Propensity score-incorporated adaptive design approaches when incorporating real-world data

Area of Science:

• Biostatistics
• Clinical Trial Design
• Real-World Data Analysis

Background:

• Real-world data (RWD) can augment prospective clinical studies.
• The propensity score-integrated composite likelihood (PSCL) method integrates RWD by forming strata based on propensity scores (PS) and down-weighting RWD information.
• PSCL was initially developed for fixed study designs.

Purpose of the Study:

• To extend the PSCL method to an adaptive design framework.
• To enable early success claims or sample size re-estimation in clinical trials using RWD.
• To propose a general strategy for implementing these adaptive features within PSCL.

Main Methods:

• Extension of the PSCL method to an adaptive design.
• Utilization of Fisher's combination test for early success claims.
• Application of the Cui, Hung, and Wang test for sample size re-estimation.

Main Results:

• A general strategy for adaptive PSCL is proposed.
• Demonstration of early success claim procedures.
• Demonstration of sample size re-estimation procedures.

Conclusions:

• The proposed adaptive PSCL strategy effectively integrates RWD for enhanced clinical trial design.
• The methods allow for flexibility in trial conduct, including early stopping or sample size adjustments.
• The approach provides a robust framework for leveraging RWD in adaptive clinical trials.