Skip to content
This GLP
← Research
Evidence

Ninety-two more days, in mice

Semaglutide started late in life extended median lifespan from 742 to 834 days in female mice — and beat a calorie-restricted group matched to how much less they ate.

Ruth Alvarez7 min read
Median lifespan, female micevehicle742 dayssemaglutide834 days92 days apart — and the comparison arm ate 24% less tooTreatment started at 20 months of age. No human data exists.

Nothing in this paper is a reason to buy a GLP-1, and it is going to be reported as one. What it is, is the first careful attempt to separate a drug’s effect on aging from the effect of simply eating less — and it is the outcome that the surrogate-marker literature keeps standing in for, measured directly, in animals.

What was done

Female mice of a single common laboratory strain were given daily semaglutide by injection starting at twenty months of age, which is late middle age for a mouse rather than the start of life. [1] Treatment continued for the rest of their lives.

Median lifespan came out at 742 days in the vehicle group and 834 days on semaglutide, compared by log-rank test. The gap of 92 days, about 12%, is this desk’s subtraction from those two published figures rather than a number the paper prints.

What it also found

Three months of treatment improved physiological function, attenuated recognized hallmarks of aging, and shifted nutrient-sensing pathways and conserved genetic regulators of aging in the direction calorie restriction shifts them.

The authors read the whole picture as semaglutide acting like a calorie-restriction mimetic, with some effects beyond what reduced intake explains — which is a mechanistic claim about mice, and a different kind of evidence from a biomarker moving in people.

Every reason to hold it loosely

Female mice only, one strain, one laboratory. Strain and sex both change lifespan results in this field routinely, and a finding in female C57BL/6 mice is a finding in female C57BL/6 mice until somebody repeats it elsewhere.

The schedule is not a human schedule either. These animals were injected daily; people take semaglutide weekly, at doses that do not map across species by any simple ratio. Drugs that extend rodent lifespan have a long history of not doing so in humans, and there is no registered trial asking this question in people.

What a reader should take from it

That the mechanism question is now genuinely open, and that it was mostly closed before this — the reasonable default was that any longevity effect of these drugs is a weight effect wearing a lab coat. This is real evidence against that default, in mice.

What it is not is a purchase. The drugs on this site are sold for weight, at prices set by sellers, to people who will take them for a year or two at most, and what happens outside a controlled study looks nothing like a mouse facility. A first-in-class result has to survive a long way before it means anything at a checkout — the same distance an early-phase weight figure has to travel.

Frequently asked

Does semaglutide extend lifespan?
In female mice of one strain, started at twenty months, median lifespan went from 742 to 834 days. Nothing equivalent has been measured in people, and no trial is asking.
Is it just because the mice ate less?
That is the question the study was built to answer. Semaglutide cut food intake by 24%, and a separate group was restricted by the same 24% — against it, the drug still did better on exploration, spatial memory and glucose control.
How much longer is 92 days for a mouse?
About 12% on the vehicle group's median. That percentage is arithmetic from the two published figures, not a number the paper reports.
Should this change what I do?
No. It was daily injection in mice at doses that do not translate across species, and rodent lifespan results have a long record of not carrying over to people.

Sources

  1. [1] Feng Y, et al. (2026). Late-life semaglutide treatment slows ageing and extends lifespan in female mice Nature. PMID 42686906

Where to get it

Best GLP-1 injections

Every injectable seller we can verify, with the price each one publishes and an honest read of what the trials measured.

Compare providers →

More in Evidence