Metformin's Longevity Trial Has Not Enrolled a Single Person. The Exercise Trials Ran Anyway, and They Keep Finding the Same Thing.
Metformin has been prescribed for more than sixty years, costs a few dollars a month, and sits in the medicine cabinet of a large share of people who take longevity seriously. The case for it was built on a striking observation: in a 2014 analysis of UK primary care records, people with type 2 diabetes taking metformin appeared to outlive matched people who did not have diabetes at all. If a drug given for a disease makes patients live longer than healthy controls, the drug is doing something beyond treating the disease. That single finding launched the TAME trial, a six-year study in more than 3,000 older adults designed to establish whether metformin delays age-related disease in people without diabetes. TAME was designed in 2015. As of 2026 it has not enrolled anyone, because no funder will pay roughly seventy-five million dollars to study a generic drug nobody can patent. What has been studied in the meantime is what happens when you give metformin to people who are training. Three randomized double-blind trials have now done that, and all three found the same thing: the drug shrank the adaptation.
Key Takeaways: What the Metformin Aging Evidence Actually Shows
Metformin's longevity reputation rests largely on a retrospective study of UK primary care records published in Diabetes, Obesity and Metabolism in 2014, in which 78,241 people with type 2 diabetes started on metformin monotherapy had adjusted median survival roughly 15 percent longer than 90,463 matched people without diabetes, at a survival time ratio of 0.85 with a 95 percent confidence interval of 0.81 to 0.90. The TAME trial, designed to test metformin against delayed development of heart disease, cancer and dementia in more than 3,000 adults aged 65 to 79 across 14 research institutions over six years, has been fully designed and drug-ready since 2015, and the American Federation for Aging Research still lists it in 2026 as awaiting the funding and participants needed to launch. The MASTERS trial published in Aging Cell in 2019 gave 1,700 mg of metformin or placebo daily to 94 adults aged 65 and older through 14 weeks of supervised progressive resistance training, and lean body mass rose 1.95 percent on placebo against 0.41 percent on metformin, a between-group difference at p equals 0.003, while thigh muscle mass rose 3.90 percent on placebo against 0.45 percent on metformin. A double-blind trial published in Aging Cell in 2019 randomized 53 adults averaging 62 years of age to metformin or placebo through 12 weeks of aerobic training and reported that metformin attenuated the gains in VO2 max and whole-body insulin sensitivity and abolished the exercise-driven increase in skeletal muscle mitochondrial respiration. A 91-person double-blind trial published in The Journal of Clinical Endocrinology and Metabolism in 2026 gave 2,000 mg of metformin or placebo daily through 16 weeks of low-intensity or high-intensity exercise, and VO2 max rose significantly in both placebo arms and in neither metformin arm. The National Institute on Aging Interventions Testing Program tested metformin at 0.1 percent of the diet in genetically heterogeneous mice and reported no significant lifespan extension from metformin alone, though metformin combined with rapamycin extended lifespan robustly. A 40-month study in adult male cynomolgus monkeys published in Cell in 2024 reported slowed multi-tissue aging clocks and roughly a six-year regression in brain aging on metformin. MET-PREVENT, published in The Lancet Healthy Longevity in 2025, gave 500 mg of metformin three times daily or placebo for four months to 72 adults with a mean age of 80.4 and probable sarcopenia, and the adjusted difference in 4-metre walk speed was 0.001 metres per second at p equals 0.96, with 12 of 35 metformin participants admitted to hospital against 3 of 36 on placebo. No randomized trial has tested metformin as an aging intervention in healthy non-diabetic middle-aged men.
Why Is Metformin Considered an Anti-Aging Drug at All?
Metformin is considered an anti-aging candidate because it acts on the same nutrient-sensing machinery that caloric restriction acts on, and because people taking it for diabetes appeared in database studies to live unusually long lives. The mechanistic half of that argument is genuinely interesting. Metformin inhibits mitochondrial complex I, which lowers cellular energy charge and activates AMP-activated protein kinase, the enzyme the cell uses to signal that fuel is scarce. AMPK activation suppresses mTOR, increases autophagy, improves insulin sensitivity and shifts the cell away from growth and toward maintenance. That is, in outline, the same direction that caloric restriction, rapamycin and prolonged fasting push a cell. Since caloric restriction is the most reproducible lifespan intervention in laboratory animals, a cheap pill that mimics part of its signalling is an obvious thing to test. The population half of the argument came later and carried more weight with the public, because it involved humans rather than mice. It is also the half that has held up least well under scrutiny, and understanding why is the whole of the metformin question.
Did Metformin Really Make Diabetics Outlive Healthy People?
That is what the 2014 study reported, and the result is real, but the comparison it makes is not the comparison most people think it is. Bannister and colleagues used the UK Clinical Practice Research Datalink to identify 78,241 people with type 2 diabetes who progressed to first-line metformin monotherapy, 12,222 who progressed to sulphonylurea monotherapy, and 90,463 matched people without diabetes, across a combined 503,384 person-years of follow-up. Unadjusted mortality ran at 14.4 deaths per 1,000 person-years in the metformin group against 15.2 in its matched non-diabetic controls, and at 50.9 in the sulphonylurea group against 28.7 in its controls. After adjustment, median survival was 15 percent shorter in the matched non-diabetic controls than in the metformin group. Read carefully, though, this is a comparison between people who were diagnosed, monitored and medicated and people who simply were not in the healthcare system for that reason. Progressing to metformin monotherapy and staying on it selects for the healthiest possible diabetic: someone with good kidney function, no metformin intolerance, disease mild enough not to need a second agent, and enough clinical contact to be prescribed anything at all. The non-diabetic comparators were matched on recorded characteristics, not screened for undiagnosed disease. The sulphonylurea arm in the same study is the useful control here, since those patients died at nearly twice the rate of their own matched controls, which tells you the database is capable of showing diabetes as harmful when the drug is not metformin. What the study cannot do is separate the drug from the kind of patient who ends up on it and stays there.
The Database Result That Started the Metformin Longevity Case
Unadjusted deaths per 1,000 person-years in the retrospective UK Clinical Practice Research Datalink analysis published in Diabetes, Obesity and Metabolism in 2014, covering 503,384 person-years of follow-up. People with type 2 diabetes on metformin monotherapy died at a slightly lower rate than their matched non-diabetic controls, while people on sulphonylurea monotherapy died at nearly twice the rate of theirs. The sulphonylurea arm shows the dataset can detect diabetes as harmful, which is what makes the metformin comparison interesting. It also cannot rule out that metformin monotherapy selects the healthiest patients in the system.
What Happened When Metformin Was Added to Resistance Training?
The training worked and the metformin group got most of the muscle gain removed. The MASTERS trial, published in Aging Cell in 2019, was a randomized double-blind placebo-controlled multicenter study that recruited healthy adults aged 65 and older, randomized 109 of them to 1,700 mg of metformin daily or matching placebo, and put all of them through 14 weeks of supervised progressive resistance training. Ninety-four completed. The hypothesis going in was that metformin would enhance the response, since it lowers inflammation and older muscle is inflamed. The result went the other way. Lean body mass rose 1.95 percent on placebo, which was significant at p below 0.001, and 0.41 percent on metformin, which was not significant at p equals 0.218, with a between-group difference at p equals 0.003. Thigh muscle mass measured by CT rose 3.90 percent on placebo against 0.45 percent on metformin. Normal-density muscle area, which is the CT measure that best tracks contractile tissue rather than fat infiltration, rose 10.5 percent on placebo against 4.16 percent on metformin at p equals 0.001. Muscle density rose 4.13 percent against 2.49 percent at p equals 0.020. Strength outcomes moved in the same direction without reaching significance, with knee extension one-rep max at p equals 0.055, isometric strength at p equals 0.082 and peak power at p equals 0.064. The muscle biopsies pointed at a mechanism: metformin increased AMPK phosphorylation and blunted activation of RPS6, a downstream marker of the mTOR signalling that drives protein synthesis. The drug did what it is supposed to do, and hypertrophy is downstream of the thing it suppresses.
Fourteen Weeks of Lifting, With and Without 1,700 mg of Metformin
Percent change from baseline in the MASTERS trial published in Aging Cell in 2019, a randomized double-blind placebo-controlled multicenter study in 94 completers aged 65 and older who all performed 14 weeks of supervised progressive resistance training. Lean body mass rose significantly on placebo at p below 0.001 and not at all on metformin at p equals 0.218, with a between-group difference at p equals 0.003. The CT measures of thigh muscle mass, normal-density muscle area and muscle density all moved the same way, at p equals 0.001 and p equals 0.020 for the two density measures.
- Placebo + training
- Metformin + training
Does Metformin Blunt Aerobic Training Too?
Yes, in every randomized trial that has looked, including one published in 2026. Konopka and colleagues published a double-blind trial in Aging Cell in 2019 that randomized 53 older adults averaging 62 years of age to placebo or metformin through 12 weeks of aerobic exercise training. Exercise worked regardless of treatment for fat mass, HbA1c, fasting insulin and 24-hour glucose variability. But metformin attenuated the improvement in whole-body insulin sensitivity and in VO2 max, and it abrogated the exercise-driven increase in skeletal muscle mitochondrial respiration entirely. The response was strikingly variable, with some individuals improving and others getting worse, and the change in insulin sensitivity tracked the change in mitochondrial respiration. The authors closed by saying that additional work is needed to understand these responses before metformin is prescribed to slow aging. Seven years later the picture had not improved. Malin and colleagues published a 91-person double-blind trial in The Journal of Clinical Endocrinology and Metabolism in 2026 that randomized adults at risk for metabolic syndrome to low-intensity exercise at about 55 percent of VO2 max or high-intensity exercise at about 85 percent, five days a week for 16 weeks, each arm with 2,000 mg of metformin daily or placebo. VO2 max rose significantly in both placebo arms and in neither metformin arm. Metformin blunted the exercise-driven increase in insulin-stimulated dilation of the conduit arteries and in microvascular blood flow, and it attenuated the reductions in fasting glucose, endothelin-1 and TNF-alpha that the placebo arms achieved. Three independent groups, three different exercise modalities, and the same direction of effect each time.
What Metformin Took Off the Training Response
A schematic reading of three randomized double-blind trials that added metformin to supervised exercise: MASTERS in Aging Cell in 2019 with 94 older adults and 14 weeks of resistance training, Konopka in Aging Cell in 2019 with 53 adults averaging 62 years and 12 weeks of aerobic training, and Malin in The Journal of Clinical Endocrinology and Metabolism in 2026 with 91 adults at risk for metabolic syndrome and 16 weeks of low or high-intensity exercise. Grey marks adaptations the trials found reduced on metformin, amber marks a signal that trended without reaching significance, and dark grey marks outcomes metformin left alone. Bar heights indicate the consistency of the finding rather than an effect size.
Has Metformin Extended Lifespan in Animals?
Not in the most rigorous mouse programme that has tested it, and the primate evidence measures aging markers rather than lifespan. The National Institute on Aging Interventions Testing Program is the gold standard for this question because it runs the same protocol at three independent sites in genetically heterogeneous mice, which removes the single-strain and single-lab problems that make most lifespan claims unreliable. Its report published in Aging Cell in 2016 tested metformin at 0.1 percent of the diet and found no significant lifespan extension from metformin alone. Metformin combined with rapamycin did extend lifespan robustly, but historical comparison with earlier rapamycin-only studies makes it hard to say how much of that came from the metformin. In the same programme, acarbose and 17-alpha-estradiol both produced clear male-specific lifespan gains, so the null result for metformin is not a case of the assay failing to detect anything. The most encouraging recent animal data come from a 40-month study in adult male cynomolgus monkeys published in Cell in 2024, which reported that metformin slowed multi-tissue aging clocks built from transcriptomics, DNA methylation, plasma proteomics and metabolomics, with roughly a six-year regression in brain aging, preserved brain structure and better cognitive performance, mediated in part through activation of the antioxidant transcription factor Nrf2. That study is a genuine advance because primates are closer to humans than mice are. It also did not measure lifespan, ran in a small number of animals, and enrolled only males, which is worth noting given how often sex-specific effects turn up in this literature.
What Did the Trial in Frail Older Adults Find?
Nothing on the primary outcome, and a troubling safety signal. MET-PREVENT, published in The Lancet Healthy Longevity in 2025, was a double-blind randomized placebo-controlled trial run in Gateshead and Newcastle that enrolled 72 adults aged 65 and older with a 4-metre walk speed below 0.8 metres per second and probable sarcopenia defined by low grip strength or slow sit-to-stand time. Mean age was 80.4 years. Participants took 500 mg of metformin or matching placebo three times daily for four months. Mean 4-metre walk speed at four months was 0.57 metres per second on metformin and 0.58 on placebo, with an adjusted treatment effect of 0.001 metres per second and a 95 percent confidence interval running from minus 0.06 to 0.06, at p equals 0.96. That is a flat result, not an underpowered hint. Tolerability was the bigger finding. All 35 participants who received metformin had at least one adverse event, generating 108 events against 77 in the placebo group, and hospital admissions occurred in 12 of 35 metformin participants against 3 of 36 on placebo. The authors concluded that metformin did not improve walk speed and was poorly tolerated in this population. This was the first randomized trial to test metformin specifically as a geroprotector in a frail non-diabetic population, and it is the closest thing yet to the question TAME was built to answer.
What Are the Practical Downsides of Long-Term Metformin?
Vitamin B12 depletion is the documented one, gastrointestinal intolerance is the common one, and lactic acidosis is the rare one that defines the contraindications. Metformin interferes with calcium-dependent B12 absorption in the terminal ileum, and the effect compounds over years. In a three-year randomized trial of 165 participants published in Nutrients in 2019, 26 of 81 people on metformin finished below the normal B12 threshold of 221 pmol per litre against 10 of 84 on placebo, with holotranscobalamin II falling as well. B12 deficiency presents as fatigue, cognitive fog and peripheral neuropathy, all of which are easy to misread as aging in someone taking a drug specifically to counter aging. Anyone on long-term metformin should have B12 and ideally methylmalonic acid measured periodically. Gastrointestinal side effects affect a substantial minority and are the main reason people stop, which is why extended-release formulations and slow titration exist. Lactic acidosis is rare but serious, and it is why metformin is contraindicated below an estimated glomerular filtration rate of 30 millilitres per minute per 1.73 square metres and generally held around acute illness, dehydration, iodinated contrast imaging and heavy alcohol use. For a man who trains, the exercise-blunting data belong in this list too. The trials that documented it used standard clinical doses of 1,700 to 2,000 mg daily in older or metabolically at-risk adults, and none of them tested whether a lower dose, intermittent dosing or dosing far from training sessions would preserve the adaptation. Nobody has run that study, so the question of timing remains open rather than answered.
Why Has the TAME Trial Never Started?
Because metformin is generic, and nobody stands to earn back the roughly seventy-five million dollars the trial would cost. TAME, short for Targeting Aging with Metformin, was designed by Nir Barzilai and colleagues and endorsed by the FDA in 2015 as a workable template for testing aging itself as an indication, using a composite endpoint of cardiovascular disease, cancer, dementia and death rather than any single disease. The design calls for more than 3,000 adults aged 65 to 79 followed for six years across 14 research institutions, coordinated by Wake Forest University School of Medicine. The American Federation for Aging Research describes the protocol as expertly prepared and the drug supply as ready, and continues to solicit philanthropic donors to launch it. That was the status in 2016 and it is still the status in 2026. The regulatory significance of TAME was always larger than metformin, since a positive composite result would establish that aging can be a trial endpoint at all, which is why the field keeps returning to it. In the absence of TAME, the largest ongoing randomized test of metformin in people without diabetes is VA-IMPACT, a Veterans Affairs trial that began enrolling in April 2023, targets roughly 7,410 participants with prediabetes and established atherosclerotic cardiovascular disease, and estimates completion in September 2029. VA-IMPACT will answer a cardiovascular question in a high-risk population rather than an aging question in a healthy one, but it is the trial that will actually report.
Planned Participants Against Participants Who Actually Enrolled
The two largest metformin studies relevant to aging have contributed no completed participants to the literature. TAME was designed in 2015 for more than 3,000 adults aged 65 to 79 across 14 institutions and has never enrolled anyone, because a generic drug attracts no commercial sponsor for a study of that size. VA-IMPACT began recruiting in April 2023 toward a target of roughly 7,410 veterans with prediabetes and established cardiovascular disease, and estimates completion in September 2029, so its enrolled figure is shown as zero completed rather than zero recruited. Everything the field currently knows about metformin and aging in randomized form comes from the four small trials at the bottom of this chart.
- Planned or randomized
- Completed with published results
How Should a Man Read the Metformin Evidence Right Now?
The mechanism is plausible, the human aging evidence is observational and confounded, and the randomized evidence that does exist mostly documents metformin working against training. Those three facts frame the decision. On the supportive side, metformin has an exceptional six-decade safety record, activates a pathway that overlaps meaningfully with caloric restriction, and produced encouraging multi-omic aging-clock results in male monkeys over 40 months. For a man with prediabetes, meaningful insulin resistance or type 2 diabetes, metformin is a first-line drug with strong evidence for its actual indication, and that case does not depend on the longevity argument at all. Against it, the survival finding that started the enthusiasm compares medicated diabetics to unscreened non-diabetics and cannot separate the drug from the patient. The most rigorous mouse lifespan programme found nothing for metformin alone. The only randomized geroprotection trial in frail older adults found no effect on walk speed and worse tolerability. And three separate randomized trials adding metformin to supervised exercise found the training adaptation reduced, covering hypertrophy, VO2 max, mitochondrial respiration and vascular insulin sensitivity. That last point matters more for a man in his forties or fifties than any of the others, because resistance training and cardiorespiratory fitness are the two interventions with the strongest evidence for preserving function with age, and the metformin trials suggest the drug takes a share of both. For a metabolically healthy man who lifts and does cardio, the honest reading is that metformin is a drug taken to slow aging that has been shown to interfere with the two things known to slow aging, and that the trial capable of settling whether it does anything else has been sitting fully designed and unfunded for eleven years.
The Metformin Aging Record on One Axis
Every strand of the metformin aging case arranged as a single schematic map. Orange marks what the evidence supports, amber marks findings that are real but narrower than they sound, slate marks documented downsides, and dark grey marks claims the evidence does not carry. Metformin is a good drug for the condition it treats. As an aging intervention for a metabolically healthy man who trains, the randomized record currently shows more interference than benefit, and the trial designed to settle the question has been sitting unfunded since 2015.
Frequently Asked Questions
Have more questions? Contact us or explore our full peptide intelligence platform.
Related Articles
Does Telomere Length Determine Lifespan? Short Telomeres Track With Earlier Death. The Genetics Say Longer Ones Raise Cancer Risk.
Does telomere length determine lifespan? Short telomeres track with earlier death, yet genetic studies show longer ones raise the risk of several cancers.
Low-Dose Aspirin Added 8 Percent to Male Mouse Lifespan. In 19,114 Healthy Older Adults, It Added No Healthy Years and Deaths Ran Higher.
Does low-dose aspirin help longevity? It extended male mouse lifespan 8%, but the ASPREE trial in 19,114 healthy older adults found no gain and more bleeding.
Is NAD+ a Peptide? No. It Is a Coenzyme Built From Two Nucleotides, and the Peptide Label Comes From the Clinic, Not the Chemistry
Is NAD+ a peptide? No. NAD+ is a coenzyme built from two nucleotides. What NAD+ is used for, what IV NAD+ studies found, and how it differs from SS-31.