A mixture gatekeeping procedure based on the Hommel test for clinical trial applications

Area of Science:

• Biostatistics
• Clinical Trial Design
• Statistical Methodology

Background:

• Clinical trials often involve hierarchically structured objectives requiring multiplicity adjustment.
• Controlling the familywise error rate (FWER) in the strong sense is crucial for valid trial outcomes.
• Existing methods may lack power or fail to account for logical dependencies between hypotheses.

Purpose of the Study:

• To propose a novel gatekeeping procedure for clinical trials with hierarchical objectives.
• To enhance statistical power in multiplicity adjustment compared to existing methods.
• To provide a flexible procedure applicable to complex clinical trial designs.

Main Methods:

• Development of a gatekeeping procedure based on the Hommel (1988) test.
• Detailed description of the procedure's application to various clinical trial designs.
• Illustrative examples using two distinct clinical trial scenarios.

Main Results:

• The proposed Hommel-based gatekeeping procedure offers improved statistical power.
• The method effectively controls the familywise error rate in the strong sense.
• The procedure is adaptable to complex hierarchical trial structures.

Conclusions:

• The Hommel-based gatekeeping procedure is a powerful and valid tool for multiplicity adjustment in clinical trials.
• This methodology enhances the efficiency of clinical trials with ordered objectives.
• The proposed procedure offers practical advantages for designing and analyzing complex clinical trials.