FDA advisers backed BPC-157. The human evidence is still remarkably thin.

What changed last week—and what did not

On July 23, FDA’s Pharmacy Compounding Advisory Committee voted 8–6, with one abstention, to recommend BPC-157 free base and BPC-157 acetate for inclusion on the section 503A Bulks List.[8] The committee was answering a compounding-policy question. It was not reviewing a new-drug application or declaring the substance safe and effective.

FDA’s public meeting page preserves the briefing documents and presentations, while the agency decides what to do with the recommendation.[2] The regulatory status remains unchanged today: BPC-157 is not FDA-approved.

Evidence ledger

The entire human record FDA found

FDA searched PubMed, Embase, ClinicalTrials.gov, safety-reporting systems, and other sources. It identified five clinical studies. They are heterogeneous enough that their participant counts should not be pooled into a single implied trial: different populations received the compound by rectal, joint, bladder, or intravenous routes.[1]

Counts describe separate reports, not unique participants in one program. FDA characterized the studies as short, small, exploratory, and limited in safety monitoring.[1]

The number “five” can sound more mature than the record is. Two reports were conference abstracts on rectal administration. The musculoskeletal paper looked backward at charts from one clinic. The bladder study had no comparator. The intravenous pilot included two people.[1][5][6]

The controlled study did not show a clear treatment advantage

The most informative efficacy report randomized 53 people with mild-to-moderate ulcerative colitis to a BPC-157 enema or placebo for two weeks; 46 completed it. The reported disease-activity score improved by 3.2 points with BPC-157 and 1.6 points with placebo. But the estimated 1.6-point between-group difference had a 95% confidence interval from −4.84 to 1.62, which crossed no difference.[1]

The abstract did not provide enough detail about eligibility, the endpoint definition, statistical methods, or follow-up for a full appraisal. Improvement within the treatment group is not proof of efficacy when the comparison with placebo is inconclusive. This trial also studied ulcerative colitis, not tendon or ligament repair.

The injury-recovery claim rests mostly on animal work

The frequently cited knee-pain report identified 17 patients in a retrospective chart review and reached 16 of them afterward. Participants received BPC-157 alone or combined with thymosin beta-4, without randomization, blinding, a placebo group, standardized imaging outcomes, or a prospectively registered protocol.[1]

Those limitations make it impossible to separate a treatment effect from natural recovery, co-interventions, selection, recall, or placebo effects. A 2025 systematic review of orthopaedic BPC-157 research found 36 eligible studies, but only one was clinical—the same knee report. Thirty-five preclinical papers do not become 35 human trials.[4]

An ex-vivo experiment on isolated human tissue or cells can answer a mechanism question without being a clinical trial. Likewise, an animal injury model can justify further development without telling us whether a marketed vial improves a patient-important outcome in people.

Safety is not established by an absence of reported events

FDA noted that no serious adverse events appeared in the five studies, but also emphasized their short duration, small samples, exploratory designs, and limited safety reporting. A two-person, two-day intravenous pilot is a first observation—not a safety database.[1][5]

FDA also found three adverse-event reports involving injected BPC-157: injection-site reaction, shortness of breath, and skin or gingival darkening. The reports cannot establish causation; concomitant peptides and missing details limit interpretation. They cannot establish incidence either, because spontaneous reporting has no reliable denominator and underreporting is expected.[1]

The agency identified no human pharmacokinetic data after oral, subcutaneous, nasal, or transdermal administration and no adequate long-term safety information. “No signal detected” in a tiny record is not equivalent to “risk excluded.”

“BPC-157” is also a product-identity problem

FDA evaluated the free base and acetate as different bulk drug substances. The withdrawn nominations called the material “BPC-157,” but their identifiers and certificates of analysis mixed information for the free base and acetate. FDA also found multiple salts and derivatives sold under the same common name and no USP, European, Japanese, or International Pharmacopoeia monograph.[1]

That matters because a study result belongs to the material and formulation actually tested. A name match on a storefront does not demonstrate sameness in active substance, content, impurity profile, sterility, stability, or clinical performance. FDA reported no approved BPC-157 product in any country in its review.[1]

One registered Phase 1 oral study planned to enroll 42 healthy volunteers, but ClinicalTrials.gov has no posted results and FDA found no associated publication.[3] A registration shows that a study was planned; without results, it cannot answer whether the product was safe, absorbed, or effective.

Questions people are asking