Enhancing External Control Arm Analyses through Data Calibration and Hybrid Designs

Abstract

Almost all external control arm analyses to contextualize findings of a single arm trial struggle with two key issues: the lack of baseline randomization, and equally important, the difference in data collection between the experimental arm with its primary data collection, and the external control arm using secondary data. We illustrate the data calibration design to remedy issues arising from differential measurements in the two arms, and discuss the hybrid design that expands an underpowered randomized internal control arm with real‐world data to mitigate the lack of randomization of the external control arm. We show how the two approaches fit into an evidence‐development strategy that naturally builds on the incremental insights gained.