Methodology

Adaptive Methods: When and How Should They be Used in Clinical Trials?

¹ Université Paris Diderot, Hôpital Saint-Louis, Paris, France
² Novartis Pharma S.A.S., Rueil Malmaison, France
³ Institut Pasteur, Paris, France

Correspondence and offprints: Brigitte Lecocq, Laboratoire Novartis, 2-4 rue Lionel Terray, 92506 Rueil Malmaison Cedex, France. E-mail: brigitte.lecocq@novartis.com

Received: 5 May 2011
Accepted: 16 May 2011

Abstract

Adaptive clinical trial designs are defined as designs that use data cumulated during trial to possibly modify certain aspects without compromising the validity and integrity of the said trial. Compared to more traditional trials, in theory, adaptive designs allow the same information to be generated but in a more efficient manner. The advantages and limits of this type of design together with the weight of the constraints, in particular of a logistic nature, that their use implies, differ depending on whether the trial is exploratory or confirmatory with a view to registration. One of the key elements ensuring trial integrity is the involvement of an independent committee to determine adaptations in terms of experimental design during the study. Adaptive methods for clinical trials are appealing and may be accepted by the relevant authorities. However, the constraints that they impose must be determined well in advance.