RESUMEN
To analyze HER2 amplification in a large cohort of diagnostic breast cancer specimens, fluorescence in situ hybridization (FISH) and immunohistochemistry (IHC) were performed on the same specimens with use of Food and Drug Administration&approved products. Procedures were standardized following the manufacturers' recommendations. Of 116,736 IHC specimens, 20% were positive for HER2. In 16,092 FISH specimens, 22.7% showed HER2 amplification. In the subset of 6556 tissues analyzed with IHC and FISH, however, 59% were positive on IHC and 23.6% were amplified on FISH. The increased frequency of positive test results is skewed by more frequent reflex FISH testing. In general, expression and amplification trended together, with the least amplification (4.1%) seen in IHC-negative cases, 7.4% amplification seen in IHC 1+ cases, 23.3% amplification seen in IHC 2+ cases, and 91.7% amplification seen in IHC 3+ cases. When FISH amplification ratios were stratified, the low FISH ratios (2.0-2.2) were most frequently seen in specimens with negative IHC results, high ratios (>5.0) were seen in IHC 3+ specimens, and intermediate levels of amplification were similar for all levels of IHC. The effect of changing the cutoff point was analyzed: removing cases with a ratio of exactly 2.0 decreased the FISH positivity rate to 22.2% in the combined IHC and FISH cohort. Sequentially moving the cutoff point to 2.2 and 2.5 affected cases at all IHC expression levels. Each change removed approximately 2% from the apparent positivity. This large database provides the distribution frequency of HER2 protein expression and gene amplification in invasive ductal and lobular breast cancer. The relationship between level of HER2 amplification and clinical outcome will require reanalysis of pivotal trial data.