ABSTRACT

This chapter derives formulas for sample size calculation for continuous, binary response, and time-to-event study endpoints under a multiarm dose-response trial. It provides the sample size formula for the determination of minimum effective dose (MED) based on Williams' test. The chapter discusses sample size estimation and related operating characteristics of phase I dose escalation trials. It also provides sample size formula based on Cochran-Armitage's trend test for binary response. In general, linear contrast tests are useful in detecting specific shapes of the dose-response curve. In practice, the focus of phase I dose-response studies emphasize safety, while phase II dose-response studies emphasize the efficacy. It should be noted that the escalation algorithm and dose intervals not only have an impact on the sample size, but also affect other important operating characteristics such as the accuracy and precision of the estimation of the maximum tolerable dose (MTD) and the number of dose-limiting toxicity (DLTs).