richter syndrome in b-cell chronic lymphocytic leukemia
TRANSCRIPT
Pathology International
2003;
53
: 195–203
Review Article
Blackwell Science, LtdOxford, UKPINPathology International1320-54632003 Japanese Society of Pathology
534April 20031455
Richter syndrome in B-CLLN. Nakamura and M. Abe
10.1046/j.1320-5463.2003.01455.xReview Article195203BEES SGML
Correspondence: Naoya Nakamura, MD, Department of Pathology,Fukushima Medical University School of Medicine, 1-Hikarigaoka,Fukushima-shi, 960-1295, Japan. Email: [email protected]
Received 8 August 2002. Accepted for publication 6 December2002.
Richter syndrome in B-cell chronic lymphocytic leukemia
Naoya Nakamura and Masafumi Abe
Department of Pathology, Fukushima Medical University School of Medicine, Fukushima, Japan
Richter syndrome (RS) is well known as a secondary high-grade lymphoma, mostly diffuse large B-cell lymphoma(DLBCL) developed in patients with B-cell chronic lym-phocytic leukemia (B-CLL). In this review, we describeclinicopathological, histological, immunophenotypical andgenetic findings of RS. The patients with RS, regardless oftransformation of pre-existing clone or
de novo
malignantclone, were resistant to conventional combined chemo-therapy and died within months of diagnosis. Moleculartechniques can provide convincing results for the clonalrelationship of RS to pre-existing B-CLL. When RS carriesa same rearrangement band or a same sequence as B-CLLby Southern blotting or nucleotide sequence analyses ofimmunoglobulin heavy and/or light chain genes, it is sug-gested to that RS transforms from original B-CLL. Theseanalyses have showed that approximately two-thirds of RScases evolved from a B-CLL clone. How and where does theB-CLL clone evolve to RS? The genetic alteration of trans-forming B-CLL clone into RS has been addressed. Abnor-malities of chromosomes 11 and 14 were most frequentlyinvolved in RS, but non-specific. In addition, RS does notinclude chromosomal translocation between Ig locus andoncogenes or rearrangements of
bcl-6
gene, both of whichwere found in some
de novo
DLBCL. Several candidates,such as mutation of p53 gene and abnormalities of cyclindependent kinase inhibitor, have been proposed to play animportant role in the transformation of a part of B-CLL. How-ever, there is still uncertainly as to how B-CLL progressesor develops into RS.
Key words:
B-cell chronic lymphocytic leukemia, diffuse largeB-cell lymphoma, Hodgkin lymphoma, immunoglobulin heavychain gene, Richter syndrome, somatic mutation
Patients with a low-grade lymphoid neoplasm who normallyhave a survival of more than several years occasionally ter-
minate as a result of an aggressive high-grade lymphoma.
1,2
This frequently results in early death. This progression isdemonstrated by histological changes of neoplastic cells.Small to medium-sized cells might change to large-sizedcells. Richter syndrome (RS) was first described in 1928and is well known as a secondary large-cell immunoblasticlymphoma that develops in patients with B-cell chronic lym-phocytic leukemia (B-CLL).
3
Subsequently, Lorholary
et al.
reported similar cases, and named them RS.
4
There havebeen a lot of case studies and reviews about RS.
5–13
A con-cept of RS has been well built, but an effective therapy thatcan cure the patient with RS is not available. In addition,RS is known to evolve in the natural course of B-CLL with-out prior chemotherapy, and both RS related to clonalevolution
14,15
and RS unrelated to the original
15
clone havebeen reported.
In this review, we describe RS, mostly diffuse large B-cell lymphoma (DLBCL), in clinicopathological, histological,immunophenotypical and genetic aspects, as well asHodgkin’s lymphoma (HL), a rare variant of RS. The clonalrelationship between B-CLL and RS has been almost under-stood, but there is still uncertainly as to how and why B-CLLprogresses or develops into RS.
CLINICAL FINDINGS
Although the frequency of RS is reported to occur in 3–10%of patients with B-CLL, Giles
et al
. reported that the true inci-dence might be higher as postmortem examinations were notperformed in most patients, thus underestimating occult dis-ease.
10
Mauro
et al
. reviewed 1011 patients with B-CLL, inwhich 22 patients terminated to RS (2.2%).
12
These patientsconstituted 18 cases with DLBCL (1.8%) and four cases withHL (0.4%). A different frequency of RS in a younger groupfrom that in an older group was found. When comparing ayounger patients’ group (
£
55 years, 204 cases) and the othergroup (
>
55 years old, 807 cases), both groups showed anelevated rate of second primary cancers (8.3 versus 10.7%),whereas the occurrence of RS was significantly higher
196 N. Nakamura and M. Abe
in younger patients (5.9%, 12 patients versus 1.2%, 10patients;
P
<
0. 00001). A significant increased risk of devel-oping secondary cancers in CLL patients regardless of age isimportant, and it is interesting that patients
£
55 years of ageshowed a higher occurrence of RS than older patients.Geographically, B-CLL is the most frequent disorder amonghematological malignancies in Western countries, so that it islikely that RS terminated from B-CLL is largely described inWestern countries. Further, B-CLL is infrequently observedin other areas, probably as a result of differences in race. Forexample, the low occurrence rate of B-CLL in immigrantsfrom Japan to USA has been reported.
16
The accurate fre-quency of RS outside Western countries is unknown butextremely rare, although RS has been reported in Japan andTaiwan.
17,18
In addition, RS initiates with the elevation of serum lactatedehydrogenase, rapid lymph node enlargement, systemicsymptoms of fever and/or weight loss, and extranodalinvolvement. Patients sometimes have hepatosplenomegalyand central nervous system (CNS) involvement. The diagno-sis of RS is usually made by a biopsy of an enlarged lymphnode. The primary neoplasms of RS in extranodal sites with-out the enlargement of systemic lymph nodes have beenreported and will be described later. The prognosis of RS isquite poor. The patients with RS are resistant to conventionalcombined chemotherapy, and die within months of diagnosis.A new and strong therapy is required for RS. Giles
et al
. con-ducted a combined chemotherapy consisting of
cis
-platinum,fludarabine and cytosine arabinoside, and another one usingcyclophosphamide instead of
cis
-platinum.
10,19
Among 11patients with RS, two patients achieved complete remissionand three patients achieved partial remission. Recently, atrial of allogeneic bone marrow transplantation in eightpatients with RS has been done and three out of eightpatients (38%) were alive and in remission at 14, 47 and67 months, respectively.
20
HISTOLOGY AND IMMUNOPHENOTYPING OF RS
Richter syndrome was originally reported as immunoblasticlymphoma.
3
The typical morphology of immunoblastic lym-phoma, which is same as DLBCL, immunoblastic variant (IB)according to the World Health Organisation classification
21
inthe biopsy of a lymph node revealed a diffuse proliferation oflarge-sized cells that have large round to moderately irregularwith vesicular nuclei, prominent nucleoli and a moderateamount cytoplasm (Fig. 1). Cases with DLBCL, centroblasticvariant (CB) showing a diffuse and monotonous proliferationof centroblasts or intermingled of immunoblasts and centro-blasts (Fig. 2) were also reported. A precious proportion of IBversus CB in the literature is not estimated because of thefew descriptions of histological findings in some literature.
There are two cases of IB and three cases of CB in our file ofRS. Para-immunoblasts are named for medium to large-sizedimmunoblast-like cells that are scattered in pseudofolliclesof B-CLL (Fig. 3). Para-immunoblsts are positive for CD20,cytoplasmic IgM and MIB-1, and are regarded as proliferatingcells, but do not show a confluent sheet in B-CLL.
21,22
Otherthan DLBCL, HL is also reported as a variant of RS, and it willbe described later.
Because IB consists of approximately 10% of all DLBCL,
23
cases of IB in RS are much higher than that in
de novo
DLBCL. Lukes and Collins documented that approximately10% of newly diagnosed patients with immunoblastic lym-phoma might have had histories of prior lymphoproliferativereaction or malignancy, including RS.
24
Immunoblastic mor-phology is frequently found in secondary malignancy, and isespecially marked in RS. It is comparable that the trans-formation of low-grade lymphoma of mucosa-associatedlymphoid tissue (MALT) shows DLBCL centroblasticmorphology.
25
Most cases of initial B-CLL express CD19, CD20, surfaceIgM and D (normally weak), as well as CD5. Some cases of
Figure 1
Richter syndrome as diffuse large B-cell lymphomaimmunoblastic variant. Immunoblasts are large-sized cells that arelarge round to moderately irregular with vesicular nucleus, a promi-nent nucleolus in the center and a moderate amount of cytoplasm.Magnification,
¥
200.
Richter syndrome in B-CLL 197
RS have the same immunophenotype as the initial B-CLL,but others do not. Some cases of RS show negative or weakpositive expression of CD5 and IgD, even though the RS isevolved from the initial B-CLL clone. In this situation, thedownregulation of CD5 and IgD occurs during transforma-tion. In addition, CD5, IgM and D might be shared on RS,which is demonstrated to exhibit a new malignant clone. Inthe latter cases,
de novo
CD5
+
DLBCL occurs in patients withCD5
+
B-CLL.
CLONAL RELATIONSHIP AND IMMUNOGLOBULIN GENE ANALYSIS
Secondary DLBCL in RS is not always derived from the cloneof the pre-existing B-CLL. It may be transformed from theclone of the initial neoplasms or
de novo
. Before the geneticmethod was available, immunophenotyping of membraneand/or the cytoplasmic immunoglobulin light chain was used,and was only a key for examining the clonal relationshipbetween B-CLL and RS. When neoplastic B cells of two dif-ferent types carried the same light chain, RS is suggested totransform from the original B-CLL. However, when neoplasticB cells expressed a different light chain, RS was a
de novo
DLBCL.
26–28
Molecular techniques can provide convincingresults for clonal relationships. Southern hybridization withthe immunoglobulin heavy chain (IgH) gene and light chain(IgL) gene was expected to be the most useful key. A rear-rangement of a same-sized band for both B-CLL and sec-ondary DLBCL supports the same cell origin,
28–31
and arearrangement of a different-sized band suggests a newmalignant event.
32–37
Miyakawa
et al
. reported that B-CLL
Figure 2
Richter syndrome as diffuse large B-cell lymphoma cen-troblastic variant (original magnification
¥
200). Centroblasts arelarge-sized cells that have a large round to moderately irregular withvesicular nucleus and several nucleoli.
Figure 3
Para-immunoblasts inpseudofollicles, B-Cell chronic lympho-cytic leukemia (B-CLL; original magni-fication,
¥
50 (right),
¥
200 (left)). Para-immunoblasts, prolymphocytes andsmall lymphocytes are intermingledin pseudofolicles. Para-immunoblastsare medium to large-sized cells thathave a round to oval nucleus with dis-perse chromatin and a central nucleus.Para-immunoblasts are immunoblast-like cells, but do not show a confluentsheet in B-CLL. Prolymphocytes areslightly larger than small lymphocytesand have more disperse chromatinand small nucleoli.
a b
198 N. Nakamura and M. Abe
with lambda light chain and RS with kappa light chain hadidentical clonal rearrangements of the IgH chain gene.
32
Thisindicates that a different immunophenotype of CLL and sub-sequent RS does not indicate biclonality. Finally, the nucle-otide sequence of the IgH and/or IgL genes prooves a clonalrelationship of RS and pre-existing B-CLL. Cherepakhin
et al
.reported a common clonal origin in a case of RS by compar-ison of the two clones’ nucleotide sequences of the IgH geneby polymerase chain reaction (PCR).
14
Matolcsy
et al
. sub-sequently analyzed complementarily determining regionthree (CDR3) of paired B-CLL/small lymphocytic lymphoma(SLL) and secondary DLBCL samples for three cases ofRS.
15
Of the three cases, one case, in which the IgH generearrangement pattern in Southern blotting was the same forboth samples, showed an identical sequence in the B-CLL/SLL and secondary DLBCL components. In the other twocases, the IgH gene rearrangement pattern was different,and one showed a non-identical sequence while anothershowed an identical sequence. Thus, analysis of the nucle-otide sequence of the CDR3 region rather than Southernblotting is more useful for the determination of whether thetwo clones are identical. A clonal evolution of B-CLL into RScan be demonstrated by confirmation of identical rearrange-ment bands in Southern blotting. However, if a different rear-rangement band was detected, a transformation of B-CLLcan not be decided. The nucleotide sequence of the CDR3region, including V-D-J rearrangement of the IgH gene, canprovide definite evidence for the comparison of the twoclones. The CDR3 region consists of V-N-D-N-J. The numberof N varies in each clone because N is inserted by Terminaldeoxynucleotidyltransferase activity during VDJ rearrange-ment in the early phase of B-cell differentiation. When thelength and nucleotide sequence of CDR3 are matched, itconcludes that the two clones are identical. There is less pos-sibility that Ig of remaining CLL clone, not RS clone, would beamplified, and it would be concluded as the same clonal ori-gin of the two.
What is a proportion of transformed RS? We have found noreports documenting a precise relationship using nucleotidesequence analysis. Two reviews discussed its proportionusing literature of RS including Southern blotting and cyto-genetic analyses.
38,39
In a 1993 review by Foon
et al.
, nineout of 14 RS showed identical IgH or IgL rearrangement.
38
In1995, Bessudo and Kipps stated that 28 out of 37 RS casesappeared to have a common clonal origin by Southern blot-ting and cytogenetic analyses.
39
Although Giles
et al
. sug-gested that the proportion of transformed RS was estimatedat 60%,
10
this figure might be less. It is considered that thedifferent rearrangement band does not mean a differentclone in RS. Approximately two-thirds of cases (possiblymore) of RS are evolved from the B-CLL clone.
The analysis of the IgH gene variable region (VH gene) hasbeen extending our understanding of the biology of B-cells. A
large amount of data on the analysis of the rearranged VHgene obtained by PCR in normal and neoplastic B-cells hasdefined the clonal origin of various B-cell neoplasms.
40,41
The pre-germinal center (GC) B-cells (naive or virgin B)show a germline sequence, and GC and post-GC B-cellsshow somatic hypermutation. Both lymphoblastic leukemia/lymphoma and mantle cell lymphoma (MCL) lack significantmutations in their rearranged VH genes, suggesting that theyare of naive B-cell origin. Somatically mutated VH geneswere found in follicular lymphoma, MALT lymphoma, DLBCLand multiple myeloma (MM), which were suggested to be aGC and post-GC cell origin. However, CD5
+
B-CLL is uniqueand exhibits germline or somatic hypermutation of rear-ranged VH gene. Cases with germline VH gene are esti-mated as approximately 60% of cases.
42
Somatic mutationanalysis of the VH gene of RS has been described in only afew reports. The RS case reported by Cherepakhin
et al
.exhibited approximately 8% somatic mutation frequencyof the VH gene.
14
We have two cases with CD5
+
RS with0.5% in somatic mutation of VH gene, almost germlinesequence,
43
and with 8.0% (unpubl. data), respectively. TheB-CLL cases could be divided into two groups, cases withunmutated VH genes and cases with mutated VH genes, andB-CLLs with unmutated VH genes have a more aggressiveform.
44
It remains unclear whether development of RS mightor might not be related to mutation patterns. Further analysisfor RS and VH gene are required.
GENETIC ALTERATION OF RS
The molecular events that underlie the appearance of RSin CLL are poorly known. Chromosomal abnormalitiesobserved in RS are often complex and included 14q
+
,
+
12,del 11q (11q23), del 13q and del 17p.
8,45–47
These abnor-malities are present in the patient with or without previoustreatment for B-CLL. Trisomy 12 and deletion 13q14 are thecommon cytogenetic abnormlities in B-CLL, and Hebert
et al
. proposed that cytogenetic studies in the patient withRS should be examined using the resected lymph node bywhich the diagnosis of RS was made.
27
Although no specificchromosomal abnormalities have been detected so far,chromonsomes 11 and 14 were most frequently involvedin RS. Zhu
et al
. reported that 11q23 was occasionallyobserved in
de novo
DLBCL, but all three cases of RS exhib-ited the deletion at 11q23.
48
Chromosomal complex changesand instabilities are present in patients with RS. Han
et al
.suggested that multiple chromosal changes, which might notbe specific, are related to the development of RS in patientswith B-CLL.
45
Although the CD5 antigen is sometimes downregulated ornot expressed in RS, it is important to compare genetic alter-ations including cytogenetic abnormalities of RS to those in
Richter syndrome in B-CLL 199
de novo
CD5
+
DLBCL. Little has been reported about kary-otypic abnormalities of
de novo
CD5
+
DLBCL. We havereported complex karyotypes of three cases with
de novo
CD5
+
DLBCL. The important abnormalities mentioned abovein RS, such as 14q
+
and del 11q23, were not present inthese three cases.
49
Moreover, Matolcsy
et al
. reported that44% of cases of
de novo
CD5
+
DLBCL exhibit
bcl-6
rear-rangement, but RS has no alteration of the
bcl-6
gene so that
de novo
CD5
+
DLBCL and RS-associated DLBCL must begenotypically distinct.
50
Thus, RS is likeky to have a differentmechanism from that of
de novo
CD5
+
DLBCL.Most of
de novo
DLBCL are believed to be derived fromGC or post GC B cells, and its malignant transformation oftenoccurs in the GC.
41
Somatic hypermutation, class switchingand receptor editing of the IgH and/or IgL genes might affectand modify the DNA of GC B cells and play a important rolein tumorgesis of GC and post-GC B-cells, resulting in B celllymphoma, such as DLBCL. For example, chromosomaltranslocation of Ig gene and oncogene, such as
bcl-6
gene inDLBCL and c-
myc
in Burkitt lymphoma (BL), are assumed tobe caused by incorrect class switching in GC B cells. Somatichypermutation of the variable region of IgH and IgL genesintroduces normal B-cells to produce antibodies with highaffinity for the antigenie, and Küppers
et al
. indicated that amechanism called affinity maturation might be involved in thegeneration of translocation because it often introduces dele-tions and duplications, causing breaks in DNA.
41
Moreover,somatic mutation of the Ig gene as well as several protoon-cogenes have been recently found in DLBCL.
42,51
Therefore,GC is the most important microenvironment for malignanttranformation of
de novo
DLBCL.The p53 gene mutation was first reported to be associated
with B-cell type RS,
52
and it was thought that its mutationcould provide a clue as to the mechanism of transformationfrom a low-grade leukemia/lymphoma to a high grade one.However, RS with or without p53 gene abnormalities hasbeen reported, and the p53 gene abnormalities have beenshown to be present and/or develop in B-CLL without evi-dence of RS.
15,52,53
Thus, p53 gene abnormalities arerelated to some cases of RS in both the development of anew malignant clone and progression of pre-existing B-CLL.Recently, abnormalities of cyclin-dependent kinase inhibitor(CDKI) have been discussed in relation to RS. The CDKIsrepresent a class of negative regulatory elements of cellgrowth that supress the kinase activity of the cyclins andCDKIs. For example, p16
INK,4a
participates in the control ofthe G1 phase by inhibiting the kinase activities of CDKIs, andplay an important role as tumor suppressor genes and thep53 gene.
54,55
Mutations of the p53 gene and/or homo-zygous deletions of ARF/INK4a locus, which codes for twodifferent proteins of p16
INK,4a
and p14
ARF
, occur in approxi-mately 60% of RS cases.
15,56–58
Pinyol
et al
. reported thatbiallelic deletion of the p16
INK,4a
gene was not found in B-
CLL, but was found in RS, which was demonstrated to havea clonal evolution by Southern blotting, and that in sevencases of RS, three cases had at least one of p53/p14
ARF
gene and p16
INK,4a
gene mutation.
57,58
This alteration mightbe acquired during the transformation process and might-have played a role in its pathogenesis. Cobo
et al
. immuno-histochemically investigated cell cycle regulators in RS.59
While the majority of CLL displayed strong p27 immunoreac-tivity, RS was p27-negative. Most CLL cases showed noexpression of the Rb protein. In contrast, all RS exhibitedstrong Rb expression. The overexpression of cyclin D1 wasalso detected in a case of CLL evolving into RS and a caseof RS. Cyclin D1 is the key oncogene of bcl-1 locus is proline-rich attachment domain 1/cyclin D1, encodes a proteininvolved in the cell cycle, and cyclin D1 overexpression basedon t11,14 (q23;q32) is known to occur in MCL.60 It is uniquethat the overexpression of cyclin D1 without such transloca-tion might be related to the transformation of RS. Recently,chromosomal imbalances in RS have been reported,61 withRS showing a higher number of gains, losses, total alter-ations and losses of 8p and chromosome 9 than B-CLL atdiagnosis. It is concluded that B-CLL has frequent chromo-somal imbalances that might increase during the progressionof the disease and transformation into RS.
Transformation of the Epstein Barr virus (EBV) appears tobe involved in the pathogenesis of several types of B- andT- cell lymphoma and HL. It is harbored in most cases ofendemic BL, pyothorax-associated lymphoma and NaturalKiller/T-cell lymphoma, frequent cases of non-Hodgkin’s lym-phoma (NHL) in patients with immunodeficiency, half of allcases of HL, and approximately 5% of conventionalDLBCL.21 The EBV-associated RS has been occasionallyreported. Ansell et al. reported that three DLBCL casesamong 25 cases of RS are harbored EBV.62 Giles suggestedthat a potential etiology for RS is long-term immunosuppres-sion with a causal relationship similar to that seen in NHLassociated with human immunodeficiency virus infectionor immunodeficiency after transplantation.10 It could beexplained by high frequency of secondary cancer in B-CLL.Few cases have examined both EBV association and theclonal relationship between RS and the pre-existing B-CLL.We reported a case of B-cell type RS associated with EBV.Secondary DLBCL occurring in the intestine was derivedfrom a de novo clone different from pre-existing B-CLL,17 andEBV might be associated with both de novo RS and trans-formed RS.
A number of alternative proposed mechanisms of trans-formation have been reported: (i) increased copy numberof c-myc oncogene by a combined modified comparativegenomic hybridization;63 (ii) RS expressed approximately 25-fold less A-myb RNA than the CLL cells;64 and (iii) alterationin 3/9 microsatellite repeats were detected in RS, but not inCLL.65
200 N. Nakamura and M. Abe
EXTRANODAL RS
Richter syndrome presents with systemic symptoms of local-ized or disseminated lymph node enlargement, as well asextranodal involvement, and is an aggressive lymphomaresistant from chemotherapy. Giles et al. reported 41% of RScases have extranodal involvement of the pleura (10.2%), theCNS (12.8%), the oropharynx (7.7%), the skin (5.1%), thegastrointestinal (GI) tract (5.1%), the lung parenchyma(2.5%), bone (2.5%) and marrow (5.1%).10 Bone marrowinvolvement of RS was infrequent. Further, DLBCL withoutenlargement of lymph nodes can occur in extranodal sites ofpatients with B-CLL. Secondary DLBCL has been reported tooccur in the following regions: cutaneous,66–68 GI tract,17,69
brain,70 lung71 and ocular.72 Generalized symptoms werenot conspicuous with weight loss and general weakness.Patients with GI-tract RS had bleeding, intestinal obstruction/ileus and acute perforation. The diagnosis of RS should becarefully made and exclude MALT lymphoma. A poor survivalwas archived in cutaneous RS,65,66 and GI-tract RS.69
Ratnavel et al. reported that a case of cutaneous RS, whichderived from an identical clone originated from B-CLL,showed an aggressive clinical course and died within4 months and a case of cutaneous RS exhibited a differentclone from B-CLL showed an indolent clinical behavior.68 Onthe other hands, six cases with primary GI tract RS describedby Parrens et al. are reported to show poor prognosis with amedian survival of 22 months.69 The series included twocases with a common clonal origin between RS and B-CLLand one case with different clones. We demonstrated a denovo RS without MALT-type low-grade lymphoma in thecolon.17 The patient died within several months. Therefore,most of the RS transformed from the B-CLL clone archivehave a poor prognosis similar to the original form of RS withnodal presentation, but a minority of RS with a localizedtumor, either of transformation or de novo, might show rela-tively good clinical behavior. The prognosis of extranodal RSmight be related to the clonal relationship to the superveningB-CLL.
RS OTHER THAN DLBCL
It is known that B-CLL rarely terminates at HL, and HL isalready called a variant of RS.10,73 However, can we acceptthat the progression of B-CLL to HL is a variant of RS? Threequestions would be applied: (i) Is the diagnosis of HL defi-nite?; (ii) Is HL supervening B-CLL aggressive clinical disor-der?; and (iii) Is the clonal evolution demonstrated in anycases? First, The presence of typical Hodgkin/Reed-Stern-berg (HRS) cells or lacunar cells in histology, as well asimmunohistochemical detection of CD15 and/or CD30 anti-gens on HRS cells, are required to make a diagnosis of HL
pre-existing B-CLL. It has been known that a small number oflarge-sized blastic cells scattered in a diffuse proliferation ofsmall mature lymphocytes in occasional cases of B-CLL. Inaddition, B-CLL occasionally exhibits a diffuse proliferation ofsmall mature lymphocytes with scattered giant cells resem-bling HRS cells74 and this histology of B-CLL mimics HL, butit should be not confused with HL. The HRS-like giant cells inthe background of B-CLL were sometimes positive for CD30,but not CD15. Second, Brecher and Banks reported eightcases with HL occurring in a course of B-CLL.73 The patientssurvived from 2 months to more than 8 years with a 50%cumulative survival of 12 months. Fayed et al. demonstrateda poor prognosis of HL supervening B-CLL.75 The patientsshowed generalized symptoms with progressive lymphaden-opathy and four out of seven cases of HL have died within1 year. Therefore, HL supervening B-CLL should be treatedas RS. Composite lymphoma of B-CLL and HL at the sameanatomical site is also excluded for RS because of its goodclinical behavior.73 Third, Ohno et al. examined the clonalrelationship by single cell PCR of IgH gene of HRS cells in HLas RS and B-CLL cells, and demonstrated the same clonalorigin of HRS cells and B-CLL cells.76 Cases of HL unrelatedto B-CLL have been reported.77 Most HL occurring inpatients with B-CLL are regarded as clonal evolution of B-CLL, whereas a small number of cases are de novo HL, rep-resenting a second malignancy presumably not clonallyrelated to CLL. Thus, HL subsequently occurring in thecourse of B-CLL is likely to be a variant of RS.
In most cases with de novo HL, HRS cells are of a B-cellphenotype and derived from GC B cells. Somatic hypermu-tation of rearranged VH gene of HRS cells were demon-strated by single cell PCR assay. CD5+ B-CLL comprisesunmutated cases and somatically hypermutated cases. TheHRS cells, usually negative for CD5, in secondary HL super-vening B-CLL have a clonal relationship with B-CLL. Wheredo HRS cells in a HL variant of RS occur? Do CD5+ B-CLLcells recruit to GC? Cases of HL underlying B-CLL are morefrequently associated with EBV than de novo HL.78–81 Themechanism of transformation of B-CLL into HL might be dif-ferent from de novo HL.
The frequency of patients with B-CLL terminating to HL isnot accurate, but Fayad et al. reviewed 1374 cases with B-CLL at a single institution and seven cases of HL occurredduring 21 years.75 Mauro et al. reported 22 patients with RSamong 1011 B-CLL, which consisted of 18 cases withDLBCL and four cases with HL.12 The rate of patients with B-CLL who developed in HL might account for approximately0.5% of cases. The report of long-term follow-up of patientswith B-CLL receiving Fludarabine regimens as initial therapy,which were produced from the same institution to Fayadet al., revealed four cases of RS found in 174 cases of B-CLL(2.3%).11 The patients of B-CLL treated with alkylatingagents might have a slightly high possibility of developing HL.
Richter syndrome in B-CLL 201
Eight cases of HD supervening B-CLL reported by Brecherand Banks were comprised of four cases of mixed cellular(MC)-HL and four cases of nodular sclerosis (NS)-HL. Fayeret al. also reported seven patients with HL developed in B-CLL comprising six cases with MC-HL and one case with NS-HL. The NS-HL cases showed a poor prognosis.
UNUSUAL SECONDARY LYMPHOID NEOPLASM
The NHLs other than DLBCL, for example, low-grade lym-phoma such as MM, seldom occur in patients with B-CLL, butcases have been reported.10 In a patient with B-CLL, MM isusually advanced at the time of presentation and the patientusually has a poor response and survival time.10 In twodifferent histological forms of B-CLL and MM, single clonalorigins were demonstrated by Southern blotting; however,no evidence at nucleotide sequence level has been foundto support this. T-cell lymphoma/leukemia has been alsoreported in patients with B-CLL. A case of T-cell acute lym-phoblastic leukemia and two cases with high grade T-cell lym-phoma are described.82,83 T cell RS developed from CLLwith a T-cell phenotype, which might be large granular lym-phocyte (LGL) leukemia, has been rarely reported and p53gene mutation was found in RS but not in CLL.84
REFERENCES
1 Lennert K, ed. Malignant Lymphoma, Lymphoplasmacytic/Lymphoplasmacytoid. 209–249 Malignant Lymphomas OtherThan Hodgkin’s Disease. Histology, Cytology, Ultrastructure,Immunology. New York: Springer-Verlag, 1978.
2 Gaidano G, Pastore C, Capello D, Cilli V, Saglio G. Molecularpathways in low grade B-cell lymphoma. Leuk Lymphoma1997; 26 (Suppl. 1): 107–13.
3 Richter MN. Generalized reticular cell sarcoma of lymph nodesassociated with lymphatic leukemia. Am J Pathol 1928; 4: 285–99.
4 Lortholary P, Boiron M, Ripault P et al. Leucemie lympoidechronique secodairement associee a une reticulpathie maligne,syndrome de Richter. Nouv Rev Fr Hemat 1964; 78: 621–44.
5 Armitage JO, Dick FR, Corder MP. Diffuse histiocytic lym-phoma complicating chronic lymphocytic leukemia. Cancer1978; 41: 422–7.
6 Trump DL, Mann RB, Phelps R, Roberts H, Conley L. Richteríssyndrome: diffuse histiocytic lymphoma in patients with chroniclymphocytic leukemia. A report of five cases and review of theliterature. Am J Med 1980; 68: 539–48.
7 Harousseau JL, Flandrin G, Tricot G, Brouet JC, Seligmann M,Bernard J. Malignant lymphoma supervening in chronic lym-phocytic leukemia and related disorders. Richter’s syndrome. Astudy of 25 cases. Cancer 1981; 48: 1302–8.
8 Robertson LE, Pugh W, O’Brien S et al. Richter’s syndrome: areport on 39 patients. J Clin Oncol 1993; 11: 1985–9.
9 Wetzler M, Kurzrock R, Goodacre AM, McLaughlin P, Ku S,Talpaz M. Transformation of chronic lymphocytic leukemia tolymphoma of true histiocytic type. Cancer 1995; 76: 609–17.
10 Giles FJ, O’Brien SM, Keating MJ. Chronic lymphocytic leuke-mia in (Richter’s) transformation. Semin Oncol 1998; 25: 117–25.
11 Keating MJ, O’Brien S, Lerner S et al. Long-term follow-up ofpatients with chronic lymphocytic leukemia (CLL) receiving flu-darabine regimens as initial therapy. Blood 1998; 92: 1165–71.
12 Mauro FR, Foa R, Giannarelli D et al. Clinical characteristicsand outcome of young chronic lymphocytic leukemia patients: asingle institution study of 204 cases. Blood 1999; 94: 448–54.
13 Pangalis GA, Angelopoulou MK, Vassilakopoulos TP,Siakantaris MP, Kittas C. B-chronic lymphocytic leukemia,small lymphocytic lymphoma, and lymphoplasmacytic lym-phoma, including Waldenström’s macroglobulinemia: a clinical,morphologic, and biologic spectrum of similar disorders. SeminHematol 1999; 36: 104–14.
14 Cherepakhin V, Baind SM, Meisenbolder GW, Kipps TJ. Com-mon clonal origin of chronic lymphocytic leukemia and highgrade lymphoma of Richter’s syndrome. Blood 1993; 82: 3141–7.
15 Matolcsy A, Inghirami G, Knowles DM. Molecular genetic dem-onstration of the diverse evolution of Richter’s syndrome(chronic lymphocytic leukemia and subsequent large cell lym-phoma). Blood 1994; 83: 1363–72.
16 Yanagihara ET, Blaisdell RK, Hayashi T, Lukes RJ. Malignantlymphoma in Hawaii-Japanese: a retrospective morphologicsurvey. Hematol Oncol 1989; 7: 219–32.
17 Nakamura N, Kuze T, Hashimoto Y et al. Analysis of the immu-noglobulin heavy chain gene of secondary diffuse large B-celllymphoma that subsequently developed in four cases with B-cell chronic lymphocytic leukemia or lymphoplasmacytoid lym-phoma (Richter syndrome) Pathol Int 2000; 50: 636–43.
18 Dunn P, Kuo TT, Tien HF. Richter’s syndrome: report of a case.J Formos Med Assoc 1995; 94: 686–8.
19 Giles FJ, O’Brien SM, Keating MJ et al. Sequential cis-platinum,fludarabine, and cytosine arabinoside (PFA) or cyclophospha-mide, fludarabine, and cytosine arabinoside (CFA) in patientswith Richterís syndrome. Blood 1996; 88(suppl.): 93a.
20 Rodriguez J, Keating MJ, O’Brien S, Champlin RE, Khouri IF.Allogeneic haematopoietic transplantation for Richter’s syn-drome. Br J Haematol 2000; 110: 897–9.
21 Jaffe SJ, Harris NL, Stein H, Vardiman JW, eds. Tumours ofHaematopoietic and Lymphoid Tissues. World Health Organi-zation Classification of Tumours. Lyon, France: InternationalAgency for Research on Cancer, 2001.
22 Nakamura N, Kuze T, Hashimoto Y et al. Analysis of the immu-noglobulin heavy chain gene variable region of 101 cases withperipheral B cell neoplasms and B cell chronic lymphocytic leu-kemia in the japanese population. Pathol Int 1999; 49: 595–600.
23 Diebold J, Anderson JR, Armitage JO et al. Diffuse large B-celllymphoma: a clinicopathologic analysis of 444 cases classifiedaccording to the updated Kiel classification. Leuk Lymphoma2002; 43: 97–104.
24 Lukes RJ, Collins RD. Tumors of the Hematopoietic System.Besthsda, Maryland: Armed Forces Institute of Pathology,1992.
25 Isaacson PG. Recent developments in our understanding ofgastric lymphomas. Am J Surg Pathol 1996; 20: S1–7.
26 Splinter TAW, Bom-Van Noorloss A, Van Heerde P. CLL anddiffuse histiocytic lymphoma in one patient: clonal proliferationof two different B cells. Scand J Haematol 1978; 20: 29–36.
27 Hebert J, Jonveaux P, d’Agay MF, Berger R. Cytogenetic stud-ies in patients with Richter’s syndrome. Cancer Genet Cytoge-net 1994; 73: 65–8.
28 Michiels JJ, van Dongen JJM, Hagemejer A et al. Richter’s syn-drome with identical immunoglobulin gene rearrangements in
202 N. Nakamura and M. Abe
the chronic lymphocytic leukemia and the supervening non-Hodgkin lymphoma. Leukemia 1989; 3: 819–24.
29 Ostrowski M, Minden M, Wang C, Bailey D. Immunophenotypicand gene probe analysis of a case of Richter’s syndrome. AmJ Clin Pathol 1989; 91: 215–21.
30 Coffrancesco E, Baldini L, Ciani A, Neri A et al. Evidence ofclonal progression in a case of Richter syndrome. Cancer 1993;71: 741–4.
31 Traweek ST, Liu J, Johnson RM, Winberg CD, Rappaport H.High-grade transformation of chronic lymphocytic leukemia andlow grade non-Hodgkinís lymphoma. Genotypic confirmation ofclonal identity. Am J Clin Pathol 1993; 100: 519–26.
32 Miyakawa K, Osada H, Yamauchi T et al. Single origin ofneoplastic B-cells with different immunoglobulin light chains ina patient with Richter’s transformation. Cancer 1990; 66: 140–44.
33 van Dongen JJ, Hooijkaas H, Michiels JJ et al. Richter’s syn-drome with different immunoglobulin light chains and differentheavy chain gene rearrangements. Blood 1984; 64: 571–5.
34 McDonnell JM, Beschorner WE, Staal SP, Spivak JL, MannRB. Richter’s syndrome with two different B-cell clones. Cancer1986; 58: 2031–7.
35 Sun T, Susin M, Desner M, Pergolizzi R, Cuomo J, Kodru P.The clonal origin of two cell populations in Richter’s syndrome.Hum Pathol 1990; 21: 722–8.
36 Tohda S, Morio T, Suzuki T et al. Richter syndrome with two Bcell clones possessing different surface immunoglobulins andimmunoglobulin gene rearrangements. Am J Hematol 1990; 35:32–6.
37 Matolcsy A, Casali P, Knowles DM. Different clonal origin of B-cell populations of chronic lymphocytic leukemia and large-celllymphoma in Richter’s syndrome. Ann NY Acad Sci 1995; 764:496–503.
38 Foon KA, Thiruvengadam R, Saven A et al. Genetic related-ness of lymphoid malignancies. Transformation of chronic lym-phocytic leukemia as a model. Ann Intern Med 1993; 119: 63–73.
39 Bessudo A, Kipps TJ. Origin of high grade lymphomas inRichter syndrome. Leuk Lymphoma 1995; 18: 367–72.
40 Muller-Hermelink HK, Greiner A. Molecular analysis of humanimmunoglobulin heavy chain variable genes (IgVH) in normaland malignant B cells. Am J Pathol 1998; 153: 1341–6.
41 Küppers R, Klein U, Hansmann ML, Rajewsky K. Cellular originof human B-cell lymphomas. N Engl J Med 1999; 341: 1520–29.
42 Klein U, Goossens T, Fischer M et al. Somatic hypermutation innormal and transformed human B cells. Immunol Rev 1998;162: 261–80.
43 Nakamura N, Hashimoto Y, Kuze T et al. Analysis of the immu-noglobulin heavy chain gene variable region of CD5-positve dif-fuse large B-cell lymphoma. Lab Invest 1999; 79: 925–33.
44 Hamblin TJ, Davis Z, Gardiner A, Oscier DG, Stevenson FK.Unmutated Ig V (H) genes are associated with a more aggres-sive form of chronic lymphocytic leukemia. Blood 1999; 94:1848–54.
45 Han T, Henderson ES, Emrich LJ et al. Prognostic significanceof karyotypic abnormalities in B cell chronic lymphocytic leuke-mia: An update. Semin Hematol 1987; 24: 257–63.
46 Catovsky D. The search for genetic clues in chronic lympho-cytic leukemia. Hematol Cell Ther 1997; 39: S5–11.
47 Brynes RK, McCourty A, Sun NC, Koo CH. Trisomy 12 inRichter’s transformation of chronic lymphocytic leukemia. Am JClin Pathol 1995; 104: 199–203.
48 Zhu Y, Monni O, Franssila K et al. Deletions at 11q23 indifferent lymphoma subtypes. Haematologica 2000; 85: 908–12.
49 Nakamura N, Kuze T, Hashimoto Y et al. CD5-positve diffuselarge B-cell lymphoma presenting in bone marrow without lym-phadenopathy. J Clin Exp Hematopathol 2001; 41: 73–80.
50 Matolcsy A, Chadburn A, Knowles DM. De novo CD5–positiveand Richter’s syndrome-associated diffuse large B cell lym-phomas are genotypically distinct. Am J Pathol 1995; 147: 207–16.
51 Pasqualucci L, Neumeister P, Goossens T et al. Hypermutationof multiple proto-oncogenes in B-cell diffuse large-cell lympho-mas. Nature 2001; 412: 341–6.
52 Gaidano G, Ballerini P, Gong JZ et al. P53 mutations in humanlymphoid malignancies: association with Burkittís lymphomaand chronic lymphocytic leukemia. Proc Natl Acad Sci USA1991; 88: 5413–7.
53 Fenaux P, Preudhomme C, Lai JL et al. Mutations of the p53gene in B-cell chronic lymphocytic leukemia: a report on 39cases with cytogenetic analysis. Leukemia 1992; 6: 246–50.
54 Serrano M, Hannon GJ, Beach D. A new regulatory motif incell-cycle control causing specific inhibition of cyclin D/CDK4.Nature 1993; 366: 704–7.
55 Hirama T, Koeffler HP. Role of the cyclin-dependent kinaseinhibitors in the development of cancer. Blood 1995; 86: 841–54.
56 Döhner H, Fischer K, Bentz M et al. P53 gene deletionspredicts for poor survival and non-response to therapy withpurine analogs in chronic B-cell leukemias. Blood 1995; 85:1580–89.
57 Pinyol M, Cobo F, Bea S et al. p16 (INK4a) gene inactivation bydeletions, mutations, and hypermethylation is associated withtransformed and aggressive variants of non-Hodgkin’s lympho-mas. Blood 1998; 91: 2977–84.
58 Pinyol M, Hernández L, Martínez A et al. INK4a/ARF locusalterations in human non-Hodgkinís lymphomas mainly occur intumors with wild-type p53 gene. Am J Pathol 2000; 156: 1987–96.
59 Cobo F, Martínez A, Pinyol M et al. Multiple cell cycle regulatoralterations in Richter’s transformation of chronic lymphocyticleukemia. Leukemia 2002; 16: 1028–34.
60 Garcia-Conde J, Cabanillas F. Mantle cell lymphoma: a lym-phoproliferative disorder associated with aberrant function ofthe cell cycle. Leukemia Suppl 1996; 2: S78–83.
61 Bea S, Lopez-Guillermo A, Ribas M et al. Genetic imbalancesin progressed B-cell chronic lymphocytic leukemia and trans-formed large-cell lymphoma (Richter’s syndrome). Am J Pathol2002; 161: 957–68.
62 Ansell SM, Li CY, Lloyd RV, Phyliky RL. Epstein-Barr virusinfection in Richter’s transformation. Am J Hematol 1999; 60:99–104.
63 Arranz E, Martinez B, Richart A et al. Increased C-MYC onco-gene copy number detected with combined modified compara-tive genomic hybridization and FISH analysis in a Richtersyndrome case with complex karyotype. Cancer Genet Cyto-genet 1998; 106: 80–83.
64 Golay J, Luppi M, Songia S et al. Expression of A-myb, but notc-myb and B-myb, is restricted to Burkitt’s lymphoma, sIg+ B-acute lymphoblastic leukemia, and a subset of chronic lympho-cytic leukemias. Blood 1996; 87: 1900–11.
65 Volpe G, Gamberi B, Pastore C et al. Analysis of microsatelliteinstability in chronic lymphoproliferative disorders. Ann Hematol1996; 72: 67–71.
66 Zarco C, Lahuerta-Palacios JJ, Borrego L, Toscano R, Gil R,Iglesias L. Centroblastic transformation of chronic lymphocyticleukaemia with primary skin involvement – cutaneous presen-tation of Richter’s syndrome. Clin Exp Dermatol 1993; 18: 263–7.
67 Cerroni L, Zenahlik P, Hofler G, Kaddu S, Smolle J, Kerl H.
Richter syndrome in B-CLL 203
Specific cutaneous infiltrates of B-cell chronic lymphocytic leu-kemia: a clinicopathologic and prognostic study of 42 patients.Am J Surg Pathol 1996; 20: 1000–10.
68 Ratnavel RC, Dunn-Walters DK, Boursier L et al. B-cell lym-phoma associated with chronic lymphatic leukaemia: two caseswith contrasting aggressive and indolent behaviour. Br J Der-matol 1999; 140: 708–14.
69 Parrens M, Sawan B, Dubus P et al. Primary digestive Richteríssyndrome. Mod Pathol 2001; 14: 452–7.
70 Mahe B, Moreau P, Bonnemain B et al. Isolated Richter’s syn-drome of the brain: two recent cases. Nouv Rev Fr Hematol1994; 36: 383–5.
71 Milkowski DA, Worley BD, Morris MJ. Richter’s transformationpresenting as an obstructing endobronchial lesion. Chest 1999;116: 832–5.
72 Hattenhauer MG, Pach JM. Ocular lymphoma in a patient withchronic lymphocytic leukemia. Am J Ophthalmol 1996; 122:266–8.
73 Brecher M, Banks PM. Hodgkin’s disease variant of Richter’ssyndrome. Report of eight cases. Am J Clin Pathol 1990; 90:333–9.
74 Momose H, Jaffe ES, Shin SS, Chen YY, Weiss LM. Chroniclymphocytic leukemia/small lymphocytic lymphoma with Reed-Sternberg-like cells and possible transformation to Hodgkinísdisease mediation by Epstein-Barr virus. Am J Surg Pathol1992; 16: 859–67.
75 Fayad L, Robertson LE, O’Brien S et al. Hodgkin’s disease vari-ant of Richter’s syndrome: experience at a single institution.Leuk Lymphoma 1996; 23: 333–7.
76 Ohno T, Smir BN, Weisenburger DD, Gascoyne RD, HinrichsSD, Chan WC. Origin of the Hodgkin/Reed-Sternberg cells inchronic lymphocytic leukemia with ‘Hodgkin’s transformation’.Blood 1998; 91: 1757–61.
77 Pescarmona E, Pignoloni P, Mauro FR et al. Hodgkin/Reed-Sternberg cells and Hodgkin’s disease in patients with B-cellchronic lymphocytic leukaemia: an immunohistological, molec-ular and clinical study of four cases suggesting a heteroge-neous pathogenetic background. Virchows Arch 2000; 437:129–32.
78 Rubin D, Hudnall SD, Aisenberg A, Jacobson JO, Harris NL.Richter’s transformation of chronic lymphocytic leukemia withHodgkin’s-like cells is associated with Epstein-Barr virus infec-tion. Mod Pathol 1994; 7: 91–8.
79 Simpson JL, Moriarty AT, Earls J, Snyder W, McCloskey DW.Hodgkin’s disease variant of Richter’s syndrome. Report of acase with diagnosis by fine needle biopsy. Acta Cytol 1997; 41:823–9.
80 Petrella T, Yaziji N, Collin F et al. Implication of the Epstein-Barrvirus in the progression of chronic lymphocytic leukaemia/smalllymphocytic lymphoma to Hodgkin-like lymphomas. AnticancerRes 1997; 17: 3907–13.
81 O’Sullivan MJ, Kaleem Z, Bolger MJ, Swanson PE, Zutter MM.Composite prolymphocytoid and Hodgkin transformation ofchronic lymphocytic leukemia. Arch Pathol Lab Med 2000; 124:907–9.
82 Lee A, Skelly ME, Kingma DW, Medeiros LJ. B-cell chronic lym-phocytic leukemia followed by high grade T-cell lymphoma. Anunusual variant of Richter’s syndrome. Am J Clin Pathol 1995;103: 348–52.
83 Novogrudsky A, Amorosi EL, Gottesman SR. High-grade T-celllymphoma complicating B-cell chronic lymphocytic leukemia.an unusual manifestation of ‘Richter’s syndrome’. Am J Hema-tol 2001; 66: 203–6.
84 Nakamura N, Kuze T, Hashimoto Y et al. Richter transformationof a T cell phenotype with p53 gene mutation. Eur J Haematol1997; 59: 331–2.