Optimal adaptive group sequential design with flexible timing of sample size determination

Area of Science:

• Biostatistics
• Clinical Trial Design
• Statistical Power

Background:

• Fixed sample size designs can lack statistical power and efficiency.
• Flexible sample size designs, such as group sequential and sample size re-estimation, offer improvements.

Purpose of the Study:

• Introduce a novel representation of sample size re-estimation as a generalized adaptive group sequential design.
• Provide a flexible framework for sample size determination in clinical studies.
• Clarify existing methodologies and address common confusions.

Main Methods:

• Developed a generalized adaptive group sequential design framework.
• Demonstrated equivalence to Cui et al.'s sample size re-estimation design.
• Illustrated design optimization and implementation with an example.

Main Results:

• The generalized design allows flexible, mid-course sample size determination.
• This approach expands the design parameter space for optimization.
• Equivalence to existing re-estimation designs provides enhanced understanding.

Conclusions:

• The proposed generalized adaptive group sequential design offers a robust and efficient alternative to fixed sample size trials.
• This flexible design enhances statistical power and allows for optimization.
• The study clarifies sample size re-estimation concepts and aids practical implementation.