Abstract P5-03-10: Development of a novel HER2 testing strategy, using image-based cell-sorting to isolate pure cell populations from FFPE upstream of FISH

Abstract Fluorescent in Situ Hybridization (FISH) guidelines defined by American Society of Clinical Oncology (ASCO) and the College of American Pathologists for determining HER2 status are set to improve accuracy and usefulness as a diagnostic marker in breast cancer. Despite these guidelines, many factors can influence HER2 testing results such as sample preparation, assay-conditions and interpretation of test results due to heterogeneous breast cancer samples. In this multi-site study, sample preparation was carried out using the DEPArray™ to recover pure tumor cell populations from formalin-fixed, paraffin-embedded (FFPE) breast tumor samples. We then compared HER2/CEP17 ratios obtained from the DEPArray™ processed samples from each laboratory to routine FISH on tissue sections. Methods: Eight breast FFPE tumor tissue biopsies were obtained from commercial tissue banks. From the paraffin tissue blocks, four consecutive tissue curls (each 50 microns thick) were prepared. One curl from each of the 8 patient samples was distributed to four different laboratories for analysis following DEPArray™ based sample preparation. After an initial disassociation of each curl into a single-cell suspension, intact cells were sorted and then recovered based on cytokeratin/ vimentin/DAPI staining using the DEPArray™. Cytokeratin+/Vimentin-/DAPI+ tumor (~250) and Cytokeratin-/Vimentin+/DAPI+ stromal (~250) recovered cells were then deposited onto glass slides prior to standard dual-color HER2/CEP17 FISH analysis for comparison to conventional HER2 FISH result. Results: Serially sectioned breast tumors from 8 negative/positive cases: 7 infiltrating ductal carcinoma (IDC) and 1 metastatic carcinoma were studied. All four sites demonstrated 100% concordance between FISH results compared to the conventional HER2 FISH result. Overall, >60% of DEPArray™ isolated cells were recovered from FFPE samples that ranged from 1- 15 years of age and reported to contain 60% to 80% tumor content. The use of pure sorted cells permitted the accurate determination of HER2 amplification status in only the tumor cells while the stromal cells consistently yielded a more normalized ratio of HER2 to centromere 17. Conclusion: The preliminary results of this multi-site study demonstrate that use of DEPArray™ for sorted pure populations is reproducible as well as reliable method for subsequent analysis of HER2 by FISH on FFPE derived tumor cells. Given that traditional FFPE-based HER2 FISH results may be influenced by the tissue sectioning procedure, tissue heterogeneity and/or the scattering of few HER2 amplified tumor cells among normal stromal cells. The DEPArray™ allows analysis of immunofluorescence images and DNA content to isolate and recover pure and intact cell populations. This isolation of pure cell populations prior to FISH analysis is attractive for achieving precise determination of HER2 status on equivocal cases. A more formal analytical validation of this approach through CLIA is currently underway. Citation Format: Gerber A, Konig L, Millner L, Strotoman L, Khurana A, Kasimir-Bauer S, Moore MW, Cotter PD, Bischoff F. Development of a novel HER2 testing strategy, using image-based cell-sorting to isolate pure cell populations from FFPE upstream of FISH [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P5-03-10.