What it is
TB-500 is a synthetic chain of seven amino acids with an acetyl group on one end, Ac-LKKTETQ. It corresponds to residues 17 to 23 of thymosin beta-4, a natural protein of 43 amino acids, so about 18 percent of it by mass, 889 daltons against 4,921.
Thymosin beta-4 has a drug name, timbetasin, development codes, a decade of registered trials and published human safety data. TB-500 has none of those. The evidence for one does not transfer to the other.
FDA says so explicitly. Acetylation irreversibly alters a peptide charge, hydrophobicity and size, so in its words the profile of the non-acetylated heptapeptide cannot be extrapolated to TB-500. That matters, because most of the fragment research used the non-acetylated version, a third distinct molecule.
FDA also found TB-500 not physically or chemically well characterised, citing naming that does not follow nomenclature standards and missing impurity data in both the literature and the certificates of analysis it reviewed. Tests for impurities, aggregates, microbiological quality and endotoxin are not in the public domain.
The substitution problem
The confusion is not confined to marketing. It is in the peer-reviewed literature.
- A 2026 review in a pain journal lists thymosin beta-4 and thymosin beta-500 as separate entities, inventing an expansion of the acronym. The 500 is a product number and there is no such peptide.
- A 2026 scoping review in a sports medicine journal treats the pair as a single search unit, which pools evidence for the two molecules into one bucket.
- A 2026 rat tendon study titles its intervention synthetic thymosin beta-4 (TB-500), treating the seven-mer as though it were the 43-mer.
- A 2025 materials science paper applies the name TB500 to a hydrogel containing the non-acetylated sequence, which is the third molecule again.
- A major research chemical catalogue lists the correct CAS number under a title calling it a thymosin beta-4 synthetic molecule, while its own datasheet gives the seven amino acid sequence. The title asserts what the datasheet denies.
- Retail listings exist advertising TB-500 while describing the contents as full-length 43 amino acids, which is the substitution running in the opposite direction.
One 2026 review in Sports Medicine handles it correctly, listing thymosin beta-4 and TB-500 as two distinct entries. That is the minority practice.
How it works
Thymosin beta-4 binds monomeric actin one to one and buffers it. That is established for the parent protein. Whether TB-500 does it is a different question, and the structural literature suggests not.
LKKTET is the contact motif, but the function that prevents polymerisation needs the segment upstream of it, and a separate study found that extending the protein first helix over the LKK segment abolishes activity. TB-500 has no first helix, and no published paper measures actin binding for Ac-LKKTETQ.
The benefits map to other parts of the protein
The canonical map of thymosin beta-4 assigns its anti-inflammatory and anti-fibrotic activity to residues 1 to 4, and its cell survival activity to residues 1 to 15. TB-500 contains neither. A domain-mapping study across seventeen combinations in mice and pigs found the cardiac benefit belongs to residues 40 to 43, not 17 to 23. No preclinical cardiac study of TB-500 exists, and the mapping contradicts the cardiac claims attached to it.
One mechanism where the fragment matches the parent, and it is not a repair mechanism
Peptides corresponding to residues 17 to 23 stimulated mast cell mediator release at or above the level of native thymosin beta-4, in mouse and human cells. That is histamine and tryptase release, not repair, and it is the only mechanism where the fragment matches the full protein.
Research evidence
Human evidence for TB-500: none
FDA, verbatim: the nomination did not include, and FDA did not find any information in the medical literature where TB-500 was administered to patients to treat any disease or condition. It also identified no human study of its pharmacokinetics or pharmacodynamics by any route.
An independent search confirms it. The correct statement for this compound is no human studies, not limited human studies.
Registered trials of TB-500: zero, and one trap
There are no registered interventional trials of TB-500 in humans. There are twenty registered trials of thymosin beta-4, which is a different molecule.
One record looks like an exception and is not. A ClinicalTrials.gov entry exists for a phase 1 and 2 study of TB-500 in atherosclerotic cardiovascular disease, listed as recruiting. Its own summary begins by stating that it is a fictional study and an example of a ClinicalTrials.gov-style record. It is a template, and exactly the kind of artefact that gets screenshotted as proof TB-500 is in trials.
The thymosin beta-4 human record, reported accurately
The parent protein has a real programme and it deserves an honest summary, because it is the thing whose credibility TB-500 borrows.
A phase 1 intravenous study in 40 healthy volunteers found no dose-limiting toxicity from 42 to 1,260 mg. A separate phase 1 in China used 0.05 to 25 micrograms per kilogram, roughly 3.5 micrograms to 1.75 mg for an adult, also well tolerated. Four orders of magnitude apart, both declared safe, so there is no reconciled human dose even for the parent protein.
On efficacy: one small phase 2 in severe dry eye, nine patients, was positive. A three-trial phase 3 dry eye programme, the last enrolling 700 people, missed its primary outcomes. A phase 3 in neurotrophic keratopathy was terminated at 18 enrolled, and two phase 2 wound trials were terminated. The injectable programme went on clinical hold in 2011 over manufacturing compliance, not safety, and never restarted. No approval anywhere.
Preclinical evidence, and what the foundation actually is
The entire preclinical basis for the TB-500 marketplace rests on one sentence in a 2003 paper: in aged mice, the actin-binding domain as a seven amino acid synthetic peptide promoted repair comparably to the parent molecule. A real result. The peptide was not acetylated, so it was not TB-500.
The one time TB-500 itself was tested for wound healing, it did not work. In scratch-wound fibroblast cultures, TB-500 free base at 50 micrograms per millilitre induced no healing. Of the parent and its metabolites, only Ac-LKKTE, a five amino acid breakdown product, showed activity. The authors concluded the previously reported activity may be due to the metabolite rather than the parent. FDA cites the same study.
That leaves open a first-order question: the marketed molecule may be a prodrug for something else, or it may be inert.
The one in vivo result that is genuinely TB-500
In July 2026 a Turkish group tested TB-500 at 60 micrograms per kilogram per day intraperitoneally in a rat Achilles transection model, 32 animals in four groups, alongside BPC-157 and a combination arm. TB-500 reached significance on load to failure and total Bonar score. BPC-157 did not.
That is the strongest in vivo evidence for TB-500 there is, with its limits: eight animals per group, one model, one species, intraperitoneal rather than the route people use, four weeks. The combination arm did not beat TB-500 alone, and that is the entire in vivo evidence base for the Wolverine blend.
Beyond that, no TB-500 study exists in any large-animal musculoskeletal model, or in any muscle injury, ligament or cartilage model.
Safety
The absences here are not gaps in an otherwise complete picture. They are the picture.
FDA found no nonclinical toxicity study of TB-500 of any kind. No no-effect level, no repeat-dose toxicity, no genotoxicity, no reproductive toxicity, no carcinogenicity. Also no pharmacokinetic study by the proposed route, and no in vivo work assessing whether it promotes wound healing at all.
There is no dose-response curve for TB-500 in any species for any endpoint. The two documented in vivo exposures are single point doses, and the longest exposure in any organism is four weeks.
Immunogenicity is FDA primary stated concern and has never been assessed in a human. The agency noted the proposed intramuscular and subcutaneous routes may pose significant risk, amplified by aggregation and peptide impurities, and that the nomination included no study assessing it.
One risk framing worth stating plainly. Thymosin beta-4 is overexpressed across a range of cancers and has been proposed as a knockdown target rather than something to supply, with RNA interference against it reducing tumour growth in mice. No TB-500 oncology data exists either way.
A vial may not contain the peptide at all
This is documented rather than suspected. The Racing Medication and Testing Consortium tested products marketed as TB-500 and reported that many contain no thymosin beta-4 or any peptide derived from it, most containing no proteins, peptides or amino acids at all. A French racing laboratory independently reported products online claiming to contain either the acetylated fragment or the full protein, the same substitution seen from the analytical side.
No systematic purity survey of the current market has been published, so the proportions are unknown. A given vial may hold the fragment, the full protein, a different peptide, or nothing peptidic.
Regulatory status
TB-500 is not an approved drug for any indication anywhere, and neither is thymosin beta-4. Its entry in the FDA substance registry is identification, not approval.
FDA proposed not adding TB-500 free base or acetate to the 503A compounding list on four grounds: inadequate characterisation, no historical compounding use, no evidence of effectiveness, unknown human safety risk. On historical use the finding was categorical, that no studies were found using TB-500 as a compounded product and no outsourcing facility has reported compounding it. It sits on the Category 2 page among substances nominated but withdrawn.
On 23 July 2026 the Pharmacy Compounding Advisory Committee is reported to have voted eight to six with one abstention in favour of TB-500, against FDA staff. The vote is advisory and rulemaking would follow. That figure comes from trade press, so treat it as provisional until the minutes publish.
The anti-doping language is cleaner than the scientific literature, unusually. The WADA prohibited list, section S2.3, reads: thymosin-beta-4 and its derivatives e.g. TB-500. By naming the parent and adding its derivatives it sidesteps the identity problem, so the fragment is caught whether or not it is the same thing. Prohibited at all times, in and out of competition. A Canadian volleyball athlete got a four-year sanction in 2024 for BPC-157 and TB-500.
Horse racing is the best-evidenced strand, because TB-500 was a veterinary product first and racing laboratories built detection methods years earlier. The Racing Medication and Testing Consortium recommends the most severe penalty category, notes FDA has not assessed the peptide in the horse, and records metabolites detectable for roughly ten hours after injection, which makes it prohibited but hard to catch on race day.
Handling and identity
The identifiers for the parent protein conflict: the FDA registry gives one CAS number and chemical catalogues widely give another. For TB-500 the CAS number is concordant. No PubChem identifier for it could be confirmed from two sources, so none is published here.
TB-500 is cleaved rapidly. In rats the two-residue Ac-LK was the main metabolite in the first six hours and the three-residue Ac-LKK stayed detectable to 72 hours. In vitro, peptides of this kind degrade at all bonds regardless of protective modifications. Any circulating half-life figure has no human source.
Other fragment names circulate and none is interchangeable with TB-500: Ac-SDKP is residues 1 to 4, AGES is 40 to 43, plain LKKTETQ is the non-acetylated 17 to 23. Retail listings exist describing TB-500 as the fragment 1 to 4, which is the wrong fragment entirely.