Key design considerations for adaptive clinical trials: a primer for clinicians

Abstract

This article reviews important considerations for researchers who are designing adaptive clinical trials. These differ from conventional clinical trials because they allow and even enforce continual modifications to key components of trial design while data are being collected. This innovative approach has the potential to reduce resource use, decrease time to trial completion, limit allocation of participants to inferior interventions, and improve the likelihood that trial results will be scientifically or clinically relevant. Adaptive designs have mostly been used in trials evaluating drugs, but their use is spreading. The US Food and Drug Administration recently issued guidance on adaptive trial designs, which highlighted general principles and different types of adaptive clinical trials but did not provide concrete guidance about important considerations in designing such trials. Decisions to adapt a trial are not arbitrary; they are based on decision rules that have been rigorously examined via statistical simulations before the first trial participant is enrolled. The authors review important characteristics of adaptive trials and common types of study modifications and provide a practical guide, illustrated with a case study, to aid investigators who are planning an adaptive clinical trial

Fig 1

Common types of adaptive trials. Sample size reassessment: if interim analysis shows worse results than expected, the sample size can be reassessed and increased to ensure that the trial is adequately powered. Response adaptive randomisation: if interim analysis shows promising results for a treatment, the allocation ratio can be modified to favours enrolment to that treatment. Adaptive enrichment: if interim analysis shows that a treatment has more promising results in one subgroup of patients, the study eligibility criteria can be modified to investigate the efficacy of the intervention in the that subgroup, with a sample size reassessment to ensure a sufficient sample size. SSR=sample size reassessment

Fig 2

Adaptive design planning process

Fig 3

Case study of a dose-response adaptive trial. VAS=visual analogue scale