Insilico starts Phase III trial for AI-discovered lung-fibrosis drug

Rentosertib's path from discovery to late-stage clinical trial illustrates how AI-driven target identification and chemistry design can accelerate drug development for age-related diseases. Insilico used its PandaOmics platform to integrate multi-omics data from fibrotic tissues, biological network analysis, causal inference, and pathway analysis to rank TNIK as a top-priority fibrosis target. The company then applied Chemistry42 to generate and optimize the small molecule, embedding this workflow in a framework that connects hallmarks of aging—fibrosis, chronic inflammation, extracellular matrix remodeling, and cellular senescence—to disease mechanisms. Earlier work published in peer-reviewed journals (Nature Biotechnology for discovery-to-clinic rationale and Nature Medicine for Phase IIa results) provided the scientific foundation for this advancement.

IPF remains a severe clinical challenge with limited disease-modifying options: median survival after diagnosis is approximately two to four years, and existing antifibrotic drugs can only slow progression. The Phase IIa dose-dependent efficacy signal—most notably the +98.4 mL improvement in forced vital capacity at the 60 mg dose—has generated interest from research institutions globally. The Phase III trial, enrolling 320 patients over 52 weeks, will now test whether this early signal translates to clinically meaningful outcomes in a larger, longer-term setting under rigorous double-blind, placebo-controlled conditions led by specialists at leading Chinese respiratory medicine institutions.