HER2 In Situ Hybridization: Validation and Implementation of Brightfield Assay in a US Academic Pathology Laboratory

Assessment of HER2 overexpression and gene (ERBB2) amplification remains an essential predictive test that determines tailored breast cancer therapy. Based on institutional needs, we recently validated the VENTANA HER2 Dual ISH DNA Probe Cocktail assay (DISH). Validation included testing 61 retrospective breast cancers, followed by 40 prospective cases (parallel testing). There was 99% concordance for binary positive/negative status when correlated with immunohistochemistry, and 87% concordance for exact ISH category (groups 1 to 5). We designed an online tool for automatic calculation of ratios and group assignment, with prompts for additional counting when needed. During the first year, 1286 DISH assays were performed, with 2.9% initial assay failures requiring repeat. Based on conservative guidelines in the first year, we sent confirmatory fluorescence in situ hybridization (FISH) in 4% of cases; 9 cases (0.7%) had discordant DISH and FISH results, all of which were near a threshold (including "low amplified" results with HER2/CEP17 ≥2, average HER2/cell 4 to 6). The average turnaround time for HER2 DISH from ordering to finalization was 3.0 days, versus 4.8 days for FISH at our institution (37.5% improvement). We encountered occasional pitfalls, including zones lacking hybridization signals, and enhanced silver dust associated with anthracosis or tattoo pigment. We also observed differences across whole slide scanner platforms. HER2 DISH advantages included the ability of pathologists to directly score slides in correlation with morphology and immunohistochemistry, improved turnaround time, and greater automation for high-volume HER2 testing, as compared with FISH. In summary, we found HER2 DISH to be an accurate and practical alternative.