GHK-Cu, route by route.

The answer, first

Why this ranking exists

The most-repeated fact about GHK traces to a 1973 thesis written four years before the molecule was identified. The whole human record is three randomised trials, two of them negative.

Most of what you have been told about copper peptides descends from a doctoral thesis written in 1973 — four years before anyone knew what the molecule was.

That is not a rhetorical flourish. The single most-repeated fact about GHK is that plasma levels fall from about 200 ng/mL at age twenty to about 80 ng/mL at sixty. It appears in the two most-cited modern reviews. In one of them the sentence carries no citation at all.

In the other, the citation resolves to A Tripeptide in Human Serum That Promotes the Growth of Hepatoma Cells and the Survival of Normal Hepatocytes. That is a 1973 UCSF PhD thesis¹⁰. GHK was not structurally identified until 1977. A 1973 dissertation could not have quantified an analyte that had not yet been defined.

The clinical claims fare no better. Peer-reviewed reviews cite Augmented effect of a copper tripeptide-containing cosmetic against photodamaged skin, Dermatologic Surgery 1999, volume 25, pages 601 to 609.

Those pages in that volume belong to a paper about vitamin and zinc levels in leg-ulcer patients, by different authors, running four pages rather than nine.

The 71-participant and 41-participant photoaging figures that circulate everywhere hang off that citation. Where they can be traced at all, they lead to conference posters shown at the American Academy of Dermatology in February 2002 and reproduced inside manufacturer product sheets. Never peer-reviewed, methods unreadable.

Here is the honest starting position. Strip out the studies where GHK-Cu was one active among many, and strip out the studies with no control arm. The entire human record for copper peptides is then three randomised controlled trials — 1992, 1994, 2006 — of which two were negative.

An independent review published in August 2026 arrives at the same place from the other direction. Searching the aesthetic-medicine literature to March 2026, it found twenty studies in total, of which eighteen were preclinical and two were randomised⁴⁴.

We cite that ratio and nothing else from it. The review's own headline clinical claim, wrinkle volume and depth reduced against controls, could not be traced back to any indexed primary study. This report does not carry it.

Forty years of mechanism is a reason to run the trial. It is not a substitute for one.

GHK-Cu, route by route

Nine route-and-outcome pairs graded separately, because the market sells them as one question and the evidence answers them differently.

Nine rows. Each one grades a route paired with the outcome it is actually sold for, because those are different questions and the market sells them as one.

### What GHK-Cu actually does — the part that is real

The mechanistic literature is legitimate science and deserves to be stated fairly. The Reims group showed collagen synthesis in cultured fibroblasts beginning at 10⁻¹² molar and peaking at 10⁻⁹ molar⁵.

In rat subcutaneous wound chambers, GHK-Cu raised collagen to 344% of control against total protein at 230%. That is a genuine preferential effect, with type I and III collagen mRNA up and an inactive control tripeptide doing nothing⁴.

Glycosaminoglycan synthesis rose biphasically, favouring dermatan and heparan sulfate, with no effect on hyaluronic acid⁸. Decorin messenger RNA went up, biglycan down⁶.

That is a coherent, replicated, forty-year story. It is also entirely fibroblasts in dishes and rats in cages. Nobody has taken a punch biopsy from a treated human face and measured hydroxyproline.

The 2025 independent review in BioImpacts puts it in one sentence: there is a surprising absence of clinical studies using them⁹.

### Elastin: the claim with no human data behind it

This is where the marketing is furthest ahead of the evidence, so it is worth being precise.

Elastin cross-links have been measured after GHK-Cu exactly once, in Maquart's 1993 rat study. The methods measured desmosine and isodesmosine by ion-exchange chromatography. The result: elastin was present in low amounts, compared with collagen, but was also increased.

Then the sentence that settles it — the increases in uronic acid and elastin were nearly parallel to that of total proteins, with the normalised data not shown⁴.

In the same paper the collagen-to-total-protein ratio rose significantly. Elastin showed no such preferential rise. More tissue was deposited and that tissue contained proportionally normal elastin. That is accretion, not elastogenesis.

Every other elastin claim in circulation traces to cell culture. The primary source behind the most-cited review's elastin sentence is a 2016 paper in a journal PubMed does not index, n=3. It uses a colourimetric assay that boils cell pellets in oxalic acid specifically to dissolve insoluble elastin into soluble α-elastin. That assay destroys the cross-linked network in order to measure it, and therefore cannot distinguish tropoelastin from mature fibre.

The reported rise was flat across four orders of magnitude of concentration, which is the signature of a non-specific effect.

And there is a harder problem underneath. Raising tropoelastin without demonstrating cross-linking is the LOXL1-null phenotype — the failure mode, not the repair. It is also, structurally, what solar elastosis looks like: accumulated, disorganised, poorly cross-linked elastotic material.

No GHK-Cu study distinguishes the two. Not one study in any species has shown GHK-Cu raising lysyl oxidase activity, which is the enzyme that would have to do the cross-linking.

Meanwhile GHK-Cu raises MMP-2, a gelatinase with elastolytic activity — and the Reims group showed that particular effect was reproduced by copper ions alone and not by the peptide⁷.

You should treat any product claiming elastin regeneration from a copper peptide as making a claim nobody has tested in a human being.

The arithmetic nobody publishes

GHK-Cu is 15.8% copper by mass, so roughly 1.9 mg carries a full day of the only approved parenteral copper allowance — into a body with no gut to regulate it.

Start with the mass. GHK is 340.38 daltons. Chelate a copper ion and you get roughly 402 daltons for the complex. Copper is therefore 15.8% of GHK-Cu by mass — about one part in six.

Now the comparator. The only FDA-approved parenteral copper product for adults over 50 kg delivers 300 micrograms of elemental copper per day²⁴. That means roughly 1.9 mg of GHK-Cu delivers a full day's approved parenteral copper allowance. Five milligrams delivers about 2.6 times it.

The obvious objection is that the oral tolerable upper intake level for copper is 10,000 micrograms per day²³, which makes those numbers look small.

That objection does not survive the physiology. Oral copper absorption is a variable-gain valve: bioavailability runs about 75% at an intake of 400 micrograms per day and falls to about 12% at 7.5 mg per day.

The upper limit assumes that valve is working. An injection has no gut.

And even with the valve intact, a metabolic-ward study using stable-isotope copper found regulation was not sufficient. At high intakes it did not prevent retention of more than 0.6 mg of copper per day²². Copper leaves the body in bile, essentially only in bile.

The approved parenteral product carries warnings to monitor hepatic and biliary function and to check serum copper and ceruloplasmin during long-term administration²⁴.

Two facts finish the picture. After an intravenous dose in rats, GHK is rapidly cleaved to the dipeptide HK and eliminated — the carrier disappears fast, the copper does not.

And the chelate is not inert. Glutathione, the most abundant reductant in human plasma, reduces the copper–GHK complex and releases the copper ion. One chemistry group proposes it as a copper donor to the CTR1 cellular transporter¹⁴.

Cyclic voltammetry puts GHK-Cu in a different class from the albumin binding site DAHK — a fast-exchanging, reducible copper carrier, not a lock¹⁵.

None of that makes GHK-Cu dangerous. It makes injecting it a copper delivery method with a quantifiable dose and no monitoring, framed by nobody selling it that way.

The regulatory position, verified. FDA's 503A bulk drug substances document, updated 14 May 2026, lists the substance in Category 1 as GHK-Cu [except for injectable routes of administration]. The exclusion appears in parentheses in the original.

The nominations were withdrawn on 22 April 2026. On 5 May 2026 one nominator clarified it intended to withdraw only the injectable route, and the non-injectable entry was reinstated. FDA has said it intends to consult the Pharmacy Compounding Advisory Committee before the end of February 2027²⁵.

On the agency's significant-safety-risks page, current as of 22 April 2026, the entry for GHK-Cu for injectable routes reads as follows. Compounded injectable drugs containing GHK-Cu may pose risk for immunogenicity due to the potential for aggregation and peptide-related impurities, and there are limited data in humans to inform safety-related considerations²⁶.

Read the parenthetical plainly. The one route the market sells is the one route the nomination abandoned.

GHK-Cu against the whole field

Twenty-one interventions on one ladder. Sunscreen and prescription retinoids grade A. Copper peptide grades D, topically and injected.

Peptides sit on the same ladder as everything else. Twenty-one rows, ordered by grade rather than by category. Here is where copper peptide lands.

The pattern is not subtle. Where a truly inert vehicle was used with blinded assessors — retinol in 2007, tretinoin in 1988 — the control arm did essentially nothing and the active effect was real.

Everywhere the control arm worked, the control was not inert: a moisturiser, a needling procedure, another active injectable, or nothing at all with unblinded before-and-after photography. The interventions that cannot produce a vehicle-controlled result are the ones sold hardest.

Reading the shelf

Three different molecules share the word peptide on an ingredient list, and below the 1% line the order of that list carries no information at all.

Three things, and they will change how you read every copper peptide product you pick up.

One. The three molecules. `Copper Tripeptide-1` is GHK-Cu. `Tripeptide-1` is GHK with no copper at all. `Palmitoyl Tripeptide-1` is a third molecule again — a palmitoylated derivative.

Several widely cited copper peptide studies actually tested tripeptide-1 without copper. Marketing collapses all three into copper peptides. Read the INCI list, not the front of the box.

Two. The 1% line. Under 21 CFR 701.3, ingredients above 1% must be listed in descending order of predominance. Ingredients at or below 1% may be listed in any order after those⁴⁰. The EU rule at Article 19 of Regulation 1223/2009 is functionally identical⁴¹.

Once you cross that line, order carries zero information — 0.9% and 0.001% look the same on a label. You can usually locate the line, because phenoxyethanol caps at 1% and xanthan gum, carbomer, sodium benzoate and potassium sorbate normally sit below it.

A product listing copper tripeptide-1 before those markers is making a real claim. One listing it after is making none.

Three. The arithmetic, so you can run it yourself. Assume a serum density near 1.0, so 30 mL weighs about 30 g. A 30 mL serum at 1% contains 300 mg of GHK-Cu.

At 0.1%, 30 mg. Running it the other way: 1 g of raw material at 1% makes about 100 g of finished product, which is roughly three and a third 30 mL jars.

At 0.1% it makes about 33 of them. Supplier technical sheets for the raw material disagree with each other by two orders of magnitude on recommended use level. One says 0.5% to 1%, another says 0.001% to 0.1%. So there is no standardised concentration for this ingredient, and no rule forcing anyone to state theirs.

A fourth wrinkle. A percentage can be quoted on the complex or on the free peptide.

One milligram of complex contains about 0.845 mg of GHK. So a stated 1% is either 1% or 0.845% depending on the basis used. That is a 15.5% swing, larger than the gap between many competing products, and nothing compels disclosure of which.

What published work actually used. The randomised facial trial stated no concentration at all¹. The venous ulcer trial used 0.4%².

A permeation study used 0.68% aqueous¹⁶. The registered 2026 wound trial uses 0.1% w/w gel²⁸. The numbers in real trials sit at the low end of what the shelf implies, and the highest-concentration products are not the ones with the trials.

None of that is an instruction to mix anything. It describes what appears on labels and in published work, so you can price the gap yourself without needing a number from us.

What actually degrades it

GHK survived fourteen days at 60 °C undamaged. What actually breaks a jar is acid pH, oxidation, thiols and microbiology — not warmth.

The storage advice you have been given is defensible, but almost nobody gives the right reason.

GHK is unusually stable. The only formal preformulation study on record ran forced degradation: in water and in pH 4.5 to 7.4 buffers at 60 °C for fourteen days, no detectable degradation. Four weeks at 40 °C in a niosome, none either.

Aqueous solubility around 325 mg/mL, logD across cosmetic pH −2.38 to −2.49, strongly hydrophilic¹². There is a chemical reason: the classically labile residues are cysteine, methionine, asparagine, glutamine and tryptophan. GHK is glycine, histidine, lysine. It contains none of them.

So the generic peptide rule — freezer for lyophilised powder, refrigerator for solution, short-dated — is a class-wide convention written for the peptides GHK is not. Cheap and prudent. Not evidence about this molecule.

What actually degrades it. Four things, in order of how much they matter in a jar.

• Oxidation. Hydrogen peroxide at 0.5% took GHK to 84.2% remaining in one hour¹². Air and time in an unsealed container are the slow version of that.
• Acid pH. This is the one to care about. The peptide survives acid fine; the copper complex does not. Coordination is three-nitrogen — the N-terminal amine, a deprotonated amide nitrogen, and the imidazole nitrogen. Protonate those and the chelate comes apart. Measured conditional stability falls roughly 87-fold for a single pH unit, from 7.4 down to 6.5¹³. Both raw-material technical sheets specify an optimum of pH 5 to 7. A low-pH vitamin C or AHA product is chemically hostile to the complex.
• Thiols and chelators. Glutathione reduces the complex and releases the copper¹⁴. Any thiol does the same. EDTA and similar chelators compete for the copper directly.
• Ascorbic acid, in both directions. This one is real chemistry, not folklore. In metal-free buffer at pH 7, ascorbate barely oxidises at all — the first-order rate constant is below 6 × 10⁻⁷ per second¹⁹. What people call ascorbate autoxidation is metal catalysis, and copper is one of the best catalysts for it. Putting a copper complex next to vitamin C degrades the vitamin C.

The two incompatibility rules that are folklore. Searches for published support that niacinamide or retinoids destabilise copper peptides return nothing.

Niacinamide is a weak monodentate donor with no evidence it displaces a tridentate chelate holding copper at picomolar conditional affinity. These two rules circulate as received wisdom and have no literature behind them.

The blue tells you one thing, not two. The colour is the copper d–d absorption band of the intact complex, at 605 to 606 nm^13,14.

Reduced copper is colourless, so when the complex is reduced the blue goes with it. Loss of blue is real evidence the copper complex is gone. The reverse does not hold: retained blue proves copper is still bound to something, not that the tripeptide is intact.

Degradation products including free histidine have been identified while copper stayed in solution¹².

So why does a home mixture need the fridge. Not because the peptide is heat-labile. Because the moment you dilute a finished moisturiser with a water-containing addition, you dilute its preservative system below the level at which it was validated.

FDA says this directly: adding water waters down preservatives that are intended to keep microorganisms from growing.

ISO 29621 puts bacterial growth generally above a water activity of 0.90, with yeasts and moulds down to 0.60 — a hydrated cream sits well above both^42,43.

A jar that has been opened daily, fingered, and never challenge-tested is a microbiology problem, and cold slows microbiology. The refrigerator is protecting you from the jar, not the peptide from the heat.

The community file

Our August 2026 sweep of the skin-peptide communities found a real positive record on both routes. Read carefully, it is a strong satisfaction signal and a weak structural one — which is exactly what the clinical file above predicts.

Topical. The dominant report is glow and plumpness — "it gives my mature skin a fresh young glow, and makes it plumper" [Reddit, Apr 2023] — sometimes within days, occasionally overnight.

An overnight change is the tell. Hydration and surface optics move in hours, collagen remodeling moves in months. So the fastest reports are evidence about the vehicle, not the peptide.

The single most useful anecdote in the sweep came with its own control. Under-eye hydration and lip lines reportedly improved while the neck did not [Reddit, Apr 2023]. That is a user-supplied negative control, which makes the report more informative and still proves nothing about mechanism. Repurchase loyalty is high, and loyalty measures satisfaction, not biology.

Injectable. Reports place texture and "thickness" changes at weeks to months ["I think my skin is thicker" — Reddit, Oct 2024]. Hair and nail stories turn up in the longer runs [hair a hairdresser reportedly noticed by week three — Reddit, Feb 2026].

The record's honest standout is a psoriasis photo gallery [Reddit, Jul 2026]. Its author discloses stress management, disease fluctuation, and little added benefit from more exposure. That is the rare anecdote that argues against itself where the data is weak.

Attribution across the injectable file is the worst in the sweep: blends, concurrent topicals, planned procedures and uncertain product identity run through nearly every report.

The published literature does not rescue the gap. A 2025 review of topical GHK notes broad cosmetic use alongside a striking absence of clinical studies establishing effectiveness^c1. The older penetration work demonstrates skin-layer retention in vitro, not a clinical outcome^c2.

FDA's compounding review flags limited human safety data and immunogenicity risk for the injectable route. So the community file lands where the ranking already sits: a real, recurring, satisfying experience — and no controlled evidence that the peptide is what produced it.

The full method for reading records like this one is in the community evidence report.

What doesn't work

Named claims, with the specific evidence that kills each one.

Named, with what kills them.

Injected GHK-Cu for skin. Not unproven. Untested. Zero human trials of any design, and the route is now excluded by name on FDA's own compounding list with a stated immunogenicity concern^25,26.

GHK-Cu for post-laser erythema. Tested and failed. Computer image analysis and blinded evaluators found no earlier resolution¹.

GHK-Cu 0.4% for venous stasis ulcers. Tested and failed against inert vehicle in 86 patients, and beaten by a cheap generic antimicrobial².

Copper peptides outperformed retinoic acid. The source is a 20-person unblinded pilot on normal, non-photodamaged skin, reporting counts out of ten, in a journal that was never MEDLINE-indexed. It is not a head-to-head efficacy trial and never was.

GHK-Cu rebuilds elastin. No human measurement exists. The rodent cross-link data tracked total protein. The cell-culture source is n=3 with no dose-response, using an assay that dissolves the cross-links it implies.

GHK resets 4,000 genes. The underlying paper is a 2012 Connectivity Map analysis of human lung tissue from COPD smokers. It surfaced GHK computationally as a signature-reversing molecule and validated it in lung fibroblasts using GHK, not GHK-Cu¹¹. It is a computational signature match in the wrong organ, repackaged.

Chelated copper is inert. Physiological glutathione reduces it and releases the copper ion¹⁴.

Oral collagen for skin aging. 23 RCTs. The effect exists only in the industry-funded and low-quality subsets³⁴.

Copper peptides with niacinamide, copper peptides with retinoids. No published support for either incompatibility. Formulator folklore.

One correction to our own file. Import Alert 66-80 is titled for GLP-1 receptor agonist bulk drug substances and names semaglutide, tirzepatide, liraglutide, exenatide, dulaglutide, orforglipron and retatrutide. It is not a general peptide import alert, and GHK-Cu is not on it.

What would change this ranking

One properly designed trial, already running, reading out in February 2027.

One thing, and it has a date. NCT07437586 is a Phase 2, randomised, quadruple-masked, vehicle-controlled, split-wound study of topical GHK-Cu gel 0.1% w/w in standardised punch-biopsy wounds, each participant serving as their own control.

Estimated enrolment 60, started 2 February 2026, estimated completion February 2027²⁸. It is the first properly designed GHK-Cu monotherapy trial in twenty years. No result has been posted and nothing can be claimed from it yet.

FAQ

References

All identifiers below were retrieved and verified. Where a claim could not be confirmed against a primary source, it has been softened or omitted rather than cited loosely.

1. Miller TR, Wagner JD, Baack BR, Eisbach KJ, 2006. Effects of topical copper tripeptide complex on CO2 laser-resurfaced skin. Arch Facial Plast Surg 8:252-9 — PMID 16847171
2. Bishop JB et al., 1992. A prospective randomized evaluator-blinded trial of two potential wound healing agents for the treatment of venous stasis ulcers. J Vasc Surg 16:251-7 — PMID 1495150
3. Mulder GD et al., 1994. Enhanced healing of ulcers in patients with diabetes by topical treatment with glycyl-l-histidyl-l-lysine copper. Wound Repair Regen 2:259-69 — PMID 17147644
4. Maquart FX et al., 1993. In vivo stimulation of connective tissue accumulation by the tripeptide-copper complex in rat experimental wounds. J Clin Invest 92:2368-76 — PMID 8227353
5. Maquart FX, Pickart L et al., 1988. Stimulation of collagen synthesis in fibroblast cultures by the tripeptide-copper complex. FEBS Lett 238:343-6 — PMID 3169264
6. Siméon A et al., 2000. J Invest Dermatol 115:962-8 — PMID 11121126
7. Siméon A et al., 2000. Life Sci 67:2257-65 — PMID 11045606
8. Wegrowski Y et al., 1992. Life Sci 51:1049-56 — PMID 1522753
9. Mortazavi SM et al., 2025. Topically applied GHK as an anti-wrinkle peptide: advantages, problems and prospective. BioImpacts 15:30071 — PMID 39963574
10. Pickart L, Margolina A, 2018. Regenerative and Protective Actions of the GHK-Cu Peptide in the Light of the New Gene Data. Int J Mol Sci 19:1987 — PMID 29986520
11. Campbell JD et al., 2012. A gene expression signature of emphysema-related lung destruction and its reversal by the tripeptide GHK. Genome Med 4:67 — PMID 22937864
12. Badenhorst T, Svirskis D, Wu Z, 2016. Physicochemical characterization of native glycyl-l-histidyl-l-lysine tripeptide for wound healing and anti-aging. Pharm Dev Technol 21:152-60 — PMID 25384620
13. Bossak-Ahmad K et al., 2020. Ternary Cu[II] complex with GHK peptide and cis-urocanic acid. Int J Mol Sci 21:6190 — PMID 32867146
14. Ufnalska I et al., 2021. Intermediate Cu[II]-thiolate species in the reduction of Cu[II]GHK by glutathione. Inorg Chem 60:18048-57 — PMID 34781677
15. Hureau C et al., 2011. X-ray and solution structures of Cu[II]GHK and Cu[II]DAHK. Chemistry — PMID 21780203
16. Li H et al., 2015. Microneedle-mediated delivery of copper peptide through skin. Pharm Res — PMID 25690343
17. Li H et al., 2016. Selected biomarkers revealed potential skin toxicity caused by certain copper compounds. Sci Rep 6:37664 — PMID 27892491
18. Philips N et al., 2010. Stimulation of MMP-1 and IL-8 by copper in dermal fibroblasts. Connect Tissue Res 51:224-9 — PMID 20053132
19. Buettner GR, 1988. In the absence of catalytic metals ascorbate does not autoxidize at pH 7. J Biochem Biophys Methods 16:27-40 — PMID 3135299
20. Lau SJ, Sarkar B, 1981. The interaction of copper[II] and glycyl-L-histidyl-L-lysine. Biochem J 199:649-56 — PMID 7340824
21. Zajda J et al., 2024. CE-ICP-MS/MS monitoring of GHK-Cu encapsulation in liposomes. Electrophoresis — PMID 39451062
22. Turnlund JR et al., 2005. Long-term high copper intake: effects on copper absorption, retention and homeostasis in men. Am J Clin Nutr — PMID 15817858
23. NIH Office of Dietary Supplements — Copper, Health Professional Fact Sheet — ods.od.nih.gov
24. Tralement trace element injection, FDA label, NDA 209376, 2020 — accessdata.fda.gov
25. FDA, Bulk Drug Substances Nominated for Use in Compounding Under Section 503A, updated 14 May 2026 — fda.gov/media/94155/download
26. FDA, Certain Bulk Drug Substances for Use in Compounding That May Present Significant Safety Risks, content current as of 22 April 2026 — fda.gov
27. FDA warning letter, Tailor Made Compounding LLC, 1 April 2020 — fda.gov
28. ClinicalTrials.gov NCT07437586 — Phase 2 split-wound topical GHK-Cu gel 0.1% w/w
29. Hughes MC, Williams GM, Baker P, Green AC, 2013. Sunscreen and prevention of skin aging: a randomized trial. Ann Intern Med 158:781-90 — PMID 23732711
30. Topical tretinoin improves photoaged skin: a double-blind vehicle-controlled study. JAMA 1988 — PMID 3336176
31. Improvement of naturally aged skin with vitamin A [retinol]. Arch Dermatol 2007 — PMID 17515510
32. Alpha hydroxy acids for photodamage, vehicle-controlled. Arch Dermatol 1996 — PMID 8651713
33. Independent meta-analysis of peptides for skin aging, 19 RCTs. Front Med 2026 — PMID 41924746
34. Effects of collagen supplements on skin aging: systematic review and meta-analysis stratified by funding source. Am J Med 2025 — PMID 40324552
35. Quantitative histometry of extracellular matrix proteins after IPL. J Drugs Dermatol 2011 — PMID 22052303
36. Radiofrequency microneedling, instrumented human study — PMID 33942050
37. Platelet-rich plasma, masked split-face versus saline. JAMA Dermatol 2018 — PMID 30419125
38. Polynucleotide versus non-crosslinked hyaluronic acid, randomised split-face, n=27. J Dermatolog Treat 2022 — PMID 32248707
39. Retracted: Pak 2014, phase III polynucleotide crow's feet trial — PMID 25473210, retraction notice J Korean Med Sci 2016;31:330, PMID 26839493. Also retracted: polycaprolactone versus polynucleotide, n=218 — PMID 39313949, retraction notice J Cosmet Dermatol 2026;25:e70794
40. 21 CFR 701.3, cosmetic ingredient labelling — ecfr.gov
41. Regulation [EC] No 1223/2009, Article 19 — eur-lex.europa.eu
42. FDA, Microbiological Safety and Cosmetics — fda.gov
43. ISO 29621:2017, Cosmetics — Microbiology — guidelines for risk assessment of microbiologically low-risk products — iso.org
44. Mokhtar J et al., 2026. The Regenerative Potential of GHK-Cu in Aesthetic Medicine. Aesthet Surg J — PMID 42619529. Cited here only for its study count; its wrinkle-reduction claim could not be traced to an indexed primary study
45. Mortazavi SM, Mohammadi Vadoud SA, Moghimi HR. Topically applied GHK as an anti-wrinkle peptide: advantages, problems and prospective. Bioimpacts 2025 — PMID 39963574
46. Hostynek JJ, Dreher F, Maibach HI. Human skin retention and penetration of a copper tripeptide in vitro as function of skin layer. Inflamm Res 2010 — PMID 20703511

Inside Your Peptides is independent. We sell nothing. We disclose every referral relationship where one exists. The report above contains no vendor link; the sourcing desk that follows it is separate, and disclosed. Grades and rankings follow the published criteria in our editorial policy, never referral terms.

This report is for research and educational purposes only. It is not medical advice. Nothing here is a dosing recommendation or a suggestion to obtain, possess or administer any compound. Several substances discussed are unapproved, prescription-only, or subject to active regulatory proceedings. Speak to a qualified clinician before making any decision about your health.