Thymosin beta-4

A cell crawls by building actin filaments at its front edge and dismantling them at the back. Thymosin beta-4 keeps a pool of loose actin ready for that work. In skin models, this shows up as faster migration of keratinocytes and other repair cells into a wound, along with less inflammation, cell death and scarring.

A short stretch near the middle of the protein, seven amino acids that bind actin, drives its effect on new blood vessels. In 2003, NIH researchers found that this motif matched the full protein’s activity in endothelial cell migration and vessel-sprouting assays, while fragments missing part of it were inactive. That stretch is what TB-500 copies.

Other parts of the protein do other jobs. In heart cells, thymosin beta-4 forms a complex with PINCH and integrin-linked kinase that switches on Akt, a survival signal. Its front end is cut by two enzymes, meprin alpha and prolyl oligopeptidase, into a four-amino-acid peptide called Ac-SDKP, which appears to carry most of the protein’s anti-fibrotic activity in animal models of liver, lung, heart and kidney scarring.

Put together, thymosin beta-4 acts as a coordinator of repair. It helps cells move, helps vessels grow, helps stressed cells survive and helps keep scarring in check. That breadth is why one protein reached trials in three very different tissues.⁷³¹⁰⁵⁴

The cornea is a good place to test a repair protein. Its surface is easy to see, wounds can be measured with a simple dye, and eye drops put the protein exactly where it is needed. The RGN-259 program, run by ReGenTree, chose two conditions where existing options often fall short.

The first was severe dry eye. In a small phase 2 trial, nine patients used RGN-259 or a vehicle drop for 28 days and were followed for another 28. At day 56, the treated eyes had 35.1% less ocular discomfort and 59.1% less corneal staining, a sign of surface damage, than the vehicle eyes. Tear breakup time and tear volume also improved.

The second was neurotrophic keratopathy, a rare disease in which the cornea loses its nerve supply and surface wounds stop healing. This became the phase III trial that anchors the whole program, and its result deserves a closer look than a headline gives it.

The patients in that trial had stage 2 disease, a persistent epithelial defect, meaning a patch of missing surface cells that will not close on its own, or stage 3, where the defect has deepened into a corneal ulcer. These are hard wounds by design. On the Mackie scale doctors use to stage the disease, the treated group improved over the following weeks, reaching significance by day 43.¹³¹²¹⁷

In the neurotrophic keratopathy trial, 6 of 10 people using RGN-259 had a completely healed corneal defect at four weeks. On placebo, 1 of 8 did. Most readers would call that a clear win. The statistics called it a trend, with a p-value of 0.0656, just above the 0.05 line that conventionally marks a result as significant.

With only 18 patients, that line is unforgiving. A single extra healed patient on the drops, or one fewer on placebo, could have moved the result to the other side. The honest reading is that the trial did not meet its main endpoint, and its small size leaves real uncertainty in both directions. It neither proves the drops work nor shows that they fail.

The secondary results lean the same way. Two weeks after treatment stopped, healing on RGN-259 was significant (p = 0.0359) and no treated patient had a recurrence, while the one placebo patient who healed had a recurrence. Comfort, foreign-body sensation and dryness improved significantly, and no significant side effects were seen.

Two more facts complete the picture. The trial was designed by employees of the developer with an independent contract research organization running and analyzing it, and the authors disclose those ties. And a second phase 3 trial, SEER-2, is recruiting. That trial, more than any forum post, will decide whether thymosin beta-4 becomes an approved eye medicine.¹²¹⁷

Chronic wounds are where thymosin beta-4’s animal record is most consistent. It sped the healing of full-thickness punch wounds in normal rats and mice, and in models that heal poorly, including steroid-treated rats, diabetic mice and aged mice. Those are the conditions in which human wounds also stall.

In people, two phase 2 trials tested topical thymosin beta-4 in venous stasis ulcers and pressure ulcers. Among patients whose ulcers healed, healing came almost a month sooner. A European study across Italy and Poland was designed with 72 patients randomized three to one to the protein or placebo, alongside compression therapy.

The most hoped-for skin use was epidermolysis bullosa, a genetic condition in which skin blisters and tears with minor friction. A 2019 review argued that faster healing, less inflammation and less scarring are exactly what these patients need. The phase 2 trial of RGN-137 gel, though, stopped after four patients for a business decision, so that question remains open.

Epidermolysis bullosa also shows why the protein’s breadth matters. Beyond speeding wound closure, thymosin beta-4 modulates cytokines, including those that drive itching, and reduces inflammation and scarring in skin models. For patients whose blisters itch, get infected and scar, a single molecule acting on several of those problems at once is the appeal.⁸⁹⁷¹⁸

In 2004, a Nature paper from UT Southwestern showed that thymosin beta-4 helps embryonic and postnatal heart cells migrate and survive. In mice with a tied-off coronary artery, treatment raised activity of the survival signals integrin-linked kinase and Akt, improved early heart muscle survival and improved heart function. The heart became the protein’s most ambitious target.

The first human phase I study, published in 2010, was designed with heart attacks in mind, and found intravenous thymosin beta-4 well tolerated. An injectable form, RGN-352, was registered for a phase 2 heart attack trial that was withdrawn before it began.

The next human heart trial came from China, using a recombinant version. A 2025 trial randomized 96 people after a major heart attack and an opened artery. Across the whole group, infarct size at 90 days was not significantly different from placebo. In a subgroup of 43 patients whose first dose came within eight hours after their artery was opened, the infarcted area was significantly smaller. The authors called for larger trials focused on early dosing.

The mouse half of that 2025 paper added a new piece to the mechanism. The recombinant protein reduced heart scarring and lowered NT-proBNP, a blood marker of heart strain, after an induced heart attack. In heart cells in a dish put through low oxygen and reoxygenation, it switched on a growth signal called ErbB2. In mice, blocking ErbB2 erased the protein’s benefit to heart function and remodeling. Two decades after the Nature paper, the pathways are still being mapped.^10141911

TB-500 is a synthetic copy of seven amino acids from thymosin beta-4, positions 17 to 23, with an acetyl cap added to one end. Anti-doping labs identified it in 2012 while analyzing products sold under that name. It contains the actin-binding motif that drives cell migration and vessel growth in lab studies.

It does not contain the front end of the protein, the source of Ac-SDKP and most of the anti-fibrotic activity, and it has never been tested in the eye, skin or heart trials described here. Those trials used full-length thymosin beta-4, in eye drops, gels or intravenous infusions.

Regulation now splits them too. FDA’s October 2026 enforcement discretion lets 503A pharmacies compound TB-500 but does not cover the full protein. So the version with the human trial record is the one a compounding pharmacy cannot currently prepare under that policy, and the version a pharmacy can prepare is the one without a trial record of its own.

The two names carry different evidence. Anyone citing the eye or heart trials in support of TB-500 is borrowing results from a larger molecule, and the fragment’s own case still rests on cell studies of its actin-binding motif.¹⁶³⁴²³

Small, short-term human trials have found it generally well tolerated, though they cannot rule out rare or long-term harms. The 2010 phase I study randomized 40 healthy volunteers in four rising-dose groups to intravenous thymosin beta-4 or placebo, first as a single dose and then daily. Adverse events were infrequent and mild or moderate, with no dose-limiting toxicities or serious adverse events.

A 2021 phase I study of a recombinant version in 84 healthy Chinese volunteers found the same pattern, with no obvious build-up after repeated daily dosing. The eye trials reported no significant adverse effects. Several of the trials involved companies developing the protein, which the authors disclose.

The open question is cancer. In a 2003 mouse study, melanoma cells engineered to overproduce thymosin beta-4 grew larger tumors, formed more than four times as many lung nodules and grew about four times as many blood vessels. That was the tumor’s own protein, not a treatment, but it is the reason a personal cancer history belongs in any conversation about it.

Other groups that should talk to a clinician first include anyone pregnant or breastfeeding, anyone with liver or kidney disease, anyone with a recent heart attack or heart condition, and tested athletes.^1415126

Platelets are where thymosin beta-4 comes from at a wound, though no study has made platelet count a test for this protein. As general baseline context, it screens for problems with clotting and the bone marrow that a clinician would want to know about. hs-CRP, high-sensitivity C-reactive protein, measures low-grade inflammation across the body, one of the forces that keeps wounds from closing.

ALT, alanine aminotransferase, is a liver enzyme that rises when liver cells are under stress, and eGFR estimates how well your kidneys filter. They give a clinician general context on liver and kidney health. They cannot establish whether thymosin beta-4 is safe for you, and whether they are useful to repeat depends on the medicine and your circumstances.

Vitals Vault starts peptide care with that kind of panel at Quest Diagnostics. A licensed clinician reviews the results and your history, explains which options are available under current rules, and retests the same markers at the end of any course.⁸

No thymosin beta-4 product is FDA-approved. RGN-259 eye drops are in a recruiting phase 3 trial, SEER-2, for neurotrophic keratopathy. The skin gel and injectable heart programs in the US stopped or never started.

FDA’s October 29, 2026 enforcement discretion covers six peptides for 503A compounding, including the TB-500 fragment but not full-length thymosin beta-4. In sport, WADA prohibits thymosin beta-4 and all its derivatives.^1718192324

Thymosin beta-4 is a common protein with an uncommon résumé. It coordinates cell movement, blood vessel growth, cell survival and scar control, and it has been tested in people on the eye, skin and heart. The results range from encouraging to inconclusive, and its most promising trial missed significance by a hair in 18 patients.

If you are weighing it against TB-500 or BPC-157, the useful question is which evidence you are relying on. For eye and skin surfaces, thymosin beta-4 has human data. For tendons and sports injuries, no controlled human trial in this family has been published, only scattered uncontrolled reports, and the next question is how a clinician and your labs can make any trial you try a measured one.¹²⁸¹¹