Nicotinamide adenine dinucleotide — NAD+ — is one of the more elegant molecules in metabolism. Every cell uses it. Levels fall with age. Restore levels in a mouse and the mouse looks younger by several metabolic measures. This is the story that turned NMN and NR into a billion-dollar supplement category.
The story is real. The clinical translation to humans is much thinner than the marketing implies. NAD+ levels do rise in humans on both NR and NMN. What does not clearly happen: measurable improvements in the functional endpoints — endurance, strength, insulin sensitivity, cognition — that would justify the price.
This is the honest walk-through of what the human trials actually show.
Why NAD+ matters, in one paragraph
NAD+ is a cofactor for hundreds of enzymes. Two families matter most for the aging conversation. The sirtuins (SIRT1 through SIRT7) are NAD+-dependent deacetylases involved in metabolic regulation, DNA repair, and stress response. The PARPs (poly-ADP-ribose polymerases) consume NAD+ during DNA damage repair. Both consume NAD+ faster under stress. Both slow down when NAD+ is scarce.
NAD+ levels in human tissue decline by roughly 20 to 50 percent between young adulthood and age 60, depending on tissue and measurement method (Massudi et al. 2012; Zhu et al. 2015 MRI-based measurements). Nobody debates the decline. The debate is whether raising NAD+ back up with an oral pill actually reverses the age-related dysfunction it correlates with.
NR versus NMN: the mechanism difference
Both are NAD+ precursors. They enter the salvage pathway at slightly different points.
- Nicotinamide riboside (NR). Enters cells directly via nucleoside transporters. Once inside, phosphorylated to NMN by NRK enzymes, then converted to NAD+. Sold as Niagen (ChromaDex). Regulatory status: recognized as a dietary supplement.
- Nicotinamide mononucleotide (NMN). One step closer to NAD+ than NR. The question of how NMN enters cells was debated for a decade — Sinclair's group argued for a specific transporter called Slc12a8 (2019 paper). Other labs contested that finding. Most researchers now believe NMN is dephosphorylated to NR before entering cells, then rephosphorylated inside. If true, oral NMN and oral NR are pharmacokinetically similar.
Practical takeaway: both compounds raise NAD+ levels in human blood. Whether one is meaningfully better than the other at raising NAD+ in specific tissues has not been settled by direct head-to-head trials.
What the human RCTs actually show
Here are the main trials most commonly cited. Read them together and the pattern is consistent: NAD+ metabolites go up; the clinical endpoints mostly do not move.
Martens et al. 2018 (NR, healthy middle-aged and older adults). 500 mg NR twice daily for six weeks. Whole blood NAD+ roughly doubled. Modest reduction in systolic blood pressure in a subgroup with elevated baseline. No significant change in arterial stiffness, physical function, or metabolic markers overall.
Elhassan et al. 2019 (NR, older men). 1000 mg NR daily for 21 days. NAD+ metabolites up. No significant improvement in insulin sensitivity, mitochondrial function, or muscle performance despite adequate exposure.
Dolopikou et al. 2020 (NR, young and old adults, acute). Single-dose NR. Some acute improvement in oxidative stress markers in older subjects. Interesting mechanistic signal, no chronic outcome data from this trial.
Yoshino et al. 2021 (NMN, prediabetic postmenopausal women, Washington University). 250 mg NMN daily for ten weeks. Modest improvement in skeletal-muscle insulin sensitivity as measured by hyperinsulinemic-euglycemic clamp. No change in body composition, blood pressure, or several other metabolic endpoints. One of the more optimistic NMN readouts, and it is still narrow.
Yamanashi trial (Irie et al. 2020, NMN, healthy men). Single ascending doses up to 500 mg. Safe, well-tolerated, raised NAD+ metabolites. Not designed to measure clinical endpoints.
Yi et al. 2023 (NMN, healthy middle-aged adults, dose-ranging). 300, 600, and 900 mg daily for 60 days. NAD+ rose dose-dependently. Six-minute walk distance improved, particularly in the 600 and 900 mg groups. This is one of the cleaner functional signals in the NMN literature, and it comes from an industry-adjacent group. The reader is entitled to some skepticism.
Igarashi et al. 2022 (NMN, healthy older men, exercise-training context). 250 mg NMN daily during a 12-week training program. Some improvement in aerobic capacity gains over training alone, though effect sizes were modest.
Dellinger et al. 2017 (Basis, an NR/pterostilbene combo). Eight weeks. Raised NAD+ by roughly 40 percent. No clinical endpoints reported.
Read across the list. The biomarker (NAD+ level) reliably moves. The endpoints people actually care about — strength, endurance, cognition, insulin sensitivity, healthspan measures — move inconsistently, with small effect sizes, in narrow populations. The largest positive signals come from industry-adjacent trials. None of these are longevity trials; nobody has run a randomized trial long enough to measure mortality or age-related disease outcomes.
Sinclair's dog data versus his human data
David Sinclair, the Harvard geneticist whose books popularized this field, has published striking preclinical data. Old mice given NAD+ precursors show improved endurance, muscle function, and vascular density. His companies have extended some of this work into dogs, where a randomized trial of an NMN-derivative reported functional improvements in old dogs.
Extrapolating from mouse and dog data to human outcomes is where the whole longevity field gets over its skis. Mice on caloric restriction live 30 to 40 percent longer. Rhesus monkeys on caloric restriction show only modest lifespan improvement. Humans have never been shown to live longer on caloric restriction. Species matter. The metabolic gap between a two-year-old lab mouse and a 55-year-old human is enormous.
None of this proves NAD+ precursors do not work in humans. It proves the mouse data is not evidence of human efficacy. The human evidence is what is on the list above.
The FDA reversal on NMN, and why oral NMN is regulatorily unstable
In late 2022, the FDA quietly reversed course on NMN. The agency had previously acknowledged NMN as a dietary ingredient. It then concluded that NMN had been "authorized for investigation as a new drug" prior to being marketed as a supplement, which under the FD&C Act triggers a preclusion. Under the current interpretation, NMN cannot legally be sold as a dietary supplement in the US.
Enforcement has been inconsistent. Some retailers pulled NMN. Many did not. Product remains on shelves. The commercial supply chain is exposed to future enforcement, class action, and reformulation risk in a way that NR — which does not share the same drug-authorization history — is not.
This is not the same thing as NMN being unsafe. Human safety data through 900 mg per day for 60 days is reassuring. It does mean the market for oral NMN sits on regulatory sand.
Cost per gram — the number the marketing hides
NR sold as Niagen runs roughly $1 to $2 per 300 mg dose, or about $3 to $6 per gram. NMN pricing has fallen since 2020 but remains higher on quality-controlled product, often $2 to $5 per gram.
Trials used doses of 250 to 1000 mg per day. A middle-of-the-road 500 mg NR daily protocol costs $50 to $100 per month. NMN at 500 mg daily runs similar. Comparable clinical exposures for a year are $600 to $1200.
For that same annual budget you could purchase five years of magnesium glycinate, three years of creatine monohydrate, a lifetime supply of protein powder, and a Withings sleep mat. Each of those has stronger clinical evidence for functional endpoints than the NAD+ precursor category currently offers.
The niacin question people never ask
Niacin (nicotinic acid) and nicotinamide are the classical NAD+ precursors. Both raise NAD+. Both cost pennies. Niacin has been in the pharmacopeia for eighty years.
The reason NR and NMN exist as products is that regular niacin has a flushing side effect and, at cardiovascular-drug doses, hepatotoxicity risk. Nicotinamide does not flush, but historically was considered inferior at raising NAD+. Modern re-analysis of that assumption is ongoing — a growing body of work suggests plain nicotinamide at modest doses may raise tissue NAD+ competitively with NR and NMN.
If you are set on trying an NAD+ precursor and cost matters, plain nicotinamide at 500 mg per day is $3 a month. The evidence gap between plain nicotinamide and branded NR is smaller than the price gap. This is not a formal recommendation — it is a note that the branded story has priced itself far above the biological plausibility of doing much better than the eighty-year-old vitamin form.
Who might reasonably try NR or NMN
Given the honest read on the evidence, the shortlist of reasonable candidates is narrow.
- Adults over 50 with the disposable income to run a personal experiment. The safety profile through 12 months of use is reassuring. If a $60 per month spend is trivial to you, this is a low-harm high-uncertainty bet.
- Endurance athletes willing to test aerobic capacity endpoints. Some of the more positive signal is on submaximal aerobic performance. If you have a way to measure your own VO2 or six-minute walk, you can generate personal data.
- Nobody who is skipping foundational interventions. If you are not sleeping seven hours, not lifting weights twice a week, not walking 8000 steps a day, not eating protein at each meal, do not spend $60 a month on NAD+. The foundational interventions have effect sizes an order of magnitude larger.
How to run a 90-day personal trial
If you have decided to try. Ninety days is a reasonable window — long enough for NAD+ levels to plateau, short enough that you have not sunk two years of budget.
- Choose one compound. NR for regulatory stability. NMN if you accept the regulatory risk. Not both.
- Choose a well-defined dose from the trial range. 300 to 600 mg per day is where most positive signal sits. Higher doses have not shown clearly better outcomes.
- Baseline for two weeks before starting. Track resting heart rate, HRV, subjective energy on a 1 to 10 scale, and a functional endpoint you can actually measure (grip strength, six-minute walk pace, or your normal weekly training benchmark).
- Run at fixed dose for 90 days. Same time each day, ideally with a light meal.
- At day 90, compare final two weeks to baseline two weeks. If the numbers moved, keep going another 90 days and see if the improvement holds. If the numbers are indistinguishable, the compound is not doing anything for you specifically, and the $180 you spent bought certainty rather than benefit. Certainty is worth something.
The honest verdict
NAD+ precursors are probably safe at consumer doses through the exposure windows humans have tested. NAD+ levels reliably rise. Clinical endpoints move inconsistently, with small effect sizes, in narrow populations, sometimes only in industry-linked trials. Longevity claims are extrapolated from mouse data that has not translated to any species in between.
The category is expensive relative to its evidence. It has been marketed with confidence that the trial data does not support. If you can afford it and enjoy the experimentation, run a disciplined ninety-day trial and use your own data to decide. If your budget is finite, spend it on the interventions with cleaner outcomes: sleep, weight training, dietary protein, magnesium, creatine.
Boring interventions still beat exciting ones. That has not changed.