Optimal promising zone designs.

Area of Science:

• Biostatistics
• Clinical Trial Design
• Statistical Methods

Background:

• Adaptive clinical trial designs, particularly those with sample size reassessment (SSR) based on interim analyses, are widely used.
• Promising zone designs specify a range for the interim test statistic to trigger sample size increases.
• Existing methods lack guidance on selecting promising zones and SSR rules, often relying on simulation for comparisons.

Purpose of the Study:

• To develop a practical framework for constructing promising zone designs in adaptive trials.
• To create intuitive designs that align with sponsor preferences.
• To achieve the gold-standard optimality defined by Jennison and Turnbull (2015).

Main Methods:

• Eliciting key preferences from trial sponsors.
• Constructing promising zone designs based on these elicited preferences.
• Utilizing an approach based on maximizing expected utility for design evaluation.

Main Results:

• Demonstration of a method to construct intuitive and optimal promising zone designs.
• Successful integration of sponsor preferences into the design process.
• Achieving the benchmark of gold-standard optimality for adaptive trial designs.

Conclusions:

• Sponsor preferences can guide the construction of effective promising zone designs.
• The proposed method offers a practical alternative to simulation-based comparisons.
• This approach enhances the intuitiveness and optimality of adaptive sample size reassessment strategies.