LL-37: The Definitive Guide

What is LL-37?

LL-37 is a small protein fragment, a peptide of 37 amino acids, that the human body makes to help fight germs. Lab-made LL-37 is sold online and discussed for infections, gut problems, wounds and skin conditions, but it has never been tested in people as the under-the-skin injection that sellers describe.

Amino acids are the building blocks of proteins, and a peptide is a short chain of them. The name describes the chain: it starts with two leucines (the amino acid with the one-letter code L) and has 37 amino acids in total. LL-37 is the only cathelicidin people make. It is cut from a larger protein called hCAP18, and white blood cells called neutrophils, skin cells and the cells lining the airways and gut release it when they meet germs or injury. Its official drug name is ropocamptide. [1], [2], [3], [4], [5]

The key term is investigational: researchers are still working out what it can do and what its risks are. Swedish companies developed it as a wound treatment and ran two skin-ulcer trials, but in 2023 the developer's board said it had limited ability to raise money for further work. The LL-37 sold online is the same chain made in a lab, usually as a powder for injection. [6], [7], [8]

Typical LL-37 Protocols

The most common protocol in the reviewed sources is 100–200 mcg once a day by subcutaneous injection for 2–4 weeks, followed by a break. [9], [10], [11], [12]

Titration Changing the amount in steps

The usual approach starts at 50 mcg a day for 3–5 days, moves to 100 mcg, and goes up to 200 mcg only if the lower amount is tolerated. The reason given is to avoid flu-like symptoms and strong injection-site reactions in the first days. A widely copied table instead climbs by 50 mcg each week, from 50 mcg in week 1 to 400 mcg by week 8. [9], [13], [14]

Breaks and cycles

After a 2–4-week course, sources describe about 2 weeks off, or a break as long as the course. Some also use a weekly pattern of 5 days on and 2 days off. [10], [11], [15]

Read the units: mcg means micrograms; 1 mg equals 1,000 mcg. Amounts refer to LL-37 itself, not the total weight of a blend. The wound trials measured LL-37 as a strength, in mg per mL of solution spread on an ulcer. A strength is not an amount per injection, so those numbers cannot be compared with this table. [6]

Each row is a complete schedule as published, so its amount and duration belong together. Low, mid and high are alternatives, not steps to move through. These schedules come from protocol sources rather than clinical trials. No study has given LL-37 to people by subcutaneous injection, so none of these amounts has been tested for safety or benefit; the cited references give the full details. [9], [16]

Cautions for every schedule above. No human safety data exist for any of these schedules. In lab and animal studies, LL-37 is a target of the immune attack in psoriasis, causes rosacea-like skin inflammation through mast cells, switches mast cells on rather than calming them, and helped some tumour types grow. That makes it a particular concern for people with psoriasis, lupus, rosacea or another autoimmune or inflammatory skin condition, a mast-cell condition or a history of cancer. It has not been studied in pregnancy, breastfeeding, people trying to conceive or anyone under 18, and it has not been shown to treat infections, so it should not delay proven care. In community reports, the most common problems are reactions at the injection site (redness, itching, swelling or burning) and feeling tired, achy or flu-like in the first few days. No published study has tested LL-37 bought online for identity, purity or sterility. See the full risks and who should be cautious. [17], [18], [19], [20], [16], [21], [22], [23]

Where protocols differ: titration, course length, breaks and the trial amounts ↓

What is LL-37 commonly used for?

People usually look into LL-37 because of an infection or inflammation that keeps coming back: frequent colds, a stubborn gut problem, a skin condition or a long illness they believe involves hidden infection. It is often described online as a "natural antibiotic". [26], [27]

That label mixes two different facts. Your body's own LL-37 does help defend the skin and other surfaces, and lab-made LL-37 kills bacteria in a dish. Neither fact shows that injecting it treats an infection in a person, and no study has tested that. The sections below separate what people hope for, what they report and what research has actually measured. [28], [29], [16]

Colds, flu and other infections

This is the most common reason people give. Protocol sources describe short courses during an active infection, sometimes combined with thymosin alpha-1, another peptide sold for immune support. Community accounts describe fewer or shorter colds and a tooth infection that hurt less within hours. [9], [30], [21], [31], [32], [22]

Those stories come with major limits. Colds usually get better on their own, several people were also taking antibiotics or other peptides, and nobody measured the infection. In a dish, blood serum weakens LL-37's germ-killing effect, and some bacteria make enzymes that break it down. A dish result is not an infection cured. [29], [33], [34]

The only human infection trial did not use the peptide at all. It gave COVID-19 patients capsules of live bacteria engineered to make LL-37, and it cannot tell us what injected LL-37 does. [35]

Gut problems: SIBO, IBS and "gut healing"

Gut problems come up often, especially small intestinal bacterial overgrowth (SIBO), irritable bowel syndrome (IBS), constipation and suspected candida. Some community accounts describe less bloating, better bowel habits or a "less reactive" gut. Others describe no change, or loose stools and an upset stomach after the first dose. [36], [37], [38], [39]

No human gut study of LL-37 exists. In one mouse study of chemically induced colitis (bowel inflammation), LL-37 injected into the abdomen lowered damage scores in a preliminary experiment, but the main experiments tested a smaller fragment instead. [40]

Lyme disease, mould illness and long-lasting symptoms

People with Lyme disease, illness they link to mould exposure, or chronic fatigue discuss LL-37 because of lab findings that it disrupts biofilms, slimy layers that protect bacteria. That biofilm finding comes from one type of bacteria in a lab dish. The community reports in this area almost always involve many treatments at once, so LL-37's own contribution cannot be picked out. [28], [41], [42], [43]

Wounds and leg ulcers

This is the one use that has been tested in people, and only on the skin. Three small-to-medium trials spread LL-37 solution or cream on hard-to-heal ulcers. The first found faster healing at the lowest strength; the largest, with 148 people, found no more ulcers closed than with placebo. See the trial results. [44], [6], [45]

Acne, eczema, rosacea and other skin problems

Here the evidence points in two directions. LL-37 helps skin cells move to close a wound in lab tests. But LL-37 injected into mouse skin is a standard way for researchers to cause rosacea-like redness and inflammation, and it is a target of the immune attack in psoriasis. In community education groups, one member described fewer acne breakouts; in a public forum, one person with psoriasis stopped after three days because of a severe flare of joint pain. [46], [18], [17], [47]

Other uses you may come across

Other claims include herpes outbreaks, mast-cell problems, cancer and a general "immune boost" before travel. The cancer research is mixed and mostly in cells and animals; the mast-cell claim runs opposite to the lab evidence, which shows LL-37 activates mast cells. How LL-37 might work covers both. [19], [48]

Where do LL-37 protocols differ?

The recurring amount is easy to summarize. The details around it are not: sources disagree on how to start, how many weeks to use it, whether to take breaks and what it is for. Here is what those differences mean when reading a protocol.

Is the number for one dose or the whole day?

Most LL-37 schedules are once daily, so the amount each time and the daily total are usually the same. That changes when a page says twice daily: 100–200 mcg twice daily, as one calculator page describes, means 200–400 mcg for the day. [49]

Blends add another problem. Some people mix LL-37 into one vial with five or six other peptides, so the LL-37 content of each injection is often unclear. A blend's total weight is not its LL-37 amount. [50], [51]

Every day, five days a week or a few times a week?

All three appear. Most pages describe once daily; others describe 5 days a week, three times a week, or every other day. Splitting the same weekly amount differently has not been shown to matter. [10], [52], [53]

Do sources start low, or build up week by week?

Two opposing strategies appear. One starts at 50 mcg for 3–5 days and stops at 200 mcg. Another climbs 50 mcg every week to 400 mcg by week 8, and at least two other pages reproduce that table. A third page describes starting at 200 mcg, moving to 500 mcg three times a week, then tapering. [9], [13], [14], [54], [55]

Those are different examples, not interchangeable pieces of one protocol. The main reason given for starting low is to reduce flu-like symptoms and injection-site reactions in the first days, which community reports describe often. There is no evidence that a slower start works better. [9]

How many weeks, and what happens between cycles?

A course is the period of repeated use; a cycle is that course plus a planned break. The most common course is 2–4 weeks, but the range is wide: 50 days in a row followed by 4 weeks off, 8–12 weeks, or one month on with several months off in a single clinic handout. One sheet simply says to stop when you feel healthy. [56], [13], [57], [26]

Community education groups mostly frame LL-37 as a short course for a specific infection, with a break afterwards, rather than something to take all the time. In the same groups, several members describe using it for two to nine months, well past what their teachers describe. One protocol page says plainly that no validated human cycle length exists. [15]

Where do the sources disagree most?

• Daily amount: most pages fall between 50 and 500 mcg, with some as low as 50–100 mcg a day, but one vendor site describes 2–5 mg a day and a ladder rising to 3–5 mg a day for Lyme disease. The same site claims no serious side effects at doses up to 10 mg a day; no human injection study exists to support that. [58], [24], [25]
• What the trials showed: several pages say clinical trials use subcutaneous injection, or that the melanoma trial showed LL-37 is safe. Neither is correct: the trials were on wounds and inside tumours. [59], [60], [61]
• Course length: short courses of 1–2 weeks sit alongside 8–12-week ones. Longer use has not been tested at all. [13], [30]
• Inconsistent pages: one calculator page contradicts its own concentration arithmetic, and one clinic handout gives two different break lengths. Check whether each number is per injection or per day. [49], [57]

What amounts did the human studies use?

The studies used very different amounts and routes from the protocols above. They are shown here so you can see that difference, not as schedules to follow.

In the leg-ulcer trials, the strength that did best was the lowest one, and the strongest solution did no better than placebo. More was not better. [44]

Do nasal sprays and creams use the same amounts?

No. One page describes a nasal spray of 25–50 mcg in each nostril once or twice a day. Homemade creams and gels are described at strengths from 0.1 to 2 mg per mL, usually without saying how much to apply. Several pages repeat the trial strengths (0.5 and 1.6 mg per mL) as if they were doses, and one lists a daily foot-ulcer schedule the trial never used. None of these has been tested in people outside the wound trials. [9], [62], [63], [64], [45]

How do micrograms convert to syringe units?

Many people meet "units" on an insulin syringe. A U-100 syringe has 100 units per millilitre, so units measure volume, not an amount of LL-37. How much LL-37 each unit holds depends on how much liquid was added to the vial.

For example, 5 mg (5,000 mcg) mixed into 2.5 mL gives 2,000 mcg per mL, or 20 mcg per unit, so 100 mcg would be 5 units. Mixed into 5 mL instead, the same 100 mcg would be 10 units. Because LL-37 doses are small, a unit or two of error is a large share of the dose, which is why a number of units copied from someone else's protocol can be wrong. [11] Doserly's reconstitution calculator works this out for any vial and shows each step.

What happens when you stop LL-37?

This has not been studied. No study or protocol source describes withdrawal effects or a rebound after stopping, and most protocol pages simply end a course without tapering. That is an absence of data, not proof that none exist. [15], [62]

• After the wound trial. People in the largest trial were asked to return 8 and 16 weeks after treatment ended, but only 63.5% attended at least one visit. Of 39 people whose ulcers had healed and who came back, 6 had ulcers that reopened: none on the lower strength, 2 on the higher strength and 4 on placebo. The authors said attendance was too low to draw conclusions, and reported no safety concerns during follow-up. [6]
• After the melanoma injections. One patient who developed wart-like skin growths during weekly injections into her tumours saw all of them clear within two months of stopping. [65]
• In animal skin. In mice given LL-37 into the skin for 2 days, rosacea-like lesions faded after stopping; after 10 days of injections, they did not fully recover. [66]
• Community accounts point both ways. One person's tender underarm lymph glands settled 7–10 days after stopping, and another's itchy injection-site swelling went away. One person described partial improvement that faded after stopping. Two people described feeling unwell or inflamed for months to years after a course; those accounts cannot show what caused the symptoms. [21], [23], [67], [68]

The breaks that protocols describe are a convention, not a response to known withdrawal effects.

Do injected, topical and other forms of LL-37 behave the same way?

No, and only two routes have ever been tested in people. Topical use means applying LL-37 to a surface such as an open ulcer. Intratumoral means injecting it straight into a tumour. A subcutaneous injection places it in the fatty layer under the skin, from where some of it may reach the blood. Each route starts in a different place and exposes different tissues.

Doserly

Where each LL-37 amount goes

Human trials and protocol sources; not a comparison of results

1. Topical (on an ulcer)

   Where it acts

   Directly on the wound bed

   Amount

   Strength: 0.5, 1.6 or 3.2 mg per mL solution, or cream 0.5 mg per g; twice weekly

   Tested in people?

   Yes: 3 wound trials

2. Intratumoral (into a tumour)

   Where it acts

   Inside melanoma skin tumours

   Amount

   250 or 500 mcg per tumour, weekly, up to 8 weeks

   Tested in people?

   Yes: 3 treated in one small trial

3. Subcutaneous injection

   Where it acts

   Fatty layer under the skin; how much reaches the blood is unknown

   Amount

   Protocol sources: 50–200 mcg per injection, once daily

   Tested in people?

   No

4. Nasal spray

   Where it acts

   Lining of the nose

   Amount

   Protocol source: 25–50 mcg per nostril

   Tested in people?

   No

5. Oral

   Where it acts

   Gut

   Amount

   No tested peptide capsule; one COVID-19 trial used live bacteria engineered to make LL-37

   Tested in people?

   Peptide not tested; a different product was tested

Conceptual comparison · Human research and protocol sources

Only on-the-ulcer and into-the-tumour use have been tested in people. The injection amounts protocols describe have not, and the units cannot be converted into one another.

The wound trials delivered LL-37 directly to the tissue they were trying to heal. The melanoma trial did the same with tumours. Protocol pages instead use subcutaneous injections to aim for an effect elsewhere in the body, such as the gut or sinuses. Whether any intact LL-37 reaches those places after a subcutaneous injection has never been measured. [6], [61], [9]

There are also reasons it might not. In lab tests, blood serum and a blood protein called apolipoprotein A-I weakened LL-37's activity, and enzymes made by bacteria cut it apart. Bioavailability, how much of a dose reaches the blood in an active form, is unknown for every route. [33], [69], [70]

How long does LL-37 last in the body?

Nobody knows for people. Half-life is the time it takes for a measured level in the blood to fall by half, and no study has measured it after LL-37 was given to a person by any route. [16], [6]

The two published figures are indirect. In mice given an injection into a vein, the level of material recognised as LL-37 fell by half in about 3.4 days, but that test also counts peptide stuck to blood proteins and fragments. In a lab test with rat blood plasma, LL-37 lasted about 4 hours (242 minutes) before half of it was broken down. Websites quote figures from "30 minutes" to several hours; none traces to a measurement in people. [71], [72], [55], [9]

A half-life also would not tell you how long any effect lasts. Dose, half-life and benefit are separate questions.

Is the LL-37 sold online the same as the trial product?

The amino-acid chain is meant to be the same, but the product can differ in ways that matter. Peptides are sold as salts: the trial used an acetate salt, while research suppliers mostly sell a trifluoroacetate (TFA) salt. In a lab comparison, the salt form changed LL-37's germ-killing activity and its toxicity to cells. A slightly different chemical form, LL-37 amide, also exists, and even the first trial's registry record lists the amide's chemical number next to the ordinary chain. [6], [73], [74], [75], [76]

Should LL-37 be injected near an infection or wound?

Some pages say to inject close to the problem area. No human study has compared injection sites. In animal wound studies, LL-37 was applied to the wound or injected around it, and the results depended heavily on how it was packaged: in mice, adding more free LL-37 to a wound did not improve healing, while the same amount released slowly from tiny particles did. [77], [11], [78]

Is there an oral LL-37 capsule?

Not one that has been tested. Some pages give absorption percentages for LL-37 capsules; no published study supports them, and one protocol page says evidence-based oral dosing does not exist. The one human oral study used live bacteria engineered to produce LL-37 in the gut, which is a different product. [24], [79], [35]

How should LL-37 be stored, and how long does it last?

No one has published a stability study of the LL-37 people buy, so storage advice comes from research-supplier data sheets, trial records and protocol pages, not a tested medicine label.

• Powder, before mixing. Research suppliers say to keep the freeze-dried powder at −20 °C. One says its powder stays stable for at least 4 years at that temperature. These statements cover those companies' research products only. [74], [80], [81]
• After mixing. Sources conflict. Research suppliers say to freeze mixed (reconstituted) solutions in portions, with one giving 1 month at −20 °C and 6 months at −80 °C, while another advises against keeping its water-based solution for more than a day. Protocol pages instead say to refrigerate at 2–8 °C and use within 14 days, 2–3 weeks or 28 days, and one says never to freeze it. [82], [81], [74], [11], [10], [49], [55]
• In the trials. The largest wound trial kept its product at 2–8 °C and used each mixed solution within 3 hours. A foot-ulcer cream stayed stable at 4 °C and 28 °C but melted at 40 °C. [83], [45]

Storage is not only about strength. In solution, LL-37 molecules tend to stick together into clusters (aggregation), which changes how they behave, and FDA lists impurities and immune reactions among its concerns for compounded LL-37. This guide does not cover mixing or injection technique; the Academy note below covers practical handling. [84], [85], [16]

What do we actually know about LL-37 in people?

Every human study that gave LL-37 itself used it on a wound or inside a tumour. None gave it by subcutaneous injection, into a vein or into a muscle, which are the routes protocol pages describe. That gap shapes everything below. [6], [61]

Feeling better after a treatment is real to the person experiencing it. Working out why they improved is a separate task. Wounds heal and infections clear over time, other treatments help, and expectations can influence symptoms. Controlled trials with a placebo group help separate those effects from the treatment itself.

What did the human studies find?

These were small studies, and three of them tested ulcers. The two leg-ulcer trials were sponsored by the companies developing LL-37. [6], [75]

Doserly

Ulcers fully closed within 13 weeks in the largest LL-37 trial

Human trial: 148 people with leg ulcers treated, 144 analysed, 2021

All patients (main result)

LL-37 0.5 mg/mL

26.5% n = 46

LL-37 1.6 mg/mL

24.7% n = 48

Placebo

25.3% n = 50

Chosen after the trial (post hoc): ulcers of at least 10 square cm

Post hoc: cannot prove a benefit

LL-37 0.5 mg/mL

28.1% n = 21

LL-37 1.6 mg/mL

19.6% n = 21

Placebo

8.1% n = 24

First trial (2014, 34 people): faster 4-week healing rate at 0.5 mg/mL; 3.2 mg/mL no different from placebo.

Human research · Estimated proportions from the published trial

About a quarter of ulcers closed in every group, including placebo. The better large-ulcer result came from an analysis chosen after the trial.

Why did the bigger trial disagree with the first one?

The two leg-ulcer trials asked different questions. The first measured how fast ulcers shrank over 4 weeks. The second asked whether ulcers closed completely within 13 weeks, which is what matters to a patient. Early shrinking did not turn into more closed wounds. [44], [6]

In HEAL, the researchers then looked separately at people whose ulcers were at least 10 square cm. In that group, 28.1% closed on the lower strength, 19.6% on the higher strength and 8.1% on placebo. The authors noted the trial was not designed to test that subgroup. A post hoc finding like this is a reason for a new trial, not proof. In mid-2023 the developer's board said it had limited ability to raise money for further development. [6], [7]

Does a wound-cream result apply to injections?

No. The wound trials put LL-37 straight onto the tissue at a set strength per area of wound. A subcutaneous injection delivers a small amount far from any target, and how much reaches the blood is unknown. Even on wounds, more was not better: the strongest solution did no better than placebo. [44], [6]

Has LL-37 been tested for infections in people?

Not as the peptide itself. The only human infection trial gave hospitalised COVID-19 patients capsules of live bacteria engineered to make LL-37 for 7 days. People who started early turned test-negative sooner, side effects were similar to the comparison group, and the amount of LL-37 each person received is unknown. It says nothing about injected LL-37 for colds, gut infections or Lyme disease. [35]

How might LL-37 work?

LL-37 does several different jobs in lab and animal studies. Some sound helpful and some are harmful, and which one happens depends on the tissue, the amount and the surroundings. That is why it is described in the research as double-edged. [4], [88]

How does it kill germs in the lab?

LL-37 carries a positive charge and folds into a spiral (a helix) that lines up along the membranes of bacteria and disrupts them like a detergent. In a lab dish it killed bacteria and stopped one species from forming a biofilm at amounts far below those needed to kill it. [29], [84], [28]

Those results are sensitive to conditions. Blood serum weakened its killing in lab tests, and bacteria such as Staphylococcus aureus and Pseudomonas make enzymes that cut LL-37 apart. In one study, bacteria briefly exposed to LL-37 became temporarily less sensitive to vancomycin, an important antibiotic. [33], [70], [34], [89]

How does it signal the immune system?

In studies of human immune cells in a dish, LL-37 attracted white blood cells through a receptor called FPR2 (also known as FPRL1), triggered release of an inflammatory signal called IL-1β, and, at low concentrations, dampened some cells' inflammatory response to bacterial products. In mice with blood infections (sepsis), a single small dose into a vein improved short-term survival, while in newborn rats a delayed, higher dose increased deaths. [90], [91], [92], [93], [94]

Does it help wounds heal?

In lab and animal studies it can. LL-37 made skin cells called keratinocytes move faster, and blocking the body's own LL-37 slowed skin repair in donated human skin. In rabbits with reduced blood flow to a leg it increased new blood vessel growth, and in mice with infected wounds it reduced bacteria but also increased scar-like tissue. [46], [5], [95], [96]

Why can the same peptide inflame the skin?

Mast cells are immune cells in the skin that release histamine and other chemicals. LL-37 switches them on through a receptor called MRGPRX2. Injected into mouse skin, it produces the redness and swelling of rosacea, and mice lacking mast cells did not develop those features. After 10 days of injections, the lesions did not fully clear. [19], [97], [18], [98], [66]

LL-37 can also carry fragments of the body's own DNA and RNA into a type of immune cell called a dendritic cell, making the immune system react as if it had found a virus. In psoriasis, about two-thirds of people with moderate-to-severe disease had immune cells that react to LL-37 itself, making it an autoantigen. Similar reactions to LL-37 have been found in lupus, psoriatic arthritis and rheumatoid arthritis. These findings concern the body's own LL-37, not an injected product, but they explain why autoimmune conditions come up as a caution. [99], [100], [17], [101], [102], [103]

Does it help or feed tumours?

Both, depending on the tissue. In cell and animal studies, LL-37 helped ovarian, lung and breast cancer cells grow or spread, while it slowed gastric and pancreatic cancer models. Colon cancer studies disagree: one found it killed colon cancer cells, another that it helped them grow. An earlier paper claiming it switches on the IGF-1 growth receptor was retracted. [20], [104], [105], [106], [107], [108], [109], [110]

Is it an "immune booster"?

That phrase oversimplifies it. The body raises its own LL-37 when needed, and vitamin D is one of the signals that switches the gene on in immune cells. In a small study of people with leg ulcers, higher blood levels of their own LL-37 went along with better healing. Neither finding shows that adding lab-made LL-37 from outside strengthens the immune system; the lab data above show it can also drive inflammation. [112], [113]

What are the risks and unanswered questions?

Is LL-37 safe?

Its safety as an under-the-skin injection has not been established, because it has never been studied that way; the only injection study put it into melanoma tumours in 3 people. The largest wound trial found side effects about as often as with placebo, but it involved a strength applied to an ulcer for weeks, not repeated injections. FDA says it lacks enough safety information to know whether LL-37 would cause harm when given to people. [6], [16]

"No serious side effects in a wound trial" is useful information. "Safe to inject" would need very different evidence.

What side effects have actually been reported?

• In the wound trials: in HEAL, 20 of 48 people on the lower strength, 24 of 49 on the higher strength and 20 of 51 on placebo had at least one adverse event. Six events in four people, one of them on placebo, were judged possibly related to treatment: wound infections, skin inflammation, leg swelling and, in the placebo group, an inflamed vein. None of the 12 serious events was judged related. One person using the cream in the foot-ulcer trial had mild skin irritation, which the researchers put down to fluid from the wound rather than LL-37. [6], [45]
• In the melanoma programme: the trial record lists no serious events, but lists a squamous cell skin cancer, a precancerous skin patch, anaemia, low white blood cell counts and an underactive thyroid among other events, without saying whether LL-37 caused them. One patient developed wart-like skin growths that cleared after stopping. [61], [65]
• In community accounts of injections: redness, itching, swelling, burning and bruising at the injection site; aches, joint pain, fatigue and flu-like feelings in the first days; loose stools; one case of facial swelling after two doses; tender lymph glands; and one psoriatic arthritis flare. These are self-reports with unverified products. [21], [114], [22], [115], [47], [39]

Could it trigger or worsen autoimmune or inflammatory skin disease?

This is the most specific concern for LL-37. It is an autoantigen in psoriasis, reactions to it are found in lupus, psoriatic arthritis and rheumatoid arthritis, and it causes rosacea-like inflammation in mouse skin. No study has tested what injected LL-37 does in people with these conditions. [17], [101], [102], [103], [18]

Community advice is split. Some people recommend LL-37 for autoimmune conditions, while community education groups that teach about autoimmunity advise avoiding it in psoriasis, lupus and rosacea. In one public account, a person with psoriasis stopped after three days because of severe joint pain they described as a psoriatic arthritis flare. [47]

Can the immune system react against injected LL-37?

Possibly, and nobody has checked. FDA lists possible immune reactions (immunogenicity) for some routes of LL-37 among its concerns. No trial has measured antibodies against LL-37 after giving it, even though some people with autoimmune disease already make them. [16], [102], [6]

Does LL-37 cause cancer?

Nobody knows. There are no long-term human data and no standard long-term animal cancer studies. The lab findings point both ways: LL-37 helped some tumour types grow and slowed others, and FDA notes findings suggesting it can promote tumours in some tissues. In the melanoma trial, one patient developed skin growths that cleared after stopping. [16], [48], [65]

Could it affect fertility?

It is possible, and untested. Outside the body, LL-37 stopped human and mouse sperm from moving, and mouse sperm treated with it failed to fertilise eggs. FDA mentions findings suggesting harmful effects on male reproduction. No study has looked at fertility after injections. [111], [116], [16]

Could it make an infection harder to treat?

Two concerns come from lab work: bacterial enzymes break LL-37 down, and brief exposure made one bacterium temporarily less sensitive to vancomycin. The bigger practical risk is delay. In community education groups, at least one person described using LL-37 in place of a prescribed antibiotic. LL-37 has not been shown to treat any infection in people. [34], [89]

Could product quality change the risk?

Yes. No published study has tested LL-37 bought online for identity, purity, sterility or bacterial toxins. FDA flags impurities and difficulty confirming the active ingredient, and Health Canada lists LL-37 among unauthorized injectable peptides it seized. The salt form and chemical form sold can also differ from the trial product. [16], [8], [73]

Is LL-37 legal? Is it FDA-approved?

Checked September 23, 2026. Rules differ by country and change quickly.

United States. LL-37 is not an FDA-approved medicine. The recent steps:

• 2020 and 2021. FDA warning letters to two compounding businesses named LL-37 among ingredients they should not have used. [117], [118]
• 2026. FDA's page on compounding ingredients that may present significant safety risks (content current April 22, 2026) lists cathelicidin LL-37 under nominations that were withdrawn, with its safety concerns still stated. The move followed the nominators' withdrawal, not a finding that LL-37 is safe. It does not appear on FDA's May 2026 list of ingredients under evaluation. [16], [119]
• Next. FDA's pharmacy compounding advisory committee will discuss LL-37, with GHK-Cu and three other substances, at a meeting before the end of February 2027; no date had been set. A committee vote is advice, not a final decision, and compounded products would still not be FDA-approved. [120], [121]

Canada. No authorized LL-37 product was found. Health Canada's August 2025 advisory lists LL-37 among unauthorized injectable peptides seized from a seller and warns they may pose serious health risks. [8]

Australia. LL-37 is not named in the June 2026 Poisons Standard, so it has no specific schedule. That does not make selling it for human use legal, and we could not directly search the register of approved products. [122]

Europe and the United Kingdom. The last LL-37 trial in the EU trials register is the HEAL wound study, which has posted its results. No authorized LL-37 product was found, but the official product databases could not be searched directly, so this is not a confirmed absence. [83]

A "research use only" label does not make a product legal to sell for human use.

Is LL-37 banned in sport?

Yes, in effect. LL-37 is not named on the World Anti-Doping Agency's list, but category S0 bans any substance not approved for human therapeutic use at all times, and the 2027 list, published on 21 September 2026, now names peptides as an example. USADA gives the same guidance for unapproved peptides. [123], [124], [125]

Who should be especially cautious?

These groups have no safety data in people. That is a reason for caution, not proof of harm.

• People with psoriasis, lupus, rosacea or other autoimmune or inflammatory skin conditions. Because of the autoantigen and mast-cell findings above. [17], [18]
• People with mast-cell conditions or frequent hives. LL-37 activates mast cells in lab studies, the opposite of claims that it calms them. [19]
• A history of cancer. Because the tumour findings point both ways and FDA notes possible tumour-promoting effects. [16]
• Pregnancy, breastfeeding and anyone trying to conceive. There are no human data, and the sperm findings above are unexplored. [111]
• Anyone with an infection that needs medical treatment. LL-37 has not been shown to treat infections, so it should not delay proven care.
• Under 18. No studies include young people. This guide is written for adults.
• Competitive athletes. Prohibited at all times under category S0. [123]
• People taking other medicines. Interactions have not been studied; see the next question.

Has LL-37 been tested alongside other medicines?

Not in people. No study has checked how injected LL-37 interacts with prescription medicines, immune-suppressing drugs or antibiotics. The lab finding on vancomycin concerns bacteria, not a drug interaction in a person. In the accounts reviewed for this guide, LL-37 was usually taken together with several other peptides, which makes any reaction harder to trace. Bring a complete list of everything you take to a pharmacist or doctor. [89], [23]

What should be monitored while using LL-37?

There is no official monitoring plan, because no regulator has approved LL-37 or written a label for it. What exists is a record of what the trials measured and the safety signals from laboratory research. It can help you prepare questions for a clinician; it is not a personal testing plan.

What did the human studies check?

• The HEAL wound trial checked the ulcer and surrounding skin for reactions at every dressing change, measured vital signs weekly and took blood for safety tests monthly, with follow-up visits 8 and 16 weeks after treatment. Blood values showed no meaningful differences between groups. [6]
• The melanoma trial took a complete blood count with a breakdown of white cells and blood chemistry (including kidney and liver markers) every week, clotting tests at the start, and photographs and biopsies of the injected tumours. [87]
• What no trial measured: LL-37 levels in the blood after treatment, and antibodies against LL-37. [6], [45]

Some protocol pages suggest blood counts, CRP (a marker of inflammation) and kidney and liver tests every few weeks. Those are suggestions from protocol sources, not requirements from a study. [10], [55]

Which signals matter, and why?

The skin and autoimmune signals come from lab and animal work and from uncontrolled reports. They show what researchers flagged, not what a person will experience. [18], [17], [22]

What do people in public communities report?

Public discussions contain improvements, no change and reactions that made people stop, often in the same thread. Almost everyone in these discussions combined LL-37 with other peptides, antibiotics or diet changes, and most used it for problems that often improve on their own. The most useful accounts say what else was happening and come back with an update. [36], [21]

What do the positive reports look like?

One person with lifelong constipation described better bowel movements on 100 mcg a day, and said it was not a long-term fix because LL-37 is not meant to be taken long term. Another, taking 100 mcg a day for 60 days for gut health, reported that a toenail fungus cleared. A person with herpes described fewer outbreaks after a few weeks of LL-37, but was also taking two other peptides, monolaurin and many supplements. [37], [51]

Others described less gut reactivity, shorter colds or relief from a gum infection within hours. These are encouraging stories, but none can separate LL-37's effect from time, other treatments and expectations. [36], [22]

How long did those changes last?

Follow-up comments often change the picture, but many accounts never get one: the constipation account gave no later update. In a Lyme disease forum, one person reported partial improvement over about three months and said they felt sick again whenever they went off it. One long-running gut account said the improvement persisted for months after stopping, but that person had finished an 8-month antibiotic course beforehand and was also taking another peptide. [37], [23], [126]

Does everyone describe an improvement?

No. Some people noticed nothing after six-week courses, including one person who took 100 mcg a day for suspected candida. A husband reported little success from his wife's use for colitis. [127], [126], [128]

Does feeling worse in the first days mean it is working?

The accounts do not support that conclusion. Feeling unwell early is one of the most common reports: aches and burning sensations on the first night, fatigue, and flu-like symptoms that eased when the dose was lowered. Many people call this "die-off" or a Herxheimer reaction, meaning germs dying and releasing toxins. [22], [23], [129]

No study shows that happens with LL-37. The same symptoms fit what lab studies show LL-37 can do directly: trigger inflammation and switch on mast cells. In community education groups, feeling rough for the first few days is described often enough that many people start low, and at least one person stopped on day nine because of nausea. Calling it a healing reaction adds an explanation the reports cannot establish. [19], [91]

What about unwanted effects?

Beyond the first days, people describe injection-site burning, itching and swelling; tender lymph glands; loose stools; a psoriatic arthritis flare; and one case of facial swelling severe enough to affect vision after two doses. Within community education groups, one person described rosacea-like spots after their first dose, and others described burning at the site lasting hours. These deserve attention as reported experiences, but they cannot show how often an effect happens or whether the peptide, the product or something else caused it. [21], [39], [47], [115]

Two people described feeling inflamed or unwell for months to years after using it, one of them across several threads. There was no medical assessment, so the cause is unknown, but these accounts are part of the picture. [67], [68]

How can you judge an LL-37 story?

Useful questions to ask: What was the problem, and was it ever diagnosed? Exactly what was used, and how much? What else was taken, especially antibiotics or other peptides? Would the problem have cleared by itself? What happened weeks later?

The accounts also show a gap between teaching and practice. Community education groups mostly describe short courses of 100–200 mcg for a specific infection, and some advise avoiding LL-37 with autoimmune conditions. Members often use it for colds, acne or prevention, sometimes for months. Selected discussions are not a survey of all users, a response rate or a substitute for controlled studies.

How does LL-37 compare with thymosin alpha-1, KPV and BPC-157?

These peptides are often combined with LL-37, sold by the same suppliers and discussed for immune, gut and healing problems. The table compares what each one is and how much human evidence exists. The route and schedule column describes what protocol sources say, not tested doses.

The real differences are in identity and evidence. LL-37 and BPC-157 each have a few small human studies, but by routes protocol sources do not use; KPV has none; thymosin alpha-1 is the only one approved as a medicine anywhere, and even its largest recent trial, in sepsis, found no clear benefit. No study has compared these peptides with each other, or tested their combinations, in people, so the table cannot say which works better for any problem. [6], [139], [135], [134]

Common questions about LL-37

Is LL-37 a natural antibiotic? Can it replace antibiotics?

It is part of the body's natural defenses, but lab-made LL-37 has not been shown to treat any infection in people. It kills bacteria in a dish, yet blood serum weakens it and bacterial enzymes break it down. The only human trials tested wounds and tumours. Using it instead of a prescribed antibiotic risks letting an infection get worse. [29], [34], [6] See what the human studies found.

How long does LL-37 take to work?

There is no reliable answer. Community reports describe changes within hours, after a few weeks or never, usually alongside other treatments and for problems that often improve on their own. In the largest leg-ulcer trial, 13 weeks of treatment closed no more ulcers than placebo. [22], [127], [6] See how long changes lasted.

Is LL-37 the same as cathelicidin? Does vitamin D raise it?

LL-37 is the active piece of the only human cathelicidin, a larger protein called hCAP18, so the names are often used together. Vitamin D is one of the signals that tells immune cells to make more of the body's own cathelicidin. Raising your own production is not the same as injecting a lab-made peptide, and neither has been shown to treat infections. [1], [112], [140]

Can LL-37 help acne or rosacea?

The evidence points the other way for rosacea. Injected into mouse skin, LL-37 is used to cause rosacea-like inflammation, and one person in a community education group described rosacea-like spots after a first dose. For acne, community reports are mixed and no study has tested it. Homemade creams have unknown strength and quality. [18], [98], [62] See why the same peptide can inflame skin.

Is LL-37 safe with an autoimmune condition?

That has never been studied, and it is the main reason for caution. LL-37 is a target of the immune attack in psoriasis, and immune reactions to it are found in lupus, psoriatic arthritis and rheumatoid arthritis. Community advice is split, with some recommending it for autoimmune conditions and others advising against it. [17], [101], [47] See who should be especially cautious.

What is the LL-37 and thymosin alpha-1 "immune stack"?

It is a combination sold and described for colds, flu and chronic infections, often with KPV and thymulin as well. Protocol pages describe LL-37 at 100–200 mcg daily alongside thymosin alpha-1 injections two or three times a week. No study has tested the combination, and a stack makes any benefit or reaction impossible to attribute. Read the thymosin alpha-1 and KPV guides. [9], [30]

Is it legal to buy LL-37?

It depends on where you live, and it is not approved anywhere we checked. It is not FDA-approved, FDA has flagged safety risks for compounded LL-37, and it was among unauthorized injectables Health Canada seized. A "research use only" label does not make a product legal or safe to use. [16], [8] See the dated legal-status summary.

Is a longer course more likely to work?

The evidence cannot show that. Course lengths vary from days to months, and nobody has tested any of them. In mice, 10 days of LL-37 in the skin caused lesions that did not fully clear, while lesions after 2 days cleared, which is one reason longer use raises questions. [15], [66]

Glossary

Plain explanations of the route, dosing and research terms used in this guide. Underlined terms in the text link here.

Adverse event

A harmful or unwanted medical event reported after someone used a treatment. A report alone does not prove the treatment caused it.

Aggregation

Peptide molecules clumping together. LL-37 naturally forms clusters in solution, which changes how it behaves; clumps in an injected product can also make an immune reaction more likely.

Amino acid

A small chemical building block. Chains of amino acids make up peptides and proteins.

Animal study

Research in animals such as rats or mice. It shows what a substance does in a living body, but results in people can differ.

Autoantigen

One of the body's own molecules that the immune system mistakenly attacks. LL-37 is an autoantigen in many people with psoriasis.

Autoimmune disease

A condition in which the immune system attacks the body's own tissues, such as psoriasis, lupus or rheumatoid arthritis.

Bioavailability

How much of a dose reaches the bloodstream in an active form. It depends heavily on the route, and it has never been measured for LL-37 in people.

Biofilm

A slimy, protective layer that bacteria build around themselves on surfaces, which can make them harder to kill.

Blend

A product or vial that combines LL-37 with other peptides. The blend's total amount is not its LL-37 amount.

Cathelicidin

A family of germ-fighting proteins made by many animals. People make only one, called hCAP18, and LL-37 is the active piece cut from it.

Cell study (in vitro)

Research on cells grown in a dish. It gives early clues about biology, but it is far from proof of benefit in people.

Compounding

When a pharmacy prepares a customized medicine from individual ingredients. In the US, federal lists control which bulk ingredients pharmacies may use.

Controlled trial

A study that compares people who receive a treatment with a similar group who do not, often receiving a placebo instead. Randomly assigning people to each group makes the comparison fairer.

Course and cycle

A course is the period of repeated use, such as three weeks. A cycle is a course plus a planned break before any further use.

Daily total

All the amounts given in one day, added together. For example, 100 mcg twice daily is a 200 mcg daily total.

Dendritic cell

An immune cell that detects signs of infection and alerts the rest of the immune system. One type, the plasmacytoid dendritic cell, reacts strongly to DNA and RNA.

Dose (amount each time)

The amount given on one occasion. A protocol lists both the dose and how often it is given.

Ex vivo

Tests on living cells or tissue taken out of the body, such as blood cells or donated skin. The donors are not treated, so it is not a human study.

Freeze-dried (lyophilized)

Dried by freezing and removing the water, leaving a powder. Most LL-37 is sold this way and mixed with liquid before use.

Half-life

The time it takes for the measured level of a substance in the blood to fall by half. It is not the same as how long an effect lasts.

Immunogenicity

The chance that a substance makes the immune system react against it, for example by making antibodies.

Intradermal

Injected into the skin itself, rather than into the fat below it. Researchers use this route to cause skin inflammation in mice.

Intramuscular (IM)

Injected into a muscle.

Intratumoral

Injected directly into a tumour. The melanoma trial of LL-37 used this route.

Intravenous (IV)

Given directly into a vein, so it enters the bloodstream immediately.

Investigational

Still being studied and not approved by a regulator, such as the FDA, to treat any condition.

Keratinocyte

The main type of cell in the outer layer of the skin. Keratinocytes make LL-37 and move in to close wounds.

Loading phase

Starting with a higher amount for a short period, then lowering it. Most LL-37 sources start low instead.

Mast cell

An immune cell in the skin and other tissues that releases histamine and other chemicals, causing redness, itching and swelling. LL-37 switches mast cells on in lab studies.

mcg and mg

Micrograms and milligrams. 1 mg equals 1,000 mcg. LL-37 protocol amounts are usually written in mcg.

Neutrophil

The most common type of white blood cell, which rushes to infections and stores LL-37's parent protein in its granules.

Oral

Taken by mouth, as a capsule, tablet or liquid.

Peptide

A short chain of amino acids. LL-37 is a chain of 37.

Pharmacokinetics

How the body absorbs, moves, breaks down and removes a substance. It has not been measured for LL-37 in people.

Placebo

A dummy treatment with no active ingredient, used as a comparison in studies.

Post hoc analysis

An analysis chosen after a study's results were known, such as looking at one subgroup. It can suggest ideas for new trials but cannot prove a benefit.

Preclinical research

Studies done before research in people: cell studies and animal studies.

Reconstitution

Mixing a freeze-dried powder with a liquid so it can be measured and used. The amount of liquid sets how much peptide each unit holds.

Research use only

A label on products sold for laboratory research. It does not mean a product is approved, tested for human use or legal to sell as a medicine.

Rosacea

A long-term skin condition causing facial redness, visible blood vessels and pimple-like bumps. Research links the body's own cathelicidin to rosacea inflammation.

Sepsis

A life-threatening reaction to an infection in which the body's response damages its own organs.

Subcutaneous (SC)

Injected into the fatty layer just under the skin. This is the route most LL-37 protocol descriptions use, although no human study has tested it.

Syringe units

Markings on an insulin syringe that measure volume. On a U-100 syringe, 100 units equal 1 mL. The amount of drug per unit depends on the vial's concentration.

Thymosin alpha-1

A lab-made copy of a 28-amino-acid peptide first isolated from the thymus gland. It is approved in some countries as Zadaxin, sold elsewhere for immune support, and often combined with LL-37.

Titration

Adjusting a dose in steps, usually starting lower and increasing if needed.

Topical

Applied to a surface, such as the skin or an open wound. The LL-37 wound trials used this route.

WADA

The World Anti-Doping Agency, which publishes the list of substances banned in sport.

Withdrawal and rebound

Withdrawal effects are symptoms caused by stopping a substance. A rebound is a problem returning worse than before after stopping. Neither has been studied for LL-37.

Supporting research: infection, wound and inflammation studies

Go deeper into the cell and animal experiments behind the claims. In the route details, intraperitoneal means into the abdominal cavity; intradermal, into the skin itself; and topical, applied to a surface.

How this guide was researched

This guide is built from a thorough review of the sources cited throughout it: published scientific studies, registered clinical-trial records, regulatory documents and published protocol descriptions. We also reviewed public online forums and community education groups where people describe their own experiences with LL-37.

The guide cites 148 sources, including 59 original scientific papers. Each type of source answers a different question. Studies show what researchers measured. Protocol descriptions show how LL-37 is typically used. Personal accounts show what individual people experienced. Every numbered citation links to its entry below, labeled by source type.

How this guide was made

Research and drafting were AI-assisted. Every cited source was checked against the original, and the guide was reviewed and edited by Doserly before publication. It has not had an independent clinical review, and Doserly does not currently have medical reviewers. Doserly makes a medication and health-tracking app and runs Doserly Academy, both of which are promoted in this guide. Read our editorial policy for how guides are researched, updated and corrected.

This guide is for educational purposes. It summarizes what the reviewed sources report so the research is easier to understand; it is not medical advice. For a deeper dive, or to check any point for yourself, go straight to the cited sources.

Explore the sources