A multi-stage drop-the-losers design for multi-arm clinical trials.

James Wason,Nigel Stallard,Christopher Jennison,Jack Bowden

doi:10.1177/0962280214550759

Abstract

Multi-arm multi-stage trials can improve the efficiency of the drug development process when multiple new treatments are available for testing. A group-sequential approach can be used in order to design multi-arm multi-stage trials, using an extension to Dunnett’s multiple-testing procedure. The actual sample size used in such a trial is a random variable that has high variability. This can cause problems when applying for funding as the cost will also be generally highly variable. This motivates a type of design that provides the efficiency advantages of a group-sequential multi-arm multi-stage design, but has a fixed sample size. One such design is the two-stage drop-the-losers design, in which a number of experimental treatments, and a control treatment, are assessed at a prescheduled interim analysis. The best-performing experimental treatment and the control treatment then continue to a second stage. In this paper, we discuss extending this design to have more than two stages, which is shown to considerably reduce the sample size required. We also compare the resulting sample size requirements to the sample size distribution of analogous group-sequential multi-arm multi-stage designs. The sample size required for a multi-stage drop-the-losers design is usually higher than, but close to, the median sample size of a group-sequential multi-arm multi-stage trial. In many practical scenarios, the disadvantage of a slight loss in average efficiency would be overcome by the huge advantage of a fixed sample size. We assess the impact of delay between recruitment and assessment as well as unknown variance on the drop-the-losers designs.

Highlights

Testing multiple experimental treatments against a control treatment in the same trial provides several advantages over doing so in separate trials
Multi-arm multi-stage (MAMS) designs are of great interest in practice, as their use means more new treatments can be tested with the same limited pool of patients
Much of the methodology about designing MAMS trials has focused on designs in which treatments are dropped early if their test statistics are below some prespecified futility boundary

Summary

Introduction

Testing multiple experimental treatments against a control treatment in the same trial provides several advantages over doing so in separate trials. Other advantages include that direct comparisons can be made between experimental treatments and that it is administratively easier to apply for and run one multi-arm clinical trial compared to several traditional trials.. Multi-arm multi-stage (MAMS) clinical trials include interim analyses so that experimental treatments can be dropped if they are ineffective; if desired, the trial can be designed so that it allows early stopping for efficacy if an effective experimental treatment is found. Magirr et al. extend Dunnett’s multiple-testing procedure to multiple stages, which we refer to as the group-sequential MAMS design. In this design, futility and efficacy boundaries are prespecified for each stage of the trial. Since there are infinitely many boundaries that do so, a specific boundary can be chosen to minimise the expected number of recruited patients at some treatment effect, or by using some boundary function such as those of Pocock, O’Brien and Fleming, or Whitehead and Stratton.

Objectives

Methods

Results

Conclusion