The TAME metformin aging trial won FDA protocol agreement back in 2015, yet more than a decade later it has not enrolled a single patient: an off-patent generic that no one finds profitable to fund

The TAME metformin aging trial won FDA protocol agreement back in 2015, yet more than a decade later it has not enrolled a single patient: an off-patent generic that no one finds profitable to fund

The TAME trial was designed by gerontologist Nir Barzilai and colleagues at Albert Einstein College of Medicine for more than 3000 people aged 65–79 across 14 sites in the United States. An analysis of the trial's fate, published on September 26, shows that more than a decade later it still has not secured the $45–75 million needed to launch.

Metformin is usually tested for its effect against a single disease. TAME tests a broader hypothesis: whether one drug can delay an entire cluster of major age-related conditions (heart attack, cancer, dementia, and death) when aging itself is treated as the target of intervention. In clinical trials this kind of target is called a composite endpoint, and in 2015 the FDA agreed to accept it as a basis for drug approval, the first time the agency had ever done so. The trial counts as a success if participants live longer without experiencing any of these four outcomes.

Metformin was chosen for the trial because of its decades-long safety record: the drug has been approved for type 2 diabetes since 1994, appears on the WHO Model List of Essential Medicines, and is taken by hundreds of millions of people worldwide.

That very safety record became the problem. Metformin is a generic, carries no patent, and costs pennies per tablet.

"The main obstacle to funding this trial is that metformin is a generic, and no pharmaceutical company expects to profit from it," Barzilai said in an interview with NPR.

Years were spent negotiating with the U.S. National Institutes of Health, which were reluctant to fund a trial with a nonstandard design. At one point a private donor nearly covered the budget and the organizers announced a start date. The donor lost the money, and the trial never began. The National Institute on Aging contributed roughly $5 million of the required $45–75 million.

While TAME waits for funding, another trial has already secured it: VITAL-H, which tests rapamycin, dapagliflozin, and semaglutide in 726 people aged 60–69 at the University of Texas Health Science Center at San Antonio. The $38 million contract was awarded by ARPA-H, a U.S. Department of Health and Human Services agency modeled on the military's DARPA and designed for high-risk projects that neither conventional grants nor the market are willing to support. All three VITAL-H drugs, unlike metformin, have commercial value: semaglutide is sold as part of one of the most profitable drug classes in the world.

Metformin lacks the one thing it would need to attract funding: the possibility for someone to profit from it. Barzilai puts the same irony differently: "This is something everyone will be able to afford." Universal affordability removes the only typical sponsor of trials at this scale, the pharmaceutical company that recoups its investment through exclusive sales. The same irony plays out in private life: an analysis of why billionaires do not fund cheap trials found that wealthy biohackers such as Sam Altman take metformin themselves, yet their money has never once covered the budget of a study that would prove metformin's benefit for everyone else.

Some researchers propose a solution: a one-time government investment of $70–100 million in reliable aging biomarkers, measurements that would let a drug's effect be assessed in 2–3 years instead of waiting six years for clinical outcomes. Infrastructure of this kind would reduce the cost of any future trial regardless of the drug's patent status.

The barrier once considered the hardest, regulatory agreement to recognize aging as a treatment target, was cleared more than a decade ago. The remaining barrier is purely financial: the same fate awaits any other cheap generic with comparable potential for as long as proving geroprotection depends on occasional donors rather than shared infrastructure.