Insilico’s AI-designed IPF drug rentosertib enters Phase III

Insilico Medicine has advanced rentosertib, an AI-designed oral TNIK inhibitor for idiopathic pulmonary fibrosis (IPF), into Phase III testing. A randomized Phase IIa trial in China enrolled 71 patients at 22 sites and treated participants for 12 weeks with daily 30 mg or 60 mg doses. Patients on the 60 mg dose had a mean forced vital capacity (FVC) increase of 98.4 mL, while the placebo group had a mean loss of 20.3 mL. Reported adverse events were described as manageable and occurred at rates comparable to baseline across study arms. The U.S. Food and Drug Administration granted the drug Orphan Drug Designation in February 2023.

Insilico used its Pharma.AI platform for the discovery process. The PandaOmics engine analyzed genomics, clinical data, literature and patent information to build network models and apply causal inference, and it prioritized TRAF2- and NCK-interacting kinase (TNIK) as a target linked to fibrosis and inflammation. The company links TNIK activity to signaling pathways that include Wnt, TGF-β, Hippo/YAP-TAZ, JNK and NF-κB, and designed rentosertib to inhibit TNIK with the intention of affecting those pathways.

After target selection, Insilico’s Chemistry42 engine used generative molecular design driven by reinforcement learning to create candidate molecules that fit the TNIK pocket and met pharmacological property constraints. The team synthesized 79 compounds and advanced the 55th candidate into preclinical testing. Insilico reports the process from project start to preclinical candidate nomination took about 18 months.

The clinical program includes proteomic analyses aimed at testing the biological effects predicted by the algorithms. Investigators applied internal proteomic aging clocks, including ProtAge, OrganAgechrono, ipfP3GPT and PAOPAC, and mortality-risk-related clocks such as PAC and OrganAgemortality. The study also used SenMayo and CellAge signatures to assess cellular senescence and compared protein changes against population trajectories from the UK Biobank.

Peer-reviewed publications document stages of the program, covering the discovery-to-clinic workflow, preclinical efficacy and Phase I pharmacokinetics, structural validation of the TNIK inhibitor chemotypes with a kinase-domain co-crystal, and the Phase IIa safety and lung-function data. Independent research has reported that TNIK inhibition produces senomorphic effects and reduces markers of extracellular matrix remodeling.

Feng Ren, PhD, co-CEO and chief scientific officer at Insilico, said: “IPF is one of the clearest clinical examples of an age-related disease in which fibrosis, chronic inflammation, extracellular matrix remodeling, and cellular senescence intersect.” Alex Zhavoronkov, PhD, founder and CEO, described rentosertib as a program that spans AI-driven target discovery, generative chemistry and clinical testing, and noted the project moved from computational prediction to human trials.

Phase III will gather larger-scale efficacy data to determine whether the lung-function gains seen with the 60 mg dose correspond to clinically meaningful benefits for patients with IPF.