GHK-Cu: what the skin studies show, and why injections are a different question

Two controlled human trials have tested GHK-Cu on skin, and they point in opposite directions. A 1994 multicentre trial found a GHK-Cu gel closed diabetic foot ulcers roughly three times faster than standard care; a 13-person trial after laser resurfacing found no objective improvement in redness, wrinkles or skin quality beyond what participants rated themselves. Both were topical. Nobody has published a human trial of injecting it.

What it is

GHK is a three-amino-acid peptide: glycine, histidine, lysine. Loren Pickart first described it in 1973 as a factor in human blood that helped liver cells survive in culture [1]. It binds copper tightly, and the copper-bound form, GHK-Cu, is the version most research uses. Levels in human plasma average about 200 ng/mL at age 20 and around 80 ng/mL by age 60 [2]; that decline is an observation, not proof that topping it up does anything.

In 1988 a French team found GHK-Cu increased collagen production by cultured fibroblasts, starting at extremely low concentrations and peaking around one nanomolar, without changing cell numbers [3]. The same group later injected it into wound chambers implanted under the skin of rats and found more collagen, glycosaminoglycans and protein accumulating, in proportion to the dose [4]. A control tripeptide did nothing. The 1988 paper also pointed out that the GHK sequence sits inside one of the chains of type I collagen, and suggested the peptide might be released when collagen is broken down at a wound, a signal from damaged tissue that repair is needed [3]. That idea, that fragments of the matrix around cells act as messengers, is still the most coherent explanation of why a three-amino-acid peptide would have any effect at all.

From there, the claims widen quickly. A 2015 review argues GHK influences thousands of human genes and describes effects on skin firmness, wrinkles, hair follicles, bone and the gut [5]. That review was written by Pickart and colleagues at Skin Biology, a commercial skin-care company [5], and the gene-expression claims come from cell data. Treat the breadth of claimed effects as a list of hypotheses.

Outside the skin, the animal work is thinner. In mice given a bacterial toxin to cause acute lung injury, GHK-Cu reduced inflammatory signalling and the number of immune cells flooding the lungs [14]. That is an acute mouse model, and nobody has tested it in people with lung disease.

The human evidence

We found two controlled human trials of GHK-Cu itself on PubMed, plus a hair trial of a related combination. All were applied to the skin.

Diabetic foot ulcers (1994)

A multicentre, randomised, evaluator-blinded trial tested a GHK-Cu gel against a placebo gel in people with diabetic neuropathic foot ulcers, alongside debridement, pressure-relieving footwear and standard care [6]. Plantar ulcers treated with the gel closed by a median 98.5% of their area versus 60.8% on vehicle (p less than 0.05), and infections were less common (7% vs 34%). The authors describe closure as three times faster than with standard care and vehicle. The gap was widest in ulcers larger than 100 mm², where median closure was 89.2% on GHK-Cu while the vehicle-treated ulcers actually grew slightly (−10.3%, p less than 0.01). The abstract doesn't give the number of patients, and the authors note the benefit depended on starting treatment straight after the wound was debrided. It is the strongest efficacy signal for GHK-Cu, and it is more than 30 years old and, as far as we can find, unreplicated.

After laser resurfacing (2006)

A US plastic surgeon randomised patients having CO2 laser resurfacing around the mouth to aftercare products with or without GHK-Cu [7]. Thirteen completed the study. Blinded assessors and computer analysis found no difference in how quickly redness faded, and no difference in wrinkles or overall skin quality at 12 weeks. The only positive finding was on a patient questionnaire: people using GHK-Cu rated their skin quality improvement higher (p = 0.04). With 13 people and an unblinded self-rating as the only win, this is a negative trial.

What the reviews found

A 2026 systematic review in Aesthetic Surgery Journal searched three databases up to March 2026 for GHK-Cu used alone in aesthetic medicine [8]. It found 20 studies, of which 18 were preclinical and two were randomised trials. One was the laser study above; the review also reports that GHK-Cu significantly reduced wrinkle volume and depth compared with controls, a result that can't have come from the laser study. Its authors conclude that the data are limited by varied methods and few well-designed trials. A separate 2025 review of GHK as an anti-wrinkle ingredient made the same point more bluntly, noting a "surprising absence of clinical studies" despite its wide use in cosmetic products [9].

You'll often see percentages from facial studies quoted in marketing and in reviews by the peptide's commercial developers, which describe tighter, firmer and less wrinkled skin [5]. We couldn't find those underlying studies as peer-reviewed trials on PubMed, so we haven't relied on their numbers.

Hair

Hair-growth claims lean mostly on cell and animal work. The one human trial we found tested copper-free GHK combined with 5-aminolevulinic acid, not GHK-Cu, in 45 men with pattern hair loss over six months [15]. Hair counts rose by 52.6 and 71.5 in the two active groups versus 9.6 on placebo, though hair length and thickness didn't differ. Because the product was a combination, the trial can't say which ingredient did the work.

Does it get into skin?

This matters for creams. In an experiment on donated human skin in diffusion cells, a 0.68% copper tripeptide solution delivered measurable copper through the full thickness of dermatomed skin over 48 hours, with a further depot left in the tissue: about 136 µg of copper per cm² passed through and 97 µg per cm² stayed in the skin [10]. That was measured as copper, not intact peptide, and in excised skin rather than living people. Reviews note that the form matters: copper binding and fatty-acid tags such as palmitoyl-GHK improve penetration, and microneedles help further [9].

The injection question

Injecting GHK-Cu under the skin, into the face or systemically has become common in some clinics and online communities. We found no published human trial of injected GHK-Cu for any purpose. A 2026 review in Sports Medicine lists it among unapproved peptides sold directly to patients for which rigorous human safety data are scarce [11]. The rat wound-chamber study involved injection [4], but it and the mouse lung work [14] are animal models of injury, not evidence about dose, safety or effect in people. Topical results, even good ones, don't transfer automatically: a peptide sitting in the upper skin is a different exposure from one circulating in the blood, and the copper it carries is a biologically active metal.

Safety and side effects reported

Topically, GHK-Cu has a long track record as a cosmetic ingredient [9], and the trials above didn't report meaningful harms [6,7]. Those trials were far too small to catch uncommon reactions such as contact allergy, so "no harms reported" here means little more than "nothing obvious in small groups". For injection there is no human safety data to report, which is different from saying it's safe. Unregulated injectable products also carry the usual risks of contamination and wrong dosing, independent of the peptide.

Legal and regulatory status

Cosmetics. GHK-Cu is widely sold in cosmetic creams and serums [9]. Cosmetics can't legally claim to treat disease.

United States. It isn't an FDA-approved drug. Nominators had put GHK-Cu forward for pharmacy compounding, and the FDA page listing substances with significant safety risks records that nomination as withdrawn [12]. The FDA's compounding advisory committee is due to consider GHK-Cu at a meeting before the end of February 2027, according to a law-firm summary of the July 2026 meeting [13].

United Kingdom. Not licensed as a medicine; it appears in cosmetic products.

What is still open

• Whether GHK-Cu creams beat a matched cream without the peptide on wrinkles, measured objectively, in a trial of more than a few dozen people.
• Whether the 1994 ulcer result would hold up in a modern trial.
• Anything at all about injected GHK-Cu in humans: blood levels, how long it lasts, safety, or effect.
• Whether the thousands of gene-expression changes seen in cells mean anything for a living person.
• Whether copper-free GHK, palmitoyl-GHK and GHK-Cu behave differently on skin, since products use all three and trials rarely compare them.

What this means

Topically, we think GHK-Cu is a reasonable cosmetic bet. It penetrates: a 0.68% copper tripeptide solution carried measurable copper through the full thickness of dermatomed human skin in 48 hours [10]. And the 1994 ulcer trial showed a healing signal that beat standard care rather than only vehicle, in a population whose wounds close slowly [6]. We wouldn't treat the 2006 laser trial as the counterweight it looks like — thirteen people, and its single positive was a self-rating [7]. Read together, the two trials describe a peptide that probably does something to injured skin and probably does nothing measurable to already-healthy skin in a small sample. For a cream, that is a defensible place to stand.

Our position on injecting it is firmer: don't. This is not reflexive caution about a grey-market peptide. The topical case rests on copper measured in skin and a healing ulcer, and neither transfers to a needle. A peptide held in the upper skin is a different exposure from one circulating in the blood, and GHK-Cu carries copper, a biologically active metal with its own dose-dependent toxicity [10]. There is no human trial of injected GHK-Cu, and a 2026 review lists it among unapproved peptides sold direct to patients whose rigorous human safety data are scarce [11]. If you buy it at all, buy it as a cream, and look for what the label leaves out: an accredited-lab purity certificate, a stated copper content, and — for anything presented as injectable — sterility, endotoxin and heavy-metal testing. A cosmetic-grade serum meets none of those standards, and unregulated injectables carry contamination and dosing risks that have nothing to do with the peptide itself.

On why the definitive trial has never happened, the reason is specific to GHK-Cu and not the usual one. It is sold as a cosmetic, and a cosmetic cannot legally claim to treat disease [9], so no regulation compels the wrinkle trial that would settle the cream question. The commercial interest sits with the peptide's own developers, whose output is reviews built from cell data rather than the trials that would test them [5]. And the one US pathway that could have forced a human dataset — pharmacy compounding — closed when the nomination was withdrawn ahead of the FDA's safety review [12], with the advisory committee due to revisit it by February 2027 [13]. This is not a gap waiting for a motivated funder. It is a gap the incentive structure holds open, and we would rather say so than pretend a trial is imminent.

Educational content only — not medical advice. Many peptides discussed on HPR are not approved for human use. Talk to a qualified clinician before making any decision about your health.