A stochastically curtailed two‐arm randomised phase II trial design for binary outcomes

摘要: Randomised controlled trials are considered the gold standard in trial design. However, phase II oncology with a binary outcome often single-arm. Although number of reasons exist for choosing single-arm trial, primary reason is that designs require fewer participants than their randomised equivalents. Therefore, development novel methodology makes more efficient value to community. This article introduces two-arm design includes stochastic curtailment (SC), allowing possibility stopping before final conclusions known certainty. In addition SC, proposed involves use block design, which allows investigators control interim analyses. approach compared existing also early stopping, through loss function comprised weighted sum characteristics. Comparisons made using an example from real trial. The comparisons show many possible functions, superior designs. Further, may be practical, by flexible One produces realisations when anticipated response rate low. this probability rejecting null hypothesis sensitive misspecification rate. considering II, we recommend preferred over other sequential