A supplement raises a molecule associated with cellular energy. That sounds like a breakthrough until we ask the less glamorous question: what, exactly, improved in the person taking it? Nicotinamide mononucleotide (NMN) has become a remarkably clear test of how easily a measurable biochemical result can be mistaken for longer life.
Originally published: 7 June 2023 · Substantially updated: . Scientific literature review and editorial update by Ilya Nikitin, Ph.D. The original author credit is retained.
The finding: oral NMN can increase blood concentrations of NAD⁺ and related metabolites. The gap: controlled human studies have not demonstrated that NMN extends lifespan, reverses ageing or consistently improves metabolic health. New randomised evidence published in September 2026 reinforces that distinction.
A biomarker is not a benefit. The strongest reproducible human observation is increased blood NAD-related metabolites. A higher reading is not proof of better mitochondrial function, preserved independence or slower ageing.
What NMN does and what NAD⁺ is for
NAD⁺ (nicotinamide adenine dinucleotide) helps cells transfer electrons in redox reactions and supports energy metabolism, DNA maintenance and signalling. NMN is an intermediate in the pathways that make NAD⁺. That gives scientists a sound reason to investigate NMN. It does not yet tell us what happens to the health of someone who takes it.
The case for NMN sounds persuasive because it moves through three ideas so quickly. NAD⁺ matters to cells. Some NAD-related measures decline with age in certain tissues or conditions. So raising a blood marker must slow ageing. The first is established biology; the second depends on where and how it is measured. The final leap still needs proof in people.
The surprising result: blood NAD⁺ may not decline with normal ageing
Even the starting assumption deserves a closer look. In May 2026, researchers reported measurements across seven independent human cohorts, using a method designed to address analytical variability. Whole-blood NAD⁺ concentrations were remarkably stable across age and several lifestyle interventions, although they responded to nicotinamide riboside supplementation. Blood NAD⁺, it seems, is not automatically an ageing clock.
This does not prove that NAD⁺ never falls with age in muscle, brain or specific cellular compartments. Whole blood is not a substitute for those tissues. It does mean that the widely repeated phrase ‘NAD falls as we age’ needs a location, a measurement method and a population before it can function as scientific evidence.
A separate 2026 Redox Biology study likewise found broadly stable blood NAD-related measures across healthy age groups. It studied blood redox biology, not whether NMN extends life; its niacin-treatment results should not be mistaken for an NMN trial. Tissue, assay and sample handling all matter. A chart of blood NAD⁺ is not a map of biological ageing.
How the promise fared in human trials
2021: a promising result, with an important limit
To understand why NMN attracted attention, rewind to 2021. In a randomised, double-blind trial, 25 postmenopausal women with prediabetes and overweight or obesity received NMN (250 mg/day) or placebo for ten weeks. The NMN group showed improved skeletal-muscle insulin sensitivity measured by a metabolic clamp (a controlled test of how muscle responds to insulin), alongside changes in insulin-signalling markers. Unlike a change in blood NAD⁺ alone, this was a measurable physiological response.
It was a promising finding, but in a small, specific population and a physiological measure, not a demonstration of diabetes prevention, fewer cardiovascular events or longer life. A published critique pointed to differences in baseline liver fat between the treatment groups. The authors replied that baseline muscle insulin sensitivity was comparable and the result remained interpretable. The disagreement did not erase the finding; it showed why independent tests mattered.
2022: more NAD⁺ and suggestive functional findings
A year later, another study offered hints of functional improvement. A 12-week randomised trial in older men used 250 mg/day. Blood NAD-related metabolites rose, and some gait-speed and handgrip outcomes showed nominally significant improvements; body composition did not improve significantly. But ‘nominally significant’ is a caution sign when many outcomes are tested: isolated positives may appear by chance. The authors called for larger trials to see whether these findings would hold up.
2024: eight trials, 342 adults, no consistent metabolic improvement
By 2024, the story could be tested against more than individual, encouraging experiments. A systematic review and meta-analysis pooled eight randomised trials involving 342 adults. Doses ranged from 250 to 2,000 mg/day and trials lasted from 14 days to 12 weeks. Across the included populations, NMN did not significantly improve fasting glucose, fasting insulin, HbA1c, HOMA-IR (an estimate of insulin resistance calculated from fasting glucose and insulin) or lipid profiles. This does not cancel the targeted physiological result from 2021, particularly as many participants in later trials were relatively healthy. It does weaken the idea of a dependable glucose benefit for most consumers.
July 2026: a wider safety and metabolic review
One important caution before the next review: the 2024 and 2026 meta-analyses partly reuse the same underlying trials. A larger stack of review papers does not mean a fresh set of participants each time.
A 2026 review in Nutrients included 15 trials of NMN or NMN-related interventions; ten contributed to pooled safety analyses, covering 383 participants (230 in NMN groups and 153 controls). The studied regimens ran from 14 days to 24 weeks, with doses from 250 to 2,000 mg/day. The authors did not find a significant increase in overall reported adverse events or a consistent improvement in weight, BMI, fasting glucose, HbA1c, lipid measures or systolic blood pressure. For any adverse event, the pooled risk difference was −0.8 percentage points (95% confidence interval: −6.1 to +4.5 percentage points). The estimate is compatible with little difference but cannot rule out a moderately higher risk.
There were hints of benefit: a small fall in diastolic blood pressure and a statistically uncertain trend in HOMA-IR. Neither establishes a clinical benefit. The trials also tested different populations and formulations, including NMN-related preparations, rather than one standardised product.
September 2026: the molecule rose, but the main outcomes did not
The next trial brought the question back to outcomes that people might actually notice. A randomised trial published on 25 September 2026 assigned adults aged at least 65 to 1,250 mg/day NMN or placebo for 24 weeks. Blood NAD⁺ and related metabolites rose. Yet the primary auditory-function endpoint did not improve significantly, nor did measured physical-performance outcomes. Changes in some secondary measures lost significance after correction for multiple testing.
It is a revealing contrast, not a universal verdict: the study cannot rule out every possible benefit in every population. Funding disclosure: MIRAILAB BIOSCIENCE paid for the trial and supplied NMN; three co-authors were company employees. The paper reports that the company was not involved in study design, data analysis or the decision to publish. That relationship belongs in the interpretation, but is not a reason to discard the findings.
Evidence at a glance
| Claim | What human evidence establishes |
|---|---|
| NMN changes NAD-related biology | Supported: blood NAD-related metabolites rise in several trials. |
| NMN improves glucose control for most adults | Not established: pooled trials did not show reliable improvement. |
| NMN improves strength or daily function | Preliminary and inconsistent; no dependable clinical effect. |
| NMN reverses biological age | Not established; no validated demonstration of reversal in people. |
| NMN extends human lifespan | Not tested in trials capable of establishing the claim. |
What a non-significant result really means
Statistical uncertainty cuts both ways. A small trial can miss a modest real effect because of its size, variability or choice of endpoints. But testing many outcomes also increases the chance of an apparently positive result. So ‘not significant’ does not mean ‘definitely no effect’, just as a scattered positive result does not establish a benefit. Neither ‘miracle’ nor ‘fraud’ captures what the human evidence shows.
What about muscle, sleep, cognition and testosterone?
Could NMN still help with particular aspects of daily life? Small studies have reported encouraging results for gait speed, grip strength and selected sleep or fatigue scores. These outcomes come closer to how people actually feel and function, but the experiments vary in what they measure and often last only weeks. Across broad populations, a consistent improvement has not been demonstrated.
NMN has also been marketed as a way to raise testosterone. Controlled human evidence does not currently support a reliable testosterone increase in men. We examined the specific studies and interpretation of a personal blood-test experience in our separate NMN and testosterone analysis. One person’s before-and-after laboratory values cannot isolate the supplement from biological variation, timing, illness or changes in behaviour.
How safe is NMN? And what does a safety trial actually prove?
Most short-duration randomised studies have reported broadly favourable tolerability, and the 2026 pooled analysis did not find a clear excess of reported adverse events or liver-enzyme elevations. This is reassuring within the observed doses, populations and periods. It is not evidence that daily use for decades is harmless.
Trial participants are typically selected, adverse events may be uncommon, and six months cannot detect very late outcomes. People with significant illness, multiple medications, pregnancy or other special circumstances are not well represented by the consumer claim that ‘NMN is naturally present in the body, so it is safe’. A natural metabolic intermediate can still have dose-dependent or context-dependent effects.
Questions about tumour biology and the effects of chronically changing NAD-related metabolism remain mechanistically interesting, but neither a cancer-prevention benefit nor a cancer-causing effect from usual NMN supplementation is established by human trials. Treating theoretical concerns as proven harms would be no more scientific than treating animal longevity findings as human proof.
Dose, timing and the price of uncertainty
- 250–2,000 mg/day: a range used in published trials, not a validated longevity dose or evidence that more is better.
- 14 days to 24 weeks: approximate duration range in the 2026 safety/metabolic review, not the length of follow-up needed to demonstrate long-term protection.
- No proven optimum: human outcome trials have not established that taking NMN in the morning, while fasting, sublingually or in a liposomal formulation extends healthspan compared with other schedules.
Manufacturers may emphasise purity, absorption or a larger rise in NAD⁺. Those are product properties, not demonstrated gains in healthy years. Cost matters: a hypothetical €40 monthly supply adds up to €480 a year. With no established longevity benefit, that is a real expense against an unproven return; it is an illustration, not a surveyed retail price.
What about pairing NMN with resveratrol or other ‘longevity stacks’?
Combining two mechanistically attractive compounds multiplies the number of hypotheses, not the amount of demonstrated benefit. Our resveratrol review applies the same discipline: exposure, laboratory effects and clinical endpoints must be kept distinct. Without well-controlled combination trials, claiming synergy for longevity is speculation.
What kind of trial would change the conclusion?
A useful experiment would recruit a clearly defined population, randomise participants to NMN or placebo, prespecify a small set of clinically meaningful outcomes, measure adherence and adverse events, and follow participants long enough to distinguish durable improvement from measurement noise.
For functional ageing, plausible outcomes include objective walking or strength performance, disability, falls or well-validated clinical endpoints. Ultimately, mortality and disease incidence are more decisive but need far larger and longer studies. A change in blood NAD⁺ can confirm that a biochemical intervention occurred; it cannot replace the outcome.
The measurement checklist: Which tissue was sampled? How reliably was the molecule measured? And did anything meaningful change for participants? A biochemical result alone cannot answer all three questions.
Bottom line
Should a healthy adult buy NMN today for anti-ageing? Current controlled human evidence does not justify recommending it for that purpose. NMN changes NAD-related biology, and a small 2021 metabolic study deserves follow-up, but pooled evidence and a 24-week trial have not demonstrated consistent functional or metabolic gains, let alone longer life. Anyone considering supplementation should weigh an uncertain benefit, ongoing cost and unknown long-term safety against that evidence.
Sources and key trials
- Trętowicz et al., Nature Metabolism (May 2026): whole-blood NAD⁺ stability across seven human cohorts.
- Yoshino et al., Science (2021): NMN and muscle insulin sensitivity in 25 women with prediabetes.
- Published comment on the 2021 trial: imbalance in baseline liver fat; authors’ response.
- Igarashi et al. (2022): NMN, NAD⁺ and selected muscle-performance outcomes in older men.
- Chen et al. (2024): meta-analysis of eight RCTs, 342 adults, metabolic endpoints.
- Yang et al., Nutrients (July 2026): safety and metabolic meta-analysis covering 15 studies.
- Igarashi et al., npj Aging (25 September 2026): 24-week 1,250 mg/day placebo-controlled trial.
- Euro et al., Redox Biology (October 2026): NAD⁺ and blood redox measurements across ages and conditions.
- Gallagher and Emmanuel (2026): review comparing NAD-related interventions in humans and rodents.
