Logical Network Perturbations Guide Precise Target Prioritization for Drug Development

Despite advances in artificial intelligence (AI), target-based drug development remains costly, complex, and imprecise. We introduce F.O.R.W.A.R.D [Framework for Outcome-based Research and Drug Development], a network-based target prioritization platform, and demonstrate its utility in the challenging landscape of Inflammatory Bowel Diseases (IBD), a chronic, multifactorial condition. F.O.R.W.A.R.D uses real-world clinical outcomes, and a machine-learning classifier trained on transcriptomic data from seven prospective randomized trials across four drugs. It defines remission at the molecular level and calculates, using network connectivity, the likelihood that targeting a given molecule will induce remission-associated gene expression. Benchmarking against 210 completed trials across 52 targets, F.O.R.W.A.R.D achieved 100% predictive accuracy--despite variability in drug mechanisms and trial designs. Single-cell RNA-seq and a prospective biobank of patient-derived organoids confirmed that the remission signature is epithelium-specific and tracks with poor outcomes. F.O.R.W.A.R.D enables in-silico phase zero trials to inform trial design, revive shelved drugs, and guide early termination decisions. Broadly applicable and iteratively refined by emerging trial data, F.O.R.W.A.R.D has the potential to reshape drug discovery--bringing foresight to hindsight, and empowering both R&D and human-in-the-loop clinical decision-making.