mTOR, autophagy, and the longevity levers that act on the same biology

TL;DR

• Does rapamycin slow aging in humans? Not demonstrated - but it has the strongest animal lifespan record in geroscience, replicated by the NIA Interventions Testing Program even when treatment starts in mid-to-late life.
• PEARL is the one substantial human trial - 48 weeks, 114 adults, primarily a safety study: null on its primary endpoint, with sex-specific secondary signals.
• A 2026 Oxford study added a human mechanism signal, not an outcome - nine older men, four months, lower markers of DNA damage and senescence in immune cells on rapamycin versus placebo. It shows the drug engages the biology in people; it does not show slower ageing.
• Nobody is funding the trial that would settle it. Rapamycin has been off-patent for years, so no commercial sponsor has a reason to run a Phase 3.
• The field's only approval is mitochondrial, not mTOR - the FDA approved SS-31 (elamipretide) on 19 September 2025, for Barth syndrome, not for ageing.
• MOTS-c has no completed interventional human trial; the Phase 2a MOTS-MET study is recruiting now, and it is prohibited in sport under WADA rules.
• Rapamycin is prescription-only, off-label for this, immunosuppressive by design, and not a Vivre Labs compound.
• Where Vivre Labs fits: access to SS-31, MOTS-c, epithalon and NAD, each with its evidence tier stated plainly, starting from a baseline panel. See Where Vivre Labs fits.

Educational content, not medical advice. No doses or dosing schedules appear. Rapamycin is a prescription medicine; any longevity use is off-label and requires physician supervision. Rapamycin is not a Vivre Labs compound.

Updated 28 September 2026: adds the January 2026 Oxford experimental-medicine study of rapamycin and immune-cell senescence markers in older men, sized and scoped as a mechanism signal rather than an efficacy result.

Updated September 2026: adds SS-31 (elamipretide) - FDA-approved in September 2025 and the clearest demonstration that this category produces evidence when a trial gets funded - and MOTS-c, with its first interventional human trial now recruiting. An earlier September 2026 update broadened the page from a single-compound verdict into the wider set of levers acting on the same maintenance-and-autophagy biology.

---

Rapamycin occupies an unusual position. Most compounds discussed in longevity circles have a mechanism, a few small studies and a great deal of enthusiasm. Rapamycin has something rarer: a genuinely robust, independently replicated animal result that serious gerontologists have spent two decades trying and failing to explain away. That result is worth understanding properly, because the pathway it points at - nutrient sensing, mTOR, autophagy, cellular maintenance - is the most productive hypothesis in aging biology right now, and rapamycin is not the only thing acting on it.

"Most studied" is routinely mistaken for "furthest along." mTOR inhibition is the most studied hypothesis in this field; it is not where the field has actually produced a licensed medicine. That distinction belongs to a mitochondrial peptide: SS-31 (elamipretide), which the FDA approved in September 2025 - mechanism, randomised placebo-controlled trial, advisory committee, approval, the entire distance. That approval is for a rare mitochondrial disease and not for ageing, and this page is careful about the difference throughout. But it happened, which is more than mTOR inhibition can say. Rapamycin, for all its animal data, is still prescription-only, off-label for this purpose, and carrying a mechanism that makes supervision non-negotiable.

So this page is not "rapamycin, plus some alternatives." It is a map of one biology - maintenance, recycling and mitochondrial quality - approached from two ends. The mTOR end has the deepest animal literature and the most open human question. The mitochondrial end has fewer decades behind it and one approved drug in front of it.

What mTOR does, and why it became the central longevity hypothesis

mTOR - mechanistic target of rapamycin - is a protein kinase that sits at the centre of how a cell decides between two states. It is a sensor: it integrates signals about nutrient availability (amino acids in particular), growth factors including insulin and IGF-1, cellular energy status and stress. When those signals say conditions are good, mTOR is active and the cell shifts toward growth and synthesis - building proteins, making lipids, dividing. When conditions are poor, mTOR activity falls and the cell shifts toward maintenance: slowing synthesis, recycling damaged components, conserving resources.

The central maintenance process it governs is autophagy - the cell's system for degrading and recycling its own damaged proteins and organelles. Active mTOR suppresses autophagy. Inhibiting mTOR releases that suppression.

That is the mechanistic rationale, and it is a strong one. Accumulated cellular damage is a plausible common substrate for several age-related pathologies; autophagy is the machinery that clears it; nutrient-sensing pathways converge on mTOR, which also links the whole thing to the caloric restriction literature - the oldest and most reproducible lifespan intervention in biology. Rapamycin, a macrolide isolated from soil bacteria on Rapa Nui and later developed as an immunosuppressant, is a direct inhibitor of mTOR complex 1.

Mechanism tells you where to look, not what you will find - this field has a long history of coherent mechanisms that did not survive contact with a randomised trial. In this case the animal data followed.

The animal record - strong, replicated, and the reason everyone is interested

Rapamycin or mTOR inhibition extends lifespan in multiple model organisms - yeast, nematodes, fruit flies and mice. In mice, the finding has been reproduced repeatedly, and crucially by the NIA Interventions Testing Program (ITP), a programme designed specifically to address the reproducibility problem in this field. The ITP tests candidate compounds simultaneously at multiple independent sites, in genetically heterogeneous mice, under protocols agreed in advance. It exists because single-laboratory lifespan results have a long history of not replicating. Most compounds submitted to it fail. Rapamycin did not.

Two further features explain why interest has persisted rather than faded.

The effect appears when treatment starts late. Lifespan extension has been reported in mice begun on rapamycin in mid-to-late life, not only from weaning. Most interventions that work in model organisms require lifelong exposure, which makes them biologically interesting and practically irrelevant. A late-start effect is what made this a translational question rather than a curiosity.

The effect is not confined to lifespan. Reported benefits across the rodent literature span cardiac function, immune function, cognition and several tissue-level ageing measures - a pattern consistent with an intervention acting on something general rather than on a single failure mode.

This is widely regarded as the most robust lifespan-extension result in geroscience, and it deserves to be stated plainly.

The human record - what PEARL actually did

The substantial entry here is PEARL - the Participatory Evaluation of Aging with Rapamycin for Longevity trial (NCT04488601), published in Aging in 2025.

PEARL was a 48-week, double-blinded, randomised, placebo-controlled, decentralised study in 114 adults aged 50 to 85, using a compounded formulation at a low, intermittent frequency. It remains the most substantial randomised placebo-controlled evidence available for rapamycin in normative-aging adults.

What it found:

• Safety was its main contribution, and that is a real contribution. Adverse events and serious adverse events were similar across all groups including placebo, and no significant differences in safety blood biomarkers were detected between the active groups and placebo over the study period. Before PEARL, the healthy-adult safety question was being answered with transplant-population data and anecdote. Now it is being answered with a randomised trial.
• The primary endpoint did not move. Visceral adiposity - the pre-specified primary outcome - showed no significant change.
• Two secondary signals were reported. Lean tissue mass and self-reported pain improved significantly in women in one active arm. Self-reported emotional well-being and general health improved in the other.
• The published conclusion was that low-dose intermittent rapamycin over 48 weeks appears relatively safe in healthy, normative-aging adults, and that whether it slows human aging remains unproven.

Credit where it is due. PEARL is a genuine randomised, placebo-controlled, double-blinded trial with a pre-specified primary endpoint, run for most of a year in a free-living population - more rigour than the overwhelming majority of interventions marketed for longevity have faced, and a null primary endpoint reported as null is itself a marker of an honestly run study.

Its limits are real and worth stating. The sample was small for any efficacy question. The duration was under a year. Every outcome was a surrogate - body composition, biomarkers, questionnaire scores - not lifespan and not disease incidence. Several positive findings were self-reported. The findings in women were subgroup results on secondary endpoints in a trial whose primary endpoint was null, which makes them hypothesis-generating rather than confirmatory. And the product studied was a compounded preparation, so questions about pharmacokinetic consistency apply.

None of that makes PEARL a bad trial. It makes it a first trial, doing the thing a first trial is supposed to do: establish that the question can be asked safely enough to keep asking.

The 2026 Oxford immune-senescence study

A second human data point arrived in January 2026, and it is easy to overstate. A team at Oxford published in Aging Cell a combined laboratory and human study of rapamycin and the ageing immune system. The laboratory work showed that low-level mTOR inhibition made human T cells more resilient to DNA damage. The human arm was a small placebo-controlled experimental medicine study: four older men on rapamycin and five on placebo, for four months. In immune cells from the rapamycin group, p21, a marker of damage-induced senescence, fell significantly compared with placebo, alongside lower markers of DNA damage and mTOR overactivity. White-cell counts did not differ over the first eight weeks, which the authors read as evidence the regimen was not broadly immunosuppressive.

What it adds is real but narrow: the first placebo-controlled evidence that rapamycin reduces markers of cellular senescence inside living older humans, which is the mechanism the mouse literature predicts. What it does not add is any clinical outcome. Nine participants, all men, measured on blood-cell markers over four months, cannot show fewer infections, better function or slower ageing, and the authors present it as a reason to run a Phase 2 trial rather than as a result. It does not move the efficacy question from "animal" to "human"; it moves the mechanism question a step closer to people. That fits the pattern of the earlier rapalog immune studies in older adults listed in the references, which remain the most developed human clinical line for this drug class.

Why the evidence in this field looks the way it does

Two structural facts shape everything in longevity research, and neither is a verdict on any particular compound.

First: you cannot run a lifespan trial in humans. Randomising middle-aged adults and following them to death would take decades, cost an implausible amount, and deliver an answer about a formulation that would be obsolete before the results arrived. No such trial exists for any longevity intervention, and none is likely to.

So every human longevity trial measures something else: biomarker panels, body composition, physical function, imaging, self-report. These are surrogate endpoints - stand-ins accepted on the assumption that moving them corresponds to moving the thing you actually care about. That assumption has not been established for aging measures; no epigenetic clock or composite aging index is currently accepted by regulators as a validated surrogate for healthspan or lifespan, a point covered properly in can you lower your epigenetic age. This is a property of the field, not a failing of any one intervention - everything here rests on surrogates, rapamycin included.

Second: nobody is going to fund the definitive trial. A Phase 3 programme costs hundreds of millions of dollars, and a commercial sponsor pays that only where there is exclusivity at the end of it. Rapamycin's core patents expired long ago - generic sirolimus has been available for years. Spermidine is a food component. NAD precursors are sold as supplements in most markets. Many of the most mechanistically interesting molecules in this space are unpatentable or off-patent, which means there is no owner with a commercial reason to spend that money, and no public funder currently spending it at that scale either.

The consequence is easy to misread. A thin trial literature here reflects who is willing to pay for trials, not a series of failed experiments. Absence of a funded Phase 3 is a statement about pharmaceutical economics, not evidence that a molecule does nothing. That cuts both ways: the honest answer to "is this proven" will stay "no" for a long time, for reasons that have nothing to do with the biology. Where evidence is genuinely early, the right response is to say so, say what the mechanism predicts, and decide with someone who can weigh it against your situation.

And that claim is testable rather than rhetorical. One compound in this article acquired a commercial sponsor - SS-31, through a rare-disease indication with orphan-drug incentives attached - and within a few years it had a randomised, double-blind, placebo-controlled trial, an FDA advisory committee vote and a marketing approval. Nothing about the molecule changed. The money changed. Its section is below.

Also worth holding: the mouse result is a statement about a genetically defined animal, in a controlled facility, on a fixed diet, dying of a narrow set of causes. Whether an intervention that compresses that mortality profile does something comparable in a genetically diverse human population dying of cardiovascular disease, dementia, diverse cancers and organ failure across an eighty-year span is a separate empirical question.

Where rapamycin sits practically: prescription-only, off-label, supervised

Rapamycin (sirolimus) is an approved prescription medicine - for the prevention of organ transplant rejection and certain other indications. It is not approved for aging, longevity or healthspan anywhere, and any such use is off-label prescribing, which means the prescribing clinician rather than a regulator is carrying the evidentiary judgement.

The reason supervision is structural rather than procedural is the mechanism itself. This drug's approved therapeutic action is immune suppression. That is not an incidental side effect to be managed; it is what it does. Deciding that a healthy person should take it means accepting that trade-off deliberately, which requires someone qualified to weigh it against that person's infection history, vaccination status, occupation, travel and planned procedures. The well-characterised class considerations from the approved indications include immunosuppression and infection risk, impaired wound healing, mouth ulcers (stomatitis), and metabolic effects including effects on lipids and glucose handling. Whether and to what degree those apply at the exposures studied in healthy adults is an individual clinical question.

The other half is monitoring. PEARL's safety result was generated inside a trial with structured monitoring and blood testing - safety observed under monitoring is not the same as safety without it. This page discusses no doses, exposures or schedules, because those belong inside a clinical relationship.

The honest summary: the animal evidence is excellent, the mechanism is coherent, the human safety data are early but not alarming at the exposures studied, and the efficacy question is open - and will stay open, because no one is funding the study that would close it.

The same biology, from the practical end

mTOR inhibition is one way into a broader idea: that cellular maintenance and recycling capacity declines with age, and that supporting it is a plausible lever on how well people age. Several other interventions act on that same biology, have accumulating human data of their own, and - unlike rapamycin - sit within reach of a programme with real monitoring behind it.

These are not consolation prizes and they are not ranked against rapamycin. They are different tools acting on overlapping machinery, and one of them has something rapamycin does not have: a regulator's signature.

SS-31 (elamipretide) - the compound that went the whole distance

On 19 September 2025 the FDA granted accelerated approval to elamipretide - SS-31 - under the brand name FORZINITY (Stealth BioTherapeutics), to improve muscle strength in patients with Barth syndrome above a minimum body weight. It is the first therapy ever approved for Barth syndrome, a rare genetic mitochondrial disorder for which patients previously had nothing.

The mechanism is genuinely distinct. Most of this category works by adding a substrate - more precursor, more metabolite, more raw material into a system assumed to be running short. SS-31 does something else. It is a mitochondrially-targeted peptide that binds cardiolipin, the signature phospholipid of the inner mitochondrial membrane. Cardiolipin gives cristae - the folded inner-membrane structures where oxidative phosphorylation happens - their shape and their ability to hold the electron-transport chain complexes in productive arrangement. Binding and stabilising it stabilises that structure and improves electron-transport efficiency: a structural-repair approach rather than a refuelling one, and the two are not interchangeable. The wider context sits in mitochondria and aging and how mitochondria work.

The evidence. The pivotal study was TAZPOWER, a randomised, double-blind, placebo-controlled crossover trial. It enrolled 12 patients. That number invites a reflexive dismissal which is, in this case, wrong: Barth syndrome is ultra-rare, and twelve patients is a meaningful fraction of the identifiable population. A crossover design is precisely what you use when the population is that small, because each participant serves as their own control. The FDA's Cardiovascular and Renal Drugs Advisory Committee voted 10–6 in favour.

The path was not smooth, and saying so is part of reporting it honestly. In October 2024, FDA review staff stated that the evidence did not establish effectiveness. The approval that followed in September 2025 is an accelerated approval - a pathway that permits authorisation on the strength of a reasonably likely surrogate or intermediate endpoint while carrying an obligation to generate confirmatory evidence afterwards. That is a real qualification, and not a reason to discount what happened.

Here is why this matters. SS-31 is a peptide of exactly the kind this whole category is built on - synthetic, targeted, mitochondrially active, the sort usually waved away as "no human evidence." And it went the entire distance: mechanism, randomised placebo-controlled trial, advisory committee hearing, FDA approval, a brand name, a label. It is the existence proof that the thin literature elsewhere in this field reflects funding, not futility. It got a sponsor, because a rare-disease indication brought orphan-drug incentives that made the economics work - and the moment it had one, the evidence appeared. Not slowly, and not ambiguously. The molecule did not become more promising in the interim. It became fundable.

That reframes the standard complaint about peptides. "There are no trials" is usually said as though it were a statement about the biology. It is very often a statement about the balance sheet.

Now the scope, which must be equally plain. FORZINITY is approved for Barth syndrome only - a rare genetic mitochondrial disorder, in a defined patient population, above a minimum body weight. It is not approved for ageing, for athletic or cognitive performance, or for general mitochondrial support in healthy adults, anywhere. Use outside that indication is off-label, sits on mechanism and early data rather than on an approval, and belongs inside a clinical relationship with monitoring attached. The approval is evidence that the mechanism is real and that the compound can clear a regulator. It is not evidence that it does anything for a healthy fifty-year-old, and this page does not claim that it is.

Both of those things are true at once, and the second does not cancel the first. The compound reference entry, covering what use of SS-31 involves and what monitoring sits around it, is on its own page. The dedicated treatment of the approval, what it covers and what it does not, is in is SS-31 FDA approved, and the mechanistic comparison with the other mitochondrial peptide below is in SS-31 and MOTS-c, the mitochondrial peptides.

MOTS-c - the peptide your own mitochondria make when you exercise

MOTS-c is a different kind of molecule again, and the honest framing is more interesting than the overclaimed one.

It is a mitochondrial-derived peptide: encoded not in nuclear DNA but in the mitochondrial genome itself, and induced by exercise - your own mitochondria release more of it when you train. This is not a foreign compound being introduced into a system; it is an endogenous signalling molecule, part of how mitochondria communicate with the rest of the body about energy status. MOTS-c is covered in its own right in MOTS-c, the mitochondrial peptide.

The animal record is strong and consistent. MOTS-c behaves as an exercise mimetic in animal models: improved insulin sensitivity, improved metabolic homeostasis, improved physical capacity - including when administration begins late in life rather than from youth. That late-start pattern is the same feature that made the rapamycin mouse data translationally interesting. The full treatment is in MOTS-c as an exercise mimetic.

The human evidence so far is observational, not interventional, and that distinction is the whole point. In a 16-week combined aerobic-and-resistance exercise intervention in breast cancer survivors, MOTS-c levels rose significantly after the exercise programme, and post-exercise levels were associated with reductions in fat mass, body weight, HOMA-IR and CRP, alongside an increase in lean mass. Separate work has compared circulating levels in professional athletes against non-athletes. Both strands tell you that MOTS-c tracks metabolic and training status in humans. Neither tells you what happens when you administer it - for that you need an interventional trial, and there are no completed interventional human trials of MOTS-c.

What there is, is a trial in progress. A Phase 2a study, MOTS-MET, testing subcutaneous MOTS-c over 12 weeks, is currently recruiting - the step this entire category is usually accused of skipping. The question is being asked in a randomised interventional design right now, and in a few years there will be an answer that does not depend on anyone's enthusiasm. That is a better position than "no evidence," and a more honest one than "proven."

MOTS-c is also prohibited in sport under WADA rules, which matters for anyone competing at a tested level.

One broader note. SS-31 and MOTS-c belong to a wider family: signalling peptides the body makes itself, or that act on structures it maintains itself, and whose availability or activity shifts with age. Vivre covers several others in that family - including epithalon and GHK-Cu - each with its own, separately assessed evidence base. The grouping is mechanistic, not promotional: these are molecules the body already uses, a different starting position from a xenobiotic drug, and it changes what the safety question looks like without changing what the efficacy question requires.

Spermidine - the closest mechanistic cousin to mTOR inhibition

If mTOR inhibition interests you because of autophagy, spermidine is the most direct place to look next. It is a dietary polyamine - richest in wheat germ, then fermented soy, aged cheeses, mushrooms, legumes and whole grains - and the body's own production of it declines with age. Mechanistically it is an autophagy inducer, best characterised through inhibition of acetyltransferases such as EP300, with a second strand implicating eIF5A hypusination and downstream TFEB activation. It is often described as a caloric-restriction mimetic - the same de-repression of autophagy as mTOR inhibition, by a different route. Work published in Nature Cell Biology in 2024 suggests spermidine is actually required for fasting-induced autophagy in yeast, worms and human cells.

The human picture has two strands worth keeping apart. Observational data are genuinely interesting: the Bruneck cohort and NHANES both link higher dietary spermidine intake to lower all-cause and cardiovascular mortality - though neither can fully separate spermidine from a wholefood-heavy diet. Interventional data are early: small pilots in older adults at dementia risk hinted at memory benefit, but the 12-month randomised placebo-controlled SmartAge trial in older adults with subjective cognitive decline was essentially null on its primary memory outcome. Cardiovascular trials including POLYCAD are ongoing.

That is where a food-derived, unpatentable compound ends up: strong preclinical work, meaningful population signal, and an interventional literature funded at the scale available money allows rather than the scale the question deserves. The full assessment is in spermidine and autophagy.

Mitophagy and urolithin A - autophagy, but for mitochondria specifically

Autophagy has a specialised branch that clears damaged mitochondria: mitophagy, running largely through the PINK1/Parkin pathway. It is the same maintenance logic applied to the organelles that do most of the cell's energy work - and because mitochondrial quality decline is one of the better-supported hallmarks of ageing, it is a sensible target.

Among the mitophagy inducers, urolithin A has the most repeated randomised human evidence. It is a postbiotic produced by gut bacteria from ellagitannins in pomegranate, berries and walnuts, and because only a minority of people carry the microbiome to make meaningful amounts, direct supplementation has a clear rationale rather than a marketing one. The trials are real and they repeat: a four-month randomised trial in middle-aged adults reported improved muscle endurance and reduced inflammatory and acylcarnitine markers; a trial in adults aged 65 to 90 published in JAMA Network Open in 2022 reported improved hand and leg muscle endurance with reduced mitochondrial biomarkers; a two-month randomised trial in 66 participants reported improved leg and hand muscle endurance with no change to diet or exercise; and an early-phase immune trial published in Nature Aging in 2026 reported expanded naive-like CD8⁺ T cells.

The honest reading is consistent signals on endurance and biomarkers, with several headline primary endpoints - peak power, six-minute walk, muscle ATP - not met. That is a modest, specific, repeatedly observed effect rather than a transformative one, and it is stated that way in urolithin A and mitophagy and in more functional detail in does urolithin A build muscle. Product identity matters a great deal here: the trial evidence is evidence about a characterised compound, not about any capsule bearing the name - which is precisely the gap that verified sourcing closes.

Mitochondrial function and NAD - the energy side of maintenance

Maintenance costs energy, which is why mitochondrial biology sits underneath most of this. Mitochondria and aging is the piece to read for the whole map: how the decline happens, and the three strategies that address it - fuelling the system, building more mitochondria through biogenesis, and repairing what is there.

NAD⁺ is the most discussed and most oversold thread in that story, and does NAD decline with age is deliberately careful about it. The short version: NAD⁺ is a shuttle rather than a fuel, blood NAD⁺ is a poor proxy for tissue NAD⁺, and the 2026 Nature Metabolism analysis pooling 237 adults across seven cohorts found whole-blood levels essentially flat across adulthood - while direct tissue measurements in skin, muscle, brain and liver still show age-related decline. NAD precursors reliably raise blood NAD metabolites; whether that translates to tissue benefit is the open question, and consumption by CD38, sirtuins and PARPs makes it a budget problem rather than a supply problem. That is worth measuring rather than assuming.

The two forms people ask about most are covered on their own pages, with those caveats carried through rather than dropped: NAD and NMN. Note the contrast with the section above: NAD precursors are the substrate-addition approach to mitochondrial decline, SS-31 the structural-repair approach. They address different failure modes, and which is relevant depends on what is actually going wrong.

Senolytics - clearing what maintenance cannot repair

Where autophagy clears damaged components inside a cell, senolytics target cells that have stopped dividing but remain metabolically active, secreting an inflammatory signature (the SASP) that contributes to the chronic inflammation of ageing. Mouse work - genetic clearance first, then drug combinations - improved physical, cardiac and kidney function and extended healthspan.

Human work is earlier. The landmark dasatinib-plus-quercetin study was a phase 1 pilot in patients with diabetic kidney disease that demonstrated mechanism engagement - reduced adipose senescent-cell burden and falling circulating IL-1α, IL-6, MMP-9 and MMP-12 - rather than clinical benefit. Roughly 14 completed human trials of that combination had been reported by early 2026, with phase 2 work underway in idiopathic pulmonary fibrosis, Alzheimer's disease and frailty. Fisetin outperforms other flavonoids in laboratory senolytic screens and Mayo Clinic's AFFIRM trial has preliminary inflammatory-marker reports. Dasatinib is a prescription leukaemia drug with a serious adverse-effect profile; this is not a category to approach casually, and the evidence on senolytics treats it accordingly.

Klotho, and how you know any of this is working

Klotho is the newest entrant - an ageing-associated protein whose gene knockout produces accelerated ageing in mice and whose overexpression extends lifespan, now in human trials with an investigational mRNA agent in an early-phase safety study. It is genuinely investigational, the human causal question is unresolved, and no intervention has been shown to raise human klotho to a level producing demonstrated benefit. It is on this list because it is worth following, not because it is available.

More practically: measurement. If you are going to work on this biology, what you measure and how much to trust it is unavoidable. What biological age tests measure and can you lower your epigenetic age cover why a single before-and-after pair on a consumer clock is one draw from a noisy distribution, and why the CALERIE result - a randomised two-year caloric restriction trial in 220 adults that moved DunedinPACE slightly and left PhenoAge and GrimAge unmoved - is the most instructive data point in the whole measurement conversation.

And the unglamorous part, which is also the best-evidenced part: exercise activates autophagy and mitophagy, drives mitochondrial biogenesis, and has decades of hard outcome data behind it. Nothing in this article competes with that. Everything in this article sits alongside it - and in the case of MOTS-c, quite literally emerges from it, since the peptide is one of the signals trained mitochondria send out. That is a useful way to hold the whole page: the training is the intervention, and most of the rest of this is an attempt to understand and support the machinery it works through.

What would make the picture clearer

"More research is needed" hides the actual requirement. For rapamycin, a real answer would need scale - samples large enough to detect modest effects and survive testing multiple outcomes; duration measured in years rather than months; pre-specified objective endpoints such as measured physical and cognitive function, incidence of age-related disease or hospitalisation, rather than biomarker panels and questionnaires; independent replication in different populations, the standard the ITP applies to mouse work; and a defined, consistent formulation. That work has not been done, and given who would have to pay for it, it may not be.

The mitochondrial side of this page is in a different position. For SS-31, the outstanding requirement is defined and obligatory: an accelerated approval carries a confirmatory-evidence commitment, so the work that would firm up the Barth syndrome indication is scheduled rather than hypothetical - and the separate question of whether the mechanism does anything in ordinary ageing has not been asked in a trial at all. For MOTS-c, the first interventional human study is recruiting now. In both cases there is a named trial with a sponsor behind it. That is what rapamycin does not have, and it is the difference between "unanswered" and "unasked."

Hence the approach this page recommends: work with what has plausible mechanism and accumulating human data, measure what you can measure, monitor what needs monitoring, and adjust. That is not a lesser version of certainty. Given the structure of this field, it is what informed practice looks like.

Where Vivre Labs fits

Rapamycin itself is not a Vivre Labs compound, and nothing on this page is a route to obtaining it. What Vivre Labs offers is access to compounds that act on the mitochondrial end of the same maintenance biology. None is a rapamycin substitute, none has been shown to slow ageing in people, and each carries a different evidence tier, stated below. We recommend starting with baseline bloodwork.

• SS-31 (elamipretide) is a cardiolipin-binding mitochondrial peptide. It is FDA-approved, by accelerated approval, for Barth syndrome only; any use for ageing or general mitochondrial support is off-label and rests on mechanism and early data rather than on that approval.
• MOTS-c is an exercise-induced, mitochondrial-derived peptide. Animal studies suggest exercise-mimetic metabolic effects, but there is no completed interventional human trial; MOTS-MET is still recruiting. It is investigational, not FDA-approved, and prohibited in sport under WADA rules.
• Epithalon is a synthetic bioregulator peptide whose evidence is mostly small Russian studies, animal work and cell culture, set out in Epitalon and the bioregulator evidence problem. It is investigational and not FDA-approved.
• NAD is the substrate-addition approach on the energy side. Precursors raise blood NAD metabolites; whether that becomes tissue benefit is the open question covered in does NAD decline with age.
• The cellular energy and longevity protocol brings SS-31, MOTS-c and epithalon into one programme. Combining them adds no evidence that the individual compounds lack; it is a structured way to discuss them with a clinician, with monitoring attached.
• A baseline biomarker panel comes first: lipids, glucose markers, kidney and liver function and the other measures that make any later change interpretable.

The conclusion above stands. For rapamycin the efficacy question is open, and for everything in this section the human evidence is early. The useful next step is a measured baseline and a conversation with a physician about whether any of it fits your situation.

Frequently asked questions

Does rapamycin slow aging in humans? Not demonstrated. It extends lifespan in mice with unusually strong replication, including when started in mid-to-late life, and that finding is real. In humans, the substantial randomised evidence is a single 48-week placebo-controlled trial in 114 adults that was primarily a safety study, reported a null result on its primary endpoint, and measured surrogate outcomes rather than lifespan or disease incidence. The published conclusion was explicit that the question remains unproven. Worth adding: the trial that would settle it would cost hundreds of millions, and rapamycin has been off-patent for years, so no commercial sponsor has a reason to run it. The question is open partly for economic reasons, not only scientific ones.

Didn't a 2026 study show rapamycin reduces senescent cells in people? It showed something narrower. A January 2026 Oxford study in Aging Cell gave rapamycin or placebo to nine older men (four and five respectively) for four months and found lower markers of DNA damage and senescence in their immune cells. That is a genuine mechanism signal in humans, and the authors use it to argue for a Phase 2 trial. It measured blood-cell markers, not health outcomes, so it does not show that rapamycin slows ageing.

What did the PEARL trial actually show? That low-dose intermittent rapamycin over 48 weeks appeared relatively well tolerated in healthy, normative-aging adults - adverse events and safety blood biomarkers were similar across active and placebo groups. Its primary endpoint, visceral adiposity, did not change significantly. Secondary findings included improvements in lean tissue mass and self-reported pain in women in one active arm, and self-reported emotional well-being and general health in the other. Those are subgroup findings on secondary endpoints in a trial with a null primary endpoint, in a small sample, over less than a year, using a compounded product. They justify a larger trial rather than a conclusion. The safety dataset is the durable contribution.

If rapamycin isn't the answer, what acts on the same biology? Several things, none of which are rapamycin substitutes but all of which touch the same maintenance-and-recycling machinery. SS-31 (elamipretide) is the one with a regulatory approval behind it - a cardiolipin-binding mitochondrial peptide, FDA-approved in September 2025 for Barth syndrome and not for ageing. MOTS-c is a mitochondrial-derived, exercise-induced peptide with a strong animal record and its first interventional human trial now recruiting. Spermidine is the closest mechanistic relative to rapamycin itself - an autophagy inducer with strong preclinical data, meaningful population-level mortality associations and an early interventional literature. Urolithin A targets mitophagy specifically and has the most repeated randomised human trials of the group, with modest but consistent endurance and biomarker effects. NAD precursors address the energy side, with a more complicated story than the marketing suggests. Senolytics clear cells that maintenance cannot repair, and are genuinely early in humans. Underneath all of it, exercise remains the best-evidenced activator of autophagy, mitophagy and mitochondrial biogenesis there is.

Is SS-31 FDA approved? Yes - but for one specific thing. On 19 September 2025 the FDA granted accelerated approval to elamipretide (SS-31) under the brand name FORZINITY, to improve muscle strength in patients with Barth syndrome above a minimum body weight. It is the first therapy ever approved for that rare genetic mitochondrial disorder. It is not approved for ageing, longevity, performance or general mitochondrial support in healthy adults, and use outside the Barth syndrome indication is off-label. The pivotal evidence was TAZPOWER, a randomised, double-blind, placebo-controlled crossover trial in 12 patients - small because the disease is ultra-rare - and the FDA's Cardiovascular and Renal Drugs Advisory Committee voted 10–6 in favour. FDA review staff had said in October 2024 that the evidence did not establish effectiveness, and the approval granted is an accelerated one, which carries a confirmatory-evidence obligation. Full detail in is SS-31 FDA approved.

Why does an approval for a rare disease matter to anyone else? Because of what it demonstrates about this field rather than about that disease. The standard objection to mitochondrial peptides is that they have mechanism and enthusiasm but no trials. SS-31 had exactly that profile until it acquired a commercial sponsor through a rare-disease indication with orphan-drug incentives attached - and then produced a randomised placebo-controlled trial, an advisory committee vote and an FDA approval. The molecule did not change; the funding did. It does not transfer efficacy from Barth syndrome to ageing - nothing does that except a trial in ageing - but it does dismantle the assumption that this class of compound cannot generate real evidence.

Does MOTS-c actually work in humans? Not established, and it would be wrong to say otherwise. MOTS-c is a peptide encoded in mitochondrial DNA and released in response to exercise, and in animal models it behaves as an exercise mimetic - improving insulin sensitivity, metabolic homeostasis and physical capacity, including when started late in life. The human data so far are observational: in a 16-week aerobic-and-resistance exercise intervention in breast cancer survivors, MOTS-c levels rose significantly after training, and post-exercise levels were associated with reductions in fat mass, body weight, HOMA-IR and CRP and an increase in lean mass; other work has compared levels in professional athletes with non-athletes. Those show that MOTS-c tracks training and metabolic status, not what administering it does. There are no completed interventional human trials. A Phase 2a trial, MOTS-MET, testing subcutaneous MOTS-c over 12 weeks, is recruiting - which is the step most of this category never reaches. MOTS-c is also prohibited in sport under WADA rules.

Why is the human evidence so thin across this whole field? Two structural reasons. The endpoint that would settle any of it is lifespan, and a human lifespan trial would take decades - so every human longevity trial measures surrogates, and no aging surrogate is currently accepted by regulators as validated. And most of the interesting molecules are unpatentable or off-patent, so no commercial sponsor stands to gain from funding a Phase 3 programme that costs hundreds of millions. Thin literature in this field reflects who pays for trials at least as much as it reflects what the molecules do. That is a reason to weigh mechanism and early human data carefully with a clinician, not a reason to assume nothing works - and equally not a licence to claim more than the data support.

Related reading

• Is SS-31 FDA approved? - what the September 2025 approval covers, and what it does not
• SS-31 and MOTS-c, the mitochondrial peptides - the two compared side by side
• MOTS-c as an exercise mimetic · MOTS-c, the mitochondrial peptide · Senolytics and "zombie cells"
• Mitochondria and aging · How mitochondria work - the cellular biology underneath most of this
• Spermidine and autophagy - the closest mechanistic relative to mTOR inhibition
• Urolithin A and mitophagy · Does urolithin A build muscle?
• Does NAD decline with age - what the human precursor trials do and do not show
• Klotho, the longevity protein - newly in human trials
• What biological age tests measure · Can you lower your epigenetic age
• What biological age tests measure · Can you lower your epigenetic age
• Are GLP-1s anti-aging drugs? - what the 2026 longevity claim rests on, and why the ageing trial has not been run
• Part of Testing & Biological Age

---

Explore it with verified supply. Every Vivre Labs lot is independently tested and the certificate of analysis is published, so you know exactly what is in the vial. Explore SS-31.

Educational content, not medical advice. Rapamycin (sirolimus) is a prescription medicine approved for the prevention of organ transplant rejection and certain other indications; it is not approved for aging, longevity or healthspan in any jurisdiction, and any such use is off-label and requires a prescription and ongoing physician supervision. Elamipretide (SS-31) is approved in the United States under the brand name FORZINITY for Barth syndrome only, by accelerated approval; it is not approved for ageing, longevity, healthspan or performance in any jurisdiction, and any such use is off-label and requires physician supervision. MOTS-c has no completed interventional human trial and no approved indication anywhere; it is prohibited in sport under WADA rules. No doses, schedules, protocols or sourcing information appear anywhere on this page, and nothing here should be read as encouragement to obtain or use this or any prescription medicine outside a clinical relationship. Compounds described as investigational are investigational, and early human evidence is identified as early. Claims for the Malaysian market are subject to MAB requirements. Position stated as of 28 September 2026.

References

• Mannick JB, Lamming DW, et al. Targeting the biology of ageing with mTOR inhibitors. Nature Aging. (Review of mTORC1 biology, nutrient and growth-factor sensing, autophagy de-repression, and the rationale and limits of translating mTOR inhibition to human ageing.)
• Harrison DE, Strong R, Miller RA, et al. Rapamycin fed late in life extends lifespan in genetically heterogeneous mice. Nature. 2009;460:392–395. (NIA Interventions Testing Program; genetically heterogeneous mice; treatment begun in late life; lifespan extension observed.)
• NIA Interventions Testing Program - multi-site, independently replicated testing of candidate longevity compounds in genetically heterogeneous mice. Rapamycin is the programme's most consistently positive lifespan result; most submitted compounds do not replicate.
• Selvarani R, Mohammed S, Richardson A. Effect of rapamycin on aging and age-related diseases - past and future. GeroScience. (Synthesis of the rodent lifespan and healthspan literature across cardiac, immune, cognitive and tissue-level outcomes; discussion of translational limits.)
• Kaeberlein M, et al. Regulation of yeast replicative life span by TOR and Sch9. Science. 2005. · Powers RW, et al. Extension of chronological life span in yeast by decreased TOR signalling. Genes & Development. 2006. (Conservation of the TOR lifespan effect across model organisms.)
• PEARL trial - Participatory Evaluation of Aging with Rapamycin for Longevity (NCT04488601). "Influence of rapamycin on safety and healthspan metrics after one year: PEARL trial results." Aging. 2025. (48-week, double-blinded, randomised, placebo-controlled, decentralised trial; 114 adults aged 50–85 completing analysis; compounded low-dose intermittent rapamycin. Adverse events, serious adverse events and safety blood biomarkers similar across groups. Primary endpoint of visceral adiposity not significantly changed. Secondary endpoints: lean tissue mass and self-reported pain significantly improved in women in one active arm; self-reported emotional well-being and general health improved in the other. Conclusion: relatively safe in healthy normative-aging adults over 48 weeks; effect on human aging unproven.) medRxiv preprint · PubMed Central
• Kell L, et al. Rapamycin exerts its geroprotective effects in the ageing human immune system by enhancing resilience against DNA damage. Aging Cell. 2026 (University of Oxford; in vitro DNA-damage assays in human T cells, ex vivo profiling of aged immune subsets, and a placebo-controlled experimental medicine study in older male volunteers, rapamycin n=4 versus placebo n=5 over four months; significant reduction in immune-cell p21 versus placebo; no difference in leukocyte counts over the initial eight weeks; authors propose Phase 2 double-blind placebo-controlled studies). PubMed Central
• Sirolimus (rapamycin) prescribing information - approved indication for prophylaxis of organ rejection in renal transplantation and certain other specified indications. Class warnings include immunosuppression and increased susceptibility to infection, impaired wound healing, stomatitis and oral ulceration, hyperlipidaemia and effects on glucose handling. No approved indication for aging, longevity or healthspan in any jurisdiction. Core patents long expired; generic sirolimus is available.
• Mannick JB, et al. mTOR inhibition improves immune function in the elderly. Science Translational Medicine. 2014. · Mannick JB, et al. TORC1 inhibition enhances immune function and reduces infections in the elderly. Science Translational Medicine. 2018. (Rapalog trials in older adults with immune-function and infection-related endpoints; the most developed human clinical line for this drug class in ageing, and still not a lifespan or healthspan claim.)
• SS-31 / elamipretide - FDA accelerated approval granted 19 September 2025; brand name FORZINITY (Stealth BioTherapeutics); indication: to improve muscle strength in patients with Barth syndrome above a minimum body weight. First therapy approved for Barth syndrome. Mechanism: mitochondrially-targeted peptide binding cardiolipin in the inner mitochondrial membrane, stabilising cristae structure and improving electron-transport efficiency. Pivotal evidence: TAZPOWER, a randomised, double-blind, placebo-controlled crossover trial in 12 patients (population size reflects the ultra-rare prevalence of the disorder). FDA Cardiovascular and Renal Drugs Advisory Committee voted 10–6 in favour. FDA review staff had stated in October 2024 that the submitted evidence did not establish effectiveness; the authorisation subsequently granted is an accelerated approval, carrying a post-approval confirmatory-evidence obligation. No approved indication for ageing, longevity, healthspan, performance or general mitochondrial support in any jurisdiction.
• MOTS-c - mitochondrial-derived peptide encoded in the mitochondrial genome; expression and circulating levels induced by exercise. Animal literature: exercise-mimetic phenotype with improved insulin sensitivity, metabolic homeostasis and physical capacity, including with administration begun in late life. Human evidence observational only: 16-week combined aerobic-and-resistance exercise intervention in breast cancer survivors reported significant post-exercise increases in MOTS-c, with post-exercise levels associated with reductions in fat mass, body weight, HOMA-IR and CRP and an increase in lean mass; separate comparisons of circulating levels in professional athletes versus non-athletes. No completed interventional human trials. MOTS-MET - Phase 2a trial of subcutaneous MOTS-c over 12 weeks - recruiting. Prohibited in sport under WADA rules. No approved indication in any jurisdiction.
• Spermidine - EP300 acetyltransferase inhibition and eIF5A hypusination/TFEB strands as autophagy-induction mechanisms; Nature Cell Biology 2024 work indicating spermidine is required for fasting-induced autophagy in yeast, nematodes and human cells. Bruneck cohort and NHANES: higher dietary spermidine intake associated with lower all-cause and cardiovascular mortality (observational; not separable from wholefood dietary pattern). SmartAge - 12-month randomised placebo-controlled trial of wheat-germ-derived spermidine in older adults with subjective cognitive decline; null on primary memory outcome. POLYCAD and other cardiovascular trials ongoing.
• Urolithin A - gut-microbial metabolite of dietary ellagitannins; mitophagy induction via the PINK1/Parkin pathway. Singh A, et al. Cell Reports Medicine. 2022 (NCT03464500; randomised placebo-controlled, middle-aged adults, approximately four months; primary endpoint of peak power output not met; muscle endurance and inflammatory/acylcarnitine biomarker improvements reported as secondary signals). JAMA Network Open. 2022 (adults aged 65–90; hand and leg muscle endurance improved; co-primary six-minute walk distance and maximal muscle ATP production not significantly different from placebo). Two-month randomised trial, n=66, leg and hand muscle endurance improvements without diet or exercise change. Nature Aging. 2026 - early-phase immune trial reporting expanded naive-like CD8⁺ T cells. Several pivotal trials manufacturer-funded.
• NAD⁺ - "Human whole-blood NAD⁺ levels do not vary with age or lifestyle interventions." Nature Metabolism. 2026 (237 healthy individuals aged 18–70, seven independent cohorts, validated UHPLC–high-resolution mass spectrometry; whole-blood NAD⁺ essentially flat across adulthood; did not correct for red-blood-cell count). Direct tissue measurements in skin, skeletal muscle, brain and liver continue to show age-related decline. McReynolds MR, et al. Cell Systems. 2021 (NAD⁺ flux maintained in aged mouse tissue despite lower standing concentration). Camacho-Pereira J, et al. Cell Metabolism. 2016 (CD38 as an age-associated NAD⁺ consumer).
• Senolytics - dasatinib plus quercetin phase 1 pilot in diabetic kidney disease: reduced adipose senescent-cell burden and decreases in circulating IL-1α, IL-6, MMP-9 and MMP-12; mechanism engagement rather than clinical benefit. Approximately 14 completed human D+Q trials reported by early 2026; phase 2 trials underway in idiopathic pulmonary fibrosis, Alzheimer's disease and frailty. Fisetin outperforms other flavonoids in laboratory senolytic screens; Mayo Clinic AFFIRM trial preliminary inflammatory-marker reports. Dasatinib is a prescription tyrosine kinase inhibitor with cytopenias, fluid retention and pleural effusion in its adverse-effect profile. No validated measure of senescent-cell burden in living humans.
• Klotho - Kurosu H, et al. Science. 2005 (α-klotho overexpression extends mouse lifespan; gene disruption produces accelerated-ageing phenotype). KL-VS variant association data non-linear and inconsistently replicated. AKL003 alpha-klotho mRNA agent - randomised, double-blind, placebo-controlled early-phase safety and tolerability study announced February 2026; investigational, no approved indication.
• CALERIE - 220 adults without obesity randomised to sustained caloric restriction or ad-libitum control for two years; epigenetic analysis published in Nature Aging 2023. DunedinPACE slowed by approximately 3%; no significant change on PhenoAge or GrimAge. Mortality benefit not demonstrated by the trial.
• On surrogate endpoints in geroscience: no epigenetic clock or composite biological-age measure is currently accepted by regulators as a validated surrogate endpoint for healthspan or lifespan; human longevity trials therefore report biomarker, functional and self-reported outcomes rather than the outcome of interest.