A review of statistical methods on testing time-to-event data

Abstract

In oncology randomized clinical trials, the time-to-event(TTE) type of endpoints such as progression-free survival (PFS) and overall survival(OS), are commonly used as the primary or key secondary endpoints for comparing the experimental treatment and active control/ placebo. In practice, the proportional hazard (PH) is usually assumed to characterize the treatment benefit over time of TTE endpoints and calculate the required sample size. With the PH assumption, the hazard ratio (HR) between treatment arms is a constant over time, and the corresponding testing hypothesis is expressed as