ABSTRACT
While binary endpoints, such as overall response, are popularly used as the primary outcome of phase II cancer clinical trials, we sometimes use time to an event, such as time to progression or recurrence, as the primary outcome as well. When the study endpoint is time to an event, the maximum likelihood estimator (MLE) for exponential survival distributions may be used to compare survival distributions among treatment arms. Sample size calculation methods for test statistics based on the MLE of exponential distributions have been proposed by Pasternack and Gilbert (1971), George and Desu (1973), and Lachin (1981). Rubinstein et al. (1981) propose to use the sample size formula derived for the MLE test for the log-rank test by showing that this formula provides a reasonable power for the log-rank test through simulations. Their simulations are limited to balanced designs only.
