🤖 AI Summary
In distributed healthcare settings, stringent privacy regulations prevent centralized aggregation of multisite clinical trial data, impeding construction of external control arms (ECAs) and thus hindering causal inference for time-to-event outcomes. To address this, we propose the first federated learning framework integrating inverse probability weighting (IPTW) for distributed causal inference. Our method enables collaborative estimation of Cox proportional hazards model parameters without sharing raw patient-level data, and introduces a distributed covariance calibration technique to improve treatment effect estimation accuracy. Evaluated on real-world comparative drug studies, the approach identified a statistically significant treatment effect (HR = 0.62, 95% CI [0.45, 0.85]) overlooked by conventional centralized analysis. This demonstrates both methodological validity and regulatory compliance, offering a scalable, privacy-preserving technical pathway for early-phase clinical development and regulatory decision-making under data governance constraints.
📝 Abstract
External control arms (ECA) can inform the early clinical development of experimental drugs and provide efficacy evidence for regulatory approval. However, the main challenge in implementing ECA lies in accessing real-world or historical clinical trials data. Indeed, regulations protecting patients' rights by strictly controlling data processing make pooling data from multiple sources in a central server often difficult. To address these limitations, we develop a new method, 'FedECA' that leverages federated learning (FL) to enable inverse probability of treatment weighting (IPTW) for time-to-event outcomes on separate cohorts without needing to pool data. To showcase the potential of FedECA, we apply it in different settings of increasing complexity culminating with a real-world use-case in which FedECA provides evidence for a differential effect between two drugs that would have otherwise gone unnoticed. By sharing our code, we hope FedECA will foster the creation of federated research networks and thus accelerate drug development.
