Discrepancies between immunohistochemistry and fluorescence in situ hybridization for HER-2 testing in breast cancer: The role of a quality assurance program on a cohort of 3564 patients

Abstract

679 Background: While determination of HER-2 status in breast cancer by fluorescence in situ hybridization (FISH) is now widely employed, its utility as part of a QC/QA program for IHC has not been established. The aim of this study was to (a) determine whether the use of FISH could improve the utility of immunohistochemistry (IHC), (b) attempt to identify sources of discrepancy between HER-2 protein, tested by IHC and HER-2 gene status, tested by FISH, and (c) design a reasonable and cost-effective testing algorithm. Methods: We commenced a QA/QC program for HER-2 testing in 1999, accruing 3564 tumors to date. As part of the program, parallel IHC testing was performed for each clinical FISH test. Routine components of this program also include random monthly review of HER-2 IHC and FISH results, targeted review of tumors with discrepant results, and intra- and inter-pathologist review of IHC and FISH scores. FISH and IHC tests were performed using FDA-approved testing kits (Vysis/Abbott and DakoCytomation, respectively). Results: With FISH to be the presumed gold standard, the sensitivity of IHC was 92.6%. The specificity of positive (3+) overexpression was 98%. The positive and negative predictive values of IHC were 94.8% and 97.4% respectively. Using some of the components of this QA/QC program, we were able to identify tumors with potential IHC-FISH discrepancy. These include tumors with (a) heterogeneity of gene status, and (b) polysomy of chromosome 17 (with/without true gene amplification). Significant discordance was only observed in the 2+ IHC group. Conclusions: Establishing a strict QC/QA program allowed us to significantly improve the sensitivity and specificity of IHC testing. According to these results, IHC can be used for screening. FISH testing should be mandatory only on tumors with 2+ IHC scores. Such program also helped identify sources of discrepant IHC and FISH results. Tumors containing either instable HER-2/neu gene or polysomy of chromosome 17 are largely the source of discrepant IHC and FISH results, and should be further investigated. Author Disclosure Employment or Leadership Consultant or Advisory Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration NexCura NexCura Genentech, Inc.