Are peptides good for eczema?

Two randomized trials have applied a peptide to eczematous skin and neither produced a difference against its control. A phase 2 trial of the antimicrobial peptide omiganan cut Staphylococcus aureus by 93.5% against vehicle and reported no significant clinical improvement. A moisturizer containing a dipeptide improved SCORAD, hydration, water loss and itch against its own baseline, but its published results state that none of those changes differed significantly from the control group. Nothing in that record supports a claim that a peptide product treats eczema.

Why do so many eczema papers mention peptides?

Because in this literature a peptide is usually something the skin makes. Five hundred and twenty-nine records join peptides to atopic dermatitis; of the eight randomized among them, six measure the body's own antimicrobial peptides — cathelicidin, beta-defensins, dermcidin, intestinal defensins — as an outcome of coal tar, vitamin D, a synbiotic, a herbal formulation or, in one case, a comedy film. The peptide is the readout, not the treatment.

Is the peptide level in eczema skin high or low?

Low, relative to other inflamed skin. The 2002 New England Journal of Medicine study that framed the question compared inflamed skin in atopic dermatitis with inflamed skin in psoriasis and found LL-37 and human beta-defensin 2 expressed at lower levels in the atopic samples. That deficiency is the reason antimicrobial peptides became an eczema research subject at all, and it is the opposite of what the rosacea literature describes, where the same class of peptide is overexpressed and implicated in causing the inflammation.

What does the omiganan trial actually prove?

That the bacterial half of the theory can be delivered on without the clinical half following. The trial reduced Staphylococcus abundance, increased microbial diversity and cut cultured S. aureus by more than ninety per cent at the highest concentration, all against vehicle, in eighty randomized patients over 28 days. Symptoms did not improve. The authors' own conclusion is that a treatment selectively targeting the microbiome does not appear to be a successful strategy in mild-to-moderate disease.

Can a cosmetic peptide serum be used on eczema-prone skin?

That question has not been studied. The one cosmetic-style formulation with a randomized trial is the dipeptide-12 moisturizer described above, and what it demonstrates is a moisturizer effect present in both arms. No trial has tested whether the palmitoyl peptides, copper peptides or acetyl hexapeptides sold in serums irritate, soothe or do nothing on eczematous skin. An absence of evidence is not a safety finding in either direction, and questions about treating a diagnosed skin disease belong with a clinician.

What is dermcidin, and why is there a Charlie Chaplin study?

Dermcidin is an antimicrobial peptide secreted by sweat glands, and it is present at reduced levels in people with atopic eczema. The 2007 study tested whether a positive emotional state changed that: twenty patients with atopic eczema watched an 87-minute humorous film, with sweat sampled before and afterwards. It is included in the count here because it carries the randomized-trial publication type and joins the two search terms — which is a useful demonstration of what a raw search total is worth before the records are read one by one.

What does have strong evidence in eczema?

The 1,499 randomized trials on atopic dermatitis are dominated by emollients, topical corticosteroids, calcineurin inhibitors and the newer biologic and JAK-directed drugs. The largest molecular literature is filaggrin, a barrier protein, with 1,151 records against 75 for cathelicidin. The point of counting these side by side is proportion: eczema is one of the most heavily trialled skin conditions there is, and the peptide contribution to it is two applications and no difference.

Peptides and Eczema: Eight Randomized Trials, and Only Two Put a Peptide on Skin

Atopic dermatitis has 1,499 randomized trials. Eight join it to peptides, and six of those measure the peptides the body makes rather than giving one. The two that applied a peptide both failed their between-group comparison.

Search for peptides and eczema and you will find serums promising barrier repair. Search PubMed and you find a large, mature literature in which peptides appear constantly — almost always as something being measured, and almost never as something being applied.

Counted on 2026-09-19: 1,499 randomized controlled trials on atopic dermatitis. 529 records join atopic dermatitis or eczema to peptides. Eight of those carry the randomized-trial publication type. We read all eight.

What the eight randomized records actually did

Study	Year	Was a peptide given?	What it tested
Omiganan phase 2 (PMID 33010325)	2022	Yes, topically	Synthetic antimicrobial peptide, 3 concentrations vs vehicle, 80 patients
PTPD-12 moisturizer (PMID 30427231)	2019	Yes, in a cream	Autophagy-enhancing dipeptide vs control moisturizer, 43 patients
Coal tar and AHR signalling (PMID 31344386)	2020	No	Coal tar; antimicrobial peptide induction was the readout
Vitamin D supplementation (PMID 23638978)	2014	No	Oral vitamin D; antimicrobial peptide production was the readout
Infant synbiotic cohort (PMID 22093109)	2012	No	Faecal defensin levels in infancy against later sensitisation
Pentaherbs formulation (PMID 18341655)	2008	No	Five-herb concoction; immunomodulatory markers
Humorous film and dermcidin (PMID 17188121)	2007	No	An 87-minute Charlie Chaplin film; sweat peptide measured
Lactation and neuropeptides (PMID 14707470)	2003	No	Suckling; plasma neuropeptide and neurotrophin levels

Six of the eight give something that is not a peptide and then measure a peptide. That is the shape of this entire field, and it is why a count of 529 records reads as a rich evidence base until the records are opened.

The two that applied a peptide, read closely

Omiganan is the serious one, and it separated the bacteria from the disease. Omiganan is a synthetic analogue of indolicidin, an antimicrobial peptide, with activity against Staphylococcus aureus — the organism whose overgrowth is the standing explanation for why eczema flares. Eighty patients with mild-to-moderate atopic dermatitis were randomized to 1%, 1.75% or 2.5% omiganan or vehicle, applied twice daily to all lesions for 28 days, with weekly clinical, microbiological and pharmacodynamic assessment of a target lesion.

The microbiology worked. Staphylococcus abundance fell, diversity rose, and cultured S. aureus dropped 93.5% against vehicle at the highest concentration (95% CI −99.2% to −28.5%, P = .02). The clinical scores did not move. The authors' conclusion is unusually direct: a treatment that selectively targets the microbiome "does not appear to be a successful treatment strategy" in this population.

That is a more useful result than a success would have been, because it tests the mechanism that peptide marketing borrows. If clearing the bacteria that provoke eczema does not improve eczema over four weeks, a claim resting on antimicrobial activity has lost its footing before the cosmetic ingredient is reached.

The second trial is the one closest to a beauty product, and its conclusion does not follow from its results. Pentasodium tetracarboxymethyl palmitoyl dipeptide-12 — a palmitoyl dipeptide of exactly the kind found in cosmetic formulas — was put into a moisturizer and tested double-blind against a control moisturizer in 43 patients with mild-to-moderate disease, assessed at baseline, week 2 and week 4 on SCORAD, corneometry, transepidermal water loss, a pruritus scale and investigator's global assessment.

The peptide group improved on SCORAD, hydration, water loss and itch against its own baseline. The control group improved on SCORAD and hydration against its own baseline. Then comes the sentence that settles it, in the paper's own results:

The mean changes in the SCORAD index score, skin hydration, TEWL, pruritus, and number of patients with improvement in IGA were not statistically different between the two groups.

The published conclusion nonetheless describes the moisturizer as "a good therapeutic option". Both groups got better because both groups were moisturised, which is what emollients do in eczema and why they are the background therapy in almost every trial in the field. A within-group improvement in a two-arm trial is not a finding about the ingredient. Anyone quoting this study as peptide evidence is quoting the paragraph above its own comparison.

The direction of the peptide is opposite in the two conditions we have counted

This is the part worth carrying away, and it emerged from setting two of our own censuses side by side.

In atopic dermatitis, the antimicrobial peptides are deficient. The 2002 New England Journal of Medicine study that opened the subject compared inflamed skin from patients with atopic dermatitis against inflamed skin from patients with psoriasis and normal skin, by immunohistochemistry, immunodot blot and quantitative PCR, and found LL-37 and human beta-defensin 2 expressed at lower levels in the atopic samples. The proposed consequence is the one clinicians see: skin that gets infected.

In rosacea, the same molecule is excessive. As set out in our count of the rosacea literature, people with rosacea express abnormally high cathelicidin in facial skin, cut into different fragments by a more active stratum corneum enzyme, and injecting those fragments into mouse skin increases inflammation.

Same peptide family, two inflammatory diseases of facial and flexural skin, opposite abnormalities. A serum cannot be correcting both. It is a useful corrective to ingredient copy that treats "supports the skin's antimicrobial peptides" as a direction-free good.

What the eczema literature is actually about

Scale, honestly reported: filaggrin appears with atopic dermatitis in 1,151 records; cathelicidin in 75. Filaggrin is a structural protein of the outer epidermis, and loss-of-function variants in its gene are the best-established genetic risk factor for the disease. The dominant molecular explanation of eczema is a broken barrier, not a missing peptide — and the treatments with the largest randomized evidence base are emollients, topical anti-inflammatories and, more recently, targeted biologics.

What would change this page

A randomized, controlled, between-group comparison of a cosmetic peptide on eczematous skin, reporting tolerability as well as efficacy, would change it. None exists. Until then the honest summary is that the peptide-and-eczema literature contains one well-conducted negative antimicrobial trial, one moisturizer trial that did not separate from its control, and six studies in which the peptide was the measurement rather than the medicine — including one where the intervention was a Charlie Chaplin film.

Readers comparing formats rather than conditions may want our entries on what a peptide moisturizer's trials measure and on how long the published clocks actually run.

Sources

Niemeyer-van der Kolk T, et al. Topical antimicrobial peptide omiganan recovers cutaneous dysbiosis but does not improve clinical symptoms in patients with mild to moderate atopic dermatitis in a phase 2 randomized controlled trial. J Am Acad Dermatol. PMID 33010325.
Kwon SH, et al. The effect of autophagy-enhancing peptide in moisturizer on atopic dermatitis: a randomized controlled trial. J Dermatolog Treat. 2019. PMID 30427231.
Ong PY, et al. Endogenous antimicrobial peptides and skin infections in atopic dermatitis. N Engl J Med. 2002. PMID 12374875.
Yamasaki K, et al. Increased serine protease activity and cathelicidin promotes skin inflammation in rosacea. Nat Med. 2007. PMID 17676051.
Kimata H. Increase in dermcidin-derived peptides in sweat of patients with atopic eczema caused by a humorous video. J Psychosom Res. 2007. PMID 17188121.
Smits JPH, et al. Targeting the cutaneous microbiota in atopic dermatitis by coal tar via AHR-dependent induction of antimicrobial peptides. J Invest Dermatol. 2020. PMID 31344386.
Hata TR, et al. A randomized controlled double-blind investigation of the effects of vitamin D dietary supplementation in subjects with atopic dermatitis. J Eur Acad Dermatol Venereol. 2014. PMID 23638978.
Record counts: NCBI PubMed E-utilities, searched 2026-09-19, title/abstract fields with the randomized controlled trial publication type where stated.