Presentation (CTI)

OBJECTIVES: The adoption of external control arms (ECAs) derived from real-world data (RWD) has gained traction as an alternative to traditional randomized controlled trials (RCTs), particularly in rare diseases, oncology, and other settings where randomization is infeasible or unethical. While regulators and HTA bodies have shown growing interest in ECAs, questions remain about their ability to reliably "control" for bias, confounding, and clinical heterogeneity.
METHODS: This abstract presents a critical assessment of the current landscape of ECA use, drawing from recent case studies, regulatory submissions, and published literature. We explore the methodological rigor of contemporary ECA approaches—ranging from historical comparator cohorts to synthetic controls and matched real-world cohorts—and evaluate their performance across key domains: selection bias, data source quality, endpoint alignment, and statistical methodology (e.g., propensity score matching, inverse probability weighting).
RESULTS: Our analysis reveals that while ECAs can approximate treatment effects under certain conditions, they are often overburdened with expectations they are not yet methodologically mature enough to fulfill. Issues such as data missingness, limited granularity, and temporal drift frequently undermine validity. Moreover, we find a gap between technical feasibility and decision-maker confidence, particularly in high-stakes regulatory or reimbursement contexts.
CONCLUSIONS: We conclude by offering a framework to assess "fit-for-purpose" use of ECAs and call for greater transparency, methodological standardization, and contextual nuance in their deployment. Target trial framework appears to be the best fit among the available frameworks. As ECAs continue to evolve, stakeholders must ask not just can we use them—but should we, and how do we define meaningful control?