A multi-agent molecular optimization framework leads to a rapid-recovery intravenous anesthetic candidate with an improved safety margin

Abstract

Lead optimization, the systematic refinement of therapeutic compounds through iterative structural modification, faces a dual challenge in modern drug discovery: navigating astronomically vast molecular design spaces while balancing conflicting demands on potency, pharmacokinetics, and safety. We present MASCOT (Multi-Agent SearCh for molecular OpTimization), a role-specialized multi-agent framework for molecular optimization. Integrated with a chemically constrained graph-editing search, MASCOT coordinates three specialized agents: a trade-off agent that reprioritizes competing objectives, a strategy agent that adapts how molecular edits are proposed, and a reflection agent that distills lessons from previous decisions. Computational experiments showed that MASCOT achieved the best performance over competing methods on six benchmark settings. On the SARS-CoV-2 main protease task, its mean docking-score improvement was 3.6 times that of the strongest baseline. Applied to the clinically used anesthetic remimazolam (RM), MASCOT prioritized RM-1, which showed a shorter liver microsomal half-life, higher brain exposure, and a larger therapeutic index than RM. Subsequent derivative design yielded RM-7. Extensive animal studies established RM-7 as a rapid-recovery intravenous anesthetic candidate with greater potency, faster functional recovery, a wider safety margin, and preserved flumazenil reversibility. These results demonstrate that multi-agent coordination can link adaptive molecular search to medicinal chemistry and experimental pharmacology. Significance Statement Lead optimization requires balancing potency, safety, pharmacokinetics, and ease of synthesis, because improving one property often compromises another. Managing these trade-offs while searching for better molecules is a central bottleneck in drug design. We show that a team of specialized AI agents can steer this search. Rather than generating final structures, the agents decide which goals to prioritize, how molecular edits should be proposed, and what should be learned from earlier attempts, while chemical rules ensure that every step remains valid. Applied to an approved intravenous anesthetic, this approach led to a candidate with faster recovery from anesthesia and a wider safety margin in animals. This study shows how coordinated AI agents can support practical drug discovery.