Longevity drugs

What a geroprotector is#

A geroprotector is a drug intended to slow biological aging itself, and so delay many age-related diseases at once, rather than treat one disease. The idea follows from the science of aging: because age is the dominant risk factor for heart disease, cancer, and dementia, slowing aging should push all of them back together.

None is approved, for three reasons. Regulators approve drugs to treat diseases, and aging is not classed as one. Proving that a drug delays death or several diseases in healthy people would take thousands of participants and many years. And the drugs with the best animal data are cheap generics, so companies have little incentive to fund such trials. The TAME trial was designed to get around this by testing metformin against a composite of age-related diseases in 3,000 adults aged 65 to 79; as of October 2026 it has produced no results. The nearest thing to an approval is in dogs, where the FDA has accepted effectiveness and safety data for Loyal's LOY-002, with a conditional approval application expected in 2027.

Risk and benefit when healthy#

The arithmetic of prevention is unforgiving. A healthy person has the least to gain from a drug in absolute terms and as much to lose from its side effects, so a preventive drug must be very safe to come out ahead (see absolute and relative risk). For geroprotectors the benefit in humans is unknown, while the harms are partly known.

That suggests an order of operations:

1. The habits with the largest effects. Not smoking, regular training, enough sleep, a mostly whole-food diet. See Exercise, Nutrition, and Sleep.
2. Proven medicines, used early and aggressively. The optimal plan treats risk factors to the targets prevention-focused physicians use, starting in your 30s or 40s rather than waiting for a 10-year risk score to climb: ApoB under about 60 mg/dL, blood pressure under 120/80, and HbA1c under about 5.5%, the range where heart risk was lowest in a large cohort. These drugs have hard outcomes behind them Strong. Each 39 mg/dL (1 mmol/L) lower LDL cut major vascular events by 22% and deaths by 10% across 26 statin trials. Each 10 mm Hg lower systolic pressure cut major cardiovascular events by 20% and deaths by 13% across 123 trials, and in SPRINT a systolic target under 120 cut deaths by 27%. Metformin cut progression from prediabetes to diabetes by 31% (lifestyle change, 58%), and semaglutide cut heart attacks and strokes in obesity with heart disease. See Heart and blood vessels and Metabolic health.
3. Experimental geroprotectors, if at all. Only after the first two are in place, with monitoring, and with the understanding that you are running an experiment on yourself.

Guideline minimum: the 2026 ACC/AHA cholesterol guideline sets LDL goals of under 100, 70, or 55 mg/dL by 10-year risk and no numeric goal at low risk, and the 2025 blood pressure guideline sets a goal of under 130/80, encouraging under 120 systolic at higher risk.

Someone with high blood pressure or an ApoB of 110 who takes rapamycin instead of a proven drug has traded a large, proven benefit for a speculative one.

The full ITP results#

The National Institute on Aging's Interventions Testing Program tests compounds in genetically diverse mice of both sexes, at three sites (the Jackson Laboratory, the University of Michigan, and UT Health San Antonio), and publishes every result. It is the most rigorous mammalian lifespan test available. The raw data are in the public ITP database. Percentages are changes in median lifespan unless noted.

Compounds that extended lifespan#

Compounds with no lifespan benefit#

Reading the results#

• Most wins are male-only. The program's own authors describe a "strong male bias" in the drugs found so far. Aged females on canagliflozin had about 20-fold higher blood levels than young males, one clue to why.
• Popular supplements failed. Resveratrol, nicotinamide riboside (an NAD+ precursor), fisetin, curcumin, fish oil, and alpha-ketoglutarate did nothing.
• Small effects are fragile. Three compounds that worked once did not hold up at other doses or starting ages.
• The standard test may miss early effects. A 2024 reanalysis using a test more sensitive to deaths before midlife found small benefits for metformin, enalapril, and 17-DMAG in males and green tea extract and CAPE in females. These are modest, early effects, not lifespan extension.
• Mice are a first filter, not proof. See animal and mouse studies.

Rapamycin#

Rapamycin (generic name sirolimus) is an FDA-approved immunosuppressant used to prevent kidney transplant rejection. It is the drug most longevity physicians would name if forced to choose a geroprotector. Weak

How it works#

Rapamycin blocks mTOR, a protein complex that senses nutrients and growth signals and tells cells whether to grow or to conserve and recycle. mTOR sits on the "deregulated nutrient sensing" hallmark and influences autophagy, the cell's recycling system. It works through two complexes: rapamycin mainly blocks mTORC1, but long, continuous exposure also blocks mTORC2, which in mice causes insulin resistance separately from the lifespan benefit. That finding is the main rationale for the weekly dosing used off-label, though it has not been tested head to head in people.

The mouse evidence#

The mouse case is the strongest for any drug. In the ITP, rapamycin started at about 20 months of age, late middle age for a mouse, still extended lifespan in both sexes. At higher doses it raised median lifespan by 23% in males and 26% in females. Alternating months on and off also worked, and in a separate lab a three-month course in middle-aged mice increased life expectancy by up to 60%, although one dose in females shifted cancers toward aggressive blood cancers.

The human evidence#

Human data are thin and mostly measure surrogates.

• Immune function. In older adults, the related drug everolimus improved the response to flu vaccine by about 20%, and a low-dose combination reduced reported infections in a follow-up trial.
• A failed Phase 3. The company behind that work, resTORbio, then tested RTB101, a different mTOR inhibitor, in 1,024 adults aged 65 and older. It did not reduce respiratory illness (26% versus 25% on placebo).
• PEARL. The first randomized trial in healthy, normally aging adults, run by the telehealth company AgelessRx and funded by donations, gave placebo or 5 or 10 mg of compounded rapamycin weekly for 48 weeks to adults 50 to 85. Among the 114 who finished, side effects were similar across groups, and the primary outcome (visceral fat) did not change. Women on the higher dose gained some lean mass and reported less pain; with many outcomes tested, these could be chance.
• The 2024 systematic review. Across 19 human studies, rapamycin and related drugs improved some immune, cardiovascular, and skin measures, with no effect on endocrine, muscle, or brain measures. Healthy people had no serious side effects attributed to the drugs, but people with age-related diseases had more infections and higher cholesterol and triglycerides.

How people take it off-label#

There is no validated longevity dose. In a survey of 333 off-label users, almost all took it once a week, and 6 mg was the most common weekly dose. Mouth ulcers were the only condition significantly more common in users than in non-users. Surveys of self-selected enthusiasts cannot detect rarer or slower harms.

Compounded versus generic#

PEARL used compounded rapamycin, and partway through, researchers found it produced about one-third the blood level of commercial sirolimus. Its 10 mg dose was roughly equivalent to 3.3 mg of the generic drug. Doses from compounding pharmacies and generic tablets are not interchangeable, so know which one you are getting.

Side effects and interactions#

At transplant doses, sirolimus carries a boxed warning for infection and lymphoma, and its label lists high cholesterol and triglycerides, impaired wound healing, mouth sores (stomatitis), protein in the urine, lung inflammation, and male infertility. Weekly low doses appear much better tolerated, but long-term data do not exist.

Metformin and acarbose#

Metformin#

Metformin is a first-line diabetes drug, safe and cheap, and for a decade it was the best-known longevity candidate. The case has weakened. Weak

• Mice. Alone, it did not extend lifespan in the ITP; combined with rapamycin it did.
• Observational data. A 2014 study suggested people with diabetes on metformin outlived matched people without diabetes. A Danish registry analysis that included twins found the opposite: metformin-treated diabetes carried 52% higher mortality than no diabetes.
• Exercise. In randomized trials in older adults, metformin blunted gains in fitness and insulin sensitivity from 12 weeks of aerobic training and reduced muscle gained from 14 weeks of strength training.
• TAME. The trial meant to settle the question has no results as of October 2026.

Metformin remains a good drug for type 2 diabetes and prediabetes; see Metabolic health. For a healthy person who exercises, the case is hard to make.

Acarbose#

Acarbose slows the breakdown of starch in the gut, flattening the glucose rise after meals; it is taken with the first bite of each meal. In the ITP it raised median lifespan 22% in males and 5% in females, even when started late. Weak

The best human outcome trial is sobering. In 6,522 Chinese patients with heart disease and impaired glucose tolerance followed for five years, acarbose did not reduce cardiovascular events (14% versus 15%), though it cut new diabetes by 18%. Gas, bloating, and diarrhea are common and were the main reason people stopped.

SGLT2 inhibitors and GLP-1 drugs#

SGLT2 inhibitors#

SGLT2 inhibitors (empagliflozin, dapagliflozin, canagliflozin) make the kidneys excrete glucose. They have strong outcome data in type 2 diabetes, heart failure, and chronic kidney disease, covered in Metabolic health. Strong (for those conditions)

As geroprotectors the evidence is mouse-only. Canagliflozin raised male median lifespan 14% with no benefit in females, and when started late it shortened female lifespan. There are no human aging trials. Weak (for aging in healthy people)

The label risks are real: genital yeast and urinary infections, dehydration and kidney strain (especially in older adults on diuretics), ketoacidosis that can occur with normal glucose, rare Fournier's gangrene, and low blood sugar with insulin or sulfonylureas.

GLP-1 drugs as geroprotectors#

GLP-1 receptor agonists such as semaglutide and tirzepatide are the most important new metabolic drugs in a generation; their clinical use is covered in Metabolic health. In 17,604 adults with obesity and heart disease, semaglutide cut heart attacks, strokes, and cardiovascular deaths from 8.0% to 6.5% over about three years. Strong (for obesity with heart disease)

Whether they slow aging itself is unknown. A veterans' cohort of 215,970 GLP-1 users with diabetes linked them to lower rates of dementia, addiction, seizures, and infections, alongside more gastrointestinal problems, pancreatitis, and kidney stones. But in two trials of 3,808 people with early Alzheimer's, oral semaglutide did not slow decline despite improving biomarkers. Most benefits likely flow from weight loss and better metabolism, which lean, healthy people do not need. Some telehealth companies now sell low-dose "microdosing" GLP-1 programs for longevity; no trial has tested that use. Weak (as a geroprotector in lean, healthy people)

Senolytics and other candidates#

17α-estradiol#

A weak, "non-feminizing" form of estrogen, 17α-estradiol is one of the ITP's most reproducible results, extending male lifespan even when started late, with no effect in females. There are no human longevity trials. Weak For menopausal hormone therapy, see Hormones.

Senolytics#

Senolytics are drugs meant to kill senescent cells, the damaged "zombie" cells that stop dividing and release inflammatory signals. The best-studied pair is dasatinib, a leukemia drug, plus quercetin, a plant compound. Weak

• Pilots. Small studies in lung fibrosis and in 12 older adults at risk of Alzheimer's found the regimen feasible and tolerable but were too small or uncontrolled to show benefit.
• The first real test. A randomized trial in 60 postmenopausal women missed its primary bone endpoint; benefits appeared only in an exploratory subgroup with the most senescent cells.
• Fisetin, the supplement senolytic, did not extend lifespan in the ITP. Small human pilots are under way.

Dasatinib is a cancer drug whose label warns of low blood counts, bleeding, and fluid around the lungs; it is not a supplement.

NAD+ precursors#

Nicotinamide riboside and NMN reliably raise NAD+, a molecule involved in energy metabolism that falls with age, but nicotinamide riboside did not extend lifespan in the ITP, and human trials have shown little functional benefit. Injectable NAD+ is sold by telehealth companies from about $79 a month (2026). Weak Details are in Supplements.

Low-dose naltrexone and methylene blue#

Low-dose naltrexone (LDN), a fraction of the dose used for addiction, is widely prescribed off-label for pain, fatigue, and "inflammation." Two randomized trials in women with fibromyalgia, one of its most common uses, found no pain benefit at 12 weeks and no meaningful benefit over 12 months. There are no aging data. Negative (for fibromyalgia pain)

Methylene blue, a dye used to treat a rare blood disorder, is marketed for mitochondria and cognition. In the ITP it did not change median lifespan. Its prescription label carries a boxed warning for serotonin syndrome, which can be fatal, when combined with SSRI or SNRI antidepressants, MAO inhibitors, or opioids. Negative (for lifespan in mice)

Plasma, reprogramming, and gene therapy#

Young plasma and plasma exchange#

The idea comes from parabiosis, experiments that joined the circulations of old and young mice and rejuvenated some tissues in the old ones. Clinics began selling infusions of young donors' plasma. In 2019 the FDA warned there is "no proven clinical benefit" and listed the risks: infections, severe allergic reactions, lung injury, and fluid overload. Negative

Therapeutic plasma exchange goes the other way: it removes your plasma and replaces it with albumin, a protein solution, on the theory that old blood carries harmful factors. The evidence is early. Weak

• In 347 people with mild to moderate Alzheimer's, plasma exchange slowed decline in daily activities, with a borderline result for cognition and benefit limited to the moderate stage.
• In 42 healthy adults over 50, biweekly exchange, especially with immunoglobulin, improved epigenetic clock scores. Assignment was by entry date rather than true randomization, the trial was only single-blinded, and clocks are unvalidated surrogates.
• Removing plasma without replacing it, as in plasma donation, did not rejuvenate clock scores and may have worsened some.

Partial reprogramming#

The four Yamanaka factors can turn adult cells back into stem cells, but full reprogramming erases a cell's identity and can cause tumors. Partial reprogramming switches the factors on briefly, aiming to restore youthful patterns of gene activity without erasing identity. Weak

The first human data arrived in October 2026. Life Biosciences' ER-100 injects a gene therapy carrying three of the factors into the eye, switched on for eight weeks with the antibiotic doxycycline. In the Phase 1 trial for glaucoma and a form of optic nerve stroke, three people with glaucoma had no serious side effects through day 56, and two had visual-field improvements. It is open-label, uncontrolled, and at its lowest dose.

Other well-funded companies are earlier. Altos Labs describes cell rejuvenation research and lists no clinical program on its website. NewLimit describes epigenetic reprogramming medicines in development, with no human trial listed. Retro Biosciences has an autophagy drug in a Phase 1 trial and is developing replacement blood and brain immune cells.

Offshore gene therapy#

Some companies sell gene therapies for "longevity" outside FDA oversight. The best-known example is a follistatin gene therapy offered in Próspera, a special economic zone in Honduras, reported in 2023. Gene therapies can trigger immune reactions, may not be reversible, and have long-term risks no one has measured in these products. Data from such clinics are not independently verified. Weak

Peptides and the gray market#

Injectable peptides are short chains of amino acids sold for healing, fat loss, growth hormone release, and "anti-aging." Common names include BPC-157, TB-500 (a fragment of thymosin beta-4), CJC-1295, ipamorelin, GHK-Cu, MOTS-c, and epitalon. Most are sold online as "research chemicals" or through compounding pharmacies. None is FDA-approved, and BPC-157, the most popular, rests almost entirely on animal studies. Weak

The FDA's list of compounding substances that may present significant safety risks explains why regulators worry. In 2026 BPC-157 and other peptides moved off its high-risk category after the groups that nominated them withdrew, but the agency's page still lists concerns: immune reactions, impurities, and little or no human safety data. It notes serious events including death after intravenous ipamorelin, and case reports of melanoma with melanotan II. As of October 2026 their legal status for compounding remains unsettled.

Growth hormone and related hormones are covered in Hormones.

Getting and deciding#

How people get these drugs#

Doctors may legally prescribe approved drugs such as sirolimus, metformin, acarbose, and SGLT2 inhibitors off-label. Most people who take them for longevity use telehealth services, which prescribe after an online intake and, often, lab tests. As of October 2026, Healthspan lists rapamycin at about $64 a month, metformin $27, acarbose $25, and an SGLT2 protocol $99, and AgelessRx lists rapamycin from about $65 a month and metformin from $25. Both also sell less-supported products such as methylene blue, LDN, or NAD+ injections.

Convenience is not the same as care. A physician who follows you over years can check baseline labs, watch for rising lipids or glucose, catch interactions with your other drugs, and tell you to stop. Before starting, ask:

• Is this generic or compounded, and what dose does that equal?
• Which labs will be checked before starting and how often after?
• Who handles side effects, and what are the stopping rules?
• Does it interact with your other medicines, and what changes around surgery or vaccines?

For building that relationship, see Building your care team.

A decision framework#

Action plan#