Peptides for Stretch Marks: 76 Randomized Trials Exist, and Not One Tests a Peptide

Stretch marks are one of the few skin complaints where the published record is genuinely large. There are 2,671 records in PubMed with striae or stretch marks in the title or abstract, and 76 of them are randomized controlled trials. That is eight times the randomized literature behind topical peptides for wrinkles.

None of the 76 tests a peptide.

What the 76 trials are about

Searched on 2026-09-15, title and abstract terms, publication type "randomized controlled trial". Anyone can re-run these counts.

The shape of that table is the finding. Stretch marks have been studied as a dermal scarring problem, and the things tested against dermal scars are the things that reach the dermis: light, needles, radiofrequency energy, injected plasma. The one topical with a real trial count is tretinoin, a prescription retinoid.

A cosmetic peptide serum is none of those, and the literature has never asked it to be.

The 21 records that do join peptides to stretch marks

Adding peptide terms to the same search returns 21 records and no trials. Reading them is more informative than counting them, and this set is an unusually clean demonstration of why.

Fourteen of the 21 are brain anatomy. They are neuroscience papers from 1980 to 2004 — galanin mapping in the rat central nervous system, cholecystokinin receptor binding in the human brain, neuropeptide Y in the chicken hypothalamus. Striae in that literature is an anatomical term for a band of nerve fibres. The word matches; the subject does not.

Three are endocrinology, and they invert the premise. In Cushing's syndrome and ectopic ACTH syndrome, violaceous stretch marks are a presenting sign of too much adrenocorticotropic hormone. A 2025 case report describes a 24-year-old woman whose appendiceal neuroendocrine tumour secreted ACTH, presenting with rapid weight gain, violaceous striae, fatigue and oedema, and urinary free cortisol more than ten times the upper limit of normal. ACTH is a peptide. In the only substantial literature connecting peptides and striae, the peptide causes them.

This is the same pattern we found in the rosacea literature, where "peptide" names something the body makes that is implicated in the disease rather than an ingredient in a bottle. It is worth stating as a general caution: a record count for ingredient + condition is not an evidence count until someone has read what the records are about.

The one cosmetic-peptide study

McGuinn and colleagues at Thomas Jefferson University published a pilot study in Skinmed in 2020: a peri-procedural tripeptide/hexapeptide serum used alongside 1540 nm non-ablative fractional laser in 10 patients with striae distensae. Their own framing is explicit — non-ablative fractional lasers have shown improvements, topical peptides have been used in rejuvenation, and no study had examined the combination for striae.

What the abstract reports: all patients reported subjective improvement; adding the serum increased objective improvement, reduced post-inflammatory hyperpigmentation and increased satisfaction. What it does not report: a number, a control arm, or a separation between what the laser did and what the serum did.

Ten patients, no control, no effect size, and the peptide applied around a procedure rather than as a treatment. That is the strongest thing anyone can cite for peptides on stretch marks, and it is fair to say so plainly rather than to omit it.

The nearest real peptide result is insulin

If the question is whether any peptide has been put on striae and measured, the answer is yes, and it is not a cosmetic one.

Ebrahim and colleagues at Al-Azhar University treated 30 patients with abdominal striae distensae using microneedling on both sides, followed by topical hyaluronic acid on the right and topical insulin on the left. Insulin is a peptide hormone of 51 amino acids. Their reasoning was that it may help restore the integrity of damaged skin and promote granulation tissue.

The authors' conclusion is that either combination can be effective and safe. Two things follow for anyone reading a peptide serum's label. The delivery route was through microneedling channels, not across an intact barrier. And the comparator was hyaluronic acid delivered the same way, with results close enough that the study separates neither ingredient from the needling — a problem we set out in full in our microneedling entry.

A 2026 study is worth naming beside it for the same reason: 20 postpartum women applied a recombinant human collagen serum to one side of the abdomen for 4 weeks, with the untreated side as the control, and the treated side improved more on elasticity, striae area and roughness. Recombinant collagen is a manufactured protein, far larger than any peptide in a cosmetic, and the study was self-controlled and open rather than blinded. It is a result about a protein serum, and it is the kind of design a peptide has never been given here.

What a stretch-mark peptide product can honestly claim

That it may hydrate and improve the feel of the skin, which any decent emollient does. That the ingredient class has trial support on other endpoints — wrinkle depth, mostly — and that this endpoint is not one of them.

What it cannot claim is a trial, because there is not one. Striae are scars rather than photoageing, and the endpoint that the peptide trials on our skin page actually measured is wrinkle depth.

The gap is also unusually cheap to close. Most people with striae have them on both sides of the body, which is why two of the better studies in this area used one side as its own control. A named peptide, a disclosed concentration, the identical vehicle on the other side, twelve weeks, instrumental measurement: it is an obvious study, and the reason to notice it has not been run is that the cosmetic industry's peptide claims do not currently depend on it.

Sources

• McGuinn KP, Ross NA, Wang JV, Saedi N. Combination tripeptide/hexapeptide serum with 1540 nm nonablative fractional laser for the treatment of striae distensae: a pilot study. Skinmed 2020;18(6):337–41. PMID 33397562.
• Ebrahim MRY, Darwish HMA, Galal SA. Efficacy of dermapen combined with topical insulin versus dermapen combined with topical hyaluronic acid in treatment of stretch marks: comparative study. Arch Dermatol Res 2024;317(1):54. PMID 39601879.
• Li C, Hou J, Yao Y, Zhou L, Chen J. Efficacy of recombinant human collagen serum for improving skin dryness and roughness in striae gravidarum: a clinical study. Afr J Reprod Health 2026;30(13):86–100. PMID 42454417.
• Desai R, Sidra FNU, Jia L, et al. Episodic severe ectopic adrenocorticotropic hormone syndrome by metastatic appendiceal neuroendocrine tumor. JCEM Case Rep 2025;3(9):luaf171. PMID 40757053.
• PubMed counts, run 2026-09-15 via the NCBI E-utilities esearch endpoint: (striae[tiab] OR "stretch marks"[tiab] OR "stretch mark"[tiab]) — 2,671 records; the same with AND "randomized controlled trial"[pt] — 76; the same with AND peptide*[tiab] — 21 records and 0 trials. Intervention counts add the intervention term in [tiab] to the trial query. PubMed.