Home > Articles > NMN human trials: what randomized studies have measured
NMN human trials: what randomized studies have measured
Prepared by: Research Desk
Prepared by: Research Desk. Sources are linked for verification.
Nicotinamide mononucleotide (NMN) is widely marketed, but the human research behind it is narrower than the marketing suggests. This page asks one question: in randomized trials in adults, what did researchers actually measure, and what did they report? It lists endpoints, study designs and limits. It makes no promises about outcomes. For the underlying biochemistry, see our NAD+ guide. For NMN's legal position in the U.S., see Is NAD+ legal?.
Why researchers run NMN trials at all
NMN is an intermediate in NAD+ synthesis, so it is studied as a precursor [2]. Research interest starts with observations in human tissue. A 2012 study of human tissue samples reported age-associated changes in markers of oxidative stress and in NAD+ metabolism [1].
NAD+ metabolism is also tightly regulated. Synthesis, recycling and consumption happen in different cell compartments and tissues, which makes it hard to target with any single molecule [2].
The research case for NMN therefore rests on a chain of assumptions. Each link has to be tested separately:
Oral precursor intake raises NAD+ or its metabolites in blood.
Higher blood levels reflect higher levels in tissues such as muscle or liver.
Tissue-level changes alter a measurable physiological function.
That functional change matters over the long term.
Among the trials discussed below, most focus on link 1, a few reach link 3, and none was long enough to address link 4.
The endpoint ladder
It is useful to sort trial outcomes by how far they are from the molecule itself:
Tolerability and safety labs: adverse events, liver and kidney markers, routine blood chemistry.
Circulating NAD+ and metabolites: levels in whole blood, plasma or blood cells.
Tissue measures: NAD+-related metabolites or gene expression in biopsied tissue.
Physiological function: insulin sensitivity, walking tests, blood pressure, lipids.
Composite and self-reported measures: "biological age" estimates and quality-of-life questionnaires.
The lower rungs are easier to measure and tend to be reported more consistently. The upper rungs are what the marketing talks about, and they are where trial results vary most.
Trial 1: muscle insulin sensitivity in women with prediabetes
A randomized, placebo-controlled trial published in Science in 2021 enrolled postmenopausal women with prediabetes who had overweight or obesity [3]. Its main endpoint was insulin sensitivity in skeletal muscle, measured with a hyperinsulinemic-euglycemic clamp. This is a demanding laboratory method, not a single blood draw.
The authors reported that NMN increased muscle insulin sensitivity and markers of insulin signaling compared with placebo [3]. The finding is narrow, and reading it accurately means keeping its scope in view:
one tissue (skeletal muscle)
one population (women with prediabetes)
a small sample
a study period measured in weeks
It is a physiological signal in a specific group. It is not evidence about other populations or about long-term outcomes.
Trial 2: a multicenter trial in healthy middle-aged adults
A 2023 trial in Geroscience was randomized, multicenter, double-blind and placebo-controlled, with a parallel-group design. It enrolled 80 healthy middle-aged adults and ran for about two months [4]. It was designed to compare more than one intake level against placebo.
The study measured:
blood NAD concentration
six-minute walking distance
a blood-based "biological age" estimate
an insulin-resistance index (HOMA-IR)
a general health questionnaire (SF-36)
safety measures
The authors reported that blood NAD concentrations rose in the NMN groups compared with placebo. They also reported changes in walking distance and found no safety concerns over the study period [4]. They did not report a clear between-group difference in the insulin-resistance index.
Three limits apply. Self-reported and composite measures are sensitive to expectation effects. "Biological age" algorithms are not standardised. And two months is short.
Trial 3: a physiologic study in overweight and older adults
A 2023 study in the Journal of Clinical Endocrinology & Metabolism tested an oral NMN formulation in middle-aged and older adults with overweight or obesity [5]. The authors described it as a physiologic study.
It measured NAD+ and related metabolites in blood alongside cardiometabolic markers such as body weight, blood pressure and lipids. The authors reported higher circulating NAD+ and metabolite levels. They framed the cardiometabolic observations as preliminary findings that call for larger and longer trials [5].
That framing matches a broader review asking whether NAD+-raising strategies affect cardiometabolic health at all. The question mark in that review's title reflects real uncertainty in the field [10].
What reviews of the NMN trial record conclude
Several reviews have looked across the human NMN studies:
Song et al. (2023) summarised published NMN clinical trials. They reported that NMN was generally well tolerated in the short-term studies available, and described the evidence on aging-related outcomes as still limited, calling for larger, longer trials [6].
Nadeeshani et al. (2022) set the "promises" of NMN against open safety questions. They noted that long-term human safety data are lacking [7].
Gallagher et al. (2026) used a PRISMA-guided systematic review to examine both preclinical and clinical evidence on NAD+ supplementation [8]. Keeping those two evidence types apart matters, because a result in cells or animals is not the same as a result in people.
NMN trials next to NR trials
Nicotinamide riboside (NR) is the other widely studied precursor. A 2018 randomized trial in healthy middle-aged and older adults reported that NR was well tolerated and raised NAD+ levels [9].
The pattern across both precursors is similar. Blood NAD+ markers move fairly reliably. Functional outcomes are less consistent.
Direct comparisons across trials are risky. Studies sample different compartments (whole blood, plasma, blood cells or muscle) and use different assays. NAD+ concentrations also differ between compartments and tissues [2]. A larger rise in one trial's blood measurement does not mean one precursor "works better" than another.
A checklist for reading any NMN trial
When a new NMN study is reported, these questions sort strong designs from weak ones:
Randomized and blinded? Open-label studies cannot separate expectation effects from real change.
Pre-specified primary endpoint? A trial that measures many outcomes will find some changes by chance.
Which sample? Whole blood, plasma, blood cells and tissue give different NAD+ numbers.
Who took part? Results in women with prediabetes do not automatically apply to healthy adults, and results in healthy adults do not automatically apply to other groups either.
How long did it run? Weeks-long trials say little about long-term safety [7].
How many participants? Small samples produce unstable estimates.
Was the comparison against placebo? Before-and-after changes alone are weak evidence.
Regulatory context in brief
Trial results are not FDA findings. Dietary supplements do not need FDA approval before they are sold, and manufacturers are responsible for safety and truthful labelling [11]. In September 2025, FDA determined that NMN is lawful in dietary supplements, according to a legal analysis of the decision [12]. That decision followed a citizen petition docket [13]. A legal status says nothing about whether any outcome in these trials holds up. The full timeline is on our legal status page.
Tools for comparing NAD+ tests
Blood NAD+ is the endpoint measured most often in these trials. That makes at-home NAD+ blood tests a natural point of interest for readers who want to see what such a measurement looks like in practice. Tests differ in ways that change what a result means, so a good choice starts with these neutral criteria:
What is measured: NAD+ alone, the NAD+/NADH ratio, or a wider metabolite panel.
Sample type: dried blood spot or other formats. This determines which reference values apply.
Validation: whether the lab publishes its method and analytical performance.
Reference ranges: how the comparison population was defined, and how large it was.
Price transparency: the full cost of kit, analysis and any repeat tests, stated up front.
Data privacy: who stores the results, for how long, and whether data are shared.
Turnaround and report format: units shown, explanations included, raw values available.
A test that scores well on validation, clear units and transparent pricing gives the most interpretable number, and the same kit used consistently is easier to compare over time than results from mixed providers.
A consumer test result cannot be placed directly beside a number from a published trial. Research labs used their own assays and sample types, and units often differ.
How this site handles affiliate relationships, sourcing and corrections is set out on our About page.
Frequently asked questions
How many randomized NMN trials have been run in humans?
Reviews describe a modest and growing set of human trials, most of them small and short [6]. Systematic reviews also separate these from the much larger body of animal and cell work [8]. Any headline claim is worth checking against the specific trial it cites.
Do NMN trials show that it raises NAD+ in people?
Several randomized trials reported higher NAD+ or NAD+ metabolite levels in blood after NMN compared with placebo [4][5]. Fewer studies measured tissue directly, and blood levels do not necessarily reflect levels inside organs [2].
Were the trials long enough to judge safety?
Most ran for weeks to a few months. Reviews describe short-term tolerability as generally acceptable in those studies [6]. They also note that long-term human safety data are still lacking [7].
Last updated: 19 October 2026.
Related on this site
NAD+ explained: biochemistry, precursors, testing and human evidence
NAD+ test buyer's guide: how at-home blood tests work and compare
Reading an NAD+ test report: units, sample types and reference ranges
NAD+ books: an evidence-graded reading list on sirtuins and aging
Nicotinamide riboside research: what human trials have measured
Not medical advice. This page reports published research and regulatory records for information only.
Research-only notice: Content on this site is provided for research and educational purposes only and is not intended for human use. It is not medical advice.
Translation notice: Some content on this site may have been machine-translated using Google Translate and may contain errors.