The idea that a pill could slow aging has moved from science fiction into mainstream research funding. Two drugs sit at the center of the conversation: rapamycin, an immunosuppressant used to prevent transplant rejection, and metformin, a first-line diabetes medication. Both have promising mechanistic rationales. Neither is proven to extend healthy lifespan in humans.
The gap between preclinical excitement and human evidence is the entire story here. It is also the reason the field has largely resisted calling these drugs longevity treatments outside of trials.
Rapamycin: the strongest preclinical case
Rapamycin inhibits mTOR, a central regulator of cell growth and nutrient sensing. In mice, it is the most reliable lifespan-extending drug ever tested, producing modest but reproducible extensions even when started in middle age. The Interventions Testing Program run by the NIA has replicated these findings across multiple strains, giving them unusual weight.
In humans, most rapamycin data come from transplant medicine, where chronic use is associated with infections, mouth ulcers, elevated lipids, and impaired wound healing. Short intermittent dosing, as promoted by clinicians like Peter Attia and researchers exploring geroprotection, has a much thinner evidence base. A small trial from Novartis testing rapalogs in older adults reported improved vaccine responses, hinting at rejuvenation of immune function, but the results have not been robustly replicated in larger trials.
The mouse data on rapamycin are stronger than for almost any other candidate. The human data are almost entirely extrapolation.
The PEARL trial, a small placebo-controlled study of intermittent rapamycin in healthy adults, reported some improvements in body composition and quality of life measures with a favorable safety profile at low doses. It was underpowered to say anything definitive about aging outcomes.
Metformin and the TAME question
Metformin has been used safely by tens of millions of people with type 2 diabetes for decades. Retrospective analyses, most notably one led by Craig Currie, suggested that diabetics on metformin had lower all-cause mortality than non-diabetics not on the drug, an intriguing signal that spurred interest in geroprotection.
Nir Barzilai has spent years trying to launch TAME, the Targeting Aging with Metformin trial, to test whether metformin delays age-related diseases in non-diabetics. TAME is designed less to prove metformin works than to establish aging itself as a treatable regulatory endpoint, opening a path for future longevity drugs. Funding has been slow and the trial has yet to fully launch at scale.
More recent evidence has complicated the story. Metformin appears to blunt some of the muscular adaptations to resistance training in older adults, which matters if muscle preservation is itself geroprotective. And the mortality benefit in non-diabetics remains unproven.
The rest of the pipeline
Beyond these two, the field is exploring:
- Senolytics like dasatinib plus quercetin, aimed at clearing senescent cells; early human trials in idiopathic pulmonary fibrosis and diabetic kidney disease are suggestive but small
- NAD precursors such as nicotinamide riboside and NMN, which reliably raise NAD levels but have not shown clear functional benefits in rigorous trials
- GLP-1 agonists, whose cardiovascular and possible neurodegenerative benefits may amount to geroprotection by another name
- Acarbose and 17-alpha-estradiol, both of which extended lifespan in ITP mice
None of these have crossed the threshold from interesting to established.
How to think about it
If you are considering off-label use of these drugs, the honest framing is that you are participating in an uncontrolled experiment on yourself. Some of the drugs have well-characterized risks. Others have unknown ones at the doses and schedules being promoted. The people who do this responsibly work with a physician, monitor labs, and accept that the evidence base is thin.
The bottom line
Rapamycin and metformin are the most credible candidates for longevity drugs, but their human evidence lags far behind the mouse and mechanistic data. The pipeline is more interesting than proven. For now, exercise, sleep, and diet remain the geroprotective interventions with the strongest human track record.