A Bayesian phase I/II platform design for co-developing drug combination therapies for multiple indications.

Area of Science:

• Oncology
• Clinical Trial Design
• Biostatistics

Background:

• Combination therapies involving targeted or immunotherapy agents are increasingly used in cancer treatment.
• Developing these combination therapies requires efficient and flexible clinical trial designs.

Purpose of the Study:

• To propose a master-protocol-based, Bayesian phase I/II platform design for co-developing combination therapies (BPCC) across multiple cancer indications.
• To enhance efficiency in dose-finding and decision-making for combination therapies.

Main Methods:

• Utilizing a single master protocol to evaluate combination drugs concurrently or sequentially across indications.
• Employing Bayesian hierarchical models to jointly model dose-toxicity and dose-efficacy relationships, enabling information sharing across indications.
• Incorporating a utility function to quantify risk-benefit tradeoffs, especially for agents with non-monotonic dose-efficacy profiles.

Main Results:

• The BPCC design allows for efficient indication-specific decision-making by borrowing strength across indications.
• Interim analyses update toxicity, efficacy, and utility models, informing dose escalation/de-escalation and optimal dose identification.
• Simulation studies demonstrate desirable operating characteristics and efficiency gains.

Conclusions:

• The proposed Bayesian platform design (BPCC) offers an efficient strategy for accelerating the co-development of novel combination cancer therapies.
• This approach facilitates robust decision-making by integrating dose-toxicity and dose-efficacy data across multiple indications.