Response adaptive randomization procedures in seamless phase II/III clinical trials.

Area of Science:

• Clinical Trial Design
• Biostatistics
• Pharmaceutical Development

Background:

• Evaluating new therapies requires efficient, ethical, and cost-effective methods without compromising data integrity.
• Seamless phase II/III clinical trials offer advantages but can be further optimized.
• Response adaptive randomization (RAR) can improve trial efficiency, but frequentist approaches for seamless trials face statistical inference challenges.

Purpose of the Study:

• To propose a framework for frequentist response adaptive randomization (RAR) designs in seamless phase II/III clinical trials.
• To address difficulties in valid statistical inference and type I error rate control within these trial designs.
• To enhance the efficiency, ethical considerations, and economic benefits of seamless phase II/III trials.

Main Methods:

• Development of a family of frequentist RAR designs tailored for seamless phase II/III trials.
• Derivation of the asymptotic distribution of parameter estimators utilizing martingale processes.
• Implementation of solutions to effectively control the type I error rate.

Main Results:

• The proposed framework provides a viable approach for frequentist RAR in seamless phase II/III trials.
• Theoretical findings regarding asymptotic distributions and type I error control are established.
• Numerical studies validate the theoretical results and demonstrate the advantages of the proposed methods.

Conclusions:

• The developed frequentist RAR designs offer a robust solution for seamless phase II/III clinical trials.
• The methods facilitate valid statistical inference and control of the type I error rate.
• This approach enhances the efficiency and ethical conduct of clinical trial development.