Rat Achilles study: BPC-157 and TB-500

A four-arm rat study published in Joint Diseases and Related Surgery on 23 July 2026 tested BPC-157, TB-500 and the two together against a control group after a surgically cut Achilles tendon was repaired. TB-500 was the only arm that reached statistical significance on maximum load to failure at four weeks, and the combination beat neither peptide on its own^[1]. Every number in this article comes from rats. There is no human efficacy data for either compound in tendon repair, and nothing here should be read as a reason to use one.

How the study was built

Thirty-two male Sprague-Dawley rats, all 12 weeks old and weighing roughly 330 grams, underwent a standardised Achilles tendon transection and repair. They were randomly assigned to four groups of eight: control, BPC-157, TB-500 (the label used for synthetic thymosin beta-4 and its short fragments), and the two peptides combined. Treatment ran for four weeks after surgery and was given intraperitoneally, meaning by injection into the abdominal cavity. The paper reports 10 micrograms per kilogram per day for BPC-157 and 60 micrograms per kilogram per day for TB-500^[1]. Those are rodent figures on a rodent schedule by a route with no everyday human equivalent, and they do not convert into a human regimen. At four weeks the tendons were harvested for either biomechanical testing or histology.

What the numbers showed

Maximum load to failure came out higher in both the BPC-157 and the TB-500 groups than in controls, but only TB-500 crossed the significance threshold (p < 0.05). On the histopathology side, total Bonar scores were significantly lower in the TB-500 group (p = 0.016), and total Movin scores were significantly lower in both the TB-500 group (p = 0.017) and the combined group (p = 0.040). Both scoring systems run the same way round: lower means better tendon architecture, better collagen alignment and fewer degenerative changes. BPC-157 on its own moved the totals in the same direction without reaching significance on any of them^[1].

Sirius red birefringence showed more organised type I collagen and an altered distribution of type III collagen in the treated animals, most visibly in the TB-500 arm, which the authors read as a shift toward a more mature matrix. Immunohistochemistry found no significant difference in type I collagen expression between groups, while type III collagen expression did differ significantly. Giving both peptides together added nothing over either one alone. The authors offer one possible reason, that the two may converge on shared downstream pathways, and then say directly that this is a hypothesis awaiting experimental confirmation^[1].

What this is not

This is not evidence that either peptide helps a human tendon. It is an eight-animals-per-group, single-timepoint, four-week experiment, and the authors label it exploratory and preliminary in their own conclusion^[1]. The abstract does not say how the eight animals in each group were split between the biomechanical rig and the histology bench, which means the number sitting behind each individual comparison is smaller than eight. A p value of 0.040, arrived at in a study running two scoring systems, four stains and two immunohistochemical targets, is a prompt to look again rather than a result to bank.

The injury model matters too. A cleanly transected and surgically repaired rat Achilles is an acute, controlled wound with a known start time. It is not tendinopathy, not a partial tear, and not the degenerative Achilles of a forty-year-old runner. Those are different biological problems, and a compound that moves a histology score in the first one does not automatically touch the others. The open questions we track are set out on the BPC-157 page.

None of this is new ground either. A 2003 paper in the Journal of Orthopaedic Research reported that BPC-157 accelerated healing of a transected rat Achilles tendon, with improved load to failure, better elasticity and better collagen formation, using the same intraperitoneal route^[2]. That was twenty-three years ago, and the human trial it should have prompted still has not been run. A 2026 review in Sports Medicine describes the pattern across this whole category of compounds: favourable tissue repair outcomes in animal models, scarce rigorous human safety data, and real potential for serious harm to patients^[3].

Where both peptides actually stand

Neither compound is an approved medicine. The US Department of Defense's Operation Supplement Safety states that BPC-157 is not a dietary ingredient, that it is an unapproved drug which cannot legally be prescribed or sold over the counter, and that there is little to no reliable scientific evidence for its safety or effectiveness in humans. It sits on the World Anti-Doping Agency Prohibited List under S0, non-approved substances^[4]. Thymosin beta-4 and its derivatives, TB-500 included, were added as examples of prohibited growth factors under section S2.3 of the 2018 Prohibited List^[5]. For a tested athlete that is a sanction risk, not a grey area. The jurisdiction-by-jurisdiction detail sits in the BPC-157 regulation section and the TB-500 regulation section.

What would change the picture

A registered human trial with a clinical endpoint and a control arm. Not a case series, not a clinic's before-and-after photographs, and not a ninth rodent model. The paper's own conclusion is the honest summary: both peptides "warrant further investigation as candidate adjuncts to tendon repair", "pending dose-optimization and longer-term studies"^[1]. That is a research agenda rather than a treatment. If you are considering either compound for an injury, the useful next step is a conversation with a clinician who knows your medical history, plus a read of what we still do not know about TB-500.