Methods and Applications of Statistics in Clinical Trials, Volume 1: Concepts, Principles, Trials, and Designs

Chapter 37

Group Sequential Designs

Gernot Wassmer

37.1 Introduction

In clinical research there is a great interest in interim analyses for ethical, economical, and organizational or administrative reasons. Interim analyses are performed within sequential designs that offer the possibility to stop a trial early with a statistically significant test result. This trial will possibly need less patients than the trial with a fixed sample size where a test decision can be made only at the end of the trial. A therapy which was shown to be superior can be applied earlier and the inferior therapy can be replaced by the better one. Furthermore, in interim analyses the quality of the performance of the trial can be assessed and possibly be improved when necessary. Interim analyses are also carried out for the assessment of the safety of the treatments under consideration, and the observation of serious adverse events can lead to an early stopping of the trial. The appointment of an Independent Data and Safety Monitoring Board (DSMB) in a clinical trial is a recommended constitution to perform these issues according to generally accepted standards of Good Clinical Practice (GCP) [1–3]. Another issue is the redesign of the trial (e.g., sample size reassessment), but it was not intended originally nor is it generally possible with “classical” group sequential designs (see Flexible Design, Interim Analysis, Stopping Boundaries).

Sequential designs were originally developed for the analysis of ...