Geroprotectors

The "longevity molecule" category — NMN, resveratrol, rapamycin, urolithin A, spermidine and the rest — is rich in mechanism and animal data but consistently disappointing in human trials. Five years on, the gap between mechanistic promise and clinical translation has widened, not narrowed.

This category — pharmacological and nutraceutical agents proposed to target the hallmarks of aging directly — is where hope most outruns evidence. The honest assessment as of 2026: mechanistic promise is rich, human translation is thin, and industry funding distorts the published literature significantly.

The class as a whole

Two truths often missing from longevity content:

  1. No supplement on the market has been shown to extend life in healthy adults at the magnitude of basic lifestyle interventions. No "longevity molecule" comes close to the effect of regular exercise, Mediterranean-pattern eating, or quitting smoking.

  2. The gap between mechanistic promise and clinical translation is widening, not narrowing, for the entire emerging-longevity category. Five years of further research has, on net, deflated rather than confirmed most claims.

That said, this is an active and evolving field. Some agents (rapamycin, SGLT2 inhibitors, GLP-1 receptor agonists) have stronger biological signals than others (resveratrol, NMN at typical over-the-counter (OTC) doses).

Several agents here are sold on mitochondrial and metabolic-health claims. The fuel-switching markers those claims rest on, how they are measured, and the one supplement with randomised evidence of actually moving them — L-carnitine — are covered in Metabolic flexibility.

Tier 1: Best human evidence among geroprotectors

Rapamycin (and rapalogs) — Moderate (mechanistic / surrogate); Weak (hard outcomes)

The closest thing to a true gerotherapeutic in 2026.

The case for:

  • Inhibiting mTOR (the mechanistic target of rapamycin, a central nutrient-sensing pathway) extends median lifespan in mice by roughly 10–25%, depending on dose and sex (females gain more), replicated across the three sites of the US National Institute on Aging's Interventions Testing Program.[1]
  • In small, short trials, related mTOR-blocking drugs — everolimus, alone or combined with an experimental drug called BEZ235, not rapamycin itself — modestly improved older adults' response to the flu vaccine and cut self-reported infections.[2] [3] But a 1,000-person phase 3 trial of BEZ235 found no reduction in respiratory illness (26% of people on the drug vs 25% on placebo), even though it again switched on antiviral gene activity.[4]
  • The largest and longest trial to date of intermittent low-dose rapamycin in healthy adults has now reported. PEARL — 48 weeks, placebo vs. 5 mg or 10 mg weekly — found the schedule well-tolerated, with adverse events no more common than on placebo (though rapamycin users appeared to have more stomach upsets) and modest secondary signals (more lean mass and less pain in women on the higher dose).[5] But it missed its primary endpoint — no change in visceral fat — and used compounded rapamycin later found to carry only about a third the bioavailability of generic sirolimus, so it establishes tolerability more than efficacy. The lean-mass and pain result came from just 8 women on the higher dose, only people who finished the trial were analysed, and every author is an employee or shareholder of AgelessRx, a company that prescribes rapamycin for longevity. Independent trials are following: one of everolimus, a close relative, in insulin-resistant adults (EVERLAST) finished in mid-2026 without results yet,[6] and a larger dosing study in people over 65 (RESTOR) runs to 2028.[7]
  • Mechanistic: directly inhibits mTOR, the most heavily targeted antagonistic hallmark.

The case for caution:

  • Still no human RCT with hard longevity outcomes; PEARL's positive results were secondary endpoints, confined to subgroups.
  • A 2024 systematic review of 19 human studies of rapamycin and related drugs, in healthy people and people with age-related disease, found improvements in some immune, heart-and-vessel and skin measures, but no significant effect overall on muscle, hormone and brain measures. No serious side effects were reported in healthy people, but people who were already ill had more infections and higher cholesterol and triglycerides.[8]
  • Side effects: mouth ulcers, lipid abnormalities, glucose dysregulation, immunosuppression.
  • Dosing is far from optimized for healthy-adult use.
  • Not approved for longevity indication; off-label use is increasing but unmonitored.

Practical: Some clinicians prescribe low-dose intermittent rapamycin (5–6 mg weekly) off-label. The one placebo-controlled trial of that dose in older adults is not encouraging. In RAPA-EX-01, 40 people aged 65–85 took 6 mg of standard sirolimus (the generic name for rapamycin) or a placebo once a week for 13 weeks, alongside a home exercise programme. Rapamycin added no benefit: the main result was not statistically significant, and backup analyses hinted that it slightly blunted the gains from exercise. The same share of people (85%) had side effects in each group, but the rapamycin group logged more of them in total (99 vs 63), including one pneumonia that may have been caused by the drug.[9] PEARL's reassuring tolerability does not carry over to this dose, because its compounded product delivered only about a third as much drug as standard sirolimus. Without longer human safety data this remains experimental — appropriate only in informed adults working with a knowledgeable physician. Bear in mind what the drug actually is: rapamycin is a licensed immunosuppressant, and the weekly-pulse schedule used in longevity practice is an attempt to get the mTOR effect without the immunosuppression, not a demonstrated way of avoiding it. It is contraindicated in pregnancy and in active infection; live vaccines should be avoided on it; it impairs wound healing (stop before elective surgery); and it is metabolised by the CYP3A4 enzyme, so it interacts with the same enzyme flagged elsewhere on this page for berberine and grapefruit. Anyone taking it needs periodic lipids, fasting glucose or HbA1c (a marker of longer-term blood sugar), and a complete blood count (CBC) — the three things it most reliably disturbs.

Metformin — Moderate

A widely studied biguanide originally for type 2 diabetes (T2D).

The case for:

  • AMPK activator (the catabolic counterweight to mTOR).
  • Long-term follow-up of the UKPDS randomised trial, and the UK CPRD records database, showed lower all-cause mortality in T2D patients on metformin than on sulfonylureas. The much-repeated claim that metformin users outlive non-diabetic controls comes from one 2014 UK records analysis[10] and has been heavily criticised for immortal-time and comparator bias — sulfonylureas are themselves associated with harm, so "better than a sulfonylurea" is a low bar. When a 2022 Danish national-register study tested the claim again, including twin pairs in which only one twin had diabetes, it found the opposite: people starting metformin had higher death rates than matched people without diabetes.[11] Treat the survival claim as an association with a plausible confound, not evidence of a geroprotective effect.
  • The TAME (Targeting Aging with Metformin) trial was designed to evaluate aging endpoints in non-diabetic adults, but it has never enrolled anyone. It has some money — about $5 million set aside by the US National Institute on Aging, plus private pledges — against an estimated $45–70 million cost, so a decade on it remains a proposal, not a study in progress, and no result is coming soon.[12]

The case for caution:

  • Some evidence that it blunts the benefits of exercise. In a 12-week trial in older adults, metformin weakened the gains in fitness (maximal oxygen uptake, VO₂max), insulin sensitivity and muscle mitochondrial function from aerobic training.[13] In the MASTERS trial, healthy over-65s doing 14 weeks of strength training gained less muscle on metformin than on placebo, although their strength gains did not differ significantly.[14] A small 2026 trial in adults at risk of metabolic syndrome found metformin altered some artery pressure-wave responses to training, with mixed blood-pressure results.[15] Concerning for active adults relying on training for longevity.
  • Vitamin B12 deficiency risk with chronic use.
  • Modest gastrointestinal (GI) side effects.
  • No approved indication for healthy non-diabetic use.

Practical: For healthy adults, the cost-benefit currently favors not using metformin off-label. The exercise interaction is concerning. For prediabetes, lifestyle change comes first: a structured programme aiming for 5–7% weight loss and at least 150 minutes a week of moderate activity. Doctors consider adding metformin for people at the highest risk — especially adults aged 25–59 with a body mass index (BMI) of 35 or more and higher fasting glucose and HbA1c, and women who have had gestational diabetes.[16] In type 2 diabetes it remains a common, cheap starting drug, but people with heart disease or high cardiovascular risk, or with kidney disease, are now steered to a sodium-glucose cotransporter-2 (SGLT2) inhibitor or a glucagon-like peptide-1 (GLP-1) drug whatever their HbA1c, and people with heart failure to an SGLT2 inhibitor, with or without metformin.[17]

SGLT2 inhibitors (empagliflozin, dapagliflozin) — Strong (in disease populations)

A newer class — sodium-glucose cotransporter-2 (SGLT2) inhibitors block glucose reabsorption in the kidney — with surprising longevity-relevant signals.

The case for:

  • Reduce deaths from any cause in large trials of people with type 2 diabetes and heart disease (EMPA-REG OUTCOME: 5.7% vs 8.3% died, about a third fewer),[18] heart failure (HF) with reduced pumping (DAPA-HF: about 17% fewer deaths)[19] and chronic kidney disease (CKD) (DAPA-CKD: 4.7% vs 6.8% died, about 31% fewer).[20] Pooled across five large heart-failure trials the drop in deaths is smaller, about 8%; the clearer benefit there is fewer hospital stays for heart failure.[21]
  • Pleiotropic effects beyond glucose: reduced cardiac remodeling, reduced renal decline, improved autophagy markers.
  • Proposed as longevity drugs for healthy adults, but untested there: the first placebo-controlled trial in healthy older adults (80 people aged 60–80, measuring artery stiffness) is still recruiting.[22]

The case for caution:

  • Approved only for T2D, HF, and CKD.
  • Side effects: increased urinary-tract and genital yeast infections, rare diabetic ketoacidosis.
  • No safety/efficacy data in healthy adults.

Practical: For longevity-only off-label use in healthy adults, the evidence is currently insufficient to recommend.

GLP-1 receptor agonists (semaglutide, tirzepatide) — Moderate (and growing)

The newest, most contested entry in Tier 1. Among these glucagon-like peptide-1 (GLP-1) receptor agonists, semaglutide slowed several validated epigenetic aging clocks in a randomized human trial — GrimAge version 2 by about 2.3 epigenetic years per calendar year over 32 weeks, in 84 adults with HIV-related fat redistribution.[23] That signal comes from a small, post-hoc exploratory analysis in an atypical population, with the clocks scored by the vendor that sells them, so read it as suggestive rather than settled. Nor is it the first intervention to move these clocks in a randomized trial: in older Europeans, omega-3 supplements slowed three such clocks slightly over three years.[24] On firmer footing, again for semaglutide specifically: in people who already had cardiovascular disease and overweight or obesity but no diabetes, it cut the combined rate of heart attacks, strokes and cardiovascular deaths by 20% (SELECT),[25] and it produces durable double-digit weight loss in non-diabetic adults. Tirzepatide's heart-attack-and-stroke outcome trial compared it with an older GLP-1 drug, dulaglutide, in people with type 2 diabetes and heart disease: it was no worse, but not shown to be better.[26] Its one placebo-controlled outcome trial so far, in people with obesity and heart failure with preserved pumping, cut cardiovascular deaths and worsening heart failure combined from 15.3% to 9.9% of patients, almost all through fewer heart-failure events; cardiovascular deaths alone were not reduced.[27] The same biology that drives those benefits also drives meaningful muscle and bone loss unless the patient eats enough protein and trains. Off-label "microdosing" for healthy adults is plausible but unproven and carries real compounding-pharmacy risks.

Practical: For an adult with overweight/obesity and cardiometabolic risk, the evidence is now strong enough to consider seriously. For a healthy normal-BMI adult chasing longevity alone, the case is interesting but not yet made — and never start without protein and resistance training in place.

→ Full treatment in GLP-1 receptor agonists: Ozempic, Wegovy, and the longevity question.

Tier 2: Mechanistically interesting, weak human evidence

NAD+ precursors: NMN and NR

Both reliably elevate blood NAD+ at 250–1000 mg/day. Whether this translates to clinical benefit in healthy adults is essentially unproven.

  • Muscle insulin sensitivity gains in postmenopausal prediabetic women with NMN 250 mg.[28]
  • No reliable blood-pressure benefit from NR. A small 2018 trial in healthy middle-aged and older adults saw systolic blood pressure (SBP) fall on 1000 mg only in a subgroup with above-normal readings, picked out after the fact, and the authors said no conclusions could be drawn from it.[29] When the same group tested it directly in people with high blood pressure, adding NR to a walking programme lowered blood pressure no more than walking alone.[30]
  • NR 2000 mg in obese men — completely negative for insulin sensitivity, mitochondrial function, body composition.[31]

Pooled, the metabolic case collapses further. Across eight randomised trials in 342 adults, NMN moved neither fasting glucose, fasting insulin, HbA1c, HOMA-IR (the standard fasting index of insulin resistance) nor blood lipids.[32] Pooled trials in people over 60 found no gain in muscle mass, grip strength, walking speed or chair-rise time with either NMN or NR.[33] Raising blood NAD+ is real; the downstream benefit is what keeps failing to appear.

A new concern: NR and NMN, like niacin, raise blood levels of 2-PY and 4-PY, the end products of NAD+ breakdown. In observational studies of heart patients, people with higher levels had more major cardiovascular events (heart attack, stroke or death) over three years, and 4-PY inflamed blood-vessel walls in mice.[34] Whether supplement-driven rises cause harm is unknown.

Regulatory note: NMN's status has been unstable. In late 2022 the US FDA concluded NMN is excluded from the dietary-supplement definition because it had earlier been authorized for investigation as a new drug (the filing behind the pharmaceutical candidate MIB-626), throwing its US over-the-counter availability into legal limbo; after an industry lawsuit the agency suspended enforcement in late 2024, and in September 2025 it reversed itself: NMN is no longer barred from supplements, though as a new dietary ingredient its makers must still notify the FDA.[35] Being legal to sell does not make it effective. NR (Niagen) has been on stable footing throughout.

Practical: The hype substantially outruns the evidence. Save the money. The broader vitamin B3 picture — niacin, nicotinamide, and whether NAD+ truly declines with age — is in B vitamins.

Urolithin A (Mitopure)

A postbiotic — a gut-bacterial metabolite of the ellagitannins in pomegranate, berries and walnuts — and the clearest dietary activator of mitophagy, the housekeeping process that tags and recycles damaged mitochondria. It also has the best industry-funded randomised-trial data of any novel longevity compound. The ATLAS trial in 88 overweight middle-aged adults found that over four months hamstring strength rose about 10–12% on urolithin A while it fell about 10% on placebo; front-of-thigh (quadriceps) strength did not change significantly. The inflammation marker C-reactive protein (CRP) fell, significantly only at the higher 1,000 mg dose, and peak oxygen uptake (VO₂) rose about 10% in that group but fell just short of significance against placebo. The primary endpoint (peak power) was missed.[36]

Read across the whole literature rather than one trial, the picture thins out: a systematic review of five randomised trials in 236 participants could pool only a single outcome, and that one was inconclusive and rated low certainty.[37]

  • Roughly 1 in 10 people make no urolithin A at all from dietary ellagitannins (pomegranate, walnuts), according to the Spanish group that has mapped these gut-bacteria patterns in people aged 5 to 90 and found age the main thing shaping them; the amount everyone else makes varies widely.[38] [39] The often-quoted figure that only about 40% convert comes from a study funded by the Mitopure maker, after a single glass of pomegranate juice.[40]
  • Most trials, including the largest, were funded by the maker (Amazentis). The few small trials without declared industry ties are mixed,[41] [42] and none studied hard outcomes such as falls, fractures or survival.

Practical: No trial has yet shown benefit in older adults: the one functional trial in that age group (Liu, JAMA Netw Open 2022, n=66, aged 65–90) missed both primary endpoints, and the maker's trial in 26 frail adults aged 65–85 (NCT06556706, structural endpoints only) completed in April 2025 with no results reported by October 2026. For sarcopenia, resistance training and enough protein come first; the pill is an optional experiment at most. For most, eating pomegranate and walnuts is the cheaper bet.

Spermidine

Rich animal data and a striking observational signal — the Bruneck cohort showed highest tertile dietary intake had 39% lower all-cause mortality.[43]

But the SmartAge trial — twelve months in 100 older adults, the largest test of the compound in people — was negative on its primary cognitive endpoint.[44]

Part of the reason may be that the pills never arrive: 15 mg/day for five days did not raise plasma or salivary spermidine at all, leaving it unclear that an oral dose reaches the circulation in the first place.[45]

Practical: Dietary intake from Mediterranean-style eating (legumes, mushrooms, aged cheese, wheat germ, natto) is preferable to wheat germ extract pills.

Quercetin

Supplements of 500–1000 mg/day are typical. A 2016 pooled analysis put the drop in systolic (top-number) blood pressure at about 3 mmHg,[46] but the trials disagree a lot, and the newest overview puts it nearer 2 mmHg and rates the certainty of the evidence very low to moderate.[47] Lecithin "phytosome" forms (e.g. Quercefit) are absorbed far better — up to about 20 times higher blood levels in a 12-person study run by the manufacturer.[48]

The senolytic dasatinib + quercetin protocol from the Mayo Clinic Kirkland group has reduced senescent-cell markers in human pilots, but dasatinib is a prescription leukemia drug — this is not a self-administration regimen. The pilots are also where the good news stops: the first randomised phase-2 readout, intermittent dasatinib plus quercetin in postmenopausal women, missed its primary endpoint.[49]

Caution: Quercetin can change how some drugs are handled. In one small study in people, 500 mg a day for two weeks raised pravastatin levels by about a quarter, through a liver-uptake transporter rather than CYP3A4.[50] Its effect on CYP3A4, the main drug-metabolising liver enzyme, is inconsistent in people and may even speed it up,[51] and there is a single case report of a stronger warfarin effect.[52] Anyone on statins, blood thinners or transplant drugs should check with a pharmacist.

Tier 3: Hyped but weak human evidence

Resveratrol

The cautionary tale of the field. Extreme bioavailability problems: it is well absorbed, but almost entirely broken down on its first pass through the gut and liver, so only traces of intact resveratrol reach the blood.[53] Two decades of human trials, disappointingly inconsistent results.

  • RESHAW (postmenopausal women, 150 mg/day for 12 months) is the most positive trial — small gains in bone density, but bone was a secondary outcome, the trial ran at a single site, and the resveratrol maker supplied the capsules and part of the funding.[54]
  • Yoshino 2012,[55] Poulsen 2013,[56] and others are largely null.

It also failed to extend lifespan in the NIA Interventions Testing Program (ITP) — the same multi-site mouse program where rapamycin and acarbose succeed.[57] Red wine contains too little resveratrol to matter (~0.3–1 mg per glass). Whole-food polyphenol diversity beats isolated resveratrol as a defensible recommendation.

Taurine

Got Science cover treatment with a 10–12% increase in median mouse lifespan, in both sexes.[58] But a 2025 Science paper directly contradicted the central premise, showing circulating taurine does not consistently decline with age in humans, monkeys, or mice.[59] A second 2025 human study reached the same conclusion.[60] Neither tested supplements, so the mouse lifespan result itself stands.

Practical: Currently speculative. No RCT in healthy 35–55-year-olds targeting longevity.

GlyNAC (glycine + N-acetylcysteine)

Most of the human data comes from the Sekhar lab at Baylor — pilot trials showing improvements in glutathione, mitochondrial function, and aging biomarkers in older adults at 100 mg/kg/day of each.

Those trials are small — 12 people per arm in the main randomised one.[61] The only placebo-controlled trial from another group was negative on its main measure. Run by Nestlé, which holds GlyNAC patents, in 114 healthy older adults, it found no rise in glutathione overall. It used a lower dose (up to 7.2 g a day of the two combined) for only two weeks, so it is not a like-for-like replication; only a subgroup picked out after the fact, with high oxidative stress and low glutathione at the start, responded.[62]

NAC alone (600–1800 mg/day) has solid evidence for reducing flare-ups of chronic obstructive pulmonary disease (COPD) and is the standard antidote for acetaminophen overdose. Glycine alone (3 g pre-bed) has small Japanese trials suggesting modest sleep benefit.

Berberine

500 mg, 2–3× daily has strong short-term evidence for glucose and lipid lowering — comparable to metformin in head-to-head trials, with LDL-cholesterol reductions of ~15 mg/dL, apparently by boosting the liver's LDL-clearance pathway (the same pathway PCSK9-inhibitor drugs act on).

But berberine has no diabetes-prevention RCTs, no cardiovascular outcome trials, and no proven longevity equivalence to metformin despite being marketed as "natural metformin."

Caution: In a small human study (healthy men, 300 mg three times a day for two weeks), berberine slowed several drug-clearing liver enzymes: blood levels of a CYP3A4 test drug rose about 40%, a CYP2D6 test-drug marker shifted about ninefold, and CYP2C9 was also inhibited.[63] It may also block P-glycoprotein, a drug-transport pump. Interactions are plausible with statins, cyclosporine, dabigatran, many antidepressants, beta-blockers such as metoprolol, warfarin, and many blood-pressure medications, so have a pharmacist check your medicines. Gastrointestinal side effects in 10–35% of users.

Low-dose lithium orotate

The human evidence is thin. Areas with more lithium in their drinking water have lower suicide rates, but these are population-level comparisons, the studies disagree a lot, and publication bias may inflate the link.[64] For dementia the picture is mixed: a 2024 systematic review found only five studies, most showing lower risk,[65] but a Scottish cohort of 37,597 people found no protection (its water lithium levels were very low).[66] A few small trials in people who already had memory problems point to slower decline: two years of low-dose lithium carbonate in 61 older adults with mild cognitive impairment,[67] and a 300 µg daily microdose for 15 months in Alzheimer's disease.[68] No trial has tested lithium in healthy adults, and none has used lithium orotate.

A landmark 2025 Nature paper showed endogenous brain lithium is depleted in mild cognitive impairment and Alzheimer's disease, and that low-dose lithium orotate reversed pathology in mice.[69]

No trial has tested the long-term safety of over-the-counter lithium in people, and products vary widely in dose and form; in a 2025 survey of 211 people taking lithium bought as a supplement, side effects and withdrawal symptoms were more common than expected.[70] Lithium orotate most likely breaks down into ordinary lithium in the stomach,[71] which is cleared by the kidneys and at prescription doses can harm the kidneys and thyroid. Anyone taking non-steroidal anti-inflammatory drugs (NSAIDs), angiotensin-converting-enzyme (ACE) inhibitors or diuretics — which can raise lithium levels — or with kidney or thyroid disease, should talk to a doctor first. The longevity evidence is purely hypothesis-generating.

Astaxanthin

The one antioxidant carotenoid with an actual lifespan signal in the gold-standard mouse program. The NIA Interventions Testing Program — which runs each compound across three sites in genetically diverse mice — found astaxanthin extended median lifespan about 12% in males (p = 0.003).[72] The catch is exactly the kind of detail the supplement market omits: the benefit was male-only — female lifespan was not significantly affected — a reminder that even a compound with a real signal can act in a sex-specific way rather than uniformly. Human data is limited to skin photoprotection (4–12 mg/day) and some metabolic markers; no human longevity outcomes.

Fisetin

A flavonoid marketed aggressively as an elite senolytic. The senescent-cell-clearing claims rest on preclinical data. Human trials are still running, some since 2018 without reported results. The only placebo-controlled trial with posted results, in 74 people with knee osteoarthritis, found two short high-dose courses no better than placebo on knee pain, function, cartilage imaging or inflammatory markers.[73] In a Mayo Clinic bone study in postmenopausal women, fisetin did not lower a marker of bone breakdown compared with no treatment (the difference was not formally tested).[74] And when the ITP tested fisetin head-on, it did not extend lifespan in either sex at the dose and schedules used.[75] Worth tracking as senolytic research, not worth treating as an established longevity supplement. See Hallmarks of aging for the senescence picture.

Calcium alpha-ketoglutarate (Ca-AKG) — Weak (human); preliminary

A caloric-restriction-mimetic angle: αKG is a central metabolite of the cell's main energy-producing (tricarboxylic acid, TCA) cycle and a cofactor for the enzymes that maintain the epigenome, and tissue αKG falls with age. Starting Ca-AKG in 18-month-old mice extended females' median remaining lifespan by about 10–17% across two groups of mice and compressed late-life illness; males gained about 10–13%, which was not statistically significant. Several authors had filed a patent application and own shares in the company that sells the Rejuvant supplement.[76] The headline human data is a retrospective analysis of 42 customers taking Rejuvant (Ca-AKG plus vitamins), reporting an average ~8-year reduction in TruAge epigenetic age over ~7 months.[77] That result is striking but uncontrolled, unblinded, and unreplicated — DNA-methylation clocks are noisy at the individual level and prone to regression-to-the-mean artifacts without a placebo arm.

Practical: Mechanistically reasonable, but the human evidence is hypothesis-generating only. A 120-person placebo-controlled trial in Singapore (ABLE) has finished collecting data, but its results are not yet published.[78]

Sulforaphane (broccoli-sprout extract) — Weak

The prototypical dietary activator of Nrf2 — the master switch that turns on the cell's own antioxidant and detoxification enzymes. Human trials exist but stop at surrogate and disease-symptom endpoints, not aging outcomes — in a 45-child autism trial the main behaviour score did not differ from placebo, with only a caregiver-rated secondary scale improving,[79] and in a 74-person prediabetes trial the extract missed its primary glucose target, lowering fasting glucose by about 0.2 mmol/L; an exploratory analysis linked response to gut bacteria.[80] No lifespan, healthspan, or aging-biomarker outcomes in humans, and bioavailability varies widely with the myrosinase content of the preparation.

Practical: Eating cruciferous vegetables (broccoli, broccoli sprouts) is a defensible whole-food bet; isolated extracts have no longevity evidence above that.

What to actually do

For a healthy 35–55-year-old, the realistic options are:

  1. Skip the geroprotector category entirely — focus on sleep, exercise, diet, and the core supplements. This captures most of the realistic benefit at much lower cost and risk.

  2. For those willing to experiment — a minimal "biology-aware" approach:

    • Optional creatine (already in the core supplements)
    • Optional spermidine via diet (legumes, mushrooms, aged cheese)
    • Optional polyphenol-rich diet (berries, EVOO, dark chocolate, tea)
    • These are food, not supplements
  3. For those drawn to pharmacological geroprotectors — work with a clinician knowledgeable in this area, not a marketing site. Rapamycin, SGLT2 inhibitors and GLP-1 receptor agonists carry the strongest biological signals — the last two with outcome trials only in people with disease — and all have meaningful side effects.

The "longevity stack" sold online — NMN, resveratrol, urolithin A, spermidine, lithium orotate, plus 10 other compounds — is currently unsupported by human RCT evidence and represents the worst kind of cocktail-of-everything approach.

Cautions and interactions

This is the one category on the site where the agents are pharmacologically active enough to hurt you, and several are prescription drugs being used off-label. The headline risks:

  • CYP3A4 is the common thread. Rapamycin, berberine, and grapefruit all act on or through the CYP3A4 enzyme (quercetin's effect on it in people is inconsistent), which also metabolises several statins, many calcium-channel blockers, several anticoagulants, and most immunosuppressants. Stacking them is how people get unexpected drug levels. Any of these alongside prescription medication needs a pharmacist's review, not a forum post.
  • Rapamycin — an immunosuppressant. Contraindicated in pregnancy and active infection; avoid live vaccines; impairs wound healing (stop before elective surgery). Monitor lipids, fasting glucose or HbA1c, and CBC.
  • Metformin — B12 depletion with chronic use (check it periodically), GI intolerance, and a blunting of the gains from exercise — fitness and mitochondrial adaptations to aerobic training, and muscle gain from strength training — that is a genuine argument against it for an active adult. Contraindicated in significant renal impairment.
  • Berberine — inhibits several drug-metabolising enzymes (CYP2D6, CYP2C9, CYP3A4) and is a genuine hypoglycaemic; combining it with metformin, a sulfonylurea, or insulin risks hypoglycaemia. Contraindicated in pregnancy.
  • Dasatinib (in senolytic protocols) — a chemotherapy agent, with the bleeding, cytopenia, and pleural-effusion profile that implies. It has no place in unsupervised use.
  • NAD⁺ precursors (NMN, NR) — the honest caution is not a known toxicity but an unknown one: no long-term human safety data, and a theoretical concern about feeding an existing occult malignancy that has not been resolved either way.
  • High-dose niacin — not separate from the NAD⁺ precursors above: niacin, NR and NMN all end up as the same breakdown products, 2-PY and 4-PY. Higher levels went with more cardiovascular events in observational data, and 4-PY caused vascular inflammation in mice; no trial has yet shown that raising them causes harm.[81]

Anyone on anticoagulants, immunosuppressants, antidiabetics, or with a cancer history should treat this entire category as prescription territory.

Further reading

  • López-Otín C et al. Hallmarks of aging: An expanding universe. Cell 2023.[82]
  • Mannick JB et al. TORC1 inhibition enhances immune function and reduces infections in the elderly. Sci Transl Med 2018.[83]
  • Ferrell M et al. A terminal metabolite of niacin promotes vascular inflammation and contributes to cardiovascular disease risk. Nat Med 2024.[84]
  • Singh A et al. Urolithin A improves muscle strength, exercise performance, and biomarkers of mitochondrial health in a randomized trial in middle-aged adults. Cell Rep Med 2022.[85]
  • Schwarz C et al. SmartAge: Spermidine and Cognition. JAMA Network Open 2022.[86]
  • Yoshino M et al. Nicotinamide mononucleotide increases muscle insulin sensitivity in prediabetic women. Science 2021.[87]
  • Moel M et al. Influence of rapamycin on safety and healthspan metrics after one year: PEARL trial results. Aging 2025.[88]
  • Stanfield B et al. Exercise and Weekly Sirolimus (Rapamycin) in Older Adults: RAPA-EX-01 Randomised, Double-Blind, Placebo-Controlled Trial. J Cachexia Sarcopenia Muscle 2026.[89]
  • Aron L et al. Lithium deficiency and the onset of Alzheimer's disease. Nature 2025.[90]
  • Asadi Shahmirzadi A et al. Alpha-ketoglutarate, an endogenous metabolite, extends lifespan and compresses morbidity in aging mice. Cell Metab 2020.[91]
  • Demidenko O et al. Rejuvant®, a potential life-extending compound formulation with alpha-ketoglutarate and vitamins, conferred an average 8 year reduction in biological aging, after an average of 7 months of use, in the TruAge DNA methylation test. Aging (Albany NY) 2021.[92]
  • Roark KM, Iffland PH. Rapamycin for longevity: the pros, the cons, and future perspectives. Front Aging 2025.[93]
  • Mannick JB, Lamming DW. Targeting the biology of aging with mTOR inhibitors. Nature Aging 2023.[94]

— § —