Aims Eligibility for anti-human epidermal growth factor receptor 2 (HER2) treatments in breast cancer requires a correct HER2 status assessment. Testing guidelines recommend fluorescence in-situ hybridisation (FISH) for samples scored as 2+ by immunohistochemistry. This study investigates the correlation between pathological features and FISH amplification in HER2 2+ breast cancer cases.

Methods 480 HER2 2+ breast cancer samples were included. The association between tumour grade, hormone receptor status, proliferation index (Ki67) and FISH amplification, using both US Food and Drug Administration (ratio ≥2) and American Society of Clinical Oncologists/College of American Pathologists cut-offs (ratio >2.2) was evaluated.

Results 90.2% of the samples were hormone receptor positive. The median Ki67 value was 23.5%; 311 (64.8%) samples showed a Ki67 value of 15% or greater. Tumour grade was evaluable in 421 cases (87.7%), 268 (55.8%) being grade 3. FISH amplification rates were 27.5% (ratio ≥2.0) and 20.8% (ratio >2.2). Grade 3 tumours were more frequently amplified than grades 1–2 tumours: 34% versus 18% (ratio ≥2.0, p<0.001) and 27% versus 9% (ratio >2.2, p<0.001). Samples with Ki67 of 15% or greater showed higher amplification rates than low Ki67 samples: 31% versus 21% (ratio ≥2.0, p=0.022) and 25% versus 12% (ratio >2.2, p=0.003). The OR for FISH amplification was significant in the case of grade 3 and high Ki67 with both cut-offs.

Conclusions In this study, high tumour grade and high Ki67 significantly predicted FISH amplification in 480 HER2 2+ breast cancer samples.