Use of anastrozole for breast cancer prevention (IBIS-II): long-term results of a randomised controlled trial

Background Two large clinical trials have shown a reduced rate of breast cancer development in high-risk women inthe initial 5 years of follow-up after use of aromatase inhibitors (MAP.3 and International Breast Cancer InterventionStudy II [IBIS-II]). Here, we report blinded long-term follow-up results for the IBIS-II trial, which comparedanastrozole with placebo, with the objective of determining the efficacy of anastrozole for preventing breast cancer(both invasive and ductal carcinoma in situ) in the post-treatment period. Methods IBIS-II is an international, randomised, double-blind, placebo-controlled trial. Postmenopausal women atincreased risk of developing breast cancer were recruited and were randomly assigned (1:1) to either anastrozole(1 mg per day, oral) or matching placebo daily for 5 years. After treatment completion, women were followed on ayearly basis to collect data on breast cancer incidence, death, other cancers, and major adverse events (cardiovascularevents and fractures). The primary outcome was all breast cancer. Findings 3864 women were recruited between Feb 2, 2003, and Jan 31, 2012. 1920 women were randomly assigned to5 years anastrozole and 1944 to placebo. After a median follow-up of 131 months (IQR 105–156), a 49% reduction inbreast cancer was observed for anastrozole (85 vs 165 cases, hazard ratio [HR] 0·51, 95% CI 0·39–0·66, p<0·0001).The reduction was larger in the first 5 years (35 vs 89, 0·39, 0·27–0·58, p<0·0001), but still significant after 5 years(50 vs 76 new cases, 0·64, 0·45–0·91, p=0·014), and not significantly different from the first 5 years (p=0·087).Invasive oestrogen receptor-positive breast cancer was reduced by 54% (HR 0·46, 95% CI 0·33–0·65, p<0·0001), witha continued significant effect in the period after treatment. A 59% reduction in ductal carcinoma in situ was observed(0·41, 0·22–0·79, p=0·0081), especially in participants known to be oestrogen receptor-positive (0·22, 0·78–0·65,p<0·0001). No significant difference in deaths was observed overall (69 vs 70, HR 0·96, 95% CI 0·69–1·34, p=0·82) orfor breast cancer (two anastrozole vs three placebo). A significant decrease in non-breast cancers was observed foranastrozole (147 vs 200, odds ratio 0·72, 95% CI 0·57–0·91, p=0·0042), owing primarily to non-melanoma skincancer. No excess of fractures or cardiovascular disease was observed. Interpretation This analysis has identified a significant continuing reduction in breast cancer with anastrozole in thepost-treatment follow-up period, with no evidence of new late side-effects. Further follow-up is needed to assess theeffect on breast cancer mortality.