A small-molecule drug designed entirely by artificial intelligence has produced the most compelling clinical evidence yet that pharmacological intervention can reverse markers of biological aging. Insilico Medicine published results yesterday in Nature Biotechnology showing that its experimental compound rentosertib reduced patients' predicted biological age by up to six years after just four weeks of treatment.
The findings come from a secondary analysis of blood samples collected during a Phase 2a trial for idiopathic pulmonary fibrosis, a progressive and often fatal lung disease. Forty-two participants consented to longitudinal proteomic profiling, which measured 2,841 proteins at baseline and at weeks two, four, and twelve. The cohort had a mean age of 67.1 years.
Six Clocks, One Direction
What makes this study unusual is the methodology. Rather than relying on a single biomarker or proprietary aging test, researchers applied six independently developed proteomic aging clocks created by different academic groups using different training approaches. The models included ProtAge, two OrganAge variants, PAC, ipfP3GPT, and PAOPAC. Despite their methodological differences, all six registered the same trend: treated patients showed lower predicted biological age compared to placebo.
The strongest signal appeared in participants receiving 30 mg twice daily. At week four, five of the six clocks showed negative effect sizes for that dosing arm, with reductions ranging from roughly 2.7 to 3.5 years across the chronological clocks. One mortality-trained organ clock, focused on arterial aging, showed even larger shifts, predicting age reductions of up to 16.57 years in some treatment arms.
Of 54 statistical comparisons across treatment arms, timepoints, and clocks, 21 reached significance with a false-discovery-adjusted Q value below 0.10. Eleven of 18 comparisons at week four showed significantly lower changes in predicted biological age.
Beyond Lung Function
The study revealed something the researchers did not anticipate. The dose associated with the greatest improvement in lung function differed from that producing the strongest age-reversal signal. The 60 mg once-daily arm showed the best forced vital capacity response, while 30 mg twice daily produced the broadest cross-clock agreement on biological age. That divergence suggests rentosertib's effects on aging-related proteins cannot be explained simply by improved respiratory function.
Rentosertib targets TNIK, or Traf2- and Nck-interacting kinase, a gene that Insilico's PandaOmics AI platform identified as being implicated in six hallmarks of aging: chronic inflammation, cellular senescence, deregulated nutrient sensing, genomic instability, stem cell exhaustion, and altered intercellular communication. The molecule itself was generated through the company's Chemistry42 generative chemistry platform. The entire program moved from target identification to preclinical candidate nomination in approximately 18 months.
What This Proves, and What It Doesn't
The study establishes a proof-of-concept framework for integrating aging biomarkers into conventional disease trials, which could accelerate the discovery of geroprotective drugs without requiring entirely new regulatory pathways. All proteomic data have been deposited with China's National Center for Bioinformation, and the analysis pipeline has been released as an open-source Python library.
But the authors are careful about what they claim. This was a small trial in severely ill patients with an age-related disease. The study cannot yet establish that rentosertib slows or reverses systemic human aging in healthy populations. Larger trials, longer follow-up, and studies outside severe disease contexts will be necessary before anyone should start talking about longevity drugs.
Rentosertib, which Insilico describes as the first drug candidate with both an AI-discovered target and an AI-designed molecule, has now advanced to a Phase 3 trial in IPF. That study, initiated in July 2026 across 47 centers in China, will enroll approximately 320 patients and track the annual rate of decline in forced vital capacity over 52 weeks.
Whether rentosertib ultimately proves useful for longevity remains an open question. But the company's approach, embedding geroscience endpoints into disease-focused trials, represents a template other drug developers may follow. Alex Zhavoronkov, Insilico's founder and co-CEO, is presenting the results today at the Nature conference "Redefining Healthcare in the Age of AI" at Sorbonne University in Paris.
Insilico Medicine (HKEX: 3696) went public in Hong Kong in December 2025, raising HK$2.277 billion in the largest biotech IPO on the exchange that year.
