Key takeaways
- LL-37 is the last 37 amino acids of hCAP18, the only human cathelicidin, encoded by the CAMP gene; after neutrophils release hCAP18, the enzyme proteinase 3 cuts LL-37 free.
- It kills many bacteria in lab tests, and in a 2001 mouse study cathelicidin protected against a necrotic skin infection; a 2001 cell study also calls LL-37 cytotoxic, able to harm the body’s own cells.
- Research links the same peptide to inflammation in psoriasis (it helps immune cells react to the body’s own DNA) and rosacea (patients had abnormally high, differently cut cathelicidin).
- In people it has been tested in small trials, mostly on chronic ulcers. A 148-patient phase IIb trial by its developer found no significant healing difference from placebo across all patients.
- FDA lists cathelicidin LL-37 among withdrawn Category 2 compounding nominations (page current 22 April 2026), citing possible immune reactions, too little safety information, and nonclinical findings on male reproduction and tumour growth.
01LL-37 at a glance
The short version, dated 18 November 2026. Every line is sourced at the end of the page.
| Question | LL-37 |
|---|---|
| What it is | An antimicrobial peptide of 37 amino acids, cut from the end of hCAP18, the only human cathelicidin protein. |
| Sequence, formula, mass | LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES: two leucines (LL), 37 amino acids, hence the name. C205H340N60O53, 4493 g/mol (PubChem). |
| Gene | CAMP, on chromosome 3 (3p21.31). Other names for the gene product include hCAP18 and FALL-39. |
| Where the body makes it | Neutrophils, a kind of white blood cell, release hCAP18, and the enzyme proteinase 3 cuts LL-37 from it. Skin makes hCAP18 after injury and in inflammation. |
| Evidence tier | T4: cell and lab studies. T3: mice. T2: human tissue and patient observations. T1, small: randomised trials on chronic leg and foot ulcers (25 to 148 people). |
| United States | No product in Drugs@FDA (18 Nov 2026). FDA lists “Cathelicidin LL-37” among withdrawn Category 2 compounding nominations (page current 22 Apr 2026). Its advisory committee will discuss it before the end of February 2027; no date set. |
| European Union and United Kingdom | No centrally authorised medicine in the EMA’s data; no result on the UK emc database (both 18 Nov 2026). |
| Australia | Not named in the Poisons Standard October 2026, and not being named does not make it approved or permitted. The TGA regulates peptide products used for a therapeutic purpose as therapeutic goods. |
| WADA | Not named on the 2026 or 2027 Prohibited List. WADA’s notes on the 2027 List say a peptide not specifically named “may still be prohibited”. |
02What is LL-37, and where does it come from?
LL-37 is a peptide the human body makes. The CAMP gene encodes a 170-amino-acid precursor, hCAP18, which a 2001 study called “the only human cathelicidin”. LL-37 is its last 37 amino acids.
How it is released. Neutrophils, a kind of white blood cell, keep hCAP18 in their granules. In the 2001 cell study, it was not cut inside the cells after they engulfed bacteria, but outside them after release, and the enzyme proteinase 3 was “solely responsible” for freeing LL-37 (T4). The same study calls LL-37 an “antibacterial and cytotoxic peptide”: it can harm the body’s own cells as well as microbes. On skin, sweat carries the protein, and skin enzymes cut it into shorter fragments named the same way, such as KR-20. How peptides like this are built: how peptides work.
03What does LL-37 do in innate immunity, by model?
Innate immunity is the body’s fast, first-line defence, which acts before antibodies form. LL-37 is one of its molecules.
In the lab (T4). LL-37 kills a broad range of bacteria, including Staphylococcus aureus and E. coli, by forming pores in their membranes. Its activity against MRSA and the yeast Candida albicans was seen in low-salt conditions and lost at higher salt, so lab killing does not show what it does in the body.
In mice (T3). A 2001 study combining mouse and bacterial genetics found that cathelicidin (called CRAMP in mice) protects mice against necrotic skin infection with group A Streptococcus. The authors called it “an important native component of innate host defence in mice”.
In people (T2). A 2002 study compared 6 patients with morbus Kostmann, an inherited lack of neutrophils, with 22 healthy controls. The LL-37 precursor was deficient in the patients’ neutrophils and undetectable in their plasma and saliva. All had severe gum disease except one, who had received a bone-marrow transplant and had near-normal LL-37. With six patients and no intervention, that is an association.
Signalling. The CAMP gene product also “functions in cell chemotaxis, immune mediator induction, and inflammatory response regulation”, NCBI’s gene summary says.
04What has research found on LL-37 in psoriasis, rosacea and wounds?
The same peptide shows up on both sides, in defence and in disease. Each row gives both halves. Context: the skin research area.
| Area | Model | What was found | What it does not show |
|---|---|---|---|
| Psoriasis | Psoriatic skin and human immune cells, 2007 (T2, T4) | LL-37 binds the body’s own DNA into clumps that plasmacytoid dendritic cells, a type of immune cell, treat as a danger signal, releasing interferon. The authors called it “the key factor” in that activation. | Here LL-37 drives inflammation. It is a mechanism in samples and cells, not a test in people. |
| Rosacea | Facial skin of people with rosacea, and mice, 2007 (T2, T3) | People with rosacea had “abnormally high levels of cathelicidin” in facial skin, cut into different fragments. Introducing those fragments into mouse skin increased inflammation. | More cathelicidin goes with disease here, not protection; mice do not show how rosacea starts in people. |
| Wounds | Human skin wounds and lab-cultured human skin, 2003 (T2, T4) | hCAP18 rose in skin after wounding and was absent at the edges of chronic ulcers. In cultured skin, antibodies against LL-37 inhibited re-epithelialisation, the regrowth of skin over the wound. | An association and a blocking experiment; neither tests whether adding LL-37 helps wounds close in people. |
05How has LL-37 been tested in people?
Small trials, mostly of LL-37 applied to chronic ulcers. Results are mixed, and company scientists led two of the four. Promore Pharma says it is developing LL-37 under the non-proprietary name ropocamptide.
| Study | People | Design | What the authors reported |
|---|---|---|---|
| Grönberg et al., 2014; authors from Pergamum AB | 34 adults with hard-to-heal venous leg ulcers | Described by its authors as the first in people: placebo run-in, then 4 weeks randomised and double-blind, LL-37 at three strengths or placebo on the ulcer | Healing-rate measures higher than placebo for the two lower strengths, significant for the lowest only; the highest strength no different from placebo. The abstract reports no safety concerns; the developer’s 2021 paper says the highest-strength group had more frequent and more severe local reactions |
| Mahlapuu et al., 2021 (phase IIb); authors from Promore Pharma, which is developing it | 148 adults with hard-to-heal venous leg ulcers, 15 sites | Randomised, double-blind, two strengths or placebo, with compression therapy | No significant improvement in healing versus placebo across all patients. A subgroup with large ulcers, picked after the fact, did better. 12 serious adverse events in 11 patients (7.4%), none judged related |
| Miranda et al., 2023; Universitas Indonesia | 25 adults with mildly infected diabetic foot ulcers | Randomised, double-blind, LL-37 cream or placebo cream, 4 weeks | A greater rise in a wound granulation index with LL-37; no difference in inflammatory markers or bacterial counts; one mild skin irritation |
| NCT02225366; MD Anderson Cancer Center | 4 people with melanoma enrolled; 3 treated, per the posted results | Phase 1/2, completed 2020; results posted on ClinicalTrials.gov | Posted results list no serious adverse events among the 3; other events include one coded as squamous cell carcinoma. A 2018 MD Anderson case report described a patient given LL-37 in a phase 1 melanoma trial (it does not give the registration number) who developed widespread skin lesions with atypical squamous cell growth, gone within 2 months of stopping it |
A subgroup picked after the fact is a lead, not a finding. How we grade trials: evidence grades.
06What have regulators said about LL-37?
United States. No product containing LL-37 appears in Drugs@FDA (searched for “cathelicidin” and “ropocamptide”, 18 November 2026). FDA’s page on compounding substances that may present significant safety risks, current as of 22 April 2026, lists “Cathelicidin LL-37” among substances previously in Category 2 whose nominations were withdrawn. The page gives no withdrawal date. Its entry reads: “Compounded drugs containing cathelicidin LL-37 may pose risk for immunogenicity for certain routes of administration and may have complexities with regard to peptide-related impurities and API characterization. FDA lacks sufficient safety-related information regarding cathelicidin LL-37 to know whether the drug would cause harm when administered to humans. Nonclinical research findings suggest detrimental effects on male reproduction and that this drug can be protumorigenic in some tissues.” Immunogenicity means provoking an unwanted immune response; protumorigenic means promoting tumour growth. LL-37 does not appear in any category of FDA’s list of nominated substances updated 14 May 2026. FDA’s advisory committee page (current as of 15 April 2026) says the Pharmacy Compounding Advisory Committee will discuss “Cathelicidin (LL-37)”, among substances “being considered for inclusion on the 503A bulks list”, at a meeting before the end of February 2027 whose time and location “will be scheduled in the coming months”. Advisory committees “make non-binding recommendations”. Background: FDA’s categories.
European Union and United Kingdom. The EMA’s data on centrally authorised medicines lists no medicine containing it, and the UK emc database returned no result for “LL-37” or “cathelicidin” (both 18 November 2026). The emc does not cover every UK licence.
Australia. LL-37 is not named in the Poisons Standard October 2026, and not being named does not make it approved or permitted. The TGA says a peptide product is regulated as a therapeutic good if, for example, it is “promoted, supplied or generally used for a therapeutic purpose”. More: are peptides legal in Australia?
Sport. WADA’s 2026 and 2027 Prohibited Lists do not name LL-37 or cathelicidin. WADA’s notes on the 2027 List say that “even if a peptide is not specifically named on the List, utmost caution should be taken as it may still be prohibited and may be harmful to health”. See peptides and anti-doping.
07What we don’t know about LL-37
- Whether adding LL-37 helps any wound heal. The largest trial described here found no difference across all patients, and its positive subgroup was found after the fact.
- Its effects outside small wound trials, or over longer periods: the longest trial described here treated for 13 weeks. FDA says it lacks sufficient safety information to know whether it would cause harm.
- Whether the nonclinical findings FDA cites, on male reproduction and tumour growth, apply to people.
- How its protective and inflammatory roles balance in a given person. The psoriasis and rosacea research shows the same peptide can drive disease.
- What a vial labelled LL-37 contains. FDA notes complexities with peptide-related impurities; only a lab report on that lot can show identity and purity: how to read a COA.
This is a plain-English summary of public documents and studies, checked on 18 Nov 2026. It is not legal or medical advice.