Start with the identity
FDA describes TB-500 as an acetylated seven-amino-acid fragment from thymosin beta-4. Full-length thymosin beta-4 is a different molecule. A paper mentioning the full-length protein is not automatically evidence for the fragment sold under the TB-500 name. [source]
This distinction affects every later question: proposed mechanism, tissue exposure, dose, safety and clinical results. The library uses the exact study compound wherever possible rather than grouping similar names into a single treatment.
Biology versus demonstrated benefit
FDA’s TB-500 review did not find sufficient nonclinical evidence to support wound-healing use of the fragment. It also identified major toxicology gaps. A plausible role for related thymosin biology is not the same as a demonstrated effect of this exact substance. [source]
Our interpretation: a fragment can retain, lose or alter properties of its parent molecule. Its identity and activity have to be demonstrated, rather than inferred from its name or from a conceptual illustration.
What human studies actually tested
Recombinant thymosin beta-4 in healthy volunteers
The NL005 study included 54 participants in a single-dose trial and 30 in a multiple-dose trial. It used intravenous recombinant human thymosin beta-4 and assessed tolerability, drug handling and antibody responses. No serious adverse events or dose-limiting toxicities were reported; other events were mild to moderate.
How to read it: This was short-term research in healthy people, not a trial of TB-500 for tendon healing. Some authors worked for the sponsor; that disclosure matters alongside the design and results. [source]
Thymosin beta-4 eye drops for dry eye
Participants received an ophthalmic formulation or placebo for 28 days. Neither primary endpoint—ocular discomfort or inferior corneal staining—showed a significant between-group difference at the specified visit. Some secondary measures favored treatment.
How to read it: Report the primary results alongside the encouraging secondary observations. An eye-drop study cannot establish that injected TB-500 repairs muscles, and a secondary signal does not erase a missed primary endpoint. [source]
The BPC-157 combination report
A small knee-pain chart review included four people receiving BPC-157 with a substance described as thymosin beta-4.
How to read it: This subgroup was not a randomized comparison of combination treatment against either component. It cannot validate products marketed as a BPC-157/TB-500 stack. [source]
What remains unknown
At the time of its assessment, FDA had not identified human clinical studies establishing TB-500 fragment safety or pharmacokinetics. Uncertain safety data do not justify describing it as safe because it resembles a naturally occurring substance. [source]
Related-compound studies also have limits. The Chinese trial’s brief follow-up and healthy-volunteer population cannot answer questions about repeated use in injured patients, interactions, rare events or pregnancy. Those would require appropriate studies. [source]
Approval and compounding are different questions
The FDA documents reviewed do not establish an approved TB-500 wound-healing drug. The 2026 advisory process considered compounding ingredients; it was not approval of an injury-recovery indication. We have not verified a later final compounding determination. [source]
Even when a drug can lawfully be compounded, the finished compounded product is not FDA-approved. Pharmacy accreditation and eligibility rules are also distinct from proof of a particular clinical claim. [source] [source]
How to evaluate a proposed protocol
First match the compound. Then match the route, population and outcome. A fragment injection, a recombinant full-length intravenous preparation and an eye drop should not share a protocol simply because their names overlap.
The evidence reviewed does not establish a general TB-500 loading phase, maintenance cycle or recovery stack. This is an evidence gap, not proof that no future use can work. A useful next step in research would be a well-characterized product studied prospectively with a comparator and clinically meaningful endpoints.
Questions to take to a licensed clinician
- Is the cited molecule the fragment or full-length thymosin beta-4?
- Does the study measure healing, symptoms, or only short-term tolerability?
- Did its main planned endpoints succeed?
- What evidence directly supports the proposed formulation and route?
Sources & review scope
This is a focused, selected-source review, not a systematic review or a complete adverse-event inventory. Regulatory sources are dated snapshots. Publication in a journal or indexing in PubMed does not mean FDA endorsement.
- FDA pharmacy-compounding review: TB-500 (2026)
Agency review of the fragment; not a final compounding determination.
- Wang and colleagues. Recombinant thymosin beta-4 phase 1 (2021)
Randomized healthy-volunteer study in China. DOI: 10.1111/jcmm.16693.
- Sosne & Ousler. Thymosin beta-4 eye-drop phase 2 trial (2015)
Randomized study; DOI: 10.2147/OPTH.S80954.
- Lee & Padgett. Knee-pain chart review (2021)
Original human report; retrospective, uncontrolled.
- FDA July 23–24, 2026 Pharmacy Compounding Advisory Committee
Dated committee materials; advisory consideration is separate from drug approval.
- FDA: Compounding questions and answers
Explains approval, product quality and compounding.
- NABP: Unapproved Prescription Drug Evaluation Policy
Pharmacy-accreditation policy; not an endorsement of these compounds.
Edition 1 adds the initial evidence summary, limitations, safety context and discussion questions. No independent clinical sign-off has been recorded.