Insilico Medicine announced that rentosertib, a drug whose target was identified by artificial intelligence, whose molecular structure was designed by generative AI, and whose clinical development is guided by AI prediction tools, has advanced to Phase III clinical trials. This marks a watershed moment for AI-driven drug discovery: the first time a medicine born entirely from AI workflows has reached late-stage human testing.

Rentosertib targets a protein called TNIK, which plays a role in fibrosis, inflammation, and cellular aging. The drug is being tested for idiopathic pulmonary fibrosis (IPF), a progressive lung-scarring disease that primarily affects older adults. After diagnosis, the median survival is typically two to four years, and current approved treatments can only slow the disease's progression, not reverse it.

How Did AI Discover a Drug That Humans Missed?

The discovery process behind rentosertib reveals how AI is reshaping pharmaceutical research. Rather than starting with a known drug target and screening thousands of compounds, Insilico's team used a biology-first approach powered by its PandaOmics platform, an AI system trained to analyze multi-omics data, biological networks, and scientific literature to identify disease-driving mechanisms.

The workflow involved several AI-powered steps:

• Target Discovery: PandaOmics analyzed fibrotic tissue data, biological networks, and aging-related biology to rank potential drug targets, identifying TNIK as a top candidate that had been largely overlooked by conventional drug discovery approaches.
• Molecular Design: Chemistry42, Insilico's generative chemistry platform, designed and optimized the small-molecule structure of rentosertib to have the properties needed for clinical development.
• Clinical Prediction: Medicine42's inClinico platform predicted and helped improve the drug's performance in clinical trials before they began.

TNIK is a serine/threonine kinase involved in multiple disease pathways, including Wnt, TGF-beta, Hippo/YAP-TAZ, JNK, and NF-kappa-B signaling. The AI system identified it as a previously underexplored target class for IPF compared with the receptor tyrosine kinase biology addressed by existing antifibrotic drugs.

What Do the Early Clinical Results Show?

Before advancing to Phase III, rentosertib completed a Phase IIa trial called GENESIS-IPF, with results published in Nature Medicine. The trial demonstrated manageable safety and tolerability across different dose levels. Most importantly, the 60-milligram once-daily dose showed a mean improvement in forced vital capacity (a measure of lung function) of 98.4 milliliters at 12 weeks.

These results attracted enough attention that the Phase III trial is now underway. The study will enroll 320 patients with IPF and run for 52 weeks, led by Professor Zuojun Xu of Peking Union Medical College Hospital as the principal investigator, with support from renowned respiratory medicine experts.

"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," said Feng Ren, Co-CEO and Chief Scientific Officer of Insilico Medicine. "Rentosertib was not discovered by starting from a conventional target and simply screening more compounds. It came from a biology-first, aging-informed AI workflow that connected TNIK to fibrotic and inflammatory disease mechanisms, and then used generative chemistry to create a drug candidate with the properties required for clinical development."

Feng Ren, Co-CEO and Chief Scientific Officer, Insilico Medicine

Why Does This Matter Beyond IPF?

Rentosertib's journey from AI discovery to Phase III represents a proof-of-concept that could reshape how pharmaceutical companies approach drug development. The drug's discovery was grounded in aging biology, reflecting a broader thesis that diseases of aging share common molecular mechanisms that can be targeted with a single drug.

Research published in Nature Aging highlighted Insilico's approach of using AI to identify targets implicated in multiple hallmarks of aging, including inflammation, extracellular matrix remodeling, and age-related disease mechanisms. This strategy could unlock a new class of therapeutics that address both disease pathology and aging-associated biology simultaneously.

The drug's development also benefited from AI-driven automation. In separate research, TNIK inhibition showed senomorphic activity in cellular senescence models, meaning it reduced aging-related markers including senescence-associated secretory phenotype (SASP) and extracellular matrix remodeling signals. While these findings do not establish rentosertib as an anti-aging therapy, they strengthen the scientific rationale for investigating TNIK at the intersection of fibrosis, inflammation, senescence, and age-related disease biology.

The Phase III trial represents a critical test: can an AI-discovered drug not only show promise in smaller trials but also demonstrate efficacy and safety in a larger, more diverse patient population over a longer timeframe? If successful, rentosertib could become the first AI-designed medicine to receive regulatory approval, validating the entire AI drug discovery pipeline and encouraging pharmaceutical companies to invest more heavily in generative AI tools for target identification and molecular design.