There is a category of supplement that does not promise better skin or more energy. It promises to slow ageing itself — to act on the mechanisms rather than the symptoms. NMN, nicotinamide riboside, resveratrol, spermidine, fisetin, urolithin A. The names are technical, the marketing cites Nature papers, and the prices reflect the implied sophistication.
The question worth asking about all of them is the same, and it is narrower than it sounds: what has been shown in humans, over what timeframe, on what outcome?
Asked that way, the field looks very different from how it is sold.
The claim that no supplement can currently make
Start with the ceiling, because everything else sits under it.
No longevity supplement has been shown to extend human lifespan in a randomised controlled trial. Not one. This is not a controversial statement — it follows from the design constraints. A trial testing whether a compound extends human lifespan would need to run for decades, with tens of thousands of participants, at a cost nobody has been willing to bear.
So every claim in this market rests on something shorter: an effect in mice, an effect on a biomarker, or an observational association. Those are three different kinds of evidence, and none of them is the thing being implied.
NMN: the biomarker moves, the outcomes mostly do not
NMN is the flagship, so it deserves the most detail.
The theory. NAD+ is a coenzyme essential to energy metabolism, DNA repair and the activity of sirtuins. NAD+ levels decline with age across multiple tissues. NMN is a precursor: take it, and NAD+ should rise. In animal models, restoring NAD+ improves mitochondrial function, tissue repair and — in invertebrates — extends lifespan.
It is a genuinely elegant chain of reasoning. The problem is at the last link.
What human trials show. The first half works. Oral NMN reliably raises circulating NAD+ — a head-to-head trial published in Nature Metabolism found NMN and nicotinamide riboside both roughly double blood NAD+ after 14 days, while plain nicotinamide does not.
The second half is where it thins out. A systematic review with meta-analysis of randomised controlled trials examined NMN’s effect on glucose and lipid metabolism, pooling 12 studies with 513 participants. It found a significant effect on blood NAD+ levels — and that most of the clinically relevant outcomes were not significantly different between NMN and control.
Individual trials have reported scattered positives: a 12-week trial in adults aged 65–75 at 250 mg/day found a modest improvement in 4-metre walking time and some sleep quality measures, as secondary outcomes, alongside large individual variation in NAD+ response.
Broader reviews reach a consistent conclusion. NR and NMN raise circulating NAD+ by roughly 130–150%, and short-term studies in older adults show minimal improvement in cognition, vascular function or muscle performance.
Safety looks reasonable in the trials conducted so far, which are short and small. Note that the FDA has taken the position that NMN was authorised for investigation as a new drug before being marketed as a supplement, which has complicated its regulatory status in the US.
The honest summary: NMN does what it says on the mechanism — it raises NAD+. What has not been demonstrated is that raising NAD+ in a reasonably healthy adult produces meaningful functional benefit. Those are different claims and the price is charged for the second.
Resveratrol: the cautionary tale
Resveratrol was the first molecule to get this treatment, and its trajectory is the most instructive thing in this article.
The story was compelling: a compound in red wine that activates SIRT1, the mechanism behind caloric restriction, explaining the French paradox. Enormous scientific and commercial enthusiasm followed.
Human trials have consistently disappointed. Results across metabolic, cardiovascular and inflammatory markers have been inconsistent, and much of the failure is attributed to bioavailability — resveratrol is metabolised so rapidly in the first pass through the liver that achievable oral doses may never reach the tissue concentrations used in cell studies.
Whether resveratrol does not work in humans, or simply cannot be delivered in useful amounts orally, remains open. Either way, the practical answer is the same, and it took roughly fifteen years and a great deal of money to arrive at.
Spermidine: interesting, and genuinely unresolved
Spermidine is the most defensible of the specifically longevity-coded compounds, which is not the same as saying it is proven.
The mechanism is autophagy — the cellular recycling process that clears damaged components, and which declines with age. Spermidine induces it, and extends lifespan across yeast, flies and mice.
The human evidence is split. Observational work has linked higher dietary spermidine intake to lower all-cause mortality in European cohorts. That is suggestive, and it carries every limitation of nutritional epidemiology: people eating spermidine-rich foods differ in many other ways.
Small randomised trials have shown modest cognitive effects in older adults with memory complaints, at around 1.2 mg/day. Modest, small, and not replicated at scale.
Worth knowing: spermidine occurs in ordinary food — wheat germ, aged cheese, mushrooms, legumes. For a compound whose human evidence comes largely from dietary intake studies, eating the foods is at least as defensible as buying the extract, and considerably cheaper.
The rest, briefly
Nicotinamide riboside (NR). Same NAD+ story as NMN, similar evidence, similar conclusion. Raises NAD+ reliably; functional outcomes in humans remain thin.
Fisetin. A senolytic — proposed to clear senescent cells. Compelling data in animal ageing models. Human evidence as of now is limited to small feasibility studies. This is a compound at an early stage being sold at a late-stage price.
Urolithin A. Acts on mitophagy, the clearing of damaged mitochondria. Clinical studies report improvements in mitochondrial and muscle biomarkers — biomarkers, not outcomes, and without the duration or scale to say more.
Metformin and rapamycin are prescription drugs, not supplements, and are the most serious candidates in the field. Rapamycin has the strongest mouse data of any compound tested, extending median lifespan by roughly 10–15% in the NIA’s Interventions Testing Program, even when started late in life. Metformin is the subject of the TAME trial, designed specifically to test healthspan outcomes in non-diabetics. Both carry real side effects and neither is something to acquire without a doctor.
How to read this market
Four questions strip most of the marketing away.
Was it done in humans? Extending lifespan in yeast, worms, flies or mice is a routine result. The overwhelming majority of compounds that do it have never shown anything comparable in people.
Was the outcome something that matters? A supplement that raises a biomarker has demonstrated that it does something. Whether that something makes any difference to your life is a separate question, usually unanswered.
How long, and how many people? Twelve weeks in forty participants tells you about safety and biomarkers. It cannot tell you about ageing, which happens over decades.
Who is the population? NMN’s better results come from trials in older adults with metabolic risk factors — not healthy adults in their forties. Evidence in one group is not evidence in another.
What this section keeps concluding
Set these compounds beside what came earlier in this category and the contrast is hard to miss.
Fibre: 15–30% lower all-cause mortality in the observational data, supported by trial evidence, and achieved by fewer than 6% of older adults in the UK. Protein at 1.2 g/kg with resistance training: a clear mechanism connecting to muscle, and muscle strength predicts mortality better than blood pressure. Cardiorespiratory fitness: a five-fold mortality difference between the top and bottom of the distribution.
None of these is patentable, brandable or exciting. All of them have human evidence that the NAD+ precursors, senolytics and autophagy inducers do not yet have.
That is not an argument that the longevity compounds will never work. Some may. Rapamycin in particular has results that deserve to be taken seriously, and the NAD+ biology is real even if the supplement version has not delivered.
It is an argument about sequence. Spending £60 a month on NMN while eating 18 g of fibre and doing no resistance training is optimising the speculative end while the established end sits untouched — and that describes a large fraction of the people buying these products.
The unglamorous interventions are unglamorous precisely because they have been tested. That is what testing does to a claim.
This article is for general information and is not medical advice. Supplements can interact with prescription medication. If you are considering any of these compounds, particularly alongside existing treatment, speak to your GP or pharmacist.
Sources
- Song Q, Zhou X, Xu K, et al. The safety and antiaging effects of nicotinamide mononucleotide in human clinical trials: an update. Advances in Nutrition, 2023;14(6):1416–1435. https://pmc.ncbi.nlm.nih.gov/articles/PMC10721522/
- Efficacy of oral nicotinamide mononucleotide supplementation on glucose and lipid metabolism for adults: a systematic review with meta-analysis of randomized controlled trials. Critical Reviews in Food Science and Nutrition, 2024. https://www.tandfonline.com/doi/full/10.1080/10408398.2024.2387324
- Christen S, Cuenoud B, et al. Comparison of NAD+ precursors in humans. Nature Metabolism, 2025.
- Yang Y, et al. An updated review on the mechanisms, pre-clinical and clinical comparisons of nicotinamide mononucleotide (NMN) and nicotinamide riboside (NR). Food Frontiers, 2025. https://iadns.onlinelibrary.wiley.com/doi/10.1002/fft2.511
- Eisenberg T, et al. Cardioprotection and lifespan extension by the natural polyamine spermidine. Nature Medicine, 2016.
- Miller RA, et al. NIA Interventions Testing Program results for rapamycin. Aging Cell.
This article is for general information and is not medical advice. If a health problem is affecting your daily life, speak to your GP.
Comments
We read every comment. Be kind, stay on topic, and please don't ask for medical advice — we can't give it.