Purpose The aim of the study was to evaluate the predictive value of HER2 and topoisomerase IIα gene (TOP2A) for the efficacy of epirubicin in the adjuvant setting of breast cancer patients. Patients and Methods In the Danish Breast Cancer Cooperative Group trial 89D, 980 pre- and postmenopausal primary patients were randomly allocated to either CMF (cyclophosphamide, methotrexate, and fluorouracil; n = 500) or CEF (cyclophosphamide, epirubicin, and fluorouracil; n = 480) times 9, between January 1990 and November 1999. Tumor tissue was retrospectively identified from 805 patients and was analyzed for HER2-positivity and for TOP2A-amplifications and deletions. Results HER2-positivity was found in 33% of the 805 investigated tumors and was not a predictive marker for epirubicin sensitivity. TOP2A changes were identified in 23% of the 773 investigated tumors: 12% had TOP2A amplifications and 11% had TOP2A deletions. We found that patients with TOP2A amplification had an increased recurrence-free (RFS; hazard ratio [HR], 0.43; 95% CI, 0.24 to 0.78) and overall survival (OS; HR, 0.57; 95% CI, 0.29 to 1.13), respectively if treated with CEF compared with CMF, and that patients with TOP2A deletions had an almost identical hazard ratio (RFS: HR, 0.63; 95% CI, 0.36 to 1.11; OS: HR, 0.56; 95% CI, 0.30 to 1.04). This is in contrast to patients with a normal TOP2A genotype for whom similar outcome was observed in both treatment arms (RFS: HR, 0.90; 95% CI, 0.70 to 1.17; OS: HR, 0.88; 95% CI, 0.66 to 1.17). Conclusion TOP2A amplification—and possibly deletion—seems to be predictive markers for the effect of adjuvant epirubicin containing therapy in primary breast cancer, but a final conclusion has to await a confirmative study or a meta-analysis.