An AI-discovered lung therapy also shifted aging markers - critical summary review

There is a family of tests known as aging clocks. They try to estimate a person's biological age, as opposed to the chronological one, by reading how well cells and tissues appear to be working. Some versions read chemical changes in the body. The ones at the centre of this story read proteins circulating in the blood, and they do not all measure the same thing: some were built to guess a person's age from those proteins, others are tied more closely to the risk of dying.

That is the backdrop against which the news arrives. In a small trial, patients taking an experimental compound for a serious lung disease came out of six of those clocks looking biologically younger than patients on placebo. The results were published in the journal Nature Biotechnology and presented at a Nature conference at Sorbonne University, in Paris. The compound was discovered with the help of artificial intelligence, which is why a respiratory-medicine result ended up circulating as an anti-aging headline.

What rentosertib is, and how AI found it

Rentosertib is an experimental therapy developed by Insilico Medicine, a drug-discovery company based in Massachusetts. It is being tested against idiopathic pulmonary fibrosis, an incurable disease that scars the lungs and progressively compromises breathing.

Artificial intelligence entered at two points. Insilico first used a tool that sifts large volumes of data to rank possible therapeutic targets, and it flagged a protein called TNIK as an element involved in aging. The company then used a second tool, which virtually screens molecules and suggests ways to optimise them, to design and refine the compound that acts on that protein.

The compound reached a phase 2a clinical trial run between 2023 and 2024 at 21 centres in China. It was randomised, double-blind and placebo-controlled, with 71 patients receiving different doses or placebo. Some of them agreed to a more detailed analysis of the proteins in their blood, and after one exclusion for missing data, samples from 42 participants formed the group in which the aging analysis was done.

Six clocks pointing in the same direction

Over 12 weeks, the researchers tracked 2,841 blood proteins and ran six proteomic clocks over the data. The clocks were built by teams at Insilico, Harvard, Peking University and other institutions, and they use different metrics and methods. All six pointed the same way: a fall in calculated biological age among the patients on rentosertib. In the placebo group, the changes were small or showed a slight increase.

The strongest effect appeared at week four, among participants taking 30 mg twice a day. Some models indicated a drop equivalent to roughly three to four years; in one measurement, the difference reached about six years. According to O Globo, the benefit diminished later in the trial.

Two other findings sit alongside the clocks. Treatment altered the trajectory of 326 of the proteins analysed, against two in the placebo group, with the widest response concentrated in that same twice-daily 30 mg dose. And when the researchers compared their data with more than 55,000 people in the UK Biobank, a large British health database, several blood proteins had moved in the opposite direction from what is normally seen with aging. The team also reported changes in processes involved in how cells function and respond to damage.

"There is a promising trend in age reversal in addition to treating the disease, and it came as a bonus for disease treatment," Alex Zhavoronkov, founder and co-CEO of Insilico Medicine and first author of the paper, told Fox News Digital. "It basically means that you are predicted to be younger using a validated set of aging clocks. We don't know what that means yet."

Why this is not proof that aging was reversed

The most important limitation is who was in the study. All 42 participants had idiopathic pulmonary fibrosis, a disease that can itself change many of the same proteins used to measure biological age. That means the improvement in their lung disease may be what moved the protein readings. The authors write that the clocks capture changes in proteins linked to aging, but do not yet allow them to determine how much of the effect comes from the disease.

The other limitations are the ones the researchers list themselves: a small sample, a 12-week window, and results based only on proteins in the blood. They acknowledge the findings do not prove the compound slows aging, keeps people healthier for longer or extends their lives, and do not support using rentosertib as an anti-aging treatment. "This is a pilot," Zhavoronkov said. "It is early [and] exploratory." Elsewhere in the same interview he added that the compound "is still investigational".

That last word matters, because the coverage travelled faster than the evidence. What was measured is a shift in blood proteins, in sick patients, read through models that are themselves still being validated. The convergence of all six clocks is what makes researchers want to look again in bigger studies.

Dual-use molecules, and a category regulators do not have

Longevity enthusiasts have long argued for repurposing existing medicines to slow aging, and attempts to develop new anti-aging drugs have frequently ended in failure. What changed the climate, according to O Globo, is the arrival of GLP-1 treatments, whose benefits now appear to reach well beyond weight loss and diabetes. Large pharmaceutical companies have grown more interested in the field, and firms such as Insilico have chosen to concentrate on dual-use therapies: compounds that treat an age-related disease while, in theory, also helping people live longer. Zhavoronkov describes the approach as using AI to hunt what he calls "molecular criminals", biological targets that may contribute to both aging and disease. Insilico says it plans similar analyses for at least half the compounds in its pipeline.

On safety, Carol Ann Satler, senior vice-president of clinical development at Insilico, told Fox News Digital that the risks and side effects were "manageable": most commonly mild diarrhoea, slightly lower blood potassium and temporary rises in liver enzymes. "Once you stop the drug, essentially, they return to normal," she said, adding that the next trial will adjust doses based on the side-effect profile so patients can stay on treatment longer.

Rentosertib has now moved into a year-long phase 3 trial of more than 300 patients in China, still for the lung disease. That distinction is the practical one. It is currently impossible to obtain regulatory approval for aging itself as a condition; Zhavoronkov has suggested that results like these could open a path for doctors to prescribe or test the compound for aging, or for age-related diseases, once it is approved for another use.

What is known — and what is still unknown

Known: the analysis was published in Nature Biotechnology, with researchers from Insilico, Harvard Medical School, Stanford University, the Broad Institute, RWTH Aachen University, Peking University and Westlake University. In 42 lung-disease patients followed for 12 weeks, six proteomic clocks all shifted toward a lower biological age, with the largest signal at week four on the twice-daily 30 mg dose, and hundreds of proteins changed trajectory against two in the placebo group. Side effects were described by the company as manageable and reversible on stopping. And the compound is in late-stage testing in China for idiopathic pulmonary fibrosis, not for aging.

Unknown: whether the same signal appears in a trial of more than 300 patients rather than a pilot of 42; how much of it reflects lungs getting better rather than a body getting younger; whether a shift in blood proteins corresponds to anything a patient would feel, live longer with or be treated for; and what happens in people who do not have the disease, since larger and longer trials including healthy older adults have not been done.

Two concrete markers are worth following instead of the headline. The first is the phase 3 read-out in China, which is where the effect either holds up outside a pilot or does not. The second is regulatory: whether any authority eventually allows rentosertib to be tested or prescribed for aging or age-related conditions after an approval for lung fibrosis. Until one of those moves, "becoming biologically younger" belongs where the researchers put it, as a laboratory hypothesis about proteins in blood, not as a treatment anyone can get.