GHK-Cu before and after: where the numbers come from and what you can measure on your own face

Almost nobody quoting the 55.8% wrinkle figure has read the paper it sits in, and the people who have would quote it differently: the same trial reports 31.6% against a real competitor cream and 55.8% only against a plain serum, and the journal that ran it is not indexed in PubMed. Every other confident GHK-Cu before-and-after figure on the internet, including the ones this page carried until today, traces to a 2002 conference poster that was never published or a 1998 pilot study on thigh skin. The one randomised controlled trial of GHK-Cu on a human face, meanwhile, measured no change at all while the people using it rated their own skin significantly better. In other words, the machines saw nothing and the people saw something, in the same trial, under blinding.

The one randomised trial on facial skin that can be read

Miller TR, Wagner JD, Baack BR and Eisbach KJ randomised patients having carbon-dioxide laser resurfacing of the skin around the mouth to a post-treatment skin-care regimen either with or without a topical copper tripeptide complex, and assessed them at 12 weeks. Thirteen patients completed the study. Computer analysis of standardised photographs and blinded clinical evaluators both found no statistically significant difference between the two groups for redness, for wrinkles, or for overall skin quality ¹. It is the only randomised, controlled, blinded facial trial we could identify. Two German pilot studies on aged skin from 2003 (Kruger et al., references 18 and 19 of the Pickart and Margolina review) sit near it in the literature, and neither was randomised or controlled.

Objective evaluation found no significant improvement in wrinkles or overall skin quality. However, patient satisfaction was significantly higher for those who used GHK-Cu skin care products after CO2 laser skin resurfacing.Miller TR, Wagner JD, Baack BR, Eisbach KJ. Archives of Facial Plastic Surgery 2006;8(4):252-259.

Two limits belong beside that result, and reporting the negative finding without them would be its own kind of dishonesty. The skin studied had been laser-resurfaced, so the trial tests GHK-Cu as an addition to healing after a procedure rather than as a serum on untreated ageing skin. And every patient improved significantly in wrinkles and skin quality regardless of which group they were in, because of the laser ¹. That within-group improvement is why the P = .04 above is described as the only significant difference between the groups. This is the closest thing the literature has to a blinded test of copper peptides on a human face, and it is still not a clean test of a cosmetic serum.

That gap is why a personal before-and-after photograph is such weak evidence about yourself. In the one place where instrument and questionnaire were both recorded under blinding, they disagreed. Any photograph you take of your own face is the questionnaire half of that experiment, running without the other half.

Where the confident percentages actually come from

Follow each widely quoted GHK-Cu figure back to its primary source and it lands somewhere other than where the quoting article says it does. We went to read each one on 15 September 2026, through Crossref, through PubMed, and by resolving the DOIs. The table records what came back.

There is no Dermatologic Surgery paper by Leyden on copper peptides, in 2005 or in any other year. The facial-cream result and the 70/50/40 comparison are two different pieces of work, from two different years, on two different parts of the body, and neither is the paper they are usually credited to. The head-to-head against retinol and vitamin C, done on the trial record that does exist, is not close. Our comparison piece puts it plainly: "Retinol wins on RCT volume: 8 pooled trials, 1,361 patients, confirmed wrinkle reduction per a 2025 meta-analysis in Dermatology Practical and Conceptual." Against that, GHK-Cu has one blinded facial trial that found nothing. GHK-Cu vs retinol and vitamin C works through the rest.

All three of those sources reach the wider internet by one route. They are references 14, 16 and 20 of a 2018 review by Loren Pickart and Anna Margolina in the International Journal of Molecular Sciences ³, and a second 2002 Leyden poster on eye creams sits at reference 15, so the 2002 material is two unpublished posters and not one. Both authors give R&D Skin Biology in Bellevue, Washington as their affiliation, and Pickart founded companies selling GHK-Cu products ³. Nearly every confident copper-peptide claim online arrives through that single review. Two claims that both come from it are not two independent sources.

The plasma figure, roughly 200 nanograms per millilitre of GHK in blood at age 20 falling to roughly 80 at age 60, appears in that same review without a reference on the sentence ³. It can be reported with that provenance attached. It cannot carry the argument it is usually made to carry, that older skin responds faster because the deficit is larger. Nobody has measured GHK-Cu response by age under blinding, so that is a guess about who responds, dressed up as a fact about age. For what does have evidence behind it in older skin, peptides for anti-ageing covers the condition and peptides after 50 covers the practical side.

What is known about the compound reaching skin

The best measurement of GHK-Cu getting into skin at all was done on human skin outside the body. Hostynek, Dreher and Maibach mounted three preparations in flow-through diffusion cells: isolated stratum corneum (the dead outer sheet of skin that keeps water in and most molecules out), heat-separated epidermis, and dermatomed skin, which is a slice thick enough to include the living layers underneath. They applied 0.68% aqueous copper as the tripeptide over a 1 square centimetre area at infinite dose, meaning enough was applied that the supply never ran down, and collected the fluid underneath for 48 hours. That is the whole of the penetration evidence: 48 hours, a slice of skin with nobody attached to it, and an instrument counting metal. Over those 48 hours, 136.2 plus or minus 17.5 micrograms per square centimetre of copper crossed the tissue, and 82 plus or minus 8.1 stayed behind in it as a reservoir ⁵.

Liposomal and nano-carrier serums are sold on the claim that their carrier gets more of the peptide in, usually with a specific multiple attached. A 2025 review in Molecules asked directly whether that can be measured yet, and answered its own question ⁸.

An analysis of the literature data reveals that the transport of liposomes containing GHK-Cu received little attention.Ogorek K, Nowak K, Wadych E, Ruzik L, Timerbaev AR, Matczuk M. Molecules 2025;30(1):136.

So a product claiming a specific penetration multiple for its carrier is quoting a number the field has no accepted method to produce. That applies to the 2016 figures in the provenance table as well, since the delivery system those figures are attributed to is the same kind of carrier the 2025 review says has not been measured.

The strongest positive human result for GHK-Cu is on a different problem from cosmetic ageing. A multicentre, randomised, evaluator-blinded, placebo-controlled trial of a GHK-Cu gel on plantar diabetic neuropathic foot ulcers, run under a standardised wound-care protocol, reported 98.5% median area percentage closure against 60.8% for the same vehicle gel without the peptide (p<0.05), closure roughly three times faster, and ulcer infection in 7% of treated patients against 34% on the plain gel (p<0.05) ⁶. An open ulcer is skin with its barrier already gone. That is close to the opposite of an intact face. That result was also produced inside a supervised wound-care protocol, and no GHK-Cu product is approved to treat a wound of any kind. A diabetic foot ulcer needs a clinician, urgently and in person, and it is not something to manage from a page like this one.

What you can hold still when you photograph your own face

A photograph of your own face cannot separate new collagen from better lighting, a different time of day, a more hydrated face and a camera held slightly closer. Several of those you can fix, and fixing them is what turns a personal before-and-after from a mood into a comparison. This is the part of the subject nobody sells, because it makes the pictures less impressive. None of what follows is a reason to start using anything. It is a method for making a photograph mean something, not a dose, a protocol or an endorsement, and it assumes a decision you have already taken with a clinician who knows your history.

1. 1One light, one place, one time of dayPick a single spot with a light source that does not change, such as a windowless bathroom with the same fixture, and shoot at the same hour every time. Side lighting exaggerates texture and flat frontal lighting hides it, so whichever you pick, never change it. A light change alone can produce what looks like a result.
2. 2Fixed camera distance and lensMark where you stand and where the phone sits. A face photographed from closer looks different in shape, and the front camera on a phone at arm's length distorts differently from a rear camera on a stand. Use the same camera, the same lens and the same distance.
3. 3Same hydration, same time since washingDehydrated skin reads as more lined. Shoot at a fixed interval after cleansing, before applying anything, and note what you drank that morning. Hydration moves fine lines more visibly over an afternoon than a cosmetic serum measurably moved them over twelve weeks in the one blinded trial [[1]].
4. 4No makeup in either frame, neutral expression in bothMakeup removed to the same degree, face relaxed in both photographs, eyes open the same amount. Expression lines are the first thing people unconsciously soften in an after shot.
5. 5A fixed cadence you set in advanceEvery four weeks, on dates you write down before you start, rather than on the days you happen to think you look good. Choosing when to take the after photograph is the single easiest way to manufacture a result.
6. 6A written log of everything elseSun exposure, sleep, alcohol, weight, any other active you started, any illness. When something changes in the photographs, this log is the only thing that lets you say what else was running at the same time.

How long before GHK-Cu works on hair?

The one controlled hair trial we could identify involving GHK tested ALAVAX, a complex of 5-aminolevulinic acid and GHK peptide, rather than GHK-Cu on its own. Forty-five men with male pattern hair loss applied it once daily for six months at one of two concentrations against placebo, with evaluations at months 1, 3 and 6. Total hair count rose by 52.6 in the 100 mg/mL group and by 71.5 in the 50 mg/mL group, against 9.6 on placebo, and each of those increases was reported at p<0.05 within its own group ⁷. The only between-group comparison the paper reports as significant is the 50 mg/mL group against placebo. The 100 mg/mL group was never shown to beat placebo, and the lower concentration outperformed the higher one, which the authors did not explain. Hair length and hair thickness showed no significant difference between the three groups at six months ⁷.

The trial counted more hairs and measured no change in how long or how thick they were, using a complex that is not the product being sold. Beyond that, there is no published GHK-Cu hair timeline at all. The widely repeated figure of 8 to 10 weeks for shedding and density traces to no trial, the earliest measurement point in that trial was month 1, and shedding rate was never an outcome measure in it. For the condition itself and what has been studied against it, peptides for hair loss is the page that owns it.

Cream, needling, injection or capsule

The route changes which evidence applies to you, more than any brand choice does. How the peptide works once it is inside, and the discovery story behind it, belong to how GHK-Cu works. What follows is only about what the human record contains for each way of getting it in.

The blends most people are photographing

GHK-Cu is rarely bought on its own. It anchors two pre-mixed blends that put three or four compounds into the same vial. A photograph taken after twelve weeks of either one is a photograph of the combination.

Skin and appearance

GLOW

70 mg vial

The skin blend: one collagen peptide and two repair peptides

50 mgGHK-Cu

10 mgBPC-157

10 mgTB-500

Skin and appearance

KLOW

80 mg vial

GLOW with an anti-inflammatory peptide added

50 mgGHK-Cu

10 mgKPV

10 mgBPC-157

10 mgTB-500

Compare every stack →

Neither blend has been tested as a combination in people, so a result from one cannot be attributed to any single part of it, and a before-and-after of a blend credited to GHK-Cu is crediting a quarter of what was applied. For how the individual compounds studied for skin compare, the best peptides for skin covers the roster, and skin ageing covers the condition.

Why did GHK-Cu do nothing for me?

Four ordinary explanations come before any genetic one, and between them they account for far more cases.

1. What the product contained. Cosmetic serums are not required to state how much copper tripeptide is in them, and many do not. A product with an unstated amount of active cannot be compared against a study that measured one, and it cannot be compared against another serum either.
2. Whether it reached the living layers of skin at all. The only quantified permeation work measured copper crossing human skin outside the body, and the field has no accepted method for measuring what a liposomal carrier adds on top of that ⁵⁸.
3. What else changed in the same twelve weeks. Sun exposure, sleep, weight, hydration, a retinoid started at the same time, a new cleanser, a change of season. Any of these moves a face more visibly than a cosmetic copper peptide has been shown to move one under blinding.
4. How the photographs were taken. Lighting, camera distance, time of day, hydration and expression each move a face, and the six steps above are the only way to take them out of the comparison.

Genetics comes after those four. No trial has ever sorted GHK-Cu responders from non-responders by genotype. What genetics can tell you here is about your starting material, not your response. COL1A1 is the gene for the main collagen your skin is built from, and a common variant in its switch region is associated with how much of it cells make at baseline. TGFB1 is the gene for a growth signal that tells repair cells to lay down new matrix after damage. Knowing where you sit on either says something about the tissue a compound would be acting on.

It does not produce a week number, and anyone offering you a genotype-specific timeline for copper peptides is describing something nobody has measured. Our which peptides your DNA points to piece sets out what that kind of reading can and cannot do.

Is GHK-Cu FDA approved?

No. It is not legally compoundable either, and that second half is what people get wrong when they hear it came off the FDA's do-not-compound list in April 2026. The distinction that matters here is between the two products GHK-Cu is sold as. Topical GHK-Cu is lawfully marketed as a cosmetic ingredient, a regulatory category that involves no FDA approval of anything, so a serum sold lawfully is not evidence that it was reviewed, and the absence of approval is not a finding against it either. Injectable GHK-Cu is an unapproved drug, and it is the injectable form that the compounding decisions below are about.

1. 22 April 2026: injectable GHK-Cu was removed from the FDA's Category 2 list of substances that should not be compounded. The reason given was that the people who had nominated it withdrew their nominations, not any finding about safety or effectiveness.
2. That removal did not make it eligible to compound, because GHK-Cu is not on the 503A positive bulks list, the list of bulk substances a compounding pharmacy is permitted to use. Injectable GHK-Cu therefore has no lawful compounding pathway today, and a provider presenting it as a settled or approved option is describing something the record does not support.
3. 23 and 24 July 2026: the Pharmacy Compounding Advisory Committee met to evaluate seven of the twelve substances removed from Category 2, among them BPC-157, TB-500, KPV, MOTS-c, DSIP (also called emideltide), Semax and Epitalon. GHK-Cu was not among them. We are not reporting the individual vote outcomes here, because we could not source them to a published meeting record, and this page spends its whole length holding other people to that standard.
4. FDA has said it intends to consult the committee on GHK-Cu, alongside Melanotan II, LL-37, dihexa acetate and PEG-MGF, before the end of February 2027. Committee votes are recommendations, not binding decisions, and the rulemaking that would follow a favourable one takes far longer than the meeting does.

Anti-doping is a separate question from the FDA one, and this page is not going to settle it. We could not open WADA's 2026 Prohibited List at source to check how GHK-Cu is treated, and a status we have not verified is exactly the kind of claim the rest of this page exists to object to, so we are not stating one in either direction. If you are subject to testing, get that answer from your own anti-doping authority before you use anything. Are peptides legal in the US in 2026 tracks the current position and carries the sourcing for the dates above.

Practical things to settle before the twelve weeks start

• Price tells you very little here. A cosmetic serum and a research vial are sold on completely different bases, and because most serums do not state how much copper tripeptide they contain, price per unit of active cannot be compared between them at all. A more expensive serum is not a stronger one, and there is no way to check that from the outside.
• A correctly complexed preparation is blue. The colour is the copper being held by the peptide, so a preparation that has lost its colour has changed chemically. How to reconstitute peptides covers general handling and storage for research material. It is not instructions for preparing GHK-Cu for human use, for which there is no published human trial and no approved route.
• Side effects reported on skin are local: redness, stinging or itching where applied, dryness or flaking early on, and a faint blue-green tint that washes off. Some people report skin looking briefly worse in the first weeks, and if it keeps getting worse rather than settling, that is a reason to stop. Irritation can stack if it is layered with retinoids or acids. Injection-site reactions and a metallic taste are reported by people injecting it. That is the route with no published human trial behind it and no sterility oversight. None of this comes from safety surveillance: GHK-Cu has never been through a safety trial for cosmetic or injectable use, so a short list of mild reports reflects absence of study as much as absence of harm. GHK-Cu benefits covers what it is used for, and the GHK-Cu monograph carries the full side-effect list and its sourcing.
• The honest unknowns: how much intact peptide reaches the cells that would read it, what repeated long-term use does in people, and whether any of the widely quoted cosmetic percentages would survive a blinded replication. The one time somebody ran that replication on a face, the objective measures came back flat ¹.

One person filming his own eight weeks is not evidence, and it is still more informative than a cropped still on a vendor's page, because you can see who he is, what he used and which parts of his face did not change.

8 Week GHK-Cu Results: Here's What Happened To MeWilliam Day · plays on YouTube