Original ArticleMiddle East Journal of Cancer; July 2015 6(3):

♦Corresponding Author:

Received: December 12, 2017; Accepted: January 6, 2018

IntroductionIntertumoral and intratumoral het-

erogeneities are fundamental motivesthat began the emerging era ofpersonalized cancer medicine.However, they are the mostchallenging and unresolved issues incancer medicine.1,2

Intertumoral heterogeneity is well-described in breast cancer. Geneexpression profiling techniques haveresulted in segregation of breasttumors into four major molecularsubtypes – luminal A, luminal B,Her2/neu, and triple negative(including basal-like). Subsequently,

efforts have been made to useimmunohistochemical stains assurrogate markers that assign tumorsto their corresponding molecularsubtypes.1,3

Despite these achievements inintertumoral heterogeneity, theproblem with intratumoralheterogeneity has not beensufficiently addressed and may beone of the most important reasonsfor unusual and unpredictedbehaviors in a number of breasttumors. Current standards for breastbiomarker assessment, such as thewidely used 2013 American Society

♦Corresponding Author: Mehdi Montazer, MDDepartment of Hematopatholo-gy, Molecular Pathology andCytogenetics, Room #210,Faculty of Medicine, Setad Sq.,Zand St., Shiraz University ofMedical Sciences, Shiraz, Iran Tel: +98 71 32301784 Fax: +98-71-32301784 Mobile: +98 912 288 9724

Email: mehdi.montazer@gmail.com

Case ReportMiddle East Journal of Cancer; October 2018; 9(4): 339-343

Intratumoral Heterogeneity in BreastCancer: A Case Report and Molecular

DiscussionAkbar Safaei, Ahmad Monabati, Maral Mokhtari, Mehdi Montazer♦

Department of Hematopathology, Molecular Pathology and Cytogenetics, Shiraz Universityof Medical Sciences, Shiraz, Iran

AbstractThe emerging era of personalized medicine makes it increasingly important to

consider intratumoral heterogeneity, which has been found in some breast cancercases. However, its identification criteria, form of reporting, and subsequent effects onthe clinical course of this disease remain controversial and not fully defined. Here, wereport and discuss a case of breast invasive ductal adenocarcinoma with substantialintratumoral heterogeneity, discrepancy between Her2/neu immunostaining and insitu hybridization, and disparity between estrogen receptor status before and afterneoadjuvant therapy.

Keywords: Breast, Estrogen receptor, ErbB-2, Genetic heterogeneity, Personalizedmedicine

Akbar Safaei et al.

Middle East J Cancer 2018; 9(4): 339-343340

of Clinical Oncologists/College of AmericanPathologists (ASCO/CAP) guidelines, mainlyfocus on the overall status of hormone receptorsand Her2/neu status. These guidelines do not statethe problem of molecular heterogeneity, with theexception of Her2/neu heterogeneity in an in situhybridization (ISH) study.4,5

Here, we present a case with significantintratumoral heterogeneity and discrepancybetween Her2/neu immunostaining and ISH.

Case ReportA 70-year-old Iranian woman was admitted

to Shahid Faghihi Hospital, Shiraz, Iran in January2017 to undergo breast tumor resection surgery.Her disease, invasive ductal carcinoma with bonemetastases, was previously diagnosed throughcore needle biopsy (CNB) in another medicalfacility approximately six months before shereferred to our hospital. She underwentpreoperative radiation and chemotherapy thatincluded trastuzumab and letrozole.

Initially, the patient presented with back pain.Spinal magnetic resonance imaging (MRI) showedmultiple ill-defined signals in the thoracic, lumbar,and sacral vertebral bodies suggestive of bonemetastases. A subsequent bone scan supportedthe MRI findings, and also revealed metastases inher ribs and skull. Chest and abdominal computedtomography (CT) scans did not detect any furthermetastases. In a search to find the primary site, apalpable mass with nipple retraction and dischargewas discovered in the left breast. The subsequent

mammography showed an ill-defined density inthe upper outer quadrant of the left breast thatmeasured 5.5 cm in maximum diameter withaccompanying prominent axillary lymphadenopa-thy. The right breast and axillae were normal perthe imaging studies.

The initial CNB results confirmed the presenceof an invasive ductal carcinoma with weaklypositive estrogen receptor (ER) staining in 10%of the tumor cells, negative progesterone receptor(PgR), and positive Her2/neu (+3). However, thefinal resected specimen contained twointermingling distinct cell populations withdifferent histomorphology and immunohistochem-istry (IHC) features. The first cell population, alarge cell component (LCC), comprisedapproximately 80% of the overall tumor cellsand was composed of large cells with relativelyabundant cytoplasm, vesicular chromatin, andconspicuous nucleoli. The second populationconsisted of a small cell component (SCC) thatrepresented approximately 20% of the tumor cellscomprised of basaloid cells with scant cytoplasm,dense chromatin, and no nucleoli (Figure 1A).Both components predominantly consisted ofsmall to medium-sized clusters. However, theLCC also showed occasional sheets and frequentsingle cells. The SCC had prominent tubuleformation, whereas the LCC lacked this feature(Figure 1B). Mitotic activity was moderate (6-10mitoses per 10 high power fields) in the LCC,whereas the SCC showed exceptional mitoses.Overall, the LCC and SCC would have been

Figure 1. A) Two distinct intermixed tumor cell populations (original magnification, 400×; H&E stain). B) Tubule formation in the smallcell component (SCC) (Original magnification: 400×; H&E stain). C) Her2/neu immunohistochemistry (IHC) showed that the large cellcomponent (LCC) expressed Her2/neu protein (score: 3+), whereas the SCC did not express Her2/neu protein (score, 1+) (Originalmagnification: 400×).

Intratumoral Heterogeneity in Breast Cancer

Middle East J Cancer 2018; 9(4): 339-343 341

considered as grades 2 and 1, respectively, basedon the Nottingham modification of the Bloom-Richardson grading system.

Immunohistochemistry assessment indicatedthat both components were negative for ER andPgR. Her2/neu was 3+ in the LCC and 1+ in theSCC. Ki67 proliferation index was 20% in theLCC, but the SCC showed extremely lowproliferation activity (less than 1%). The highmolecular weight cytokeratin (CK5/6) and p53were non-reactive in both components (Figure1C).

Dual color Her2/neu chromogenic ISH (CISH)probes showed amplification of the Her2/neugene in both components. The average number ofHER2/neu signals per cell was 6, which madesmall clusters. The HER2/CEP17 ratio wasapproximately 3 (Figure 2).

Necrosis, lymphovascular invasion, and nippleor skin involvement were absent. However, all 33dissected axillary lymph nodes had macrometas-tases involvement with substantial extranodalextension. Interestingly, the only type thatmetastasized was the LCC.

Unfortunately, the patient sustained a fall witha femur fracture three weeks after her discharge.Her condition deteriorated thereafter and sheexperienced bedsores, renal failure due to a urinarytract infection, and sepsis. She passed away inMarch, 2017.

DiscussionHere, we reported a breast invasive ductal

carcinoma with noticeable intratumoralheterogeneity comprised of two different tumorcell populations with distinctions in histomor-phology, Her2/neu status, and molecularsubtyping.

The LCC cells were ER-/PgR-/Her2+(3+)/20% Ki67, whereas the SCC cells were ER-/PgR-/Her2- (+1) with an extremely lowproliferation index. These components wereclassified as Her2/neu and triple-negative(suspicious to basal-like), respectively, based onmolecular subtyping of breast cancer that usedIHC as a surrogate marker. Interestingly, SCCdid not express CK5/6 and was not categorized asbasal-like; unexpectedly, subsequent CISHrevealed that SCC was Her2/neu gene amplifiedand Her2/neu in nature.3

Her2/neu gene amplification is present inapproximately 1% of patients who lack Her2/neuprotein expression (IHC 0/1+).4,6 The ASCO/CAPstates that this phenomenon is most likely due toan erroneous immunostaining technique. Here,we have performed the Her2/neu IHC twice,followed by a review by two independentmolecular pathologists. Batch and internal controlsalso showed thorough staining patterns. Therefore,the possibility of a technical error seemed doubtful.Seol et al. have stated that this discrepancy is not

Figure 2. A) Another region of the tumor (H&E stain, original magnification: 400×). B) Its corresponding dual color chromogenic in situhybridization (CISH) which shows amplification of the Her2/neu gene. Green and red dots represent the Her2/neu gene and chromosome17 centromere (original magnification: 1000×). *: Small cell component (SCC); Arrows: Large cell component (LCC).

Akbar Safaei et al.

Middle East J Cancer 2018; 9(4): 339-343342

attributed to technique and it represents truebiological heterogeneity.7 Varga et al. attributedthis finding to variations in IHC results withinHer2/neu gene amplified cases. Thus, they haveencouraged the idea of performing ISH instead ofIHC on all breast carcinoma specimens.6 Webelieve that mechanisms underlying such findingsare derived from differences in Her2/neu geneexpression, at either the mRNA or protein levelsthat may prevent an amplified Her2/neu genefrom being ultimately expressed as an Her2/neuprotein on the cell’s surface.

Of note, +1 or +3 Her2/neu tumors do notneed reflex testing or further ISH study accordingto ASCO/CAP guidelines. Here, we performed theCISH assay for our own interests. It was clear thatSCC would have been falsely categorized as triplenegative (with no need for trastuzumab therapy)instead of the Her2/neu subtype (the maincandidate for trastuzumab targeted therapy) if wehad not performed ISH. In our opinion, this is ashortcoming of the ASCO/CAP guidelines thatmay overlook such patients.4

The LCC, as expected by its molecular char-acteristics, had an ominous behavior. In thispatient, there was involvement in all dissectedregional lymph nodes with extensive extranodalextension by the LCC. In contrast, the SCC did notshow any aggressive behavior in the currentspecimen. However, in this age of personalizedmedicine, the clinical importance of heterogeneousareas such as SCC should not be ignored. Vargaet al. have shown that Her2/neu amplified caseswith an IHC score of 1+ had similar overallsurvival as IHC score 2+/3+ patients withconcurrent gene amplification.6 Seol et al. havestated that intratumoral heterogeneity is a sign ofgenetic instability across tumor cell populations;as with other solid tumors, it harbors a worseprognosis.7 There is a lack of comprehensiveevidence on whether patients with discordantIHC and ISH results would benefit fromtrastuzumab. Therefore, ASCO/CAP does notcurrently support this idea. Overall, we agreewith Makroo et al. and advocate enrollment ofthese patients in future clinical trials.8

Low proliferative tumor populations, such asSCC, may not respond well to chemotherapyregimens and there is a greater chance for toleranceof primary treatments. In our opinion, if this casecould have survived the LCC, then it was theSCC that could become a therapeutic challenge.Unfortunately, our patient passed away after aboutnine months.

There was a discrepancy between ER resultsfrom the preliminary CNB which indicated weaklypositive ER (10%) and the post-neoadjuvantexcisional specimen (ES) that was ER negative.It was not clear to us whether this finding wasrelated to the type of specimen (CNB versus ES)or the time at which the specimen was obtained(before versus after neoadjuvant therapy).However, both options were reasonable. Previousstudies have shown that the agreement in ERbetween CNB and EB is not complete. Accordingto Tamaki et al., a discordancy rate of 4% isexpected.9 In addition, it has been welldemonstrated that hormone receptors andHer2/neu status may change throughoutneoadjuvant therapy. Niikura et al. have reportedthat 4.6% of ER-positive tumors become ER-negative after treatment.10 Regardless of thediscrepancy, it is wise to monitor molecular char-acteristics of a tumor during neoadjuvant therapyin order to appropriately guide the succeedingtargeted therapy.

According to the 2013 ASCO/CAP guidelines,it is not required to report Ki67 staining, (whichis necessary for molecular subtyping) and Her2gene heterogeneity. The guidelines do not addressthe importance of molecular subtyping.5 In ouropinion, the exact underlying mechanisms andprognostic consequences of intratumoralmolecular heterogeneity and discrepant Her2/neuassessments have yet to be fully realized. Werecommend including these subjects in the nextupdate of the ASCO/CAP guidelines in order toprovide unified definitions for further clinicaltrials and to aid clinicians in the practice ofpersonalized medicine.

Intratumoral Heterogeneity in Breast Cancer

Middle East J Cancer 2018; 9(4): 339-343 343

Acknowledgements We would like to thank Dr. Yahya Daneshbod

who kindly provided us the core needle biopsyspecimen.

Conflict of InterestNone declared.

References1. Cho SH, Jeon J, Kim SI. Personalized medicine in

breast cancer: a systematic review. J Breast Cancer.2012;15(3):265-72.

2. Verma M. Personalized medicine and cancer. J PersMed. 2012;2(1):1-14.

3. Tang P, Tse GM. Immunohistochemical surrogates formolecular classification of breast carcinoma: A 2015update. Arch Pathol Lab Med. 2016;140(8):806-14.

4. Wolff AC, Hammond ME, Hicks DG, Dowsett M,McShane LM, Allison KH, et al. Recommendations forhuman epidermal growth factor receptor 2 testing inbreast cancer: American Society of ClinicalOncology/College of American Pathologists clinicalpractice guideline update. Arch Pathol Lab Med.2014;138(2):241-56.

5. Fitzgibbons PL, Dillon DA, Alsabeh R, Berman MA,Hayes DF, Hicks DG, et al. Template for reportingresults of biomarker testing of specimens from patientswith carcinoma of the breast. Arch Pathol Lab Med.2014;138(5):595-601.

6. Varga Z, Tubbs RR, Moch H. Concomitant detectionof HER2 protein and gene alterations by immunohis-tochemistry (IHC) and silver enhanced in situhybridization (SISH) identifies HER2 positive breastcancer with and without gene amplification. PLoSOne. 2014;9(8):e105961.

7. Seol H, Lee HJ, Choi Y, Lee HE, Kim YJ, Kim JH, etal. Intratumoral heterogeneity of HER2 geneamplification in breast cancer: its clinicopathologicalsignificance. Mod Pathol. 2012;25(7):938-48.

8. Makroo RN, Chowdhry M, Kumar M, Srivastava P,Tyagi R, Bhadauria P, et al. Correlation between HER2gene amplification and protein overexpression throughfluorescence in situ hybridization and immunohisto-chemistry in breast carcinoma patients. Indian J PatholMicrobiol. 2012;55(4):481-4.

9. Tamaki K, Sasano H, Ishida T, Miyashita M, TakedaM, Amari M, et al. Comparison of core needle biopsy(CNB) and surgical specimens for accurate preoperativeevaluation of ER, PgR and HER2 status of breastcancer patients. Cancer Sci. 2010;101(9):2074-9.

10. Niikura N, Tomotaki A, Miyata H, Iwamoto T, KawaiM, Anan K, et al. Changes in tumor expression ofHER2 and hormone receptors status after neoadjuvantchemotherapy in 21,755 patients from the Japanese

breast cancer registry. Ann Oncol. 2016;27(3):480-7.