Keywords: Multiple comparisons, Clinical trials, Gatekeeping procedures
This presentation will focus on multiplicity issues arising in confirmatory Phase III clinical trials with "multi-dimensional" multiplicity problems. Case studies will be presented to define common multi-dimensional problems in trials utilizing multiple endpoints and doses. In order to protect the overall Type I error rate in complex multiplicity problems of this kind, multiple testing procedures that account for the hierarchical structure of the endpoint and dose-control tests, known as gatekeeping procedures, are used. This presentation will discuss key methods for building gatekeeping procedures and considerations arising in their application, e.g., use of trial-specific information and criteria for selecting most relevant and powerful gatekeeping procedures. The methods and principles discussed in this presentation will be illustrated using a Phase III trial with multiple clinical objectives.