Matrixyl: what the studies show, status and safety
Summary
Matrixyl is the trade name of Palmitoyl Pentapeptide-4, also written Pal-KTTKS: a five-part fragment of a collagen precursor with a fatty acid attached. It is a skin-care ingredient, applied to the skin at a few parts per million, and it has never been approved or studied as an injection anywhere. One vehicle-controlled trial of the peptide on its own exists, in 93 women, run by the company that sold the cream, and it published no effect sizes. The only permeation study found 0.6 per cent of the applied dose in the dermis, and the American expert panel's safety review, first drafted in 2023, has never reached a conclusion.
Key findings at a glance
- Only 0.6 per cent reaches the dermis. After 24 hours on skin, 14.6 per cent of the applied peptide was still in the skin: 8.3 per cent in the outer horny layer, 5.6 per cent in the epidermis, 0.6 per cent in the dermis [1]. The dermis holds the cells the whole idea depends on.
- The efficacy case rests on one trial. Ninety-three women, 3 parts per million, 12 weeks, one half of the face against the other. It reported improvement in lines and wrinkles, but published no effect sizes and no p values, and all six authors worked for the company selling the cream [2].
- No registered study exists. Searches of the trial register on 20 September 2026 under five different names returned nothing at all [3].
- The safety review is unfinished, not passed. The American expert panel's first draft assessment of this ingredient group went to its meeting of 11 and 12 September 2023, and no conclusion has appeared since. The draft records that no reproductive or developmental toxicity studies, no cancer studies and no repeat-dose studies were found or submitted [4].
- It is nonetheless everywhere. Industry figures from 2023 put Palmitoyl Pentapeptide-4 in 239 cosmetic products, 223 of them leave-on. In 2012 the figure was 51 [4].
- The amounts are minute. The raw material holds about 100 parts per million of peptide, and finished creams end up with up to roughly 8 parts per million [4].
- Nobody has ever injected it in a study. No market has approved an injectable form, and no published human or animal work gives it into the body [3], [5], [6].
What it is
Matrixyl is a lipopeptide: five amino acids, Lys-Thr-Thr-Lys-Ser, with a sixteen-carbon fatty acid tied to the front end. The peptide half is not invented. It is residues 212 to 216 of the tail piece of the precursor of type I collagen [7].
Katayama and colleagues had found in 1991 that a larger piece of that tail made connective-tissue cells produce more matrix. In 1993 they narrowed it to this five-part stretch, which they called the smallest sequence that still worked [7]. The fatty tail came later, to make a water-loving peptide greasy enough to get past the outer skin.
The raw material is made by Sederma in France, part of Croda International, and has been sold since 1999 [4], [8]. It is a cosmetic ingredient with one declared job, keeping skin in good condition [9]. It is applied to the skin and nothing else.
Quick facts
| Other names | Palmitoyl Pentapeptide-4 (the official ingredient name), Palmitoyl Pentapeptide-3 (the former one), Pal-KTTKS, C16-KTTKS, Lipopentapeptide 3, Oristar POPP [10] |
| Class | Synthetic lipopeptide from a collagen precursor; a skin-conditioning cosmetic ingredient, not a hormone [9], [10] |
| Sequence | Lys-Thr-Thr-Lys-Ser, with palmitic acid on the front end and a free acid group at the back [7], [10] |
| Formula and mass | C39H75N7O10, 802.06 g/mol [11] |
| Route | On the skin only; no injectable form approved or studied anywhere [3], [5] |
| Status | Allowed as a cosmetic ingredient everywhere checked; approved as a medicine nowhere [5], [6] |
| Maker | Sederma, part of Croda International; on the market since 1999 [4], [8] |
| CAS | 214047-00-4 [11] |
| PubChem CID | 9897237, taken from two other databases because PubChem itself was unreachable on 20 September 2026 [10], [11] |
| UNII | KK181SM5JG [10] |
How it works
The proposed idea is a feedback loop. When collagen is broken down, small fragments are released, and one of them may tell the cells of the dermis to make more. Every step of that has been shown in dishes, and none of it in people.
- It may signal the cells that build the matrix. The plain five-part peptide raised the production of type I collagen, type III collagen and fibronectin in connective-tissue cells, without changing total protein output [7]. That is a cell-culture result, not a skin result.
- Self-assembly may be part of it. The fatty version clumps into ordered ribbons above a certain concentration. Collagen production in human skin and cornea cells rose in step with concentration, and did so close to that threshold, which the authors take to mean the two are linked [12].
- A growth factor may be involved. In rat tendon cells the plain peptide without the fatty tail raised type I collagen, slowed the breakdown of its messenger RNA (p = 0.021) and increased TGF-beta in the culture fluid in step with the dose (p = 0.001) [13]. Rat cells in a dish, again, and not the molecule that goes into creams.
- The effect may run in both directions. At 0.1 micromoles per litre the fatty peptide cut the share of cells carrying scar-type stress fibres from 75 ± 7.1 to 38.6 ± 16.1 per cent (three replicates per condition, p < 0.05). At 0.5 micromoles per litre the effect was no longer significant, and the authors warned that the balance against the peptide's scar-promoting side needs care [14].
The obvious objection is arithmetic. If only 0.6 per cent of an applied dose reaches the dermis [1], very little of any of this can be happening there.
A 2011 review already noted the surprising absence of penetration data for a substance in such wide use [15]. A 2022 review put the general problem bluntly: without the ability to cross the skin, cell-level potency is useless [16].
This separates Matrixyl from injected peptides entirely. A bioinformatics safety exercise grouped it with ingredients that echo skin matrix proteins rather than with hormone-like peptides [17]. It is a piece of a structural protein, not a messenger the body sends through the blood.
What the studies found
The literature is large and beside the point. A PubMed search on 20 September 2026 returned 44 papers [18] and Europe PMC 131 records [19], but most concern materials science: ribbons, liposomes, microneedles, hydrogels and analytical methods. Human efficacy data amount to three papers, only one of which tests the peptide on its own.
Wrinkles and fine lines
Human RCT Ninety-three white women aged 35 to 55 used a moisturiser with 3 parts per million of the peptide on one half of the face and the identical moisturiser without it on the other, for 12 weeks [2]. The study was double-blind and randomised side to side.
It reported a significant improvement against the control for wrinkles and fine lines, by image analysis and by trained assessors. Self-assessment also improved. Other measures showed only what the authors called directional effects, and the peptide was well tolerated by the skin.
What it did not report is the problem. No effect sizes, no absolute numbers, no p values and no confidence intervals appear in the abstract, and the full text is not openly available [2]. All six authors worked at Procter & Gamble, which sold the tested cream. Nearly every claim made for Matrixyl since traces back to this one study.
Crow's feet
Human RCT Twenty-one Indonesian women aged 26 to 55 were split into three groups: an acetyl hexapeptide-3 cream, a Palmitoyl Pentapeptide-4 cream and a placebo, twice daily around the eye for eight weeks [20]. That is roughly seven people per group.
The authors report that improvements were found in several participants on both peptides, and that the pentapeptide appeared to do better than the hexapeptide. No p values or effect sizes accompany that. The concentration used is not stated, and the authors describe their own work as a first study needing a larger sample.
Multi-ingredient products
Human RCT Two double-blind randomised split-face studies, of 42 and 35 women aged 30 to 70, measured wrinkles around the eye with three-dimensional surface imaging after four weeks [21]. Participants applied 0.5 g of product twice a day.
The products contained niacinamide, two peptides including Pal-KTTKS, and carnosine. There was no arm with the pentapeptide alone, so nothing in the result can be assigned to it.
How much gets in
Laboratory data This is the strongest single finding on the page. Using excised skin from hairless mice in diffusion cells and mass spectrometry, researchers followed both the plain peptide and the fatty version for 24 hours, and watched the fluid beneath the skin for 48 [1].
Neither appeared in that fluid, so neither crossed the skin. The plain peptide was not detectable in any skin layer at all. The fatty version was found in all of them: of the 50 µg/cm² applied, 4.2 ± 0.7 µg/cm² sat in the horny layer, 2.8 ± 0.5 in the epidermis and 0.3 ± 0.1 in the dermis.
Excised skin, 24 hours, measured by mass spectrometry. 14.6 per cent stayed in the skin in total; none crossed into the fluid below. Source [1].
The fatty tail does earn its place, though. The same study measured how fast each version is chewed up by skin enzymes, and the difference is large.
Whole-skin homogenate, 60 minutes. The plain peptide was almost gone; the fatty version was not. Source [1].
Laboratory data Pre-treating excised human skin with microneedles raised the fluorescence signal for the peptide by a factor of two to twenty-two [22]. That is a measurement of what damaged skin lets through, not of what a cream does.
Cells and animals
Laboratory data In connective-tissue cells from human skin and human cornea the fatty peptide raised collagen production in step with concentration, peaking near the point where it starts forming ribbons [12]. A 2026 follow-up reproduced this at 0.0062 per cent by weight, and a parallel paper found cell survival dropping below 70 per cent above certain concentrations [23], [24].
Animal data A rat wound model ran seven groups over 21 days. Healing rose from 63.5 per cent in the untreated control to as much as 81.81 per cent in the treated groups (p < 0.05 to p < 0.001). A patch and a cream each carrying 1 mg did better than the positive control dressing (p < 0.05) [25].
Animal data A second rat study, on burns, used the peptide in a cellulose dressing with and without stem cells, and reported faster closure [26].
Laboratory data One study cuts the other way. A series of variants of the peptide was tested on skin cells; none was toxic and three encouraged growth, but those same three showed no relationship between concentration and effect in the collagen and DNA synthesis assays [27].
What is still unknown
- Whether any of it works in people. One vehicle-controlled single-ingredient trial exists, it is manufacturer-run, and nobody independent has repeated it [2].
- What the effect size is. The key trial published none, so the size of the reported improvement is not on public record [2].
- How much reaches living cells in human skin. The only permeation numbers come from mouse skin in a laboratory dish [1].
- What happens beyond 12 weeks. The longest controlled study ran 12 weeks; the marketing speaks of six months [2], [8].
- Reproduction, development and cancer. The 2023 draft assessment records that no such studies were found in the literature and none were submitted by industry; the same is true of repeat-dose toxicity [4].
- How the body handles it. The whole toxicokinetic section of that draft consists of the one laboratory permeation experiment. There is no measurement of systemic exposure in a person [4].
- Anything at all about injection. No human data, no animal data on giving it into the body, no purity standard for such a form [3], [4].
Side effects and safety
On the skin, at the concentrations used, the picture is unremarkable. What is missing is a finished independent assessment: the American expert panel's first draft on this ingredient group went to its September 2023 meeting, and no conclusion has been published since [4].
- Well tolerated in the 12-week trial, in the authors' words, with no event rates given [2].
- One mild reaction in ten. In an acute irritation test of a branded mixture holding 0.01 per cent of the peptide, one of ten participants showed very slight redness; the irritation index was 0.10 [4].
- Not sensitising. A repeat-insult patch test in 106 people, on a preparation of the other sequence Pal-KTSKS, and a second in 51 people on a branded Pal-KTTKS mixture found neither irritation nor sensitisation [4].
- Not mutagenic. A bacterial test of Pal-KTTKS up to 5,000 micrograms per plate was negative; a second bacterial test, of Pal-KTSKS, was also negative but showed signs of cell toxicity at the test conditions. A micronucleus test in human white blood cells was negative up to 1,000 micrograms per millilitre with metabolic activation, 750 without and 400 under continuous treatment [4].
- Not phototoxic. The substance absorbs no ultraviolet light between 290 and 400 nanometres [4].
- Eye irritation is unsettled. One membrane test rated a branded mixture a moderate eye irritant, another rated a different preparation slightly irritating, while a corneal model and a rabbit test found nothing [4].
- One hormone signal, not reproduced. A yeast-based screen of a preparation holding 0.12 per cent of the Pal-KTSKS sequence showed weak oestrogen-like activity at an EC10 of 6.9 × 10⁻³, with cell toxicity at the two highest concentrations. A second yeast screen of the same preparation found neither [4].
Two qualifications matter more than any single result. First, a large share of the modern toxicity work in the 2023 file was done on Pal-KTSKS, a different sequence sold under the same ingredient name, not on Pal-KTTKS [4].
Second, the gaps are not incidental. The 2023 draft states in plain words that no reproductive or developmental toxicity studies, no cancer studies and no repeat-dose studies were found or submitted, and its section on how the body handles the substance contains nothing but the single laboratory permeation experiment [4].
No medicines authority has issued a warning about it. Searches of the American warning-letter database under three names returned no results, and the recall database returned nothing [28], [29]. That reflects a substance nobody has had to police, not a clean bill of health.
Doses used in studies
The doses below are simply what the cited studies gave, listed so that the results can be understood. They are not advice on how to use anything.
Read the concentration column carefully. Every human figure here is a concentration in a cream, measured in parts per million, which is millionths by weight. None of them is an injection dose, and no injectable form of this substance has been approved or studied anywhere [3], [5].
| Study | Model or population | Concentration | Route | Duration | Ref |
|---|---|---|---|---|---|
| Robinson 2005 | 93 women, 35 to 55, photoaged facial skin | 3 ppm in a moisturiser, 0.0003 per cent | On the skin, face | 12 weeks | [2] |
| Aruan 2023 | 21 Indonesian women, 26 to 55 | Cream, concentration not stated, twice a day | On the skin, around the eye | 8 weeks | [20] |
| Kaczvinsky 2009 | 42 and 35 women, 30 to 70 | Finished products with several actives, peptide content not stated; 0.5 g twice a day | On the skin, around the eye | 4 weeks | [21] |
| Industry figures in the 2023 expert file | Market products | Raw material about 100 ppm; up to about 8 ppm in the finished product; meeting note "3, 5, maximum 10 ppm"; highest concentration on record for this ingredient 0.00061 per cent, in eye lotions and face and neck products in 2013 | On the skin | Continuous use | [4] |
| Jones 2013 and de Mello 2026 | Human dermal skin cells | Near the ribbon-forming threshold; 0.0062 per cent by weight in the 2026 work | Added to the culture fluid | Culture period | [12], [23] |
| Park 2017 | Skin cells in three-dimensional collagen lattices | 0 to 0.5 µmol/l; active at 0.1, no longer significant at 0.5 | Added to the culture fluid | Culture period | [14] |
| Kachooeian 2022 | Rats, skin wounds | Patch with 0.1 mg or 1 mg; cream with 1 mg | On the wound | 21 days | [25] |
The 0.05 per cent figure that circulates for eye make-up belongs to Myristoyl Pentapeptide-4, the fourteen-carbon relative reviewed in the same file, and not to this substance [4].
One number trips people up constantly. The raw material is not the peptide: it is a watery solution holding roughly 100 parts per million of it, alongside glycerin, butylene glycol and thickeners [8].
A label boasting of a high percentage of Matrixyl is describing that solution. An analytical study also showed that the formulation itself changes how much of the peptide is available in a cream [30].
Development and approval status
Matrixyl has never been in drug development. There is no registered trial, no phase 1 study and no development sponsor; as a cosmetic ingredient it has been on the market for a quarter of a century [3], [4].
The one regulatory process it has been through is a cosmetic safety review, and that review is still open. The substance first appeared in a 2013 draft report on palmitoyl oligopeptides, which the panel tabled so that the ingredients could be regrouped by peptide sequence rather than by what the peptide was attached to. The pentapeptides came back as their own report in August 2023, for the panel meeting of 11 and 12 September 2023. That draft carries no conclusion, and none has been published since [4].
From a collagen fragment to an unfinished safety review
- 1991A piece of the collagen precursor is shown to drive matrix productionReported within the 1993 paper, ref [7]
- 1993KTTKS identified as the smallest sequence that worksResidues 212-216, J Biol Chem, ref [7]
- 1999The fatty version reaches the market as MatrixylSederma, later part of Croda, refs [4], [8]
- 2005The one vehicle-controlled trial is published93 women, 3 ppm, 12 weeks, ref [2]
- 2013-2014An earlier review of palmitoyl peptides is tabled and regroupedPalmitoyl Oligopeptides, ref [4]
- 2014The permeation study puts 0.6 per cent in the dermisExcised skin, mass spectrometry, ref [1]
- September 2023The first draft assessment of the pentapeptides goes to the panelNo conclusion published since, ref [4]
- 20 September 2026Registers checked: cosmetic everywhere, medicine nowhereRefs [3], [5], [6]
- 1993 — the five-part sequence is identified as the smallest piece of the collagen precursor that still stimulates matrix production [7].
- 1999 — the palmitoylated version goes on sale as a cosmetic raw material [4], [8].
- 2005 — the split-face trial in 93 women is published [2].
- 2013 and 2014 — the substance sits in a broader review of palmitoyl peptides, which is tabled and then regrouped by peptide sequence [4].
- 2014 — the permeation study measures what actually gets into each skin layer [1].
- 2023 — the first draft assessment of the pentapeptide group reaches the panel in September; nothing has been concluded since [4].
- 2026 — no approval, no restriction and no registered trial anywhere checked [3], [5], [6].
| Market | Status | Note | Ref |
|---|---|---|---|
| United States | Cosmetic | Expert safety review opened in 2023, never concluded; no medicine approved | [4], [5] |
| European Union | Cosmetic | Absent from every annex of the cosmetics regulation | [31], [6] |
| Germany | Cosmetic | EU rules apply directly | [31] |
| United Kingdom | Cosmetic | No medicine entry | [32] |
| Australia | Cosmetic | Not scheduled in the Poisons Standard | [33] |
| Canada | Cosmetic | No medicine entry | [34] |
| Switzerland | Cosmetic | No medicine entry | [35] |
Every search behind this table was run on 20 September 2026. The European check was thorough: the consolidated text of the cosmetics regulation, some 2.16 million characters, contains the word pentapeptide not once [31].
Australia is worth a second look, because it usually catches peptides. The June 2026 Poisons Standard holds no entry for this one [33]. Collagen itself is prescription-only there, but only in preparations for injection or implantation, which leaves fragments applied to skin outside.
The substance also sits outside the American compounding lists entirely. It was never put forward for pharmacy compounding, unlike BPC-157, TB-500 or GHK-Cu, which appear among withdrawn nominations [36].
The grey area is elsewhere. Three products holding the peptide are self-registered in the American label database as over-the-counter drugs, with no approval number. One of those labels promises to remove scars and improve skin elasticity [37], [38]. Claims like that turn a cosmetic into an unapproved medicine, whatever its chemistry.
Anti-doping
Matrixyl is not banned in sport. The full text of the 2026 Prohibited List contains no mention of pentapeptides, palmitoyl compounds, KTTKS or Matrixyl, and the annex to German anti-doping law names none of them either [39], [40].
The only class worth discussing is S0, which covers any pharmacological substance that no health authority anywhere approves for human treatment. This substance does lack any such approval. But a cosmetic ingredient applied to the skin as intended is not a pharmacological substance, so the class does not bite [39].
That reasoning depends on the route. A cream holding a few parts per million triggers nothing and requires no declaration. Anyone injecting it would be outside every scenario the evidence covers, in doping terms as in every other. The wider picture is at peptides banned in sport.
Compared with related peptides
| Peptide | What it is | Route | Human evidence | Status |
|---|---|---|---|---|
| Matrixyl (Pal-KTTKS) | Five-part collagen precursor fragment with a fatty tail | On the skin [3] | One vehicle-controlled trial, n = 93, no effect sizes [2] | Cosmetic everywhere checked; safety review unfinished [4] |
| GHK-Cu | Three-part peptide bound to copper | On the skin and, on the grey market, injected | Not assessed on this page | Cosmetic; also on the withdrawn-nomination list for compounding [36] |
| SNAP-8 | Eight-part peptide aimed at muscle signalling | On the skin | Not assessed on this page | Cosmetic; approved as a medicine nowhere |
| Matrixyl 3000 | Palmitoyl Tripeptide-1 plus Palmitoyl Tetrapeptide-7 | On the skin | Not assessed on this page | Cosmetic; its ingredients were reviewed in 2014 [41] |
| Pentapeptide-4 | The same five amino acids without the fatty tail | On the skin | Not assessed on this page | Cosmetic; one reported use against 239 [4] |
The rows to read together are the first and the last. They differ by one fatty acid, and that difference decides everything measurable: the plain peptide was undetectable in every skin layer, while the fatty one reached all of them [1].
The fourth row is the one that causes most trouble. Matrixyl 3000 shares a brand and contains neither of these molecules [41]. Its 2014 safety review is routinely quoted as if it applied here, and it does not.
GHK-Cu makes a fair neighbour in one respect and a misleading one in another. Both are short peptides derived from collagen chemistry and sold for skin. But GHK-Cu carries a copper ion, is injected on the grey market and has been nominated for pharmacy compounding and withdrawn [36]. Matrixyl has never left the cosmetics shelf.
Common misconceptions
- "Matrixyl and Matrixyl 3000 are the same thing." They share a brand name and no chemistry. Matrixyl is Pal-KTTKS; Matrixyl 3000 is Palmitoyl Tripeptide-1 with Palmitoyl Tetrapeptide-7; Matrixyl synthe'6 is Palmitoyl Tripeptide-38 [41].
- "The Cosmetic Ingredient Review declared it safe." It did not. Its first draft assessment of this ingredient group went to the panel in September 2023 and no conclusion has been published since. The claim usually rests on the 2014 report about the Matrixyl 3000 ingredients [4], [41].
- "Palmitoyl Pentapeptide-4 on a label always means Pal-KTTKS." The 2023 report covers two different sequences under that one name, Pal-KTTKS and Pal-KTSKS, from different makers. A good deal of the safety data belongs to the second one [4].
- "Palmitoyl Pentapeptide-3 is a different ingredient." It is the former official name of the same substance, and it still appears on older labels and in older papers [10].
- "Myristoyl Pentapeptide-4 is the same substance." It is not. That one carries a fourteen-carbon tail and has its own entry. The single trial registration that surfaces in searches, an eyelash serum study, tested that molecule and not this one [42].
- "There are clinical trials of Matrixyl." None is registered. Five separate search terms returned nothing on 20 September 2026. The published human work consists of unregistered studies commissioned by the cosmetics industry [3].
- "It is proven more effective than vitamin C and as effective as retinol." That comparison comes from the raw-material maker's own marketing. No peer-reviewed publication supporting it was found, nor were the six-month comparative data it refers to [8].
- "Cell studies show it builds collagen, so it works on skin." The cell studies are real and the objection is physical: 0.6 per cent of an applied dose reaches the dermis, where those cells live [1], [16].
Frequently asked questions
What is Matrixyl?
Matrixyl is a trade name for Palmitoyl Pentapeptide-4, usually written Pal-KTTKS. Five amino acids, Lys-Thr-Thr-Lys-Ser, carry a sixteen-carbon fatty acid at the front end. The five-part stretch is a piece of the precursor of type I collagen, identified in 1993, and the fatty tail was added to help it cross the outer skin.
Is Matrixyl the same thing as Matrixyl 3000?
No. They share a brand and nothing else. The original Matrixyl is Pal-KTTKS. Matrixyl 3000 contains two different molecules, Palmitoyl Tripeptide-1 and Palmitoyl Tetrapeptide-7, and Matrixyl synthe'6 is Palmitoyl Tripeptide-38. Studies and safety assessments of one of them are not evidence about the others.
Does Matrixyl actually reduce wrinkles?
The evidence is thinner than the marketing suggests. Exactly one trial tested the peptide on its own against the identical cream without it: 93 women, 3 parts per million, 12 weeks. It reported a significant improvement in lines and wrinkles, but published no effect sizes, no p values and no confidence intervals, and all six authors worked for the company selling the cream.
How much Matrixyl reaches the dermis?
About 0.6 per cent of what is applied, in the one study that measured it. Of the applied dose, 14.6 per cent stayed in the skin after 24 hours: 8.3 per cent in the outer horny layer, 5.6 per cent in the epidermis and 0.6 per cent in the dermis. None of it passed through the skin into the fluid below. The dermis is where the cells the peptide is meant to act on sit.
Is Matrixyl ever injected?
No. There is no approved injectable form of Palmitoyl Pentapeptide-4 in any market checked, and no published study has injected it into a person. There are no animal data on giving it into the body either, and no pharmacopoeia purity standard for an injectable grade exists. Everything known about it concerns creams applied to the skin.
Has the Cosmetic Ingredient Review declared Matrixyl safe?
No. The panel's first draft assessment of this ingredient group was released in August 2023 for the meeting of 11 and 12 September 2023, and no conclusion has been published since. The draft itself records that no reproductive or developmental toxicity studies, no cancer studies and no repeat-dose studies could be found or were submitted. The widespread claim of a clean bill of health usually confuses this with the 2014 report on the Matrixyl 3000 ingredients.
What concentration of Matrixyl is used in cosmetics?
Very little. The raw material is a watery solution holding roughly 100 parts per million of peptide, and the finished cream ends up with up to about 8 parts per million. A raw-material chemist told the expert panel in 2014 that normal use was 3, 5, at most 10 parts per million. The trial cream contained 3 parts per million.
Is Matrixyl approved as a medicine anywhere?
No. Register searches in the United States, the European Union, the United Kingdom, Canada, Australia and Switzerland were run on 20 September 2026 and came back empty. As a cosmetic ingredient it is allowed everywhere checked and restricted nowhere. A skin product that claims to treat a disease falls under medicines law regardless of its chemistry.
Is Matrixyl banned in sport?
It is not named on the 2026 Prohibited List, and it does not appear in the annex to German anti-doping law. A face cream holding a few parts per million of it triggers nothing. The catch-all class for unapproved drug substances covers pharmacological substances, which a cosmetic applied to the skin is not.
Why does this page list Matrixyl concentrations?
Because the results cannot be read without them, and because the scale is the point. Every figure is a concentration in a cream, a culture dish or an animal experiment, measured in millionths by weight rather than milligrams. None of them is an injection dose, and none of them is advice.
Sources
- Choi YL, Park EJ, Kim E, Na DH, Shin YH (2014). Dermal stability and in vitro skin permeation of collagen pentapeptides (KTTKS and palmitoyl-KTTKS). Biomolecules & Therapeutics 22(4):321-7. PMID 25143811. PMCID PMC4131521. DOI 10.4062/biomolther.2014.053. Excised skin in diffusion cells, 24 hours, LC-MS/MS; layer-by-layer recoveries and enzymatic stability.
- Robinson LR, Fitzgerald NC, Doughty DG, Dawes NC, Berge CA, Bissett DL (2005). Topical palmitoyl pentapeptide provides improvement in photoaged human facial skin. International Journal of Cosmetic Science 27(3):155-60. PMID 18492182. DOI 10.1111/j.1467-2494.2005.00261.x. Double-blind, vehicle-controlled split-face study, n = 93, 3 ppm, 12 weeks; all six authors at Procter & Gamble, Miami Valley Laboratories, Cincinnati.
- ClinicalTrials.gov API v2, queries for "palmitoyl pentapeptide-4", "Matrixyl", "KTTKS", "pal-KTTKS" and "palmitoyl pentapeptide", run 20 September 2026: total count 0 for each. https://clinicaltrials.gov/api/v2/studies
- Cosmetic Ingredient Review (2023). Safety assessment of Myristoyl Pentapeptide-4, Palmitoyl Pentapeptide-4, and Pentapeptide-4 as used in cosmetics. Draft report for panel review, released 18 August 2023, panel meeting 11-12 September 2023; full PDF including raw-material dossiers and, as historical annexes, the 2013 and 2014 meeting transcripts of the earlier Palmitoyl Oligopeptides review from which this ingredient was regrouped. Source of the use frequencies, the concentrations, the toxicology and the record that no reproductive, developmental, cancer or repeat-dose data exist. The draft itself states no conclusion; the transcripts of 2013 and 2014 concern the predecessor grouping, not this report. https://www.cir-safety.org/sites/default/files/Pentapeptides.pdf
- openFDA, Drugs@FDA interface, query
openfda.substance_name:"PALMITOYL PENTAPEPTIDE-4", run 20 September 2026: NOT_FOUND. https://api.fda.gov/drug/drugsfda.json - European Medicines Agency, complete register of centrally authorised human and veterinary medicines (
medicines-output-medicines-report_en.xlsx), downloaded 20 September 2026: no hits for pentapeptide, KTTKS or Matrixyl; control queries for somatropin and tesamorelin returned entries. https://www.ema.europa.eu/documents/report/medicines-output-medicines-report_en.xlsx - Katayama K, Armendariz-Borunda J, Raghow R, Kang AH, Seyer JM (1993). A pentapeptide from type I procollagen promotes extracellular matrix production. Journal of Biological Chemistry 268(14):9941-4. PMID 8486721. Identifies KTTKS, residues 212 to 216, as the minimum active sequence and reports the 1991 subfragment work.
- Croda Beauty, Matrixyl product page, retrieved 20 September 2026. Ingredient declaration of the raw-material solution and the marketing claims about six-month testing, vitamin C and retinol; no peer-reviewed publication of those comparative data was found. https://www.crodabeauty.com/en-gb/products/product/5598-matrixyl
- COSMILE Europe, ingredient entry PALMITOYL PENTAPEPTIDE-4, retrieved 20 September 2026. Declared function "skin conditioning: maintains the skin in good condition", based on the official European ingredient database, which was not machine-readable on that date. https://cosmileeurope.eu/inci/detail/10184/palmitoyl-pentapeptide-4/
- FDA Global Substance Registration System, substance record PALMITOYL PENTAPEPTIDE-4, UNII KK181SM5JG, status approved, retrieved 20 September 2026 through the GSRS API. Synonyms, codes, molecular formula and structure. https://gsrs.ncats.nih.gov/api/v1/substances(KK181SM5JG)
- Wikidata item Q18386276, queried through the SPARQL endpoint on 20 September 2026: formula C39H75N7O10, CAS 214047-00-4, PubChem CID 9897237, InChIKey WSGCRSMLXFHGRM-DEVHWETNSA-N. PubChem itself returned server-busy errors throughout that day.
- Jones RR, Castelletto V, Connon CJ, Hamley IW (2013). Collagen stimulating effect of peptide amphiphile C16-KTTKS on human fibroblasts. Molecular Pharmaceutics 10(3):1063-9. PMID 23320752. DOI 10.1021/mp300549d.
- Tsai WC, Hsu CC, Chung CY, Lin MS, Li SL, Pang JH (2007). The pentapeptide KTTKS promoting the expressions of type I collagen and transforming growth factor-beta of tendon cells. Journal of Orthopaedic Research 25(12):1629-34. PMID 17593541. DOI 10.1002/jor.20455.
- Park H, An E, Cho Lee AR (2017). Effect of palmitoyl-pentapeptide (Pal-KTTKS) on wound contractile process in relation with connective tissue growth factor and alpha-smooth muscle actin expression. Tissue Engineering and Regenerative Medicine 14(1):73-80. PMID 30603464. PMCID PMC6171572. DOI 10.1007/s13770-016-0017-y.
- Abu Samah NH, Heard CM (2011). Topically applied KTTKS: a review. International Journal of Cosmetic Science 33(6):483-90. PMID 21535443. DOI 10.1111/j.1468-2494.2011.00657.x. Notes the absence of skin penetration data and the small number of clinical studies.
- Mortazavi SM, Moghimi HR (2022). Skin permeability, a dismissed necessity for anti-wrinkle peptide performance. International Journal of Cosmetic Science 44(2):232-248. PMID 35302659. DOI 10.1111/ics.12770.
- Bjerke DL, Li J, Gao Y, Hu P, Lintner K, Hakozaki T (2026). A framework for the safety evaluation of peptides in cosmetics. Current Research in Toxicology 10:100291. PMID 41953401. PMCID PMC13054063. DOI 10.1016/j.crtox.2026.100291. Declares a Procter & Gamble interest.
- PubMed E-utilities, search term
KTTKS OR "palmitoyl pentapeptide", run 20 September 2026: 44 records. https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi - Europe PMC REST API, search term
"pal-KTTKS" OR "palmitoyl pentapeptide-4" OR "Matrixyl", run 20 September 2026: 131 records. https://www.ebi.ac.uk/europepmc/webservices/rest/search - Aruan RR, Hutabarat H, Widodo AA, Firdiyono MTCC, Wirawanty C, Fransiska L (2023). Double-blind, randomized trial on the effectiveness of acetylhexapeptide-3 cream and palmitoyl pentapeptide-4 cream for crow's feet. Journal of Clinical and Aesthetic Dermatology 16(2):37-43. PMID 36909866. PMCID PMC10005804.
- Kaczvinsky JR, Griffiths CE, Schnicker MS, Li J (2009). Efficacy of anti-aging products for periorbital wrinkles as measured by 3-D imaging. Journal of Cosmetic Dermatology 8(3):228-33. PMID 19735523. DOI 10.1111/j.1473-2165.2009.00444.x.
- Mohammed YH, Yamada M, Lin LL, Grice JE, Roberts MS, Raphael AP, Benson HA, Prow TW (2014). Microneedle enhanced delivery of cosmeceutically relevant peptides in human skin. PLoS One 9(7):e101956. PMID 25033398. PMCID PMC4102460. DOI 10.1371/journal.pone.0101956.
- de Mello LR, Castelletto V, Seitsonen J, Hamley IW (2026). Nanostructure and collagen-stimulating activity of cationic pentapeptide lipopeptides. Journal of Peptide Science 32(8):e70111. PMID 42317129. PMCID PMC13280649. DOI 10.1002/psc.70111.
- Pelin JNBD, de Mello LR, Siqueira RAGB, de Souza Alves M, Castelletto V, Almeida JF, Seitsonen J, Lopes PS, Leite-Silva V, Andreo-Filho N, Hamley IW (2026). Novel palmitoyl pentapeptides: self-assembly, collagen stimulation, antioxidant activity, and influence of skin microbiota health. ACS Omega 11(13):20691-20702. PMID 41970822. PMCID PMC13063098. DOI 10.1021/acsomega.5c12526.
- Kachooeian M, Mousivand Z, Sharifikolouei E, Shirangi M, Firoozpour L, Raoufi M, Sharifzadeh M (2022). Matrixyl patch vs Matrixyl cream: a comparative in vivo investigation of Matrixyl effect on wound healing. ACS Omega 7(28):24695-24704. PMID 35874243. PMCID PMC9301720. DOI 10.1021/acsomega.2c02592.
- Rasouli M, Shahghasempour L, Shirbaghaee Z, Hosseinzadeh S, Abbaszadeh HA, Fattahi R, Ranjbari J, Soleimani M (2024). Mesenchymal stem cell therapy using Pal-KTTKS-enriched carboxylated cellulose improves burn wound in rat model. Archives of Dermatological Research 316(7):353. PMID 38850353. DOI 10.1007/s00403-024-03082-1.
- Tałałaj U, Uścinowicz P, Bruzgo I, Surażyński A, Zaręba I, Markowska A (2019). The effects of a novel series of KTTKS analogues on cytotoxicity and proteolytic activity. Molecules 24(20):3698. PMID 31618846. PMCID PMC6832239. DOI 10.3390/molecules24203698.
- U.S. Food and Drug Administration, warning letters full-text search for "pentapeptide", "Matrixyl" and "KTTKS", run 20 September 2026: no results for any term. https://www.fda.gov/inspections-compliance-enforcement-and-criminal-investigations/compliance-actions-and-activities/warning-letters
- openFDA, enforcement interface, query
pentapeptide, run 20 September 2026: NOT_FOUND. https://api.fda.gov/drug/enforcement.json - Chirita RI, Chaimbault P, Archambault JC, Robert I, Elfakir C (2009). Development of a LC-MS/MS method to monitor palmitoyl peptides content in anti-wrinkle cosmetics. Analytica Chimica Acta 641(1-2):95-100. PMID 19393372. DOI 10.1016/j.aca.2009.03.015.
- Regulation (EC) No 1223/2009 on cosmetic products, consolidated text of 1 January 2025, downloaded from EUR-Lex on 20 September 2026 and searched in full (2.16 million characters): the word pentapeptide does not occur, so the substance appears in none of annexes II to VI. https://eur-lex.europa.eu/eli/reg/2009/1223/2025-01-01
- electronic Medicines Compendium (United Kingdom), search for "pentapeptide", run 20 September 2026: "No search results for pentapeptide". The MHRA product database itself was not machine-readable that day. https://www.medicines.org.uk/emc/search?q=pentapeptide
- Therapeutic Goods (Poisons Standard — June 2026) Instrument 2026, F2026L00633, Australian Federal Register of Legislation, full text downloaded and searched 20 September 2026: no hits for pentapeptide, KTTKS, Matrixyl, tripeptide or tetrapeptide; control entries for GHRP-6, pralmorelin and sh-Oligopeptide-1 present; collagen scheduled only in preparations for injection or implantation. https://www.legislation.gov.au/F2026L00633/asmade
- Health Canada, Drug Product Database API, active-ingredient queries for "palmitoyl" and "pentapeptide", run 20 September 2026: empty lists; control query for somatropin returned entries. https://health-products.canada.ca/api/drug/activeingredient/
- Swissmedic, list of authorised medicines (Zugelassene Arzneimittel HAM, XLSX), downloaded 20 September 2026: no hits for pentapeptid, palmitoyl, KTTKS or Matrixyl; control queries Genotropin and Insulin positive. The list carries product names rather than active-substance names throughout. https://www.swissmedic.ch
- U.S. Food and Drug Administration, certain bulk drug substances for use in compounding that may present significant safety risks; page content current as of 22 April 2026, retrieved 20 September 2026. No entry for pentapeptide, KTTKS or Matrixyl; GHK-Cu, BPC-157 and TB-500 appear among withdrawn nominations. https://www.fda.gov/drugs/human-drug-compounding/certain-bulk-drug-substances-use-compounding-may-present-significant-safety-risks
- DailyMed Web Services v2 (National Library of Medicine),
drugnames.jsonandspls.jsonfor PALMITOYL PENTAPEPTIDE-4, retrieved 20 September 2026: three label entries, all self-registrations without an approval number. https://dailymed.nlm.nih.gov/dailymed/services/v2/spls.json - openFDA, drug label interface, query
openfda.substance_name.exact:"PALMITOYL PENTAPEPTIDE-4", run 20 September 2026: one hit, a stretch-marks cream registered as a human over-the-counter drug with the purpose "repair, firming, anti-wrinkle". https://api.fda.gov/drug/label.json - World Anti-Doping Agency, the 2026 Prohibited List, full text searched 20 September 2026: no occurrence of pentapeptide, palmitoyl, KTTKS, Matrixyl, tripeptide, tetrapeptide or collagen; classes S0 and S2.3 read in full.
- Annex to the German Anti-Doping Act (AntiDopG), gesetze-im-internet.de, retrieved 20 September 2026: no mention of pentapeptides, palmitoyl compounds, KTTKS, Matrixyl or collagen peptides. https://www.gesetze-im-internet.de/antidopg/anlage.html
- Cosmetic Ingredient Review (2014). Safety assessment of Tripeptide-1, Hexapeptide-12, Palmitoyl Tripeptide-1 and Palmitoyl Tetrapeptide-7 as used in cosmetics; the panel record specifies the combination as 100 ppm tripeptide-1 and 50 ppm tetrapeptide-7. Cited here through the 2023 draft report and its meeting material, source [4]; the review site's own ingredient search was not machine-readable on 20 September 2026.
- ClinicalTrials.gov, registry record NCT06125730, a single-group study of an eyelash serum, Dime Beauty Co., n = 35, completed 10 February 2024; full registry record retrieved 20 September 2026. The declared formula contains Myristoyl Pentapeptide-4, not Palmitoyl Pentapeptide-4. https://clinicaltrials.gov/api/v2/studies/NCT06125730
Cite this page
The facts on this page were checked on 20 September 2026, and every register search behind the status table was run on that date. The part most likely to change is the American expert review, which has been open since September 2023 and could still be concluded.
myPeptides Research & Editing. (2026). Matrixyl: what the studies show, status and safety. Version 1.0, 20 September 2026. myPeptides Peptide Register. Retrieved from https://mypep.app/peptides/matrixyl
How pages in this register are compiled and graded is described under methodology; the full register is at peptides.
| Version | Date | Change |
|---|---|---|
| 1.0 | 2026-09-20 | Initial publication |
Last verified: 20 September 2026. Next review: on any conclusion of the open safety assessment, on any independent replication of the 2005 trial, or on any change in cosmetic or medicines listing.