Every second product page aimed at thinning male hair says the same thing in the same order: DHT is the problem, this peptide addresses DHT. We went looking for the study that measures it. It does not appear to exist.
What exists instead is a stack of laboratory work, one trial of a combination product, and a mechanistic paragraph that has been read as a clinical result for over a decade. This page traces the claim rather than the ingredients — for the per-peptide evidence tiers, see peptides for hair growth, and for who was enrolled in these studies in the first place, peptides for hair growth in women.
What the database holds
Searched on 2026-09-10, peptide[tiab] AND (5-alpha-reductase[tiab] OR dihydrotestosterone[tiab] OR DHT[tiab]) AND (hair[tiab] OR follicle[tiab] OR alopecia[tiab]) returns 17 records. Read one by one, most are not about hair products at all. The set includes nesfatin-1 in lizard testis, insulin-like 3 in rat ovarian follicles, endothelin-1 in granulosa cells, steroid concentrations in seminal plasma, and receptors in uterine tissue. In those papers "follicle" means an ovarian follicle and "peptide" means a hormone, and their presence in the count is an artefact of two common words sharing a search.
Strip those out and the hair-relevant androgen literature for peptides is roughly five papers:
- Peptide AC 2 (Sci Rep 2019), isolated from a fermented water chestnut extract, rescuing DHT-treated human dermal papilla cells — cells in a dish.
- CXXC5 and prostaglandin D2 (Cells 2023), a mechanism of DHT-induced alopecia and a target for a peptide that interferes with it — cells and mice.
- DHT inhibiting hair growth in mice via IGF-I (Growth Horm IGF Res 2011) — animal.
- Sphinganine (Clin Cosmet Investig Dermatol 2016), a potent inhibitor of 5-alpha-reductase type I in an in vitro assay. Sphinganine is a sphingolipid, not a peptide; the peptide in that paper is HBD2, which the scalp makes itself.
- Capixyl's trial paper (J Cosmet Sci 2013), discussed below.
Not one of them measures an androgen in a person using a hair product.
The sentence the category rests on
The Capixyl paper is the origin of most anti-DHT peptide marketing, so it is worth quoting its structure precisely.
Thirty volunteers with recessing hair were randomised against placebo and treated daily for four months, with TrichoScan measurements: anagen hair up 13%, telogen density down 29%, anagen/telogen ratio up 46% over baseline, against −2%, +23% and −33% in the placebo group. Then, separately, the authors write that investigation of the potential mechanisms "pointed at" inhibition of 5-alpha-reductase activity, reduction of inflammatory reactions, and stimulation of extracellular matrix protein synthesis near the follicle.
Two things follow from reading it that way. The androgen claim is a mechanistic inference reported next to the trial, not an endpoint measured in the volunteers — no testosterone or DHT was reported in those 30 people. And the product tested was, in the paper's own description, an active made by combining a Trifolium pratense flower extract and a biomimetic peptide. Red clover is an isoflavone source, and isoflavones are where an anti-androgen argument would ordinarily start. The trial cannot separate the extract from the peptide, so the strongest honest statement is that a red-clover-plus-peptide combination beat placebo on a follicle-cycle ratio in 30 people, in a study written by the R&D department of the company selling it.
That distinction survives into the shelf. A product marketed to men as a DHT-blocking peptide is usually carrying a botanical alongside the peptide, and the botanical is doing the androgen work in the story.
The scale problem, in one line
| Search (PubMed, 2026-09-10) | Records |
|---|---|
minoxidil[tiab] AND androgenetic alopecia[tiab] |
765 |
finasteride[tiab] AND androgenetic alopecia[tiab] |
560 |
peptide[tiab] AND (5-alpha-reductase OR dihydrotestosterone OR DHT)[tiab] AND (hair OR follicle OR alopecia)[tiab] |
17, most of them not about hair |
The finasteride label trials enrolled 1,553 men. The 5% minoxidil randomised trial enrolled 393. The placebo-controlled peptide trials enrolled 22, 30, 35 and 45. This is not a close comparison being decided on nuance; it is two literatures that differ by two orders of magnitude in size, and a marketing convention that presents them as competing options.
What the best peptide study in men actually showed
The strongest design in the category is also the smallest: 22 Japanese men with androgenetic alopecia, four months, double-blind and randomised against the same lotion without the peptide, published in J Cosmet Dermatol in 2016 by authors at the Shiseido Global Innovation Center with a hospital co-author.
Its result is unusual and easy to misreport. The endpoint was not how many hairs there were. It was the proportion of hairs at or above 60 μm — the paper's threshold for a thick hair — which rose significantly more in the peptide group than in placebo (P = 0.006), while the proportion of vellus hairs under 40 μm fell (P = 0.029). Blinded reviewers grading photographs on a standard baldness scale also scored the peptide group better (P = 0.020). No treatment-related adverse dermatological effects were noted.
That is a real, blinded, placebo-controlled result — in 22 men, from the company that developed the ingredient, never independently replicated, and measured on calibre rather than count. What "proportion of thick hair" does and does not mean is worth understanding before comparing it with anything else, which is why we took the endpoint apart on peptides for hair thickness.
Reading a men's hair product honestly
Three questions separate a defensible claim from a decorative one, and all three can be answered from the label and the cited study.
Was the peptide tested, or the formula? If the cited trial tested a combination — a peptide with a botanical, or a spray with three actives — nothing in the result belongs to the peptide by itself.
Was the endpoint hair, or a marker? Enzyme inhibition in a dish, receptor expression on a plucked hair, or gene expression in cultured cells are all upstream of the thing being sold. The GPIGS trial measured hairs on heads. Most of the category does not.
Was it a leave-on product? Every trial above used a lotion applied daily for four to six months. Rinse-off formats face a contact-time problem that no amount of active concentration solves, which is the subject of our peptide shampoo page.
Who wrote it, and did anyone else repeat it? Manufacturer authorship is not disqualifying — it is how most cosmetic ingredient science gets funded — but a manufacturer-authored study that has never been replicated is a weaker foundation than the same study replicated by an independent group. In this category, none of the trials we opened has been independently replicated.
Applied to the shelf, those questions leave a narrow set of statements standing: one blinded trial in 22 men on hair calibre, one placebo-controlled trial of a peptide-plus-extract combination in 30 volunteers on follicle-cycle ratios, an unpublished 35-man dossier, and a large amount of laboratory work about an androgen pathway that no human study of these products has measured.
Sources
- Loing E et al. A new strategy to modulate alopecia using a combination of two specific and unique ingredients. J Cosmet Sci 2013;64(1):45-58.
- Iwabuchi T et al. The topical penta-peptide Gly-Pro-Ile-Gly-Ser increases the proportion of thick hair in Japanese men with androgenetic alopecia. J Cosmet Dermatol 2016;15(2):176-84.
- The peptide AC 2 isolated from Bacillus-treated Trapa japonica fruit extract rescues DHT-treated human dermal papilla cells. Sci Rep 2019.
- CXXC5 Mediates DHT-Induced Androgenetic Alopecia via PGD2. Cells 2023.
- Gerlach N et al. Effect of the multifunctional cosmetic ingredient sphinganine on hair loss in males and females with diffuse hair reduction. Clin Cosmet Investig Dermatol 2016;9:191-203.
- PubMed counts run 2026-09-10 with the queries printed above.