Abstract:
|
Cancer vaccine or immunotherapy trials have shown a delayed onset of effect for the experimental therapy, manifested as a delayed separation of the survival curves, thus a violation of the proportional hazards assumption. Recently, significant amount of work has been focused on the power loss incurred by the standard log-rank test, and various alternative tests have been proposed. In this work, we first present an R package that facilitates the sample size and power analysis for a weighted log-rank test that we recently developed to treat the problem, and the methods are extended to existing tests such as those in Fleming-Harrington class. Then we turn our focus to an estimand of the delayed treatment effect. We discuss the maximal likelihood estimator and propose alternative approaches to estimate the delay and the effect.
|