Keywords: event-free survival, non-proportional hazard, log-rank test,acute myeloid leukemia
In late phase randomized oncology clinical trials, the time-to-event (TTE) endpoints are commonly adopted as the primary endpoints for establishing efficacy evidence of investigational therapies. The overall survival (OS) is always considered as the gold standard in the oncology field, however the OS data can take years to mature and can be influenced by the use of post-study salvage therapies and supportive care. Therefore, to accelerate the development process and to better characterize the treatment effect of new investigational therapies, other time-to-event points such as event-free survival (EFS) are used as the surrogate endpoints for OS. Motivated by a confirmatory clinical trial on investigating an novel target therapy for treating acute myeloid leukemia (AML), we present our findings and considerations on the trial design with a special form of EFS as the primary endpoint, which is defined with a large number of events accrued on Day 1.