Résumé : Purpose: The purpose of this study is to evaluate HER-2 and topoisomerase IIα (topo IIα) as candidates for predicting the activity of anthracyclines in the adjuvant treatment of breast cancer patients. Experimental Design: In this study, we evaluated HER-2 and topo IIα gene aberrations by fluorescence in situ hybridization in a series of 430 primary breast cancer samples. Samples came from node-positive breast cancer patients randomly treated either with one of two anthracy-cline-based regimens [full-dose epirubicin-cyclophospha-mide (HEC) and moderate-dose epirubicin-cyclophospha-mide (EC)] or with cyclophosphamide, methotrexate, and 5-fluorouracil (CMF) in the context of a Phase III adjuvant therapy trial. Event-free survival comparisons were performed between the three study arms in the subgroups of HER-2-amplified and nonamplified tumors. An explorative analysis was also performed to evaluate the predictive value of topo IIα in the cohort of HER-2-amplified patients. Results: HER-2 amplification was observed in 73 of the 354 evaluable samples (21%), whereas topo IIα amplification was found in 23 of the 61 evaluable HER-2-amplified tumors (38%). The three event-free survival comparisons were CMF versus HEC, CMF versus EC, and EC versus HEC. Hazard ratios (HRs) and 95% confidence intervals (CIs) were as follows: (a) CMF versus HEC, HR = 1.42 for HER-2-amplified tumors (95% CI, 0.54-3.76; P = 0.48) and 0.84 for HER-2-nonamplified tumors (95% CI, 0.49-1.44; P = 0.53); (b) CMF versus EC, HR = 1.65 for HER-2-amplified tumors (95% CI, 0.66-4.13; P = 0.29) and 0.66 for HER-2-nonamplified tumors (95% CI, 0.39-1.10; P = 0.11); and (c) EC versus HEC, HR = 0.93 for HER-2-amplified tumors (95% CI, 0.31-2.77, P = 0.90) and 1.33 for HER-2-nonamplified tumors (95% CI, 0.82-2.14; P = 0.25). Observed HRs suggest that the anthracycline-based therapy could be more effective than CMF in the subgroup of HER-2-amplified patients, whereas treatments could be equally active in the HER-2-nonamplified cohort. topo IIα evaluation suggests that the superiority of anthracyclines over CMF in HER-2-amplified patients could be confined to the subgroup of topo Ilα-amplified tumors. Conclusions: HER-2 could have a predictive value for the activity of anthracycline-based regimens in the adjuvant therapy of breast cancer patients. The predictive value of HER-2 would most likely be related to the concomitant amplification of the topo IIα gene.