Clinical Research · global
Pulmonary fibrosis drug turns back protein clocks: rentosertib study reveals clues to anti-aging effects
All six models detected a decline in predicted biological age among patients receiving the drug, but improvements in lung disease remain difficult to separate from changes in aging. This analysis of clinical samples also tests whether “aging clocks” can become reliable indicators of drug effects.
Could treating a disease closely associated with aging also change the trajectory of how the body ages? A study of patients with idiopathic pulmonary fibrosis has found clues in blood proteins that warrant further investigation: patients receiving the drug candidate rentosertib showed a decline in predicted biological age across six models. However, key questions remain before this can be considered evidence of “age reversal” in humans.
The study was published in *Nature Biotechnology* on September 7, 2026, and the Broad Institute’s publication record also includes the same findings. The study used samples from a 12-week, randomized, double-blind, placebo-controlled phase 2a trial. The original trial included 71 patients, while this protein analysis included 42 participants who provided complete sampling data, with an average age of approximately 67 years.
Developed by Insilico Medicine, rentosertib is a TNIK kinase inhibitor whose target identification and molecular design were assisted by artificial intelligence. Its intended uses include treating pulmonary fibrosis. According to a company announcement published by News-Medical, the team first used AI to screen for targets associated with disease and aging, then designed candidate molecules using a generative chemistry platform. Human samples now offer an opportunity to examine its biological effects, but AI’s involvement in the design does not itself guarantee efficacy.
The researchers measured 2,841 proteins in serum and analyzed them using six “proteomic aging clocks.” These models use protein profiles to estimate age or a state associated with mortality risk; their readings are not direct measurements of the body’s age. Overall, the six models pointed to a lower predicted biological age, with the group receiving 30 milligrams twice daily showing the most consistent signal, particularly at week 4. Differences across models, doses, and time points did not all reach statistical significance.
To understand why the readings changed, the team further examined the biological pathways involving these proteins and identified changes associated with cellular senescence, metabolism, and antifibrotic effects. The abstract included in the Broad Institute’s record describes these findings as potential clues to anti-aging effects and proposes incorporating aging assessments into disease-focused trials.
The greatest obstacle to interpretation comes from the very disease the drug was intended to treat. Improvements in pulmonary fibrosis could themselves alter blood proteins, leading the clocks to produce younger age estimates; multiple models could also be influenced by the same disease signals. The authors explicitly caution that proteomic clocks alone cannot disentangle the effects of disease improvement from effects on aging itself. The small sample and follow-up of only 12 weeks also limit what the findings can tell us about long-term health.
The value of this work lies in bringing an anti-aging hypothesis into a testable clinical framework. The next questions are whether the clock changes persist, whether they correspond to actual improvements in patients’ function and health, and whether they hold across different populations. Until that evidence is available, a younger protein profile should be viewed as a research signal, insufficient to support uses for life extension or anti-aging.