TB-500: Almost Every Number You Have Seen Belongs to a Different Molecule

Search TB-500 and you will find a peptide with a research history going back to the 1990s, human trials, and a mechanism involving actin and cell migration.

Search a little harder and the research history belongs to a different compound.

What it is

Thymosin beta-4 is a naturally occurring peptide, 43 amino acids long, present in most human cells. It has a genuine research literature.

What vendors sell as TB-500 is not that. It is a synthetic fragment corresponding to residues 17 through 23 of thymosin beta-4, sequence LKKTETQ, with an acetyl group attached to the front end. A doping control laboratory synthesised it and characterised it by high resolution and triple quadrupole mass spectrometry to establish what was actually circulating in the market. (PMID 22962027)

Seven amino acids out of 43.

Citing the parent's research as evidence for the fragment is like quoting seven words out of a paragraph and expecting them to carry the paragraph's meaning. Sometimes a fragment does carry the activity. That is a finding you have to demonstrate, not an assumption you get for free.

The two molecules are trivially easy to tell apart on paper, which matters because you can do it yourself on a certificate of analysis.

What is sold as TB-500 is C38H68N10O14, average molecular weight 889.0, monoisotopic mass 888.49. (PubChem CID 62707662, CAS 885340-08-9) Full length thymosin beta-4 is C212H350N56O78S, average molecular weight 4,963. (PubChem CID 45382195)

One is about five and a half times the mass of the other. A certificate cannot fudge that. If a vendor tells you TB-500 is thymosin beta-4 and their own report lists a molecular weight near 889, the report is describing the fragment regardless of what the marketing says.

Two details worth having if you are checking one. The 889.0 figure is the neutral mass, not an m/z, so a certificate showing molecular weight 889.0 alongside a measured peak near 889.5 is internally consistent rather than contradictory. That gap is one proton. And because the material is usually supplied as an acetate or trifluoroacetate salt carrying some water, the powder in the vial weighs more per mole than the peptide does, which is why net peptide content on an honest certificate comes in below 100 percent.

The FDA has said this outright. In a footnote listing vendor websites, it notes that several of them describe TB-500 as a synthetic version of thymosin beta-4 and use the two terms interchangeably, then states plainly that thymosin beta-4 and TB-500 are not the same substance.

The line between them is not only chemical. It is regulatory, and it is mechanical. Under 21 CFR 600.3(h)(6), a protein is any amino acid polymer with a defined sequence greater than 40 amino acids. The 43 residue parent clears that line. The seven residue fragment does not come close. They are different molecules and different regulatory objects.

No receptor has been established for either one. Beta-thymosins are described in the primary literature as acting outside the cell by a mechanism that is not known. (PMID 21106936) The 2026 scoping review describes actin sequestration as the primary molecular function, with everything else downstream, rather than naming a receptor. (DOI 10.3390/app16126202) For the fragment specifically there is no binding data of any kind.

The molecule that was studied in humans is not the one being sold

Thymosin beta-4 has been given to people. Twice, in published work, at doses that do not resemble each other.

One randomised placebo controlled study gave healthy volunteers intravenous thymosin beta-4 at 42, 140, 420 and 1,260 mg, first as a single dose and then daily for 14 days, reporting no dose limiting toxicities or serious adverse events. (PMID 20536472)

A later first in human study of a recombinant version dosed intravenously at 0.05 to 25 mcg/kg as a single dose, and 0.5 to 5 mcg/kg daily for 10 days. (PMID 34346165)

Put those side by side. The human dose range for the parent molecule spans from micrograms per kilogram to 1,260 mg. That is roughly five orders of magnitude, depending on which preparation you mean.

For the fragment being sold as TB-500, the FDA's literature search found no article describing administration to a human, and no case reports of human use. There is no published human dose, no half life, no peak concentration, no bioavailability figure, by any route.

So there is no dose to inherit and nothing to scale from.

There is one listing that looks like it changes this, and it does not. A ClinicalTrials.gov record, NCT07487363, describes a Phase 1/2 randomized double blind placebo controlled dose escalation study of TB-500 in stable cardiovascular disease. 80 participants. Industry sponsor. Status: recruiting.

Its own summary opens with the sentence "This fictional study is an example of a ClinicalTrials.gov-style record."

It is not alone, and the wider picture belongs to Morgan McSweeney, who documented it on 31 July 2026. The same sponsor filed eight records across February and March 2026, all listing one site and one contact, all marked recruiting, claiming 3,399 participants between them, covering the grey market peptide catalogue: TB-500, BPC-157, MOTS-c, GHK-Cu, Melanotan II, tesamorelin. One describes itself as an example record. One has "(Mock Study)" in its title. Two are copies of real Eli Lilly obesity trials, reusing the original titles and Lilly's own internal molecule codes. Lilly's response was that it has no relationship with the sponsor and did not authorise the use of its protocol identifiers or the SURMOUNT-1 name. (Dr Noc)

Note what this means for anyone checking sources. A registry number carries the authority of an official government domain, and third party trial aggregators scrape these records and republish them without the disclaimer. The number will look like a citation everywhere it appears except at the source.

Even taken at face value, the record names only a "thymosin beta 4 17-23 fragment" and never specifies acetylation, so it would not have described the molecule actually being sold.

What the animal research actually used

This is the part that determines how you should read every other number.

Every rat dose in circulation was generated with full length thymosin beta-4. One study administered 2, 12 and 18 mg/kg and estimated 3.75 mg/kg, and its own abstract describes the test article as a 5K actin binding peptide, meaning the roughly 5 kilodalton parent. (PMID 25060418) Another used 6 mg/kg intraperitoneally in a rat embolic stroke model, with the first dose 24 hours after stroke induction and four more every three days. (PMID 20627173)

Topical thymosin beta-4 at 5 mcg in 5 microlitres of buffer, twice daily in mice, sped up corneal healing and reduced inflammatory signalling. (PMID 11950239) After coronary artery ligation in mice, thymosin beta-4 improved early myocyte survival and cardiac function. (PMID 15565145)

All parent molecule.

You cannot convert a dose for one molecule into a dose for a different molecule. That is not a difficult calculation. It is not a calculation at all.

Work using the fragment itself exists, and it is thin. A rat metabolism study, discussed below. (PMID 38382158) A 2025 peptide hydrogel in a corneal alkali burn model, which its authors describe as the first ocular application of TB500, though earlier rabbit corneal work with the unacetylated fragment exists. (PMID 41359360) A doping control paper in which horses received a single dose containing 10 mg of N-acetylated LKKTETQ, which is the fragment rather than the parent. (PMID 23084823) And a 2003 study administering the unacetylated fragment to mice.

Then there is the number that summarises the whole situation. A 2026 scoping review screened 1,772 records and included 80 studies. Most of them evaluated thymosin beta-4 rather than TB-500. The most represented tissue categories were skin and soft tissue wounds, blood vessels, cornea and bone. Musculoskeletal work was sparse, and direct TB-500 evidence appeared in a single included study. (DOI 10.3390/app16126202)

One study. In the tissue category the compound is almost entirely sold for.

The finding that complicates everything above

Assume for a moment that the fragment does something. There is still a problem with the thing in the syringe.

When researchers tracked what happens to TB-500 in rats, it came apart fast. Ac-LK was the dominant metabolite in the first six hours. Ac-LKK persisted and was still detectable at 72 hours. So far, unremarkable. Peptides get degraded.

Then they took the metabolites they had identified and screened them for wound healing activity in fibroblasts. The only one showing significant activity against control was Ac-LKKTE, a five residue breakdown product. Not the seven residue peptide that was administered. The authors' own conclusion is that the wound healing effects attributed to TB-500 may be due to that metabolite rather than to the compound itself. (PMID 38382158)

If that holds, then what people are injecting is closer to a shipping container than to cargo.

That is not automatically a problem. Plenty of real drugs work that way, and a compound that converts into an active form is still a compound that works. What it does mean is that every dosing schedule built around how long TB-500 persists in the body is timing the wrong molecule.

What is actually in the vial

An independent analysis of TB500 and TB1000 products bought over the internet found their contents were not systematically consistent with their descriptions. (PMID 36482504)

The veterinary side is worse. UC Davis's Kenneth L. Maddy Laboratory analysed a number of substances labelled TB-500, reported through the Racing Medication and Testing Consortium. Many contained no thymosin beta-4 or any peptide derived from it. Most did not contain any proteins, peptides or amino acids at all. Exactly one sample actually contained N-acetylated LKKTETQ.

Read that again slowly. A test that finds no amino acids in a peptide product has not found a low quality peptide. It has found something that was never a peptide.

The mislabelling starts at the source. The equine product label declares its active ingredient as "Thymosin Beta 4 10mg," while the compound generically sold under the TB-500 name is the acetylated heptapeptide. Those are the two different molecules, named interchangeably, on the label itself.

This is what turns the evidence gap from an academic problem into a practical one. Even a reader who wanted to reason carefully from the animal data cannot, because the label does not reliably tell you what is in the bottle.

Where the community schedule comes from

The convention that circulates is roughly 2 mg twice weekly for 4 to 6 weeks, then 2 mg once weekly as maintenance. Sources publishing it concede its status. One states directly that practical cycle planning for TB-500 is community derived, not validated by any published human trial of the fragment, and that no FDA approved dosing label exists.

There are two partial explanations, and honesty requires reporting that neither one closes the case.

The vial. TB-500 is most commonly sold in 10 mg vials. The most frequently repeated figures are simple fractions of that unit. 2 mg is one fifth. 2.5 mg is one quarter, and also half of a 5 mg vial. 5 mg is half, and also a whole 5 mg vial. Two of the weekly totals land exactly on vial boundaries.

But the pattern is selective rather than universal, and the entry would be wrong to overstate it. Intermediate figures like 3 mg and 4 mg are not clean fractions. Neither are the daily microdosing rungs. And 2 mg twice weekly, the single most repeated figure of all, produces 4 mg per week, which is neither a whole vial nor a tidy fraction of one. The packaging explains some of the round numbers. It does not explain the schedule.

The horse label. A veterinary equine product sold as TB-500 instructs: administer subcutaneously, one vial each week for six consecutive weeks as a loading dose, then one vial per month thereafter, with dosing after intense work at seven day intervals.

Four structural features of the human convention are already fully assembled there. The subcutaneous route. The load then maintain architecture. A six week loading block. A fixed whole vial amount rather than a weight based dose.

I want to be careful about what that does and does not establish. Two documents sharing four structural features, where one predates the other, is a strong hint and not a proof. I could not find a published account of anyone transposing the veterinary schedule into human use, and I am not going to assert one. What I can say is that the human schedule has a closer and more specific antecedent in a horse product label than it has in any study of any kind.

Nobody derived this from a trial. There has never been a trial.

What nobody knows

Whether the fragment does anything in humans. No one has published giving it to a person.

What either molecule binds to. No receptor for the parent, no binding data at all for the fragment.

Whether the injected compound or one of its breakdown products is the active agent.

Any toxicology, in any species. Not acute, not repeat dose, not genotoxicity, not reproductive, not carcinogenicity. The category is empty, and empty is different from reassuring.

What is in a given vial, without testing it yourself.

On the cancer question, which is the concern most readers arrive with. There is a real and statistically robust finding, and it is narrower than the way it gets repeated. Researchers used a virus to force melanoma cells that were already malignant to overproduce full length thymosin beta-4. Mean metastatic lung nodules rose from 10.9 to 46.7, with tumour vessel number up more than fourfold. (PMID 14625258)

That is genetic overexpression of the parent protein inside an existing tumour. It is not somebody injecting a heptapeptide. Writing "TB-500 causes cancer" would be unsupported. So would writing that the concern has been debunked, because no carcinogenicity study of TB-500 exists in any species. The question has not gone unanswered. It has gone unasked.

Regulatory status

The compound was originally nominated for the FDA's compounding list by a pharmacy network, and that nomination was later withdrawn. The FDA continued evaluating it anyway, on its own initiative, which is how the literature search described throughout this piece came to exist.

Its conclusions were the absences already listed. No published human administration. No toxicology in any species. And the explicit statement that thymosin beta-4 and TB-500 are not the same substance.

On the anti-doping side, this one is not ambiguous. TB-500 is named on the WADA Prohibited List by that exact trade name. The listed wording is "Thymosin-β4 and its derivatives e.g. TB-500," which appears under S2.3, Growth Factors and Growth Factor Modulators. (2026 Prohibited List)

Four things follow from that placement.

It is prohibited at all times, in competition and out. There is no window in which it is permitted.

It is a non-Specified Substance, because the whole S2 class is. That is the more serious category. The default sanction is four years rather than two.

Because WADA names the parent and its derivatives in a single clause, the fragment versus full peptide distinction that matters so much everywhere else in this piece makes no difference here. Both are covered.

And it has been listed since the 2018 List, effective 1 January 2018. It appears nowhere in the 2015, 2016 or 2017 lists.

One correction worth making, because it is repeated constantly. TB-500 is not banned under S0, the non-approved substances class. S0 by its own terms only reaches substances not addressed anywhere else on the list, and TB-500 is addressed explicitly under S2.3. BPC-157 is the one under S0. The two get swapped in secondary write ups routinely.

A therapeutic use exemption is not categorically barred for this class, and no WADA document says it is. In practice it is not obtainable, because the exemption standard requires the substance to be an indicated treatment for the condition, and TB-500 holds marketing approval nowhere. That is reasoning, not a WADA pronouncement, but the conclusion is the same. If you are tested, this is disqualifying.

My take

This is not a compound with weak evidence. That would be a different and more ordinary problem.

This is a compound with a naming problem sitting on top of an evidence vacuum. The research belongs to a larger molecule. The dosing schedule traces more cleanly to a veterinary label than to any study. The vials frequently do not contain what they claim, sometimes containing no peptide at all. And the form being injected may be a precursor to whatever is actually doing the work, if anything is.

Every one of those is separately fixable. A trial of the fragment would resolve the first. Analytical standards would resolve the third. None of them have been done.

What bothers me most is not the uncertainty. It is that the uncertainty is completely invisible in how the compound is sold. "TB-500, also known as thymosin beta-4" is not a simplification. It is the whole error, printed on the label.

If you have run TB-500, I would like to know whether you had a certificate of analysis for your specific batch, and what molecule it actually named. Given what the Maddy Laboratory found, that answer is more interesting than anything about the dose.

---

This content is for educational and informational purposes only and is not medical advice. Peptides discussed are research compounds and may not be approved for human use. Nothing here should be used to diagnose, treat, cure, or prevent any disease. Always consult a qualified healthcare professional before starting any peptide, supplement, or protocol. Individual responses vary. Do not self-administer compounds without proper medical supervision.