Does TB-500 Have Human Trials? The 2026 Evidence

The honest answer is “almost none” — and the trials people cite to say otherwise are usually a different molecule, a different route, or a study that didn’t actually succeed.

Abstract: Does TB-500 have human trials? For the injected 7-amino-acid fragment people actually buy, there are no completed human efficacy trials and only one registered safety study, with no results yet. Nearly every “human data” claim borrows a trial of full-length thymosin β4 — usually an eye drop — and applies it to a different molecule by a different route. This guide maps the entire human evidence base and gives you a three-question test to grade any “TB-500 is clinically proven” claim yourself.

If you have searched “does TB-500 have human trials,” you have probably seen two opposite answers within the same page of results: confident vendor copy claiming “Phase 3 human data” and “clinically studied,” next to forum posts insisting “it’s animal-only.” Both are pointing at real facts and drawing the wrong conclusion. The truth sits in the gap between them, and once you see how the evidence is structured, the contradiction disappears.

The short version: thymosin β4 — the parent molecule — genuinely has human trials. The injected TB-500 fragment that sits in a research vial essentially does not. Almost every reassuring “human data” claim is built by quietly swapping one for the other. This article lays out the full record, grades each study honestly by provenance, and hands you a simple framework to never be fooled by the swap again. It is written for research and educational use only, not medical advice.

The short answer to “does TB-500 have human trials?”

No completed human trial has tested injected TB-500 for the tissue-healing uses it is marketed for. As of 2026, exactly one human study of the fragment is even registered — a Phase 1/2 cardiovascular safety trial (NCT07487363) — and it has posted no results. Everything else that gets cited as TB-500’s “human evidence” is a trial of full-length thymosin β4 (TB4), a structurally different and much larger molecule, usually delivered as a topical eye drop or an intravenous infusion in a clinic rather than the subcutaneous self-injection people actually use.

That single substitution — fragment for full protein, injection for eye drop — is the entire trick. To explain it cleanly, two definitions have to be nailed down first, because the marketing depends on them staying blurry.

TB4 (thymosin β4) is the full 43-amino-acid protein found in nearly every cell, where it acts as the body’s main actin-sequestering molecule — Goldstein and colleagues memorably described it as an actin-binding protein that “moonlights” to repair injured tissue. TB-500 is only its 7-amino-acid actin-binding fragment (the sequence LKKTETQ, residues 17–23). The fragment carries one of the parent’s several functional regions; it is missing the anti-inflammatory and anti-scar residues, the cell-survival region, and the antimicrobial and signaling segments. So a TB4 result cannot be assumed to hold for TB-500 — and that is precisely the assumption nearly every “human trial” claim quietly makes. We unpack the molecular difference in depth in TB4 vs TB-500: why the full peptide beats the fragment.

The whole human evidence map, sorted

Here is every category of human study that gets cited for “TB-500,” with what it actually tested and how it grades. Read the last column first: it tells you whether the study is about the thing in the vial.

What’s cited	Molecule	Route / setting	Result, graded	Is it injected TB-500?
RGN-259 / SEER-1 (neurotrophic keratopathy)	Full-length TB4	Topical eye drop	Phase 3; primary endpoint missed (p = 0.0656, n = 18) [clinical]	No
RGN-259 / SEER-3 (neurotrophic keratopathy)	Full-length TB4	Topical eye drop	Phase 3; primary endpoint failed [company-reported]	No
RGN-259 / ARISE-1/2/3 (dry eye)	Full-length TB4	Topical eye drop	>1,600 pts; mixed sign/symptom endpoints, no approval [company-reported]	No
Recombinant TB4 Phase 1 (healthy volunteers)	Full-length TB4	IV, in clinic	Well tolerated, no DLT, safety only — no efficacy [clinical]	No
Cardiac stem-cell pilot	Full-length TB4	Cell pre-treatment, ~10 pts	Very small early signal [educational/secondary]	No
Dermal wound Phase 2	Full-length TB4	Topical gel	~73 pts; faster healing reported [educational/secondary]	No
Musculoskeletal “healing” reputation	TB-500 fragment	SubQ injection	Animal models + anecdote only [animal]	Yes — but not human
NCT07487363 (stable ASCVD)	TB-500 fragment	Injection, in trial	Phase 1/2 safety, registered 2026, no results [registered]	Yes — but no data yet

Every single row that involves the injected fragment is either animal data or a registered-but-unreported safety trial. Every row with real human efficacy data uses the full-length protein, and almost always a topical or intravenous route. There is no row that is both “the fragment you inject” and “a finished human efficacy trial.” That empty cell is the honest answer to the question.

The three-question test for any “TB-500 human trial” claim

You do not need to memorize trial names to defend yourself. Any time a page tells you TB-500 is “clinically studied” or “has Phase 3 data,” run the claim through three questions. A legitimate proof has to pass all three; the marketing version usually fails at least one.

1. Is it the same molecule? Full-length 43-amino-acid TB4 is not the 7-amino-acid TB-500 fragment. If the study used “thymosin β4,” “Tβ4,” or “RGN-259,” it tested the whole protein. A result for the parent does not transfer to the fragment any more than a result for the fragment transfers up to the parent.

2. Is it the same route? A drop on the surface of the eye, or an infusion given in a clinic, is a completely different pharmacology from a subcutaneous injection aimed at a tendon or muscle — different tissue, dose, distribution, and exposure. A four-week corneal-healing endpoint says nothing about a hamstring.

3. Did the trial actually hit its endpoint? “Phase 3” describes a trial’s stage, not its success. The single most-quoted “thymosin β4 works in humans” number (SEER-1’s 60% vs 13% healing) missed statistical significance, and the larger confirmatory trial failed. A trial can be large, rigorous, and still negative.

Apply this and the picture collapses to one line: the only genuinely human, controlled efficacy data for thymosin β4 is for the wrong molecule given by the wrong route, and even some of those trials did not meet their endpoints.

Walking the “human data” that actually exists

It is worth being precise about each piece, because the individual studies are real science — they are just mis-cited.

The eye drops: where “Phase 3” comes from

The flagship human program is RGN-259, a 0.1% full-length thymosin β4 eye drop developed by RegeneRx and partners for dry eye disease and neurotrophic keratopathy. Its most-cited result, SEER-1, reported complete corneal healing in 6 of 10 treated patients versus 1 of 8 on placebo — the “60% vs 13%” line. What the citation skips: the trial enrolled only 18 patients, closed early, and missed its primary endpoint at p = 0.0656 (Kang et al., 2022). A larger European confirmatory trial, SEER-3, then failed its primary endpoint outright, and the three-trial ARISE dry-eye program (over 1,600 patients) produced only mixed sign-and-symptom signals and no FDA approval. This is the single richest vein of human thymosin β4 evidence, and it is a topical full-length drug with an unconverted track record. The full breakdown lives in our RGN-259 eye-drop explainer.

The systemic safety study: real, but not about benefit

The closest thing to a “TB4 injected into the body” human trial is a first-in-human Phase 1 of recombinant full-length TB4 in healthy volunteers (Wang et al., 2021): single ascending doses in 54 subjects and multiple ascending doses in 30, well tolerated, with no dose-limiting toxicity and no tumorigenesis over six months [clinical]. It is genuinely useful — it establishes a wide human safety margin for the parent protein — but it tested no healing outcome and, again, used the full-length molecule, not the fragment. It was explicitly run as a stepping stone toward an acute-myocardial-infarction study.

The cardiac and skin signals: small and secondary

Educational summaries also cite a small (~10-patient) cardiac pilot in which TB4-pretreated stem cells were associated with better heart function, and a ~73-patient dermal Phase 2 reporting faster chronic-wound healing with a topical TB4 gel. Both are full-length TB4, both are early or topical, and the versions in circulation come through secondary educational synthesis rather than primary literature [educational/secondary]. They belong in the “interesting, not decisive” column — and they are still not the fragment.

The fragment itself: animal data plus one safety trial

Strip away everything that is actually about the parent protein, and what remains for injected TB-500 is an animal literature plus one registered human study.

The animal record is the real basis for TB-500’s reputation: rodent models showing roughly 2–3× faster muscle repair (dose-dependent), stiffer and stronger fracture healing, and functional recovery after stroke or traumatic brain injury. That body of work is suggestive, but animal data is exactly where the animal-to-human dose conversion problem bites hardest — a strong rodent effect at a milligram-per-kilogram dose tells you little about a human result.

The one human study is NCT07487363, registered in 2026: a Phase 1/2, placebo-controlled, dose-escalation safety trial of the TB-500 (17–23) fragment in adults with stable atherosclerotic cardiovascular disease, with three dose cohorts, 3:1 randomization, eight weeks of dosing, and a four-week follow-up. Two things make it a poor source for “TB-500 is proven.” Its primary endpoint is safety, not efficacy, so even a clean readout would say the fragment is tolerated, not that it heals anything. And it targets the cardiovascular system, not the musculoskeletal uses TB-500 is sold for. It also currently has no posted results. We cover exactly what it can and cannot establish in our piece on the first human TB-500 trial — which, notably, arrived the same year as the first human BPC-157 RCT, marking the moment both flagship healing peptides finally entered controlled human testing.

What a real TB-500 human trial would have to show

It is worth stating plainly what would actually change the answer, so you know what to watch for rather than what to take on faith. A study that genuinely demonstrates “TB-500 works in humans” would need to: use the defined 17–23 fragment (not full-length TB4, and confirmed by mass spectrometry); deliver it by the route people use (subcutaneous injection, for a systemic target); test a healing or functional efficacy endpoint (an imaged tendon, a measured return-to-function), not just a safety or biomarker readout; and meet that endpoint in a randomized, controlled, adequately powered design. Nothing registered to date does all four. NCT07487363 gets the molecule and route right but stops at safety. Until a trial closes that gap, the correct grade for injected TB-500’s efficacy in humans is “untested,” not “disproven” and certainly not “clinically proven.”

Where that leaves TB-500 in 2026

For a serious researcher, the practical status is unchanged by all of this. TB-500 is not FDA-approved for any human use and is sold for research and educational purposes only. It sits on the WADA Prohibited List under Section S2 (growth factors), banned in sport at all times, with a detection window of roughly a month — a relevant fact for any tested athlete. Because the human efficacy case rests on the parent molecule and the fragment’s own data is animal-plus-one-safety-trial, the burden of quality control falls entirely on the buyer.

The fragment-versus-full-length problem makes one sourcing step non-negotiable: a vial labeled “TB-500” should be confirmed as the intended molecule by mass-spec identity, not trusted on its label, since the relabeling can run in either direction. Our guide to reading a peptide COA covers the exact purity-versus-identity checks to demand. The same theoretical safety ceiling applies regardless of evidence: the repair mechanisms that make β-thymosins interesting — angiogenesis, cell migration, anti-apoptosis — are also the mechanisms by which a repair signal could, in principle, support an existing tumor, which is why a personal or family cancer history is a standard reason to avoid this class (see our peptide side-effects and safety guide). You can review the materials discussed here on the TB-500 and BPC-157 + TB-500 blend pages, which exist for research use.

The bottom line is not that TB-500 is fake or that thymosin β4 is uninteresting — the parent protein has more serious human science behind it than most research peptides. It is that the two should never be quietly merged. When someone says TB-500 “has human trials,” the accurate translation is: the full-length protein has been studied in humans, mostly topically, with a mixed record; the injected fragment has not. Hold those apart and you can read the entire field without being sold.

FAQ

Does TB-500 have human clinical trials? Not for efficacy. As of 2026 there are no completed human trials of the injected TB-500 fragment for tissue healing, and only one registered human study (NCT07487363), a Phase 1/2 cardiovascular safety trial with no posted results. The “human data” usually cited is for full-length thymosin β4, a different molecule.

Is TB-500 the same as thymosin β4 (TB4)? No. TB4 is the full 43-amino-acid protein; TB-500 is its 7-amino-acid fragment (residues 17–23). The fragment reproduces only one of the parent’s functional regions, so TB4 trial results do not automatically apply to TB-500.

What about the “Phase 3” data I keep seeing? That refers to RGN-259, a full-length TB4 eye drop. Its main neurotrophic-keratopathy result missed statistical significance (p = 0.0656) and a larger confirmatory trial failed, while its dry-eye program earned no approval. “Phase 3” describes the trial’s stage, not a success.

Is there any human safety data for TB-500? There is strong human safety data for full-length thymosin β4 (including IV dosing studies), but that is the parent molecule. The fragment’s own first human safety trial (NCT07487363) is ongoing with no results.

Is TB-500 FDA-approved or legal to use? No. TB-500 is not FDA-approved and is sold for research/educational use only. It is also on the WADA Prohibited List (Section S2), banned in sport at all times.

So does TB-500 work? For humans, it is best described as untested rather than proven or disproven. The injury-repair case rests on animal models and anecdote; the genuine human efficacy evidence belongs to the full-length protein by other routes.

References

1. Wang, H., et al. (2021). A first-in-human, randomized, double-blind, single- and multiple-dose, phase I study of recombinant human thymosin β4 in healthy Chinese volunteers. Journal of Cellular and Molecular Medicine, 25(18), 8843–8851. doi:10.1111/jcmm.16693. PMID 34346165. — [clinical — full-length TB4, IV, safety]
2. Kang, S., et al. (2022). 0.1% RGN-259 (Thymosin β4) Ophthalmic Solution Promotes Healing and Improves Comfort in Neurotrophic Keratopathy Patients in a Randomized, Placebo-Controlled, Double-Masked Phase III Clinical Trial (SEER-1). International Journal of Molecular Sciences, 24(1), 554. doi:10.3390/ijms24010554. PMID 36613994. (NCT02600429.) — [clinical — full-length TB4, topical; primary endpoint missed]
3. ReGenTree, LLC / RegeneRx Biopharmaceuticals (2021). Additional Results from ARISE-3 and Pooled Data from Three Phase 3 Clinical Trials Using RGN-259 for the Treatment of Dry Eye. Company press release, 14 May 2021. — [company-reported topline; not peer-reviewed]
4. HLB Therapeutics, reported via Ophthalmology Times. RGN-259 misses primary endpoint in European Phase 3 SEER-3 trial of neurotrophic keratopathy. — [company-reported topline; not peer-reviewed]
5. Bock-Marquette, I., et al. (2004). Thymosin β4 activates integrin-linked kinase and promotes cardiac cell migration, survival and cardiac repair. Nature, 432(7016), 466–472. doi:10.1038/nature03000. PMID 15565145. — [preclinical / animal]
6. ClinicalTrials.gov. TB-500 (Thymosin Beta 4 17–23 Fragment) for Cardiovascular Biomarkers in Stable ASCVD. Identifier NCT07487363 (registered 2026). — [clinical — registered Phase 1/2 safety, no results]
7. Goldstein, A. L., Hannappel, E., & Kleinman, H. K. (2005). Thymosin β4: actin-sequestering protein moonlights to repair injured tissues. Trends in Molecular Medicine, 11(9), 421–429. — [review]
8. World Anti-Doping Agency (2026). The 2026 Prohibited List, Section S2 (Peptide Hormones, Growth Factors and Related Substances) — thymosin-β4 and its derivatives, e.g. TB-500. — [regulatory]
9. Educational study-synthesis (licensed-MD secondary source): TB4/TB500 deep dive, summarising thymosin β4 human ocular, dermal and cardiac data. — [educational/anecdote, not primary literature]

---

Research and educational use only. This article summarises registered trials and published or educational sources for informational purposes; it is not medical advice and not a recommendation to use TB-500 or thymosin β4, which are not FDA-approved for human use. A registered clinical trial is not evidence of safety or efficacy. Evidence is graded honestly: where a study used a different molecule or route, was small, missed its endpoint, or is company-reported rather than peer-reviewed, it is labelled as such. Consult a qualified clinician before acting on any health information. — Indexa Labs Research Team �������