Analytical Evaluation of the 2-in-1 Adaptive Design for Binary Endpoints

Area of Science:

• Biostatistics
• Clinical Trial Design
• Drug Development

Background:

• Accelerating drug development is crucial for unmet medical needs, often involving skipping Phase 2 trials.
• The 2-in-1 adaptive design allows trials to adjust size based on interim analysis.
• Existing statistical methods for 2-in-1 designs primarily focus on continuous or time-to-event endpoints, neglecting binary outcomes.

Purpose of the Study:

• To propose novel formulas for evaluating type I error rate and power in 2-in-1 adaptive designs with binary endpoints.
• To provide methods that do not rely on normal approximation for binary data.
• To assess the generalizability of existing 2-in-1 design operating characteristics to binary endpoints.

Main Methods:

• Development of exact formulas for type I error rate and power calculations for 2-in-1 adaptive designs with binary endpoints.
• Avoidance of Monte Carlo simulations for precise operating characteristic evaluation.
• Application of proposed formulas to a clinical trial for pyruvate kinase deficiency.

Main Results:

• The proposed formulas accurately evaluate type I error rate and power for 2-in-1 adaptive designs with binary endpoints.
• Numerical investigations confirm that operating characteristics are comparable to those for continuous or time-to-event endpoints.
• The formulas are applicable to any statistical test used for binary endpoints.

Conclusions:

• The developed formulas offer an exact and efficient method for analyzing 2-in-1 adaptive designs with binary endpoints.
• These findings address a significant gap in statistical methodology for adaptive clinical trial designs.
• The study demonstrates the utility of the proposed methods in a real-world clinical trial setting.