Predicting HER2 amplification in breast cancers with equivocal HER2 immunohistochemistry

Accurate determination of human epidermal growth factor receptor 2 (HER2) status is essential for guiding targeted therapy in breast cancer. Immunohistochemistry (IHC) is routinely used for initial screening. Although reflex FISH testing is recommended for all IHC 2+ tumors, only a subset ultimately demonstrate HER2 amplification. We sought to identify clinicopathologic predictors of HER2 amplification among breast cancers with equivocal HER2 IHC results. A single-institution retrospective review was conducted of breast cancer cases from 2018 to 2024 with HER2 IHC 2+ results and corresponding FISH testing. Demographic and clinicopathologic variables were evaluated using univariate and multivariable logistic regression. Receiver operating characteristic (ROC) analysis, area under the curve (AUC), and Youden Index calculations were used to assess predictive performance and identify optimal Ki-67 thresholds. A total of 417 patients were included in the predictive model analyses. Overall, 16.6% of IHC 2+ cases were HER2 FISH positive. Ki-67 proliferation index was the only independent predictor of HER2 amplification on multivariable analysis (p < 0.05), whereas tumor grade, mitotic index, and ER/PR status were not independently associated. ROC analysis demonstrated moderate discrimination (AUC 0.73). A Ki-67 cutoff of 20% achieved 96% sensitivity, 97% negative predictive value, and a 4% false-negative rate. Tumors with Ki-67 < 20% were almost uniformly FISH negative, while those with higher Ki-67 values demonstrated a progressively increased probability of HER2 amplification. Ki-67 was the strongest independent histopathologic predictor of HER2 amplification among IHC 2+ breast cancers. Although these findings are not intended to replace guideline-recommended reflex FISH testing, they provide a foundation for future development and validation of adjunctive risk-stratification models.