Adaptive Trial Designs

Reviewed and revised 30 March 2015

OVERVIEW

An adaptive clinical trial involves a study design in which key characteristics are adjusted while enrollment in the trial is ongoing using prospectively defined decision rules and in response to information arising from the data accumulating in the trial

• Adjustable key characteristics include:  randomization ratio, number of treatment groups, number and frequency of interim analyses, and the patient subpopulation being considered
• All possible changes must be pre-specified to achieve standard statistical operating characteristics, including controlling the trial’s false-positive rate
• Understanding the performance characteristics of an adaptive trial requires extensive numerical simulation under varying assumptions regarding true treatment effects

RATIONALE

• Knowledge regarding the relative effectiveness of the treatments involved accumulates over the course of a clinical trial, beginning with a state of equipoise and having high confidence near the end
• Fixed assignment, as occurs in a traditional RCT, ensures that this information is ignored and a fixed proportion of patients will receive a potentially inferior therapy (assuming there is a difference between the comparison therapies)
• Interim information available in a trial can be used to improve the outcomes of trial participants, especially those who enroll later in the trial thus increasing the probability that future trial participants are assigned to the study group with a better expected outcome
• Study outcomes may also be obtained more efficiently

PROS AND CONS

Advantages

• An adaptive clinical trial design can be used to increase the likelihood that study participants will benefit by being in a clinical trial (e.g. if there is a potential interaction between baseline characteristics and the treatment effect, different allocation ratios can be used for different patient subgroups)
• fewer patients needed
• avoids problems resulting from errors such as misestimates of the optimal dose, or overly optimistic or pessimistic estimates in placebo group primary outcome event rate
• more efficient
• less expensive

Disadvantages

• complexity of study design and trial conduct
• longer design phase (involving numerical simulations)
• concerns about the introduction of bias
• understanding of funding agencies and peer reviewers
• lack of experience and knowledge among trial stakeholders

Video

Roger Lewis provides an overview of Adaptive Trial Designs here:

References and Links

Journal articles

• Chow SC, Chang M. Adaptive design methods in clinical trials – a review. Orphanet J Rare Dis. 2008 May 2;3:11. PMC2422839.
• Kairalla JA, Coffey CS, Thomann MA, Muller KE. Adaptive trial designs: a review of barriers and opportunities. Trials. 2012 Aug 23;13:145. PMC3519822.
• Lee JJ, Chu CT. Bayesian clinical trials in action. Stat Med. 2012 Nov 10;31(25):2955-72. PMC3495977.
• Meurer WJ, Lewis RJ, Tagle D, Fetters MD, Legocki L, Berry S, Connor J, Durkalski V, Elm J, Zhao W, Frederiksen S, Silbergleit R, Palesch Y, Berry DA, Barsan WG. An overview of the adaptive designs accelerating promising trials into treatments (ADAPT-IT) project. Ann Emerg Med. 2012 Oct;60(4):451-7. PMC3557826.
• Meurer WJ, Lewis RJ, Berry DA. Adaptive clinical trials: a partial remedy for the therapeutic misconception? JAMA. 2012 Jun 13;307(22):2377-8. PMID: 22692168.

FOAM and web resources

• FET — Adaptive and Bayesian Trial Design in EM Research by Roger Lewis (2012)
• ICN — Adaptive Trial Designs (2014)