The present meta-analysis examines randomized trials of third-generation aromatase inhibitors (AIs) as alternatives to tamoxifen in three treatment settings: monotherapy, sequenced therapy and extended therapy. Eleven randomized controlled trials (RCTs) were chosen based on their similarity in terms of study design and included 34,070 post-menopausal women who had undergone surgery for estrogen-sensitive early breast cancer. DFS was significantly improved by AI monotherapy (Hazard Ratio (HR): 0.89, p = 0.001), sequenced therapy (HR: 0.7, p < 0.00001) and extended therapy (HR: 0.62, p < 0.00001). All of the patients benefited significantly from sequenced therapy (HR: 0.81, p = 0.003), and hormone receptor-positive patients benefited from AI monotherapy (HR = 0.92, p = 0.046) with respect to OS. AI monotherapy conferred significantly lower risks for thromboembolic events (OR = 0.61; p < 0.001) and endometrial cancer (OR = 0.26; p < 0.001) compared with tamoxifen monotherapy; however, there was a greater risk of cardiovascular events (OR = 1.20; p = 0.030). Sequenced therapy was also superior in terms of endometrial cancer but was inferior with respect to fractures, thromboembolic and cardiovascular events. (C) 2013 Elsevier Ltd. All rights reserved.