F.O.R.W.A.R.D: A Data-Driven Framework for Network-Based Target Prioritization in Drug Discovery.

Developing new medicines is expensive, slow, and often unpredictable—even with modern advances in artificial intelligence. To help change this, researchers created F.O.R.W.A.R.D (Framework for Outcome-based Research and Drug Development), a new platform that uses real clinical results and advanced machine-learning to identify the most promising drug targets. They tested F.O.R.W.A.R.D in one of the most difficult areas of medicine: Inflammatory Bowel Diseases (IBD), a chronic condition driven by many different biological pathways. The platform learns what successful treatment looks like at the molecular level using data from seven clinical trials of four different IBD drugs. It then analyzes how likely it is that targeting a specific molecule will shift a patient’s gene activity toward a “remission-like” state. When they compared F.O.R.W.A.R.D’s predictions with the outcomes of 210 completed clinical trials covering 52 drug targets, it correctly identified which targets would succeed every time—even though those trials varied widely in their design and drug mechanisms. Additional testing using single-cell sequencing and patient-derived organoids showed that the remission signal arises from the gut epithelium and is linked to poor disease outcomes when absent. F.O.R.W.A.R.D makes it possible to run in-silico (“virtual”) phase-zero trials, helping researchers design better studies, prioritize the most promising targets, salvage drugs that might otherwise be abandoned, and decide early when a trial is unlikely to succeed. Because it can continuously improve as new clinical data emerge, F.O.R.W.A.R.D offers a powerful new way to guide drug discovery—bringing greater foresight to research and supporting smarter, more human-centered decision-making in medicine.