What 18 Trials of NMN, NR, Urolithin A and Spermidine Were Built to Test

Eight trials tested whether NMN, NR, urolithin A or spermidine could make the body work better. One met its goal. One passed a looser test.

---

Why a Trial's Own Goal Matters

NMN, nicotinamide riboside (NR), urolithin A and spermidine are four of the best-known supplements sold for healthy aging. Each has promising animal research behind it, which we cover separately for NAD+ precursors, urolithin A and spermidine. Each has also been tested in people.

"Clinically studied" on a label tells you a trial happened. It does not tell you what the trial was built to find out, or whether it found it.

Every trial names its main question before it starts. That question is its primary endpoint. Everything else a trial measures is secondary, and the more secondary questions a trial asks, the more likely one of them comes up positive by chance. We explained this in detail when rapamycin's first longevity trial missed its primary endpoint.

So we asked two questions of every trial of these four supplements that our own articles cite. What was it built to test? And did that thing move?

How We Chose and Read the 18 Trials

The rule was simple. We took every human trial of NMN, NR, urolithin A or spermidine cited anywhere on Longevity Science Daily as of 11 September 2026. That gave 19 papers describing 18 trials, because one small spermidine trial published its memory results and its safety results separately (7, 8).

For each trial we recorded the main goal as the paper states it. Where a paper names no primary endpoint, we used the one in its public registration, the record a trial files with a registry such as ClinicalTrials.gov. We read 14 papers in full and 5 from the abstract and registration, because their full text is paywalled. Every result quoted here was checked word for word against the saved source.

This is not a systematic review. A PubMed search on 11 September 2026 tagged 22 NMN, 31 NR, 25 urolithin A and 41 spermidine records as randomized trials in people, and many of those test something else. The 18 below are the better-known trials, not all of them.

Three Questions a Trial Can Ask

A supplement trial can ask three kinds of question. Each is harder to pass than the one before.

1. Is it safe, and does it get into the blood?
2. Does it change a measure inside cells or tissue?
3. Does it change how the body works? Walking, strength, blood sugar control, memory.

The third is the reason people buy these supplements. The first two are steps on the way there.

Archival Photograph, Early 19th century

Louis asked the third question about bloodletting nearly two centuries ago (20). Here is where these 18 trials landed on it.

The Scorecard

A P value appears in several rows. It is the chance of seeing a gap at least this large if there were genuinely no effect. Below 0.05 is conventionally called statistically significant. Who paid for each trial is set out further down, in the Who paid section and under Funding Transparency.

Built to change how the body works

Trial	People, dose and length	Main goal	Did it move?
NR, 2018 (1)	40 men with obesity, 2,000 mg a day, 12 weeks	insulin sensitivity	No
NR, 2020 (2)	13 adults with overweight or obesity, 1,000 mg a day, 6 weeks each on NR and placebo	insulin sensitivity and muscle mitochondrial function	No. Glucose uptake P = 0.98
NMN, 2021 (3)	25 women with prediabetes, 250 mg a day, 10 weeks	muscle insulin sensitivity	Yes. Up 25 percent on NMN, unchanged on placebo
NR, NICE trial, 2024 (4)	90 adults with peripheral artery disease, 1,000 mg a day, 6 months	six-minute walk	Yes for NR alone, under the trial's looser rule: 17.6 meters more than placebo, one-sided P = 0.08. No for NR with resveratrol
Urolithin A, 2022 (5)	66 adults aged 65 to 90, 1,000 mg a day, 4 months	six-minute walk, and energy output in a hand muscle	No, on both. The goals were changed after the trial began
Urolithin A, ATLAS, 2022 (6)	88 adults aged 40 to 65 with overweight, 500 or 1,000 mg a day, 4 months	peak power on a bike test	No
Spermidine pilot, 2018 (7, 8)	30 adults aged 60 to 80 worried about their memory, 3 months	memory	No formal test
Spermidine, SmartAge, 2022 (9)	100 adults aged 60 to 90 worried about their memory, 12 months	memory	No. P = 0.47

Built to change a measure in cells or tissue

Trial	People, dose and length	Main goal	Did it move?
NR, 2019 (10)	12 men aged 70 to 80, 1,000 mg a day, 21 days each on NR and placebo	NAD+ and mitochondrial function in muscle	No. Muscle NAD+ was 210 against 197 pmol/mg, P = 0.22
Urolithin A, 2025 (11)	50 adults aged 45 to 70, 1,000 mg a day, 4 weeks	immune cell types and their metabolism	Partly. One group of naive-like CD8 T cells rose, P = 0.0437. T memory stem cells did not change

Built to test safety or blood levels

Trial	People, dose and length	Main goal	Did it move?
NR with pterostilbene, 2017 (12)	120 adults aged 60 to 80, 8 weeks	safety	Passed. Adverse events were similar across groups
NR, 2018 (13)	30 adults aged 55 to 79, 1,000 mg a day, 6 weeks each on NR and placebo	NAD+ and four related molecules in blood cells	Partly. NAD+ rose about 60 percent against placebo. Three of the five molecules did not reach statistical significance
NMN, 2023 (14)	80 adults aged 40 to 65, 300, 600 or 900 mg a day, 60 days	NAD in blood	Yes, at all three doses, P of 0.001 or lower
NMN (MIB-626), 2023 (15)	32 adults aged 55 to 80 with overweight or obesity, 14 days	none named in the abstract	NAD rose
Urolithin A, 2019 (16)	older adults, one dose or 4 weeks	safety	Passed, by the authors' account. The abstract gives no numbers
Urolithin A against pomegranate juice, 2022 (17)	100 adults aged 18 to 80, one dose of each, open label	urolithin A in blood after 24 hours	Yes. 2.4 times higher on the supplement
Spermidine, 2024 (18)	37 men aged 50 to 70, 40 mg a day, 28 days	safety	Passed on safety. Blood polyamine levels barely changed
Spermidine and vaccines, 2026 (19)	40 adults over 65 after a third COVID vaccine, 6 mg a day, 13 weeks	safety, and antibody levels	Safety passed. The rise in antibodies was not statistically significant

What the Pattern Says

Safety and blood levels are the easy part

The four trials built to test safety reported no safety concern, over four to 13 weeks (12, 16, 18, 19). NAD rose in all four trials that measured it in the blood (12, 13, 14, 15). A urolithin A supplement put 2.4 times more urolithin A into the blood than a glass of pomegranate juice (17).

That is useful to know. These products get absorbed, and short courses caused no measured harm in the people studied. A label that says "clinically studied" can mean no more than this.

Spermidine is the exception on absorption. A 40 mg daily dose for four weeks changed blood polyamine levels very little (18).

The blood is also not the muscle. In 12 men aged 70 to 80, three weeks of NR did not raise NAD+ in their muscle (10). The trial also found no detectable difference in how well their muscle mitochondria used oxygen. A related molecule did rise, which is why the paper's title says NR "augments the aged human skeletal muscle NAD+ metabolome".

Changing how the body works is where they stall

Eight trials aimed at the question people care about. This is how they came out.

• One met its goal. NMN raised muscle insulin sensitivity by 25 percent in 25 women with prediabetes over 10 weeks, and it did not change on placebo (3). It is one small trial in one group of people, and none of the trials here repeats it.
• One passed a looser test. In 90 people with peripheral artery disease, NR alone added 17.6 meters to a six-minute walk compared with placebo (4). The trial set its own bar in advance at a one-sided P below 0.10, and it passed at 0.08. At the conventional 0.05 it would have missed. NR combined with resveratrol missed the trial's own bar.
• Five missed. Two NR trials found no statistically significant improvement in insulin sensitivity (1, 2). Two urolithin A trials found no statistically significant improvement in walking, hand muscle energy or peak power (5, 6). The largest spermidine trial found no statistically significant change in memory after a year (9).
• One ran no formal test. The 2018 spermidine pilot reported its memory result as an effect size (7). Its 95 percent confidence interval, the range of values compatible with this data, ran from minus 0.01 to 0.35. Zero is the no-difference point, so no difference at all still fits.

The headline often comes from a secondary result

In several of these trials, the result in the title or the conclusion is not the one the trial was built to test.

• The 2022 urolithin A trial in older adults concludes that urolithin A benefited muscle endurance (5). Endurance was a secondary endpoint by the end. It had started as a primary endpoint, and its advantage over placebo was no longer statistically significant at four months, because the placebo group improved too.
• The ATLAS trial's title says urolithin A "improves muscle strength". Strength, about 12 percent higher, was a secondary result, and the paper calls all of its endpoints exploratory (6).
• In the 2023 NMN trial, walking distance also improved, and a blood-based age score stayed flat on NMN while it rose on placebo (14). That age score is a laboratory score, a stand-in for aging rather than aging itself. We explain the limits of such scores in what a biological age test measures. The paper counts both results as secondary.
• The 2026 spermidine trial's title says spermidine "boosts vaccine responses" (19). Across everyone, the rise in antibodies was not statistically significant. The gains came from people whose antibodies were low at the start, and those people were 8 of the 20 on spermidine but only 2 of the 18 on placebo.
• SmartAge's authors point to possible benefits for verbal memory and inflammation from exploratory analyses, and say those need confirming at higher doses (9).

The goal on file and the goal in the paper can differ

Trials register their plans publicly, ideally before the first person enrolls. In six of these 18, the registration and the paper do not line up.

• The 2022 urolithin A trial replaced its primary endpoints after the study began, and the paper says so (5, NCT03283462).
• The 2025 urolithin A trial measured immune cell metabolism with a different method from the one it registered, and the paper says so (11, NCT05735886).
• The 2023 NMN trial registered blood NAD, the six-minute walk and a health questionnaire as primary outcomes, plus a second part of the study testing telomeres. The paper names blood NAD as its only primary outcome and does not mention the telomere part (14, NCT04823260).
• The 2018 NR trial in blood cells registered adverse events as its primary outcome. The paper calls five NAD-related molecules its primary outcomes. The trial was registered in late September 2016, as the study was ending (13, NCT02921659).
• The 2017 NR trial registered blood pressure, safety blood tests and heart rate as its primary outcomes. The paper calls safety its primary objective, then calls NAD+ "the major efficacy endpoint" (12, NCT02678611).
• The 2026 spermidine trial began recruiting in September 2021. Its registration was first submitted in June 2022 (19, NCT05421546).

None of these makes a result wrong. Each is a reason to read the registration next to the paper.

Who paid

A company that sells the compound paid for 9 of the 18 trials, fully or in part. In 7 more, a company supplied the product, and in 3 of those, authors held a financial interest in, or a paid role with, that same company. The last 2 list company staff among their authors and do not state their funding in the abstract.

How involved the funder was varied widely. In the 2022 urolithin A trial, the funder "had a role in the design and conduct of the study; collection, management, analysis, and interpretation of the data" (5). In the 2020 NR trial, the company that supplied the capsules "had no role in the study design, data collection, analysis, or preparation of the manuscript" (2).

Company money does not make a result false. It can shape which questions a trial asks, and which result ends up in the title.

What This Means If You Take One of These

If you take NMN, NR or urolithin A, the trials here support two things. The supplement raises its target in your blood, and short courses caused no measured harm in the people studied.

They do not yet support taking one of these four to walk farther, get stronger, control blood sugar better or protect your memory. The one clear positive result on how the body works, for NMN, came from 25 women with prediabetes over 10 weeks. For the wider picture, see the supplement landscape.

Three questions help with any supplement study you read about:

1. What was it built to test? Look for "primary endpoint" or "primary outcome" in the abstract, or search for the trial on ClinicalTrials.gov.
2. Did that thing move? If the headline is a secondary result, give it less weight.
3. Who paid? Look for the funding statement near the end of the paper.

If you have diabetes, peripheral artery disease or memory problems, talk to your doctor before starting any of these. Several of these trials studied exactly those conditions, and none tested a supplement as a replacement for treatment.

Frequently Asked Questions

Do NMN or NR supplements work?

They raised NAD in the blood in every trial here that measured it (12, 13, 14, 15). Whether that changes how the body works is less clear. Five trials here aimed at function or muscle. One NMN trial improved muscle insulin sensitivity in 25 women (3). One NR trial passed a looser test on walking in people with peripheral artery disease (4). Three found no statistically significant improvement (1, 2, 10). How the two precursors differ is covered in NMN vs NR.

Does urolithin A improve muscle?

Not on the main measures so far. Both trials built to improve muscle performance missed their primary endpoints (5, 6). Each reported a secondary benefit: better endurance at two months in one, and about 12 percent more muscle strength in the other. Amazentis, the company that sells it, paid for both.

Does spermidine improve memory?

The largest trial here found no statistically significant difference. SmartAge gave 100 adults aged 60 to 90 with memory worries spermidine or placebo for a year, and memory scores did not differ to a statistically significant degree, P = 0.47 (9). The smaller 2018 pilot before it reported a promising effect size but ran no formal test (7). Both used about 1 mg a day, and the SmartAge authors suggest testing higher doses.

What is a primary endpoint?

It is the one outcome a trial names, before it starts, as the question it was built to answer. Everything else it measures is secondary. The more secondary outcomes a trial checks, the more likely one looks positive by chance, which is why the primary endpoint carries the most weight.

Are these supplements safe?

The four trials built to test safety reported no safety concern, over four to 13 weeks (12, 16, 18, 19). The longest trial here, SmartAge, ran for a year and reported adverse events balanced between the groups (9). That is not long-term safety data. If you take other medicines, ask your doctor first.

Sources

1. Dollerup OL, Christensen B, Svart M, et al. A randomized placebo-controlled clinical trial of nicotinamide riboside in obese men: safety, insulin-sensitivity, and lipid-mobilizing effects. Am J Clin Nutr. 2018;108(2):343-353. PMID: 29992272. DOI: 10.1093/ajcn/nqy132
2. Remie CME, Roumans KHM, Moonen MPB, et al. Nicotinamide riboside supplementation alters body composition and skeletal muscle acetylcarnitine concentrations in healthy obese humans. Am J Clin Nutr. 2020;112(2):413-426. PMID: 32320006. DOI: 10.1093/ajcn/nqaa072
3. Yoshino M, Yoshino J, Kayser BD, et al. Nicotinamide mononucleotide increases muscle insulin sensitivity in prediabetic women. Science. 2021;372(6547):1224-1229. PMID: 33888596. DOI: 10.1126/science.abe9985
4. McDermott MM, Martens CR, Domanchuk KJ, et al. Nicotinamide riboside for peripheral artery disease: the NICE randomized clinical trial. Nat Commun. 2024;15(1):5046. PMID: 38871717. DOI: 10.1038/s41467-024-49092-5
5. Liu S, D'Amico D, Shankland E, et al. Effect of Urolithin A Supplementation on Muscle Endurance and Mitochondrial Health in Older Adults: A Randomized Clinical Trial. JAMA Netw Open. 2022;5(1):e2144279. PMID: 35050355. DOI: 10.1001/jamanetworkopen.2021.44279
6. Singh A, D'Amico D, Andreux PA, 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;3(5):100633. PMID: 35584623. DOI: 10.1016/j.xcrm.2022.100633
7. Wirth M, Benson G, Schwarz C, et al. The effect of spermidine on memory performance in older adults at risk for dementia: A randomized controlled trial. Cortex. 2018;109:181-188. PMID: 30388439. DOI: 10.1016/j.cortex.2018.09.014
8. Schwarz C, Stekovic S, Wirth M, et al. Safety and tolerability of spermidine supplementation in mice and older adults with subjective cognitive decline. Aging (Albany NY). 2018;10(1):19-33. PMID: 29315079. DOI: 10.18632/aging.101354
9. Schwarz C, Benson GS, Horn N, et al. Effects of Spermidine Supplementation on Cognition and Biomarkers in Older Adults With Subjective Cognitive Decline: A Randomized Clinical Trial. JAMA Netw Open. 2022;5(5):e2213875. PMID: 35616942. DOI: 10.1001/jamanetworkopen.2022.13875
10. Elhassan YS, Kluckova K, Fletcher RS, et al. Nicotinamide Riboside Augments the Aged Human Skeletal Muscle NAD+ Metabolome and Induces Transcriptomic and Anti-inflammatory Signatures. Cell Rep. 2019;28(7):1717-1728.e6. PMID: 31412242. DOI: 10.1016/j.celrep.2019.07.043
11. Denk D, Singh A, Kasler HG, et al. Effect of the mitophagy inducer urolithin A on age-related immune decline: a randomized, placebo-controlled trial. Nat Aging. 2025;5(11):2309-2322. PMID: 41174221. DOI: 10.1038/s43587-025-00996-x
12. Dellinger RW, Santos SR, Morris M, et al. Repeat dose NRPT (nicotinamide riboside and pterostilbene) increases NAD+ levels in humans safely and sustainably: a randomized, double-blind, placebo-controlled study. NPJ Aging Mech Dis. 2017;3:17. PMID: 29184669. DOI: 10.1038/s41514-017-0016-9
13. Martens CR, Denman BA, Mazzo MR, et al. Chronic nicotinamide riboside supplementation is well-tolerated and elevates NAD+ in healthy middle-aged and older adults. Nat Commun. 2018;9(1):1286. PMID: 29599478. DOI: 10.1038/s41467-018-03421-7
14. Yi L, Maier AB, Tao R, et al. The efficacy and safety of β-nicotinamide mononucleotide (NMN) supplementation in healthy middle-aged adults: a randomized, multicenter, double-blind, placebo-controlled, parallel-group, dose-dependent clinical trial. Geroscience. 2023;45(1):29-43. PMID: 36482258
15. Pencina KM, Lavu S, Dos Santos M, et al. MIB-626, an Oral Formulation of a Microcrystalline Unique Polymorph of β-Nicotinamide Mononucleotide, Increases Circulating Nicotinamide Adenine Dinucleotide and its Metabolome in Middle-Aged and Older Adults. J Gerontol A Biol Sci Med Sci. 2023;78(1):90-96. PMID: 35182418
16. Andreux PA, Blanco-Bose W, Ryu D, et al. The mitophagy activator urolithin A is safe and induces a molecular signature of improved mitochondrial and cellular health in humans. Nat Metab. 2019;1(6):595-603. PMID: 32694802. DOI: 10.1038/s42255-019-0073-4
17. Singh A, D'Amico D, Andreux PA, et al. Direct supplementation with Urolithin A overcomes limitations of dietary exposure and gut microbiome variability in healthy adults to achieve consistent levels across the population. Eur J Clin Nutr. 2022;76(2):297-308. PMID: 34117375. DOI: 10.1038/s41430-021-00950-1
18. Keohane P, Everett JR, Pereira R, et al. Supplementation of spermidine at 40 mg/day has minimal effects on circulating polyamines: An exploratory double-blind randomized controlled trial in older men. Nutr Res. 2024;132:1-14. PMID: 39405978. DOI: 10.1016/j.nutres.2024.09.012
19. Alsaleh G, Ali M, Kayvanjoo AH, et al. Spermidine Mitigates Immune Cell Senescence and Boosts Vaccine Responses in Healthy Older Adults-A Pilot Study. Aging Cell. 2026;25(6):e70545. PMID: 42169618. DOI: 10.1111/acel.70545
20. James Lind Library. Louis PCA (1835). Recherches sur les effets de la saignée dans quelques maladies inflammatoires. jameslindlibrary.org

Funding Transparency

LSD is editorially independent. We receive no funding from pharmaceutical, supplement or longevity companies. These are the funding ties behind each trial above, taken from each paper's own funding and conflict statements, or from the trial registration where the full paper was unavailable.

• Source 1 (NR, 2018): Funding is not stated in the abstract, and the full text was unavailable. ChromaDex supplied the nicotinamide riboside, according to the trial registration.
• Source 2 (NR, 2020): Dutch Heart Foundation, the European Union's Horizon 2020 programme and the European Research Council. ChromaDex provided the capsules and had no role in design, analysis or writing.
• Source 3 (NMN, 2021): US National Institutes of Health grants. Oriental Yeast provided the NMN and placebo capsules. One author receives part of the patent-licensing fees from MetroBiotech and Teijin through Washington University.
• Source 4 (NR, NICE, 2024): American Heart Association. ChromaDex manufactured the NR and placebo pills.
• Source 5 (urolithin A, 2022): Amazentis SA, which sells urolithin A as Mitopure. The funder had a role in the design, the analysis, the interpretation and the decision to publish. One author was an Amazentis employee during the study, and another chaired its board.
• Source 6 (urolithin A, ATLAS, 2022): Amazentis SA. Six authors are Amazentis employees.
• Sources 7 and 8 (spermidine pilot, 2018): German federal research funding and the German Research Foundation. The Longevity Labs provided the capsules. Four authors have a financial interest in the company.
• Source 9 (spermidine, SmartAge, 2022): German federal research funding. The Longevity Labs supplied the extract and provided funds to develop it during the study. Three authors held equity and advisory roles with the company, and another was its chief executive while the trial ran. The paper states the funders had no role in design, analysis or writing.
• Source 10 (NR, 2019): UK Medical Research Council, Wellcome, a Marie Sklodowska-Curie grant, the Roy J. Carver Trust and the US National Institutes of Health. ChromaDex supplied the NR. One author holds patents licensed to ChromaDex, owns its stock and advises it.
• Source 11 (urolithin A, 2025): Amazentis SA provided the product and study funds, and co-designed the trial. The paper states the funders had no role in data collection, analysis or writing. Three authors are Amazentis employees, and one is its president.
• Source 12 (NR with pterostilbene, 2017): Elysium Health, which markets the product as Basis. Six of the seven authors are Elysium employees who own shares in the company.
• Source 13 (NR, 2018): US National Institutes of Health. ChromaDex provided the pills and partial funding.
• Source 14 (NMN, 2023): Fully funded by Aba Chemicals and Abinopharm, which make the NMN tested. Three authors are their employees.
• Source 15 (NMN, MIB-626, 2023): Funding is not stated in the abstract, and the full text was unavailable. Two of the seven authors list Metro International Biotech as their institution.
• Source 16 (urolithin A, 2019): Funding is not stated in the abstract, and the full text was unavailable. Five of the nine authors list Amazentis as their institution.
• Source 17 (urolithin A against pomegranate juice, 2022): Amazentis SA. Five authors are Amazentis employees, and others sit on its board or scientific advisory board.
• Source 18 (spermidine, 2024): Sponsored by Chrysea Labs. Two authors are Chrysea employees, and the other four received funding from the company.
• Source 19 (spermidine and vaccines, 2026): Arthritis UK, Wellcome, UK SPINE, The Longevity Labs, the Helmholtz Association, the University of Graz and the Austrian Science Fund. The Longevity Labs supplied the product, and the paper states it "has had no input into the trial design, results, analysis or manuscript". Two authors consult for the company.
• Source 20: a historical record, with no funding relationship.

Related Reading

• NAD+ Supplements, the fuller story on NAD+ precursors and why a blood level is not the same as a benefit
• NMN vs NR, how the two NAD+ precursors differ
• Urolithin A and Mitophagy, what urolithin A does inside cells
• Spermidine and Longevity, the autophagy research behind spermidine
• Rapamycin's First Longevity Trial Missed Its Primary Endpoint, the same question asked of a drug
• The Supplement Landscape, which supplements have human evidence behind them at all

---

A trial's main question is public, dated, and usually one search away. Check it before the headline.

---

Written with the help of AI tools, shaped and verified by humans. Every result in this article was checked against the trial's own paper, abstract or registration.

This article is for information only and is not medical advice.