Leveraging External Data in Rare Disease Trials Through Individualized Discounting Within the Power Prior: A Case Study in Hereditary Angioedema

• 成果类型: Article; Early Access
• 署名作者: Zhu, Claire R.; Alt, Ethan M.; Ibrahim, Joseph G.
• 署名单位: University of North Carolina; University of North Carolina Chapel Hill; University of North Carolina School of Medicine
• 刊物名称: JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
• ISSN/ISSBN: 0162-1459; 1537-274X
• DOI: 10.1080/01621459.2026.2653257
• 发表日期: 2026-05-31
• 关键词: Bayesian statistics; clinical trials; historical data; power prior; Propensity scores; rare disease; propensity-score; clinical-trials; prior distributions; INFORMATION

摘要

An increasingly common approach in clinical trials involves leveraging external data to supplement the sample size of randomized controlled trials. Although designing studies with concurrent data allows direct treatment comparisons, recruiting an adequate number of participants, especially in the context of rare diseases, can be challenging, time-intensive, and financially burdensome. Incorporating external data offers a cost-effective solution to enhance trial data; however, potential non-exchangeability between external and current data introduces significant challenges. To address this, we propose the Discounting Individuals Power Prior (DIPP), a novel extension of the power prior methodology. The DIPP introduces an individual-specific local discounting parameter alongside a global discounting parameter. It employs a spike-and-slab type prior, where the probability is tied to the assumed degree of exchangeability between external and concurrent trial data, with this degree treated as a random variable. We evaluate the DIPP's performance through comparisons with the power prior, the normalized power prior, and a propensity score-based approach in simulation studies. Furthermore, we demonstrate its utility in a detailed case study of real-world clinical trial data for an inherited disorder. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.