Compound Physician required WADA prohibited Part of Peptides

TB-500

Synthetic thymosin beta-4 fragment studied for tissue repair. Animal data promising; human evidence sparse.

The story

Thymosin beta-4 entered the scientific record as an immunology molecule, not a repair molecule. It was purified from calf thymus as one component of "thymosin fraction 5," and in 1981 Teh-Lin Low, Su-Kuang Hu and Allan L. Goldstein published its complete amino acid sequence in PNAS: 43 residues, molecular weight 4,982, with an acetyl group blocking the amino terminus. They described it as a thymic hormone acting on lymphoid stem cells, characterized by its ability to induce terminal deoxynucleotidyl transferase in thymocytes (DOI 10.1073/pnas.78.2.1162). Nothing in that first description concerns tendons, muscle, or athletic recovery.

The biology moved over the following two decades. Tβ4 was recognized as the major G-actin–sequestering protein in eukaryotic cells, present in essentially all cells and body fluids, and its activities were mapped onto short internal sequences. A 2010 review by Gabriel Sosne, Ping Qiu, Allan Goldstein and Michael Wheater set out the map that everything downstream rests on: the amino-terminal Ac-SDKP tetrapeptide blocks inflammation and reduces fibrosis; a 15-residue amino-terminal region promotes cell survival; and "a short sequence containing LKKTETQ, the central actin-binding domain (aa 17-23) plus 1 additional amino acid (Q), promotes angiogenesis, wound healing, and cell migration" (DOI 10.1096/fj.09-142307). That sentence is the load-bearing scientific claim underneath every product sold as TB-500.

Clinical development belonged to RegeneRx Biopharmaceuticals (Rockville, Maryland) — a name corroborated beyond any registry field, since RegeneRx is a publicly quoted issuer (OTCQB: RGRX) publishing its own investor communications and is covered independently in trade press — which took synthetic full-length Tβ4 into trials in three formulations: RGN-259, a preservative-free ophthalmic solution; RGN-137, a topical dermal gel; and RGN-352, an injectable solution. The dermal program ran first — randomized, double-blind, placebo-controlled dose-response Phase 2 studies in pressure ulcers and venous stasis ulcers, each n=72, plus an FDA Office of Orphan Products–funded epidermolysis bullosa study terminated for "lack of patient availability and expiration of study drug" (NCT00382174; NCT00832091; NCT00311766 — T2, registry records). A later RGN-137 trial terminated at four patients for "Business Decision"; that record is registered under the name Lenus Therapeutics, LLC, not under the RegeneRx lineage, and this file does not assert who ran it (NCT03578029 — T2).

The eye program became the largest. A 2011 single-centre Phase 2 in moderate-to-severe dry eye (n=72) missed both primary endpoints while reporting significant secondary findings (DOI 10.2147/OPTH.S80954); a 9-patient severe dry eye trial reported significant improvements in signs and symptoms (DOI 10.1097/ICO.0000000000000379). Three large trials followed — ARISE-1 (n=317), ARISE-2 (n=601) and ARISE-3 (n=700) — each registered under the name ReGenTree, LLC, the joint venture named in the sponsor field of those records; the RGN-259 arms did not separate from placebo on the primary endpoints of ARISE-2 and ARISE-3 — per the sponsor's own registry-submitted results, never published, cited here only against the same lineage's public claim that the programme was positive in all three Phase 3s (NCT02974907; NCT03937882 — T2 records admitted under the R-C2b against-interest exception). A parallel neurotrophic keratopathy program, registered under the same name, produced one small positive-trending Phase 3 (SEER-1, n=18) and one miss (SEER-3, reported June 2025), with SEER-2 still recruiting (NCT05555589; Ophthalmology Times).

The injectable program — the one that would matter most for the way TB-500 is actually used — essentially never happened. The Phase 1 intravenous study in healthy volunteers was withdrawn, "never initiated due to contract manufacturing issues" (NCT00743769), and the RGN-352 Phase 2 in acute myocardial infarction was withdrawn with "trial never initiated" (NCT01311518). Injectable Tβ4 was instead carried forward by the sponsor registered under the name Beijing Northland Biotech Co., Ltd., with Phase 1a (n=54) and 1b (n=30) studies in Chinese healthy volunteers and two placebo-controlled Phase 2 trials after percutaneous coronary intervention (NCT04555824; NCT05984134).

TB-500 has a separate origin. It did not come out of RegeneRx, and it is not the 43-residue protein. It surfaced as a veterinary preparation, and the first published characterizations are doping-control papers. In 2012 Ho and colleagues wrote that "a veterinary preparation known as TB-500 and containing a synthetic version of the naturally occurring peptide LKKTETQ has emerged," and built an equine urine and plasma assay for it (DOI 10.1016/j.chroma.2012.09.043). The same year, a German group identified the N-terminally acetylated 17-23 fragment, Ac-LKKTETQ, inside TB-500 product by high-resolution mass spectrometry and synthesized it as a reference standard, describing TB-500 as "a product suspected to possess doping potential" (DOI 10.1002/dta.1402). The route into human use ran from horse racing through anti-doping laboratories into bodybuilding and then general wellness marketing, where it is now sold alongside BPC-157 as the "Wolverine stack" (MDLinx). Thymosin-ß4 "and its derivatives e.g. TB-500" have been named on the WADA Prohibited List under growth factors (WADA 2026 Prohibited List), and in July 2026 TB-500 was one of seven peptides voted on by FDA's Pharmacy Compounding Advisory Committee (FDA).

  1. 1981 Low, Hu & Goldstein publish the complete 43-amino-acid sequence of bovine thymosin beta-4 (MW 4,982), characterized as a thymic hormone
  2. 2006–2009 RegeneRx runs placebo-controlled Phase 2 dose-response trials of topical Tβ4 in pressure and venous stasis ulcers (n=72 each); the Phase 1 IV study is withdrawn, "never initiated due to contract manufacturing issues"
  3. 2010 Sosne, Qiu, Goldstein & Wheater localize angiogenesis / wound-healing / cell-migration activity to the LKKTETQ actin-binding domain, residues 17-23
  4. 2012 TB-500 first characterized in the scientific literature — as a veterinary preparation containing acetylated LKKTETQ, in two doping-control papers
  5. 2021-03-18 ARISE-3 (n=700) topline: the sponsor concedes the primary endpoints were not met — a concession against its own interest, which is the only reason a T3 release carries it here — while reporting a significant secondary endpoint (ocular grittiness); the sponsor's registry filing says the same, per the sponsor's own registry-submitted results, never published (R-C2b)
  6. 2022 SEER-1 Phase 3 in neurotrophic keratopathy published: complete healing in 6/10 on RGN-259 vs 1/8 on placebo (p=0.0656), terminated at n=18
  7. 2025-04-07 CCES announces a four-year sanction of a U SPORTS volleyball athlete for use of BPC-157 and TB-500
  8. 2025-06 SEER-3 Phase 3 in neurotrophic keratopathy misses its primary endpoint; sponsor attributes it to a stronger-than-expected placebo response
  9. 2026-07-23 FDA Pharmacy Compounding Advisory Committee votes narrowly to recommend TB-500 for the 503A bulks list for wound healing; FDA scientists had opposed all seven nominated peptides

What it does

A synthetic thymosin beta-4 fragment studied for tissue repair. Animal data is promising; human evidence is sparse and the supplier market is largely unregulated.

A synthetic thymosin beta-4 fragment, hypothesized to support cell migration in tissue repair.

  • tissue repair
  • injury recovery

What the evidence shows

Every clinical trial ever run under this compound's name studied a different molecule. All of them tested the 43-residue parent protein thymosin beta-4, not the seven-residue Ac-LKKTETQ fragment sold as TB-500 — FDA itself uses the separate name "Thymosin Beta-4, Fragment (LKKTETQ)" and states it has identified no human exposure data for it, so no human has ever been given the marketed molecule in a study. The parent protein's own trial record is larger than a reader would expect, and the sponsor's own filings contradict the sponsor's own account of it: against a release claiming the programme was positive in all three Phase 3 dry-eye trials, ARISE-2 (n=601) and ARISE-3 (n=700) show no separation from placebo on the primary endpoints — per the sponsor's own registry-submitted results, never published — and that same release concedes, against its own interest, that ARISE-3 "did not meet its primary outcome measures." Those filings are cited only against that claim, under the R-C2b against-interest exception, and never as a free-standing efficacy finding. The two dermal ulcer trials tell the same story: they closed 22% of ulcers in the treated arm and 22% in the placebo arm — per the sponsor's own registry-submitted results, never published, admitted under R-C2b as amended against the sponsor's implied position. That record was built for eyes and skin — dry eye, neurotrophic keratopathy, pressure and venous stasis ulcers, plus intravenous infusion after myocardial infarction — and never for the tendon, ligament, muscle or athletic-recovery uses that drive essentially all real-world interest; no formulation of thymosin beta-4 is approved by any regulator anywhere. The limitation that matters most is therefore not thin data but misattributed data: reasoning from the parent protein's trials to TB-500 changes the molecule, the route and the organ in a single step.

Overall evidence strength: Very low certainty

  • ARISE-2 and ARISE-3 (2018, 2021) — dry eye Randomized controlled trial

    The sponsor lineage's public claim is that its programme was positive in all three Phase 3 dry-eye trials. Its own filings for the two biggest of them say otherwise: the eye drops did not beat placebo on what the trials set out to measure — in ARISE-3, change in inferior corneal staining was −0.41 on drug and −0.46 on placebo, and change in ocular discomfort was −0.40 in both arms, per the sponsor's own registry-submitted results, never published. Those values are carried here only against that claim, under the R-C2b against-interest exception, and for nothing else. This dossier's reading, labelled as inference rather than record: that does not show the molecule is inert, only that on the endpoints regulators care about it did not separate. Nothing further is claimed for these results — neither trial has ever been published, so no T1 source characterises either one.

    Source

  • Sosne & Ousler (2015) — Phase 2 dry eye, controlled adverse environment (2015) Randomized controlled trial

    Neither of the two things the trial promised in advance to measure — ocular discomfort and inferior corneal staining at day 29 — differed from placebo, while several endpoints measured alongside them did.

    Source PMID 26056426

  • Sosne et al. (2022) — SEER-1, neurotrophic keratopathy (2022) Randomized controlled trial

    Six of ten treated patients had their corneal defect fully heal versus one of eight on placebo — a large difference that missed statistical significance (p=0.0656) because the trial was far too small, and the single healed placebo patient relapsed while no treated patient did.

    Source PMID 36613994

  • RegeneRx dermal wound program (2006–2012) — pressure ulcers, venous stasis ulcers, epidermolysis bullosa Randomized controlled trial

    This is the human wound-healing evidence the whole "Tβ4 repairs tissue" story rests on — and it has never been published. All three trials ran to completion or termination and posted results to the registry; no peer-reviewed publication of any of them exists, so no T1 source characterises what they found. The registry-posted healing counts are near-identical between drug and placebo — venous stasis ulcers closed in 4 of 18 placebo and 12 of 54 treated patients (22% in both), and pressure ulcers in 3 of 18 versus 8 of 53 — per the sponsor's own registry-submitted results, never published, admitted under R-C2b as amended against the sponsor's implied position, being a wound-healing programme it developed, funded and carried through three trials. They are cited for that purpose only and never as a free-standing efficacy finding in either direction. What remains is what T2 licenses, and it is not nothing: three randomized, double-blind, placebo-controlled trials were run in these indications, they finished, and their results have sat unpublished ever since.

    Source

  • Beijing Northland Phase 1a/1b and Phase 2a/2b (2017–2023) — intravenous recombinant Tβ4 Randomized controlled trial

    This is the only place systemic injected thymosin beta-4 has been formally dose-escalated in humans — and none of the four studies has posted or published results, so what those doses did is not in the public record.

    Source

  • Biçer et al. (2026) — rat Achilles tendon, BPC-157 vs TB-500 vs both (2026) Randomized controlled trial

    In rats, TB-500 produced a statistically significant improvement in tendon strength and architecture at four weeks — better than BPC-157 in the same experiment — and adding BPC-157 conferred no additional benefit, which is the first direct test of the combination people actually buy.

    Source PMID 42542926

  • Evidence syntheses (2026) (2026) Preclinical / mechanistic

    Three independent 2026 reviews converge: 67% of the class literature is preclinical and the musculoskeletal claims "remain unsubstantiated by current human trials" (DOI 10.1177/03635465261464420); for TB-4 and TB-500 specifically "human orthopaedic data are lacking" (DOI 10.1177/03635465251357593); and a Sports Medicine review lists Tβ4 and the TB-500 fragment as two separate entities (DOI 10.1007/s40279-026-02437-0).

    Source

Common misconceptions

“TB-500 is thymosin beta-4 — that's just the research name for it.”

They are not the same molecule, and that difference is the single most consequential fact in this dossier. - Thymosin beta-4 (Tβ4) is a naturally occurring 43-amino-acid protein, Ac-SDKPDMAEIEKFDKSKLKKTETQEKNPLPSKETIEQEKQAGES, MW 4,982, acetylated on serine-1, sequenced in 1981 (DOI 10.1073/pnas.78.2.1162). It is the highest-expressed β-thymosin in human cells and the principal G-actin–sequestering protein (DOI 10.2174/1389203724666221201093500), and it has a registered clinical program. - TB-500 as chemically characterized is a seven-residue synthetic peptide, Ac-LKKTETQ — residues 17–23 of Tβ4, the central actin-binding domain — with, in the analysts' words, "artificial acetylation of the N-terminus" (DOI 10.1016/j.chroma.2012.09.043; DOI 10.1002/dta.1402). That acetyl group is not where the acetyl group sits on natural Tβ4, and the peptide is roughly a sixth of the protein by residue count. FDA treats them as distinct and uses the fragment name explicitly: its bulk-substances page addresses "thymosin Beta-4, Fragment (LKKTETQ)" and states FDA "has not identified any human exposure data drug products containing Thymosin Beta-4, Fragment" (FDA). A 2026 Sports Medicine review likewise lists "Tβ4 (thymosin beta-4)" and "TB-500 (thymosin beta-4 fragment)" as two separate substances (DOI 10.1007/s40279-026-02437-0). The fragment is not inert — that is what makes the confusion durable. LKKTETQ is where much of Tβ4's angiogenesis, cell-migration and wound-healing activity has been localized (DOI 10.1096/fj.09-142307), and in human hepatic stellate cells the 17-23 fragment, but not the N-terminal 1-15 fragment, reproduced Tβ4's anti-fibrogenic blockade of Akt phosphorylation (PMID 30063851). But partial mechanistic overlap is not equivalence: Tβ4's other documented active sites — the anti-inflammatory Ac-SDKP tetrapeptide and the cell-survival N-terminal region — are simply absent from the fragment. As the analysts who built the first quantification assay for it wrote in 2024, "the biological effects of TB-500, however, have not been documented" (DOI 10.1016/j.jchromb.2024.124033). Honest answer: Human clinical data on thymosin beta-4 does not transfer to TB-500. Different molecule, different length, chemically different N-terminus, never given to a human in a registered trial, and no bridging pharmacokinetic study connects the two. The confusion is not purely a consumer error, either: products are marketed under the TB-500 name in both forms — some listings advertise a "full-length 43 amino acid" peptide under that label, others the fragment — so the name on the vial does not determine the molecule inside it. Reasoning "thymosin beta-4 has Phase 3 trials, therefore TB-500 is well studied" makes three substitutions at once: molecule, route, and indication.

“There are positive Phase 3 trials, so this works.”

There are Phase 3 trials, and their pre-specified primary endpoints were largely not met. ARISE-2 (n=601) and ARISE-3 (n=700) show no separation from placebo on the primaries — per the sponsor's own registry-submitted results, never published (NCT02974907; NCT03937882 — T2 records admitted under the R-C2b against-interest exception, cited here only against the sponsor's own "positive in all three Phase 3s" claim and never as a free-standing efficacy finding); the sponsor's own release concedes ARISE-3 "did not meet its primary outcome measures"; and SEER-3 missed its primary endpoint in June 2025 (Ophthalmology Times — T1-adjacent journalism relaying that sponsor's statement). What is true is that the sponsor reports several pre-specified secondary symptom endpoints as statistically significant, ocular grittiness among them (RegeneRx — T3, per sponsor announcement); the p-values it gives for them are deliberately not reproduced here, because a T3 sponsor announcement may never carry an efficacy figure. What is also true is that a small neurotrophic keratopathy trial showed a large, non-significant healing difference (6/10 vs 1/8, p=0.0656) — that figure is reproduced, because its source is T1 peer-reviewed literature (PMID 36613994). Note whose account this is on both sides: the "positive in all three Phase 3s" framing and the registry filings that contradict it come from the same sponsor lineage — RegeneRx, corroborated independently as a public issuer; ReGenTree, LLC and the ARISE and SEER records registered under that name; the GtreeBNT partner; and HLB Therapeutics — so they are one source, not a claim and its check. Both halves are printed here unreconciled. After roughly two decades and more than 1,800 randomized participants, no formulation of thymosin beta-4 is approved by any regulator anywhere.

“It heals tendons and muscle — that's what it's for.”

Not one human study of Tβ4 or TB-500 has enrolled a musculoskeletal injury. The human program is eyes, skin ulcers and heart attacks. The tendon claim rests on animal work, most directly a 2026 rat Achilles transection study in which TB-500 significantly improved load-to-failure and histological architecture at four weeks (DOI 10.52312/jdrs.2026.2951). A 2026 scoping review found 67% of the class literature was preclinical and the musculoskeletal claims "remain unsubstantiated by current human trials" (DOI 10.1177/03635465261464420). The indication with the most human evidence is dry eye. The indication with the most real-world use is soft-tissue injury. They do not overlap.

“It's a natural peptide your body already makes, so it's safe.”

Tβ4 is genuinely endogenous — unlike the corresponding claim for BPC-157, that part is not disputed. But endogenous does not mean benign at pharmacological exposure, and Tβ4's own literature supplies the reason for caution: reviews document increased Tβ4 expression in a number of metastatic tumors, associated with increased VEGF, and describe it as "a potential molecular target for tumor therapy" precisely because it facilitates "tumor metastasis and angiogenesis" (DOI 10.1615/critreveukargeneexpr.v22.i2.30; DOI 10.3109/07357907.2012.756111). ⚠️ The direction is not consistent. In multiple myeloma, lower Tβ4 expression was associated with worse outcomes and overexpression reduced proliferation (DOI 10.1111/j.1749-6632.2010.05470.x). No study in any species has tested whether administered Tβ4 or TB-500 promotes or inhibits tumor growth in a living animal with cancer. Open, not resolved; neither confident answer is supportable.

“It won't show up on a test — and anyway, a fragment isn't the banned substance.”

Both halves are wrong. Detection methods for Ac-LKKTETQ and its metabolites in urine and plasma were published in 2012, before most of the current marketing existed (DOI 10.1016/j.chroma.2012.09.043), and a validated UHPLC-Orbitrap method now quantifies TB-500 and its metabolites, with Ac-LKK detectable at 72 hours after dosing in rats (DOI 10.1016/j.jchromb.2024.124033). On the rules: the WADA Prohibited List names "Thymosin-ß4 and its derivatives e.g. TB-500" explicitly under growth factors (WADA 2026 Prohibited List), and Sport Integrity Australia lists TB-500, TB4 and thymosin beta-4 as names for one prohibited substance for which a Therapeutic Use Exemption "would not be granted … in any medical condition" (Sport Integrity Australia). The distinction that matters scientifically gives no protection at all in anti-doping. ---

Who it's for, who should skip

Experimental only. Status is gray_zone — removed from Category 2 in April 2026 and PCAC-recommended, but unlawful pending rulemaking. Any use belongs in a physician-supervised program with third-party purity testing.

Running it

No dosing protocol is published for this compound — there is no lawful supervised channel for the use this guide covers. What follows is monitoring, expectations, and safety framing only.

Form
injectable

What to expect

  • First 1–2 weeks — No complete corneal healing in either arm at day 8; 3/10 vs 1/8 by day 15 (NCT02600429). In dry eye, ARISE-3 measured its primaries at day 15; the posted values are carried only in and, under R-C2b, against the sponsor's own claim — they are not restated here as a free-standing expectation (NCT03937882 — T2, status) (Confidence: Trial-documented; the dry-eye result is cited only in/ under R-C2b)
  • 4 weeks — Complete corneal healing 6/10 vs 1/8, p=0.0656 (PMID 36613994); a later, larger trial in the same indication missed its primary endpoint (Ophthalmology Times) (Confidence: Trial-documented, conflicting between trials)
  • 12 weeks (topical, chronic wounds) — The dermal ulcer trials ran to day 84 and posted results to the registry; neither was ever published. Closure rates were essentially identical to placebo at day 84: 12/54 vs 4/18 and 8/53 vs 3/18 — per the sponsor's own registry-submitted results, never published, admitted under R-C2b as amended against the sponsor's implied position (Confidence: Trial exists; result carried under R-C2b as amended)
  • Long-term — The longest human exposure on record is 84 days of once-daily topical gel. Nothing in any formulation extends beyond that (Confidence: No data)
  • Anything at all for the fragment — Nothing — no onset, plateau, duration or cessation data exists in humans (FDA) (Confidence: No data)

Response signs

Likely working

  • Reduced ocular discomfort, grittiness, dryness on ophthalmic Tβ4 — Sponsor-characterized secondary endpoints (T3) — no figure is carried for them; the corresponding primaries were not met in the largest trials, per the sponsor's own registry-submitted results, never published (R-C2b) · RegeneRx — T3, per sponsor announcement; DOI 10.2147/OPTH.S80954 (T1)
  • Visible closure of a corneal epithelial defect over ~4 weeks — Trial-documented, underpowered (n=18), contradicted by a later Phase 3 · PMID 36613994 Trial-documented
  • A chronic skin ulcer closing faster than expected — Trial-documented — and the trials found no advantage over placebo. The two dermal ulcer trials posted closure counts to the registry and were never published; the counts show 22% closure in both the treated and placebo arms — per the sponsor's own registry-submitted results, never published, admitted under R-C2b as amended · NCT00832091 — T2, existence and status Trial-documented
  • Faster return to activity after tendon, muscle or ligament injury; reduced stiffness — Anecdotal, low confidence — the dominant claim in circulation, with no human trial support · DOI 10.1177/03635465261464420 Anecdotal

Adverse — seek review

  • Injection-site reaction that spreads, warms or persists; any hive-like or systemic allergic pattern — Regulatory concern, not trial-documented for this route — FDA's stated worry is immunogenicity from aggregation and peptide-related impurities · FDA Trial-documented
  • Fever, rash, malaise after a dose — Inference, not documented — flagged because measured gray-market purity has ranged from 5% to 75% · ECRI/ISMP, 2026 Inference

Common / neutral

  • No perceptible acute effect of any kind — Consistent with the record: no human trial describes an acute subjective signature · DOI 10.1007/s40279-026-02437-0

Getting it right

  • Monitoring: Establish a baseline before, not after. Nothing in the human record identifies a marker that moves with this compound; without a pre-exposure baseline, a later abnormality can be neither attributed nor excluded. mitigate side effect
  • Monitoring: Immunogenicity is the pharmacologically specific concern, and it is unmeasured. It was an explicit objective of the only intravenous human program, and those results were never published (NCT04555824); FDA raises it directly for the fragment. mitigate side effect
  • Monitoring: Age- and history-appropriate cancer surveillance is the reasonable response to an unresolved angiogenesis/metastasis question — not because a signal has been seen in a treated person, but because the question has never been asked in a living animal with a tumor. mitigate side effect
  • General safety: Product identity is the dominant risk here, ahead of pharmacology. The name "TB-500" is used for two chemically different molecules, so even an honest seller and an honest buyer can disagree about what is in the vial. Independent identity testing — not a seller-supplied certificate of analysis — is the only thing that resolves it. mitigate side effect
  • General safety: Gray-market product quality is documented and poor: tested peptide products have shown purity from 5% to 75%, with arsenic and lead above toxicity thresholds. This is a market-level finding, not attributed to TB-500 specifically (ECRI/ISMP, 2026). mitigate side effect
  • General safety: Athletes and service members face strict liability. The documented sanction in was for use, with no positive test, and ran four years. mitigate side effect
  • The physician conversation: Which molecule is it — the 43-amino-acid protein or the seven-residue fragment? This is the first question for this compound, and it usually cannot be answered from the label. mitigate side effect
  • The physician conversation: What is the specific injury or condition, and has it been imaged or diagnosed? (The human evidence base is eyes and skin ulcers, not tendons.) mitigate side effect
  • The physician conversation: What conservative options with randomized-trial support have been tried to completion? (.) mitigate side effect
  • The physician conversation: What is the personal and family cancer history? (is open, and the answer changes how open it should feel.) mitigate side effect
  • The physician conversation: What would count as it working, by when — and what happens if that date passes with no change? mitigate side effect
  • The physician conversation: Is there any competitive sport, collegiate, or military testing obligation? (— this is a four-year-sanction compound.) mitigate side effect

Safety

  • highly variable product purity; anaphylaxis risk with injection of unknown compounds

Stacking & alternatives

  • Complements BPC-157 — The canonical pairing, sold as the "Wolverine stack" (MDLinx). The only head-to-head experiment found no additive benefit, with TB-500 outperforming BPC-157 in rats (DOI 10.52312/jdrs.2026.2951). See
  • Alternative to GHK-Cu — Same tissue-repair intent, different mechanism; reviewed alongside TB-500 in the same 2026 scoping review and found equally unsupported in humans (DOI 10.1177/03635465261464420)

Access & cost

Legally unsettled. Access, where it exists at all, runs through a physician — not a consumer purchase. See the status note above.

No lawful channel to price. As of 2026-08-28 there is no lawful US route to obtain TB-500 for human use: no approved product, no USP monograph, and no placement on the 503A Category 1 bulks list, with the July 2026 advisory recommendation still requiring notice-and-comment rulemaking before any compounded product could be lawfully made (AJMC; FDA). There is therefore no legitimate telehealth, compounding-pharmacy, or brand price to report. A gray market nonetheless exists and has been documented in physician-facing reporting, with products sold online under "research use only" framing and no consistent relationship between the label, the molecule, and the price (MDLinx). No INTERNAL pricing subsection appears in this dossier: we found no TB-500-specific price figure in mainstream journalism, and the compilation brief's vendor-source rule excludes seller-derived figures. If a journalist-reported figure surfaces, it belongs in a marked, not here. ---

Questions

Is TB-500 the same thing as thymosin beta-4?
No. Thymosin beta-4 is a 43-amino-acid protein; TB-500, as characterized in seized and marketed product, is a synthetic seven-residue fragment of it (Ac-LKKTETQ, residues 17–23) with an artificially placed N-terminal acetyl group (DOI 10.1016/j.chroma.2012.09.043). FDA addresses the two separately and states it has identified no human exposure data for the fragment (FDA). Products are nonetheless sold under the TB-500 name in both forms, so the label does not settle what is in the vial. See.
Does TB-500 actually work?
For soft-tissue injury in humans, that has never been tested. It works in rats — a 2026 controlled study found significantly greater load-to-failure and better collagen architecture in repaired rat Achilles tendons (DOI 10.52312/jdrs.2026.2951) — and the parent protein has a large human program in eyes and skin whose primary endpoints were largely not met (NCT02974907; NCT03937882). Two 2026 sports-medicine reviews conclude human musculoskeletal data does not exist (DOI 10.1177/03635465261464420; DOI 10.1177/03635465251357593).
Oral, nasal, or injectable — which form works?
None has been compared to another in a human being. Ac-LKKTETQ is a seven-residue peptide with short-lived metabolites — in rats the principal early metabolite was the dipeptide Ac-LK (DOI 10.1016/j.jchromb.2024.124033) — and no published human bioavailability study exists by any route. Every registered human exposure to any form of thymosin beta-4 has been topical to skin, topical to eye, or intravenous. Treat oral, nasal and transdermal claims as marketing, and note that injectable claims are equally unsupported in humans for this fragment.
What are the side effects?
Not established for TB-500, because no human has taken it in a study. For full-length Tβ4 the randomized ophthalmic and dermal trials are the only tolerability record that exists, and the most-quoted characterization of ARISE-3's safety comes from the sponsor's own press release. A press release is not sufficient support for a safety figure, so no adverse-event numbers are reproduced here. What would settle it: the registry-posted adverse-event tables for ARISE-2 and ARISE-3, or a peer-reviewed publication of those trials — neither of which has been read into this dossier. Even if it were settled, it would describe an eye drop, not a systemic injection, and not this molecule. FDA raises a specific concern that the compounded fragment "may pose risk for immunogenicity for certain routes of administration due to the potential for aggregation as well as peptide-related impurities" (FDA).
Will it show up on a drug test, and for how long?
Yes. Validated methods detect Ac-LKKTETQ and its metabolites in urine and plasma, with Ac-LKK detectable up to 72 hours after dosing in rats (DOI 10.1016/j.jchromb.2024.124033); a separate urinary marker exists for full-length Tβ4 at 0.19 ng/mL (DOI 10.1002/dta.3552). It is prohibited at all times under WADA S2.3 with no Therapeutic Use Exemption available (Sport Integrity Australia). See.
Is the BPC-157 + TB-500 stack better than either alone?
The one experiment that has tested that says no: "combined BPC-157 and TB-500 treatment did not confer additional benefits compared to either agent alone," with TB-500 alone outperforming BPC-157 on the biomechanical endpoint (DOI 10.52312/jdrs.2026.2951). No human data on the combination exists beyond a four-patient subgroup in an uncontrolled chart review (PMID 34324435). See. ---

Last reviewed 27 August 2026. Regulatory status verified 27 August 2026.