LL-37: research, side effects and dosage in studies

In short

LL-37 is the human body's own antimicrobial peptide, made in the skin, airways and gut. Given from outside it has been studied mainly for hard-to-heal leg ulcers, and the largest trial showed no benefit [2]A. It has no marketing authorisation as a medicine.

• The first human study (34 participants) looked strong: at the lowest concentration the healing rate was about six times that of placebo [1]A.
• A phase 2b trial (148 participants) found no difference from placebo in the whole study population [2]A.
• Most human studies of LL-37 do not give the peptide; they measure the body's own level as a marker, often in vitamin D studies.
• In cell studies LL-37 is toxic to human cells at 1-10 micromolar [4]D. That is why the dose is not a simple matter.
• There is no approved dosage, and no LL-37 product has a marketing authorisation.

What is LL-37?

LL-37 is the only human antimicrobial peptide of the cathelicidin family, and it is made of 37 amino acids [5].

The body does not make LL-37 as such but as a precursor called hCAP-18 or CAP-18, a protein 170 amino acids long. LL-37 and several shorter peptides, such as LL-23, KR-20 and RK-31, are cut from it when needed [5]. The gene is called CAMP.

LL-37 is produced widely in the body: in the skin, airways, gut, testes and immune cells [6]D. Skin peptides more generally have their own page: peptides in skincare. LL-37 is part of innate immunity, the body's first line of defence, which works without first having to learn to recognise a pathogen.

This is an important difference from the other substances on this site. LL-37 is not foreign to the body; every person has it all the time. So the question is not whether LL-37 acts in the body, but what happens when more of it is given from outside. The two are easily confused, and this page keeps them apart.

LL-37 is not a steroid, a hormone or anything like growth hormone. An example of another research-stage peptide with evidence at a completely different level is BPC-157. Basics: What are peptides?. All substances: Peptides A-Z.

Evidence at a glance

Each cell gives the kind of study and its source. The letter is the evidence level, not the result.

"-" = no such study is known.

What is LL-37 studied for?

Given from outside, LL-37 is studied mainly for wound care; as the body's own peptide it is measured as a marker in many diseases. None of these is an approved use.

Studies of LL-37 given from outside focus on three things: hard-to-heal venous leg ulcers [1]A [2]A, diabetic foot ulcers [3]A and one trial in COVID-19 where the peptide was given by mouth [7]A. The registry also holds a small melanoma study in which LL-37 was injected into tumours [15]E, not peer-reviewed.

The body's own LL-37 level has been measured far more widely. Vitamin D studies, for example, have followed whether the body's own LL-37 level rises [8]B. As a marker it is measured in tuberculosis, lung diseases, periodontitis, atopic dermatitis, rosacea and prosthetic joint infections, among others.

Keep this distinction in mind when you meet claims online about the benefits of LL-37. Much of the research shows that a disease changes the LL-37 level, not that giving LL-37 treats the disease.

How does LL-37 work?

LL-37 works in two ways: it kills microbes directly, and it regulates the immune response.

The direct effect comes from its structure. LL-37 is an amphipathic helical peptide: one side seeks water and the other fat. It attaches to the bacterial membrane and forms pores in it, so the cell leaks and dies. Under physiological conditions the peptide also forms multimers, which affects how it binds to surfaces [6]D. In cells it has been shown to act against both gram-positive and gram-negative bacteria and some viruses [4]D.

Regulating the immune response is more complex. LL-37 binds and inactivates bacterial endotoxins, draws immune cells to the site of inflammation and mediates calcium entry into cells through membrane receptors [4]D. According to reviews it also promotes blood vessel formation and wound closure [6]D. Another substance on this site linked to wound healing and the skin is GHK-Cu.

An essential and often overlooked point: LL-37 has both pro-inflammatory and anti-inflammatory effects [4]D. This sets it apart from peptides studied as one-way calming agents for inflammation, such as KPV. The same review notes that very high concentrations, about 300 micromolar, have been measured in psoriasis lesions, and about 1 micromolar in gum fluid in periodontitis [4].

All of the mechanism above rests on cell and laboratory studies. In humans, outcomes such as a change in wound size have been shown, not the mechanism itself.

Has LL-37 been studied in humans?

Yes, but the human studies fall into two completely different groups: studies where LL-37 was given as a treatment, and a much larger body where the body's own LL-37 was measured.

LL-37 given as a treatment

Four peer-reviewed treatment studies in which LL-37 was given to humans are known. There is also one case report and one registered injection trial.

• First human study, leg ulcers (2014). 34 participants with a hard-to-heal venous leg ulcer. A three-week open placebo run-in was followed by a four-week randomised double-blind phase in which LL-37 was applied twice a week at 0.5, 1.6 or 3.2 milligrams per millilitre, and then four weeks of follow-up. The healing rate constant was about six times that of placebo at 0.5 and about three times at 1.6 (p=0.003 and p=0.088). Mean ulcer area shrank by 68% at the lowest and 50% at the middle concentration. The highest concentration, 3.2, did not differ from placebo. No local or systemic safety concerns were seen [1]A.
• Phase 2b trial HEAL LL-37, leg ulcers (2021). A multicentre trial with 148 participants, double-blind and placebo-controlled, with three arms: LL-37 at 0.5 and 1.6 milligrams per millilitre and placebo, all on top of compression therapy. Mean age was 67.6 years, median ulcer duration 20.3 months and mean wound size 11.6 square centimetres. In the whole study population, healing did not improve significantly compared with placebo. A post hoc subgroup analysis found a significant benefit in people whose wound was at least 10 square centimetres at randomisation. The product was well tolerated at both concentrations. The authors themselves say the finding needs a new, adequately sized trial [2]A.
• Diabetic foot ulcer (2023). A randomised double-blind trial in Jakarta: LL-37 cream or placebo cream twice a week for four weeks on ulcers with mild infection. According to the registry there were 40 participants [10]E, not peer-reviewed. Formation of granulation tissue was better in the LL-37 group on every measurement day, 7, 14, 21 and 28 (p=0.031, 0.009, 0.006 and 0.037). The inflammatory mediators IL-1 alpha and TNF alpha did not fall, and the number of bacteria in the wound did not fall significantly [3]A. So the wound healed faster, but the mechanisms by which it was expected to heal did not show in the measurements.
• LL-37 by mouth in COVID-19 (2023). 238 hospitalised adults, randomised and placebo-controlled but open-label, so participants and staff knew who received what. The product was LL-37 made with recombinant DNA technology in the bacterium Lactococcus lactis. Time to viral clearance was 9.80 days (standard deviation 2.67) in the LL-37 group and 14.04 days (5.89) in the placebo group, p<0.01. No serious adverse events were seen [7]A. The open-label design and a single study centre are important limitations.
• LL-37 injected into melanoma. In a US phase 1-2 trial LL-37 was injected into skin melanoma metastases once a week for up to eight weeks. According to the registry four patients took part, and the trial ended in 2020 [15]E, not peer-reviewed. A case report on skin harms in one patient has been published, see What are the side effects of LL-37? [14]B. No peer-reviewed results paper from the trial is known.

The last finding of the first study matters: the effect did not grow with concentration but disappeared at the highest one [1]A. That fits the cell finding that LL-37 is toxic to cells at high enough concentrations [4]D.

The body's own LL-37 as a marker

Most of the literature belongs to this group, and it says nothing about the benefits of giving LL-37.

An example of the design: in an open trial, healthy volunteers were given phenylbutyrate and vitamin D, and the researchers measured whether LL-37 production in their own macrophages went up. It did, and 500 milligrams of phenylbutyrate twice a day together with 5000 units of vitamin D worked best [8]B. No LL-37 was given here; the aim was to make the body produce more of it.

Beyond this, LL-37 levels have been measured in dozens of studies on vitamin D, probiotics, exercise and various diseases. These are observational or mechanistic studies, not treatment trials.

Promising, but only in cell studies

This finding comes from cells and laboratory work and has not been shown in humans.

• Biofilms, fungi and viruses no human studies

  In the laboratory LL-37 also acts against biofilms, fungi and some viruses, and it has been put forward as a starting point for new kinds of antimicrobial drugs in a time of antibiotic resistance [6]D [9]D.

  If it held in humans: peptides like LL-37 could offer an option where ordinary antibiotics fail.

  Why it might not translate: in the laboratory the concentration and conditions are controlled. In the body the peptide meets salts, proteins and enzymes that weaken it, and the concentration at which it kills microbes is close to the one at which it damages the body's own cells, 1-10 micromolar [4]D. This narrow window may be one reason why LL-37 given from outside did not prove effective in the largest human trial.

  Evidence now: there are many reviews, but no new registered human trial of giving LL-37 is known after the phase 2b trial was published [10]E, not peer-reviewed.

Who studies LL-37?

Treatment studies of LL-37 have been small academic and company-led projects, and basic research is spread around the world.

The leg ulcer trials were published in a wound care journal. Three authors of the phase 2b trial worked for the Swedish company Promore Pharma, which is developing LL-37 under the name ropocamptide [2]A. The diabetic foot ulcer trial was done in Indonesia [3]A and the COVID-19 trial in China [7]A. The melanoma trial was sponsored by the M.D. Anderson Cancer Center in the US [15]E, not peer-reviewed.

Basic research and reviews come from Sweden [4], Italy [6] and Canada [9], among others.

Who wrote this page: the editorial team

This page was written by the Molecular Recovery editorial team. The draft was made with the help of AI, and the claims were checked against the sources (AI policy). The page has not been medically reviewed. How we work: About us.

What are the side effects of LL-37?

In the wound trials LL-37 was well tolerated on the skin [1]A [2]A, but in cell studies it is toxic to human cells [4]D, and after injections into a tumour one patient developed widespread skin lesions [14]B. Peptide harms more generally: side effects in studies.

Reported in humans

The first leg ulcer trial found no local or systemic safety concerns [1]A. In the phase 2b trial the product was well tolerated and safe at both concentrations [2]A. The trial by mouth found no serious adverse events during the hospital stay or follow-up [7]A.

A melanoma patient who received eight weekly LL-37 injections into the tumour developed, about 45 days after the start, several wart-like bumps and a blistering lesion on the trunk and limbs. Biopsies showed inflammation and atypical squamous cell growth. The lesions cleared within two months of stopping the injections [14]B. This is a case report of one patient.

These observations cover use on a wound for four weeks, one product taken by mouth and one treatment injected into tumours. No human study of LL-37 injected under the skin or into muscle is known.

In cell studies

LL-37 is toxic to several types of human cells at 1-10 micromolar, especially to infected cells [4]D. According to the review the peptide can also locally destroy the host's own cells weakened by infection [4].

What has not been studied

There are no data on long-term use: the longest treatment periods in the published wound trials were four weeks [1]A [3]A. There are no pharmacokinetic data on exposure of the whole body.

Interactions and who has no data

No published data are known on using LL-37 during pregnancy or breastfeeding or in children. No human study of drug interactions is known.

The participants were adults, and in the phase 2b leg ulcer trial the mean age was 67.6 years [2]A. No treatment studies in pregnancy are known. LL-37 levels have been seen to rise in the foetal membranes and uterine muscle after spontaneous labour [11]B, but that is a finding about the body's own peptide and says nothing about the safety of giving it from outside.

The picture in cancer is especially unclear. According to a review, the role of LL-37 in cancer runs both ways: in different cancer types it has been described as both promoting and restraining growth [12]. This is no argument in either direction, but a reason for caution.

Evidence against and what is unknown

The strongest evidence against is the size and order of the trials: a small study showed a strong result, a larger one did not [1, 2]A.

Evidence against

• Small study first, large one after. The first study (34 participants) showed a strong result [1]A. The larger phase 2b trial (148 participants) found no difference from placebo in the whole study population [2]A. In small studies, chance and selection easily produce impressive numbers that do not survive a larger trial. The positive finding remained only in a post hoc subgroup analysis, and a subgroup chosen after the fact is weaker evidence than a pre-specified main outcome.
• Dose response. In the first study the highest concentration, 3.2 milligrams per millilitre, did not differ from placebo [1]A. The highest concentration did not work, which fits the cell-study finding that at high concentrations LL-37 also damages the body's own cells [4]D.
• Mechanism. In the diabetic foot ulcer trial the wound healed faster, but inflammatory mediators did not fall and the bacterial count did not fall significantly [3]A. The assumed mode of action did not show in the measurements.

What the evidence does not support

• "LL-37 is an approved medicine": it is not. No LL-37 product has a marketing authorisation in Finland, the EU or elsewhere.
• "Injected LL-37 helps": no human study of LL-37 injected under the skin or into muscle is known. The only injection trial was of injections into tumours in four melanoma patients [15]E, not peer-reviewed.
• "LL-37 calms inflammation or boosts immunity": according to the review it has both pro- and anti-inflammatory effects [4]D, and there are no human studies for these uses.
• "Raising your own LL-37 equals taking LL-37": these are different things, and most of the literature concerns the first [8]B.

LL-37 dosage in studies

LL-37 has no approved dosage, because no product has been approved as a medicine. Below are the concentrations and doses used in studies. They are not recommendations.

What the numbers do not tell you. The wound trial figures are concentrations applied to the wound, not injections or whole-body doses. They cannot be converted into an injected dose, because LL-37 injected under the skin or into muscle has not been studied in humans. The dose response was not linear either: the highest concentration studied did not work [1]A, and in cell studies the peptide damages human cells at 1-10 micromolar [4]D. More is not better. The table with sources: LL-37 doses in studies. Why the vial calculator does not cover this substance: LL-37 and the calculator. How to read study doses in general: doses in studies.

Is LL-37 legal in Finland?

LL-37 has no marketing authorisation in Finland, so it may not be sold or marketed as a medicine.

In Finland, Fimea is responsible for marketing authorisations and supervision of medicines; an overview is on the page Peptides and the law in Finland. Fimea has separately warned against using injectable peptides bought online [13]regulator. Bringing them back from travel: Travelling with peptides.

The label "research use only" is a classification, not a permit. It means the product is not intended or made for human use and has not passed the quality, safety and efficacy checks required of a medicine.

LL-37 has one extra point. Because it is the body's own peptide, an idea circulates online that it would therefore be safer or more natural than other research peptides. That does not follow: the body makes its own peptide in tightly controlled amounts in the right place, and a dose from outside does not reproduce that. The same review that describes the peptide's defence roles also describes its toxicity to the body's own cells [4]D.

This is not legal advice. Disclaimer.

How to assess an LL-37 product

For LL-37 sold as a research chemical you can check the certificate of analysis, but the length of the peptide makes judging purity harder than usual.

LL-37 is a 37-amino-acid peptide cut from the 170-amino-acid precursor hCAP-18. The precursor also yields several other peptides, such as LL-23, KR-20 and RK-31 [5]. This matters for quality: shortened forms made during synthesis and storage are chemically close to the right product, but they are not it.

Reading a certificate of analysis (COA) has its own page: How to read a COA. Two things decide. The first is the method: liquid chromatography, HPLC, gives purity, and mass spectrometry (MS) tells whether the molecule is the right one. A long peptide needs both, because one missing amino acid may not show in the purity figure. The second is that the batch number matches the vial.

The certificate says nothing about sterility, endotoxin content or what happened to the product in transport. Handling and storage: Mixing and storage.

The future: what comes next?

No new registered human trial of giving LL-37 is known.

The authors of the phase 2b trial themselves say that the subgroup finding in large wounds would need a new, adequately sized trial [2]A. No such trial was in the clinical trial registry on 14 Oct 2026 [10]E, not peer-reviewed. The registry record NCT04098562 of the diabetic foot ulcer trial has status unknown, last updated in 2019 [10]E, not peer-reviewed.

In practice the development of LL-37 as a treatment looks stalled, even though basic research and reviews remain lively.

FAQ

Read the research yourself: sources

Every source opens the original publication or the regulator's page. The level shows what the source itself reports.

1. Grönberg A, Mahlapuu M, Ståhle M, Whately-Smith C, Rollman O. Treatment with LL-37 is safe and effective in enhancing healing of hard-to-heal venous leg ulcers: a randomized, placebo-controlled clinical trial. Wound Repair Regen. 2014;22(5):613-621. PMID 25041740 A
2. Mahlapuu M, Sidorowicz A, Mikosinski J, et al. Evaluation of LL-37 in healing of hard-to-heal venous leg ulcers: A multicentric prospective randomized placebo-controlled clinical trial. Wound Repair Regen. 2021;29(6):938-950. PMID 34687253 A
3. Miranda E, Bramono K, Yunir E, et al. Efficacy of LL-37 cream in enhancing healing of diabetic foot ulcer: a randomized double-blind controlled trial. Arch Dermatol Res. 2023;315(9):2623-2633. PMID 37480520 A
4. Svensson D, Nilsson BO. Human antimicrobial/host defense peptide LL-37 may prevent the spread of a local infection through multiple mechanisms: an update. Inflamm Res. 2025;74(1):36. PMID 40063262 (review)
5. UniProt. P49913 (CAMP_HUMAN), cathelicidin antimicrobial peptide: 170 amino acids; cleaved into LL-37, FALL-39, LL-29, LL-23, FF-33, RK-31, KS-30 and KR-20 (record read 14 Oct 2026). UniProt
6. Xhindoli D, Pacor S, Benincasa M, et al. The human cathelicidin LL-37: A pore-forming antibacterial peptide and host-cell modulator. Biochim Biophys Acta. 2016;1858(3):546-566. PMID 26556394 (review)
7. Zhao Y, Zhang H, Zhao Z, et al. Efficacy and safety of Oral LL-37 against the Omicron BA.5.1.3 variant of SARS-COV-2: A randomized trial. J Med Virol. 2023;95(8):e29035. PMID 37605995 A
8. Mily A, Rekha RS, Kamal SM, et al. Oral intake of phenylbutyrate with or without vitamin D3 upregulates the cathelicidin LL-37 in human macrophages: a dose finding study for treatment of tuberculosis. BMC Pulm Med. 2013;13:23. PMID 23590701 B
9. Bhattacharjya S, Zhang Z, Ramamoorthy A. LL-37: Structures, Antimicrobial Activity, and Influence on Amyloid-Related Diseases. Biomolecules. 2024;14(3):320. PMID 38540740 (review)
10. ClinicalTrials.gov. Studies with LL-37 as the intervention (registry status 14 Oct 2026), including NCT04098562 (LL-37 cream for diabetic foot ulcers, phase 2, status unknown, last update 2019). Registry entries, not peer-reviewed. ClinicalTrials.gov E, not peer-reviewed
11. Lim R, Barker G, Lappas M. Human cathelicidin antimicrobial protein 18 (hCAP18/LL-37) is increased in foetal membranes and myometrium after spontaneous labour and delivery. J Reprod Immunol. 2015;107:31-42. PMID 25435436 B
12. Chen X, Zou X, Qi G, et al. Roles and Mechanisms of Human Cathelicidin LL-37 in Cancer. Cell Physiol Biochem. 2018;47(3):1060-1073. PMID 29843147 (review)
13. Fimea. Fimea warns: do not use injectable peptides purchased online (29 Jun 2026). Fimea regulator
14. Dolkar T, Trinidad CM, Nelson KC, et al. Dermatologic toxicity from novel therapy using antimicrobial peptide LL-37 in melanoma: A detailed examination of the clinicopathologic features. J Cutan Pathol. 2018;45(7):539-544. PMID 29665030 B
15. ClinicalTrials.gov. NCT02225366: Intratumoral Injections of LL37 for Melanoma (phase 1-2, sponsor M.D. Anderson Cancer Center, completed 2020, 4 participants, results posted in the registry). Registry entry, not peer-reviewed. ClinicalTrials.gov E, not peer-reviewed

Sources read on 13 Oct 2026 and checked on 14 Oct 2026 (PubMed, ClinicalTrials.gov, UniProt). Dose details are as given in the abstracts of the publications.

Evidence levels: A = randomised human trial or meta-analysis · B = observational human study or case series · C = animal study · D = cell or laboratory study · E = not peer-reviewed · regulator = a regulator's assessment or decision, not research evidence. The marker official means official product information (summary of product characteristics or label); it is not research evidence. For a review, the level shows what kind of research the statement rests on. No level: statements that something has not been studied, reviews' interpretations, named hypotheses and the general rules in our certificate-of-analysis guide.

Content is general information about research, not medical advice. If you have symptoms, contact a healthcare professional.