Augmenting Treatment Arms With External Data Through Propensity-Score Weighted Power Priors: An Application in...

Area of Science:

• Biostatistics
• Clinical Trial Methodology
• Real-World Evidence

Background:

• Real-world data (RWD) integration in clinical trials necessitates advanced statistical methods to manage confounding.
• Existing hybrid methods often focus on augmenting control arms with historical data.
• Augmenting treatment arms using expanded access data presents unique challenges for evidence synthesis.

Purpose of the Study:

• To develop and validate a novel statistical method, ProPP, for integrating RWD from expanded access programs into clinical trial analysis.
• To address confounding in RWD by combining propensity score weighting and Bayesian dynamic borrowing.
• To improve the precision and reliability of evidence synthesis for investigational medicines.

Main Methods:

• Developed the ProPP (Propensity score and Power prior) method, combining propensity score weighting and modified power prior.
• Propensity score weighting estimates average treatment effect with constraints on external patient weights.
• Utilized Bayesian dynamic borrowing to address unmeasured confounding.

Main Results:

• The ProPP method demonstrated favorable performance in simulations compared to existing hybrid methods regarding precision and type I error rate.
• The method was successfully illustrated using individual patient data from a vemurafenib trial and expanded access program for metastatic melanoma.
• ProPP offers a double safeguard against prior-data conflict in evidence synthesis.

Conclusions:

• The ProPP method provides a conceptually simple and user-friendly approach for integrating expanded access data into clinical trial evidence.
• This method represents a valuable addition to existing evidence synthesis techniques for combining trial and RWD.
• ProPP enhances decision-making by robustly incorporating real-world data into the evaluation of investigational medicines.