LL-37 dosage and evidence

What is LL-37?

How does LL-37 work?

It breaks open the outer skin of germs. It also changes how immune cells act, how wounds repair and how much swelling there is.

What have studies in people found about LL-37?

LL-37 has been given to people in two trials for leg ulcers caused by poor vein circulation. The same Swedish company ran both. The first, in 34 patients, reported faster healing at the two lower doses and nothing at the highest. The second was built to confirm it in 148 patients. Across everyone, it found no difference from a dummy treatment.

• The larger trial: 148 patients got 0.5 mg/mL, 1.6 mg/mL or a dummy treatment, all on top of compression bandaging. Their average age was 67.6. Their ulcers had been open for about 20.3 months and averaged 11.6 square centimetres. Across everyone, LL-37 did not heal ulcers better than the dummy. Looking back afterwards, several healing measures were better in people whose ulcer was at least 10 square centimetres. Both strengths were well tolerated.
• The first trial in people: 34 patients with hard-to-heal leg ulcers first had 3 weeks on a dummy treatment. Then they got LL-37 on the wound twice a week for 4 weeks, at 0.5, 1.6 or 3.2 mg/mL, or the dummy, and were followed for 4 more weeks. Healing was about six times faster than the dummy at 0.5 mg/mL, and about three times faster at 1.6 mg/mL, though that second result could still be chance. Average ulcer size fell 68% and 50% in those two groups. The strongest dose, 3.2 mg/mL, did no better than the dummy, so the dose effect ran backwards.
• Melanoma: a small early study at MD Anderson Cancer Center injected LL-37 into melanoma tumours (NCT02225366). It finished in November 2020 with 4 people enrolled and has posted no results.
• Almost every other human study listed under LL-37 measures the body's own supply of it in the blood, usually in vitamin D trials. They do not give the peptide to anyone.

What this does not tell you: The hopeful result comes from 34 people at one of three strengths. The strongest dose did no better than the dummy. That pattern fits a chance finding as well as a 'sweet spot' dose. The trial built to confirm it did not. A look back at a subgroup of large wounds is an idea for the next trial, and the authors say so themselves. Both trials put the peptide on an open wound. Nothing here describes injected, swallowed or inhaled LL-37. No trial has tested it against an infection, the property the peptide is famous for.

The state of the research

The evidence, in charts

The studies we cite

• Study in people2014
  Treatment with LL-37 is safe and effective in enhancing healing of hard-to-heal venous leg ulcers: a randomized, placebo-controlled clinical trial

  The first human test of LL-37 on wounds looked strongly positive. Thirty-four people with leg ulcers that had refused to heal had the lowest strength applied to the wound, or a dummy. Ulcers on the drug closed roughly six times faster. Nothing worrying turned up on safety. Thirty-four people is a first look, not an answer, and the larger trial that followed did not repeat the result. Read this alongside that one, never on its own.

  First-in-man trial in 34 patients: 0.5 mg/mL topical LL-37 raised the healing rate about sixfold versus placebo, with no safety concerns.

  PMID 25041740 Wound Repair and Regeneration
• Study in people2021
  Evaluation of LL-37 in healing of hard-to-heal venous leg ulcers: a multicentric prospective randomized placebo-controlled clinical trial

  The bigger trial tested LL-37 properly and it did not work. A hundred and forty-eight people with stubborn leg ulcers used it or a dummy across several hospitals. Healing was no better on the drug than on the dummy. A benefit showed up only when the researchers went back afterwards and looked at people with large wounds. Picking a group after seeing the results is how false findings are made. The early sixfold result did not hold up.

  Phase 2b in 148 patients: no significant improvement in healing versus placebo in the full population; a benefit appeared only in a post hoc subgroup with large wounds.

  PMID 34687253 Wound Repair and Regeneration
• Study in people2020
  Induction of Antitumor Response in Melanoma Patients Using the Antimicrobial Peptide LL37

  This trial ran, enrolled four people, and has never reported. Doctors at a cancer centre injected LL-37 directly into melanoma tumours in an early safety study, and it closed in November 2020. No results have been posted since. Four people could not have answered the question even if the numbers had been published. A completed trial that reports nothing is still worth recording, because a reader should know the attempt was made and left blank.

  An investigator-initiated Phase 1/2 trial of intratumoral LL37 injection in melanoma, completed November 2020 with 4 participants enrolled. No results have been posted.

  NCT02225366 ClinicalTrials.gov, M.D. Anderson Cancer Center
• Review of studies2006
  LL-37, the only human member of the cathelicidin family of antimicrobial peptides

  A review of LL-37, a 37 amino acid peptide. It kills a wide range of germs, blocks a bacterial toxin, draws immune cells in and helps new skin cover wounds.

  Reviews the 37-residue amphipathic helix, its broad-spectrum antimicrobial activity, LPS neutralization, chemoattraction and wound re-epithelialization roles.

  PMID 16716248 Biochimica et Biophysica Acta
• Lab (cells in a dish)2007
  Plasmacytoid dendritic cells sense self-DNA coupled with antimicrobial peptide

  In a dish, LL-37 turned the body's own DNA into an alarm signal for immune cells, as seen in psoriasis. That is why it may help drive autoimmune disease.

  LL-37 converts inert self-DNA into a trigger of interferon production in psoriasis, the basis for its double-edged role in autoimmunity.

  PMID 17873860 Nature