Futility Analyses for the MCP-Mod Methodology Based on Longitudinal Models

Area of Science:

• Biostatistics
• Clinical Trial Design
• Pharmacometrics

Background:

• Futility analyses are crucial for efficient clinical trial management.
• The MCP-Mod methodology requires robust interim analysis tools.
• Utilizing longitudinal data can improve decision accuracy.

Purpose of the Study:

• To derive formulas for predictive and conditional power within the MCP-Mod framework.
• To evaluate the performance of futility analysis using longitudinal models versus completer-only models.
• To demonstrate the application of the proposed methodology in a real-world dose-finding study.

Main Methods:

• Derivation of formulas for predictive and conditional power for MCP-Mod.
• Simulation studies to assess repeated sampling properties of decision rules.
• Comparison of longitudinal and completer-only models for interim decision-making.

Main Results:

• The proposed futility analysis methods for MCP-Mod perform adequately.
• Longitudinal analysis demonstrates superior performance over completer-only analysis.
• The benefits of longitudinal analysis are pronounced with higher recruitment speed and data correlation.

Conclusions:

• The developed methodology provides a reliable approach for futility analyses in MCP-Mod.
• Longitudinal data utilization in interim analyses enhances decision-making accuracy in dose-finding studies.
• The findings support the adoption of longitudinal models for more efficient clinical trial conduct.