Effect of anastrozole and tamoxifen as adjuvant treatment for early-stage breast cancer: 10-year analysis of the ATAC trial
Jack Cuzick,Ivana Sestak,Michael Baum,Aman Buzdar,Anthony Howell,Mitch Dowsett,John F Forbes,ATAC/LATTE investigators
DOI: https://doi.org/10.1016/S1470-2045(10)70257-6
Abstract:Background: The Arimidex, Tamoxifen, Alone or in Combination (ATAC) trial was designed to compare the efficacy and safety of anastrozole (1 mg) with tamoxifen (20 mg), both given orally every day for 5 years, as adjuvant treatment for postmenopausal women with early-stage breast cancer. In this analysis, we assess the long-term outcomes after a median follow-up of 120 months. Methods: We used a proportional hazards model to assess the primary endpoint of disease-free survival, and the secondary endpoints of time to recurrence, time to distant recurrence, incidence of new contralateral breast cancer, overall survival, and death with or without recurrence in all randomised patients (anastrozole n=3125, tamoxifen n=3116) and hormone-receptor-positive patients (anastrozole n=2618, tamoxifen n=2598). After treatment completion, we continued to collect data on fractures and serious adverse events in a masked fashion (safety population: anastrozole n=3092, tamoxifen n=3094). This study is registered as an International Standard Randomised Controlled Trial, number ISRCTN18233230. Findings: Patients were followed up for a median of 120 months (range 0-145); there were 24,522 woman-years of follow-up in the anastrozole group and 23,950 woman-years in the tamoxifen group. In the full study population, there were significant improvements in the anastrozole group compared with the tamoxifen group for disease-free survival (hazard ratio [HR] 0·91, 95% CI 0·83-0·99; p=0·04), time to recurrence (0·84, 0·75-0·93; p=0·001), and time to distant recurrence (0·87, 0·77-0·99; p=0·03). For hormone-receptor-positive patients, the results were also significantly in favour of the anastrozole group for disease-free survival (HR 0·86, 95% CI 0·78-0·95; p=0·003), time to recurrence (0·79, 0·70-0·89; p=0·0002), and time to distant recurrence (0·85, 0·73-0·98; p=0·02). In hormone-receptor-positive patients, absolute differences in time to recurrence between anastrozole and tamoxifen increased over time (2·7% at 5 years and 4·3% at 10 years) and recurrence rates remained significantly lower on anastrozole than tamoxifen after treatment completion (HR 0·81, 95% CI 0·67-0·98; p=0·03), although the carryover benefit was smaller after 8 years. There was weak evidence of fewer deaths after recurrence with anastrozole compared with tamoxifen treatment in the hormone-receptor-positive subgroup (HR 0·87, 95% CI 0·74-1·02; p=0·09), but there was little difference in overall mortality (0·95, 95% CI 0·84-1·06; p=0·4). Fractures were more frequent during active treatment in patients receiving anastrozole than those receiving tamoxifen (451 vs 351; OR 1·33, 95% CI 1·15-1·55; p<0·0001), but were similar in the post-treatment follow-up period (110 vs 112; OR 0·98, 95% CI 0·74-1·30; p=0·9). Treatment-related serious adverse events were less common in the anastrozole group than the tamoxifen group (223 anastrozole vs 369 tamoxifen; OR 0·57, 95% CI 0·48-0·69; p<0·0001), but were similar after treatment completion (66 vs 78; OR 0·84, 95% CI 0·60-1·19; p=0·3). No differences in non-breast cancer causes of death were apparent and the incidence of other cancers was similar between groups (425 vs 431) and continue to be higher with anastrozole for colorectal (66 vs 44) and lung cancer (51 vs 34), and lower for endometrial cancer (six vs 24), melanoma (eight vs 19), and ovarian cancer (17 vs 28). No new safety concerns were reported. Interpretation: These data confirm the long-term superior efficacy and safety of anastrozole over tamoxifen as initial adjuvant therapy for postmenopausal women with hormone-sensitive early breast cancer.