Statistics Seminar: Cornelia Kunz

Cornelia Kunz, University of Warwick

Friday 17 May 2013, 1400-1500
A54, Postgraduate Statistics Centre Lecture Theatre

Adaptive treatment selection in seamless phase II/III trials using short-term endpoints

Adaptive seamless phase II/III designs with treatment selection at an interim analysis have become increasingly more attractive due to their potential to save development costs and to shorten time-to-market of a new treatment. Different methods have been proposed for selection of the treatment group that will continue along with the control group to the second stage of the trial. If the primary endpoint is observed only after long-term follow-up it may be desirable to use short-term endpoint data at the interim analysis to select a treatment. In other cases, at least some long-term endpoint data might be available at the time of the interim analysis which might be used together with the short-term endpoint data to estimate the treatment effect upon which the treatment selection can be based.

While appropriate methods to combine data from different stages of the trial ensure control of the family-wise type I error rate in the strong sense, the power of the different approaches for treatment selection differs depending on several assumptions. In this talk we present the results of a formal comparison of different methods for treatment selection based on analytical results and a simulation study, together with a summary of the strengths and weaknesses of the approaches. Based on these results,we show how existing methods can be improved, increasing both the probability of selecting the most effective treatment and the power.