Abstract
PURPOSE: In early breast cancer, HER2-directed therapies are approved for the treatment of patients with HER2-positive invasive breast cancer as defined by HER2 protein overexpression, or HER2 gene amplification with HER2/CEP17 ratios ≥2.2. Beyond this cut-off, however, it is unknown whether the efficacy of HER2-directed therapy improves with increasing HER2/CEP17 ratios. We evaluated whether quantitative assessment of the HER2/CEP17 ratio predicts pathological complete response (pCR) and event-free survival (EFS) in patients treated with neoadjuvant HER2-based regimen in the prospective phase III NeoALTTO trial. PATIENTS AND METHODS: 455 women with HER2-positive early breast cancer, who had received neoadjuvant trastuzumab and/or lapatinib, together with 12 cycles of weekly paclitaxel, were included in this analysis. The HER2/CEP17 ratio in the primary tumor samples was correlated with pCR and survival outcome. RESULTS: The median HER2/CEP17 ratio was 5.1 (range: 1.1-100.0), and ratios were not associated with age, hormone receptor (HR) status, or any other clinicopathological variable analyzed. The log HER2/CEP17 ratio significantly predicted pCR in both univariate (OR: 1.83; 95 % CI: 1.11-3.01, p = 0.0176) and multivariate analysis (OR: 1.79; 95 % CI: 1.07-2.99, p = 0.0257). Higher HER2/CEP17 ratios were, however, not associated with improved EFS (adjusted HR = 0.795; p = 0.3537). A pCR prediction model including HER2/CEP17 ratio, treatment arm, and HR status improved the predictive strength of treatment arm alone from a ROC AUC value of 0.60-0.69. CONCLUSION: In patients treated with HER2-based neoadjuvant therapy, quantitative analysis of the readily available pretreatment HER2/CEP17 ratio by FISH is predictive of pCR.