The risk profile of exogenous LL-37 is unusually poorly characterized for a compound this widely circulated. The known hazards are mechanistically credible and the unknown ones are large: the peptide damages human cell membranes at concentrations close to those that kill bacteria, it is an established autoantigen in psoriasis and lupus, and no route, dose, or duration of systemic use has ever been established as safe in people. Every controlled human trial on record has been topical.
No route, dose, or duration of systemic LL-37 has ever been established as safe in humans, and the entire controlled human record consists of topical application to hard-to-heal venous leg ulcers.
Selectivity in LL-37 is a matter of charge and cholesterol, not of any recognition system, which makes it a gradient rather than a switch. Bacterial membranes present anionic phospholipids that draw a peptide carrying a net charge near +6, while mammalian plasma membranes present mostly zwitterionic phosphatidylcholine and contain cholesterol that stiffens the bilayer and resists insertion. The preference is real and it is narrow, and ordinary variation in local tissue concentration after a bolus can span it on its own.
| Property | Bacterial membrane | Mammalian membrane |
|---|---|---|
| Outer-leaflet charge | Anionic, with LPS or teichoic acid | Zwitterionic phosphatidylcholine |
| Cholesterol content | Absent | Present, stiffens the bilayer |
| Reported effective range | 0.18 to 1.5 micromolar in vitro | DNA fragmentation at 6 to 20 micromolar |
| Under 100 to 150 mM NaCl and serum | Potency suppressed, sometimes tenfold | Little protective effect |
Reported minimal effective concentrations sit near 0.26 micromolar for Escherichia coli, 0.18 micromolar for Pseudomonas aeruginosa, and 1.5 micromolar for Staphylococcus aureus under dilute low-salt assay conditions, while significant DNA fragmentation in mammalian cells has been recorded at 6 and 20 micromolar.
This is the best-documented reason for caution and the one least appreciated by people who know LL-37 only as an antimicrobial. Self-DNA released from dying cells is normally invisible to the immune system, and LL-37's cationic charge condenses it into nuclease-resistant aggregates that plasmacytoid dendritic cells take up efficiently and read through TLR9 and TLR7, driving sustained type I interferon production. For a person already carrying an interferon signature, the mechanism of the peptide and the mechanism of the disease are the same mechanism.
LL-37 condenses self-DNA and self-RNA into nuclease-resistant complexes that activate plasmacytoid dendritic cells through TLR9 and TLR7, the mechanism established in Nature in 2007 as an initiating step in psoriasis and documented again in systemic lupus erythematosus.
There is remarkably little, and all of it topical. Two controlled trials in venous leg ulcers both reported the peptide well tolerated, yet neither was built to detect the harms that matter for systemic use: autoimmune induction over months, tumor promotion over years, cumulative injection-site pathology, or antibody responses against an endogenous human sequence. The absence of reported harm there reflects the absence of looking.
The controlled human safety record for LL-37 consists of two topical venous leg ulcer trials in 34 and 148 patients, and the larger Phase IIb study found no significant healing benefit across its full study population.
LL-37 does not have one effect on cancer. It has opposite effects in different tissues, which is worse than a straightforward hazard because it removes any general rule to reason from. Almost all of the evidence sits in cell lines and xenograft models, with human data limited to expression studies in resected tissue that establish association rather than causation, so no magnitude of risk can be stated for a person taking the peptide.
| Dimension | Ovarian, lung, breast, melanoma | Colon and gastric |
|---|---|---|
| Tumor expression | Markedly overexpressed versus normal tissue | Falls as disease progresses |
| Direction of effect | Proliferation, migration, invasion, angiogenesis | Restoration suppresses tumor growth in models |
| Signaling implicated | Formyl peptide receptor 2, EGFR transactivation, MAPK, PI3K/Akt | Protective role in gut epithelium |
| Evidence level | Cell lines, xenografts, human expression studies | Cell lines and animal models |
LL-37 is overexpressed and growth-promoting in ovarian, lung, and breast carcinoma through formyl peptide receptor 2 signaling with EGFR transactivation, while cathelicidin expression falls and appears protective in colon and gastric tissue, so a systemic dose carries no single predictable direction of effect.
No drug regulator has approved LL-37 for any indication, in any formulation, anywhere. That absence means no approved label, no approved dose, no approved route, and no manufacturer carrying regulatory responsibility for the material's quality or its effects. What follows is an accountability gap rather than a paperwork gap.
LL-37 holds no marketing authorization from any major regulator and cannot lawfully be compounded under 503A or 503B, and the FDA's April 2026 move of cathelicidin LL-37 to its nominated-but-withdrawn list is not a safety clearance, since the agency still publishes the risks it identified, including protumorigenic activity and detrimental effects on male reproduction.
The injection-site reaction reported after a cationic peptide is not an allergy, which is why it can appear on the very first dose. LL-37 is a potent agonist at MRGPRX2, a G protein-coupled receptor on connective-tissue mast cells that responds to basic, amphipathic ligands by charge rather than by antigen recognition. Skin-prick and specific-IgE testing cannot predict who reacts, because the pathway those tests interrogate is not the pathway involved.
LL-37 activates connective-tissue mast cells through MRGPRX2 by charge rather than by antigen recognition, an IgE-independent pathway that can produce wheal, flare, and itch within five to twenty minutes of a first dose and that skin-prick or specific-IgE testing cannot predict.
The clearest exclusion the mechanism predicts is a diagnosed autoimmune or interferon-driven inflammatory condition, because for those patients the mechanism of the peptide and the mechanism of the disease are the same mechanism. The list below inverts how contraindication lists are normally read: with an approved drug, a short warning section reflects a large body of study that found few problems, while with LL-37 a short list reflects the absence of study.
Mechanism alone predicts exclusion for diagnosed autoimmune or interferon-driven disease, cancer history or unresolved findings, pregnancy and breastfeeding, mast cell disorders, and concurrent immune-modulating therapy, and that list is short because the study is absent rather than because the risks were investigated and found few.
Even if the peptide itself were benign, the vial would not be. Solid-phase synthesis of a 37-residue sequence is a long chain of coupling steps, and purification to 95 or 98 percent by HPLC still leaves several percent of the mass as deletion sequences, truncated chains, and side-chain-modified variants whose biology nobody has examined.
Because LL-37 binds lipopolysaccharide avidly, an endotoxin-contaminated preparation is not a peptide plus a separate contaminant but a peptide already complexed with an immune trigger, and research suppliers operate under none of the sterility, potency, or beyond-use-dating requirements that bind licensed compounders.
Route determines exposure, and exposure decides which of LL-37's liabilities actually come into play. Topical application to a wound is the most defensible use and the only one with meaningful human data behind it, because it concentrates the peptide where a high local level is physiologically normal and keeps systemic absorption low. Frequency compounds all of it: a single topical application resolves, while repeated injection into the same tissue plane asks a question about cumulative local pathology that no one has answered.
| Criteria | Topical to wound | Subcutaneous | Inhaled or nebulized |
|---|---|---|---|
| Human data | Two controlled ulcer trials | None | None |
| Local concentration | High where physiologically normal | High and uncontrolled, in mast-cell-dense tissue | Direct contact with susceptible airway epithelium |
| Systemic exposure | Low | Low and brief, via apoA-I binding and proteolysis | Uncharacterized |
| Observed ceiling | 3.2 mg/mL lost the benefit lower concentrations showed | Not established | Not established |
Topical wound application is the only route of LL-37 supported by controlled human data, while subcutaneous injection creates a high uncontrolled depot concentration in tissue dense with connective-tissue mast cells and apolipoprotein A-I binding with rapid proteolysis leaves systemic exposure both low and brief.
Educational use only. This article describes what the published scientific and clinical literature reports about LL-37. It is not medical advice, and it does not recommend, prescribe, or tell anyone to use anything described here. The regulatory status shown at the top of this page reflects what the record showed on the date given there and can change. mdpep.com does not sell any substance described here, does not endorse human use of it, and does not direct anyone to obtain it.
This is not guidance for your situation. Nothing here accounts for your medical history, your current medications, or anything else specific to you, and none of it should be used to make a decision about your own health.
Affiliate disclosure. Some links on this site are affiliate links, and mdpep.com may earn a commission when they are used. That never affects what is written here, it is not an endorsement of any vendor, and it is not a statement that anything described on this page is available to buy.
Every claim here ties to a named primary source and a date, written as plain documentation with nothing for sale. That is how MD PEP covers the whole peptide market.
