A dose-schedule finding design for phase I-II clinical trials

Area of Science:

• Clinical Trials
• Biostatistics
• Pharmacology

Background:

• Traditional dose-finding methods risk misidentifying optimal treatments.
• Treatment efficacy depends on both dose and administration schedule.
• Phase I/II trials require robust designs for dose-schedule optimization.

Purpose of the Study:

• To propose a novel Phase I/II clinical trial design for identifying optimal dose-schedule combinations.
• To develop a Bayesian dynamic model for joint dose and schedule effects.
• To create a dose-schedule-finding algorithm for sequential patient allocation.

Main Methods:

• Utilized a Bayesian dynamic model to jointly analyze dose and schedule effects.
• Developed a sequential allocation algorithm for patient enrollment.
• Applied the design to a Phase I/II trial of a gamma-secretase inhibitor in solid tumors.
• Conducted simulations to evaluate the design's operating characteristics.

Main Results:

• The proposed Bayesian model effectively borrows strength across dose-schedule combinations.
• The algorithm facilitates sequential patient allocation to promising combinations.
• Simulations demonstrated the design's ability to identify optimal dose-schedule combinations.
• The design was successfully applied to a real-world clinical trial scenario.

Conclusions:

• The proposed Phase I/II trial design offers an effective approach for optimizing dose-schedule combinations.
• This Bayesian method allows flexible modeling of dose and schedule interactions.
• The developed algorithm aids in efficient patient allocation and optimal combination selection for cancer therapies.