LL-37: A real human peptide with thin, mostly negative clinical data

LL-37 stands out from most of the peptides people ask about, because it is not something invented in a lab. It is a molecule your own body makes as a first line of immune defense. Marketing leans on that fact to suggest the peptide is both safe and proven, but the real story is messier. As a natural immune signal, LL-37 is one of the best-studied antimicrobial peptides (germ-fighting molecules) in biology. As a drug you might buy and use, though, the human evidence is small, early, mixed, and in the one solid trial, it didn’t work. This page keeps those two things apart: the well-understood biology, and the much weaker case for using LL-37 as a treatment.

What it is

LL-37 is the only human member of the cathelicidin family of antimicrobial peptides. It is a short chain of 37 amino acids (the building blocks of proteins). It is cationic (carries a positive charge) and amphipathic, which means one side of the molecule likes water and the other side repels it. That two-faced shape lets it wedge itself into the outer membranes of microbes and tear them apart.

Your body doesn’t make the finished peptide directly. A gene called CAMP first builds a larger starter protein named hCAP18. Then an enzyme (a molecule that cuts other molecules), called proteinase 3, snips off the active piece, LL-37, inside neutrophils (a type of white blood cell) and along the skin and the moist linings of the body. Once free, the peptide folds into a helix-break-helix shape, basically a spiral with a bend near its twelfth building block.

Its biology genuinely cuts both ways. On top of directly killing bacteria, LL-37 acts as an “alarmin,” an internal alarm signal: it calls in immune cells (neutrophils, monocytes, and T cells) through a receptor called FPR2/FPRL1, nudges the release of cytokines (chemical messengers that coordinate inflammation), and can encourage new blood vessels to grow and skin to close over a wound. But the same membrane-breaking power that hurts bacteria can also hurt your own cells. And LL-37 can latch onto a person’s own DNA, forming clumps that set off inflammation through a sensor called TLR9, a chain of events linked to the skin disease psoriasis.

The claims

Two of the claims about LL-37 rest on real clinical development. The first is healing wounds when applied to the skin, specifically venous leg ulcers (slow-healing sores on the lower leg caused by poor blood return). The second is treating cancer by injecting it straight into a tumor, which has been studied in melanoma (a type of skin cancer).

Beyond those two, the wider research looks at LL-37 for fighting bacteria, breaking up biofilms (slimy communities of bacteria), and fighting fungi and viruses, plus tuning the immune system in general. Some of this work suggests reusing it against cancer, but the cancer findings clash with each other: LL-37 appears to fight tumors in some tissues and to feed them in others, including melanoma.

Separately, a wave of gray-market vendor and “peptide reference” sites push LL-37 for immune support, gut health, chronic infection or biofilm, and anti-aging. These products are usually sold “for research use only,” which is not a legal basis for using them in people. None of those marketed uses is backed by an approval or by proper human trials, so those claims should be read as advertising, not evidence.

The evidence

There are three human studies that matter here, and reading them in order tells the real story.

The first was a Phase I/IIa trial in venous leg ulcers (Grönberg, 2014). It was positive, but very small. It enrolled 34 patients. After a placebo run-in period, they got four weeks of LL-37 applied to the skin twice a week, at three different doses. The two lower doses sped up healing, but the highest dose did no better than placebo, an odd pattern where more isn’t better. The study was run by the company developing the drug.

The second study is the one that counts the most. This Phase IIb trial (Mahlapuu, 2021) randomly assigned about 149 patients across several treatment centers. It was double-blind (neither patients nor doctors knew who got the real drug) and placebo-controlled, and everyone also got standard compression therapy. It failed its main goal. Wounds fully and confirmed closed in 26.5% of the low-dose group, 24.7% of the high-dose group, and 25.3% of the placebo group, basically the same across the board. A later, after-the-fact look at large ulcers hinted at some benefit, but those after-the-fact subgroups can only suggest ideas to test later; they don’t prove anything. The trial was funded by Promore Pharma, and several of the authors worked for the company.

The third human study is in melanoma (NCT02225366), run at MD Anderson. It enrolled just four patients, and its main goal was to find the best biological dose based on side effects. In plain terms, it was a safety and “can we even do this” study, not a test of whether the peptide shrinks tumors.

The animal and lab data are extensive: killing microbes and breaking up biofilms, helping wounds heal and blood vessels grow, and both helping and harming tumors depending on the tissue. None of that tells us whether it helps people.

The honest summary: no properly sized, confirming randomized trial supports any main use of LL-37 as a drug. The one well-sized trial came back negative. The positive results are small, paid for by the developer, and inconsistent in how they respond to dose. There is no human evidence for taking it by mouth or for whole-body use, and poor stability, harm to the body’s own cells, and high production cost have all slowed development.

Safety and side effects

In the controlled skin-wound trials, LL-37 was generally well tolerated. The Phase IIb study reported local reactions such as redness, swelling, and warm skin in a fair number of patients, mostly mild to moderate, with no major whole-body safety warning sign.

The melanoma program produced one documented skin-toxicity case: a 63-year-old woman developed multiple wart-like bumps and a blistering rash about 45 days after starting weekly injections into the tumor. A biopsy (a small tissue sample examined under a microscope) showed an inflammatory buildup with eosinophils (a type of white blood cell tied to allergic and inflammatory reactions), and the lesions cleared up within roughly two months of stopping (Dolkar, 2018).

Two concerns rooted in how the molecule works deserve attention. First, in the lab LL-37 damages the body’s own cell membranes (red blood cells, lymphocytes, fibroblasts) at concentrations close to the levels it needs to kill microbes, which is a real obstacle to giving it throughout the body. Second, LL-37 stuck to a person’s own DNA can switch on immune cells through TLR9, a process tied to psoriasis. The wider research also discusses links to other inflammatory conditions. And there’s the cancer ambivalence noted above: in several tissues LL-37 can push tumors to behave more aggressively, which is part of why regulators have flagged it.

There is no long-term human safety data for whole-body or ongoing use.

Legal and regulatory status

LL-37 is not FDA-approved for any use. It used to sit on the interim 503A “Category 2” list of bulk drug substances that may carry significant safety risks for pharmacy compounding. In April 2026, FDA announced it was removing twelve peptides, including cathelicidin LL-37, from Category 2 because those nominations were voluntarily withdrawn. Important point: being taken off that list does not mean approval. It just reflects withdrawn nominations and a review that’s still pending. LL-37 was not part of the July 2026 advisory committee meeting and is set for a separate review by the end of February 2027. The “peptides are legal again” spin you’ll see from some vendors is promotional and overstated.

There is no record of LL-37 or ropocamptide being approved as a marketed drug in the EU, UK, or anywhere else; the Swedish developer never reached approval after the negative Phase IIb.

For athletes, LL-37 does not appear by name on the WADA 2026 Prohibited List, and it is not a peptide hormone or growth factor. But because it is an outside, non-approved drug substance, it could plausibly be swept up by the S0 “Non-Approved Substances” catch-all. No clear ruling was found either way, so don’t assume it’s allowed.

In practice, LL-37 is sold online as a “research use only” lyophilized (freeze-dried) powder, with no approved human use, no USP monograph (an official quality standard), and no guarantee of manufacturing quality. At the consumer level, there’s no way to verify its purity, its identity, or its endotoxin levels (bacterial contaminants that can cause fever and other reactions).

Bottom line

LL-37 is a real, well-studied human immune peptide, and that biology is solid. But as a treatment you’d actually use, it sits at the weak end of “preliminary human.” The supporting trials are small and run by the developer, the one properly sized trial failed its main goal, and the cancer work is a four-patient dose-finding study. Add in real worries about harm to the body’s own cells, inflammatory autoimmune mechanisms, and tumor-feeding effects that depend on context, and the picture is of a molecule that’s fascinating in the lab but unproven and not approved as a drug. Marketing that sells it as a safe, established immune booster is running well ahead of the evidence.

Sources

• Discovery of FALL-39/LL-37 (Agerberth, PNAS 1995)
• UniProt P49913 — human cathelicidin (CAMP) precursor
• NMR structure, helix-break-helix conformation (PMC5873590)
• FPR2/FPRL1 receptor for LL-37 (De Yang, J Exp Med 2000)
• Phase I/IIa venous leg ulcer trial (Grönberg, 2014, PMID 25041740)
• Phase IIb venous leg ulcer trial, negative primary endpoint (Mahlapuu, 2021, PMC9298190)
• Melanoma Phase 1/2 intratumoral study (NCT02225366)
• Dermatologic toxicity case report (Dolkar, 2018, PMID 29665030)
• LL-37, self-DNA, TLR9 and psoriasis mechanism (PMC3346901)
• LL-37 dual role in cancer review (PMC9445486)
• LL-37 and melanoma local invasion (PMC10046113)
• FDA 503A bulk drug substances framework
• FDA 2026 peptide reclassification analysis (Frier Levitt)
• FDA removal of 12 peptides from Category 2 (Orrick)
• WADA 2026 Prohibited List