Longevity · 12 min read · Published September 8, 2026

Peptides for Longevity: What the Evidence Actually Shows in 2026

A pillar review of every peptide and peptide-adjacent compound sold for aging, epitalon, thymalin, MOTS-c, humanin, SS-31, GHK-Cu, klotho, FOXO4-DRI, the growth-hormone axis, GLP-1 drugs, senolytics and rapamycin, with the best human study, the animal data and the regulatory status of each in one table.

Do any peptides actually extend lifespan?

In humans, none has been shown to. Lifespan data come from animals: epitalon raised maximum lifespan 12.3 percent in 54 mice (2003), klotho overexpression extended mouse life (2005), and rapamycin, not a peptide, extended mouse lifespan 9 to 14 percent from 600 days (2009). Human data are biomarkers, small pilots and one open-label Russian cohort.

Regulatory status, stated plainly: as of September 8, 2026 nothing on this page is approved by any major regulator for slowing aging or extending life. Some are approved for other things (elamipretide for Barth syndrome, semaglutide for obesity and diabetes, rapamycin for transplant rejection, tesamorelin for HIV lipodystrophy). Several were in the July 23 to 24, 2026 FDA compounding votes, covered in the FDA status tracker, which are not approvals. Evidence tiers used below: human RCT, human pilot or observational, animal, in vitro, none.

What is the evidence for epitalon and thymalin?

Two 2003 Khavinson-group papers. In mice, monthly epitalon from age 3 months until death (54 per group) left mean lifespan unchanged but raised maximum lifespan 12.3 percent and cut leukemia six-fold. In humans, an open-label study of 266 people over 60, followed 6 to 8 years, reported mortality 2.5-fold lower with thymalin plus epithalamin. Neither has been independently replicated.

Epitalon (Ala-Glu-Asp-Gly) is a synthetic tetrapeptide modeled on the pineal extract epithalamin; thymalin is a thymus extract. The mouse paper is careful: 'treatment with Epitalon did not influence food consumption, body weight or mean life span of mice', while the last 10 percent of survivors lived 13.3 percent longer and bone-marrow chromosome aberrations fell 17.1 percent (Biogerontology, 2003). Evidence tier: animal. The human cohort reported mortality reductions of 2.0 to 2.1-fold for thymalin, 1.6 to 1.8-fold for epithalamin and 2.5-fold for both, and 4.1-fold in a group treated annually for 6 years (Neuroendocrinology Letters, 2003). Evidence tier: human observational, open-label, single institute. It used epithalamin, the extract, not the synthetic epitalon in the vials, and a mortality reduction that size would be unprecedented for any intervention, which argues for replication rather than belief.

Mitochondrial peptides: MOTS-c, humanin and SS-31

MOTS-c prevented diet-induced obesity and insulin resistance in mice (Cell Metabolism, 2015) and has no completed controlled human trial. Humanin overexpression extended C. elegans lifespan, and 18 centenarian offspring had higher circulating humanin than 19 controls (Aging, 2020). SS-31 (elamipretide) alone has a human RCT: accelerated approval on September 19, 2025 for Barth syndrome, a rare mitochondrial disease.

The pattern across all three: elegant biology, strong rodent data, human evidence that is observational or confined to a rare disease. MOTS-c, a 16-amino-acid peptide encoded in mitochondrial DNA, activated AMPK and protected mice from age-dependent and high-fat-diet insulin resistance in its discovery paper (Cell Metabolism, 2015); evidence tier: animal. Our MOTS-c evidence review tracks the first registered human trial. Humanin showed lifespan extension in worms, healthspan gains in middle-aged mice given the HNG analogue twice weekly, and the centenarian-offspring association (Aging, 2020); evidence tier: animal and human observational. Elamipretide reached approval only in Barth syndrome, on a knee-extensor strength endpoint with a confirmatory trial required (FDA press release, September 19, 2025); evidence tier: human RCT in disease. Its failed Phase 3 in mitochondrial myopathy is covered in our SS-31 approval post.

GHK-Cu, klotho and FOXO4-DRI

GHK-Cu has topical skin data and a 2015 review claiming it modulates thousands of genes; no human aging trial exists. Klotho overexpression extended mouse lifespan (Science, 2005) and one dose improved memory in aged monkeys (2023); no human trial exists. FOXO4-DRI restored fitness, fur and kidney function in aged mice (Cell, 2017) and has never been given to a human.

GHK-Cu is a copper-binding tripeptide whose plasma level falls with age. The 2015 review behind most marketing lists collagen effects, wound healing in several species and cosmetic outcomes, and proposes that GHK 'is capable of up- and downregulating at least 4,000 human genes' (Biomed Research International, 2015); evidence tier: human topical trials for skin, none for aging. Klotho is the best-supported aging target in animals: overexpression extended mouse lifespan by suppressing insulin and IGF-1 signaling (Science, 2005), and a single low dose improved spatial memory in aged rhesus monkeys for weeks (Nature Aging, 2023); evidence tier: animal, primate. Our klotho explainer covers what is sold under the name. FOXO4-DRI pries p53 away from FOXO4 so senescent cells commit apoptosis; in 2017 it 'restored fitness, fur density, and renal function' in fast-aging and naturally aged mice (Cell, 2017); evidence tier: animal, no human exposure data.

Why growth hormone peptides run against the longevity biology

Because less GH signaling, not more, associates with longer life. Ames dwarf mice lacking GH live longer than normal mice. Adults with GH-receptor deficiency, followed 22 years, had one non-lethal cancer and no diabetes versus 17 and 5 percent in relatives (2011). A 2007 review of 18 GH trials in 220 elderly found small body-composition changes and more adverse events.

Sermorelin, ipamorelin, CJC-1295, MK-677 and tesamorelin all raise growth hormone and, downstream, IGF-1. The anti-aging pitch dates from the 1990s; the biology has moved the other way since. Brown-Borg's report that Ames dwarf mice, which lack GH, are exceptionally long-lived was the first demonstration that a single gene change could extend mammalian lifespan (Nature, 1996); evidence tier: animal. The Guevara-Aguirre cohort of roughly 100 people with GH-receptor deficiency, monitored 22 years, 'exhibited only one nonlethal malignancy and no cases of diabetes, in contrast to a prevalence of 17% for cancer and 5% for diabetes in control subjects' (Science Translational Medicine, 2011); evidence tier: human observational. GH given to healthy older adults, across 18 study populations, added 2.1 kg of lean mass, removed 2.1 kg of fat, and caused significantly more adverse effects, leading the reviewers to reject it as an anti-aging therapy (Annals of Internal Medicine, 2007); evidence tier: human RCT, pooled. The secretagogue peptides have never been tested for lifespan in any species.

Do GLP-1 drugs slow biological aging?

One trial suggests they might. A 2026 Nature Communications analysis of a 32-week randomized, placebo-controlled Phase 2b trial in adults with HIV-associated lipohypertrophy (45 semaglutide, 39 placebo) found semaglutide slowed PhenoAge by 4.9 years per year, GrimAge V2 by 2.3 and DunedinPACE by 9 percent. The analysis was post hoc, the cohort HIV-specific, and epigenetic clocks are surrogates, not outcomes.

This is the most-discussed longevity result of 2026 and the caveats are the authors' own: 'the post hoc design, modest sample size, HIV-specific cohort, and 32-week follow-up limit generalizability' (Nature Communications, 2026). Evidence tier: human RCT, exploratory endpoint. It matters here because semaglutide is a peptide, is approved, and has more randomized human data than everything above it combined. Whether clocks moving over 32 weeks means anything a person would notice is unknown, and a clock can move because fat and inflammation fell rather than because aging did. Tirzepatide and retatrutide are compared in our GLP-1 field guide.

Senolytics and rapamycin: the non-peptide benchmarks

Senolytics have one human result: in 9 patients with diabetic kidney disease given dasatinib 100 mg plus quercetin 1,000 mg for 3 days, senescent cells in fat and skin fell within 11 days (2019). Rapamycin extended mouse lifespan 9 to 14 percent from 600 days (2009); the 48-week PEARL trial (2025) found no change in its primary endpoint, visceral fat.

Neither is a peptide, but both are the benchmarks peptide vendors invoke. The Mayo Clinic pilot of dasatinib plus quercetin was open-label and tiny, and the first demonstration that senescent cell burden can fall in humans (EBioMedicine, 2019); evidence tier: human pilot. Rapamycin has the deepest animal record: the 2009 Nature study showed lifespan extension at three independent test sites in genetically heterogeneous mice even when feeding started at 600 days (Nature, 2009); evidence tier: animal. In people, everolimus improved influenza vaccine response by about 20 percent in elderly volunteers (Science Translational Medicine, 2014), and PEARL randomized healthy adults to placebo, 5 mg or 10 mg of compounded rapamycin weekly for 48 weeks: adverse events were similar across groups, visceral fat did not change, and lean mass and self-reported pain improved in women on 10 mg (Aging, 2025); evidence tier: human RCT, negative primary endpoint. How to read that trial is argued in the rapamycin dosing debate.

The bottom line

Sort the table by human evidence and the peptides marketed hardest for longevity land at the bottom: epitalon, thymalin, GHK-Cu, FOXO4-DRI and the GH secretagogues have no controlled human aging data at all. The compounds with real human trials are approved drugs being studied for a new purpose, semaglutide and rapamycin, plus elamipretide in one rare disease. The strongest animal lifespan results belong to genetic manipulations and to rapamycin, none of which comes in a vial sold as a peptide. That is where the evidence stood in September 2026. Related reading: NMN vs NR vs NAD and new compounds of 2026.