BPC-157 is everywhere online. The human evidence is almost nowhere.

Few compounds we cover have a wider gap between how often they are talked about and how much they have actually been tested in people than BPC-157. It is a fixture of podcast wellness segments, a staple of “recovery stack” content, and one of the most searched research peptides on the internet. The published human evidence for it would fit on an index card.

That gap is the story. Not because BPC-157 is obviously dangerous, and not because the preclinical work is fake — some of it is genuinely interesting — but because the confidence with which it gets recommended is completely out of proportion to what has been demonstrated in humans.

What it is

BPC-157 (“Body Protection Compound 157”) is a synthetic pentadecapeptide — a chain of 15 amino acids — derived from a sequence found in a human gastric protein. In animal models it has shown effects on tendon, ligament, muscle, and gut-lining healing, with proposed mechanisms involving nitric oxide signaling and angiogenesis. The rodent literature, much of it from a single research group, is the entire basis for the “heals everything” reputation.

The entire human evidence base, in one paragraph

A 2025 systematic review in orthopaedic sports medicine screened 544 articles published between 1993 and 2024 and found 36 studies worth including — of which 35 were preclinical and one was clinical. Across everything published, fewer than 30 humans have taken BPC-157 in any documented study, spread across small pilots, case reports, and observational series with no control groups. As of mid-2026 there is no completed, peer-reviewed, randomized controlled trial showing BPC-157 is effective — or safe — for any indication. A Phase II randomized trial in hamstring strain (NCT07437547) is registered but has not reported results, and a 2025 intravenous safety pilot enrolled just two participants.

That is not a summary of the strong studies versus the weak ones. That is the whole shelf.

Why the animal data doesn’t carry the claim

Preclinical promise is a reason to run human trials, not a substitute for them — and the history of medicine is a graveyard of compounds that healed rats and did nothing (or harm) in people. BPC-157 has the specific problems that make animal-to-human extrapolation risky: most of the evidence comes from one lab, the injury models and doses don’t map cleanly onto how people use it, and the marketing describes it with the grammar reserved for approved drugs — “for tendon repair,” “for gut health,” “for recovery” — a syntax the human evidence does not yet earn. It belongs squarely in the research-peptides-marketed-as-wellness category, where the mechanistic story is plausible and the clinical story is mostly absent.

What the FDA actually did — and the date to watch

The regulatory picture is easy to get wrong because it moved recently and in a counterintuitive direction. In September 2023 the FDA placed BPC-157 in Category 2 of its interim 503A bulk-substances review — the “significant safety risks” bucket that blocks compounding pharmacies from preparing it. Then, on April 22, 2026, the FDA removed BPC-157 from Category 2.

It is tempting to read that removal as a partial green light. It is the opposite. FDA removed it because the nominations to add it to the bulks list were withdrawn — not because the agency found it safe. There is still no compliant 503A pathway for routine compounding of BPC-157, and the FDA has stated it has insufficient safety information to know whether compounded BPC-157 would harm people for the proposed routes of use. The substance also remains banned at all times for athletes under category S0 of the WADA Prohibited List.

The next real milestone is a Pharmacy Compounding Advisory Committee (PCAC) review scheduled for July 23, 2026 — the first substantive public evaluation of where BPC-157 belongs. Anyone actually following the regulatory status of this peptide should watch that date rather than the April removal.

The gray-market problem sits on top of all of this

Because there is no approved product and no compliant compounding pathway, essentially all consumer BPC-157 comes from research-chemical vendors selling vials “for research use only.” Those products carry no verification of identity, purity, dose accuracy, or sterility, and the same is true of the other injectable research peptides frequently stacked with it. So the real-world question is not only “does BPC-157 work in humans” — it is “does the vial even contain what the label says,” and for gray-market peptides the honest answer is that nobody checked.

The bottom line

BPC-157 is a genuinely interesting preclinical compound and an almost-entirely-unproven human one. That combination — heavy animal data, near-zero human data, heavy internet enthusiasm — is precisely the profile that generates the most confident recommendations and deserves the most skepticism. If the July 23 PCAC review or the registered Phase II trial changes the picture, we will update the BPC-157 reference page and say so with sources. Until then, the evidence gap is the headline.