A federal health agency is betting that simulated, always-on clinical trials can shave years off drug development.
The Department of Health and Human Services unveiled a five-year program called SURPASS, short for Simulation-augmented, Real-time Platform Adaptive Seamless Trials. It will run through the Advanced Research Projects Agency for Health, or ARPA-H.
The premise is to replace the rigid three-phase trial model with designs that learn as data arrives. Organizers say the goal is to evaluate drugs faster, at lower cost, and with fewer participants, while steering more patients into studies likely to help them.
SURPASS has three workstreams. The first builds a phaseless design engine that folds digital twins and predictive models into trial planning, simulating outcomes before a study opens. The second develops a continuous inference engine meant to shrink the need for large control groups without weakening the evidence. The third creates an agentic operations layer to automate trial startup, data collection, and monitoring.
ARPA-H paired SURPASS with companion projects. STACK aims to convert research-naive clinics into trial-ready sites. COMMONS is building a national consent framework for sharing data across the trial ecosystem.
Program manager Daria Fedyukina said the current system carries too many stops and too much duplicated effort. She said the new frameworks should let teams learn in real time, adapt when needed, and generate stronger evidence faster.
HHS did not say how much money backs the effort. Officials have framed the initiative as a bid to keep the United States competitive with China in biomedical research.
