A two-stage adaptive clinical trial design with data-driven subgroup identification at interim analysis.

Area of Science:

• Biostatistics
• Clinical Trial Design
• Translational Medicine

Background:

• Randomized controlled trials (RCTs) are crucial for treatment efficacy assessment.
• Identifying patient subgroups with differential treatment effects can improve trial outcomes.
• Biomarker-driven subgroup identification is an emerging area in clinical research.

Purpose of the Study:

• To propose a two-stage adaptive clinical trial design for efficacy assessment using time-to-event outcomes.
• To develop methods for data-driven identification of efficacy subgroups based on biomarkers.
• To integrate subgroup selection with adaptive trial decision-making for futility, continuation, or sample size adjustment.

Main Methods:

• A two-stage adaptive design with interim and final analyses.
• Utilizing subgroup-finding algorithms to identify patient groups with enhanced treatment effects.
• Employing conditional power calculations for interim decision-making.
• Applying combination tests and closed testing procedures for final efficacy assessment.

Main Results:

• Simulation studies demonstrated the utility of the proposed adaptive design.
• The approach effectively identified subgroups with differential treatment effects.
• The data-driven subgroup selection improved efficacy assessment in clinical trials.

Conclusions:

• Incorporating data-driven subgroup selection into adaptive clinical trial designs offers significant benefits.
• This methodology enhances the ability to detect treatment effects in specific patient populations.
• The proposed adaptive design provides a flexible and efficient framework for modern clinical trials.