The study was designed to have 90% power to detect a hazard ratio for death of 0.75 (a 25% reduction in risk) in the TAS-102 group as compared with the placebo group, with a one-sided type I error rate of 0.025. Given the treatment assignment ratio of 2:1 (TAS-102:placebo), we calculated that 800 patients had to be enrolled in the study, and at least 571 events (deaths) would be required for the primary analysis.