Semaglutide longevity signals grow in mice and DNA clocks but human proof is missing

Semaglutide longevity claims have moved from biohacker forums into peer reviewed journals, yet the evidence still stops short of healthy people. Two 2026 papers anchor the debate: a post hoc look at DNA aging clocks inside a placebo controlled semaglutide trial in people with HIV, and a study in old female mice in which the drug lengthened median lifespan. A feature in The Scientist on 26 October heard from the researchers behind both. Their shared view: the drug may slow some aging biology in people at risk, but proof that it extends healthy human lives does not exist.

Why GLP 1 aging is now a drug developer question

The case for GLP 1 aging research is indirect. Semaglutide, the Novo Nordisk peptide sold as Ozempic and Wegovy, belongs to a class with reported gains in blood sugar, blood pressure, heart and kidney function, all tied to diseases that cut lives short. Kidney protection is one area where semaglutide already holds the firmest kidney evidence in its class, from a dedicated outcome trial. According to The Scientist, researchers from Eli Lilly and Novo Nordisk discussed GLP-1 drugs aimed at aging at the 2025 Aging Research and Drug Discovery meeting, and longevity influencers soon took up the idea. The feature also reports a rise in off-label use by health focused people trying to lower their biological age, a use that no regulator has approved.

At this year's meeting, Novo Nordisk scientist Nikolaj Roed moved the pitch from lifespan to healthspan, the years a person keeps physical health, independence and clear thinking. "Whether we can call GLP-1s longevity [drugs], we can still discuss, but it's certainly playing a role in some of the important hallmarks of aging," he said in his talk. Alex Zhavoronkov, chief executive of Insilico Medicine, told the magazine he already takes a GLP-1 drug for longevity despite being of normal weight, while conceding the evidence is not complete. On our board, semaglutide stays approved for its current uses and nothing about aging is on any label. The table below sets out the main pieces of evidence the feature draws on and where each falls short.

Epigenetic clocks slowed in a placebo controlled HIV trial

Michael Corley of the University of California, San Diego, worked with Grace McComsey of Case Western Reserve University on blood samples from a phase 2b trial of weekly semaglutide in adults with HIV associated lipohypertrophy, a build up of fat around the organs. The parent trial's primary endpoint was fat distribution, not aging. The team ran epigenetic clocks, which read DNA methylation patterns in blood, at the start and after 32 weeks and compared drug with placebo. The paper, published in Nature Communications on 19 May 2026, reports slower epigenetic aging across multiple clocks in the semaglutide group. The earlier preprint version lists paired clock data for 45 people on the drug and 39 on placebo, a modest sample for a secondary question.

Corley said the result does not show the drug prevents aging in healthy people, only that it may slow aging slightly in people prone to faster aging.

Corley told The Scientist the slowing was "the shock to me," because people in this group would normally age about 64 weeks over a 32 week span. Corley said the result does not show the drug prevents aging in healthy people, only that it may slow aging slightly in people prone to faster aging. He believes the data justify "a very rigorous follow-up study." His group is now checking the same participants at 52 weeks to see whether the clock changes reverse once treatment ends, since, Corley said, McComsey sees weight and related illness return in patients who stop. Two caveats matter for semaglutide longevity claims here. The analysis was added after the trial was designed, and epigenetic clocks are surrogate markers rather than measures of how long or how well people live.

Caloric restriction mimetic or more in old mice

Danica Chen of the University of California, Berkeley, tested the idea in animals without disease, which she says sets the work apart from earlier studies in sick mice or patients. In the NIH funded study, her team gave semaglutide or a control to 20 month old female mice, roughly 60 human years by The Scientist's estimate. Continued treatment extended lifespan, the Nature paper on aged mice reports. According to the journal's press release, median lifespan reached 834 days on semaglutide against 742 days for controls. Three months of treatment also improved movement and exploration and reduced markers of several hallmarks of aging, including inflammation, cellular senescence and stem cell decline.

The sharper question was whether semaglutide is only a caloric restriction mimetic, copying the one intervention with solid evidence of slowing aging. The drug cut food intake by 24 percent, the paper reports, so a second group of mice ate a diet restricted to match. Exercise tolerance and muscle function came out almost the same in both groups. Semaglutide did better on exploratory behaviour, spatial memory and blood sugar control, results Chen said surprised her. Corley, who was not involved, called it "one of the strongest pieces of preclinical evidence so far" that GLP-1 receptor activation can affect aging itself. He also flagged the limit: all the animals were female and came from a single inbred strain, so nobody knows yet how far the effect generalises.

What a semaglutide longevity trial would need to show

Turning semaglutide longevity from a hypothesis into a label claim needs an endpoint regulators accept. "There is a good reason why there is no longevity drug approved," Zhavoronkov said, adding that current drugs might work on aging and "we just don't know." He and Corley both point to the proteomic aging clock, which reads plasma proteins and was linked to death and disease risk in a 2024 Nature Medicine study, as a possible surrogate. Roed said at the 2026 meeting that such clocks improve on GLP-1 drugs in ongoing trials, but that the lifespan gain might be only about two years, presuming lifelong use. That figure is a company scientist's estimate from unpublished data, not a trial result. Lilly and Novo Nordisk declined to comment for the feature.

Many basic questions remain open. Nobody knows whether benefits would appear in metabolically healthy people, how early treatment would need to start, what amount would suit a healthy adult, or whether use would have to be lifelong. Staying on the drug is already hard in routine care, as our report on early semaglutide discontinuation showed. Corley also notes that some people respond poorly to GLP-1 drugs, and he expects future work to favour combinations, perhaps tailored to aging, as the incretin pipeline grows. Healthy users face known risks such as lean muscle loss, which matters most in frail older adults. Any use outside approved indications belongs with a prescriber and licensed medicines, not unregulated copies, and this report is research context rather than medical advice.

Conclusion

Semaglutide longevity now has two serious data points: slower DNA clocks in a small post hoc human analysis and longer median life in old female mice. Neither shows that healthy people will live longer. The flat signal reflects that gap. "The jury is still out on longevity," Corley said, calling for human trials that measure function and survival rather than disease risk alone. The next checkpoints are his 52 week follow up and any published proteomic clock data from company trials.