BPC-157

BPC-157 has been given to 30 people in three published human studies, none with a control group: a retrospective phone survey of 16 knee pain patients, a pilot in 12 women with interstitial cystitis, and a two-person intravenous tolerability check. All three share a first author, and the two that looked for benefit measured how patients rated their own symptoms. The deeper record comes from rats and cultured cells. That record is large and unusually consistent: transected Achilles tendons in rats healed stronger and faster, chemically induced stomach ulcers in rats came out smaller, and rat tendon cells in a dish migrated and survived stress better.

It has a large following among athletes. The consistency is also where caution belongs, because a 2025 review of the musculoskeletal literature noted that every published BPC-157 study reports a beneficial effect, and flagged possible publication bias. A record with no negative results is a reason for more scrutiny, not less.

FDA's published safety concerns about it still stand. It left the agency's Category 2 list because the nominators withdrew it, not because FDA cleared it, and WADA names it in the prohibited text.

BPC-157 is a synthetic chain of 15 amino acids, sequence GEPPPGKPADDAGLV, with a molecular weight of 1419. It is described as a partial sequence of a larger "body protection compound" first isolated from human gastric juice, which is why older papers call it a gastric pentadecapeptide. It has also appeared in the literature under the codes PL 14736 and PLD-116.

It is short lived in the body. The 2025 review puts its half-life at under 30 minutes, while the tissue changes reported in animals lasted weeks to months, a mismatch the review attributes to downstream signalling that keeps running after the peptide is gone. That explanation is a hypothesis drawn from animal work, not a measured result in people.

What It Does to Tendon Cells

The cleanest mechanistic study in the packet used fibroblasts taken from rat Achilles tendon. Explants cultured with BPC-157 grew outward faster. The cells did not multiply any faster, but more of them survived when stressed with hydrogen peroxide, and they migrated and spread across the dish more readily, in a dose-dependent way. The authors tied this to increased phosphorylation of two adhesion proteins, FAK and paxillin, without any change in how much of either protein the cells made.

So in that model it acts on movement and survival rather than growth. That fits a repair story, since a torn tendon needs cells to travel into the gap, but it is one cell type from one species.

The Wider Pathway Picture

Review-level work assembles a broader map from many animal and cell studies: increased VEGFR2 activity and nitric oxide production through the Akt and eNOS pathway, which drives new blood vessel growth; ERK1/2 activation in endothelial cells, where blocking ERK abolished the pro-migratory effect; and lower TNF-alpha, IL-6 and interferon-gamma, with macrophages shifting toward a reparative profile. The review singles out tendons and the junction between muscle and tendon because they have a poor blood supply.

Those same pathways are why the review lists theoretical risks alongside the benefits. Angiogenesis is also what tumours need, and sustained high nitric oxide has its own harms. Neither risk has been measured in people, and neither has been ruled out.

A rat Achilles tendon is not a human rotator cuff. The tendon work cut a rat's tendon cleanly and followed it for 14 days. A long-running human tendon problem that is loaded every day is a different situation, and no source here tested one.

Thirty uncontrolled patients is not proof it works in people. It is easy to cite the knee and bladder pilots that way. They had no comparison group, so they cannot show it. What they can show is that about 30 people received it without reported harm, and even that is a small window.

Feeling better is not tissue repair. The knee study asked patients by phone how their pain had changed; its authors list follow-up MRI as work for future studies. Nothing in the human record shows a tendon, ligament or cartilage healing.

One research group is not replication. All three human studies share a first author. Until an independent team runs a controlled trial, the positive results have not been tested against anyone else's methods.

Faster cell migration is not faster recovery. More tendon cells crawling across a dish is a real finding in rat fibroblasts. Whether that turns into a tendon that returns to full strength sooner in a person depends on loading, rehabilitation and the injury itself, and none of that has been tested.

Gut models have not been tested on human gut disease. Shrinking a chemically induced ulcer in a rat is a different thing from treating ulcers, reflux or inflammatory bowel disease in people. The one human programme for bowel disease named in these sources has no published results here.

Human safety rests on 30 people, followed for between three days and about a year. In the cystitis pilot nobody dropped out and no adverse events were reported; the review notes those women were checked for fever, rash, nausea, vomiting, worsening urinary symptoms and pain with intercourse. The knee study reported no adverse effects but did not describe how it looked for them. The two intravenous participants were followed over three days, reported no side effects and showed no measurable change in tested blood markers.

FDA's position is published and unchanged. The agency states that compounded drugs containing BPC-157 "may pose risk for immunogenicity for certain routes of administration and may have complexities with regard to peptide-related impurities and active pharmaceutical ingredient (API) characterization," and that "the agency lacks sufficient information to know whether the drug would cause harm when administered to humans." Immunogenicity means the body mounting an immune response against the peptide or impurities in it, so it is a product-quality problem as much as a pharmacology one.

The theoretical concerns raised in the review come from the mechanism: unwanted blood vessel growth, toxic breakdown products, and excess nitric oxide. Rat data pointing the other way exists, but none of it was collected in people.

For someone whose training lives or dies on tendons and gut tolerance, BPC-157 is worth understanding as a piece of repair biology. It points at real levers: blood supply to poorly vascularised tissue, cell migration into a wound, and the inflammatory clean-up phase. Those are the same processes that rehabilitation and sensible training load already work on.

What the evidence actually shows is narrow. In rats and rat cells, BPC-157 changes how injured tendon and stomach tissue heal. In people, the record is self-rated relief in uncontrolled pilots, and nobody has run the controlled study that would say whether it does anything at all, at what risk, or for whom. The gap between those two statements is the whole story, and no quantity of favourable rat papers closes it.

Not approved, and not cleared by FDA. It has no FDA approval for any use. It came off the Category 2 list on or about 2026-04-22 because the nominators withdrew the nomination, not because FDA resolved its safety concerns, and those concerns remain published. FDA's own briefing document going into the 23-24 July 2026 Pharmacy Compounding Advisory Committee meeting proposed that BPC-157 not be added to the 503A list. FDA has published no minutes, transcript or written vote tally from that meeting, so no outcome is stated here. A committee recommendation is not an FDA decision and would not change any legal status; rulemaking on the full record decides that.

Not permitted in tested sport. WADA prohibits it at all times under S0, Non-Approved Substances, and names it in the class text itself: "including but not limited to BPC-157, 2,4-dinitrophenol (DNP), ryanodine receptor-1-calstabin complex stabilizers [e.g. S-107, S48168 (ARM210)] and troponin activators (e.g. reldesemtiv and tirasemtiv)". S0 substances are Specified Substances.

Not a proven tendon, knee or gut treatment in humans. Its tendon and ulcer results come from rats. The human knee and bladder studies measured self-rated symptoms without a comparison group, and no controlled human trial has tested any of these uses.

Not a body-made molecule you are topping up. It is a synthetic fragment modelled on a protein from gastric juice. Its origin story does not make it a natural supplement, and says nothing about how a person handles the synthetic version.

BPC-157 is the compound I get asked about most, and I understand why. The rat data is genuinely striking, and if you have nursed a tendon for months you want it to be true.

I am not going to tell you the biology is fake, because it is not. A tendon that heals stronger in a controlled animal experiment, across more than one lab, is real information. But I also notice that every paper on it comes out positive, that the human record is three pilots from one research group with no comparison group, and that FDA still says it cannot tell whether it causes harm. That combination is the profile of a compound that has not been properly tested yet.

Where I land: worth following, worth reading, and not something I would put in my own body on today's evidence. If a real controlled trial ever reports, I will rewrite this page around it, in whichever direction it points.