What causes discrepancies in HER2 testing for breast cancer? A Japanese ring study in conjunction with the global standard

Abstract

We assessed interinstitutional and interobserver consistency of human epidermal growth factor receptor type-2 (HER2) testing using immunohistochemical analysis and fluorescence in situ hybridization (FISH) in a set of 20 breast cancer samples among 10 institutions in Japan and a Herceptin adjuvant study participating laboratory in Germany and identified factors that may lead to discordant results.We found a good agreement in immunohistochemical HER2 scoring between the coordinating institution and 10 participating laboratories (kappa = 0.718) and excellent agreement for FISH (kappa = 0.900). The results of a comparison between 10 Japanese laboratories and the German laboratory was good for immunohistochemical studies (kappa = 0.713) and excellent for FISH (kappa = 0.887). FISH retesting of equivocal samples (2+ immunohistochemically) improved agreement. Discrepancies between results were attributed to the evaluation process in 33.0% of the samples, staining procedures in 25.0%, and a combination of the two in 41.7%. Evaluation of samples according to the American Society of Clinical Oncology/College of American Pathologists guideline increased the number of 2+ immunohistochemical scores. By performing FISH retesting for these samples, consistency among multiple institutions could be archived. The quality of the staining procedures performed and the consistency of evaluations require regular assessment.