Peptides for healing are studied for tendon, ligament, gut and skin repair after an injury. Most of the evidence is animal and cell work, not human trials. Here is what each one has been tested in, and where it stands with FDA. Last reviewed 10 November 2026.
6 peptides studied for tissue healing after an injury. Tap any card for the full guide.
The best-known 'healing peptide' in the research community, prized for tendon and gut repair claims. Backed almost entirely by animal data, not human trials.
Read the guideA synthetic fragment of thymosin beta-4. Its evidence is animal and cell work, and no published human trial has tested the fragment.
Read the guideThe full-length parent molecule behind TB-500, with more actual human trial exposure, mostly in eye and cardiac research.
Read the guideA copper-carrying tripeptide with real cosmetic-industry backing. One of the few peptides on this list with legitimate topical clinical data.
Read the guideA small anti-inflammatory fragment of alpha-MSH, studied mostly in gut-inflammation animal models.
Read the guideA naturally occurring antimicrobial peptide the body already makes. Studied clinically as a chronic-wound treatment.
Read the guideSix peptides come up most in research on healing after an injury: BPC-157, TB-500, full-length thymosin beta-4, GHK-Cu, KPV and LL-37. For most of them the evidence is animal or cell work, not human trials. Only two have a published randomised, placebo-controlled trial in human wounds, and both were applied to the skin: thymosin beta-4 and LL-37, each in venous leg ulcers. None of the six is an FDA-approved drug, for healing or for anything else.
This page is about tissue healing after an injury: wounds, tendons and ligaments, the gut lining, skin. Recovery from training is a different question with a longer list, and it has its own page: peptides for recovery. What follows is general information, not medical advice. It has no doses and no protocols, because for most of these compounds no human dose has ever been established.
| Peptide | Studied for | Best evidence so far | FDA status |
|---|---|---|---|
| BPC-157Body Protection Compound 157 | Tendon, ligament, muscle and gut injury | Animal and cell studies. Three small human reports, none with a control group. | Not approved. Advisory committee recommended it for compounding in July 2026. No rule yet. |
| TB-500Thymosin Beta-4 fragment | Muscle, tendon and skin repair | Animal and cell studies. No published human trial of the fragment. | Not approved. Advisory committee recommended it for compounding in July 2026. No rule yet. |
| Thymosin Beta-4 (full length)Tβ4 | Skin ulcers and the surface of the eye | Human trials: a phase 2 trial in venous ulcers (73 patients) and a small phase 3 eye-drop trial. | Not approved. Investigational. |
| GHK-CuCopper peptide | Skin wounds and skin ageing | Animal wound studies. Human data is cosmetic. A phase 2 wound trial is recruiting. | Not an approved drug. Non-injectable routes are in FDA's Category 1 for compounding. |
| KPVLys-Pro-Val | Gut inflammation | Mouse colitis models and cell studies. No published human trial. | Not approved. Advisory committee recommended it for compounding in July 2026. No rule yet. |
| LL-37Cathelicidin | Chronic wounds | One randomised human trial in venous leg ulcers (34 patients), topical. | Not approved. Committee review due before the end of February 2027. |
BPC-157 is a synthetic chain of 15 amino acids, based on part of a protein found in human gastric juice. It is the peptide most people mean when they say "healing peptide", and the animal record is large. A 2025 narrative review in Current Reviews in Musculoskeletal Medicine describes improved healing in rats with a cut Achilles tendon, a cut knee ligament and a cut quadriceps muscle. In a 2011 lab study on rat Achilles tendon cells, BPC-157 sped the outgrowth of tendon fibroblasts and increased their survival under stress and their migration. It did not directly change how fast the cells multiplied.
The human side is short. The same review found three human studies: a chart review of knee injections with no control group, a 12-person pilot in a bladder condition, and an intravenous safety pilot in two people. It called the human data "exceedingly sparse" and concluded that BPC-157 "should be considered investigational". A randomised trial in hamstring strain is recruiting and has no results. The full list sits on our BPC-157 and TB-500 research page.
TB-500 is a short synthetic fragment of thymosin beta-4. Its direct evidence is animal and cell work on muscle, tendon, skin and heart tissue. FDA says it "has not identified any human exposure data" for products containing the fragment. People often borrow the human results for the parent protein to fill that gap. They do not transfer: a fragment is a different molecule.
Thymosin beta-4 is the natural 43-amino-acid protein that TB-500 is cut from, and it has been tested in people. A phase 2, double-blind, placebo-controlled trial at eight European sites put a topical form on venous ulcers in 73 patients. Safety was comparable to placebo, and the authors wrote that one strength "may have the potential to accelerate wound healing". That is a cautious phrase, and the trial was not built to prove an effect. A 2012 review by researchers in the field also cites faster healing of skin punch wounds in rats and mice, including diabetic and aged mice.
The other human data is in the eye. A phase 3 trial of a thymosin beta-4 eye drop for neurotrophic keratopathy, a corneal condition, reported complete healing at four weeks in 6 of 10 treated patients against 1 of 8 on placebo. That difference did not reach statistical significance (p = 0.0656), and several of the authors worked for the company developing the drop. All of this is topical. None of it tests an injection for a tendon or a muscle.
GHK-Cu is a three-amino-acid peptide bound to copper, found naturally in human plasma. A 2015 review by Pickart and colleagues cites wound-healing effects in rats, mice, pigs and dogs. The human evidence is mostly cosmetic: skin firmness, elasticity and fine lines, from products applied to the skin. A phase 2 trial of a GHK-Cu gel on small standardised skin wounds in healthy adults is registered on ClinicalTrials.gov. It plans to enrol 60 people, was recruiting when we checked on 10 November 2026, and has posted no results.
KPV is three amino acids from the tail end of alpha-MSH, a hormone involved in controlling inflammation. Interest in it is about the gut lining. In a 2008 study, KPV reduced intestinal inflammation in two mouse models of colitis. There are no published human trials, and FDA says it "has not identified any human exposure data on drug products containing KPV".
LL-37 is an antimicrobial peptide your own skin and immune cells make. It has one real human wound trial. In a randomised, placebo-controlled, first-in-human study, 34 people with hard-to-heal venous leg ulcers had LL-37 applied to the wound twice a week for four weeks, at one of three strengths. The two lower strengths improved measures of healing, with mean ulcer area falling 68% and 50%. The highest strength did no better than placebo. More was not better, which fits what is known about this peptide: raised LL-37 is linked to rosacea and psoriasis. FDA's listed concern is separate, nonclinical findings that "suggest detrimental effects on male reproduction".
Sorted by tissue, the picture is uneven. People searching for the best peptides for injury recovery usually have one tissue in mind, so here it is that way round.
A few compounds from the wider recovery list are left out here on purpose. ARA-290 has phase 2 human data, but for nerve pain. Teriparatide and abaloparatide are FDA-approved for osteoporosis. Pentosan polysulfate is FDA-approved for a bladder pain condition and is not a peptide. All of them are on the recovery page.
No. None of the six is an FDA-approved drug for healing, and none has an approved form for another use. What has changed since 2024 is narrower: whether a compounding pharmacy may make them on a prescription.
The Category 2 page tracks each peptide, and the July 2026 vote explainer covers what a recommendation does and does not change. A vial sold online as "research use only" is none of the above. It is not an approved drug and it is not a pharmacy product.
Strong animal results are a reason to run human trials. They are not proof that something works in people, and most drug candidates that look good in rodents fail later. A few specific limits are worth knowing.
An injury that is not healing is a reason to see a clinician.
No human study has compared these peptides head to head, so nothing ranks one as best. What exists is separate research on separate tissues: BPC-157 and TB-500 in animal tendon, ligament and muscle models, KPV in mouse colitis, GHK-Cu in animal skin wounds, and topical thymosin beta-4 and LL-37 in small human trials on venous leg ulcers. None of them is an FDA-approved drug.
In animals and cell cultures, several of them have improved measures of repair. In people, the evidence is thin. BPC-157 has three small human reports with no control groups. TB-500 and KPV have not been tested in a published human trial. The controlled human trials of thymosin beta-4 and LL-37 were small, topical and early phase. Whether any of these peptides improves healing in people is not established.
No. BPC-157, TB-500, thymosin beta-4, GHK-Cu, KPV and LL-37 are not FDA-approved drugs for healing or for any other use. An FDA advisory committee recommended BPC-157, TB-500 and KPV for pharmacy compounding in July 2026, but that vote is advisory, and at our last check on 15 October 2026 FDA had not issued a rule. Products sold as research use only are not approved drugs.
BPC-157 and TB-500 are the two that come up most. BPC-157 improved healing in rat Achilles tendon and knee ligament models, and a 2025 systematic review of its orthopaedic research counted 35 preclinical studies and one clinical study. The direct evidence for TB-500 is animal and cell work. Neither has a published controlled human trial for a tendon or ligament injury.
Nobody knows, because no study has compared them. Both rest on animal and cell research. BPC-157 has a sizeable animal literature plus three small human reports without control groups. TB-500 itself has not been tested in a published human trial, though its parent protein, thymosin beta-4, has. Results for thymosin beta-4 do not transfer to the fragment, and neither compound is FDA-approved.
Two have a published randomised, placebo-controlled trial in human wounds, and both were applied to the skin, not injected. Thymosin beta-4 was tested as a topical treatment in 73 people with venous ulcers, and LL-37 in 34 people with hard-to-heal venous leg ulcers. Both trials were early phase and small. A thymosin beta-4 eye drop has also been through a small phase 3 trial for a corneal condition.
Start with the law and the product, not the claims. Is it legal to buy peptides sets out where a prescription, a compounding pharmacy and a "research use only" vial each stand. How to vet a peptide supplier covers what to check before any order.
Peptide AI is building a marketplace with two ways to buy: on a prescription through licensed telehealth, for what can legally be prescribed, or from vetted suppliers with an independent lab report you read before you order. It is not open yet. Early access is here.
These are the studies and regulatory pages behind the claims on this page. Compound-level detail is on the individual compound pages linked above. Last reviewed 10 November 2026.
Where a compound on this page has no human trial listed here, that is because we did not find a published one for healing. That absence is stated on the page, not filled in.
Peptide AI joins your doses, labs, and wearables so you can see which compounds actually move your numbers.