Optimizing Trial Designs for Targeted Therapies

Area of Science:

• Clinical trial design
• Decision theory
• Biostatistics

Background:

• Developing targeted therapies requires identifying patient populations with a positive benefit-risk balance.
• Pivotal clinical trials assess treatment efficacy in overall or pre-specified subpopulations.
• Optimizing trial design is crucial for efficient drug development.

Purpose of the Study:

• To derive optimized clinical trial designs using a decision theoretic framework.
• To maximize utility functions considering sample size, trial population, and statistical procedures.
• To incorporate prior knowledge of drug efficacy and model diverse utility perspectives.

Main Methods:

• Utilized a decision theoretic framework to maximize utility functions.
• Incorporated a two-dimensional prior distribution for prior efficacy knowledge.
• Modeled utility functions from both sponsor and public health perspectives.
• Employed numerical optimization to determine optimal trial designs.

Main Results:

• Developed a method for optimizing pivotal clinical trial designs.
• Demonstrated the ability to optimize sample size, population selection, and multiple testing procedures.
• Showcased optimized designs considering cost, expected benefit, and prior efficacy data.
• Presented examples of optimized designs from sponsor and public health viewpoints.

Conclusions:

• The decision theoretic framework provides an optimized approach to clinical trial design for targeted therapies.
• This method allows for the selection of appropriate patient populations and trial parameters to ensure a positive benefit-risk balance.
• Accounting for prior knowledge and diverse utility functions enhances the efficiency and relevance of trial designs.