Insilico Takes AI-Discovered IPF Drug Into Phase III Without Approval Dates

Insilico Medicine is moving its AI-discovered drug rentosertib into Phase III trials for idiopathic pulmonary fibrosis (IPF), after a randomised Phase IIa study showed a measurable improvement in lung function.

In the 60 mg once-daily group, patients recorded a mean forced vital capacity gain of +98.4 mL over 12 weeks, compared with a 20.3 mL loss in the placebo group.

The oral drug targets TNIK, or the TRAF2- and NCK-interacting kinase, in an effort to address the fibrosis and inflammation underlying IPF.

IPF causes severe scarring of lung tissue, and patients typically have a median survival of two to four years after diagnosis.

The trial enrolled 71 patients across 22 Chinese clinical sites.

Investigators tested 30 mg and 60 mg daily doses against placebo, while safety profiles remained manageable and adverse events stayed within expected baseline rates across the trial arms.

The U.S. Food and Drug Administration granted the asset Orphan Drug Designation in February 2023.

How the AI system selected the target

Insilico’s Pharma.AI platform began with PandaOmics, which processed genomics, clinical-trial outcomes, academic literature and patent intelligence to build biological network models.

Causal-inference methods identified TNIK as a central node in IPF-related fibrosis and inflammation, including activity across Wnt, TGF-β, Hippo/YAP-TAZ, JNK and NF-κB signalling pathways.

The approach also used a hallmarks-of-aging framework, scoring targets against aging mechanisms, chronic inflammation and extracellular-matrix remodelling.

Feng Ren, Insilico’s co-CEO and chief scientific officer, said rentosertib emerged from a biology-first, ageing-informed workflow rather than from screening a conventional target against a larger compound library.

Chemistry42 then used Generative Tensorial Reinforcement Learning to design molecules for the target protein pocket while balancing structural fit with pharmacological requirements.

The system synthesised exactly 79 physical molecules, and the team selected the 55th iteration for preclinical testing.

Insilico said the process reduced the time from project initiation to preclinical candidate nomination to 18 months.

The programme’s clinical and research record

Proteomic analyses in the IPF trial included biological-age clocks such as ProtAge, OrganAgechrono, ipfP3GPT and PAOPAC, alongside mortality-risk clocks including PAC and OrganAgemortality.

Researchers also used SenMayo and CellAge signatures to examine senescence-related biology.

UK Biobank age-associated trajectories served as external comparison datasets.

Peer-reviewed work in Aging and Disease found that pharmacological TNIK inhibition produced senomorphic activity and reduced indicators of extracellular-matrix remodelling.

Nature Biotechnology published the discovery-to-clinic progression, while the Journal of Medicinal Chemistry reported structural validation of novel TNIK inhibitor chemotypes using a TNIK kinase-domain co-crystal structure.

Nature Medicine documented the Phase IIa safety and lung-function data.

Alex Zhavoronkov, Insilico’s founder and CEO, described the programme as having progressed through target discovery, molecular design, preclinical validation, Phase I safety and randomised Phase IIa data before entering Phase III development.

The next trial will test whether the computationally selected target and molecule can produce confirmatory clinical efficacy.

Phase III enrolment size, completion timing, regulatory filing dates, partner economics and any approval timetable were not disclosed in the source.