TB-500 and thymosin beta-4: the tendon evidence

Two catalogue lines carry the name TB-500 at different amounts, the 10 mg line and the 5 mg line, and the literature is not settled on what that name denotes. A 2026 scoping review in sports medicine searched PubMed for six emerging peptides and listed the pair as one entry, “thymosin beta-4 or TB-500”, which reads as one compound (Tewari et al., PMID 42578445). A separate 2026 narrative review listed them as two different things: “Tβ4 (thymosin beta-4), and TB-500 (thymosin beta-4 fragment)” (Mendias and Awan, PMID 41966639).

That one word — fragment — is the whole argument. The molecule with the best-described biology is a specific 43-residue peptide, and whether the material sold under the catalogue name is that peptide, a fragment of it, or something else the papers do not settle. In PubMed, thymosin beta-4 in title and abstract returns 602 records; the exact string “TB-500” returns 28. Most of the science sits under one name while the commercial name has its own thin literature.

The molecule, as far as biochemistry goes

The best-established facts about thymosin beta-4 are structural rather than therapeutic. Beta-thymosins are a family of highly conserved polar peptides of about 5 kDa, present at high concentration in almost every cell, and thymosin beta-4 is their principal intracellular G-actin sequestering member: it binds monomeric actin in a 1:1 complex and acts as an actin buffer, preventing polymerisation into filaments while supplying a pool of monomers when the cell needs them (Huff et al., PMID 11311852).

The same review is blunt about the extrapolation. Effects are attributed to thymosin beta-4, to its oxidised form, or to the fragment acSDKP possibly generated from it — induction of metalloproteinases, chemotaxis, angiogenesis, inhibition of inflammation — and it adds that nothing was known about the molecular mechanisms mediating any of the effects attributed to extracellular beta-thymosins (PMID 11311852). An effect asserted in one paper and mechanistically unexplained in a review of the field is not yet a mechanism.

The animal and cell biology is substantial. Smart and colleagues found that thymosin beta-4 secreted from the myocardium drives epicardium-derived cells toward endothelial and smooth muscle fates in cultured adult explants (PMID 17495252) — real experimental work, and also explant culture. That paper also notes the peptide was already in multicentre phase 1 trials for cardiovascular disease, which tells you where the human effort went.

The tendon evidence is one rat study, and its authors call it exploratory

The most recent source behind tendon claims about TB-500 is a 2026 study in rats (Biçer et al., PMID 42542926). Thirty-two male Sprague-Dawley rats, twelve weeks old, about 330 g, each had a standardised Achilles tendon transection and repair, then were randomly assigned to four groups of eight: control, BPC-157, TB-500, and the combination. Treatment was intraperitoneal and daily for four weeks, and tendons were harvested at four weeks.

The results are specific, and narrower than the marketing. The TB-500 group reached statistical significance for maximum load to failure against control (p < 0.05) and had significantly lower total Bonar scores (p = 0.016); the TB-500 and combination groups shared significantly lower total Movin scores (p = 0.017 and p = 0.040). The BPC-157 group was numerically lower on total scores without reaching significance. Immunohistochemistry found no significant differences in collagen type I expression between groups; collagen type III differed, consistent with the histochemistry. Combining the two agents conferred no additional benefit over either alone.

The authors call it what it is: an exploratory rat model study, in which both compounds “were associated with” improved histopathological parameters during early Achilles tendon repair, with TB-500 additionally significant on the biomechanical measure. Eight animals per group, four weeks, a young healthy male rat, and a surgically repaired tendon — not the biological object a degenerating tendon in an adult human is.

The human trials exist — in the heart and in leg ulcers, not in tendons

Three human records are worth reading carefully, and not one of them measures a tendon.

The first is a 2010 randomised, placebo-controlled intravenous study in healthy volunteers (Ruff et al., PMID 20536472). Four cohorts of ten subjects each received a single injection of placebo or synthetic thymosin beta-4, at ascending reported amounts of 42, 140, 420 or 1260 mg, then continued the same regimen daily for fourteen days after a safety review. Adverse events were infrequent and mild or moderate, with no serious adverse events and no toxicity that limited the amount administered. This establishes tolerability and pharmacokinetics in ten healthy people per cohort, and measures no therapeutic outcome at all.

The second is topical, on venous ulcers, and the closest thing to a real efficacy signal (Guarnera et al., PMID 20536470). A double-blind, placebo-controlled, escalating-amount study across eight European sites — five in Italy, three in Poland — enrolled and randomised 73 patients. The reported safety profile was acceptable and comparable to placebo, and a 0.03% amount “may have the potential to accelerate wound healing”, with complete healing within three months in about a quarter of patients, especially where ulcers were small to moderate. The earlier protocol paper describes the design in progress: three groups of 24, randomisation in a 3:1 ratio, 84 days of treatment (PMID 17495250). This is chronic skin ulcers, not tendon.

The third is the most instructive, because of what it does not show. A 2025 paper reports a randomised, placebo-controlled, double-blind trial in 96 patients with acute ST-elevation myocardial infarction after primary PCI (Zhang et al., PMID 41229390). Infarcted area at 90-day follow-up was significantly reduced in the group that received its first amount within eight hours after PCI — that subgroup being 43 patients. But the overall difference in infarcted area between the treatment and placebo groups across all 96 patients was not statistically significant, and the authors conclude that further rigorous randomised studies are needed. A positive subgroup inside a null result is a hypothesis, not a finding.

What a tendon search actually returns

The absence claim, done properly. Every count on this page was run on PubMed on 2 October 2026, with every term tagged to title and abstract. thymosin beta-4 AND tendon returns three records: the sports-medicine scoping review (42578445), the rat Achilles study (42542926), and an electrospun PLGA/PLA yarn study carrying thymosin beta-4, which released the peptide over 28 days and had an additive effect on human adipose-derived stem cell migration, proliferation and tenogenic differentiation (31753373) — a construct in a dish, not a treatment trial. Adding human, patient or trial terms cuts the list to two, and both survivors are the review and the yarn study. thymosin beta 4 AND (randomized or randomised) AND (muscle or tendon or strength) returns zero, as does thymosin beta 4 AND (muscle or muscular) filtered to the clinical trial publication type. Hyphenated and unhyphenated forms were run separately: on some compounds the hyphen decides the result, but here both spellings returned the same 602 records.

So: 602 records on the compound, three on tendons, and no human tendon trial among them.

What the data do not show

This page deliberately does not tell you whether TB-500 heals tendons, because no study cited here measured that in a person. It does not carry the rat findings across to humans, it does not treat the naming dispute as settled, and it does not read the 43-patient subgroup of a 96-patient trial as a result.

What the evidence does show is narrower and more interesting than the marketing: a molecule with a solid structural role in actin handling, a rich animal literature, a genuine rat tendon study its own authors call exploratory, a small human tolerability study, and two clinical programmes that reached venous ulcers and myocardial infarction rather than tendons. Two 2026 reviews, one scoping and one narrative, reach a compatible conclusion: the claimed musculoskeletal benefits of these peptides remain unsubstantiated by current human trials, rigorous human safety data are scarce, and they should not stand in for orthopaedic care.

Sourcing note

For researchers, TB-500 appears in the catalogue at exactly two amounts, 10 mg and 5 mg, and those are the two lines in question. The naming problem is worth resolving before any comparison between a vial and a published result: a study of a defined 43-residue peptide is not automatically a study of the material described in the reference list.

Research use only. This page describes published animal, in vitro and clinical literature. It is not medical advice, not a treatment recommendation, and not a suggestion for human use. Nothing we supply is for human or veterinary use.