Hit to development candidate in 10 months: Rapid discovery of SGR-1505, a novel, potent MALT1 inhibitor

MALT1 is a validated target for lymphocyte regulation disorders, but second-generation allosteric MALT1 inhibitors faced permeability, efflux, and solubility challenges while needing to maintain on-target potency for use in relapsed/refractory B-cell malignancy combination therapy. The team combined physics-based modeling and machine learning: WaterMap analysis of displaceable binding-site waters, De Novo Design evaluating 8.2 billion compounds computationally, FEP+ modeling of over 1,700 molecules in the first three months, and multiparameter optimization scoring over 5,000 ideas, all coordinated via LiveDesign. This yielded 78 synthesized compounds in the lead series (129 program-wide) and identified development candidate SGR-1505 within 10 months; Phase 1 results showed a favorable safety profile with encouraging preliminary efficacy, including monotherapy responses in chronic lymphocytic leukemia and Waldenström macroglobulinemia patients.

The customer is anonymised in Schrödinger’s own case study — not withheld by us.

Schrödinger is a physics-plus-machine-learning computational platform for molecular design; all top-20 pharma companies by revenue are customers of its drug-discovery software.

This sits in the Build stage of the AI use-case lifecycle: Named examples from the foundation-model labs and platforms pharma teams use to actually construct, train and generate the model or application.