Using a Combination of Early and Primary Endpoint Data in Group-Sequential and Interrupted Clinical Trials

• Activity Number: 135 - The Use of Baseline Covariates and Early Read-Outs in Adaptive and Group Sequential Designs: Worth the Effort?
• Type: Topic Contributed
• Date/Time: Monday, August 8, 2022 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Nigel Stallard*
• Companies: University of Warwick
• Keywords: Adaptive trial design; Surrogate endpoints; Missing data; Coronavirus pandemic

Abstract

In a group-sequential trial interim analyses are used to decide whether to stop the trial early, or to modify the trial in some way, based on inference on treatment effects based on the data already observed. Often the primary endpoint can be observed only after long-term follow-up, so that at the time of an interim analysis primary endpoint data are available for only a relatively small proportion of the patients randomised. Data on some short-term endpoint may, however, be available, both for these patients and for others for whom the long-term primary endpoint has not yet been observed. This talk will describe methods that enable these short-term data to be used in combination with the long-term primary endpoint data to draw inference on primary endpoint treatment effects. This leads to increased precision in estimation of the treatment effects, and hence to improved decision-making. Analysis of a combination of short-term and long-term data similar to that seen at an interim analysis could also arise if a clinical trial has been interrupted, for example due to the SARS-CoV2 pandemic. The application of similar methods is this setting will also be discussed.