Peptides for skin: what the human studies show
Reviewed 22 sourcesPublished by PepVault
Peptides for skin, by the human data: Argireline, palmitoyl pentapeptide-4, GHK-Cu and SNAP-8 creams, injected BPC-157 and GHK-Cu, and how to patch test.

- Last reviewed
- Sources
- 22
- Questions
- 6
- Compounds
- 6
The short answer
The skin peptides with human trials are creams, and the trials are small, short and mostly run by the makers. Over 8 weeks, a GHK-Cu serum cut wrinkle volume 24%, against 15% for its plain serum, in 20 women; palmitoyl pentapeptide-4 and Argireline beat placebo in 93 and 60 people. Injected GHK-Cu and BPC-157 have no human skin trial. Patch test any new product for 7 to 10 days.
Evidence: Small randomized trials in people for topical GHK-Cu, palmitoyl pentapeptide-4, Argireline and a SNAP-8 patch, most funded by the ingredient's maker; no human trial of any injected peptide on skin; BPC-157 skin data in mice and rats only.

What kinds of skin peptides are there?
A peptide is a short chain of amino acids, the building blocks of proteins. Skincare uses dozens of them, and a dermatology review sorted the ones tested on aging skin into four groups by what they are meant to do1:
- Signal peptides are designed to tell skin cells to make more collagen, the protein that gives skin its firmness. Palmitoyl pentapeptide-4 (pal-KTTKS), the active in the ingredient sold as Matrixyl, is the best known919.
- Neurotransmitter-inhibitor peptides are meant to soften expression lines by damping the nerve signal that makes small face muscles contract, the same signal botulinum toxin blocks. Argireline (acetyl hexapeptide-8) is the main one, and SNAP-8 (acetyl octapeptide-3) is Argireline's six amino acids with two more added1020.
- Carrier peptides deliver a metal to the skin. GHK-Cu, a three-amino-acid peptide bound to copper, is the example.
- Enzyme-inhibitor peptides are meant to slow the enzymes that break collagen down1.
The other world people mean by skin peptides is injected or swallowed: GHK-Cu and BPC-157 injected, melanotan for tanning, and collagen peptides by mouth. They come after the creams below.

Can a peptide in a cream get into the skin?
Only a small one, and only a little of it. The outer layer of skin, the stratum corneum, is a wall of dead cells sealed in fat. Dermatologists Bos and Meinardi argued that its practical limit is 500 daltons, a unit of molecular weight: almost every drug that works through skin is under it, and larger molecules do not pass2. Most skin peptides sit above that line.
| Peptide | Molecular weight | Against the 500-dalton line |
|---|---|---|
| GHK (the peptide in GHK-Cu) | 340 daltons; about 400 with its copper3 | Under |
| Palmitoyl pentapeptide-4 | 802 daltons4 | Over |
| Argireline (acetyl hexapeptide-8) | 887 daltons5 | Over |
| SNAP-8 (acetyl octapeptide-3) | 1,075 daltons20 | Over |
| BPC-157 | 1,419 daltons6 | Well over |

Food and Drug Administration scientists measured what happens to Argireline. They put a cream with 10% Argireline on skin from human donors in a lab for 24 hours: most of it washed off the surface, 0.22% of the dose stayed in the stratum corneum, 0.01% reached the living layer under it, and none was found in the dermis, the deeper layer where collagen is made7. A 2025 review of Argireline concluded that this low penetration limits what it can do, and that whether it reaches the nerve-muscle junction at all is uncertain18. Makers work around the barrier. A palmitoyl group, a fatty tail, makes a peptide more oil-loving, and the GHK-Cu trial below packed its peptide in fat-based nano-carriers because GHK-Cu dissolves in water and resists the fatty outer layer8.

What did the human trials of skin peptides find?
Every trial below measured facial wrinkles, mostly by 3D imaging of the skin or by silicone casts of it, and all but the collagen drink put the peptide on the skin. None lasted longer than 12 weeks.
| Peptide | How applied | Best human study | What was measured | Result | Who funded it | Evidence level |
|---|---|---|---|---|---|---|
| GHK-Cu | Serum in nano-carriers, twice daily | Badenhorst 2016: 40 women aged 40 to 65, 8 weeks, randomized, double-blind, split-face; 20 compared it with the plain serum, 19 with a Matrixyl 3000 product8 | Wrinkle volume and depth by 3D imaging | Volume down 24% from baseline vs 15% on the plain serum, a 55.8% relative difference (p < 0.001), and 31.6% more than the Matrixyl 3000 product (p = 0.004); depth down 32.8% more than the plain serum (p = 0.012)8 | Snowberry New Zealand, with a Callaghan Innovation grant8 | One small trial |
| Palmitoyl pentapeptide-4 | Moisturizer with 3 ppm, split-face | Robinson 2005: 93 women aged 35 to 55, 12 weeks, double-blind, placebo-controlled9 | Wrinkles and fine lines by image analysis and expert graders | Significant improvement over the same moisturizer without it; well tolerated9 | Authors at Procter & Gamble9 | One trial |
| Argireline (acetyl hexapeptide-8) | Cream around the eyes, twice daily | Wang 2013: 60 adults, 4 weeks, randomized 3 to 1 against placebo11 | Global wrinkle grading; roughness on skin casts | Wrinkles improved in 48.9% vs 0% on placebo; every roughness measure fell (p < 0.01)11 | Not stated in the abstract | Two small studies |
| SNAP-8 (acetyl octapeptide-3) | Dissolving microneedle patch, overnight, mixed with a vitamin C form and another active | Shin 2024: 24 adults, 28 days, patch on one eye, plain patch on the other12 | Eye wrinkles, elasticity, water loss | Better than the plain patch on all three, with no side effects12 | Authors at Raphas, the patch's maker12 | One small trial of a mix |
| Collagen peptides | Drink, 2.5 g once daily | Proksch 2014: 114 women aged 45 to 65, 8 weeks, double-blind, placebo-controlled13 | Eye wrinkle volume; collagen in skin samples | Wrinkle volume 20% lower than placebo; procollagen I 65% higher in a subgroup13 | Not stated in the abstract | Limited trials, by the authors' own count13 |
Argireline's first study, by the team that designed it, reported wrinkle depth down by up to 30% after 30 days of a 10% cream in healthy women, with no placebo group in the abstract10. The SNAP-8 patch carried other actives and needles that pierce the outer layer, so its result belongs to the whole patch, not to SNAP-8 alone12. GHK-Cu's skin trials, including a laser-recovery trial that found no measured difference, have their own guide: GHK-Cu for skin and hair.
Commonly reported: smoother texture and softer fine lines after one to two months of daily use, and little stinging compared with retinoids or acids, which is why many people layer a peptide serum with them. These are user reports, not measurements.
How strong is the evidence for skin peptides?
Real but thin. The trials are positive, and three things limit them. They are small: 24 to 114 people. They are short: 4 to 12 weeks, so nobody knows what a year adds. And most were run or paid for by a company selling the product or the ingredient, as the table's funding column shows. A 2021 systematic review of over-the-counter anti-aging ingredients from the University Hospital Zurich notes that the evidence behind them is often lacking14. A November 2026 systematic review of peptides used after laser and microneedling found six studies, with benefits for crusting, redness and recovery, held back by small samples and short follow-up15.
The closest to a head-to-head is a Procter & Gamble trial in 196 women: a three-product regimen with peptides, niacinamide and retinyl propionate, a mild relative of retinol, beat prescription 0.02% tretinoin on wrinkles at 8 weeks, matched it at 24 weeks in the smaller group that continued, and irritated less21. It tested the whole regimen, so no trial ranks a single peptide against a retinoid.
What about injected peptides for skin, like GHK-Cu and BPC-157?
As of November 2026, neither GHK-Cu nor BPC-157 has been tested by injection on skin in a published human trial. GHK-Cu's only human injection data is a scalp mixture for hair that cannot isolate it; the rest comes from animals, which the GHK-Cu guide covers. BPC-157's skin results come from mice and rats: in burned mice, a BPC-157 cream improved every healing measure the Zagreb group scored, including new skin cover by two weeks and the strength of the healed skin16, and in rats with chemical burns a BPC-157 cream closed the wounds faster22. No trial has tested BPC-157 on a person's wound or skin. Its human data, which is not about skin, is in the BPC-157 human data guide.
Melanotan, injected for tanning, works on pigment rather than wrinkles, and our guide melanotan 1 vs melanotan 2 sets out its trials and side effects. Collagen peptides by mouth are a food protein, not a signal peptide; the guide on collagen versus research peptides covers what eaten collagen does in the body.
How do you patch test a new peptide product?
The American Academy of Dermatology advises testing any new skincare product before it goes on your face17:
- Choose a spot about the size of a quarter where the product will not be rubbed or washed off, such as the underside of your arm or the bend of your elbow17.
- Apply the normal amount you would use, as thick as you would use it, and leave it on as long as you normally would17.
- Repeat twice a day for 7 to 10 days17.
- If your skin reacts, gently wash the product off as soon as you can and do not use it again; a cool compress or petroleum jelly can calm the spot17.
- If you see no reaction after 7 to 10 days, start using the product as directed17.
PepVault's side-effect log keeps each reaction with its date and a severity from 1 to 5, and the doctor's report lists them, so a dermatologist can see what happened and when. Log it in PepVault
Some ingredients often paired with peptides, such as retinol and glycolic acid, irritate the skin in a normal, temporary way17. A severe reaction that does not settle with those steps is a reason to see a dermatologist17.
Sources
- Gorouhi F, Maibach HI. Role of topical peptides in preventing or treating aged skin. Int J Cosmet Sci. 2009. PubMed 19570099
- Bos JD, Meinardi MM. The 500 Dalton rule for the skin penetration of chemical compounds and drugs. Exp Dermatol. 2000. PubMed 10839713
- PubChem. Glycyl-histidyl-lysine (GHK), CID 73587: molecular weight 340.38; copper complex CID 165429100: 401.91. pubchem.ncbi.nlm.nih.gov
- PubChem. Palmitoyl pentapeptide-4 (pal-KTTKS), CID 9897237: molecular formula C39H75N7O10, molecular weight 802.1. pubchem.ncbi.nlm.nih.gov
- PubChem. Acetyl hexapeptide-3, synonym acetyl hexapeptide-8 (Argireline), CID 71587772: molecular formula C35H62N14O11S, molecular weight 887.0. pubchem.ncbi.nlm.nih.gov
- PubChem. BPC-157, CID 9941957: molecular formula C62H98N16O22, molecular weight 1419.5. pubchem.ncbi.nlm.nih.gov
- Kraeling ME, Zhou W, Wang P, Ogunsola OA. In vitro skin penetration of acetyl hexapeptide-8 from a cosmetic formulation. Cutan Ocul Toxicol. 2015 (US Food and Drug Administration). PubMed 24754410
- Badenhorst T, Svirskis D, Merrilees M et al. Effects of GHK-Cu on MMP and TIMP expression, collagen and elastin production, and facial wrinkle parameters. J Aging Sci. 2016;4:166. doi.org
- Robinson LR et al. Topical palmitoyl pentapeptide provides improvement in photoaged human facial skin. Int J Cosmet Sci. 2005 (The Procter & Gamble Company). PubMed 18492182
- Blanes-Mira C et al. A synthetic hexapeptide (Argireline) with antiwrinkle activity. Int J Cosmet Sci. 2002. PubMed 18498523
- Wang Y et al. The anti-wrinkle efficacy of argireline, a synthetic hexapeptide, in Chinese subjects: a randomized, placebo-controlled study. Am J Clin Dermatol. 2013. PubMed 23417317
- Shin JY et al. Clinical safety and efficacy evaluation of a dissolving microneedle patch having dual anti-wrinkle effects with safe and long-term activities. Ann Dermatol. 2024 (Raphas Co., Ltd.). PubMed 39082657
- Proksch E et al. Oral intake of specific bioactive collagen peptides reduces skin wrinkles and increases dermal matrix synthesis. Skin Pharmacol Physiol. 2014. PubMed 24401291
- Imhof L, Leuthard D. Topical over-the-counter antiaging agents: an update and systematic review. Dermatology. 2021. PubMed 32882685
- Padmanabhan R et al. Peptides as cosmeceuticals for post-procedural wound healing: a systematic review. J Cosmet Dermatol. 2026. PubMed 43047393
- Mikus D et al. Pentadecapeptide BPC 157 cream improves burn-wound healing and attenuates burn-gastric lesions in mice. Burns. 2001. PubMed 11718984
- American Academy of Dermatology. How to test skin care products (last updated August 10, 2021). aad.org
- Zdrada-Nowak J, Surgiel-Gemza A, Szatkowska M. Acetyl hexapeptide-8 in cosmeceuticals: a review of skin permeability and efficacy. Int J Mol Sci. 2025. PubMed 40565185
- Gomes A et al. Boosting cosmeceutical peptides: coupling imidazolium-based ionic liquids to pentapeptide-4 originates new leads with antimicrobial and collagenesis-inducing activities. Microbiol Spectr. 2022. PubMed 35950860
- PubChem. SNAP-8 (acetyl octapeptide-3), CID 76283482: Ac-Glu-Glu-Met-Gln-Arg-Arg-Ala-Asp-NH2, molecular weight 1075.2. pubchem.ncbi.nlm.nih.gov
- Fu JJ, Hillebrand GG, Raleigh P et al. A randomized, controlled comparative study of the wrinkle reduction benefits of a cosmetic niacinamide/peptide/retinyl propionate product regimen vs. a prescription 0.02% tretinoin product regimen. Br J Dermatol. 2010 (The Procter & Gamble Company). PubMed 20374604
- Huang T, Zhang K, Sun L et al. Body protective compound-157 enhances alkali-burn wound healing in vivo and promotes proliferation, migration, and angiogenesis in vitro. Drug Des Devel Ther. 2015. PubMed 25995620
Newsletter · Weekly
The weekly peptide briefing.
New guides, the week’s peptide news, one tool. Free, one email a week, unsubscribe anytime.