NAD+: The Number Moves. The Outcome Has Not Been Bought.

The short version

NAD+ is not a peptide and not a drug. It is a coenzyme — a small helper molecule the body makes for itself and uses everywhere, in every cell, all the time. Its main job is carrying electrons through the reactions that turn food into usable energy. Its second job is being consumed as fuel by a set of enzymes that repair DNA, regulate genes and manage inflammation.

Tissue levels of NAD+ fall with age. That single observation is the engine behind an entire supplement category, because the inference is irresistible: if the level falls and the level matters, raising it should help.

This page exists because the second half of that sentence is where the money runs out. Raising the level is genuinely cheap and genuinely well-demonstrated — several controlled trials have done it, and the blood number goes up in a tidy dose-dependent way. Turning that raised number into a measured improvement in how long or how well people live is a completely different purchase, and as of the most recent human synthesis it has not been made.

What it is

NAD+ is nicotinamide adenine dinucleotide: two nucleotides, nicotinamide mononucleotide and adenosine monophosphate, joined through a pair of bridging phosphate groups, carrying a pyridine nicotinamide ring at one end and an adenine ring at the other. Its oxidised form is NAD+ and its reduced form is NADH. The molecular formula is C21H27N7O14P2.

It is not classified as an approved drug. NAD+ and its precursors are sold as dietary supplements, while injectable and intravenous forms are typically compounded rather than manufactured to an approved standard. The regulatory position of nicotinamide mononucleotide has itself been contested, with the FDA taking the view that it is excluded from the dietary-supplement definition because it was investigated as a drug — an unresolved marketplace question rather than a scientific one.

The practical distinction that matters for reading the literature is between NAD+ itself and its precursors. Oral NAD+ is poorly taken up by cells intact, and most researchers treat precursors — nicotinamide riboside (NR) and nicotinamide mononucleotide (NMN) — as the rational oral approach. Nearly every human trial cited on this page tested a precursor, not NAD+.

How it works

Two roles, and they compete with each other for the same pool.

The first role is redox chemistry. NAD+ shuttles electrons through glycolysis, the TCA cycle and oxidative phosphorylation, which is how cells convert fuel into ATP. In this role NAD+ is recycled rather than consumed, cycling continuously between its oxidised and reduced forms.

The second role is as a consumed substrate for signalling enzymes. The sirtuins (SIRT1 through SIRT7) use NAD+ to remove acyl groups from proteins, influencing gene regulation and mitochondrial function. PARP1 uses it in the DNA-damage response. The ectoenzymes CD38 and CD157 consume it too. A foundational review identifies precisely these enzyme families as the major consumers competing for the NAD+ pool, and frames restoring NAD+ as a candidate strategy against age-related disease ^[9].

The mechanistic case for why levels fall points at CD38. In mouse work, CD38 is the principal NAD+-consuming enzyme whose activity rises with age, driving the age-related fall in tissue NAD+; deleting CD38 protected mice against that decline, preserved SIRT3 activity and improved mitochondrial and metabolic health with age ^[11]. That is a clean mechanism — and it is a mouse mechanism, which is a note about price, not a dismissal.

What the research shows

The human evidence divides sharply into two piles, and telling them apart is the whole point of this page.

The surrogate is bought, cleanly and repeatedly. In healthy overweight adults, nicotinamide riboside at 100, 300 and 1000 mg per day for eight weeks raised whole-blood NAD+ by 22%, 51% and 142% respectively, with no flushing and no significant difference in adverse events from placebo at any dose; NR did not raise LDL cholesterol or disrupt one-carbon metabolism ^[10]. In a multicentre double-blind randomised trial in middle-aged adults, oral NMN at 300 to 900 mg per day for 60 days raised blood NAD+ dose-dependently against placebo at both day 30 and day 60, with 600 mg per day identified as the optimal dose and no safety issues at any dose; the trial also reported improved walking distance and quality-of-life scores ^[7]. Two independent precursors, two controlled designs, one consistent result: the number moves.

One physiological outcome has been bought, in a narrow population. Ten weeks of oral NMN at 250 mg per day improved muscle insulin sensitivity, measured by hyperinsulinaemic-euglycaemic clamp, and remodelled insulin signalling in prediabetic postmenopausal women — with no change in body composition or HbA1c ^[8]. This is a real, mechanistically specific human result. It is also a single study, in one narrowly defined group, on a physiological measure rather than a clinical event.

The destination is not bought. A 2025 narrative review of the human clinical evidence on NAD+ precursor supplementation in ageing concluded that human trials have shown limited efficacy, that age-related NAD+ decline has been consistently observed only in a limited number of human studies, and that data on tissue-specific NAD+ dynamics remain sparse — underscoring the need for more clinical study rather than reliance on rodent extrapolation ^[6].

The most striking mechanism remains preclinical. Using human myocardium and an established murine model of heart failure with preserved ejection fraction, oxidised NAD+ repletion restored activity of the ketogenic enzyme HMGCS2 through deacetylation, increased fatty-acid oxidation and rescued cardiac function — and failed to rescue the model when cardiomyocyte HMGCS2 was knocked down ^[12]. The knockdown control is what makes this good science: it demonstrates the pathway is necessary rather than merely present. It is still a mouse model with human tissue alongside, not a trial.

Reported effects, cautions and safety

The corpus behind this desk carries no structured set of community-reported effects for NAD+, which is itself informative. Where semaglutide and GHK-Cu both have large, patterned bodies of user reporting, the NAD+ conversation is dominated by marketing rather than by users comparing notes. What circulates about intravenous NAD+ clinics is promotional material, and any claim sourced that way is anecdotal, not clinical evidence — with the added problem that anecdote sold by a seller is the cheapest evidence of all.

The documented concerns come from the review and regulatory record instead.

Oral NAD+ itself is poorly absorbed intact, and some researchers argue that plain oral NAD+ capsules are largely ineffective for that reason — a formulation problem, not a toxicity one. Intravenous NAD+ wellness therapy is marketed aggressively on minimal controlled evidence; infused NAD+ clears rapidly from plasma, and infusions can cause chest or abdominal discomfort, flushing and nausea if run too fast. Compounded injectable NAD+ carries contamination risk, and the FDA has issued a Class I recall of a compounded NAD+ injection for elevated bacterial endotoxin — the most serious recall class, and a documented event rather than a hypothetical.

A theoretical concern exists that boosting NAD+ could support the metabolism of an existing cancer, since NAD+ sustains proliferating cells; its role in oncology is dual and context-dependent, and caution is advised in cancer populations. Supplement-grade products vary widely in purity and actual content, and third-party testing is not guaranteed. Finally, much of the strongest anti-ageing data comes from rodents, and the 2025 human review is explicit that rodent extrapolation is not a substitute for clinical study ^[6]. Nothing here is medical advice and no dose is recommended to anyone; the doses named above are the doses used in the cited trials.

What this evidence is worth

NAD+ is the most instructive entry in this hub, because it fails on exactly one of the four levers and that single failure changes everything.

Sample sizes are modest but real. Durations are short but adequate for the endpoint measured. Randomised placebo comparison is present in the key trials ^[7][10]. The lever that gives way is the fourth: what was measured. Blood NAD+ is a surrogate. It is cheap to move, it moves reliably, and moving it is not the same as making anyone healthier — which is the finding the 2025 review reports, in as many words ^[6].

This is the most common way peptide and supplement evidence gets overvalued, and it is worth naming precisely. A marker rising is a genuine result. It licenses the statement this compound does something measurable in humans. It does not license this compound works, because working is a claim about outcomes and no outcome was purchased.

On this desk's arithmetic, NAD+ is therefore neither a bargain nor a bad deal. It is an honest partial purchase, frequently resold as a complete one. The muscle insulin-sensitivity result in prediabetic postmenopausal women ^[8] is the closest thing here to a genuine clinical endpoint, and the disciplined reading is that it is a promising single finding in a narrow population, not a foundation.