Peptide LexiconAll compounds

GHK-Cu: A Copper-Carrying Tripeptide With One Randomized Trial, and It Was Negative

GHK-Cu is a three-amino-acid fragment of human collagen that carries a copper ion. Exactly one randomized controlled trial is indexed for it, in 13 patients, and it found no objective benefit. What the compound is, and what is and is not established.

Never approved as a medicineCosmetic ingredientNaturally occurring in humans

Caroline S · Published 2026-09-07 · Updated 2026-09-08

Length

3 amino acids

Sequence

Glycyl-L-histidyl-L-lysine, complexed with copper(II)

Origin

Occurs in human plasma; released from collagen on tissue injury

Last reviewed

2026-09-07

What is it good for?

Its claims are about skin and connective tissue — wrinkles, firmness, wound healing, hair. The laboratory and animal case is substantial. The human case is one randomized trial in 13 patients, which found no objective benefit, and one registration that does not survive checking (see below).

Illustration: A deep green glass vial with a blue liquid and a copper cap on a white desk.
Illustration

What it is

GHK-Cu is one of the smallest compounds in this library: a peptide of three amino acids, carrying a copper ion.

The peptide is GHK — glycine, histidine, lysine, in that order. It is not synthetic in origin. It occurs naturally in human blood plasma, and it is released from collagen when tissue is damaged. That observation is what started the interest: a fragment that appears exactly when the body is repairing itself looks, on the face of it, like a repair signal.

The Cu is a copper(II) ion held by the peptide. This part is not decorative. Copper is genuinely required by several enzymes that build and remodel connective tissue, and a small molecule that can carry copper into skin is doing something chemically real.

So the starting premise is respectable, which is worth saying plainly before the evidence section, because the evidence section is where this entry gets uncomfortable.

What it is said to be good for

Skin, mostly: wrinkles, firmness, elasticity, wound healing and hair. It appears in a great many cosmetic products and is sold separately as a research chemical.

What the research shows

There is a lot written about GHK-Cu. PubMed indexed 130 records on 2026-09-07 for the compound and its synonyms, 47 of them tagged as human research. Much of the most-cited material consists of reviews and laboratory studies of gene expression, collagen and tissue remodelling — real work, done carefully, in cells and animals.

There is one randomized controlled trial. Exactly one, out of those 130 records, is indexed as a randomized controlled trial: Miller and colleagues, Archives of Facial Plastic Surgery, July–August 2006. Thirteen patients completed it. They had circumoral skin resurfacing with a carbon dioxide laser and were randomised to aftercare with or without GHK-Cu products.

It was negative on every objective measure. In the authors' own words, the copper tripeptide products offered no significant reduction or resolution of post-treatment redness, and objective evaluation found no significant improvement in wrinkles or overall skin quality. Computer analysis and blinded evaluators both found no difference between the groups.

One result was positive, and it is worth reporting exactly rather than dismissing. On the patients' own validated questionnaire, satisfaction with overall skin quality was significantly higher in the GHK-Cu group (P = 0.04). Every patient in the study improved — laser resurfacing does that — but only the people using GHK-Cu rated the improvement higher.

A difference that a user reports and that a blinded evaluator and a computer cannot see is a real finding about perception. It is not a finding about skin. Both halves of that sentence belong in any honest account of this compound, and most accounts print only one of them.

What is being tested now. ClinicalTrials.gov returns three interventional studies for GHK-Cu, and two of them are not what they look like: one is a study of a facial treatment device, and one measures circulating GHK during use of an adhesive patch. One is filed as administering the compound: NCT07437586, a Phase 2 randomized, double-blind, vehicle-controlled split-wound study of a topical GHK-Cu gel, 60 planned participants.

Corrected 2026-09-08. An earlier version of this entry counted that registration as a trial in progress. Its sponsor, Hudson Biotech, has filed eight ClinicalTrials.gov registrations in total (registry API, 2026-09-08): all list one facility and start dates inside a two-week window in February 2026, and together they cover MOTS-c, BPC-157, TB-500, GHK-Cu, Melanotan II, tesamorelin, tirzepatide and retatrutide. Two of them carry Eli Lilly's own protocol identifiers — a registry search for I8F-MC-GPHK returns Lilly's completed tirzepatide study NCT04184622 alongside Hudson Biotech's filing, and J1I-MC-GZBF returns Lilly's completed retatrutide study NCT04881760 alongside Hudson Biotech's. None of the eight has posted results. This entry therefore does not treat that registration as evidence that a trial programme is under way. Peptifact's MOTS-c dosage reference sets out the full table of eight and how to reproduce the two identifier searches. So the honest position is that nothing is being tested that we can verify.

A note this library applies to every registry entry, and which matters more here than usual: a ClinicalTrials.gov record is a description written by the study's own sponsor. Nobody verifies it before it appears. It becomes evidence when results are published, and not before.

Why the evidence is thin when the ingredient is everywhere

The obvious question, once the trial count is on the table, is how a compound sold this widely for this long can have one small randomized trial behind it.

The answer is regulatory, and it is worth understanding because it applies to a great many things in this library. GHK-Cu is used as a cosmetic ingredient, and a cosmetic is held to a standard of safety and honest labelling. A drug is held to a standard of demonstrated effect, which means trials, endpoints and a regulator reading the results.

Nobody has had to run the trials, because nothing GHK-Cu is sold as has required them. Decades of use is evidence about safety and about commercial appeal. It is not evidence of effect, and it never accumulates into evidence of effect no matter how long it goes on.

The same regulatory route explains the library's other widely-sold tripeptide. All 57 US drug-database records for glutathione turn out to be homeopathic remedies and skin-whitening cosmetics — categories that reach a shelf without ever demonstrating an effect.

Where it stands

Approved as a medicine: nowhere.

Regulated as: a cosmetic ingredient, in the products most people encounter it in.

Randomized human evidence: one indexed trial, 13 completers, negative on every objective endpoint.

Available as: cosmetic products, and separately as a research chemical.

Related entries in this library: this entry answers what GHK-Cu is and what is known about it. Questions about which products contain it, or which are worth buying, are not questions this library answers.

Related entries in this library: argireline is the other widely sold cosmetic peptide, and its only rigorous trials were run by the NIH for an eye-muscle disorder rather than for wrinkles. The octapeptide marketed as a needle-free alternative has a smaller file still — two papers, neither of which isolates it from the other ingredients it was tested with.

Last checked

2026-09-07. The PubMed counts and the ClinicalTrials.gov records were retrieved on that date through the NCBI E-utilities and the registry API; the 2006 trial's methods and results are quoted from its published abstract.

What the research shows

Effects on collagen and tissue remodelling in the laboratory

Cell and animal work; a large literature

Visible improvement in wrinkles or redness after laser resurfacing

Human; the one randomized trial found none objectively

Wound healing when applied topically

One registered trial that does not withstand checking; nothing reported

Bars show how much of the evidence is in humans, not how well anything works.

Where it stands

Approved as a medicine

Nowhere.

Regulated as

A cosmetic ingredient in the products most people meet it in — a safety-and-labelling standard, not an efficacy standard.

Randomized human evidence

One indexed trial (2006, 13 completers), negative on every objective endpoint.

In progress

NCT07437586, Phase 2 topical gel vs vehicle in wounds, 60 planned — filed by a sponsor whose eight registrations do not withstand checking (see below). No results.

Frequently asked questions

What is GHK-Cu?

A very small peptide — three amino acids, glycine, histidine and lysine — bound to a copper ion. The peptide part, GHK, occurs naturally in human blood plasma and is released from collagen when tissue is injured, which is the observation that started interest in it. GHK-Cu is the copper-carrying form, and copper is genuinely required by several enzymes involved in building and remodelling connective tissue.

What is GHK-Cu said to be good for?

Skin: wrinkles, firmness, wound healing and hair. Those claims come from a large body of laboratory and animal work on collagen, gene expression and tissue remodelling. Whether they translate into a visible effect on a human face is a separate question, and it is the question with the least evidence behind it.

How many randomized trials of GHK-Cu are there?

One indexed on PubMed, as of 2026-09-07 — a 2006 study in Archives of Facial Plastic Surgery, in 13 patients who completed it, testing GHK-Cu skin products after carbon dioxide laser resurfacing. For a compound sold as widely as this one, a single small randomized trial across twenty years is the central fact about its evidence.

What did that trial find?

It was negative on everything it measured objectively. Computer analysis and blinded evaluators found no statistically significant difference in how quickly redness resolved, and no significant improvement in wrinkles or overall skin quality. Every patient improved — laser resurfacing works — but the groups did not differ. The one significant result came from the patients' own questionnaire, where satisfaction with overall skin quality was higher in the GHK-Cu group (P = 0.04). A difference visible to the user but not to a blinded evaluator or a computer is a real result, and it is a result about perception.

Is GHK-Cu approved as a medicine?

No, nowhere. It is widely used as a cosmetic ingredient. That distinction decides what a product has to prove: a cosmetic must be safe and honestly labelled, while a drug must show it does what it claims in trials. GHK-Cu has been sold for decades under the first standard, which is why so much is written about it and so little of that writing is trial evidence.

Is copper tripeptide-1 the same thing as GHK-Cu?

Yes. Copper tripeptide-1 is the INCI name — the standardised term a cosmetic label must use — for the same molecule this entry describes. A product listing copper tripeptide-1 in its ingredients and one advertising GHK-Cu on the front of the box are naming one compound twice. There is no separate substance and no separate evidence base: everything on this page applies to both names.

Is anyone testing GHK-Cu properly now?

One registered trial is filed as administering it: NCT07437586, a Phase 2 randomized, double-blind, vehicle-controlled split-wound study of a topical GHK-Cu gel, 60 planned participants. It has not reported, and it does not survive checking. Its sponsor, Hudson Biotech, has filed eight ClinicalTrials.gov registrations in total (registry API, 2026-09-08): all list one facility and start dates inside a two-week window in February 2026, and together they cover MOTS-c, BPC-157, TB-500, GHK-Cu, Melanotan II, tesamorelin, tirzepatide and retatrutide. Two of them carry Eli Lilly's own protocol identifiers — a registry search for I8F-MC-GPHK returns Lilly's completed tirzepatide study NCT04184622 alongside Hudson Biotech's filing, and J1I-MC-GZBF returns Lilly's completed retatrutide study NCT04881760 alongside Hudson Biotech's. None of the eight has posted results. This entry therefore does not treat that registration as evidence that a trial programme is under way. A registry entry is a sponsor's description of a study it intends to run; nobody verifies it before it appears, and it becomes evidence only when results are published.