Metformin is a cheap, generic type 2 diabetes drug that has become one of the most talked-about candidates for slowing human ageing. The theory is biologically plausible and the animal data are intriguing, but the honest summary is this: no trial has yet shown that metformin extends lifespan or healthspan in people who don’t have diabetes. The landmark study designed to answer that question — the TAME trial — has spent years waiting on funding rather than producing results.
Here is what the evidence for metformin for longevity actually supports, where it falls short, and why a muscle-building debate complicates the picture for otherwise healthy adults.
How metformin might affect ageing
Metformin is a biguanide that has been used for decades to lower blood glucose in type 2 diabetes. Its relevance to ageing comes from the pathways it touches, several of which overlap with interventions already linked to longer life in the lab, such as calorie restriction and exercise.
- AMPK activation. Metformin switches on AMP-activated protein kinase (AMPK), a cellular energy sensor. When AMPK is active, cells shift towards a conservation-and-repair mode: burning fat, improving insulin sensitivity and triggering autophagy, the process by which cells clear out damaged components.
- mTOR inhibition. AMPK activation indirectly dampens the mechanistic target of rapamycin (mTOR), a growth-and-nutrient-sensing pathway whose suppression is strongly tied to longevity in animals. Metformin also appears to inhibit mTOR through routes that don’t depend on AMPK.
- Mitochondrial and metabolic effects. Metformin partially inhibits mitochondrial complex I, reduces glucose production in the liver, and is associated with lower oxidative stress and chronic inflammation.
These are the same mechanisms that make the drug interesting to researchers exploring whether ageing itself can be treated as a modifiable process — a question at the heart of the emerging field. (For the bigger picture, see our guide to what longevity medicine actually is.)
What the animal evidence shows — and its limits
In simple organisms, the signal is real. In the nematode worm C. elegans, metformin has extended median lifespan substantially, driven by AMPK-related signalling, with effects comparable to dietary restriction.
In mammals, the picture is far muddier. Some rodent studies reported modest lifespan gains, but the most rigorous test is less flattering. The US National Institute on Aging’s Interventions Testing Program (ITP), which evaluates compounds across multiple independent labs in genetically diverse mice, found that metformin alone did not significantly extend lifespan at the dose tested. Rapamycin, by contrast, produced robust extension in the same programme. That contrast is a recurring theme in metformin’s longevity story: promising in theory, inconsistent in practice.
The human evidence is observational — and contested
Much of the excitement around metformin traces to a single 2014 UK observational study (Bannister and colleagues), which reported that people with type 2 diabetes taking metformin on its own had lower mortality than matched non-diabetic controls — a striking claim, since you would expect diabetics to fare worse.
That finding has not held up well. Critics identified serious confounding: in real-world prescribing, metformin is typically the first drug given at diagnosis, and patients are switched to or given additional drugs only as their disease worsens. Comparisons between drug groups therefore partly reflect disease severity rather than the drug’s intrinsic benefit. When researchers attempted to replicate the broad protective effect in an independent cohort, the result did not reproduce.
Beyond that study, large observational datasets associate metformin use with lower rates of some cancers, cardiovascular events and cognitive decline. But association is not causation. People prescribed metformin differ systematically from those who aren’t, and no amount of statistical adjustment fully removes that bias. The bottom line: the human data are suggestive, not conclusive, and they come almost entirely from people who already have diabetes or prediabetes.
The TAME trial: a test of the idea, not just the drug
The Targeting Aging with Metformin (TAME) trial was conceived to cut through this uncertainty. Led by Dr Nir Barzilai at the Albert Einstein College of Medicine, it was designed to enrol roughly 3,000 older adults (around ages 65–79) across multiple sites and follow them for about six years.
Its cleverest feature is the endpoint. Rather than testing whether metformin prevents one disease, TAME uses a composite outcome — the time until a first major age-related condition such as cardiovascular disease, cancer, dementia or death. The deeper goal is regulatory as much as scientific: to establish ageing as a target that drug trials can legitimately address, creating a template for future therapies.
The obstacle has been money. Because metformin is an inexpensive generic, no pharmaceutical company stands to profit, so the usual funding route is closed. The trial’s cost has been estimated in the tens of millions of dollars, and despite backing from the American Federation for Aging Research, philanthropists and a comparatively small allocation from the National Institute on Aging, it has struggled for years to secure full funding. As of this writing, TAME has not launched at scale, and until it reports, the central question about metformin and human ageing remains open.
Who has metformin actually been studied in?
Overwhelmingly, in people with type 2 diabetes or prediabetes. That matters, because a drug that improves outcomes in metabolically unhealthy people may do little — or even something unhelpful — in someone who is already lean, fit and insulin-sensitive. Evidence in metabolically healthy adults taking metformin purely for prevention is thin, which is precisely the gap TAME was meant to fill.
The exercise-blunting debate
For active people, there’s a genuine trade-off worth knowing about. In the double-blind, placebo-controlled MASTERS trial (2019), older adults who took metformin while doing supervised resistance training gained less muscle and lean mass than those on placebo. Mechanistically this fit expectations: metformin increased AMPK signalling and showed a trend towards blunting the mTORC1 signalling that drives muscle growth after training.
Other work has raised similar concerns about metformin dampening some of the cardiometabolic and mitochondrial improvements that come from aerobic exercise. The irony is pointed: the same AMPK/mTOR effects that make metformin attractive for longevity may partly cancel out the benefits of exercise, itself one of the best-evidenced longevity interventions. Given how strongly measures like grip strength and muscle mass predict long-term health, this is not a trivial consideration for someone who trains.
A note on responsible use
Metformin is a prescription medicine, not a supplement, and it is not approved anywhere as an anti-ageing drug. It commonly causes gastrointestinal side effects, can deplete vitamin B12 with long-term use, and — very rarely — carries a risk of lactic acidosis, particularly where kidney function is impaired. Taking it off-label for longevity is a decision to make with a clinician who can weigh your individual metabolic profile, monitor appropriately and account for any trade-off with exercise goals.
Frequently asked questions
Does metformin extend lifespan in humans?
There is no trial evidence that it does. The human data are observational and confounded, the most rigorous mouse study found no lifespan extension from metformin alone, and the trial designed to settle the question — TAME — has not yet reported.
What is the TAME trial and has it started?
TAME (Targeting Aging with Metformin) is a proposed large, multi-site trial using a composite age-related-disease endpoint to test whether metformin delays ageing. It has faced persistent funding difficulties and, as of this writing, has not launched at scale.
Can metformin interfere with exercise gains?
Possibly. The MASTERS trial found that metformin blunted muscle and lean-mass gains from resistance training in older adults, and other research suggests it may dampen some benefits of aerobic exercise. For people who train, that trade-off is worth discussing with a doctor.
Is metformin safe to take for longevity?
It is a well-characterised drug in diabetes, but it is not approved for ageing and carries side effects including GI upset and long-term B12 depletion. Any off-label use for longevity should be medically supervised.
The takeaway
Metformin sits at an awkward crossroads: a mechanistically credible, cheap and familiar drug whose longevity case rests on worm data, inconsistent mouse results and confounded human observations — with the one trial that could provide real answers still stalled for lack of funding. For people with diabetes it remains a valuable medicine. For healthy adults chasing a longer healthspan, the current evidence doesn’t justify treating it as a proven anti-ageing therapy, and the exercise-blunting data give active people a concrete reason for caution. The sensible stance is interest without overclaiming — exactly the posture the science supports today.