Every human study is a cream. It is sold in a vial.
What it actually is
GHK is three of the building blocks the body uses to make proteins, joined in a row: glycine, histidine and lysine. A short chain like that is called a peptide, and this one is already present in your own body. The Cu on the end is a single copper atom that the chain grabs and carries about, rather than a mineral stirred into the mix, and copper makes up 15.8 percent of the finished thing by weight, a figure worked out from the molecule's weight of 401.91 rather than measured in any paper. It has a real human research record, which is rare for anything in this reference. All of that record was rubbed on skin, while what is sold now is a powder in a small glass bottle that people mix with liquid and inject.
What it is supposed to do
The story that sells it is a decline with age. Blood levels are said to run about 200 nanograms in a millilitre of blood at age 20 and about 80 by age 60, so putting it back is supposed to restore what time took away. That sentence sits in a reviewed article with no reference attached to it, and no original source for it could be found. In animals the effects are real but they do not all point one way: rats grew more connective tissue on wounds made for the experiment, and in repaired knee ligaments the benefit was one the authors themselves called transient. Two reviews of injectable peptides published in 2026, one in sports medicine and one in bone and joint surgery, landed in the same place, that there is no clinical data and that the condition to use it for, the amount, how often and for how long are all unknown.
What people take it for
The published human work is all skin and wounds: ulcers on the feet of people with diabetes, ulcers on the legs, and the skin around the mouth after laser resurfacing. It is also quoted constantly for hair, and the study behind that claim is about a different molecule. The vial people inject is bought on the age decline story rather than on any of the trials. That is what it is sold and talked about for, not what it has been shown to do, since I could not find a published study putting it into a person by injection.
The dose question
What circulates
1 to 2 mg a day under the skin, some say 3
What was studied
No injection into a person that I could find
Every published human study applied it to skin, and the 1994 gel's strength is not stated
There is nothing to compare against, because I could not find a published study injecting it into a person by any route. The human evidence that does exist is a cream on a wound, and even the 2 percent strength quoted for it on vendor pages is not in the paper. The circulating numbers do match the syringe: at 25 mg/mL, 1 mg is 4 units, 2 mg is 8, 3 mg is 12.
Reported because it is what people use. Nothing here recommends any amount.
How long it lasts
Nobody has published one. I could find no measurement of how long it lasts in a person by any route, and no published study putting it into a person by injection.
Route studied
On the skin, as a cream or a gel, in every published human study. In animals, the only dose figure that names both a species and a route is 15 milligrams for every kilogram of body weight, in aged mice.
Route used
Injected under the skin, from a powder mixed with liquid.
How to check you have the right molecule
The name to check is AHK-Cu. The paper most often held up as evidence for this and hair is actually a study of AHK-Cu, the same idea built with alanine in the first position instead of glycine, one letter apart in writing and a different molecule in the bottle, and a generic title lets one pass for the other. So read the three building blocks named on the certificate, glycine then histidine then lysine, rather than the title of whatever is linked as proof. On the chemistry, the whole complex weighs 401.91 and copper is 15.8 percent of that by weight, though that percentage is arithmetic from the weight rather than something a paper measured, and this entry publishes no test of any vial on sale, so there is no purity figure here to hold a certificate against. If a seller points you at a trial record, note that ClinicalTrials.gov has five records for GHK and the only one giving GHK-Cu as the study drug is NCT07437586, which this entry says outright not to cite: it sits inside a cluster of eight registrations under one sponsor at one hospital that Morgan McSweeney documented on 31 July 2026, two of them copies of real Eli Lilly obesity trials.
Buying a vial on the strength of a cream. The one controlled trial that worked applied a gel to diabetic foot ulcers in 1994, closing wounds by a median of 98.5 percent against 60.8 percent for the same base with nothing added, and that study is on every vendor page. The two controlled trials that measured nothing, on leg ulcers in 1992 and on skin after laser resurfacing in 2006, are on none of them. Even the 2 percent gel strength quoted for the positive one is not in the paper, which never states its concentration. And the amounts people inject were not derived from any of that: mix a 50 milligram bottle with 2 millilitres of liquid and you get 25 milligrams in every millilitre, which makes 1 milligram 4 marks on an insulin syringe, 2 milligrams 8 marks and 3 milligrams 12. The amounts that circulate are the easiest marks to read.
What is known about harm
I could find no recorded side effects from injecting this, because I could find no published study that injected it into a person. That means untested rather than clean. The open question is copper. A 2 milligram dose carries roughly 320 micrograms of elemental copper, against a recommended daily intake of 900 micrograms and an upper limit of 10,000. Both of those describe copper eaten in food, and eating it is limited twice over, by how much the gut takes up and by how much the liver bothers to get rid of. An injection goes past both. Those limits exist because taking too much copper over a long period has known consequences, and there is no published work on blood copper, on ceruloplasmin, the blood protein that carries copper, or on copper in the liver, in a single person who injected this and nothing else. Its legal standing in the United States is not settled here either: three pages at the United States Food and Drug Administration timed out and sellers contradict each other.
The arithmetic
GHK-Cu is sold in 50 mg vials. Mixing errors put people an order of magnitude off using a perfectly good vial, and the arithmetic is simple enough to check yourself.
Vial
Water
Concentration
In 10 units
50 mg
1 mL
50 mg/mL
5 mg
50 mg
2 mL
25 mg/mL
2.5 mg
50 mg
3 mL
16.67 mg/mL
1.67 mg
10 units is 0.1 mL on a U-100 insulin syringe, whatever is in it. For any other combination, or to see the arithmetic worked out, use the calculator.
Everything on this page is a summary. The citations, the studies and the reasoning live in the full entry.