Dashnaiv Peptides
Compounds·immunomodulators & host defense·cathelicidin antimicrobial peptide; membrane-active, immunomodulatory

LL-37 peptide: the antimicrobial your body already makes, what three trials found, why it drives psoriasis, and what the injectable vials rest on

What 2,826 indexed publications and 38 randomized trials actually state about ll-37, sentence by sentence, each tied to its source and labeled by the kind of evidence it is.

Human clinical trial Reviewed 32 sources Updated

At a glance

Evidence availability
Human clinical evidence
Strongest tier in the indexed literature
Indexed publications
2,826
Europe PMC, title or abstract, 2026-09-14
Human studies in ledger
3
3 randomized; study count, not efficacy proof
Approval
none found
no registered development programme
Routes reported
oral, topical
from studies in this ledger
Studied in
Humans
25 primary studies in the evidence table
Reviewed by Adam Mirando, PharmD, on . What changed

What LL-37 is, and why the body already makes it

LL-37 is a peptide in the immunomodulators & host defense class (cathelicidin antimicrobial peptide; membrane-active, immunomodulatory). Europe PMC indexes 2,826 publications naming it or a listed alias in a title or abstract, including 38 randomized controlled trials and 23 clinical trials of any design, as of . The strongest evidence tier in that literature is human clinical trials.

This page reports what those studies state, sentence by sentence, with each statement tied to its source and labeled by the kind of evidence it is. It does not recommend use, and it does not translate study doses into anything personal.

What the evidence level means

Publication counts show what is indexed, not how strong the evidence is. The ledger below separates human clinical records from preclinical, analytical, and in-vitro work.

Human clinical studies in this ledger33 randomized · 0 other human clinical
Evidence strengthNot yet assessedSeparate from evidence availability
Efficacy signalNot yet assessedRead the study result, not the tier alone

Interpretation boundary: An indexed publication count is a discovery measure. It does not establish efficacy, safety, approval, or a personal treatment plan.

LL-37 in two minutes

What it is. The one antimicrobial cathelicidin peptide humans make, stored in neutrophils and produced by skin, gut and airway linings. It kills bacteria by breaking their membranes, disrupts biofilms in the laboratory, and signals to immune cells.

What the research actually shows. 2,826 indexed publications, most about the body's own LL-37. Three trials gave it to people: a 148-person topical trial on venous leg ulcers that missed its primary endpoint with a positive subgroup, a diabetic-foot-ulcer cream trial that improved a granulation index over four weeks, and a 238-person trial of an oral bacterium engineered to make LL-37 that shortened viral clearance in COVID-19. Twenty other 'human studies' in the ledger measured LL-37 after vitamin D, probiotics or infection and gave none.

The part vendors leave out. LL-37 in excess is part of what causes psoriasis, rosacea and lupus: it binds the body's own DNA and RNA and turns them into immune triggers.

Status. Not approved anywhere; a topical drug programme (ropocamptide) reached phase 2b; sold as a research chemical.

Where the evidence is thinnest. Injection. No study, in any species in the ledger, has injected LL-37, measured how long it survives in blood, or tested it against a biofilm infection in a person.

Editorial synthesis Written from general pharmacology, the development record and what is commonly reported; not a cited finding. Cited findings on this page carry their own tier.

How LL-37 works: the only human cathelicidin, in defence and in disease

Written from general pharmacology and the development record, and marked as editorial synthesis; the measurements behind it are quoted in the biomarker section.

  • LL-37 is a 37-amino-acid peptide beginning with two leucines, cut by proteinase 3 from the C-terminus of hCAP18, the only cathelicidin gene humans have. Neutrophils carry it in their granules; skin, gut and airway epithelium make it on demand, and vitamin D switches the gene on, which is why so many of the 'human studies' in the ledger are vitamin D trials measuring LL-37 rather than giving it. It is an amphipathic helix that sits in bacterial membranes and breaks them, kills a broad range of bacteria, fungi and enveloped viruses, binds and neutralises endotoxin, and blocks or disrupts biofilms at concentrations below those that kill free bacteria.Editorial synthesis
    Editorial synthesis from general knowledge
  • Its second job is signalling. LL-37 pulls immune cells to a wound through the formyl-peptide receptor FPR2, drives new blood vessels and skin regrowth, and triggers neutrophils to throw out extracellular traps. That second job is also where it turns on its owner: in psoriasis, LL-37 made in excess binds the skin's own DNA and RNA and converts them into a signal that plasmacytoid dendritic cells read as infection, pouring out interferon and inflaming the plaque; in rosacea, abnormal processing yields fragments that inflame vessels; in lupus, LL-37-DNA complexes feed the autoantibody cycle. The same molecule is defence at the right dose in the right place and disease at the wrong one.Editorial synthesis
    Editorial synthesis from general knowledge
  • What is not known is anything about injecting it. The drug programmes that took LL-37 into people chose a wound gel (ropocamptide) and an engineered oral bacterium precisely because the peptide is degraded fast by proteases in blood and tissue and because a systemic antimicrobial peptide is a systemic membrane-active molecule. No pharmacokinetic, dose-finding or safety study of injected LL-37 exists in any species in the ledger.Editorial synthesis
    Editorial synthesis from general knowledge

What happened in the human studies?

This record's ledger holds 3 primary human studies, of which 3 are trials. Few enough to show in full: each card quotes what its abstract reported about LL-37. Read them before any other section on this page.

  • Randomized controlled trial humans 2023 Human clinical trial
    For Oral LL-37, as early as treated in 6 days, the adjusted hazard ratio (HR) for a primary event of nucleic acid negative outcome was 6.27-fold higher than 7-day-later (HR: 6.276, 95% confidence interval [CI]: 3.631-10.848, p < 0.0001), and the adjusted HR of Oral LL-37 within 6 days is higher than placebo (HR: 2.427 95% CI: 1.239-4.751, p = 0.0097).
    Source pmid-37605995 · quoted verbatim from the abstract
  • Randomized controlled trial humans 2023 Human clinical trial
    The increase in granulation index was consistently greater in the LL-37 group on days 7, 14, 21, and 28 (p = 0.031, 0.009, 0.006, and 0.037, respectively).
    Source pmid-37480520 · quoted verbatim from the abstract
  • Randomized controlled trial humans n = 148 2021 Human clinical trial
    Efficacy analysis performed on the full study population did not identify any significant improvement in healing in patients treated with LL-37 as compared with the placebo.
    Source pmid-34687253 · quoted verbatim from the abstract

Trial by trial: what LL-37 did in people

The three studies that gave LL-37 to people, then the kind of study that fills the rest of the ledger.

Human studies that administered LL-37.
StudyParticipantsForm, dose, durationPrimary resultSafety
HEAL LL-37 (Mahlapuu 2021), phase 2b148 adults with hard-to-heal venous leg ulcers, median 20 months old, 11.6 cm² meanTopical, 0.5 or 1.6 mg/mL, with compressionNo significant improvement in healing vs placebo in the full population; post hoc benefit in wounds of 10 cm² or moreWell tolerated and safe at both strengths
Diabetic foot ulcer cream (Miranda 2023)Adults with mildly infected diabetic foot ulcers, JakartaCream twice a week, 4 weeksGranulation index higher at every week; bacterial counts lower until day 28; cytokines unchangedNot detailed in the abstract
Oral LL-37 for Omicron (Zhao 2023)238 adult inpatients, single centre, open labelOral Lactococcus lactis engineered to secrete LL-37, vs placeboViral RNA cleared in 9.8 vs 14.0 days when started within six daysNo severe adverse events
What the other twenty 'human studies' in the ledger are.
Study typeExamples in the ledgerWhat was givenWhat LL-37 was
Vitamin D trialsChildren's respiratory infections (2024); sepsis (2024, 2025); HIV and tuberculosis (2017, 2019)Vitamin D, phenylbutyrateA blood or airway biomarker measured as an outcome
Probiotic and synbiotic trialsInfant colic (2019); B. infantis (2026); L. fermentum (2026); synbiotics (2026)Bacteria or prebioticsA biomarker of mucosal immunity
Disease-marker studiesTuberculosis serum levels (2017); periprosthetic joint infection (2020); nasal polyps (2019); asthma (2015)NothingA marker of infection or inflammation, and in nasal polyps a driver of it
Other interventionsButyrate enemas in shigellosis (2012); exercise (2025)Butyrate; cyclingA marker of the intervention's effect

The ledger's 'human clinical trial' tier is true and misleading at once: the trials are real, and almost none of them gave anyone LL-37.

Editorial synthesis Written from general pharmacology, the development record and what is commonly reported; not a cited finding. Cited findings on this page carry their own tier.

What the evidence shows

One row per primary study in this record's ledger. Fields are read from each paper's indexing and abstract. "Analytical method" marks papers about detecting or measuring the compound rather than about its effects.

25 primary studies in this record's ledger, strongest evidence first. A dash means the abstract did not state the value.
StudyDesignnSpeciesDoseRouteDurationReported outcomeTier
Chen 2026 Clinical trial (LL-37 measured, not given) — Humans——12 weeks; 12 week Synbiotic vs. placebo: URTI incidence reduced from 85.4 to 55.4% (adjusted risk ratio (aRR) 0.64, 95% CI 0.43-0.95; ARR 30.0%, NNT = 3.3; p = 0.028); pneumonia from 17.7 to 7.7% (RR 0.40, 0.17-0.94; p = 0.034);… Human clinical trial
Song 2026 Randomized controlled trial (LL-37 measured, not given) 24 Humans——8 weeks Furthermore, B. infantis BI45 supplementation significantly increased serum levels of immunomodulatory markers Immunoglobulin A (IgA) and antimicrobial peptide LL-37 compared to the placebo (p < 0.05) and modulated the… Human clinical trial
Jin 2026 Randomized controlled trial (LL-37 measured, not given) 49 Humans——— In a randomized, double-blind clinical trial (n = 49), daily intake of 3 × 10 10 CFU of LF61 for 8 weeks increased serum levels of the antimicrobial peptide LL-37 by 12.3% (p 3), activated the carbohydrate metabolism… Human clinical trial
Ito 2026 Randomized controlled trial (LL-37 measured, not given) — Humans——— However, salivary lysozyme, lactoferrin, human β-defensin-2, and LL-37 secretion rates did not change from baseline to after exercise in both trials. Human clinical trial
Gonçalves 2025 Randomized controlled trial (LL-37 measured, not given) 207 Humans——— There was a significant difference in 25vitD levels between groups on the third [ p = 0.002] and seventh [ p Conclusion : Supplementation with higher doses of cholecalciferol may contribute to a significant increase in… Human clinical trial
Reyes 2024 Randomized controlled trial (LL-37 measured, not given) 101 Humans——6 months Baseline 25(OH)D was 24.9 ± 6.1 ng/ml, with 23% having 25(OH)D Conclusions While weekly VD3 supplementation, in doses equivalent to 800 IU and 1600 IU daily, was associated with improved 25(OH)D levels in preschoolers,… Human clinical trial
Ashoor 2024 Randomized controlled trial (LL-37 measured, not given) — Humans——— The (day 7-day 0) change in serum PCT and LL-37 levels and SOFA score were significantly different in Group I ( P = .010, P P P P P = .025). Human clinical trial
Zhao 2023 Randomized controlled trial — Humans—oral6 days; 7 day For Oral LL-37, as early as treated in 6 days, the adjusted hazard ratio (HR) for a primary event of nucleic acid negative outcome was 6.27-fold higher than 7-day-later (HR: 6.276, 95% confidence interval [CI]:… Human clinical trial
Miranda 2023 Randomized controlled trial — Humans——— The increase in granulation index was consistently greater in the LL-37 group on days 7, 14, 21, and 28 (p = 0.031, 0.009, 0.006, and 0.037, respectively). Human clinical trial
Fujita 2022 Clinical trial (LL-37 measured, not given) — Humans——— There were no significant differences in the background characteristics between the 33 NTM and 36 BE patients recruited. Human clinical trial
Mahlapuu 2021 Randomized controlled trial 148 Humans—topical— Efficacy analysis performed on the full study population did not identify any significant improvement in healing in patients treated with LL-37 as compared with the placebo. Human clinical trial
Vargas 2020 Randomized controlled trial (LL-37 measured, not given) 20 Humans——60 years; 90 days The increased ASL antimicrobial activity seen during the summer-fall was abrogated by adding the LL-37 neutralizing antibody. Human clinical trial
Banke 2020 Clinical trial (LL-37 measured, not given) 8 Humans——— For statistical analysis, SPSS 26.0/R3.6.3 (p Results The AMPs LL-37 and HBD-3 were significantly elevated (up to 20×) in synovial membranes from PJI compared to aseptic loosening or primary arthroplasty. Human clinical trial
Nocerino 2020 Randomized controlled trial (LL-37 measured, not given) — Humans——28 days The rate of infants with reduction of ≥50% of mean daily crying duration was higher in infants treated with BB-12, starting from the end of 2nd week. Human clinical trial
Missailidis 2019 Randomized controlled trial (LL-37 measured, not given) 167 Humans——16 weeks Vitamin D + phenylbutyrate treatment significantly increased 25-hydroxyvitamin D levels ( p p = 0.02), whereas treatment failed to significantly impact LL-37-levels in multiple regression analysis. Human clinical trial
Cao 2019 Clinical trial (LL-37 measured, not given) — Humans——— Nasal polyps had higher NETs% compared with that of controls (23.01% ± 3.43% vs 4.52% ± 1.33%, P Conclusion Neutrophil extracellular traps are significantly increased in nasal polyps and LL-37 induces NET formation in… Human clinical trial
Majewski 2017 Randomized controlled trial (LL-37 measured, not given) 31 Humans——— We established that the mean level of LL-37 was statistically significantly higher in TB patients than that in patients with Gram-positive bacteria-induced pneumonia (p < 0.001), in patients with Gram-negative… Human clinical trial
Han 2017 Randomized controlled trial (LL-37 measured, not given) 30 Humans——— Plasma concentrations of free 25(OH)D over time were correlated with total 25(OH)D levels (r= 0.82; P 3 dose than with the lower dose. Human clinical trial
Thijs 2015 Randomized controlled trial (LL-37 measured, not given) — Humans——— Levels of neutrophil α-defensins (human neutrophil peptides 1-3; HNP1-3) and lipocalin 2 (LCN2; also known as NGAL) were significantly lower in asthmatics, but no differences in LL-37 and SLPI were detected. Human clinical trial
Koziel 2014 Clinical trial (LL-37 measured, not given) — Humans——— Also, the ability of citrullinated LL-37 to quench macrophage responses to lipoteichoic acid and poly(I:C) signaling via TLR2 and TLR3, respectively, was significantly reduced. Human clinical trial
Raqib 2012 Randomized controlled trial (LL-37 measured, not given) 40 Humans——3 days Stool concentration of LL-37 remained significantly higher in the Intervention group on days 4 and 7. Human clinical trial
Zhang 2026 Animal study — ———— Short hairpin RNA (shRNA)-mediated Stac silencing significantly enhanced EV71 infection, while Stac overexpression markedly reduced it. Animal, preclinical
Giraldo-Lorza 2026 In vitro study (LL-37 measured, not given) — Humans——— ITC measurements confirmed a higher binding affinity of LL-37 for sphingomyelin-rich regions, whereas fluorescence permeabilization assays demonstrated a progressive increase in membrane resistance to peptide-induced… Mechanistic, in vitro
Hou 2026 In vitro study — ———— The double cooperative effect between two major antimicrobial peptides, LL-37 and HNP1, where their combination enhances antimicrobial efficiency while reducing cytotoxicity, offers a promising strategy for developing… Mechanistic, in vitro
Ramotar 2026 In vitro study (LL-37 measured, not given) — Humans——— LL-37-mediated increase in COX-2 expression was significantly higher than that mediated by citLL-37. Mechanistic, in vitro

Doses reported in studies

Each entry is a sentence quoted from the study's abstract, with the study's design, sample size and year in front of it. Comparator doses in the same sentence are not attributed to ll-37.

  • In HEAL LL-37, 148 patients with hard-to-heal venous leg ulcers received topical LL-37 at 0.5 or 1.6 mg/mL, or placebo, with compression.Human clinical trial
    Source pmid-34687253
  • Diabetic foot ulcers were treated with LL-37 cream or placebo cream twice a week for four weeks.Human clinical trial
    Source pmid-37480520
  • 238 COVID-19 inpatients received an oral recombinant Lactococcus lactis engineered to produce LL-37, or placebo.Human clinical trial
    Source pmid-37605995

Every LL-37 dose in the trials, in one table

Every figure here is a topical concentration or a probiotic formulation. No study has injected LL-37 into a person.

Doses and forms of LL-37 administered in human studies.
StudyFormConcentration or doseScheduleSite
HEAL LL-37 (2021)Topical gel with compression0.5 mg/mL and 1.6 mg/mLApplied at dressing changes over a 13-week treatment phase (full text)Venous leg ulcer
Diabetic foot ulcer (2023)CreamNot stated in the abstractTwice a week, 4 weeksFoot ulcer
Oral LL-37 (2023)Engineered Lactococcus lactis secreting LL-37Bacterial dose, not a peptide doseDaily, during admissionGut

Figures for injected LL-37 circulate on forums and vendor pages. None has been given to a person in any study, so none is reproduced here; the table holds every form a study actually used. The gap is not a missing number but a missing route: nobody has put this peptide into a human bloodstream on purpose and watched.

Editorial synthesis Written from general pharmacology, the development record and what is commonly reported; not a cited finding. Cited findings on this page carry their own tier.

Side effects: what the topical trials recorded, and what the biology warns

The topical trials called it well tolerated at both strengths. The concern with injecting it comes from biology, not trials: LL-37 disrupts human cell membranes at high concentrations and drives inflammation in psoriasis, rosacea and lupus, where the body already makes too much. Events and their frequency as each study reported them, with the denominator where the abstract gives one.

  • Topical LL-37 was well tolerated and safe at both strengths in 148 leg-ulcer patients.Human clinical trial
    Source pmid-34687253
  • No severe adverse events were seen with the oral LL-37 bacterium during hospitalisation and follow-up.Human clinical trial
    Source pmid-37605995
  • LL-37 drives neutrophils to form extracellular traps, which are increased in nasal polyps; the peptide is part of the inflammation there, not a remedy for it.Observational, human
    Source pmid-31046155
  • No study has injected LL-37 into a person, so the injectable product has no safety data at all. The biology is the caution: LL-37 is amphipathic and disrupts human cell membranes at high concentrations, and in psoriasis, rosacea and lupus the body's own overproduction of it is part of the disease, because it binds self-DNA and RNA and turns them into immune triggers. Whether injected LL-37 reaches concentrations that matter is unknown; whether that is reassuring depends on the peptide surviving long enough to do anything, which is also unknown.Editorial synthesis
    Editorial synthesis from general knowledge · primary document to be added to the ledger

Who LL-37 is discussed for, and the cautions that recur

Studied: adults with chronic venous or diabetic wounds, topically; COVID-19 inpatients, orally via an engineered bacterium. Never studied: anyone by injection or nasal spray; anyone with Lyme disease, mould illness, SIBO, biofilm infection or chronic sinusitis. Community: people with those diagnoses, under functional-medicine protocols.

No trial excluded anyone for the injected route because none used it. The cautions come from the peptide's biology:

  • Psoriasis, rosacea, lupus and other autoimmune or inflammatory skin disease. LL-37 overproduction is part of the disease mechanism in all three. Adding more is the opposite of what the biology suggests.
  • Inflammatory sinus and airway disease. LL-37 drives neutrophil extracellular traps in nasal polyps; the nasal spray is aimed at exactly the tissue where that was shown.
  • Cell toxicity. An amphipathic membrane-active peptide is not fully selective for bacteria; at high local concentrations it damages human cells, which is why the drug programme used a gel on a wound.
  • 'Die-off' reasoning. Symptoms after injection are read by users as bacteria dying. No study supports that reading, and the same symptoms describe an inflammatory reaction to the peptide.
  • Antibiotic replacement. No trial has tested LL-37 against an infection in a person; using it instead of antibiotics for a real infection has no evidence and an obvious cost.
  • Tested athletes. Unapproved substances are prohibited under WADA S0 at all times.

Editorial synthesis Written from general pharmacology, the development record and what is commonly reported; not a cited finding. Cited findings on this page carry their own tier.

Biomarkers measured in studies

Laboratory and clinical measures the studies report tracking, including which hormones did and did not change.

  • Clinical trial humans 2026 Human clinical trial
    Secondary outcomes included pneumonia, diarrhea, eczema, antibiotic use, gut microbiota (16S rRNA V3-V4 sequencing), and fecal immune biomarkers (calprotectin, sIgA, HBD-2, LL-37).
    Source pmid-42598364 · quoted verbatim from the abstract
  • Randomized controlled trial humans n = 24 2026 Human clinical trial
    Furthermore, B. infantis BI45 supplementation significantly increased serum levels of immunomodulatory markers Immunoglobulin A (IgA) and antimicrobial peptide LL-37 compared to the placebo (p < 0.05) and modulated the gut microbiota by enriching beneficial short-chain fatty acid producers.
    Source pmid-42118446 · quoted verbatim from the abstract
  • Randomized controlled trial humans n = 49 2026 Human clinical trial
    In a randomized, double-blind clinical trial (n = 49), daily intake of 3 × 10 10 CFU of LF61 for 8 weeks increased serum levels of the antimicrobial peptide LL-37 by 12.3% (p 3), activated the carbohydrate metabolism pathway (p = 0.002), and maintained stable α-diversity in the microbiome (Shannon index p > 0.05).
    Source pmid-41655688 · quoted verbatim from the abstract
  • Randomized controlled trial humans 2023 Human clinical trial
    This study analyzes the efficacy of LL-37 cream in enhancing wound healing rate and decreasing the levels of IL-1α, TNF-α, and the number of aerobic bacteria colonization in DFU with mild infection.
    Source pmid-37480520 · quoted verbatim from the abstract
  • Clinical trial humans 2022 Human clinical trial
    Additionally, the Hospital Anxiety and Depression Scale (HADS) scores and chronic inflammatory biomarkers of the alveolar region (surfactant protein [SP]-A, SP-D, and human cationic antibacterial protein [hCAP]/LL-37) were assessed and compared between NTM-LD and BE patients.
    Source pmid-36329435 · quoted verbatim from the abstract
  • Randomized controlled trial humans n = 148 2021 Human clinical trial
    In contrast, a post hoc analysis revealed statistically significant improvement with LL-37 treatment in several interrelated healing parameters in the subgroup of patients with large target wounds (a wound area of at least 10 cm 2 at randomization), which is a known negative prognostic factor for healing.
    Source pmid-34687253 · quoted verbatim from the abstract

Reported timelines

Onset, peak and duration figures as each study reported them.

  • Granulation improved more with LL-37 cream at every weekly measurement over four weeks.Human clinical trial
    Source pmid-37480520
  • Oral LL-37 started within six days of diagnosis shortened viral RNA clearance from 14 to about 10 days.Human clinical trial
    Source pmid-37605995

What is measured over time, and what is not

The trials measured wounds and viruses over weeks. Granulation improved from day 7 and at every weekly measurement to day 28 under LL-37 cream; the leg-ulcer trial ran a 13-week treatment phase after a three-week placebo run-in and found no overall difference in healing; viral RNA cleared in about 10 days instead of 14 with the oral bacterium when started early. Nothing at all has been measured for injected LL-37: not how long it survives in blood, where it goes, what concentration it reaches, or what it does at any time point. 'When to take LL-37' and 'how long until it works' have no answers in the literature because the route has never been studied.

Editorial synthesis Written from general pharmacology, the development record and what is commonly reported; not a cited finding. Cited findings on this page carry their own tier.

Study durations

Treatment and follow-up periods as stated in each abstract.

  • Four weeks of twice-weekly cream in the diabetic-foot trial, with wounds measured weekly.Human clinical trial
    Source pmid-37480520
  • The venous-ulcer patients had a median ulcer duration of 20 months at entry; the treatment phase followed a placebo run-in.Human clinical trial
    Source pmid-34687253

Routes: topical gel and cream, an oral bacterium, and nothing injected

Topical gel on leg ulcers, cream on foot ulcers, and an engineered oral bacterium in the trials; nasal sprays and subcutaneous injection, the products people buy, have never been studied in people. Routes of administration named in each study.

  • Randomized controlled trial humans 2023 Human clinical trial
    administration by oral route reported
    Recombinant LL-37 Lactococcus lactis (Oral LL-37) was designed to prevent progression of COVID-19 by targeting virus envelope, however, effectiveness and safety of Oral LL-37 in clinical application was unclear.
    Source pmid-37605995 · quoted verbatim from the abstract
  • Randomized controlled trial humans n = 148 2021 Human clinical trial
    administration by topical route reported
    This clinical trial HEAL LL-37 was a phase IIb double-blind, randomized, placebo-controlled study, with the aim to evaluate the efficacy and safety of a new drug LL-37 for topical administration, in combination with compression therapy, in 148 patients suffering from hard-to-heal venous leg ulcers.
    Source pmid-34687253 · quoted verbatim from the abstract

What people report outside the literature

Forum reports describe subcutaneous injection for Lyme disease, mould illness, SIBO, biofilms and chronic infections, often with die-off symptoms and injection-site pain. None of it has a study, and the route has none either. What circulates in user communities and clinic marketing, reported so a reader knows the figures and their origin. None of it has been tested in a trial, and none of it is a suggestion.

  • LL-37 circulates for 'biofilm infections', Lyme disease and its co-infections, mould illness, SIBO and gut dysbiosis, chronic sinusitis and general 'immune support', injected subcutaneously in daily or cycled patterns, sometimes as a nasal spray, often alongside KPV, BPC-157 or thymosin alpha-1 under functional-medicine protocols. Reports describe injection-site pain and swelling, 'die-off' symptoms read as evidence of effect, fatigue, and in some cases flares of skin or gut inflammation. None of this comes from a study: no human study has injected LL-37 or sprayed it into a nose, and the laboratory biofilm results were never carried into a patient. Figures for this compound circulate on forums and vendor pages. None has been tested in a human study by the route people use, so none is reproduced here; the table above holds every dose a study actually gave.Editorial synthesis
    Editorial synthesis from general knowledge
Method note · shared across compounds

Reconstitution mathematics

Lyophilised peptide is supplied as a mass in a vial and dissolved in a stated volume of diluent. Two figures follow from that and nothing else:

  • Concentration = mass in vial ÷ diluent volume. A 5 mg vial in 2 mL is 2.5 mg/mL, or 2,500 mcg/mL.
  • Volume for a given mass = that mass ÷ concentration. 250 mcg at 2,500 mcg/mL is 0.1 mL. On a U-100 insulin syringe, where 1 mL is 100 units, that is 10 units.

Study doses on this page are quoted as published, in the units the authors used. Converting between mcg, mg and IU, or between a mass and a syringe volume, is arithmetic; deciding a mass is not, and this page does not do it. The reconstitution calculator performs the conversion with the formula shown.

Method note · shared across compounds

Reading a certificate of analysis

A certificate of analysis is a laboratory report on one batch. Reading it means checking four things in order.

  1. Identity. Mass spectrometry should report an observed molecular mass matching the peptide's theoretical mass within the instrument's stated tolerance. A certificate with no mass spectrometry, or one that lists only the expected value, has not confirmed identity.
  2. Purity. HPLC purity is the area of the main peak as a share of all peaks, at a stated wavelength. It says nothing about what the other peaks are, and it is not a measure of quantity.
  3. Quantity. Net peptide content, by amino acid analysis or UV, is the fraction of the vial's mass that is peptide rather than counter-ions and water. A 5 mg vial at 80% net content holds 4 mg of peptide, and every concentration figure above changes accordingly.
  4. Provenance. The batch number on the certificate should match the vial, the laboratory should be named and independent of the seller, and the test date should be recent.

This page rates no supplier and links to none. It describes how to read the document.

Method note · shared across compounds

Equipment described in studies

Published human studies of injectable peptides report the same small set of materials: single-use insulin syringes with fixed needles, most often U-100 with 29 to 31 gauge needles; bacteriostatic water or sterile water for injection as diluent, with the choice stated in the methods; alcohol swabs; and refrigerated storage of reconstituted solution. Animal studies more often report intraperitoneal or intragastric administration and state the vehicle used. Where a study in this record's ledger names its materials or vehicle, that sentence appears under storage or routes above; where it does not, nothing is assumed here.

Storage and handling

LL-37 is a laboratory reagent with published handling data from reagent suppliers: lyophilised, stored frozen, reconstituted in low-binding tubes because the peptide sticks to plastic and aggregates in salt. That is laboratory guidance for cell-culture use, not for a vial sold for injection, whose contents are unverified.

Editorial synthesis Written from general pharmacology, the development record and what is commonly reported; not a cited finding. Cited findings on this page carry their own tier.

Common mistakes in how LL-37 is discussed

  1. Counting studies that measured LL-37 as studies that gave it. Twenty of the ledger's twenty-three human rows measured the body's own peptide after vitamin D, probiotics or infection. Three gave it, none by injection.
  2. Carrying laboratory biofilm results into people. LL-37 disrupts biofilms in a dish. No human biofilm infection has been treated with it in a study.
  3. Ignoring the disease-driver biology. The peptide vendors sell for 'immune support' is a known driver of psoriasis, rosacea and lupus when the body makes too much of it.
  4. Quoting a subcutaneous dose. The human trials used a gel, a cream and a bacterium. The microgram figures come from forums.
  5. Calling the wound trial positive. HEAL LL-37 missed its primary endpoint; the positive finding was a post hoc subgroup of large wounds, and the programme's next step is undocumented.
  6. Reading 'die-off' as efficacy. Feeling worse after injecting a membrane-active inflammatory peptide has a simpler explanation than dying bacteria.

Editorial synthesis Written from general pharmacology, the development record and what is commonly reported; not a cited finding. Cited findings on this page carry their own tier.

LL-37 vs thymosin alpha-1, KPV, BPC-157 and antibiotics

LL-37 beside the compounds it is stacked or compared with, from their own records where one exists.
CompoundWhat it isHuman evidenceStatus
LL-37The human cathelicidin antimicrobial peptideThree trials: two topical wound trials (one negative), one oral bacterium for COVID-19; nothing injectedNot approved; topical programme reached phase 2b; research chemical
Thymosin alpha-1Thymic immune-regulating peptide; 1,079 indexed publications65 randomized trials; approved abroad for hepatitisApproved in about 35 countries; not in the US
KPVThree-amino-acid anti-inflammatory tail of α-MSH; 138 indexed publications2 human studies; mouse and cell work onlyNot approved; research chemical
BPC-157Gastric repair pentadecapeptide; 228 indexed publications2 small uncontrolled human studiesNot approved; WADA S0
AntibioticsSmall molecules that kill or stop bacteriaSeventy years of randomized trials for every common infectionApproved; the standard LL-37 is sold against

The functional-medicine stack of LL-37 with KPV and thymosin alpha-1 puts an antimicrobial, an anti-inflammatory and an immune regulator together on the theory that they cover each other's gaps. Only thymosin alpha-1 has trials, none of the three has a trial for the conditions the stack is used for, and no study has combined any two of them.

Editorial synthesis Written from general pharmacology, the development record and what is commonly reported; not a cited finding. Cited findings on this page carry their own tier.

Other compounds in its class

Same class in the registry: Thymosin alpha-1. Each row shows what that compound's own record states; nothing is inferred across rows.

2 compounds in the immunomodulators & host defense class. Each row states what that compound's own record says; nothing is inferred across rows.
CompoundTierPublicationsRCTsRecord
LL-37 Human clinical trial 2,826 38 draft
Thymosin alpha-1 Human clinical trial 1,079 65 draft

Regulatory status: an investigational wound drug abroad, a research chemical in the United States

Registry facts only. Jurisdiction-level status, compounding categories and sport prohibitions need their own cited documents before they appear here.

  • Not approved anywhere. Promore Pharma developed topical LL-37 as ropocamptide for hard-to-heal leg ulcers through phase 2b (HEAL LL-37, 2021), which missed its primary endpoint; the programme's status since is not documented in the ledger. An engineered oral bacterium producing LL-37 was tested in a single-centre Chinese COVID-19 trial. In the United States LL-37 is sold as a research chemical; it does not appear on FDA's 503A compounding lists as far as this record can determine, and a physician-reviewed page's statement that it is 'available as a compounded prescription' could not be verified. Not named on the WADA Prohibited List; as an unapproved substance it falls under S0.Editorial synthesis
    Editorial synthesis from general knowledge · primary document to be added to the ledger

Open questions and limitations

What is genuinely unknown, stated plainly. A missing field is not evidence that the outcome is safe, effective, or established.

  • No study in any species in the ledger has injected LL-37 or measured its pharmacokinetics, the route the product is sold for.
  • Twenty of the ledger's twenty-three human rows measured endogenous LL-37 rather than administering it; the tier reflects trial design, not treatment evidence.
  • The status of Promore Pharma's ropocamptide programme after the 2021 phase 2b, and the psoriasis and rosacea mechanism papers, are not in the ledger.

Questions people ask

What is LL-37?

LL-37 is a 37-amino-acid peptide the human body makes itself, the only member of the cathelicidin family of antimicrobial peptides in humans. It is cut from a precursor protein (hCAP18) in neutrophils, skin and the lining of the gut and airways, kills bacteria by punching holes in their membranes, and signals to immune cells. As a drug it has been tested as a wound gel (ropocamptide) and as an engineered oral bacterium; the injectable vials sold online have never been studied.

Does LL-37 cause or treat psoriasis?

LL-37 is part of what drives psoriasis, not a treatment for it. In psoriatic skin it is overproduced, binds to the body's own DNA and RNA, and turns those complexes into triggers for the interferon-producing cells that inflame the plaque; it acts similarly in rosacea and lupus. No study has given LL-37 to a person with psoriasis, and the mechanism argues against it.

Does LL-37 break biofilms?

In the laboratory, yes: LL-37 inhibits biofilm formation and disrupts existing biofilms of several bacteria at concentrations that do not kill the free-living cells, and that property is well documented in reviews. No human study has tested it against a biofilm infection, and the injectable route has no study at all; the biofilm claim on vendor pages is a laboratory result carried to a use it has never been tried in.

What is the half-life of LL-37?

Short and unmeasured for the injected route. LL-37 is degraded quickly by proteases in blood and tissue, which is one reason the drug programmes chose topical and oral-bacterial delivery; no human pharmacokinetic study exists, and no abstract in the ledger states a figure.

Can LL-37 help with weight loss?

No study has looked, in any species, and nothing in its biology points that way. LL-37 is an antimicrobial and immune-signalling peptide; the weight-loss question comes from its being sold beside peptides that are.

Is LL-37 more effective than antibiotics?

No trial has compared them. LL-37 kills a broad range of bacteria in the laboratory by disrupting membranes, a mechanism bacteria find harder to resist, which is why drug developers are interested in it and its analogues. In people it has been tested only as a wound gel, where the endpoint was healing, not infection, and as an oral bacterium for a virus. Antibiotics have seventy years of trials; LL-37 has three.

Has LL-37 been tested in humans?

Three times as a treatment: a 148-person phase 2b trial of topical LL-37 gel on venous leg ulcers, which missed its primary endpoint; a small randomized trial of LL-37 cream on diabetic foot ulcers, which improved a granulation index over four weeks; and a 238-person open-label trial of an oral bacterium engineered to secrete LL-37 in COVID-19, which shortened viral clearance. The other twenty human studies in the ledger measured the body's own LL-37 and gave none. No study has injected it.

Is there an LL-37 dose?

For a wound gel, 0.5 or 1.6 mg/mL applied at dressing changes; for a cream, twice a week for four weeks; for the oral bacterium, a probiotic dose. For injection, none: no study in any species in the ledger has injected LL-37, so the microgram figures on vendor pages have no study behind them.

What are the side effects of LL-37?

The topical trials called it well tolerated at both strengths and the oral trial saw no severe events. For injection there are no data. The biological caution is that LL-37 damages human cell membranes at high concentrations and drives inflammation in psoriasis, rosacea, lupus and nasal polyps when the body overproduces it.

Does LL-37 cause psoriasis?

It is part of the mechanism. In psoriatic skin LL-37 is overproduced, binds the skin's own DNA and RNA, and turns those complexes into a signal that plasmacytoid dendritic cells read as infection, releasing interferon and inflaming the plaque. Similar processes operate in rosacea and lupus. No study has given LL-37 to anyone with these diseases.

Does LL-37 kill biofilms?

In the laboratory it inhibits biofilm formation and disrupts existing biofilms of several bacteria at concentrations below those that kill free-living cells. No human biofilm infection has been treated with LL-37 in a study, and the injected route has not been studied at all.

Is LL-37 FDA approved?

No, and it is not approved anywhere. Promore Pharma took a topical form, ropocamptide, through phase 2b for leg ulcers; the trial missed its primary endpoint. In the United States it is sold as a research chemical.

What is the difference between LL-37 and cathelicidin?

Cathelicidins are a family of antimicrobial peptides found across mammals; humans have one gene, whose precursor protein is hCAP18, and LL-37 is the 37-amino-acid active peptide cut from it. In humans the terms are used interchangeably; in other species 'cathelicidin' means a different peptide.

Can LL-37 be taken as a nasal spray?

No study has tested it. The one relevant human finding runs the other way: LL-37 drives neutrophil extracellular traps in nasal polyps, where it is part of the inflammation. Vendor nasal sprays are untested products aimed at the tissue where the peptide was shown to inflame.

Does vitamin D increase LL-37?

Yes; vitamin D switches on the cathelicidin gene, which is why many trials in the ledger measured LL-37 after vitamin D supplementation. Results are mixed: some found higher LL-37 or antimicrobial activity, others no change in the outcome that mattered. Those are studies of the body's own LL-37, not of giving it.

Is LL-37 the same as thymosin alpha-1 or KPV?

No. Thymosin alpha-1 is a thymic immune regulator approved abroad for hepatitis; KPV is a three-amino-acid anti-inflammatory fragment of α-MSH with mouse data only; LL-37 is the body's antimicrobial cathelicidin. They are stacked together in functional-medicine protocols on theory, and no study has combined any two of them.

What is ropocamptide?

The drug name for topical LL-37 developed by Promore Pharma for hard-to-heal venous leg ulcers. Its phase 2b trial, HEAL LL-37, found no overall healing benefit in 148 patients but a post hoc benefit in large wounds; the programme's status after 2021 is not documented in the ledger.

Why is LL-37 sold for Lyme disease?

Because it is antimicrobial in the laboratory and Lyme disease is bacterial, and because functional-medicine protocols adopted it for 'biofilm' infections. No study has tested LL-37 against Borrelia in a person or animal in the ledger, and the injected route has never been studied in anyone.

Sources

Full citations. Every claim above links to one of these by its id.

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10
  11. 11
  12. 12
Show the remaining 20 sources
  1. 13
  2. 14
  3. 15
  4. 16
  5. 17
  6. 18
  7. 19
  8. 20
  9. 21
  10. 22
  11. 23
  12. 24
  13. 25
  14. 26
  15. 27
  16. 28
  17. 29
  18. 30
  19. 31
    Antibacterial peptides: basic facts and emerging concepts.
    pmid-12930229 · · peer-reviewed
  20. 32
    Cathelicidins, multifunctional peptides of the innate immunity.
    pmid-12960280 · · peer-reviewed

Reference card

Reference card · generated /compounds/ll-37
Compound
LL-37, immunomodulators and host defense
Evidence tier
Human clinical trial
Indexed publications
2,826 · 38 RCTs · 23 other clinical trials
Approval
no registered development programme found
Routes reported
oral, topical
Reviewed
Adam Mirando, PharmD,

Study figures are as published in each source and are not recommendations. Print or save this card with its date.

Last reviewed and what changed

Newest first. These are the record's own revision dates, and the same dates feed the sitemap.

  1. · Reviewed by Adam Mirando, PharmD, on 2026-09-25.
  2. · Written under the sequencing rule after research/intents/ll-37.json: guide (8 sections incl. a given-versus-measured trial table), FAQ to 12 plus 8 from the map, dose, duration, timeline and adverse-event claims from the three administration trials, mechanism (defence and disease driver), reported-use, regulatory; twenty evidence rows relabelled as studies that measured LL-37 rather than gave it; misdrafted interactions removed; intent-driven H1 and title. Triage: escalation claims drawn from an in vitro calorimetry paper ('isothermal titration') removed.
  3. · Claims drafted extractively from 32 ledger sources by scripts/draft_claims.py: 22 claims, 25 evidence-table rows. Status researched -> draft.
  4. · Claims drafted extractively from 32 ledger sources by scripts/draft_claims.py: 29 claims, 25 evidence-table rows. Status researched -> draft.
  5. · Evidence fetched from Europe PMC and ChEMBL by scripts/fetch_evidence.py; claims not yet written.