Assessment of HER2 status in invasive breast cancer with increased centromere 17 copy number by fluorescence in situ hybridization

Abstract

Background: A subset of breast cancers shows increased copy numbers of chromosome 17 centromere on in situ hybridization (ISH). However, recent studies have revealed that true polysomy 17 is a rare event in breast cancer, and that an increased copy number of centromere 17 represents amplification or copy number gain in and around the centromeric region. In such instances, the use of chromosome enumeration probe targeting centromere 17 (CEP17) in HER2 ISH is limited

thus, alternative methods for precise assessment of HER2 status are necessary. Performing ISH using probes for other genes on chromosome 17 as additional reference genes has been proposed by 2013 ASCO/CAP guidelines as well as several previous studies. In this study, we applied this method to breast cancers with increased CEP17 copy numbers (≥ 2.6) and compared it with conventional methods based on the 2007 and 2013 ASCO/CAP guidelines. Methods: After reviewing all HER2 fluorescence in situ hybridization (FISH) reports recorded from June 2004 to December 2011 at Seoul National University Bundang Hospital, we identified 300 cases (29.6%) with CEP17 copy number ≥ 2.6 from 1013 breast cancers. We performed FISH with probes for RARA, SMS, and TP53 genes on 253 breast cancers that had available tissue blocks, using tissue microarrays. If one or more gene had a mean copy number <2.6 the largest number for that gene(s) was chosen as an alternative to the CEP17 copy number, and we re-graded the HER2 status based on HER2: alternative gene ratio. After re-grading the HER2 status, we selected 8 cases which represent the various patterns of copy number alterations on chromosome 17, and performed high-resolution array-based comparative genomic hybridization (aCGH) to confirm the genomic copy number variation. Results: Of the 243 cases in which re-grading were possible, only 2 had copy numbers ≥2.6 for RARA, SMS and TP53 gene. Of 151 breast cancers classified as HER2 non-amplified by the 2007 ASCO/CAP guidelines using the HER2:CEP17 ratio (<1.8), 42 (27.8%) were re-graded as amplified and 33 (21.8%) as equivocal after FISH using additional reference genes. Of the 101 HER2 non-amplified cases by the 2013 ASCO/CAP guidelines, 2 (2.0%) were reclassified as amplified and 24 (23.8%) as equivocal. Of 46 equivocal cases, 35 (76.1%) were re-graded as amplified. After re-grading, amplified cases were significantly increased, and the concordance between HER2 FISH and immunohistochemistry decreased. Of the 8 cases analyzed by aCGH, six were upgraded from non-amplified to amplified by additional FISH studies. However, only 3 cases were proven to have HER2 amplification on aCGH. Two cases which were assumed to have true polysomy 17 by additional FISH studies were proven not to be polysomic. We also reviewed the pathologic features of the cases whose HER2 status were upgraded to be amplified by additional FISH, but some pathologic features were not matched with those of HER2-amplied tumors. Conclusion: Using additional reference genes in combination might be an option for accurate HER2 evaluation in breast cancer with increased CEP17 copy numbers. However, it has some limitations. It can cause over-grading of HER2 status, when the tumor has loss of new reference genes. Especially three genes that we used in current study (SMS, TP53 and RARA) were not suitable for alternative reference gene when used independently. Moreover, copy number alterations detected by additional FISH and those by aCGH were not well-correlated. Thus, use of alternative genes on chromosome 17 such as SMS, RARA and TP53 instead of CEP17 is not still suitable to be applied in daily practice. Additional studies to search the most stable gene that rarely shows copy number alteration will be needed.