An experimental drug designed with artificial intelligence showed signs of reducing biological age in a small clinical trial of patients with a serious lung disease, according to research published Wednesday in Nature Biotechnology. The study found that blood-based aging clocks shifted toward younger predicted ages in participants receiving the drug, though the researchers cautioned against interpreting the results as literal age reversal.
The drug, rentosertib, is being developed to treat idiopathic pulmonary fibrosis, a condition that causes progressive scarring of the lungs. Researchers analyzed blood samples from 42 participants in a 12-week Phase 2a trial and applied six different aging clocks based on changes in blood proteins. The participants, who had an average age of 67.1 years, received one of three rentosertib regimens or a placebo. Blood samples were collected at the start of the study and after two, four and 12 weeks.
What the aging clocks revealed
All six aging clocks showed a shift toward a younger predicted biological age among people receiving the drug. The placebo group showed little change or slight increases. The strongest overall signal appeared after four weeks, particularly among patients taking 30 milligrams twice a day.
For another treatment group receiving 60 milligrams once a day, four of the clocks estimated biological-age reductions of between 2.71 and 3.46 years at week four. The researchers stressed that these figures reflect changes in blood-based markers and should not be interpreted as patients literally becoming several years younger.
Separating disease improvement from aging effects
A central question is whether the apparent effect on aging was simply the result of improvements in the patients' lung disease. The researchers found that the dose producing the greatest improvement in lung function was not the one showing the strongest aging-clock response. This disconnect suggests the two effects may operate through different mechanisms, though the evidence remains indirect.
The team also compared the protein changes with data from more than 55,000 older adults and found that the twice-daily regimen tended to move some age-related protein patterns in the opposite direction from normal aging. This finding adds weight to the possibility that the drug affects aging-related biology rather than just disease symptoms.
Limitations and next steps
The study involved a small number of people, lasted only 12 weeks, and focused exclusively on patients with pulmonary fibrosis. These constraints make it difficult to separate the drug's effects on the disease from any possible effects on aging itself. The authors said further studies, including research in people without the lung condition, will be needed to determine whether rentosertib genuinely affects the human aging process.
Rentosertib was designed with the help of AI, a method increasingly used in drug development to identify promising molecular candidates faster than traditional approaches. The integration of machine learning into early-stage drug discovery represents a growing area of interest for pharmaceutical researchers, with several AI-designed compounds now advancing through clinical trials. For scientists working at this intersection, structured training in applying AI to research workflows has become a practical necessity as the field matures.
Why this matters for research scientists
This study illustrates both the promise and the methodological challenges of using aging clocks as surrogate endpoints in clinical trials. For researchers in drug development, the key takeaway is that blood-based biomarkers can produce signals that look compelling but require rigorous disentangling from disease-specific effects. The finding that the dose optimizing lung function differed from the dose optimizing aging-clock readings is a concrete example of why mechanism-of-action studies must accompany biomarker data. Replicating these results in healthy populations and over longer time horizons will determine whether the observed shifts represent a genuine geroprotective effect or a transient response to treatment.
Your membership also unlocks:
