Intralaboratory reproducibility of HER2 testing in breast cancer by immunohistochemistry and comparison of results obtained by different assays

Abstract

To investigate the intralaboratory reproducibility of immunohistochemistry (IHC) testing for HER2 status in breast cancer, and to evaluate the factors which influence the reproducibility. The concordance between monoclonal antibody CB11 and HercepTest was also assessed. HER2 overexpression on paraffin sections from thirty-seven cases of breast invasive ductal carcinoma was evaluated using CB11 and the evaluation procedure had been repeated for five times scored the tests together according to the HercepTest and new American Society of Clinical Oncology/College of American Pathologists (ASCO/CAP) grading schemes by 2 experienced pathologists together. Reproducibility rates of the five rounds were assessed using Kappa statistic, and the results from two scoring systems were compared. HercepTest kit was applied to the same cases afterward and the results were compared with CB11. Substantial intralaboratory reproducibility was achieved among 5 rounds tests. Excluding the influence effect of changing antibody lots, the intralaboratory reproducibility was closed to the perfect threshold (Kappa = 0.7858, HercepTest scheme). The results derived from the two grading schemes had an almost perfect agreement (Kappa = 0.8549). The concordance (positive vs. negative) between CB11 and HercepTest was 83.78%. Laboratory work with strict supervision and more experience will ensure a reliable testing consistency. Reproducibility analysis could be adopted to evaluate the intralaboratory staining quality on HER2 testing. Different antibody lots bring some influence to the intralaboratory reproducibility, but not significant. CB11 could be accepted to screen HER2 status in routine practice after testing validation.