Thymosin Beta-4: Real Human Trials, But the Best Indication Has Missed Two Phase 3s

Thymosin beta-4 stands out from most “research peptides” for one simple reason: it has actually gone through real, registered, peer-reviewed human trials. That is its honest claim to fame. But it is also where the story gets complicated, because its most-studied use has now fallen short of its main goal in two separate late-stage (Phase 3) trials. On top of that, people constantly mix it up with TB-500, a different molecule you’ll find on the gray market that has much weaker evidence behind it.

What it is

Thymosin beta-4 (Tβ4 for short) is a protein your body makes naturally. It is made of 43 amino acids (the small building blocks that make up proteins) and weighs about 4,982 daltons (a tiny unit for measuring molecular mass). You’ll find it in nearly every human cell. It is built from a gene called TMSB4X, and when it floats loose in fluid it has no fixed shape (scientists call this “intrinsically disordered”), with one end chemically capped.

Its main known job inside a cell is something called actin sequestration — basically, keeping a building material in reserve. Actin is a building block of the cell’s internal scaffolding, and it comes in two forms: loose single pieces (G-actin) and long assembled chains (F-actin). Tβ4 grabs the loose pieces one at a time and holds onto them, keeping a backup supply and stopping them from snapping together into chains on their own. Detailed imaging (crystallography, a method that maps a molecule’s exact 3D shape) shows the peptide caps both ends of each loose actin piece so it can’t join a chain. A short stretch of it called the LKKTET motif is the classic part that does the binding, though almost the whole peptide ends up touching the actin.

Beyond that actin job, Tβ4 released outside the cell is reported to help cells move around, encourage new blood vessels to form, calm inflammation, and reduce scarring. These extra “side gig” activities are the reason people hope it could help with wound healing and tissue repair — but they have mostly been worked out in lab dishes and animals, not in people.

The claims

Real drug companies have aimed at a small, focused set of uses: healing the surface of the eye — both a condition called neurotrophic keratopathy (where the cornea loses its normal nerve supply and won’t heal) and dry eye disease — using eye drops, and healing skin and wounds using a topical gel (one rubbed onto the skin). An injectable version was tried for heart-attack and tissue repair, but that effort stalled.

The claims you see on the gray market and in off-label chatter are far bigger — and they are almost always tied to the TB-500 fragment, not the full protein. Sellers and forums promise faster healing of tendon, ligament, and muscle injuries, less inflammation, quicker recovery, anti-aging benefits, and hair growth. These are guesses stretched from animal data and from how the molecule works in theory, not results from human trials measuring real outcomes.

Here is the key point to keep straight: TB-500 is not the full Tβ4 protein. TB-500 is a man-made seven-amino-acid snippet (Ac-LKKTETQ, positions 17–23) that contains the actin-binding part. The human trials described below all tested the full 43-amino-acid protein (called timbetasin), not the snippet sold online. Vendors often point to these trials as if they prove TB-500 works; they don’t. And just to add to the confusion, there’s yet another separate snippet called Ac-SDKP (goralatide) — also its own distinct molecule, and also not TB-500.

The evidence

Safety / Phase 1. The main human safety data comes from Ruff and colleagues (2010), an early-stage (Phase 1) trial that was randomized and placebo-controlled (meaning some volunteers unknowingly got an inactive look-alike for comparison). They gave synthetic Tβ4 by IV (intravenously — straight into a vein) to healthy volunteers: four groups of ten, with single doses of 42, 140, 420, and 1,260 mg, then daily dosing for 14 days. It was well tolerated, with no dose-limiting toxicities (side effects bad enough to force lowering or stopping the dose) and no serious adverse events, and the amount in the blood rose in step with the dose. A separate first-in-human trial of a lab-grown (recombinant) version in Chinese volunteers (Wang 2021) used much smaller IV doses (micrograms per kilogram of body weight) with daily dosing for 10 days, and again found only mild-to-moderate side effects and no dose-limiting toxicity. Both trials were small, short, and run by or tied to the companies funding them.

Eye surface — the best efficacy signal, still preliminary. A mid-stage (Phase 2) dry-eye trial (Sosne 2015) of the eye-drop version reported less eye discomfort and less corneal staining (a dye test that reveals damage on the eye’s surface) — but it included only nine patients, so at best it can suggest a hypothesis, not prove one. The full Phase 3 dry-eye program (the ARISE trials) gave mixed results: ARISE-3 did not hit its pre-chosen co-primary endpoints (the main measures of success it set out to beat in advance), with only a secondary symptom (a gritty feeling in the eye) reaching statistical significance (a result unlikely to be due to chance). And those results came mainly through company press releases rather than fully peer-reviewed papers.

The neurotrophic keratopathy program is the one most often cited. The published Phase 3 (SEER-1, 2022) is where the frequently quoted “60% vs 12.5%” healing figure comes from — but that was a missed primary endpoint (p=0.0656, meaning it fell just short of the agreed bar for statistical significance), in a trial stopped early at just 18 of a planned 46 patients because this rare disease enrolled people too slowly. A secondary endpoint measured at day 43 did reach significance. A separate European Phase 3 (SEER-3) also missed its primary endpoint, which the sponsor blamed on a stronger-than-expected placebo response (patients on the dummy treatment improving more than predicted). A US trial (SEER-2) started enrolling in 2023.

Wound / dermal. A Phase 2 trial in venous stasis ulcers (slow-healing leg sores caused by poor blood return in the veins) (Guarnera 2010) was run as a placebo-controlled, step-up-the-dose study across European sites. A trial in epidermolysis bullosa (a genetic disease that makes skin blister and tear very easily) was stopped around 2012 because not enough patients signed up.

Heart and stroke — human data essentially absent. The injectable heart-attack trial was withdrawn and never even started. Animal studies in the 2000s (notably papers in Nature in 2004 and 2007) reported that Tβ4 could rally repair cells on the heart’s surface and help grow new blood vessels. But later work (Zhou 2011) found that giving Tβ4 after a heart attack increased the density of tiny blood vessels (capillaries) without turning those cells into new heart muscle — so the regeneration claim is actually disputed in the situation that matters clinically (after an injury), not simply confirmed. A widely cited stroke “dose-response” paper is a study in rats, not humans — a real citation trap to watch for. There is no completed human trial showing it works for stroke or the heart.

What’s missing. There is no positive, repeated, properly sized Phase 3 for any use; the strongest candidate, neurotrophic keratopathy, has missed its primary endpoint in two Phase 3 trials. There are no independent (non-company) trials confirming it works, no long-term safety data, and basically no dedicated human efficacy data for the TB-500 fragment that most people actually buy.

Safety and side effects

Within the trials, the short-term human safety looks reassuring. IV Tβ4 up to 1,260 mg (both as a single dose and over 14 days) was well tolerated, with no dose-limiting toxicities and no serious adverse events; the lab-grown version caused only mild-to-moderate effects; and the eye drops were consistently described as safe and well tolerated, with side-effect rates similar to placebo.

But the limits of this data matter just as much as the good news. It comes from small groups, short time frames, carefully picked patients, and mostly company-run studies. There is no long-term human safety data, and none in children or during pregnancy. There’s also a theoretical worry: a substance that encourages new blood vessels, cell movement, and cell survival could, in principle, also help a tumor grow. That has not been studied in humans — so it is an open question, not a proven harm. Separately, gray-market products (usually TB-500) carry real-world risks of wrong labeling, impurities, and contamination that have nothing to do with the molecule itself. None of this is medical advice.

Legal and regulatory status

Thymosin beta-4 is not approved for any use in the US, EU, UK, or Australia. The FDA granted it orphan-drug designation for neurotrophic keratopathy in 2013 — a status meant to encourage development of treatments for rare diseases — but that designation is not the same as an approval. A 2011 clinical hold (a regulator-ordered pause) on the injectable heart-attack program was about manufacturing problems (cGMP, the FDA’s good-manufacturing rules) at a contract factory, not a safety problem; that heart trial was later withdrawn anyway. China’s lab-grown version is in clinical trials, with no confirmed regulatory approval.

For US pharmacy compounding (when a pharmacy custom-mixes a drug), the rules are unsettled and have shifted around a lot — check the current FDA 503A bulk drug substances list rather than trusting any vendor’s claim.

In sport, Tβ4 and its close relatives are banned at all times by WADA (the World Anti-Doping Agency) under Section S2 (peptide hormones, growth factors, and related substances), a ban in force since the 2012 Prohibited List; it showed up in high-profile Australian doping cases. Athletes should check the exact wording of the current year’s list themselves. Products sold online (mostly as TB-500) are labeled “research use only,” are not pharmaceutical grade, and have no legal path for personal therapeutic use in the US.

Bottom line

Thymosin beta-4 is a real molecule with real, registered human trials — which already sets it apart from most peptides marketed for “healing.” But the human evidence is thin and uneven. The safety record is short-term and company-run, the eye-surface program has missed its main goals in two Phase 3 trials, and the heart-and-stroke story rests on animal work that is disputed for the very situation that counts. There is essentially no human evidence that the TB-500 fragment most buyers use actually works. Treat the sweeping recovery and anti-aging claims as unproven.

Evidence grade: 6/10 · Preliminary.

Sources

• Ruff et al. A randomized, placebo-controlled, single and multiple dose study of intravenous thymosin β4 in healthy volunteers. Ann N Y Acad Sci, 2010.
• Ruff 2010 trial record (NCT00743769).
• Wang et al. A first-in-human, randomized, double-blind, single- and multiple-dose phase I study of recombinant human thymosin β4 (PMC).
• Sosne et al. SEER-1: Phase 3 trial of RGN-259 in neurotrophic keratopathy. Int J Mol Sci, 2022 (PMC).
• HLB Therapeutics misses primary endpoint in Phase 3 SEER-3 trial of RGN-259 (Ophthalmology Times).
• Phase III ARISE-3 trial of RGN-259 fails to meet endpoint in dry eye syndrome (Medthority / RegeneRx).
• Guarnera et al. Thymosin β4 in venous stasis ulcers (Phase 2). Ann N Y Acad Sci, 2010.
• Epidermolysis bullosa Phase 2 trial, terminated (NCT00311766).
• Acute myocardial infarction trial, withdrawn (NCT01311518).
• Smart & Riley. Thymosin β4 induces adult epicardial progenitor mobilization and neovascularization. Nature, 2007.
• Zhou et al. Thymosin β4 treatment after myocardial infarction does not reprogram epicardial cells into cardiomyocytes (PMC).
• Dose-response study of thymosin β4 for acute stroke (rat model, PMC).
• Structural basis of actin sequestration by thymosin-β4. EMBO J, 2004 (PMC).
• Low, Hu & Goldstein. Complete amino acid sequence of bovine thymosin β4. PNAS, 1981.
• Progress on the Function and Application of Thymosin β4 (review, Frontiers in Endocrinology, 2021).
• RegeneRx gets FDA orphan-drug status for Tβ4 in neurotrophic keratopathy (2013).
• RegeneRx Phase 2 AMI trial on clinical hold due to GMP compliance issues (Fierce Biotech, 2011).
• WHO recommends INN for thymosin beta-4 as “timbetasin” (2018).
• World Anti-Doping Agency. The Prohibited List.