Leveraging real-world data to conduct externally controlled trial for rare diseases with count-type endpoints:...

Area of Science:

• Clinical Trials Methodology
• Real-World Data Analysis
• Rare Disease Research

Background:

• Randomized controlled trials for rare diseases face challenges due to small patient populations and ethical concerns.
• Real-world data (RWD) offers a viable alternative for generating supporting evidence when traditional trials are infeasible.
• External control arms (ECAs) constructed from RWD are occasionally used for single-arm trials, but patient eligibility at multiple time points complicates their design.

Purpose of the Study:

• To propose a novel externally controlled trial design that accommodates multiple entry points for patients from RWD.
• To recommend statistical methods for handling confounders, small sample sizes, within-subject correlation, and count data overdispersion.
• To present a blinding process suitable for this novel trial design.

Main Methods:

• Development of a new externally controlled trial design allowing multiple entries per patient from RWD.
• Formulation of statistical methods addressing measured confounders, limited sample size, within-subject correlation, and overdispersion.
• Simulation studies based on a rare disease case to evaluate design and method performance against single-entry approaches.

Main Results:

• The proposed design with multiple RWD entries demonstrated controlled type I error rates.
• The methods showed robustness against certain model misspecifications.
• A moderate improvement in statistical power was observed compared to selecting a single entry point per ECA patient.

Conclusions:

• The novel externally controlled trial design with multiple RWD entries enhances trial performance.
• This approach provides a reliable and efficient method for rare disease drug development.
• The proposed statistical methods and blinding process support the implementation of this design.