Evaluating response-adaptive randomization procedures for recurrent events and terminal event data using a composite...

Area of Science:

• Biostatistics
• Clinical Trials
• Epidemiology

Background:

• Recurrent and terminal events are common in clinical studies.
• Composite endpoints are useful for evaluating treatment efficacy, especially in costly, long-term studies.
• Joint frailty modeling is a standard approach for analyzing recurrent event data with a terminal event.

Purpose of the Study:

• To develop target-driven response adaptive randomization strategies for clinical trials.
• To utilize a composite endpoint based on joint frailty modeling for improved treatment allocation.
• To enhance ethical considerations in clinical trials by minimizing patient exposure to suboptimal treatments.

Main Methods:

• Implementation of a balanced randomized design as a baseline.
• Investigation of response adaptive randomization strategies.
• Application of joint frailty modeling for composite endpoints involving recurrent and terminal events.

Main Results:

• The proposed response adaptive randomization procedures effectively reduce the number of participants receiving inferior treatment.
• The strategies achieve desired optimal treatment targets compared to traditional balanced randomization.
• An R Shiny application is available for sample size calculation and allocation probabilities.

Conclusions:

• Response adaptive randomization with composite endpoints offers an ethical and efficient approach to clinical trial design.
• These methods are particularly beneficial for studies with recurrent and terminal events.
• The developed procedures provide a practical tool for optimizing treatment allocation in clinical research.