TB-500 vs. BPC-157: What the Evidence Actually Shows

TB-500 and BPC-157 are sold side by side, priced similarly, and marketed for the same thing: faster recovery from soft-tissue injury. Anyone comparing them is really asking a simple question — if I can only buy one, which one has more behind it?

That question has a real answer, and it is not the one either product page gives. This article compares the two compounds head to head on identity, mechanism, evidence volume, human data, pharmacokinetics, and regulatory status. It is not about combining them: our Wolverine stack article already covers what happened when someone tested the two together, and this article does not repeat that result.

First, what each one actually is — and one of these is not what it says

BPC-157 is a synthetic 15-amino-acid peptide, sequence GEPPPGKPADDAGLV, described in the literature as a partial sequence of a larger protective compound identified in human gastric juice. It has one identity, and the thing sold under that name is the thing studied under that name.

TB-500 is a different situation, and this is the most important single fact in the comparison. It is sold as thymosin beta-4, and it is not thymosin beta-4. Two independent laboratory papers — one from a human doping-control lab, one from an equine one — identified what is actually in a TB-500 vial: "the N-terminal acetylated 17-23 fragment of human thymosin beta 4 (Ac-LKKTETQ)³," a synthesized seven-residue piece of the natural 43-amino-acid protein, with an acetyl group added to the front end. The equine paper explains why that particular seven residues were chosen: "(17)LKKTETQ(23) is the active site within the protein thymosin β4 responsible for actin binding, cell migration and wound healing⁴."

So the comparison is not "peptide A versus peptide B with equivalent pedigrees." It is a whole synthetic peptide with its own three-decade research literature, against a fragment marketed under its parent protein's name. A 2026 sports-medicine review treats them as separate substances for exactly this reason, listing "Tβ4 (thymosin beta-4)" and "TB-500 (thymosin beta-4 fragment)" as two distinct entries in its own summary⁸. The parent protein has a real, decades-long human trial record — and it belongs to the parent, not to the fragment. Our separate thymosin beta-4 review covers that record on its own terms, precisely so this article does not borrow it here.

How much research exists behind each one

This is where the two diverge most sharply, and it is measurable rather than impressionistic.

For BPC-157, a 2025 systematic review searched PubMed, Cochrane, and Embase from 1993 through mid-2024 and reported the total directly: 36 studies, split as "35 preclinical studies, 1 clinical study⁵." A second, independent narrative review reached the same count of human work from its own separate search: "Only three pilot studies have examined BPC-157 in humans⁶." That is a thin human record sitting on top of a genuinely large, decades-old animal literature spanning gastric ulcers, tendon, ligament, muscle, and bone models.

For TB-500, our own evidence review documents the full PubMed search: 17 total results for the term, and outside of the review literature, nearly every one is an analytical-chemistry or doping-detection paper rather than an efficacy study. There is no TB-500 equivalent of BPC-157's thirty-five-study preclinical base. Three separate 2026 reviews each say so in their own words — "human orthopaedic data are lacking⁷," "rigorous human safety data are scarce⁸," "there is a current lack of clinical trials⁹."

The one experiment that tested both

Exactly one study has ever put BPC-157 and TB-500 in the same experiment, under the same conditions, in the same animals — which makes it worth reporting carefully, and worth being equally careful about what it cannot establish.

A 2026 Turkish orthopedic group took 32 male Sprague-Dawley rats, each about 330 grams and 12 weeks old, performed a standardized Achilles tendon transection and repair on every animal, and split them into four groups of eight: an untreated control, BPC-157 at 10 µg/kg/day, TB-500 at 60 µg/kg/day, and a combined group. Every dose was given intraperitoneally, daily, for four weeks, after which the tendons went to biomechanical testing or histology¹.

The results, in the study's own words. On biomechanics: "Biomechanical testing revealed higher maximum load to failure values in the BPC-157 and TB-500 groups compared to controls, reaching statistical significance in the TB-500 group (p < 0.05)¹." On tissue architecture: "significantly lower total Bonar scores in the TB-500 group (p = 0.016) and significantly lower total Movin scores in the TB-500 and BPC + TB groups (p = 0.017 and p = 0.040, respectively) relative to controls¹" — lower scores meaning better tendon structure and collagen alignment. And on the BPC-157 arm specifically: "The BPC-157 group showed numerically lower scores without reaching statistical significance for total scores¹."

Read at face value, TB-500 outperformed BPC-157 in that experiment. Now read the design, because three features of it make "TB-500 is the better one" an unsupportable conclusion from this single result:

• •The two arms were not dose-matched. TB-500 was given at 60 µg/kg/day and BPC-157 at 10 µg/kg/day — a sixfold difference. A comparison in which one compound gets six times the dose of the other is not a fair test of the compounds; it is a test of those two specific doses.
• •Eight rats per group. With that group size, "did not reach statistical significance" often means the study lacked the power to detect a real effect, not that no effect existed. The BPC-157 arm did produce numerically better numbers than control on both measures.
• •One injury model, one route, four weeks. The study's own authors call it "this exploratory rat model study," describe the findings as "preliminary," and state they warrant further investigation "pending dose-optimization and longer-term studies¹." That is the investigators' own assessment of what they produced, and it is the right one.

The study also included a combined-treatment arm. What that arm found is a different question from this article's, and our Wolverine stack review covers it in full rather than this page restating it.

Mechanism: two different proposed routes to the same claim

The mechanistic stories behind the two compounds are genuinely different, which is part of why they get marketed together.

BPC-157's proposed mechanism is the better characterized of the two at the cell level. A 2011 rat study found the peptide significantly accelerated tendon-explant outgrowth, improved fibroblast survival under oxidative stress, and dose-dependently increased both fibroblast migration and phosphorylation of FAK and paxillin — two proteins central to how a cell grips and crawls across a wound surface¹⁰. The review literature adds VEGFR2 activation, nitric-oxide signaling through the Akt-eNOS axis, ERK1/2 engagement, and reduced inflammatory cytokines⁵⁶. All of that was measured in cultured rat cells and animal tissue, not in a living human.

TB-500's proposed mechanism is inherited rather than measured. The claim is that Ac-LKKTETQ carries the actin-binding, cell-migration activity of the parent Tβ4 protein's active site — which is a reasonable hypothesis and the reason those seven residues were chosen⁴. The one study that actually tested it complicates it. A 2024 Korean doping-control lab synthesized authentic standards for TB-500 and each of its metabolites and screened them head-to-head in a fibroblast wound-healing assay, opening its own paper with the observation that "the biological effects of TB-500, however, have not been documented²." Its result: "Ac-LKKTE only showed a significant wound healing activity compared to the control²" — a metabolite, not TB-500 itself. Its conclusion states the implication directly: the previously reported wound-healing activity of TB-500 "may be due to its metabolite Ac-LKKTE rather than the parent form²." That is one in vitro study, not a verdict. It is also the only experiment anyone has run that isolated TB-500 from its own breakdown products, and it did not find the parent compound to be the active one.

Human evidence, and the pharmacokinetic gap nobody mentions

On human data the comparison is short. BPC-157 has three published human studies — a retrospective knee-pain chart review, a 12-woman interstitial-cystitis pilot, and a two-patient intravenous safety pilot — all from the same lead investigator, the same single journal, and the same private clinic, with combined enrollment under 30, no control group anywhere, and no blinding anywhere¹¹¹²¹³. A 2026 orthopedic review is blunt about the strongest of them: "A single human case series reported improvements in pain after intra-articular knee injections of BPC-157, although significant methodological flaws and a lack of controls limit its applicability and reliability⁷." TB-500 has zero published human studies of any kind. Three is a very small number. Zero is a different kind of number.

There is a second asymmetry that neither product page mentions and neither compound's own review article draws out. BPC-157 has a formal pharmacokinetic profile: a 2022 study ran a complete ADME workup in rats and beagle dogs, reporting an elimination half-life under 30 minutes, linear dose-proportional kinetics, and measured absolute bioavailability after intramuscular injection of roughly 14-19% in rats and 45-51% in dogs¹⁴. It is animal data, and it is real measurement. TB-500 has no equivalent. What exists instead is detection chemistry — methods built so anti-doping labs can find it in urine and plasma after a horse or an athlete has been given it⁴, and a metabolite-mapping study². Nobody has published how much of a TB-500 dose reaches circulation, how long it persists, or how it clears. For a compound whose one biological-activity screen suggested a metabolite might be doing the work, that gap is not a footnote.

Regulatory status: effectively identical, and unresolved for both

Neither compound has an FDA-approved product, and their compounding-law histories have moved in lockstep. Both sat in FDA's Category 2 list of bulk substances presenting significant safety risks from 2023, both were removed from that category on April 15, 2026 through the original nominator withdrawing rather than any FDA safety reversal, and both went before FDA's Pharmacy Compounding Advisory Committee on July 23, 2026 as candidates for the 503A Bulks List — BPC-157 evaluated for ulcerative colitis, TB-500 for wound healing¹⁵. Neither is on a finalized list today. What that decision actually decided covers the full timeline.

For a tested athlete the answer is simpler and more settled than the compounding question. BPC-157 is explicitly named on the World Anti-Doping Agency's prohibited list under category S0, non-approved substances — banned at all times, in and out of competition, with no therapeutic use exemption available, because S0 substances are categorically ineligible for one¹⁶. The 2026 orthopedic review states the same of both compounds: TB-4 and TB-500 "both remain banned substances in sports⁷."

The bottom line

The honest head-to-head does not produce a winner, and here is precisely why. BPC-157 has more of everything measurable: a larger animal literature (thirty-five preclinical studies to TB-500's handful), three published human studies to TB-500's zero, a characterized cell-level mechanism, and the only formal pharmacokinetic profile either compound has. TB-500 beat BPC-157 in the one experiment that tested both — at six times the dose, in eight rats per group, in one tendon model over four weeks, in a study its own authors call exploratory and preliminary.

More evidence is not the same as good evidence. BPC-157's thirty-five preclinical studies still sit on a human base of three uncontrolled pilots from one clinic, which the systematic review that counted them rates at "level IV and level V" — the lowest tiers, reserved for case series and expert opinion⁵. Neither compound has a completed controlled human trial. Neither has an FDA-approved product or an established dose. Whichever one a seller is steering you toward, the accurate description of the evidence behind it is the same: promising animal work, no human proof, and no agency-verified product to check a vial against.

For what each compound's own literature actually reports, see BPC-157's full evidence review and TB-500's. For what doses studies administered — again, not recommendations, since no completed human trial has established one — see our BPC-157 dosage article and our TB-500 dosage and side-effects review. If you already have a number in hand and just want the vial-to-syringe arithmetic, the BPC-157 and TB-500 reconstitution calculators do that math without validating the number, the product, or the choice.