Confirmatory efficacy testing for individual dose-placebo comparisons using serial gatekeeping procedure in...

Area of Science:

• Clinical Trials
• Biostatistics
• Pharmacology

Background:

• Dose-finding trials are crucial for determining optimal drug doses for regulatory approval.
• Multiple Comparison Procedures-Modeling (MCP-Mod) is a standard approach in dose-finding.
• Existing extensions like closed MCP-Mod face challenges with model misspecification.

Purpose of the Study:

• To evaluate a novel serial gatekeeping procedure for confirmatory efficacy testing in dose-finding trials.
• To compare the performance of serial gatekeeping with closed MCP-Mod.
• To assess the ability of these methods to claim individual dose efficacy alongside dose-response signals.

Main Methods:

• Utilized serial gatekeeping procedures (fixed sequence, Holm, Hochberg, step-down Dunnett) to integrate MCP-Mod with pairwise dose-placebo comparisons.
• Controlled family-wise error rate in the strong sense.
• Conducted simulation studies to compare statistical power and sample size requirements.

Main Results:

• Serial gatekeeping demonstrated comparable statistical power to closed MCP-Mod in detecting at least one effective dose.
• Both methods efficiently pursued efficacy claims with minimal sample size increase (<20%) under monotonic dose-response assumptions.
• Serial gatekeeping offers advantages in implementation simplicity and interpretability.

Conclusions:

• Serial gatekeeping provides a robust and practical alternative for confirmatory efficacy testing in dose-finding trials.
• This approach facilitates simultaneous claims of dose-response signals and individual dose efficacy.
• Adopting serial gatekeeping can potentially accelerate drug development programs.