 225tepidemiology, clinical, and laboratory

featuresabsolute basophilia/eosinophilia/

leukocytosis, 219splenomegaly and purpura, 219symptoms at diagnosis, 219WHO criteria for diagnosis, 219

historical perspective and current relevance,217–218

BCR–ABL1 fusion, study, 218imatinib, treatment of CML, 218molecular diagnostics, impact, 218Philadelphia chromosome, discovery

of, 217–218

Index 345

timeline of landmark developmentsin CML to use of targetedtherapy, 218f

management of, integration of moleculardiagnostic testing in, 231–232

quantitative RT-PCR in disease monitoring,226–228

resistance to tyrosine kinase inhibitortherapy, 229–231

structure and pathogenesis of BCR–ABL1,220–222

ABL1 tyrosine kinase activity, study inmouse models, 221–222

abnormalities in blast-phase/accelerated-phase, 222

breakpoints in ABL1, results, 220breakpoints in BCR, 220–221chimeric nature of BCR-ABL1

fusion, 221Philadelphia chromosome from

reciprocal translocation, 220, 220ft(9;22) mechanism, 220

Chronic myeloproliferative disorders, 82–85BCR–ABL1-negative myeloproliferative

neoplasmschronic neutrophilic leukemia, 84essential thrombocythemia, 84polycythemia vera, 84primary myelofibrosis, 84

molecular and cytogenetic findings, 83tmolecular tests for BCR–ABL1

diagnosis of CML, 83other BCR–ABL1-negative disorders,

84–85Philadelphia chromosome, reciprocal

translocation ofdiagnosis of CML, 83

Chronic neutrophilic leukemia (CNL), 58CISH, see Chromogenic in situ hybridization

(CISH)Classical Hodgkin lymphoma (CHL), 200–201CLL, see Chronic lymphocytic leukemia

(CLL)CLL, molecular pathology of

cytogenetic abnormalitiesapoptosis, 278–279B-cell receptor, 270–277chromosomal translocations resulting

into BCL2 rearrangement, 263–264deletion 17p13, 260–261deletion 6q, 261deletion 13q14, 257–259deletion 11q22-q23, 259–260

epigenetic changes, 266FISH/cytogenetics, identification

methods, 256, 258fhigh-throughput molecular methods to

assess CLL, 266–2708q24 gain, 262role of the microenvironment, 279–280stimulation with CD40 ligand

expressing cells and IL-4, 256stimulation with CpG-

oligodeoxynucleotides andIL-2, 256

t(2;14)(p16;q32), 265t(14;19)(q32;q13), 262–263translocations involving chromosome

14q32, 266Trisomy 12, 260Trisomy 3q27, 261

incidence in western/Asian countries, 256lymphocytosis in CLL/lymphadenopathy

in SLL patients, 256Clonal abnormality, (ISCN 2009), 47Clonality assessment, 25–31

IGH receptor gene, 26, 27fIg/TCRs, characteristics and recombination

process, 26–28, 27tlimitations, 30–31malignant or reactive/benign cells,

characterization, 25–26techniques

PCR-based assessment of TCR, 30PCR-based methods, analysis of the

IGH gene, 28–29, 29fsomatic hypermutation process, 29–30Southern blotting, 28

CLP, see Common lymphoid progenitor (CLP)CML, see Chronic myelogenous leukemia

(CML)CMP, see Common myeloid progenitor (CMP)Common hematolymphoid chromosomal

abnormalities (ISCNnomenclature), 49t

Common lymphoid progenitor (CLP), 327Common myeloid progenitor (CMP), 327Comparative genomic hybridization (CGH),

41, 43, 47, 82, 109, 120, 158, 191,194, 260, 267, 267f

See also Array comparative genomichybridization (array-CGH)

Complementarity-determining regions (CDRs),27f, 29, 160f, 161, 296, 298

Complement-dependent cell-mediatedcytotoxicity (CDCC), 296–297

346 Index

Complement-dependent cytotoxicity(CDC), 296

Conventional karyotyping, 69, 81, 128–130,132, 135, 139, 257

Core binding factor AMLs (CBF-AML), 133Corticosteroids, 118CRLF2, see Cytokine receptor-like factor 2

(CRLF2)CXCR4, see Chemokine receptor 4 (CXCR4)Cyclophosphamide, doxorubicin, vincristine,

and prednisone (CHOP), 302Cytogenetic abnormalities in CLL

apoptosis, 278–279B-cell receptor, 270–277

pharmacological inhibition of BCRsignaling, 277

somatic mutations of Ig variable regiongenes, 271–274

stereotypy, 274–277chromosomal translocations resulting into

BCL2 rearrangement, 263–264deletion 17p13, 260–261deletion 6q, 261deletion 13q14

miR-15a/miR-16-1 deletion, role inpathogenesis, 258–259

13q14, site of a tumor suppressor gene,258–259

Rb1 deletion, role in pathogenesis, 258deletion 11q22-q23

ATM gene mutations, poor prognosis,259–260

ATM gene, role in CLL pathogenesis,259

ATM-mutant CLL cases, 259CUL5 (ubiquitin-dependent apoptosis

regulation), 260NPAT (cell cycle regulation), 260PPP2R1B (component of the cell cycle

and apoptosis regulating PP2A),260

epigenetic changes, 266high-throughput molecular methods to

assess CLL, 266–2708q24 gain

MYC abnormalities, 262MYC overexpression in CLL, 262, 263f

role of the microenvironment, 279–280t(2;14)(p16;q32), 265, 265ft(14;19)(q32;q13), 262–263, 264ftranslocations involving chromosome

14q32, 266Trisomy 12, 260

Trisomy 3q27BCL6 gene, repression of transcrip-

tion, 261BCL6 gene, translocations in, 261

Cytogenetic tests, 81conventional karyotyping, 81FISH, 81

Cytokine receptor-like factor 2 (CRLF2),111–112

DDasatinib, 16, 230, 245, 277, 308–310, 314Degradation of nucleic acids, 5–6Deletion 17p13, 260–261Deletion 6q, 261Deletion 13q14, 257–259Deletion 11q22-q23, 259–260Delta-like ligand, 117Denaturing high-performance liquid

chromatography (DHPLC),266–267

“de novo” DNA methyltransferases, 335Deoxynucleotide triphosphates (dNTPs),

14, 18DEPC, see Diethylpyrocarbonate (DEPC)Detection window, 12DHPLC, see Denaturing high-performance

liquid chromatography (DHPLC)Diethylpyrocarbonate (DEPC), 8Diffuse large B-cell lymphoma (DLBCL),

70–72, 90t, 93–94, 168, 172–174,181t, 330

BCL6 alterations in, 172–174groups based on pathogenic

mechanisms, 174other genetic alterations in, 174–175

other lymphomas of large B cells, 173tspecific diffuse large B-cell

subtypes, 172tsubtyping of

germinal center B-cell (GC) type,171–172

post-germinal center or activated B-cell(ABC) type, 171

DiGeorge syndrome critical region gene 8(DGCR8), 325

Diploid cell, 48DLBCL, see Diffuse large B-cell lymphoma

(DLBCL)DNA sequencing

NGS, 17–18pyrosequencing, 16–17

drawbacks, 16–17

Index 347

Sanger sequencing, 14–15limitations, 15

DNTPs, see Deoxynucleotide triphosphates(dNTPs)

Down syndrome, 110–113Dried blood spots (guthrie cards), 5Drosha, 325Dual-color break-apart (DCBA) probe, 43Dual-color, dual-fusion (DCDF) LSI probes,

52–53Dye-primer chemistry, 15

EEarly T-cell precursor (ETPs), 120EATL, see Enteropathy-associated T-cell

lymphoma (EATL)EDTA, see Ethylenediaminetetraacetic acid

(EDTA)EGFR, see Epithelial growth factor receptor

(EGFR)Electrophoresis, 10–13

agarose gels, use of, 11capillary electrophoresis, advantages, 11

Enteropathy-associated T-cell lymphoma(EATL), 97, 197

Epithelial growth factor receptor (EGFR),294, 297

Essential thrombocythemia, 58, 82, 83t, 84,110, 216, 221, 232, 307, 313

Ethidium bromide, 10–11, 11f, 22Ethylenediaminetetraacetic acid (EDTA), 4ETPs, see Early T-cell precursor (ETPs)Exportin 5, 325Extranodal marginal zone B-cell lymphoma

(MALT-type), 73Extranodal natural killer-/T-cell lymphoma,

197–198

FFACS, see Fluorescent antibody cell sorting

(FACS)FBXW7, 119Fiber FISH, 43Ficoll-Hypaque, 3FISH, see Fluorescence in situ hybridization

(FISH)FISH, basic principles of, 50–51

analysis on metaphase chromosomesinterphase FISH, 51

analysis on paraffin-embedded tissuesection

advantage/disadvantage, 51Flow cytometry, 1, 25, 80–81, 85–88, 91, 95,

104, 186, 272

FLT3-LM and FLT3-TKD, 143–145PCR detection of FLT3-LM, 144fstructure of FLT3 gene, 144f

Fluorescence dye-terminator chemistry, 15Fluorescence in situ hybridization (FISH), 42

See also Chromogenic in situ hybridization(CISH)

Fluorescence resonance energy transfer(FRET) technology, 15

Fluorescent antibody cell sorting (FACS), 3Fluorometric methods, 9–10Follicular lymphoma (FL), 71–72, 92, 166–169

and BCL2 t(14;18), 166–168BCL2, antiapoptotic effect, 166BCL2/IGH rearrangement at MBR,

167, 167fcytogenetics/FISH/PCR, detection

methods, 168grade 3B follicular lymphomas, 168+18q mechanism, 168

genetic abnormalities in grade 3B FL,168–169, 169t

additional abnormality by routinecytogenetics, 168

gene expression profiling, 169transformation to higher grade

lymphoma, 168Formalin, 5–6, 28, 30FRA, see Fragile sites (FRA)Fragile sites (FRA), 326Framework regions (FRs), 161French–American–British (FAB) scheme, 41FRET, see Fluorescence resonance energy

transfer (FRET) technology

Gγ-secretase inhibitors (GSIs), 117GARDASIL R©, 316Gastrointestinal stromal tumors (GISTs), 245,

308–309GATA1, 110–112G-banded bone marrow karyotype, 47fGCB-DLBCL, see Germinal center B-cell-like

phenotype (GCB-DLBCL)Gel electrophoresis, 10, 11f, 12, 14, 19, 25, 192Gene expression profiles (GEP), 82, 93–94,

169, 190–191, 198–199, 201Gene therapy, 316Gene transfer process, 316GEP, see Gene expression profiles (GEP)Germinal center B-cell-like phenotype

(GCB-DLBCL), 331

348 Index

Giemsa banding, 42GISTs, see Gastrointestinal stromal tumors

(GISTs)GITC, see Guanidine isothiocyanate (GITC)Guanidine isothiocyanate (GITC), 8

HHairpin RNA precursor (pre-miRNA), 325Hairy and enhancer-of-split analog-1 (HES1),

117, 119Helicobacter pylori, 73, 93, 175Hematologic malignancies, targeted therapy

angiogenesis-inhibiting drugs, 314apoptosis-inducing drugs, 315cancer vaccines, 315–316challenges/changes in clinical practice,

317–318gene therapy, 316molecular genetic signatures, clinical use,

293–294small molecule drugs, 303–310

ATRA, 310–311examples for AML, 311–312examples for treatment of CML, 314examples for treatment of MPN,

312–313targeted therapy, 294–296

advantages over traditional cancertherapies, 295

conventional cytotoxicchemotherapeutic agents, 295

definitions, 294–295“magic bullet” therapy, 294molecular targeted therapy, 294National Cancer Institute

classification, 295test and the drug criterion by FDA,

294–295therapeutic monoclonal antibodies in

targeted therapy, 296–301examples, 302–303

Hematolymphoid disordersarray-based genomic profiling of, 74–75

disadvantage, 74FISH analysis, 74SNP analysis by molecular

allelokaryotyping, 74SNP analysis of AML/MDS

samples, 75uniparental disomy of chromosomes,

74–75bone marrow/leukemic blood, cytogenetic

analysis of

culture harvesting, slide preparation,and staining, 46

historical perspectives, 42–43karyotype and cytogenetic

nomenclature, 47–48microscopic analysis, guidelines for,

46–47specimen collection and storage, 43–44specimen processing and tissue culture,

44–45bone marrow/leukemic blood, FISH

analysis ofadvantages/disadvantages of FISH,

54–55clinical indications, 51FISH, basic principles of, 50–51FISH probes used, types of, 52–54

chromosomal rearrangements in neoplasia,molecular mechanisms, 48–50

diagnosis of, cytogenetic analysis inarray CGH, genome-wide study, 41chromosome analysis, 40–41FISH, 41myeloid/lymphoid neoplasms

classification, 41PCR, 41

lymphoid disorders, diagnostic/prognosticcytogenetic markers

B-lymphoblastic leukemia/lymphomawith recurrent genetic abnormali-ties, 63–66

chronic lymphocytic leukemia/smalllymphocytic leukemia, 67

Hodgkin lymphoma, 74non-Hodgkin’s lymphoma, 70–73plasma cell myeloma, 67–70T-lymphoblastic leukemia/lymphoma,

66–67myeloid disorders, diagnostic/prognostic

cytogenetic markersAML, 59–63MDS, 59MPN, 55–58

Hematopoietic stem cells (HSCs), 103, 222,245, 298, 309

Hepatosplenic T-cell lymphoma (HSTL),196–197

HES1, see Hairy and enhancer-of-splitanalog-1 (HES1)

High-throughput molecular methods to assessCLL, 266–270

array-CGH, 267–268BAC array-based CGH, 268

Index 349

CGH-a technique applied to CLL cases,268, 268f

oligonucleotide-based array-CGH, 268principles of, 267f

gene expression profiling, 269gene expression levels in CLL, 270f

microRNA, 269MLPA, 269sequence and mutation analysis, 266–267

array-based mutation analysis, 267DGGE, 267DHPLC, 266SSCP, 267

SNP-arrays, 268–269High-throughput sequencing, 17

See also Next-Generation Sequencing(NGS)

Hodgkin lymphoma (HL), 74, 90, 97–98, 173,173t, 199, 239, 331

CHL, 200–201NLPHL, 200

HSCs, see Hematopoietic stem cells (HSCs)HSTL, see Hepatosplenic T-cell lymphoma

(HSTL)Humanized type of mABs, 298Hybridization probes, 22–23, 246Hydrolysis probes, 23Hyperdiploid cell, 48Hypodiploid cell, 48

IIGH, see Immunoglobulin heavy-chain gene

(IGH)IKAROS, 108–109, 108f

chromatin remodeling, 108DNA-binding and nuclear localization, 108

IKZF1, 107–109, 108f, 113IL-7 alpha receptor, 112Imatinib mesylate (Glivec R©/Gleevec R©), 308

dasatinib, 310nilotinib or AMNI07, 309–310treatment of CML, 308–309

imatinib-based therapy, problems, 309treatment of GISTs, 309

Immunoglobulin heavy-chain gene (IGH), 160,160f, 163f

Immunoglobulin (Ig), 26, 27t, 259Insulin-like growth factor receptor (IGFR), 327Interferon-α (alpha) therapy, 223International Prognostic Scoring System

(IPSS), 59Interphase FISH, 51IPSS, see International Prognostic Scoring

System (IPSS)

JJAK2, 14, 21–24, 44, 55, 83t, 84–86, 111–112,

146, 194, 216, 216t, 232–238, 303,307, 313, 332

JAK2 R683 mutation, 111–112JAK2 V617F mutation, 14, 21–22, 44, 84,

233–234, 236Janus kinase, 110, 307–308“Just another kinase,” see Janus kinase

KKappa-deleting element (Kde), 161f, 162Kinase inhibitors, see Small molecule drugsKinases, categories, 307Kit receptor, 245f, 328

LLATE-PCR, see Linear-after-the-exponential

PCR (LATE-PCR)Leukemia predisposition, 113Leukemia stem cell, 114Linear-after-the-exponential PCR

(LATE-PCR), 25Locus-specific identifier (LSI) probes, 52, 223LOH, see Loss of heterogeneity (LOH)Loss of heterogeneity (LOH), 327LPL, see Lymphoplasmacytic lymphoma

(LPL)LSI probes, see Locus-specific identifier (LSI)

probesLumiliximab, 300T, 302Lymphadenopathy, 45t, 91, 193, 195, 199, 244,

256, 261Lymphoblastic lymphoma, 66, 103, 239,

241, 243Lymphocytosis, 45t, 256, 302Lymphoid disorders, diagnostic/prognostic

cytogenetic markersB-lymphoblastic leukemia/lymphoma,

63–66chronic lymphocytic leukemia/small

lymphocytic leukemia, 67Hodgkin lymphoma, 74non-Hodgkin’s lymphoma, 70–73plasma cell myeloma, 67–70T-lymphoblastic leukemia/lymphoma,

66–67Lymphoid neoplasms classification

FAB scheme (1976), 41REAL (1994), 41WHO classification (1997), 41

Lymphoplasmacytic lymphoma (LPL), 165,181–182

350 Index

MmABs, see Monoclonal antibodies (mABs)“Magic bullet” therapy (Paul Ehrlich), 294Major break point cluster region (M-bcr), 107Major breakpoint region (MBR), 167,

167f, 173Malignant lymphomas

molecular/cytogenic findings, 90–91ALCL, 95–96Burkitt lymphoma, 94diagnostic problems, 91–92DLBCL, 93–94FL, 92MCL, 92MZL, 93NK-cell lymphomas, 96–97PCN, 94–95T-cell lymphomas, 95

MALT lymphomas, see Mucosa-associatedlymphoid tissue (MALTlymphomas)

Mammalian target of rapamycin (mTOR), 120Mantle cell lymphoma (MCL), 45t, 51, 54t,

70–71, 73, 90t, 92, 169–170,305t, 315

clinical implications, 171cyclin D1 dysregulation, detection of

testing for t(11;14)(q13;q32),methods, 170

genetic abnormalities in, 170MAP kinases, see Mitogen-activated protein

(MAP) kinasesMarginal-zone lymphomas (MZLs), 93, 159t,

175–178clinical implications, 177detection of MALT lymphoma

translocations, 177genetic abnormalities in, 176–177

frequencies (%) of MALT lymphomatranslocations and trisomies, 176t

MALT, site of occurence, 176other MZLs

splenic MZLs, 178Mast cell disease, 216, 244–246, 245f

mastocytosis, 244diagnosis of, 244increased risk of myeloid neoplasms,

244M-bcr, see Major break point cluster region

(M-bcr)m-bcr, see Minor break point cluster region

(m-bcr)MBR, see Major breakpoint region (MBR)

MCL, see Mantle cell lymphoma (MCL)MDS, see Myelodysplastic syndromes (MDS)Megakaryopoiesis, 328Methylation-specific PCR, 24–25Methyltransferase enzymes, 13MF, see Mycosis fungoides (MF)“Microarray era,” 43Micro-RNAs (miRNAs), 54t, 109, 147, 158,

258, 269, 279, 325–335Minimal residual disease (MRD), 3, 20–21, 23,

28, 51, 57, 94, 120, 128–129, 131,133, 135, 163, 180, 184, 189, 218,223, 227

Minor break point cluster region (m-bcr), 107miR-230, 109–110miR-10a, 328, 330t, 333–334miR-130a, 328miRNAs, see Micro-RNAs (miRNAs)miRNAs in hematologic malignancies

biogenesis of miRNAs, 325–327, 326fmiRNAs and human cancer,

relationship, 326–327miRNAs in leukemias, 332–335miRNAs in lymphomas, 329–332miRNAs in normal hematopoiesis,

327–329miRNAs in leukemias, 330t, 332–335

high expression of TCL1, effects onCLL, 333

miR-15a and miR-16, role as TSGs inCLL, 332–333

miR-191 and miR-199a, prognosticimplications in AML, 334–335

miR-17-92, role in ALL vs. normal CD19+B cells, 333–334

miR-17-92, role in CML, 333miRNAs in lymphomas, 329–332, 330t

miR-143 and miR-145, role, 331miR-106a-363 polycistron overexpression,

role, 331miR-106b-25 and miR-17-92, control of

apoptosis, 331miR-17-92 cluster overexpression, role,

331miRNA expression in Hodgkin lymphoma,

331–332miR-155, role in lymphomagenesis

ABC-DLBCL vs. and GCB-DLBCL, 331

B-cell-specific miR-155 transgenicmouse model, study, 329

miR-155 knockout (KO) mice models,study, 329

Index 351

pediatric Burkitt’s lymphoma, 329miRNAs in normal hematopoiesis, 327–329

C57BL6 mouse model transplantedwith mice MPPs overexpressingmiR-155

miR-155 block of erythrocytic/megakaryocytic differentiation, 328

megakaryocyte differentiation, 328miR-181a expression, effects, 328miR-150, inhibition of B-cell

development, 329miRNA expression patterns in erythrocyte

precursors, 327–328miRNAs as “fine tuners,” 329miR-223, role in human granulopoiesis

MEF2c/IGFR, effects, 327miR-424, role in myeloid hematopoiesis,

327silencing of miR-451 in zebra fishembryo

model, results, 328Mitogen-activated protein (MAP) kinases, 107,

110, 269Mitogens, 45MLPA, see Multiplex ligation-dependent probe

amplification (MLPA)“Molecular allelokaryotyping,” 74Molecular and cytogenetic procedures, 81

CGH, 82generation of GEP by array analyses, 82sequencing of DNA, 82

Molecular Machines and Industries, 4Molecular pathology of AML

AML-associated mutations, 129tAML with recurrent genetic abnormalities

AML with balanced translocations/inversions, 130–133

AML with gene mutations, 139–142core binding factor AML, 133–137gene expression and prognosis in


significance, 143–145other recurrent translocations, 137–139rare subtypes of AML, 139

mutations and translocations associatedwith other myeloid neoplasms, 146

Molecular pathology of B-cell/T-celllymphomas, see Burkitt Lymphoma(BL); Molecular testing for B-NHL;Molecular testing for T-NHL

Molecular pathology of CLLcytogenetic abnormalities

apoptosis, 278–279

B-cell receptor, 270–277chromosomal translocations resulting

into BCL2 rearrangement, 263–264deletion 17p13, 260–261deletion 6q, 261deletion 13q14, 257–259deletion 11q22-q23, 259–260epigenetic changes, 266FISH/cytogenetics, identification

methods, 256, 258fhigh-throughput molecular methods to

assess CLL, 266–2708q24 gain, 262role of the microenvironment, 279–280stimulation with CD40 ligand

expressing cells and IL-4, 256stimulation with CpG-

oligodeoxynucleotides andIL-2, 256

t(2;14)(p16;q32), 265t(14;19)(q32;q13), 262–263translocations involving chromosome

14q32, 266Trisomy 12, 260Trisomy 3q27, 261

incidence in western/Asian countries, 256lymphocytosis in CLL/lymphadenopathy

in SLL patients, 256Molecular pathology of MPN

BCR–ABL1-negative “classic”myeloproliferative neoplasms

effect of JAK2 allelic burden,235–236

epidemiology, clinical, and laboratoryfeatures, 232–233

JAK2 mutation detection methods,236–237

JAK2 mutations in PV, 234–235JAK2 – unresolved issues, 237–238JAK2 V617F’s contribution to diagnosis

of MPNs, 233–234MPL, 238

CMLdiagnostic testing, 222–223disease monitoring/response to therapy,

223–226epidemiology, clinical, and laboratory

features, 219historical perspective and current

relevance, 217–218management of, integration of

molecular diagnostic testing in,231–232

352 Index

Molecular pathology of MPN (cont.)quantitative RT-PCR in disease

monitoring, 226–228resistance to tyrosine kinase inhibitor

therapy, 229–231structure and pathogenesis of

BCR–ABL1, 220–222mast cell disease, 244–246PDGFR and FGFR1 abnormalities

FGFR1, 243–244PDGFRA, 240–241PDGFRB, 241–243spectrum of eosinophilia-related

disorders, 239–240tyrosine kinases in, 216t

“Molecular targeted therapy,” 294Molecular techniques in hematopathology

step 3: assessment of nucleic acid qualityand quantity, 9–10

step 2: nucleic acid extraction, purification,and storage

extraction techniques, 7–9step 4: selected techniques

allele-specific PCR, 20–21asymmetric PCR, 25clonality assessment, 25–31DNA sequencing, 14–18electrophoresis, 10–13methylation-specific PCR, 24–25nested PCR, 21PCR, 18–20Q-PCR, 23–24real-time PCR, 21–23restriction enzymes, 13–14restriction site PCR, 25RT-PCR, 20

step1: specimen collection and processingcell enrichment and selection

techniques, 3–4patient identification and labeling, 2source-specific requirements for nucleic

acid integrity, 4–6standard precautions and safety, 2

Molecular testing for B-NHLB-cell

biology and maturation, 158–163clonality testing, 163–165somatic hypermutation testing, 166

Burkitt Lymphoma, 178–181and MYC, 178–180other genetic abnormalities in, 180–181

detection of MALT lymphomatranslocations, 177

DLBCL, 172–174BCL6 alterations in, 172–174genetic alterations in, 174–175

follicular lymphoma, 166–167and BCL2 t(14;18), 166–168other genetic abnormalities in, 168–169

lymphoplasmacytic lymphoma, 181–182MCL, 169

clinical implications, 171cyclin D1 dysregulation, detection

of, 170genetic abnormalities in, 170

MZLs, 175–176clinical implications, 177genetic abnormalities in, 176–177other MZLs, 177–178

Molecular testing for T-NHLAITL, 191

genetic abnormalities in, 191–192ALCL, 186–189

detection of ALK dysregulation,189–190

genetic abnormalities in, 190–191EATL, 197extranodal natural killer-/T-cell lymphoma,

197–198HSTL, 196–197MF, 192

genetic abnormalities in, 193–195molecular staging of, 193

PTCL-NOS, 198–199SS, 195

genetic abnormalities in, 195–196T-cell

biology and maturation, 182–184clonality testing, 184–186

Molecular tests, 81–82DNA/RNA analyses, 82“negative” finding cases, FISH analysis, 82residual disease detection, 82structural abnormalities, detection of, 81

Monoclonal antibodies (mABs), 3, 273, 276,295–298, 299t, 302–303, 308, 317

Monosomy, 48MPN, see Myeloproliferative neoplasms

(MPN)MPN, molecular pathology of

BCR–ABL1-negative “classic”myeloproliferative neoplasms

effect of JAK2 allelic burden, 235–236epidemiology, clinical, and laboratory

features, 232–233

Index 353

JAK2 mutation detection methods,236–237

JAK2 mutations in PV, 234–235JAK2 – unresolved issues, 237–238JAK2 V617F’s contribution to diagnosis

of MPNs, 233–234MPL, 238

CMLdiagnostic testing, 222–223disease monitoring/response to therapy,

223–226epidemiology, clinical, and laboratory

features, 219historical perspective and current

relevance, 217–218management of, integration of

molecular diagnostic testing in,231–232

quantitative RT-PCR in diseasemonitoring, 226–228

resistance to tyrosine kinase inhibitortherapy, 229–231

structure and pathogenesis ofBCR–ABL1, 220–222

mast cell disease, 244–246PDGFR and FGFR1 abnormalities

FGFR1, 243–244PDGFRA, 240–241PDGFRB, 241–243spectrum of eosinophilia-related

disorders, 239–240tyrosine kinases in, 216t

MPP, see Multipotent hematologic progenitor(MPP)

MRD, see Minimal residual disease (MRD)mTOR, see Mammalian target of rapamycin

(mTOR)Mucosa-associated lymphoid tissue (MALT

lymphomas), 73, 90t, 93, 175Multiplex ligation-dependent probe

amplification (MLPA), 269Multiplex PCR, 20Multipotent hematologic progenitor

(MPP), 327MYC and Burkitt Lymphoma, 178–180Mycosis fungoides (MF), 192–195

genetic abnormalities in, 193–195molecular staging of, 193

Myelodysplastic disordersMDS group, entities, 85“proliferative” clinical features, 85refractory cytopenia of childhood, 85–86

“cytopenic” features, 85

genetic abnormalities, 86morphological analysis, 85patients with RARS, 865q deletion, 86

Myelodysplastic syndromes (MDS)complex karyotype identified in

a 83-year-old female withpancytopenia, 60f

conventional cytogenetic analysischromosomal changes in MDS, 59

FISH analysis in, 59“Myeloid cytogenetic markers,” 63Myeloid/lymphoid neoplasms classification,

FAB scheme, 41Myeloproliferative neoplasms (MPN), 55–58,

82–85chronic eosinophilic leukemia/idiopathic

hypereosinophilic syndrome, 58chronic neutrophilic leukemia, 58CML, BCR/ABL1 positive, 55–57

karyotyping, FISH, and RT-PCR, 55–57Philadelphia chromosome, generation

of, 55, 56ftreatments, 56–57

cytogenetic abnormalities in, 57fessential thrombocythemia, 58polycythemia vera, 57–58primary myelofibrosis, 58See also Chronic myeloproliferative

disordersMZLs, see Marginal-zone lymphomas (MZLs)

NNested PCR, 21Neurofibromatosis, 113Next-Generation Sequencing (NGS), 14,

17–18advantages over automated Sanger-based

methods, 17drawbacks, 17high-throughput sequencing of single

DNA, 18novel applications, 18

NGS, see Next-Generation Sequencing (NGS)Nilotinib, 230, 304t, 308–310, 314NK-cell lymphomas, 91, 96–97, 198NLPHL, see Nodular lymphocyte-predominant

Hodgkin lymphoma (NLPHL)NOD/SCID mouse, 114–116Nodular lymphocyte-predominant Hodgkin

lymphoma (NLPHL), 74, 159t,173, 200

Noncoding RNAs (ncRNAs), see Micro-RNAs(miRNAs)

354 Index

Non-Hodgkin’s lymphoma (NHL)anaplastic large-cell lymphoma, 73Burkitt lymphoma, 72–73diffuse large B-cell lymphoma, 72extranodal marginal zone B-cell lymphoma

(MALT-type), 73follicular lymphoma, 71–72mantle cell lymphoma, 73splenic marginal zone lymphoma, 73

Non-receptor tyrosine kinases (non-RTKs), 303examples, 303–307

Non-RTKs, see Non-receptor tyrosine kinases(non-RTKs)

“Non-specific” abnormalities, seeChromosomal abnormalities

NOTCH1, 106, 106t, 117–120, 118fNPM1, 139–141

exon–intron structure of, 140fNucleic acid

degradation of, 5–6extraction techniques

inorganic (chaotropic salt–silicacolumn), 8

organic (phenol–chloroform), 7–8integrity, source-specific requirements for

bone marrow aspirates/wholeblood/body fluids, 4

dried blood spots (guthrie cards), 5fixed, paraffin-embedded tissue, 5–6fresh tissue, 5

quality/quantity, assessment of, 9–10fluorometric methods, 9–10gel electrophoresis, qualitative

assessment, 10, 11fspectrophotometric methods, 9

storagelong-term storage, considerations, 9stability of DNA in storage, 8stability of RNA in storage, 8–9

OOncolytic virotherapy, 316

PPalindromic sequences of nucleotides

(P nucleotides), 161PALM Microlaser Technologies, 4Paraffin, 5–7, 28, 42, 51, 55, 70–71, 82, 91,

165, 170, 177, 193, 273fPCN, see Plasma cell neoplasms (PCN)PCR, see Polymerase chain reaction (PCR)PDGFR and FGFR1 abnormalities

FGFR1, 243–244PDGFRA, 240–241

PDGFRB, 241–243spectrum of eosinophilia-related disorders,

239–240Peripheral T-cell lymphoma, not otherwise

specified (PTCL-NOS), 198–199Ph+ B-ALL, 107Phenol–chloroform, 7–8Philadelphia chromosome, 42, 55, 56f, 83, 107,

217–220, 220f, 222–224, 232, 293,309, 333

Phosphatase and tensin analog (PTEN), 119Phosphatidylinositol 3-kinase (PI3K), 107,

110, 118f, 119–120, 189, 222, 234,235f, 238, 245, 272, 277, 307–308

Phosphatidylinositol-3, 4, 5 trisphosphate(PIP3), 119

Photomultiplier tube (PMT), 12PI3K, see Phosphatidylinositol 3-kinase

(PI3K)PIP3, see Phosphatidylinositol-3, 4, 5

trisphosphate (PIP3)Plasma cell myeloma, 45t, 51, 67–70, 69f, 95,

170Plasma cell neoplasms (PCN), 68, 94–95, 165PLZF protein, see Promyelocytic leukemia

zinc finger (PLZF) proteinPML–RARA translocation, detection of,

130–133conventional karyotyping, 130exon–intron structure of PML and

RARA, 132fFISH probes, 130, 131fmolecular minimal residual disease

testing, 133RQ-PCR, results, 131–133, 132fRT-PCR, 130–131

PMT, see Photomultiplier tube (PMT)P nucleotides, see Palindromic sequences of

nucleotides (P nucleotides)Polyacrylamide gels, 12, 15Polyadenylated precursor (pri-miRNA), 325Polycythemia vera, 44, 57–58, 82, 83t, 84, 110,

216, 216t, 232–233, 307, 313Polymerase chain reaction (PCR), 2–4, 18–20,

41, 50, 82, 246, 269Post-germinal center B-cell malignancies, 29P210 protein, 221Precursor B-lymphoblastic lymphoma

(B-LBL), 103Precursor lymphoid malignancy

altered NOTCH signaling in T-ALL,117–119

Index 355

B-lymphoblastic leukemia witht(9;22)(q34;q11.2)

ALL lacking BCR–ABL1 expression,findings, 109

BCR–ABL1 fusion mRNA transcript,expression of, 107

BCR–ABL1 signaling and alteredIKAROS function, 108–109

genomic organization of the IKZF1gene, 108f

homodimerization of BCR–ABL1,results, 107

M-bcr, BCR break point inadult/pediatric B-ALLs, 107

mice deficient in IKAROS, defects, 108miRNAs, role, 109–110Ph+ B-ALL, clinical outcomes, 107Ph+ B-ALL, SNP analysis, 108

cancer stem cells in precursor B-ALL:definitions and controversies,113–116

cytogenetic/molecular lesions, 105–106B-ALL and T-ALL, genetic lesions,

105, 106tchromosomal translocation, effects on

gene expression, 106“type A mutation,” 106“type B mutation,” 106

down syndrome-associated ALLDS-ALL/ALL, cytogenetic analyses in

non-DS children, 110GATA1 mutations in TMD/AMKL,

110–111, 111fJAK2 mutations, 111JAK2 R683 mutation, 111pathogenesis of, role of CRLF2 in,

111–112identification of novel ALL subtypes, 120pathogenesis of T-ALL, role of PTEN/PI3K

array-based comparative genomichybridization, 120

mTOR, downstream target ofPI3K-AKT, 120

PTEN, negative regulator of PI3Ksignaling, 119

role of aberrant NOTCH1 andPI3K/AKT signaling, 118f,119–120

precursor lymphoblasticleukemia/lymphoma, geneticfactors

ALL development in children, riskfactors, 113

transplacental leukemic “metastasis,”113

WHO classification, 104–105, 104tPrecursor T-lymphoblastic leukemia

(T-ALL), 103Precursor T-lymphoblastic lymphoma

(T-LBL), 103Primary myelofibrosis (PMF), 58, 82, 83t, 84,

110, 216, 216t, 232, 313pri-miRNA, 325Promyelocytic leukemia zinc finger (PLZF)

protein, 328PTEN, see Phosphatase and tensin analog

(PTEN)Pyrosequencing, 14, 16–17, 231

Q8q24 gain, 262Q-PCR, see Quantitative real-time PCR

(Q-PCR)Quantitative real-time PCR (Q-PCR), 23–24,

128–139, 141, 268Quinacrine banding, 42

RRARS, see Refractory anemia with ring

sideroblasts (RARS)Real-time PCR, 21–23

signal detection options, 22–23Receptor tyrosine kinases (RTKs), 240,

303, 312examples, 303

Recombination-activating gene (RAG)proteins, 160f, 161

Refractory anemia with ring sideroblasts(RARS), 85–86, 234

Restriction enzymes (REs)defense mechanism of, 13DNA digestion by, 13methyltransferase enzymes, role, 13recombinant DNA technology, role in, 13variations in fragment pattern by

recognition sites, results, 13–14Restriction site PCR, 25Reverse transcription PCR (RT-PCR), 20, 24f,

57, 128–130, 132–133, 135–136,139, 145, 177, 189, 192, 196,222–223, 226f, 228, 231–232, 241,243, 246, 270f

The “Revised European–AmericanClassification of LymphoidNeoplasms” (REAL), 41

RISC, see RNA-induced silencing complex(RISC)

356 Index

Rituximab, 174–175, 277, 298, 299t, 302RNA-induced silencing complex (RISC), 326RNA stabilization tubes (PAXgene series), 4RTKs, see Receptor tyrosine kinases (RTKs)RT-PCR, see Reverse transcription PCR

(RT-PCR)RUNX1, 20, 53t, 60, 65, 88t, 104t, 106, 113,

129t, 133–135, 134fRUNX1T1, 20, 53t, 60, 88t, 129t, 133–134,

134, 134f

SSanger, 14Sanger chain termination methods, 14, 231Sanger sequencing, 14–15, 14–16, 231Sequencing by synthesis, see PyrosequencingSequencing, definition, 14Sequencing of DNA, 13–15, 17, 82Sézary syndrome (SS), 195

genetic abnormalities in, 195–196‘Shadow peak,’ 12SHIP, see Src homology 2 domain-containing

inositol-5-phosphatase (SHIP)Signal transducers and activators of

transcription (STATs), 107Signal transduction inhibitors, see Small

molecule drugsSilica column, 7–8Single nucleotide polymorphisms arrays

(SNP-arrays), 43, 74–75, 108, 267f,268–269

Single-stranded binding protein (SSB), 16–17SLL, see Small lymphocytic lymphoma (SLL)Small lymphocytic lymphoma (SLL), 67, 166,

255–256Small molecule drugs, 303–310

ATRA, 310–311deregulation of phosphorylation patterns,

effects, 303examples for AML, 311–312, 312texamples for treatment of CML, 312–313,

313texamples for treatment of MPN, 312–313FDA approved drugs, 304t–306thalf-life of, 308kinases

categories, 307conformations in activation loop of,

307–308multi-targeting approach

imatinib mesylate, treatment of CML,308–309

oral administration of drugs, 308

stages in clinical course of CMLaccelerated phase (AP), 309blast crisis (BC), 309chronic phase (CP), 309

tyrosine kinases, enzyme groups innon-RTKs, 303RTKs, 303

Small molecule drugs vs. mABs, 307–308Small molecule inhibitors, see Small molecule

drugsSNP-arrays, see Single nucleotide

polymorphisms arrays (SNP-arrays)Somatic hybridization technique/hybridoma

technology, 297–298Somatic hypermutation (SHM), 29–30, 159t,

161–166, 174, 179, 200, 259, 261,271f, 273, 275

Southern blotting technique, 26, 28Specimen collection and processing

cell enrichment and selection techniquesdensity-gradient centrifugation

methods, 3laser capture microdissection, 3preparation of leukocyte-rich layer, 3selective erythrocyte lysis, 3

nucleic acid integrity, source-specificrequirements for

bone marrow aspirates, whole blood,and body fluids, 4

dried blood spots (guthrie cards), 5fixed, paraffin-embedded tissue, 5–6fresh tissue, 5

patient identification and labeling, 2standard precautions and safety, 2

Specimen harvesting, 46Spectral karyotyping, 43Spectrophotometers, 9Spectrophotometric methods, 9Splenic marginal zone lymphoma, 73, 90t, 96,

166, 177Src homology 2 domain-containing

inositol-5-phosphatase (SHIP), 329SS, see Sézary syndrome (SS)SSB, see Single-stranded binding protein

(SSB)STATs, see Signal transducers and activators of

transcription (STATs)Stereotyped BCR, 274SYBR green, 10, 22

TT-ALL, see Precursor T-lymphoblastic

leukemia (T-ALL)

Index 357

TaqMan probes, 23, 236“Targeted therapy,” 294T-cell

biology and maturation, 182–184diversity of TCRs (TRA/TRB/

TRD/TRG), 182–183, 183f, 184fTCR delta (TRD), uniqueness,

182, 183fclonality testing

detection using primer designs, 185flow cytometric immunophenotyping

with antibodies, 186PCR, false-positive and false-negative

results, 186PCR for TCR gene rearrangement, 185predominant peak compared to

polyclonal background, 185–186Southern blot testing, 184

T-cell non-Hodgkin lymphoma (T-NHL),182–199

T-cell receptors (TCRs), 13, 26, 27t, 66, 70, 87,164, 182, 184f, 185, 193, 196–197,240, 272

TCRs, see T-cell receptors (TCRs)TdT, see Terminal deoxynucleotidyl transferase

(TdT)Terminal deoxynucleotidyl transferase (TdT),

27, 27f, 103, 160, 161fTherapeutic mABs, in targeted therapy,

296–301administration by intravenous

injections, 296anti-cancer action of, approaches,

296–297approved by FDA, 299t–301tchimeric and humanized types of

mABs, 298examples

AME-133, 302apolizumab, treatment of CLL,

302–303bevacizumab, treatment of colon

cancer, 302bispecific monoclonal antibodies,

treatment of CLL, 303combination treatment regimens of

mABs, 302GA101, 302lumiliximab, 302mABs plus chemotherapy, CHOP, 302rituximab and alemtuzumab, treatment

of NHL/CLL, 302

small molecules directed to TRAIL,treatment of NHL, 302

veltuzumab, 302fully humanized antibodies, development

of, 298general features, 298immunoglobulin polypeptide chains in, 296rituximab, treatment of NHL, 298somatic hybridization technique/hybridoma

technology, 297–298water-soluble proteins, 296

Thymic stromal-derived lymphopoietin(TSLP), 112

T-LBL, see Precursor T-lymphoblasticlymphoma (T-LBL)

T-lymphoblastic leukemia/lymphoma, 66–67,103, 104t, 164, 185, 241, 243

TMD, see Transient myeloproliferativedisorder (TMD)

t(2;14)(p16;q32), 265, 265ft(9;22)(q34;q11.2), see Philadelphia

chromosomet(14;19)(q32;q13), 262–263, 264fTRAIL, see Tumor necrosis factor-related

apoptosis-inducing ligand (TRAIL)Transient myeloproliferative disorder (TMD),

110Translocations involving chromosome 14q32,

266Transplacental leukemic “metastasis,” 113Treatment of AML, examples for, 311–312,

312tTreatment of CML, examples for, 313, 313tTreatment of MPN, examples for, 312–313Trisomy, 48Trisomy 12, 54t, 67, 73, 258f, 260, 263, 264f,

274Trisomy 3q27, 261TRIzol-LS R©, 8TSLP, see Thymic stromal-derived

lymphopoietin (TSLP)Tumor necrosis factor-related apoptosis-

inducing ligand (TRAIL), 259,302

Tumor-specific signature, 31Tumor suppressor genes (TSGs), 195, 198,

258–259, 316, 327Tyrosine kinase inhibitor therapy, 229–231Tyrosine kinases, 55, 58, 65, 73, 98, 106–110,

134, 143, 144f, 187, 189, 195,216–218, 216t, 221, 229–233,239–245, 272–273, 295, 303,307–309, 312

358 Index

UUniparental disomy (UPD), 74–75, 145,

235, 269UPD, see Uniparental disomy (UPD)

VVacutainer CPT Mononuclear Cell Preparation

Tube, 3Veltuzumab, 301t, 302

WWaldenström’s macroglobulinemia, 182Walter Flemming, 42WCP probes, see Whole-chromosome paint

(WCP) probesWHO classification (2008), cytogenetic/

molecular testsacute leukemia, 86–89

with no specific cytogenetic findings,89–90

chronic myeloproliferative disorders,82–85

diagnosis of hematolymphoidneoplasms, 97

diagnostic workups, process, 80–82clonal population tests/detection, 81evaluation of stained smears and tissue

sections, 80

flow cytometry analysis, 80immunohistochemical staining

procedures, 81molecular and cytogenetic procedures,

see Cytogenetic tests; Moleculartests

Hodgkin lymphoma, 97–98malignant lymphomas, 90–91

ALCL, 95–96Burkitt lymphoma, 94diagnostic problems, 91–92DLBCL, 93–94FL, 92MCL, 92MZL, 93NK-cell lymphomas, 96–97PCN, 94–95T-cell lymphomas, 95

myelodysplastic disordersrefractory cytopenia of childhood,

85–86other hematolymphoid neoplasms, 98

WHO Classification of Tumors of Hematopoi-etic and Lymphoid Tissues,41, 80

Whole-chromosome paint (WCP) probes,52–53

Wilms Tumor 1 (WT1), 145

the eye › public › books › biomed › hematopathology... · 2018-11-24 · editor domnita...

Documents