This library covers peptides. NAD+ is not one, and the most useful sentence this entry can offer is that plain statement, because almost everyone who arrives here searching "NAD peptide" believes otherwise.
What it is
NAD+ — nicotinamide adenine dinucleotide — is C₂₁H₂₇N₇O₁₄P₂, molecular weight 663.4, PubChem CID 5892. The name describes the structure exactly: two nucleotides, one carrying the base adenine and one carrying nicotinamide, joined through a bridge of two phosphate groups.
Nucleotides are the class of building block that DNA and RNA are made from. They are not amino acids, and the links between them are not peptide bonds. There is no sequence to write down and no residue count to give.
| A peptide | NAD+ | |
|---|---|---|
| Built from | Amino acids | Nucleotides |
| Joined by | Peptide bonds | A phosphate bridge |
| Has a sequence | Yes | No |
| Example | BPC-157, 15 residues | Two nucleotides, 663.4 Da |
So why is it here? Because of where it is sold rather than what it is. NAD+ is offered by the same clinics that offer injectable peptides, given by the same routes, priced in the same way and described in the same vocabulary of energy, recovery and ageing. In that market "peptide" has drifted from a chemical description into a category of injectable wellness product. This library carries two other entries for the same reason — C-peptide, which is a laboratory measurement nobody takes, and GLP-3, which is a name for something that does not exist in humans.
What it does, in two different ways
NAD+ has two jobs, and confusing them is where most of the reasoning about supplementing it goes wrong.
As an electron carrier it is recycled. The reactions that pull energy out of glucose, fat and amino acids hand their electrons to NAD+, turning it into NADH; the machinery that makes cellular fuel takes those electrons back, turning it into NAD+ again. In this role the molecule is a shuttle. It is not used up, and a cell's supply is not consumed by ordinary metabolism.
As a substrate it is destroyed. Several enzyme families — those involved in DNA repair and in cellular signalling — do not merely borrow NAD+ but break it apart in the course of their work. This role does consume the molecule, and it is why cells must keep manufacturing more.
That second role is the entire basis of commercial interest in raising NAD+ levels. It is a real piece of biochemistry. What does not follow automatically from it is that adding NAD+ from outside reaches the right compartment of the right cells in a useful amount — which is what the research is supposed to establish.
The registry, read rather than counted
On 2026-09-18, an intervention search for nicotinamide adenine dinucleotide on ClinicalTrials.gov matched 264 studies. That number appears in a great deal of marketing.
Reading the first 200 of those records and classifying what each one actually gives a participant:
| What the study administers | Records |
|---|---|
| Nothing NAD-related at all | 141 |
| Nicotinamide riboside (a precursor) | 38 |
| NAD or NADH itself | 9 |
| Niacin or nicotinamide (precursors) | 8 |
| Nicotinamide mononucleotide (a precursor) | 4 |
Seventy per cent of the matches give no NAD agent of any kind. The interventions in that group include letrozole, voxelotor, and olaparib paired with an experimental partner drug — the term is matching somewhere else in the record, frequently because NAD is something being measured as an outcome rather than something being given.
This library has documented the same trap on humanin, where five of seven registry records were observational and both interventional ones administered something else entirely, and on GLP-3, where eleven PubMed hits for an abbreviation resolved to a fish peptide, a worm gene, a proteoglycan and a study-duration label. A search result is a string match. It becomes an evidence count only after someone has read what the records are about.
Of the 200 read, 188 are interventional and 12 observational.
And what is tested is mostly not NAD
Set the 141 aside and look at the 59 that do give something. Nicotinamide riboside accounts for 38 of them — more than four times the number giving NAD itself. Add niacin, nicotinamide and nicotinamide mononucleotide and the precursors outnumber NAD by roughly five to one.
The distinction is not pedantic. Precursors are compounds the body converts into NAD, and in this research they are almost always taken by mouth. What is sold in clinics is NAD+ injected or dripped. Evidence that an oral precursor moves a laboratory measurement is not evidence about an intravenous infusion of a different molecule, and the two are routinely quoted as though they were one body of work.
The conditions studied spread widely and thinly: ageing (8 records), healthy volunteers (6), multiple myeloma (5), Parkinson's disease (5), COVID-19 (5), then acute kidney injury, heart failure, alcohol use disorder and mitochondrial myopathies at three or four each. No indication dominates — the shape of a field interested in a mechanism rather than in a disease.
Where it stands
NAD+ is not an approved treatment for any indication in the United States. It is supplied through compounding pharmacies and administered in clinics, which is a different regulatory position from approval: a compounded preparation can be dispensed without the controlled evidence of benefit that an approval requires. The absence of approval is not evidence that it does nothing; it means the question has not been settled in the way approval settles it.
Questions about whether to have any of it belong with a clinician. This entry exists to establish what the molecule is.
Related entries
- The peptide bond — the link NAD+ does not contain
- What peptides are — the definition this molecule falls outside
- C-peptide — a second thing in this library that nobody takes
- GLP-3 — a third disambiguation, and the strangest
- Humanin — where the registry-counting trap was first measured here