Back in July we looked at human epigenetic data suggesting semaglutide slows biological ageing markers. Interesting, but correlational. This month the field got something more concrete: a Nature paper showing that semaglutide given late in life actually extended lifespan in mice - and that it held its own against calorie restriction, the most established lifespan-extending intervention in the biology of ageing playbook.
The study, led by Danica Chen’s lab at UC Berkeley and published in Nature on September 2, is the most direct test yet of the idea that GLP-1 drugs act as calorie restriction mimetics. The results suggest they do - and that in a few measurable ways, they might do things that simply eating less does not.
The study
The researchers worked with 20-month-old female C57BL/6 mice. In mouse terms that is well into old age - the equivalent of starting the intervention when ageing is already well underway, not preventing it in youth. That design choice matters. Plenty of interventions look good when started early in a long life; far fewer rescue anything when the clock has mostly run.
Two comparisons drove the paper. First, mice received semaglutide for three months and were assessed against untreated animals for physiological function and molecular markers of ageing. Second, in a head-to-head arm, one group got semaglutide for five months while another was placed on a 24 percent calorie-restricted diet calibrated to match how much the drug-treated mice were actually eating. That matching is the clever part: since semaglutide suppresses appetite, the obvious confound is that any benefit just comes from eating less. Matching intake isolates whatever the drug does beyond reduced calories.
What happened to the mice
The short version: the treated animals aged more slowly and lived longer.
On function, semaglutide-treated mice showed better muscle and cognitive performance than untreated controls, along with gene-activity profiles pointing to reduced inflammation and better maintenance of tissue repair and regeneration - several hallmarks of ageing moving in the right direction at once.
On lifespan, mice treated until death had a median lifespan nearly 100 days longer than untreated animals. In a mouse that already starts the study old, that is a substantial extension of the remaining tail of life.
The head-to-head against calorie restriction is where it gets genuinely interesting. Most physiological measures came out similar between the two groups - both interventions preserved baseline function and blunted age-associated decline. But semaglutide separated from calorie restriction in a few places:
- Exploratory behaviour improved beyond the animals’ own starting levels, where calorie restriction did not
- Spatial memory showed more favourable trajectories under the drug
- Blood-sugar control improved above baseline with the drug
- Metabolic rate stayed largely unchanged in treated mice, while it slowed in the calorie-restricted group - meaning the drug’s effects did not simply ride on the metabolic slowdown that eating less produces
What the Research Says
- Semaglutide treatment started at 20 months of age in female mice improved physiological function, attenuated hallmarks of ageing and extended median lifespan by nearly 100 days (Feng et al., Nature, September 2026).
- In direct comparison with matched 24 percent calorie restriction, semaglutide reproduced most of calorie restriction’s functional benefits and exceeded it on exploratory drive, spatial memory and glucose control.
- The authors frame GLP-1 receptor activation as functioning as a calorie restriction mimetic, while noting the drug produced effects beyond those attributable to reduced calorie intake alone.
- Rafael de Cabo, senior investigator at the NIH’s National Institute on Aging, authored an accompanying commentary noting that if GLP-1 drugs slow ageing itself, their wide range of clinical benefits is exactly what you would expect to see.
- This is mouse data. It does not show that Ozempic, Wegovy or any GLP-1 drug extends human lifespan, and the authors say so.
Why the calorie restriction framing matters
Calorie restriction has been the reference standard in ageing biology for nearly a century. Reduce intake enough, and lab animals across species live longer and age slower. The catch, in both mice and humans, is the price tag: chronic hunger, reduced energy, slowed metabolism, and in humans real problems with muscle and bone at aggressive restriction levels.
A drug that reproduces the ageing benefits of eating less without actually requiring the deprivation - and without slowing metabolic rate - is the textbook definition of a calorie restriction mimetic. That is the framing the Nature paper earns, and it is why this study landed where it did.
The finding also gives the field a mechanistic story for something that has been piling up empirically for a few years now: GLP-1 drugs keep showing benefits in places that have nothing obvious to do with weight. Cardiovascular outcomes, kidney disease, sleep apnoea, neuroinflammation models, alcohol use disorder research. If a single drug class is nudging the underlying ageing process, broad multi-system effects are not a surprise - they are the prediction. As de Cabo put it, most chronic diseases are deeply rooted in ageing, so a drug that slows it should look like it helps everything.
The caveats
There are several, and they are the usual suspects plus a couple specific to this study.
First, female mice only. The lifespan experiment was run in female animals, which is standard in much ageing research but leaves the male question open. Sex differences in both GLP-1 pharmacology and longevity interventions are well documented.
Second, mice are not people. Mouse lifespan interventions fail in humans often enough that nobody serious is claiming a human longevity effect from this paper. Chen and colleagues are explicit that clinical research in people - potentially including healthy older adults without obesity or diabetes - would be needed to know whether any of this translates. The ScienceDaily coverage of the paper notes a post-hoc analysis of the SLIM LIVER trial as one early human hint, with the clear caveat that far more work is required.
Third, dose translation is murky. The mice received sustained high-dose GLP-1 activation relative to their body size, and nobody knows what dose, duration or age of initiation would matter in humans - or what the trade-offs would be.
Fourth, and worth saying plainly for this community: nothing here is a reason to take semaglutide for longevity. The compound remains approved for diabetes and weight management indications, the longevity hypothesis is preclinical, and the grey-market version of this experiment carries exactly the risks we covered in this week’s TGA counterfeit story.
Where this fits
Stack this against the July epigenetic data and a pattern emerges: the mouse lifespan result is mechanistic support for the human biomarker signals, not just a curiosity. The Chen lab’s work gives ageing researchers a defined pathway to interrogate - GLP-1 receptor activation as a nutrient-sensing intervention - and nutrient sensing is one of the most productive frameworks in the field.
For the Australian research community following GLP-1s, the honest summary is this: the “these drugs do more than weight” thesis now has lifespan data in old mice, molecular ageing signatures in humans, and a plausible unifying mechanism. What it does not have is any human clinical evidence for longevity, and anyone claiming otherwise is selling something. Our semaglutide research page tracks the published data, and our July analysis of the epigenetic ageing findings is the natural companion read.
If you want to argue about the calorie restriction comparison - and people absolutely will - the Grey Highway Telegram group is the place for it.
Sources
Source: Feng, Y., Barthez, M., Wang, Y. et al. “Late-life semaglutide treatment slows ageing and extends lifespan in female mice.” Nature (2026). DOI: 10.1038/s41586-026-10940-7. nature.com/articles/s41586-026-10940-7
Source: ScienceDaily - Scientists find Ozempic may slow aging itself (September 2026), based on materials provided by the NIH Office of the Director. NIH support: grants R01AG063404, R01AG063389, R01AG082105.
Source: de Cabo, R. “Weight-loss drug slows ageing in female mice.” Nature News & Views (2026). nature.com/articles/d41586-026-02468-7
Disclaimer: This article is for educational and informational purposes only. It does not constitute medical advice, therapeutic recommendations, or endorsements of any compound. Grey Highway is a research-education community. We do not sell, supply, or promote the use of research compounds. Always consult a qualified healthcare professional regarding health decisions. For Australian regulatory information, visit the TGA website.