Metformin, Anti-Aging and Cancer: What the Evidence Actually Shows

Metformin has been a mainstay diabetes drug for decades. It is cheap, generic and widely considered safe. In the last several years, it has also become one of the most talked-about "longevity" drugs online. Podcasts and social media credit it with slowing aging. They also credit it with cutting cancer risk. Some of that interest is well-founded. Metformin is genuinely one of the most-studied repurposed drugs in aging research. But the strongest, most rigorous data available tells a more complicated story than the online buzz suggests.

The aging claim: what NIA's own mouse studies found

The National Institute on Aging runs the most rigorous test for anti-aging drug claims. It is called the Interventions Testing Program (ITP). This multi-site study feeds candidate compounds to genetically diverse mice. It measures, directly, whether they live longer. Metformin has been through it.

NIA published its ITP results. Metformin alone was tested at 1,000 parts per million, starting at 9 months of age. It produced no increase in median or maximum lifespan in mice. That's not a fluke of one strain or lab. The ITP specifically uses three independent test sites and genetically diverse mice to rule that out.

The same cohort tested a combination of metformin and rapamycin together. That combination did extend lifespan in both male and female mice. Rapamycin, tested on its own across several other ITP cohorts, has repeatedly extended mouse lifespan by itself. So the pattern is this. Metformin alone showed nothing. But it performed well paired with a drug already known to work. That is a reasonable basis for research into combinations. But it doesn't support metformin itself as an anti-aging drug. At least, not in the best available animal data.

The human trial that would answer this — hasn't started

The definitive way to know whether metformin slows aging in people is a large, randomized human trial in humans. One has been designed for exactly that purpose. It is called TAME (Targeting Aging with Metformin). Researchers at Albert Einstein College of Medicine proposed it, with the American Federation for Aging Research. The goal is to track whether metformin delays age-related disease in thousands of older adults.

As of late August 2026, TAME still hasn't launched. It remains in a "raising funds" stage. That is roughly a decade after it was first proposed. Researchers involved have said publicly the core obstacle is money. Metformin is an old, cheap, generic drug. So no drug company has a commercial reason to fund the trial. That trial would settle whether metformin actually slows aging in humans. Until TAME or something like it runs to completion, claims that metformin "slows human aging" get ahead of the evidence. No test at that level of rigor has answered the question yet.

The cancer claim: a real signal, still under investigation — and a null result where it's been tested directly

Multiple observational studies compare outcomes across large populations, rather than randomly assigning a treatment. These studies found that people with diabetes who take metformin are diagnosed with cancer somewhat less often. They also tend to have somewhat better outcomes than similar patients on other diabetes medications. That pattern shows up across several cancer types and more than one dataset.

But studies like these can't fully rule out something else. The kind of patient prescribed metformin, not the drug itself, may explain part of the difference. A 2026 review published via the NIH's PubMed database concluded plainly that "evidence supporting a role for metformin in cancer prevention remains inconclusive," even while noting the signal is worth continued study. That study is happening right now. NCI's clinical trials registry lists active studies testing metformin as a cancer-prevention agent. These include trials in lung cancer, in oral cancer prevention for patients with precancerous lesions, and in breast cancer prevention for women at elevated risk. None has reported results showing that metformin prevents cancer in general.

It's also worth being clear about what happened when metformin was tested as a cancer treatment. This was a proper randomized trial, in people who already had cancer, not an observational comparison. The largest such trial, MA.32, randomized 3,649 patients with high-risk breast cancer. They got either metformin or a placebo for five years. It found no improvement in disease-free survival. The hazard ratio was 1.01. That is essentially a coin flip. Serious non-blood-related side effects were somewhat more common on metformin. That result is a useful reminder. When the earlier, more encouraging pattern from observational studies was tested properly, it did not hold up. Prevention and treatment are different questions. For treatment, the strongest trial available found nothing.

What this means in practice

Metformin is a real drug. It has a well-established safety profile for its approved use. It's also a legitimate, active subject in aging and cancer research. That part of its reputation is earned. The strongest available evidence does not yet support one leap: calling it a "proven anti-aging drug" or "proven cancer preventive." NIA's own gold-standard mouse studies found no lifespan benefit from metformin alone. The human trial built to test the aging question directly has never enrolled a single patient. And the one large randomized trial that tested it against existing cancer found no benefit at all.

None of this is a reason to avoid a medication prescribed for diabetes. Off-label use for other purposes is a decision to make with a physician. That physician should know the patient's full history. But the drug isn't, at this point, backed by the kind of definitive lifespan or cancer data that some online discussion implies.

Sources

• National Institute on Aging, Interventions Testing Program — Supported Interventions — https://www.nia.nih.gov/research/dab/interventions-testing-program-itp/supported-interventions
• American Federation for Aging Research, TAME — Targeting Aging with Metformin — https://www.afar.org/tame-trial
• ClinicalTrials.gov, NCT04931017 (metformin for lung cancer chemoprevention) — https://clinicaltrials.gov/study/NCT04931017
• ClinicalTrials.gov, NCT05237960 (metformin for oral cancer prevention) — https://clinicaltrials.gov/study/NCT05237960
• NIH/PubMed, 2026 review on metformin and cancer prevention evidence