Comparison of automated and manual determination of HER2 status in breast cancer for diagnostic use: a comparative methodological study using the Ventana BenchMark automated staining system and manual tests.

This study was performed to test the validity of manual and automated HER2 tests in one hundred routinely formalin-fixed and paraffin-embedded diagnostic breast carcinoma tissues. Immunohistochemical (IHC) and fluorescence in situ hybridization (FISH) assays for HER2 were separately carried out in two institutes of pathology specialised in diagnostics of breast diseases. Manual immunostaining was performed by the Dako-HercepTest. Automated IHC and FISH were carried out in the Ventana BenchMark platform by using the Pathway-CB11 antibody and the INFORM(R) HER2 probe, respectively. Positivity rates varied between HercepTest (26%), automated CB11 IHC (23%) and automated FISH (22%). Overall concordance between positive (2+, 3+) and negative (0; 1+) results of manual and automated IHC was 97%, between automated FISH and IHC 92%, and between automated FISH and HercepTest 89%. The frequency of 2+ IHC scores was 13% using the BenchMark and 14% with the HercepTest; 6/12 and 8/14 of the respective cases were not amplified by FISH. Automated FISH was not interpretable in 11 of 100 specimens. In the 89 informative cases, automated IHC resulted in increased specificity (92% vs. 88%), increased positive predictive value (73% vs. 64%) and increased efficiency (92% vs. 89%). We conclude that automation improves the accuracy of HER2 detection in diagnostic breast carcinoma tissues and provides a new approach for the global standardization of clinical HER2 tests.