Nobody can change the birth year on their passport. But a new study suggests a drug built for a scarring lung disease may shift how old a body looks from the inside, at least on paper.
Patients with a serious lung condition took an experimental pill designed with the help of artificial intelligence. Their blood was later scored by six separate “aging clocks,” and all six said the same thing: the treated patients looked biologically younger than expected, while those on a dummy pill did not.
The interesting half is not the size of the number. It is the agreement, and the honest admission from the researchers about what they still cannot tell apart.
What the blood tests showed
- The analysis, published on 7 September 2026 in Nature Biotechnology, used stored blood from 42 people with idiopathic pulmonary fibrosis, average age 67.1.
- Researchers measured 2,841 proteins in each sample and ran them through six independently built proteomic aging clocks.
- All six clocks pointed toward a lower biological age in people taking rentosertib.
- At week 4, the 60 mg once-daily group read 2.71 to 3.46 years younger on all four clocks that estimate age.
- The drug shifted 326 proteins, against just 2 in the placebo group.
| Finding | Plain-English meaning |
|---|---|
| All 6 aging clocks moved toward a younger age on the drug | Models built by different teams, from different proteins, reached the same direction of answer, which makes a one-model fluke less likely. |
| 2.71 to 3.46 years younger at week 4 (60 mg once daily) | On the age-estimating clocks, treated patients’ blood looked roughly three years younger than it should for their age. |
| 326 proteins changed on the drug vs 2 on placebo | The pill had a broad effect on the body’s protein signals, not a tiny one. |
| No further shift between weeks 4 and 12 | The younger reading held steady rather than growing, so it plateaued early. |
| Only 42 patients, all with lung disease | This is a small, early signal, and the effect could come from a healthier lung rather than slower aging. |
A lung drug that was also aimed at aging
Idiopathic pulmonary fibrosis, usually shortened to IPF, slowly scars the lungs until breathing becomes hard work. It mostly affects older adults, and its cause is unknown. The two approved treatments slow the decline in lung function. They do not stop or reverse it.
Rentosertib comes from Insilico Medicine. The company says AI helped it pick the drug’s target, a protein called TNIK, and design the molecule that blocks it. Insilico had flagged TNIK as linked to several hallmarks of aging, not only to lung scarring. That is why the team asked whether a lung drug might show up on aging tests too.
The blood came from an earlier phase 2a trial of 71 patients at sites in China, run over 12 weeks. Patients were randomly assigned to one of three doses or a placebo. That trial’s main goal was safety, and it met it. The press release for the new paper also notes that the 60 mg once-daily group gained 98.4 mL of lung capacity over the trial, while the placebo group lost 20.3 mL.
For the new analysis, 42 of those patients who had agreed to protein testing gave blood at the start and at weeks 2, 4 and 12. The researchers compared their protein profiles against more than 55,000 reference profiles from the UK Biobank.
Why six clocks beat one
An aging clock is a model that looks at a blood sample and estimates how old the body behaves, or how likely a person is to die. There is no gold standard. A single clock can be thrown off by the particular proteins it relies on or by the maths behind it.
So the team did not trust one. They used six, built by different groups using different methods. Four estimate age. Two estimate mortality risk.
The clearest results came at week 4. The 30 mg twice-daily dose gave the most consistent response, showing up on both the age clocks and the mortality clocks. The 60 mg once-daily dose produced the clean 2.71 to 3.46 year figure on the age clocks, but did not significantly move the mortality clocks.
Some exploratory models that look at individual organs gave much larger numbers. Treat those with real caution. Organ-level clocks are newer, and the researchers did not name one figure as the overall amount of change. The more solid claim is the direction, not any single headline number.
The part coverage should not skip
Here is the catch, and the researchers state it plainly: they cannot tell whether the drug slowed aging or simply calmed a sick lung.
IPF is itself an inflammatory disease, and inflammation leaves fingerprints in blood proteins. A drug that eases the disease would be expected to push many of the same markers that aging clocks read, without touching aging anywhere else in the body. In the paper’s words, full disentanglement is not achievable within an IPF group.
A few more limits matter:
- It is small. Forty-two people, split across three doses and a placebo group, is a thin base for any firm conclusion.
- It is short. Twelve weeks cannot show whether anyone stays healthier, and the younger reading stopped growing after week 4. The authors suggest this may be a plateau, possibly tied to the short window.
- It was a second look, not a new trial. The aging question was asked of samples collected for a lung study, so it is exploratory by design.
- The drug maker ran it. Insilico sponsored the trial and several authors, including first author and co-CEO Alex Zhavoronkov, work for the company. The senior author, Vadim Gladyshev, is an academic aging researcher at Harvard Medical School, and the company has released its analysis pipelines as open-source software, which lets outsiders check the work.
- A younger clock is not a longer life. No regulator approves a drug on a change in proteomic age. These readings are model predictions, not proof of longer, healthier living.
The results did hold up when the team removed the six patients with more serious side effects. That makes a pure data artefact less likely, but it does not solve the lung-versus-aging question.
Why this still counts as a step forward
Most aging research lives in mice. Human trials of “anti-aging” drugs are hard to run because aging takes decades to show up as disease. This study points to a cheaper route: take a drug already being tested for an age-related illness, and read aging clocks from the blood samples that trial collects anyway.
If that approach holds up, one trial could answer two questions at once. Does the drug treat the disease, and does it nudge the body’s aging signals too? The authors call this dual-purpose trial design, and it is arguably the most useful idea in the paper.
For people living with IPF, the lung results matter on their own. A treatment that gains lung capacity rather than just slowing the loss would be a meaningful advance, whatever it does to any aging clock. Insilico says rentosertib has moved into a larger phase 3 trial focused on lung function.
What to do with this news
If you have IPF, this does not change your treatment today. Rentosertib is investigational and not approved. Ask your lung specialist whether any open trials suit you, and keep taking your prescribed therapy.
If you are interested in aging, be wary of anyone selling a blood test or supplement on the strength of this study. It did not show that anyone aged more slowly or lived longer. It showed that six computer models agreed on a direction in 42 sick patients over three months. That is a promising lead, and it needs larger and longer trials, including in healthy people, before it means anything for the rest of us.
Sources:
- Nature Biotechnology, “Integration of proteomic aging clocks in a phase 2a clinical trial supports simultaneous geroprotective assessment” (https://www.nature.com/articles/s41587-026-03286-y)
- News-Medical, “Rentosertib shifts multiple proteomic aging clocks toward younger age profiles in IPF trial” (https://www.news-medical.net/news/20260910/Rentosertib-shifts-multiple-proteomic-aging-clocks-toward-younger-age-profiles-in-IPF-trial.aspx)
- ScienceAlert, “AI-Designed Drug Makes Patients’ Blood Look Biologically Younger, Study Shows” (https://www.sciencealert.com/ai-designed-drug-makes-patients-blood-look-biologically-younger-study-shows)
- EurekAlert (Insilico Medicine press release), “Nature Biotechnology | Insilico’s AI-driven IPF candidate rentosertib shows potential for biological age reversal, as assessed by six proteomic aging clocks” (https://www.eurekalert.org/news-releases/1142848)
This article is for general information and is not medical advice. Talk to a qualified clinician before making any changes to your or your child’s care.