Phase 3 randomized study of adjuvant anastrozole (A), exemestane (E), or letrozole (L) with or without tamoxifen (T) in postmenopausal women with hormone-responsive (HR) breast cancer: The FATA-GIM3 trial

515 Background: Uncertainty still exists regarding the optimal schedule of adjuvant treatment of breast cancer with aromatase inhibitors (AI) and no trial has ever compared all the three AI. Methods: FATA-GIM3 is a multicenter, open label, 2x3 factorial phase 3 randomized study of adjuvant A, E and L upfront (UP - for 5 years) or sequentially (SEQ - for 3 years after 2 years of T) in postmenopausal HR breast cancer pts. Two comparisons were planned: UP vs SEQ and A vs E vs L. DFS (including local or distant relapse, second breast or non-breast cancer, DCIS and death, whichever came first) was the primary end-point; 2% at 5 yrs (corresponding to a HR of 0.79) was defined as the minimum difference required to declare superiority of UP vs SEQ. With two-tailed alfa 0.05, power 80%, 669 events and the enrolment of 3600 patients were planned. Following Data Monitoring Committee advice, final analysis was performed after 5yrs median follow-up. For each comparison a Cox regression model was applied adjusted by stratification factors and stratified by the other treatment factor. Analyses were based on intention-to-treat. Results: from 3/2007 to 7/2012, 3697 patients were enrolled at 76 centres. Median age 64, pT1 69.7% , pN0 64.3%, ER and PgR positive 88.9%, HER2 positive 8.9%, previous chemotherapy 38.3%. At 60 months median follow-up, 401 events were reported. 5yrs DFS was 89.8 with UP and 88.5 with SEQ (delta 1.32%, 95% CI -0.90-3.54; HR 0.89, 95% CI 0.73-1.08; P=0.23). 5yrs DFS was 90.0 with A, 88.0 with E and 89.4 with L (P=0.19). Conclusions: in the FATA-GIM3 trial there is a small non statistically significant DFS advantage for UP vs SEQ. No significant difference is evident among the three AI. Supported by the FARM5K3MEE AIFA grant from the Italian Drug Agency. Clinical trial information: NCT00541086.