An enhanced exact permutation rank-based inferential seamless phase 2/3 design.

Area of Science:

• Clinical Trials
• Biostatistics
• Pharmacometrics

Background:

• Seamless phase 2/3 adaptive designs combine dose optimization and confirmatory analysis, accelerating drug development.
• Challenges include type I error inflation and unknown correlations between short-term biomarkers and long-term efficacy.
• Existing multiplicity adjustment methods can be conservative, leading to reduced statistical power.

Purpose of the Study:

• To propose an efficient permutation rank-based method for seamless phase 2/3 adaptive designs.
• To address type I error inflation and power loss associated with dose selection in adaptive trials.
• To preserve the correlation structure between biomarkers and efficacy endpoints without explicit estimation.

Main Methods:

• A novel permutation rank-based method is proposed.
• The method preserves the correlation structure between biomarkers and efficacy endpoints.
• Simulations and an oncology case study were used for validation.

Main Results:

• The permutation rank-based method robustly controls family-wise error rates.
• The proposed method achieves uniformly higher statistical power compared to existing methods.
• The method demonstrated efficacy in an oncology trial using progression-free survival.

Conclusions:

• The permutation rank-based method offers an efficient solution for seamless phase 2/3 adaptive designs.
• This approach enhances statistical power while maintaining type I error control.
• The method provides a valuable tool for optimizing drug development in clinical trials.