What is LL-37? What it does, benefits and what the evidence shows

LL-37 is the active piece of cathelicidin, an antimicrobial peptide that the human body makes on its own. It is not an approved drug anywhere we could find, and its human evidence is thin: one small first-in-human trial and one larger trial, both of a topical formulation applied to chronic leg ulcers, plus a tiny cancer study. The larger trial found no significant benefit in the full group of patients, only a signal in a subgroup picked after the fact. There is no good human evidence for the things LL-37 is mostly marketed for on the grey market, such as fighting infections or general immune support.

What LL-37 is and how it works

LL-37 is the C-terminal fragment of a larger protein called human cationic antimicrobial protein 18 (hCAP-18), and its name comes from its first two amino acids, two leucines, and its length of 37 (Coffelt et al., PNAS, 2009). It belongs to the cathelicidin family of host-defense peptides. A review of the research describes it as far more than an antibiotic: beyond antibacterial, antifungal and antiviral activity, it attracts immune cells, modulates immune responses, and has been linked to wound healing, blood vessel growth and control of cell death (Vandamme et al., Cellular Immunology, 2012).

That range of activity is why it gets called a "factotum" peptide, and it is also why it is hard to predict. A molecule that recruits immune cells, promotes new blood vessels and affects cell survival can help in one tissue and cause trouble in another, which matters for the risks section below.

The version tested in the clinic is a synthetic copy formulated as a gel to apply to a wound. Its drug developer, Promore Pharma of Sweden, uses the name ropocamptide for it (Promore Pharma press release, November 19, 2020). The leg ulcer trials below used that topical product on open wounds, and the melanoma study injected LL-37 into tumors. None tested a peptide injected under the skin for general wellness.

What people use it for

What the research actually shows

Venous leg ulcers: two trials, one mixed result

The first-in-human trial, published in 2014, enrolled 34 people with venous leg ulcers that were hard to heal. After a run-in period on placebo, participants were randomized to one of three concentrations of LL-37 gel or to placebo. The healing rate was roughly six times placebo at the lowest concentration (statistically significant, p = 0.003) and roughly three times placebo at the middle one (not statistically significant, p = 0.088), while the highest concentration showed no difference from placebo, and the authors reported no safety concerns with local or systemic adverse events (Grönberg et al., 2014). The drop-off at the top concentration matters: more was not better, and nobody has shown that it would be for any other route or use.

The follow-up was a Phase 2b trial called HEAL LL-37, with 148 patients in three arms. In the peer-reviewed paper, the analysis of the full study population "did not identify any significant improvement in healing" compared with placebo: complete wound closure was 26.5% at the lower concentration, 24.7% at the higher one and 25.3% on placebo. A post hoc analysis, meaning one run after the results were in, found significant improvements only at the lower concentration and only in the subgroup with ulcers of 10 cm2 or larger, and the authors concluded that this called for a further study designed to test that group properly (Mahlapuu et al., Wound Repair and Regeneration, 2021).

The company's own announcement was more upbeat. It reported that among patients with larger ulcers, about 28% at the better-performing concentration reached complete healing versus about 8% on placebo, and that no serious side events were considered related to the study drug (Promore Pharma press release, November 19, 2020). The company itself noted the next day that this subgroup analysis was not part of the original statistical analysis plan (Promore Pharma clarification, November 20, 2020). Subgroup findings like this are worth following up, but they are how many drugs that later fail first look promising. We did not find a completed Phase 3 trial of LL-37 for this use, and the most recent status we could confirm from a primary source is the 2020 to 2021 Phase 2b reporting above.

Infection: strong in the lab, unproven in people

The reason LL-37 has its reputation is its antibacterial, antifungal and antiviral activity, which reviews such as Vandamme et al. (2012) describe as part of the natural peptide's biology. We did not find a human trial showing that giving someone LL-37 prevents or treats an infection. The cathelicidin-related trials we did find on ClinicalTrials.gov mostly measure natural LL-37 levels or test vitamin D and probiotics, which are different interventions.

Melanoma: a tiny dose-finding study

MD Anderson Cancer Center ran a Phase 1/2 study of LL-37 injected directly into melanoma lesions, meant to find a tolerable dose and see whether it could stimulate an immune response against the disease. The registry record lists 4 enrolled participants and a completion date in November 2020, while its posted results describe only 3 treated participants in two dose groups, one of whom stopped for lack of efficacy (ClinicalTrials.gov NCT02225366). We could not find a peer-reviewed publication of the results, and so few patients cannot show whether it works. It belongs in the "early human data" category only in the loosest sense.

General wellness, gut, joints and immunity

We found no human trial of LL-37 for any of these. The claims come from LL-37's role as a natural immune molecule, not from a test of taking it as a supplement.

Side effects and risks

In the leg ulcer trials, the topical gel was described as well tolerated, with no serious adverse events assessed as related to the study drug (Mahlapuu et al., 2021). That is reassuring for a gel on a wound. It says little about a peptide injected under the skin, which no trial we found has tested.

The bigger concern is that LL-37's biology cuts both ways, and the lab literature contains findings that should give a buyer pause:

• Tumors. Ovarian tumors overexpress the LL-37 precursor, and LL-37 stimulated ovarian cancer cell growth and invasion in cell experiments (Coffelt et al., International Journal of Cancer, 2008). In mice, blocking LL-37 reduced ovarian tumor growth, and the authors concluded that it can help tumors recruit supporting cells and blood vessels (Coffelt et al., PNAS, 2009). These are lab and mouse findings about the peptide the body makes inside tumors, not proof that a vial of LL-37 causes cancer. They are also not proof that it doesn't.
• Autoimmunity. LL-37 was identified as the factor that turns the body's own DNA into a trigger for immune activation in psoriasis, and the authors suggested this pathway may drive that autoimmune disease (Lande et al., Nature, 2007). Anyone with an autoimmune condition should read that finding carefully.
• Cell survival. LL-37 protected human skin cells from induced cell death in the lab, an effect the authors linked to psoriasis (Chamorro et al., Journal of Investigative Dermatology, 2009).

None of this is a demonstrated effect of injectable LL-37 in people. It is a reason to be cautious about a compound that is already active in inflammation, angiogenesis and cell survival, with no long-term human safety data for non-topical use.

On the grey-market side, the risks our data shows are the ordinary ones: what is in the vial versus the label. With only two vials on record, we cannot say more than what follows.

What the lab tests show

As of September 29, 2026, our database holds 2 vials sold as LL-37, from 2 different vendors, with lab reports dated April 2025 and August 2026. Purity measured 98.8% on one and 99.4% on the other. The measured amounts came in about 2% and about 11% under the label. Neither vial met our failure definition (empty, under 95% pure, more than 20% under label or more than 50% over).

Two vials is not a sample. It says these two batches measured high purity and reasonably close to their labeled amount (a purity result alone does not confirm identity), and nothing about how any vendor performs on this compound. We don't have enough lab data on LL-37 to rank vendors for it, so here are the highest-ranked vendors across all compounds right now. A vendor's overall record is a signal, not a guarantee for a compound they have never had tested.

Buying smarter

Because LL-37 has almost no lab data behind it, the most useful thing a buyer can do is add to it. Check whether a vendor has any record for this compound on our LL-37 results page, and treat a posted certificate of analysis as a starting point, not proof (see how to read a peptide COA and the red flags in fake peptide COAs).

Testing your own vial is worth it for a compound with this little history. How to get a peptide tested covers identity, purity and quantity testing, and endotoxin in research peptides explains when the add-on matters.

If your report comes back failing, stop using that batch, tell anyone who bought from the same run, bring the report to the vendor, and submit it to VendorVerify. For related compounds that people pair with LL-37 in the wound-repair conversation, see our guides to BPC-157, KPV and GHK-Cu, and browse the full peptide hub.

Database figures are from independent lab reports in VendorVerify as of September 29, 2026. This isn't medical advice. LL-37 is an investigational compound with no approved use that we could find, and vials sold as LL-37 are sold for research use only.