Hypothesis testing for two-stage designs with over or under enrollment.

TitleHypothesis testing for two-stage designs with over or under enrollment.
Publication TypeJournal Article
Year of Publication2015
AuthorsZeng, Donglin, Fei Gao, Kuolung Hu, Catherine Jia, and Joseph G. Ibrahim
JournalStat Med
Date Published2015 Jul 20
KeywordsBiostatistics, Clinical Trials, Phase II as Topic, Computer Simulation, Confidence Intervals, Humans, Models, Statistical, Neoplasms, Sample Size

Simon's two-stage designs are widely used in cancer phase II clinical trials for assessing the efficacy of a new treatment. However in practice, the actual sample size for the second stage is often different from the planned sample size, and the original inference procedure is no longer valid. Previous work on this problem has certain limitations in computation. In this paper, we attempt to maximize the unconditional power while controlling for the type I error for the modified second stage sample size. A normal approximation is used for computing the power, and the numerical results show that the approximation is accurate even under small sample sizes. The corresponding confidence intervals for the response rate are constructed by inverting the hypothesis test, and they have reasonable coverage while preserving the type I error.

Alternate JournalStat Med
Original PublicationHypothesis testing for two-stage designs with over or under enrollment.
PubMed ID25809924
PubMed Central IDPMC4636905
Grant ListP01 CA142538 / CA / NCI NIH HHS / United States
R01 GM047845 / GM / NIGMS NIH HHS / United States