AI-Designed Drug Rentosertib Reverses Biological Age by 3-4 Years in Clinical Trial

Insilico Medicine Achieves Clinical Milestone with AI-Designed Drug

Insilico Medicine has published groundbreaking findings in Nature Biotechnology showing that rentosertib, an AI-designed drug, reduced biological age markers in a Phase IIa clinical trial.1

1

The study, conducted with researchers from Harvard Medical School, Stanford University, The Broad Institute, RWTH Aachen University, Peking University, and Westlake University, evaluated 42 patients from a 71-participant trial originally designed to treat idiopathic pulmonary fibrosis.5

5

This represents the first clinical evaluation of a drug candidate featuring both an AI-discovered target with relevance to aging biology and an AI-designed molecule, distinguishing it from repurposed generic drugs like rapamycin or metformin.1

1

Six Proteomic Aging Clocks Show Consistent Age Reversal

Researchers evaluated serum proteome profiles across 2,841 proteins using six independently developed proteomic aging clocks: ProtAge, OrganAge (chronological and mortality variants), PAC, ipfP3GPT, and PAOPAC.1

1

Despite relying on diverse methodologies and training criteria, all six aging clocks consistently indicated a reduction in predicted biological age among patients treated with rentosertib over the 12-week trial period.2

2

The peak effect appeared at week four in participants receiving 30mg twice daily, with approximately 3-4 years reversal in biological age and up to 6 years on certain aging clocks.1

1

At week four, patients taking 60mg daily showed reductions of roughly 2.7 to 3.5 years on four chronological-age clocks.2

2

AI-Driven Drug Discovery Process Behind Rentosertib

Insilico Medicine, founded in 2014 by Alex Zhavoronkov, incorporated aging biology into its AI-driven drug discovery strategy from the outset.1

1

Using its AI-powered target discovery platform, Insilico identified TNIK as a novel TNIK inhibitor implicated in aging and fibrosis, scoring highly in 6 hallmarks of aging.1

1

The company's generative chemistry platform, Chemistry42, then designed the small-molecule candidate rentosertib targeting TNIK.1

1

The program progressed from target identification to preclinical candidate nomination in approximately 18 months.1

1

Insilico built neural networks to analyze health records, blood tests describing microscopic proteins, and academic papers, then created a second system analyzing protein structures to generate entirely new molecules.3

3

Dual Benefits for Lung Disease and Aging Biology

The Phase IIa trial, reported in Nature Medicine in June 2025, met its primary safety endpoint and demonstrated promising dose-dependent efficacy trends.1

1

In the 60mg once-daily group, patients experienced a mean improvement in Forced Vital Capacity of +98.4mL, compared with a mean decline of -20.3mL in the placebo group.1

1

Idiopathic pulmonary fibrosis, which typically manifests around age 65, progressively scars lung tissue and reduces oxygen transfer to the bloodstream.3

3

In healthy individuals older than 65, Forced Vital Capacity declines at 20-50mL per year, making it a potential physiological biomarker of aging.1

1

The study showed that rentosertib's anti-aging effects operate partially independent of its respiratory benefits.1

1

Limitations and Path to Regulatory Approval

The study's authors acknowledge they cannot fully separate effects on aging from improvements in lung disease.2

2

The trial sample size was modest at 42 participants with an average age of about 67, and aging clocks are not always reliable predictors.3

3

Vadim Gladyshev, a Harvard Medical School professor who helped build one of these proteomic aging clocks, called it "the first study that shows, very clearly, that predicted biological age can be reduced," but noted the small sample size.3

3

Seven patients discontinued due to liver toxicity, with four also taking nintedanib, the existing standard of care.5

5

Rentosertib remains investigational without regulatory approval, and no medicines regulator licenses drugs based on changes in proteomic age.5

5

Future Research and Broader Implications for Longevity Research

Insilico has announced a Phase 3 trial targeting lung disease, expected to enroll 320 participants and measure changes in lung function over 52 weeks.2

2

Establishing whether the AI-generated drug benefits people without idiopathic pulmonary fibrosis will require further research.2

2

The findings provide a proof-of-concept framework for integrating biological age markers into standard clinical trials to accelerate discovery of geroprotective drugs and longevity therapeutics.1

1

Insilico deposited proteomic data with the China National Center for Bioinformation and released analysis pipelines as open-source software, making claims checkable by independent researchers.5

5

Alex Zhavoronkov will present results at the Nature conference: Redefining Healthcare in the Age of AI at Sorbonne University in Paris on September 8, 2026.1

1

The work represents a milestone in efforts to improve healthcare using generative AI techniques that underpin systems like ChatGPT, pushing beyond short-term treatments into longevity research.3

3