Background: The 33-month median follow-up of the ATAC (anastrozole and tamoxifen, alone or in combination) trial showed a potential interaction between anastrozole and previous chemotherapy; however, this was much smaller at 47 months' median follow-up. When the effects of different chemotherapy regimens were evaluated at that time, the apparent interaction was limited to patients who had received cyclophosphamide, methotrexate, and 5-fluorouracil (CMF).
Methods: In this retrospective analysis of 68-month data, we investigated the impact of prior chemotherapy, including different chemotherapy regimens, on time to recurrence. The chemotherapy regimens were 1) CMF only; 2) anthracycline-containing regimens (anthracycline or anthracycline and CMF); and 3) other chemotherapy regimens, including taxane-containing combinations.
Results: No evidence was found for an interaction between prior chemotherapy and anastrozole (hazard ratio [HR] 0.89 vs. 0.74 for those with or without prior chemotherapy, respectively; P = .21 for interaction). For those with prior chemotherapy, the HR of anastrozole when compared with that of tamoxifen shifted from 0.98 (95% confidence intervals [CI], 0.76-1.28) at 47 months' median follow-up to 0.89 (95% CI, 0.71-1.12) at 68 months' median follow-up and was closer to the overall treatment effect (HR, 0.79; 95% CI, 0.70-0.90). No differences according to type of chemotherapy were seen, and a benefit for anastrozole was also now apparent for patients receiving prior CMF (HR, 0.89; 95% CI, 0.63-1.24).
Conclusions: On the basis of the 5-year Completed Treatment Analysis, the ATAC trial does not indicate that the relative treatment benefits of anastrozole differ significantly between patients who received prior chemotherapy and those who did not.
Copyright 2006 American Cancer Society.