TB-500 is sold as a recovery peptide and discussed as though its side effects were catalogued. They are not. When FDA's reviewers evaluated it in 2026 they found no human study of it, no animal toxicity study and no adverse-event report, and wrote that the risks in humans are unknown. What circulates as "TB-500 side effects" is mostly borrowed from trials of thymosin beta-4, a much larger protein given by different routes for different conditions. This page sets out both records separately, with sources and dates, and says what neither can answer. It reports; it does not reassure or warn, and a decision about a symptom or about use belongs with a clinician. The dose figures that circulate, and where they come from, are on our TB-500 dosage page. How brands appear on this site is on our disclosure page.
What FDA looked for, and did not find
TB-500 was reviewed for FDA's Pharmacy Compounding Advisory Committee meeting of July 23–24, 2026 (FDA briefing document, TB-500-related bulk drug substances, evaluation dated 2026-05-15). The reviewers searched PubMed, Embase, ClinicalTrials.gov, FAERS and the agency's food and supplement complaint system. The result reads as a list of absences:
| What was searched for | What FDA found |
|---|---|
| Clinical studies of TB-500, any route | None — "no articles were found in which TB-500 was administered to humans" |
| Human pharmacokinetics | None |
| Published case reports of harm | None |
| FAERS reports, through 2025-03-26 | None |
| Food/supplement complaints, 2004 to 2025-03-10 | Two, about "blended TB-500 and BPC-157", with no safety assessment |
| Acute and repeat-dose toxicity studies | None |
| Genotoxicity, reproductive and carcinogenicity studies | None |
| Immunogenicity or aggregation studies | None |
The nominator cited three papers in support; all three were studies of thymosin beta-4, not TB-500. FDA's conclusion: "There is a lack of clinical and nonclinical safety information on the use of TB-500 … potential safety risks associated with the use of these substances in humans are unknown." It recommended against adding TB-500 to the list of substances pharmacies may compound. The committee's advisory vote went the other way, as covered on our Category 2 list page; an advisory vote adds no safety data.
The one concern FDA did spell out
With no observations to report, the review's only specific safety discussion is theoretical. Peptides given by injection can provoke an immune response, and subcutaneous injection is among the routes most associated with it. FDA wrote that TB-500, "a peptide that is administered through injectable ROAs (SC and IM), … may pose a significant risk for immunogenicity, potentially amplified by aggregation as well as potential peptide-related impurities." The consequences it describes range from antibodies with no visible effect to severe reactions, including antibodies that neutralise the body's own version of the peptide. No study has checked whether any of this happens with TB-500. It is a stated unknown, not a finding.
Why thymosin beta-4 data does not transfer
Thymosin beta-4 is a 43-amino-acid protein present in most human cells. TB-500 is Ac-LKKTETQ: seven residues copied from the protein's actin-binding region, with an acetyl group added. FDA's pharmacology reviewers were explicit that the two cannot be treated as one. Acetylation "irreversibly alters their charge, hydrophobicity, and size," so the fragment's profile "cannot be directly extrapolated" even from the non-acetylated seven-residue sequence, let alone the full protein.
The one direct comparison FDA located points the same way. In a 2024 laboratory study (Rahaman and colleagues), TB-500 free base at 50 µg/mL did not speed the closing of scratch wounds in cultured fibroblasts, while one of its breakdown products, Ac-LKKTE, produced a small but significant effect. The pharmacokinetic work that does exist is animal and in-vitro: in horses given 10 mg under the skin, blood levels peaked at 0.05–0.08 ng/mL one to two hours later and were unmeasurable by six to ten hours, with the peptide trimmed from its tail end into shorter fragments in horses, rats and human serum and tissue preparations alike.
This matters for side effects because a borrowed safety record is only as good as the match between the molecules, the routes and the doses. For TB-500 all three differ.
What the thymosin beta-4 trials recorded
Because these are the numbers people quote, they are worth reading accurately. Every trial below gave the full protein, and all but one applied it to the eye or skin rather than injecting it.
| Trial (registry) | Who and how | Adverse events posted |
|---|---|---|
| NCT03937882 | 699 with dry eye; eye drops four times daily, ~4 weeks | No serious AE either arm. Instillation-site pain 23/350 drug vs 16/349 placebo; reduced visual acuity 12 vs 7; irritation 5 vs 2 |
| NCT02974907 | 601 with dry eye; eye drops, up to day 29 | Serious AE 3/299 drug (diverticulitis, vertigo, hyperglycaemia) vs 2/302 placebo |
| NCT00832091 | 72 with venous leg ulcers; topical gel at three strengths | Serious AE 3/55 drug vs 1/17 placebo; raised sedimentation rate in 6/55 on drug, 0/17 on placebo |
| NCT00382174 | 71 with pressure ulcers; topical gel up to 84 days | Serious AE 13/53 drug vs 5/18 placebo, in a frail, ill population; no pattern named |
| Ruff 2010 (PubMed 20536472) | 40 healthy volunteers; 42–1,260 mg IV, single then daily for 14 days | "Infrequent, and mild or moderate"; no serious AE or dose-limiting toxicity |
Two readings of that table are ours. First, the largest body of human thymosin beta-4 data is eye drops, where the most common complaint is stinging at the eye; it says little about a protein, let alone a fragment, injected under the skin. Second, the one intravenous study was 40 people for two weeks — ten per dose cohort, placebo included — which is a first-in-human tolerability study, not a safety record for repeated use.
FAERS tonight
We ran our own query of FDA's adverse-event database through openFDA on 2026-09-26 (data last updated 2026-07-30). The query and the split between "suspect" and "concomitant" products are part of the result, so both are stated.
- "TB-500" (and "TB 500"): one report, received 2026-05-27. TB-500 appears only as one of three peptides in a single blended vial (with GHK-Cu and BPC-157), coded as concomitant; the suspect products were chorionic gonadotropin and Ozempic. The report codes a malignant lymphoid neoplasm and "product dispensed by unauthorised provider". It cannot attribute anything to TB-500.
- "Thymosin beta-4": three reports. In one it is one of about thirty concomitant products; one codes "drug ineffective" alongside a multiple-sclerosis drug; one describes malaise and a product-label and dispensing error in a person also using sermorelin, BPC-157 and semaglutide.
- "TB500" (no space or hyphen): none.
So the database that holds 20 reports naming BPC-157 (see our BPC-157 side-effects page) holds, in effect, nothing attributable to TB-500 — an absence of reports from an unregulated product with no reporting obligation, which is not the same as an absence of harm. Our peptide side-effects overview sets that count against thirteen other compounds.
A trial record that describes itself as fictional
A ClinicalTrials.gov search for TB-500 returns one record naming the fragment: NCT07487363, a "phase 1/2 … dose-escalation study of TB-500 (Thymosin Beta 4 17-23 Fragment)" in adults with atherosclerotic heart disease, first posted 2026-03-23 and listed as recruiting at a hospital in Shenzhen, sponsor "Hudson Biotech". Its brief summary opens: "This fictional study is an example of a ClinicalTrials.gov-style record." We report the wording as it stands on 2026-09-26 and draw no conclusion about the sponsor; the practical point is that the one registered trial of the fragment has posted nothing, and its own text says not to read it as a real study. When results from any genuine trial of TB-500 are posted, this page will change.
The blend problem
Much of what is sold as TB-500 is sold in combination — with BPC-157 as the "Wolverine" stack, or with GHK-Cu and BPC-157 as "Glow" — and both FDA's complaint records and the one FAERS report above involve blends. A reaction to a blended vial cannot be assigned to any one ingredient, which is why the component records have to be read one at a time. Our Wolverine stack page covers what is documented about the combination.
What the record does not contain
- No human study of TB-500 by any route, so no incidence figure for any side effect.
- No animal toxicity study of any duration, so no organ-level signal to watch for and no basis for a "safe" dose.
- No carcinogenicity or reproductive study.
- No measurement of antibodies or aggregation in any TB-500 product.
- No data on long-term or repeated use.
Anyone weighing TB-500, or wondering whether a symptom is connected to it, should take the question to a clinician, bringing the product's label and certificate of analysis — how to read one is covered separately.
Sources and dates
- FDA. Evaluation of TB-500-Related Bulk Drug Substances, Pharmacy Compounding Advisory Committee briefing document, July 23–24, 2026 meeting (evaluation dated 2026-05-15). fda.gov/media/193349. Read 2026-09-26.
- ClinicalTrials.gov records NCT03937882, NCT02974907, NCT00832091, NCT00382174 (posted results) and NCT07487363. Read 2026-09-26.
- Ruff D, Crockford D, Girardi G, Zhang Y. A randomized, placebo-controlled, single and multiple dose study of intravenous thymosin beta4 in healthy volunteers. Ann N Y Acad Sci. 2010;1194:223–9. PMID 20536472.
- Wang WC, et al. Thymosin β4, a potential marker of malignancy and prognosis in hepatocellular carcinoma. Scand J Gastroenterol. 2023;58(4):380–91. PMID 36269095.
- openFDA drug adverse event endpoint, queries for "TB-500", "TB 500", "TB500" and "thymosin beta-4", run 2026-09-26 (dataset updated 2026-07-30).
- PubMed search "TB-500 OR TB500", 2026-09-26: 33 records, several unrelated (tuberculosis, chemistry); none a human study of the fragment.