AI-Developed Lung-Fibrosis Drug Rentosertib Shows Promising Anti-Aging Signals in Blood Protein Profiles

The intersection of artificial intelligence and longevity science has yielded a fascinating new data point, as researchers evaluating an experimental lung-fibrosis medication discovered unexpected markers of youthfulness in patients’ blood protein profiles. Published in the prestigious journal Nature Biotechnology, a recent secondary analysis of a Phase 2a clinical trial revealed that six distinct biological aging clocks detected younger protein profiles in participants treated with rentosertib. While the findings introduce exciting possibilities regarding the dual utility of targeted disease treatments as potential geroprotectors, scientists remain cautious, noting the complex challenge of distinguishing systemic anti-aging effects from localized responses to fibrosis treatment.

The discovery highlights an evolving strategy in modern pharmacology: leveraging drugs originally designed to target age-related pathologies to see if they concurrently slow down broader biological aging processes. Because many chronic diseases share common underlying mechanisms with aging—such as cellular senescence, chronic inflammation, and metabolic dysfunction—interventions tailored for specific conditions may offer systemic benefits. However, standard clinical observations are rarely sufficient to capture these sweeping systemic changes. To overcome this, researchers increasingly rely on advanced biological age clocks, computational tools designed to track chronological age or predict mortality risks using molecular data.

Background and Development of Rentosertib

Rentosertib was developed by Insilico Medicine, a prominent longevity biotechnology firm utilizing proprietary artificial intelligence-driven drug discovery platforms. The drug functions as an inhibitor of TNIK (TRAF2- and NIK-kinase), an enzyme increasingly implicated in various aging-related pathways, including cancer, metabolic dysfunction, and tissue fibrosis.

The primary target for rentosertib during its initial development was idiopathic pulmonary fibrosis (IPF), a debilitating and progressive condition characterized by the accumulation of excessive scar tissue in the lungs, which severely compromises respiratory function. IPF is profoundly age-related, with a mean onset age of approximately 65 years. The disease pathobiology heavily involves biological hallmarks of aging, notably cellular senescence and aberrant tissue remodeling.

To test the safety and efficacy of rentosertib, researchers previously conducted a 12-week, randomized, double-blind, placebo-controlled Phase 2a clinical trial. The study evaluated three distinct dosing regimens alongside a placebo control group: 30 milligrams administered once daily, 30 milligrams administered twice daily (split dose), and 60 milligrams administered once daily. Out of the 71 original participants enrolled in the primary trial, the subsequent proteomic aging clock analysis successfully incorporated blood samples from 42 participants who completed the protocol.

Proteomic Clocks Reveal Younger Biological Profiles

To evaluate the systemic impact of rentosertib beyond lung tissue, the research team employed six different proteomic aging clocks. Unlike traditional epigenetic clocks that rely on DNA methylation, proteomic clocks analyze the abundance of specific proteins circulating in the blood. These tools offer enhanced interpretability, allowing scientists to pinpoint precisely which proteins drive the algorithm’s readings and connect those shifts directly to established biological pathways.

The six clocks utilized in the study were developed by various research teams using extensive data cohorts from the UK Biobank, including one clock formulated internally by Insilico. Four of these diagnostic instruments were trained to track chronological age, while the remaining two were engineered to predict mortality risk.

The analysis produced notable results: all six clocks registered lower predicted biological ages in patients receiving rentosertib, though the magnitude and consistency of the effect varied depending on the dosage and timeline. Following four weeks of treatment, the four chronological age-based clocks indicated that patients in the 60-milligram single-daily-dose group were, on average, approximately three years younger biologically than they were at baseline. These shifts achieved statistical significance when compared against the placebo cohort. Conversely, the two mortality-prediction clocks did not register statistically significant changes for this specific dose at the four-week mark.

The most consistent reductions across both chronological and mortality-based clocks were observed in the group receiving 30 milligrams twice daily (the split-dose regimen). Interestingly, the most pronounced decreases in predicted biological age occurred by week four, with subsequent readings plateauing over the remaining eight weeks of the trial.

Researchers emphasize a vital caveat regarding these metrics: these shifts demonstrate that patient blood protein profiles converged toward signatures typically observed in younger populations. However, they do not definitively prove that patients physically reversed years of aging or gained an extended lifespan.

Dissecting the Data: Anti-Aging Versus Disease Treatment

Beyond the aggregate clock readings, a deep dive into the proteomic data revealed that rentosertib administration altered the levels of 326 distinct blood proteins. These modulated proteins are fundamentally linked to fibrosis, metabolic regulation, and cellular stress responses. In the higher-dose cohorts, the majority of these protein shifts persisted throughout the duration of the trial or emerged later in the treatment window, indicating that the drug continued exerting biological influence even after the numerical age clocks stabilized.

A central challenge in interpreting these findings involves identifying the precise drivers behind the younger age estimates. A prominent fibrosis-associated protein known as LTBP2 heavily influenced the outputs of all six clocks. This heavy reliance raised an immediate scientific question: do the synchronized clock readings reflect a genuine, systemic slowing of aging, or are they simply capturing a localized therapeutic response to reduced lung scarring?

To untangle this relationship, the research team cross-referenced changes in predicted biological age against objective measurements of patient lung function. The two metrics did not closely track one another. This divergence suggests that improved respiratory function alone cannot fully account for the dramatic shifts observed in the protein clocks, though researchers acknowledge that standard pulmonary function tests do not capture every systemic facet of IPF.

Further comparative analyses provided additional clues. When researchers matched the treatment-induced protein shifts against baseline differences between younger and older participants within the UK Biobank, they found that proteins typically elevated in older individuals tended to decrease under rentosertib treatment, while proteins associated with youth increased. This restorative pattern was statistically significant in the split-dose regimen. Furthermore, proteins tied to cellular senescence—which typically accumulated in the placebo group—decreased among treated participants, implying a potential suppression of senescence-associated secretory phenotype (SASP) activity, even though direct tissue measurements of senescent cells were not performed.

Expert Perspectives and Broader Implications

The study marks a significant methodological step forward in clinical trial design for both longevity science and traditional pharmacology. By embedding advanced molecular biomarkers directly into standard phase trials, researchers can glean unprecedented insights into secondary drug effects.

“This study illustrates how biomarkers of aging can be incorporated into conventional clinical trials,” said Vadim Gladyshev, a professor of medicine at Harvard Medical School and co-author of the study, in an interview. “If validated more broadly, such approaches could allow us to learn not only whether a treatment affects a particular disease, but also whether it influences some aspects of aging biology.”

Despite these encouraging observations, the scientific community remains measured in its conclusions. Because biological pathways governing fibrosis, metabolic health, and cellular senescence are deeply intertwined, definitively isolating a broad geroprotective effect from targeted anti-fibrotic efficacy remains exceptionally difficult.

Future clinical investigations will likely need to test AI-discovered compounds in broader populations or specifically designed longevity trials to confirm whether drugs engineered for age-related conditions can genuinely alter the fundamental trajectory of human aging. For now, rentosertib stands as a compelling proof-of-concept, demonstrating how artificial intelligence and multi-omic profiling can illuminate the hidden biological intersections between specific diseases and the aging process itself.