Aims A few reports have assessed HER2 status in breast cancer by both dual-probe fluorescence in situ hybridisation (FISH) and immunohistochemistry (IHC) in an unselected and consecutive fashion, but CEP17 and HER2 copy number were not evaluated separately in these studies. Therefore, the aim of this study was to perform FISH testing for HER2 in a large number of breast tumours, irrespective of the IHC scores, which were also determined in all cases.

Methods Both FISH and IHC were applied to 200 tumours from 196 consecutive patients who underwent resection of primary breast cancer with the sentinel procedure and/or axillary dissection. Not only the ratio, but also mean HER2 and CEP17 copy number were determined and used in statistical analyses to evaluate relationships between FISH, IHC and clinicopathological features.

Results The amplification status based solely on HER2 signals was 98% concordant with results of dual-probe FISH. In non-amplified tumours, the mean CEP17 and HER2 copy number correlated, possibly because of cell cycling. Amplified tumours were histopathologically more aggressive than non-amplified tumours, and features of aggressiveness increased with the mean HER2 copy number. In both amplified and non-amplified tumours, a gene dosage effect was observed: an increase in the mean HER2 copy number was associated with a higher IHC score.

Conclusions This working method and analysis enabled new insights to be obtained into the pathobiology of HER2 in breast cancer. The findings may be helpful in optimising the methodology of HER2 testing.