Study 3004 was held by three different sponsors, leading to several major protocol amendments. Under Wyeth’s original protocol (April 2009), the study was designed to enroll 3850 patients in order to observe the 337 DFS events necessary to detect a hazard ratio of 0.70 with 90% power and a one-sided significance level of 0.025. There were two planned interim analyses at approximately 135 (for futility only) and 236 (for futility and efficacy; efficacy boundary: p-value