Predicting analysis time in events-driven clinical trials using accumulating time-to-event surrogate information

Area of Science:

• Biostatistics
• Clinical Trial Design
• Medical Oncology

Background:

• Precise prediction of event accrual is vital for interim and final analyses in time-to-event clinical trials.
• Overall survival (OS) is a common primary endpoint in oncology, necessitating accurate timing for analyses based on a set number of deaths.
• Correlated surrogate endpoints, such as time-to-progression (TTP), are often collected but not utilized for improving analysis time predictions.

Purpose of the Study:

• To develop a novel method for predicting clinical trial analysis timings by incorporating surrogate endpoint information.
• To enhance the precision of predicting event accrual for time-to-event endpoints, specifically overall survival (OS) in oncology.
• To propose a general parametric model for OS and TTP that accounts for the impact of disease progression on survival.

Main Methods:

• A general parametric model for overall survival (OS) and time-to-progression (TTP) is proposed, assuming disease progression influences OS.
• A Bayesian prediction procedure is developed to leverage surrogate information for improved analysis time predictions.
• Progression-free survival is handled separately as it can be derived from OS and TTP data.

Main Results:

• Simulations were conducted to evaluate the performance and accuracy of the proposed Bayesian prediction method.
• The study demonstrates the potential of borrowing strength from surrogate data to refine predictions of analysis timings.
• The proposed method offers a more precise approach to forecasting event accrual compared to existing techniques.

Conclusions:

• The developed Bayesian method effectively utilizes surrogate endpoint data (e.g., TTP) to improve the prediction of analysis timings for time-to-event endpoints like OS.
• This approach enhances the precision of event accrual prediction, which is critical for efficient clinical trial management and interpretation.
• The findings have significant implications for optimizing the design and execution of oncology clinical trials and potentially other therapeutic areas.