NMN: Human Trials, Safety and Regulatory Status

NMN is a molecule involved in NAD synthesis. Several short human trials, many manufacturer-linked, report higher blood NAD-related measures, while functional benefits remain mixed. A small trial found improved muscle insulin sensitivity in a narrow population; it does not establish general anti-aging benefit. Product-label testing has found discrepancies, and financial independence must be assessed separately. Long-term safety and interactions remain uncertain. People with cancer, pregnancy or relevant medicines should seek clinical guidance before considering it (Yoshino et al. 2021; Wen 2024).

Table of contents

• Evidence summary
• What NMN is
• Forms and formulations
• How NMN works
• What works and what does not
• Benefits with evidence grades
• Risks and reported side effects
• Medicines and interaction uncertainties
• Who should seek clinical guidance
• Dosage and how to take NMN
• Infographics with full text versions
• Related research
• Frequently asked questions
• Sources and funding notes

Evidence summary

What NMN is

Nicotinamide mononucleotide (NMN) is an intermediate in NAD synthesis, also reported in small amounts in some foods. NAD participates in energy metabolism and is used by enzymes involved in DNA repair and other cellular processes. Age-associated changes vary by tissue and study; a mechanism or low NAD measurement does not show that NMN supplements reverse human aging (Song 2023, Advances in Nutrition).

NMN differs chemically from nicotinamide riboside, niacin and nicotinamide, which enter NAD synthesis through different routes. It was marketed as a US dietary supplement in the 2010s under a regulatory framework, rather than being “unregulated.” A November 2023 trade-media report quoted Antlerzz marketing-firm estimates of a 176% rise in US NMN-related searches over the preceding two years. That dated search estimate measures neither current supplement use nor clinical benefit (Longevity.Technology, 2023).

Several short trials report increased blood NAD-related measures after oral NMN, but assays, doses and tissue responses differ. The reviewed evidence does not establish reversed aging, longer lifespan or disease prevention. Animal findings can explain research hypotheses; they cannot substitute for human clinical outcomes (Freeberg 2023; Wen 2024).

Where the marketing narrative comes from

Public advocacy and personal supplement routines are not human efficacy evidence. David Sinclair has a commercial stake: the Pencina 2023 original trial disclosed that he was a Metro International Biotech consultant and equity owner. A 2026 supplement-seller account of his routine is commercial secondary material, not an independent verification of his protocol or NMN’s benefits. This article uses trial results and their disclosures rather than a personal stack to judge clinical claims.

Metro International Biotech developed a proprietary crystalline NMN drug candidate, MIB-626, and asked FDA to retain NMN’s supplement exclusion in a November 2023 submission. The company’s own letter documents its commercial and regulatory position; it is an interested-party account, not independent evidence of efficacy or proof that it alone determined FDA policy. FDA’s later decision explains the statutory timing test without publicly disclosing confidential IND authorization dates. (Metro letter; FDA petition decision)

Forms and formulations

How it works

NMN participates in pathways that build NAD, which supports redox reactions and serves as a substrate for enzymes such as sirtuins and PARPs. Age-associated NAD changes and potential supplementation effects depend on the tissue, population and assay; biochemical plausibility is not an established anti-aging treatment (Song 2023).

In Yoshino’s 2021 trial, 250 mg/day for 10 weeks increased blood-immune-cell NAD and selected muscle NAD-related metabolites, while muscle NAD itself did not increase. Muscle insulin sensitivity rose 25% ± 7% within the NMN arm, with a significant treatment interaction, in 25 postmenopausal women with prediabetes. Another mechanistic trial by Pencina used 2,000 mg/day for 28 days: blood NAD rose, but muscle NAD and the measured physical/metabolic performance endpoints did not improve. Different tissue measurements should not be collapsed into a general organ-restoration claim.

Human absorption mapping remains incomplete. Animal transporter proposals do not establish how much intact oral NMN reaches human tissues versus conversion into other precursors. Several short human studies report downstream blood NAD-related changes, using different methods; Yi’s trial measured serum total NAD plus NADH with a colorimetric assay (Freeberg 2023; Yi 2023).

Blood NAD-related measurements are not interchangeable with NAD inside muscle, brain or liver. Yoshino and Pencina reported increased circulating measures without increased muscle NAD by their respective methods. Neither study established restored youthful function across organs. A circulating biomarker is not a validated substitute for the clinical benefit being marketed.

NAD synthesis uses several precursors, including tryptophan, niacin, nicotinamide and nicotinamide riboside. Sharing biochemical pathways does not mean an ordinary dietary allowance is equivalent to an experimental NMN dose or that the compounds have equal clinical effects. The cited human NAD-precursor review does not establish a clinical winner from a matched NMN-versus-NR or niacin comparison.

What works and what does not

Blood NAD-related measures — increases in several short trials

Yi 2023, online in 2022 randomized 80 adults to 300, 600 or 900 mg/day or placebo for 60 days. Serum total NAD plus NADH increased versus placebo at days 30/60; 600 and 900 mg gave the largest responses without a significant difference between those doses. Okabe and Igarashi reported whole-blood NAD increases using different protocols. These findings do not establish a universal dose response or benefit in every tissue.

Insulin sensitivity — one positive signal, not yet replicated at scale

The Yoshino et al. 2021 Science trial directly measured skeletal-muscle insulin sensitivity with NMN, and it found a 25% ± 7% improvement from baseline in 13 NMN-treated postmenopausal prediabetic women, with a significant treatment interaction and 12 placebo controls over 10 weeks (Yoshino et al. 2021). That is a genuine, mechanistically coherent finding — but it is a single 25-person trial in one narrow population (postmenopausal, prediabetic, overweight/obese women), and independent human-trial evidence is insufficient to generalize this to men, younger adults, or people without prediabetes.

Physical performance (strength, walking speed, endurance) — mostly non-significant or modest

Wen’s 2024 review reported ten studies but included Connell’s non-NMN tryptophan/niacin/nicotinamide trial and did not perform a meta-analysis. Its 437-person census, dose range and descriptive before/after averages should not be treated as pooled NMN results. The original NMN trials differed in population, endpoints and design; selected findings are described separately below.

• Igarashi 2022 randomized 42 older men, but a supplier error left only 10/arm with usable 12-week data. The 1.30 versus 1.60 m/s values are placebo versus NMN endpoint means, not a before/after increase. Exploratory models gave gait p=0.033 and left-grip p=0.019 without multiplicity correction; the unadjusted gait-change comparison was p=0.066. Muscle mass did not improve.
• Akasaka 2023’s original abstract describes 14 older men with diabetes and impaired physical performance, 250 mg/day for 24 weeks, with no between-group improvement in grip or walking speed and no significant exploratory-outcome differences. Teijin/NOMON affiliations appear in the publisher record; complete methods and funding/conflict sections were not available for verification here.
• Kim 2022 studied 108 older Japanese adults, not an obese-only population. At 250 mg/day for 12 weeks, the sit-to-stand interaction was p=0.04, but all four timing/treatment groups improved. Sleep, fatigue and performance results do not establish a dependable NMN-specific functional benefit. Mitsubishi funded the trial, supplied product and paid publication costs; three authors were employees.
• Pencina 2023 randomized 30 overweight/obese adults to 2,000 mg/day (1,000 mg twice daily) or placebo for 28 days. Blood NAD increased, but measured strength, fatigability, aerobic capacity and insulin sensitivity did not improve. Selected weight, blood-pressure and lipid findings used nominal, unadjusted P values. Metro funded the study; company employment, consultancy and equity relationships were disclosed.

Liao 2021 reported selected ventilatory-threshold improvements in 48 runners receiving 300/600/1,200 mg/day alongside training for six weeks, without improved VO2max. Threshold oxygen and power outcomes must be distinguished. Yi 2023 reported greater six-minute walking distances and higher SF-36 scores by day 60, with no significant 600-versus-900 mg advantage. Liao received public funding plus GeneHarbor-supplied product; Yi was manufacturer-funded. These different protocols do not establish a generally effective performance dose.

Sleep quality — no consistent human signal

Irie 2020 tested single 100/250/500 mg doses in ten men, with five-hour laboratory/metabolite monitoring and sleep questionnaires before/after the intervention day; sleep scores did not change. This was not a five-hour sleep trial. Kim 2022 studied 108 older Japanese adults for 12 weeks: overall sleep-quality interaction was not significant. Drowsiness had an interaction signal, but both evening NMN and evening placebo improved; sit-to-stand improved in all groups. The results do not establish NMN as a sleep aid or an optimal dosing time.

Longevity and “reversing aging” — not established in humans

The cited human trials do not establish lifespan extension, validated reversal of aging or disease prevention. Short surrogate measurements cannot answer those questions. Mouse findings, including the 2016 institutional report and the 2018 vascular-aging report, are discussed as preclinical research rather than evidence that NMN extends human life.

Telomere length — one very small trial, not confirmed

Niu 2021 gave eight men 300 mg/day dissolved in water for 90 days without blinding or a placebo group, and reported increased relative telomere measurements in blood immune cells. This small uncontrolled observation cannot establish a causal effect, lifespan extension or an anti-aging benefit. Chinese public grants were reported; two authors were ERA Biotech employees. Human product-supply terms were not fully specified.

The FDA regulatory saga in full

Before FDA’s 2022 drug-preclusion determination, NMN was marketed in US dietary supplements. CIRS, a commercial regulatory-services firm, reported a May 2022 no-objection response to SyncoZymes’ NDI notification and a subsequent reversal. NDI notification is not FDA approval of a supplement’s efficacy or blanket approval of all products (CIRS 2022).

FDA’s October 11 and November 4, 2022 letters to Kingdomway stated that NMN was excluded under the drug-preclusion clause. FDA later set both aside after changing how prior supplement marketing is treated in the statutory timing test. This chronology is confirmed directly by its December 2, 2025 letter. (FDA December 2025 letter)

The practical fallout was swift and significant for the supplement industry. In February 2023, Amazon announced it would remove NMN dietary supplements from its platform starting March 13, 2023, citing FDA’s determination directly (NutraIngredients, February 2023). Industry trade groups pushed back immediately: the Natural Products Association (NPA) and the Alliance for Natural Health USA filed a joint Citizen Petition in March 2023 asking FDA to reverse its position or, at minimum, commit to non-enforcement while the issue was resolved. FDA reaffirmed its November 2022 position through the spring and summer of 2023, and in July 2023 explicitly rejected a congressional request (from Rep. Jeff Duncan) for a public hearing on the matter, telling stakeholders to use the formal Citizen Petition process instead (NutraIngredients-USA, July 2023).

Metro’s November 2023 letter requested continued exclusion and described its own NMN drug-development program. The NPA’s August 2024 complaint records its lawsuit against FDA and DOJ; a complaint is a party’s allegation, not a ruling. December 2025 trade reporting describes an October 2024 litigation stay and interim enforcement deprioritization. Those details are secondary reporting here, not verification of the complete court record.

On September 29, 2025, FDA concluded that NMN is not excluded from the dietary-supplement definition under the drug-preclusion clause. The agency accepted evidence of US supplement marketing as early as 2017, before relevant drug-investigation authorization, and reconsidered its earlier view that the prior marketing had to be lawful to count. This decision addresses the exclusion test; separate safety, NDI, adulteration and labelling requirements remain. (FDA petition decision, pages 19–23; FDA December 2025 letter)

NutraIngredients reported ingredient-supplier letters in December 2025 and Amazon relisting in October 2025. The October report predates those December letters and cannot be their consequence. The NPA attributed the outcome to its lawsuit; that is an interested trade association’s account, not independent proof of causation (NutraIngredients, December 2025; NutraIngredients, October 2025; NPA press release).

Venable’s client-facing legal analysis discusses possible challenges to FDA’s interpretation following Loper Bright and continuing consumer-protection obligations. Such commentary provides legal context, not assurance that every NMN product is lawfully marketed. Sellers and buyers need the current official rules applicable to the particular product and jurisdiction (Venable, October 2025).

NMN vs. nicotinamide riboside: no established clinical winner

NR is a related NAD precursor with separate trials and regulatory assessments. A longer research history does not establish clinical superiority. Manufacturer support in either field needs source-specific disclosure; it does not make a result unusable or independently financed (Freeberg 2023).

Several studies of NMN and NR report increased blood NAD-related measures, but results differ by assay, preparation, population and endpoint. The 2023 human NAD-precursor review does not establish equal clinical effects or a universally superior compound. Short tolerability findings and selected physiological results do not settle comparative disease-prevention, longevity or functional benefits. Its authors disclosed NIH support and ChromaDex material-transfer agreements, so the review is not a blanket independent-finance clearance.

Benefits by claim

Risks and reported side effects

Reviewed short trials did not identify treatment-related serious events, but mild events occurred. Okabe reported one NMN-related mild abdominal-pain episode. Fukamizu’s four-week 1,250 mg/day trial had no events judged treatment-related; its negative Ames result is a bacterial assay, not a human cancer test. Wen’s mixed 437-person set includes a non-NMN trial and cannot supply an NMN-specific adverse-event rate (Wen review).

The trials assessed here were generally short and do not establish year-long use or safety in pregnancy, lactation, active cancer or significant organ disease. Funding and author conflicts are disclosed separately; short tolerability findings must not be promoted as proof of long-term safety. (Wen 2024 review)

The cancer question — a documented concern, not a documented harm

NAD metabolism matters in cancer biology, but Fujita’s 2024 surgical-patient study did not administer NMN. A 2026 pancreatic-cancer cell/mouse study found reduced sensitivity to oxaliplatin, 5-fluorouracil and gemcitabine with NMN. It did not test radiation or establish effects in human patients. Institutional coverage supports discussing supplements with an oncologist during cancer care; it does not make this a proven clinical interaction or a universal contraindication after any past cancer.

Medicines and interaction uncertainties

The cited trials and reviews do not establish a formal clinical drug-interaction profile for NMN. Limited short trials cannot certify compatibility with a medicine class. The examples below identify questions for the prescriber; mechanisms and missing data are not proof that a specific interaction occurs (human NAD-precursor review).

Who should seek clinical guidance

• People receiving cancer care — discuss NMN with the oncology team because of preclinical chemotherapy findings. A past cancer diagnosis alone is not an established universal contraindication (original abstract; research coverage).
• Pregnancy or breastfeeding — the reviewed adult trials and EFSA’s specified-ingredient safety conclusion do not cover these groups; do not infer safety from adult tolerability (EFSA 2026).
• Children and adolescents — the cited adult trials do not establish pediatric safety or benefit. EFSA’s assessed use is for adults (EFSA 2026).
• Significant liver or kidney disease — the reviewed trials do not establish safety in major organ impairment, and several excluded important chronic conditions. Seek clinical advice rather than extrapolate healthy-adult findings (Fukamizu; Okabe).
• Immunosuppressive treatment — ask the prescriber about the actual medicines and supplement. The cited evidence does not characterize a clinical interaction; a general immune-metabolism mechanism is not proof of one (Freeberg review).
• Anyone expecting longer lifespan or “de-aging” — the reviewed human evidence does not establish these outcomes; a blood biomarker change is insufficient (human review).

Dosage and how to take NMN

NMN has no established nutritional RDA or clinically proven anti-aging dose. Reviewed protocols include single 100/250/500 mg doses in Irie, 1,250 mg/day for four weeks in Fukamizu, and 2,000 mg/day as 1,000 mg twice daily for 28 days in Pencina. Other cited trials used 250 mg/day or 300/600/900 mg/day. These research doses do not establish a recommended range, optimal dose or safety for routine long-term use.

Short trial doses are not a recommended range or proof of safety for routine use. EFSA’s assessment concerns a specified high-purity ingredient at up to 300 mg/day in adults excluding pregnancy and lactation. For any contemplated use, check applicable regulation, formulation, batch testing and clinical risks; do not infer benefit from a blood biomarker alone (EFSA 2026).

The product-quality problem

Sandalova’s blinded study tested 18 NMN and five urolithin-A products. Two NMN products with dose claims and one without an NMN label claim were not quantifiable; extraction limitations complicate some formulations. Funding was unstated. ConsumerLab’s public summary describes a separate ChromaDex commercial assay in which 14 of 22 products had less than 1% of the claimed NMN. It is not ConsumerLab’s own independent 22-product experiment; neither sample estimates market-wide prevalence.

Global regulatory status outside the United States

The UK Food Standards Agency’s RP-2116 record lists Effepharm’s novel-food application as in progress at risk assessment, last modified 1 September 2026. That status does not establish completed authorisation. Check current applicable rules for the specified ingredient/use rather than infer legality from online availability.

EFSA adopted its NMN opinion on 4 March 2026, first published it on 11 May and amended it on 8 July. It found Effepharm’s specified ≥99%-pure ingredient safe at up to 300 mg/day for adults excluding pregnancy/lactation, using applicant-supplied proprietary data; it did not assess anti-aging efficacy. The Commission register records draft D118035/01 submitted 10 September for a 24 September 2026 committee meeting. This draft and the EFSA opinion do not themselves verify that final authorisation is in force.

Rules and permitted uses differ by jurisdiction and specified ingredient. Regulatory safety assessment, authorisation and evidence of clinical benefit answer separate questions; online availability does not settle any of them. (Primary source)

Infographics with full text versions

Infographic 1: The FDA’s NMN regulatory reversal, step by step

Infographic 2: What NMN human trials actually measured vs. what marketing claims

Related research

NMN’s proposed metabolic and insulin-sensitivity effects connect to Pure City Research’s reviews of related conditions: for readers focused on diabetes prevention and blood sugar management, the insulin-sensitivity signal from the Yoshino et al. trial is the most relevant human data point, though it should not replace established interventions. For readers evaluating NMN alongside cardiovascular risk factors, see Pure City’s heart disease prevention guide for evidence and funding limitations on cardiometabolic supplements. Readers considering NMN for sleep support should first review Pure City’s sleep prevention guide, since the reviewed NMN trials did not establish a reliable sleep benefit. Anyone considering NMN while managing a current or past cancer diagnosis should read Pure City’s cancer prevention guide alongside a conversation with their oncology team, given the unresolved NAD+/tumor-metabolism question discussed above.

Frequently asked questions

Sources and funding notes

Animal and non-human evidence excluded

In-vitro evidence used

Human efficacy conclusions are based on human trials, with commercial support and author ties disclosed. Animal and cell work is used only to explain theoretical mechanisms and safety questions; bacterial Ames findings are laboratory data. None proves human cancer benefit, cancer harm or long-term safety. (Wen 2024 review)

Human studies are distinguished from animal and cell research. Manufacturer funding, product supply and author patent or commercial ties are disclosed separately from study design and outcomes. Company-linked results are reported as context and do not establish independent efficacy, general safety or longer lifespan. Long-term outcome and interaction evidence remains limited.

• FDA — NMN citizen-petition decision, September 2025
• FDA — Kingdomway NDI letter, December 2, 2025