The short version

Four things, before anything else.

• The definitive modern trial is negative. TESTS — 1,106 patients, 22 centres, double-blind, placebo-controlled, published in the BMJ in 2025 — found no mortality benefit in sepsis. Hazard ratio 0.97, p = 0.82.
• The trial everyone cites as positive did not meet its primary endpoint. ETASS, the 361-patient study quoted in almost every vendor description, reported 28-day mortality of 26.0% against 35.0% — at p = 0.062. That is a miss, and it is reported as a miss in the paper itself.
• It is not FDA approved. It holds orphan designations, which are not approvals. FDA's own register says, in those words, “Not FDA Approved for Orphan Indication.” It sits on FDA's Category 2 do-not-compound list.
• For what it is actually sold for online — immune support, chronic fatigue, Lyme, longevity — there is essentially nothing. For Lyme specifically we found zero clinical trials of any design.

What thymosin alpha-1 is

A 28-amino-acid fragment of a larger protein, capped at one end — and that cap is the whole molecule's identity.

Thymosin alpha-1 is the N-terminal 28 residues of prothymosin alpha, released by an enzyme called legumain. The parent protein is acetylated at its N-terminus, so the fragment is too. This is a genuine post-translational modification, not a synthetic convenience.

It is a strongly acidic peptide, and almost nothing written about peptide handling applies

Count the charges. Three aspartates, six glutamates, one free C-terminal carboxyl — ten anionic groups. Against that, four lysines and nothing else basic: no arginine, no histidine. And the alpha-amino group, which would normally carry a positive charge, is blocked by the acetyl.

Why the acetyl cap is not cosmetic

• It blocks the only primary amine, protecting the peptide from the enzymes that chew peptides from the N-terminal end.
• A des-acetyl impurity is a different molecule — 42 Da lighter, one charge different, different chromatographic behaviour. Material supplied without the acetyl is not thymosin alpha-1, whatever the label says.

Disordered in water, structured only at an interface

In ordinary aqueous buffer this peptide has no stable secondary structure. The published NMR structure had to be solved in 40% trifluoroethanol, a membrane-mimicking solvent, where residues 14–26 form a helix. Anyone quoting “helical structure” for thymosin alpha-1 in water is describing a solvent that is not in the vial.

One useful consequence, and it runs in our favour: the sequence contains no tryptophan, tyrosine, phenylalanine, methionine or cysteine. Photodegradation and oxidation — the two things most peptide storage advice is written around — are genuinely low risks here. The expected degradation route is deamidation at Asn28 and the aspartates, which is driven by pH and temperature, not light.

Sepsis — the largest trial ever run, and it is negative

This is the section that matters most, because sepsis is where the strongest modern evidence sits and the direction of that evidence is the opposite of what is usually claimed.

The meta-analysis, and why it does not rescue the picture

A 2025 meta-analysis pooled 11 randomised trials, 1,927 patients, and reported an overall odds ratio of 0.73 (p = 0.003) for 28-day mortality. That figure circulates widely. Here is what sits underneath it:

Six of the eleven trials were single-centre studies with between 20 and 106 patients. The pooled effect is carried entirely by them. Restrict to the trials that were well-conducted, or to the ones run across more than one hospital, and the effect disappears. The same paper's trial sequential analysis states plainly that “the current sample size is inadequate.”

That is a textbook small-study effect, and it is the reason a well-powered double-blind trial was worth running. It was run. It was negative.

Hepatitis B and C — the oldest indication

This is where the compound came from and where its one real marketing authorisation lives. It is also thinner than its reputation.

Hepatitis B

The most-cited support is a 2008 meta-analysis comparing thymosin alpha-1 against interferon alpha. Its total evidence base is four randomised trials and 199 patients, restricted to English-language publications, with no formal validity assessment — a limitation the Centre for Reviews and Dissemination noted in its own critique of the paper.

The effect appears after treatment stops, not during it — which is at least consistent with an immunomodulator rather than a direct antiviral. The registration dossier's pooled analysis (n = 223) reports 36% response against 19% at twelve months.

Worth knowing: there is no completed Cochrane review of thymosin alpha-1 for hepatitis B. The 2022 Cochrane record often cited as one is a protocol — a statement of intent to review, not a review.

Hepatitis C — the largest trial in the field was negative

You will very often see “41% versus 26%, p = 0.048” attributed to this trial. That figure is real, and it is the per-protocol completers subgroup — the patients who finished 48 weeks of a demanding regimen. Analysing only the people who completed treatment is exactly the analysis that intention-to-treat exists to guard against, because the patients who drop out are not a random sample of the patients who enrol.

The honest sentence is: the largest hepatitis C trial of thymosin alpha-1 was negative on its primary analysis, and a secondary analysis of completers was positive. Anyone quoting only the second half is quoting half a result.

COVID-19 — what seventy-six patients became

A single unadjusted retrospective series became, in circulation, a proven mortality benefit. The larger and better-controlled studies that followed do not support it.

What it is actually sold for — and what stands behind that

Almost nobody buying this compound online is treating hepatitis B. It is sold for immune support, for chronic fatigue, for chronic Lyme, for ageing. Here is what exists for each.

Chronic fatigue syndrome / ME

ME/CFS was one of twelve indications formally nominated to FDA when the case was made to allow thymosin alpha-1 to be compounded. FDA reviewed the evidence for each and found it insufficient across all twelve. There is no controlled trial of thymosin alpha-1 in ME/CFS. One registered trial exists in cancer-related fatigue, which is a different condition.

Longevity, immunosenescence, “restoring the ageing immune system”

A 2025 review devoted entirely to thymosin alpha-1 and ageing reports no clinical trial in ageing, longevity, immunosenescence, or immune support in healthy older adults. What exists is preclinical and mechanistic work. The review's own conclusion is that further research is needed to validate long-term efficacy and safety.

General immune support in healthy people

No trial exists. Every human study of this compound — every single one — has been run in a defined disease population: hepatitis, sepsis, COVID, cancer, dialysis, transplant. There is no dataset describing what it does in a healthy adult.

The one claim that holds

There is a real, specific, population-bounded use with genuine evidence behind it — and it is not the one being sold.

As an adjuvant to vaccination in immunocompromised and elderly people, thymosin alpha-1 has supporting data and a national marketing authorisation.

• Gravenstein 1989 — double-blind, placebo-controlled, elderly men. Influenza antibody response was augmented against placebo.
• Carraro 2012 — haemodialysis patients given an adjuvanted H1N1 vaccine. Seroconversion 88–89% against 53% on placebo, p < 0.01.

This matters for how the compound gets described. “Improves vaccine response in dialysis patients and the elderly” is supportable. “Boosts immunity” is a generalisation of it that the evidence does not carry — the studies were done in people whose immune systems were measurably impaired, and there is no equivalent work in anyone else.

Where it stands — and the Canadian position

Four jurisdictions, four different answers, and one of them is frequently misreported.

Canada

Health Canada's 9 April 2026 advisory on injectable peptides bought online lists fourteen example products. Thymosin alpha-1 is not among them.

Health Canada's position, verbatim: “most injectable peptides are regulated as prescription drugs in Canada”, and products it has not authorised “are illegal to be sold in Canada.” There is no authorised thymalfasin product in Canada.

United States — not approved, and specifically restricted

• Not approved. FDA's orphan-drug register carries a designation from March 2000 and states, in terms: “Not FDA Approved for Orphan Indication.” A designation is a research incentive, not a marketing approval, and the two are constantly conflated.
• On the Category 2 do-not-compound list since 29 September 2023, for substances that “may present significant safety risks”. FDA's stated reason: “may pose significant risk for immunogenicity for certain routes of administration and may have complexities with regard to peptide-related impurities and API characterization.”
• December 2024: FDA reviewers recommended against adding it to the list of substances permitted in compounding, citing “lack of evidence to support the effectiveness.”

Europe

No centralised EU marketing authorisation. Italy authorises it — two products, Zadaxin and Jitide — for enhancing influenza vaccine response.

Sport — and the distinction that is routinely got wrong

The word “thymosin” appears exactly once in the 2026 Prohibited List, under S2.3 Growth Factors: “Thymosin-β4 and its derivatives e.g. TB-500.”

Sentient Biogenix supplies thymosin alpha-1 for laboratory research use only. It is not for human or veterinary use.

Reconstitution — where the standard advice collides with the chemistry

The usual instruction for a 10 mg vial exceeds every published solubility figure for this compound. That is worth understanding rather than ignoring.

What the approved product does

Note the design. The fill sits below the solubility ceiling, and it is buffered on top of that. The mannitol is at 31 times the mass of the active, doing three jobs: bulking the cake so it is physically coherent, protecting the peptide through freeze-drying, and bringing the solution to roughly 274 mOsm/L — essentially isotonic. None of that is incidental for a subcutaneous injection.

Published solubility

The 300-fold gap between water and buffer is not an error. A peptide carrying net −6 drifts its own pH acidic as it dissolves, protonates its own carboxylates, loses charge and falls out of solution. Held near pH 7 with counter-ions available, it stays ionised and dissolves freely. This is why the approved product is a peptide in phosphate buffer, not a peptide in water.

The arithmetic for a 10 mg research vial

What is actually in the vial — and it is not the counterion

For most peptides the counterion is the big correction. Here it is small. With only four lysines and a net negative charge at neutral pH, acetate can account for at most about 7% of the mass, and realistically much less.

One further trap: research material is supplied variously as the acetate salt, the trifluoroacetate salt, or unstated — and every supplier prints the free-base formula and molecular weight regardless. The two salts share a CAS number, so the CAS on a certificate tells you nothing about which one you have. Residual trifluoroacetate is cytotoxic at low concentrations and is a known confounder in cell work.

Storage — and the number nobody has measured

The formulated drug is a refrigerated product. Neat research powder is a frozen one. Those are not the same instruction and they are constantly conflated.

Do not cite the drug product's fridge temperature as authority for storing unformulated powder. The mannitol and buffer stabilise the peptide in a way neat lyophilisate is not.

Let a frozen vial reach room temperature before opening it. Peptides are hygroscopic and condensation lands straight on the cake — and here moisture is not merely a handling nuisance, because water content is one of the terms in the peptide-content arithmetic above. A vial that has taken up atmospheric moisture contains less peptide per milligram than it did.

In solution — the absence is the finding

The figure most often quoted, “2 to 7 days at 4 °C”, appears even in an FDA document. Traced back, it matches one chemical supplier's generic catalogue storage text near-verbatim, including the unusual range and the storage temperature — and that same boilerplate appears across that supplier's whole product line. It is not a thymosin alpha-1 determination.

The only actual measurements we could find anywhere are in a formulation patent, and they cover a mannitol/phosphate product for ten hours at room temperature.

Note also that the approved product's diluent is sterile water, not bacteriostatic — and that bacteriostatic water is unbuffered, mildly acidic and hypotonic, which is the opposite of the manufacturer's formulation in all three respects. Our general 28-day storage convention is a sensible default for handling this compound, but it should be described as an in-use window rather than a stability finding, because for this molecule nobody has measured the second thing.

Reported effects

In registration-quality trials the tolerability record is genuinely clean. The question is whether that record transfers to material that is not registration-quality.

What the trials found

• ETASS (n = 361): no treatment-related severe adverse event, no discontinuation for intolerance.
• TESTS (n = 1,106): no significant difference in safety outcomes between groups.
• Shehadeh (n = 49): no local injection reactions reported at all, serious adverse events balanced between arms and none judged related.
• Product labelling reports under 1% drug-related adverse events — injection-site discomfort, rarely erythema, rash, arthralgia.

What FDA raised, which is absent from essentially all vendor descriptions

Also on the record, and worth knowing:

• One fatal case of immune haemolytic anaemia reported following use alongside donor lymphocyte infusion.
• Potential worsening of graft-versus-host disease.
• ALT flares and thyroid function abnormalities documented.
• Contraindicated where immunosuppression is deliberate — transplant recipients — unless benefit clearly outweighs risk. An immunostimulant given to someone whose immune system is being suppressed on purpose is working against the treatment.

The distinction to hold onto: the clean safety record belongs to buffered, characterised, pharmaceutical-grade drug given in monitored trials. FDA's concern is specifically about the characterisation and aggregation behaviour of non-registered material given subcutaneously. Those are not the same product and the safety data does not automatically transfer.

Pharmacokinetics and how it is dosed

A two-hour half-life, dosed twice a week. That combination tells you something about how it is thought to work.

Worth stating plainly: with a two-hour half-life and no accumulation, the drug is gone from serum long before the next dose. The rationale for twice-weekly dosing is downstream immune signalling, not sustained plasma exposure. That is a coherent pharmacological position, but it does mean plasma concentration tells you very little about whether a dose is doing anything.

Questions we actually get

Is this the same as TB-500?

No, and the confusion matters. TB-500 is a fragment of thymosin beta-4 — a different molecule, a different function, and expressly prohibited in sport under S2.3. Thymosin alpha-1 is an immunomodulator and is not named anywhere on the 2026 Prohibited List. They share a first word and almost nothing else.

Is it FDA approved?

No. It holds orphan designations, which are research incentives, not approvals — FDA's own register says “Not FDA Approved for Orphan Indication.” It is also on FDA's Category 2 do-not-compound list. At least one peer-reviewed review repeats the approval error, which is how it reaches vendor copy.

Does it work for sepsis?

The largest and best-designed trial — 1,106 patients, double-blind — says no. Hazard ratio 0.97, p = 0.82. The trial usually cited as positive missed its own primary endpoint at p = 0.062.

What about for Lyme, or chronic fatigue?

For Lyme there are no clinical trials at all, of any design. For ME/CFS, FDA reviewed the evidence and found it insufficient. Neither claim has a controlled trial behind it.

How much water should I use for a 10 mg vial?

More than you have probably been told. Every published solubility figure puts the ceiling at 1–2 mg/mL in plain water, so 5–10 mL is where the arithmetic lands. If a vial does not go clear at 1 or 2 mL, that is chemistry, not a bad vial.

How long does it last once reconstituted?

Nobody has measured it. The manufacturer says use immediately; the “2 to 7 days” figure in circulation traces to a supplier's generic catalogue text. Our 28-day convention is a reasonable in-use window, but it is a sterility convention, not a stability finding for this molecule.

Why does my vial look empty?

A few milligrams of peptide lyophilised without a bulking agent forms a thin, wispy cake that can be almost invisible. Note that Health Canada's seizure list distinguishes “Thymosin-α1” from “Thymosin-α1 without Mannitol” as separate products — so whether there is a bulking agent in the vial varies by supplier, and it changes both the appearance and the tonicity of the resulting solution.

So why do you sell it?

Because it is a legitimate research compound with sixty years of literature behind it and researchers have their own reasons for wanting it. What we will not do is tell you it is FDA-approved, that it is proven in sepsis, or that it does anything for Lyme disease. The largest trial is negative, the regulator says it is not approved, and for the indication it is most often sold for there is no trial at all. You should know all three before you buy anything from us.

References

1. Wu J, Pei F, Zhou L, et al. The efficacy and safety of thymosin α1 for sepsis (TESTS): multicentre, double blinded, randomised, placebo controlled, phase 3 trial. BMJ 2025;388:e082583. DOI 10.1136/bmj-2024-082583. PMID 39814420. — the paper this page turns on. Correction at BMJ 2025;389:r1098.
2. Wu J, Zhou L, Liu J, et al. The efficacy of thymosin alpha 1 for severe sepsis (ETASS): a multicenter, single-blind, randomized and controlled trial. Crit Care 2013;17(1):R8. DOI 10.1186/cc11932. PMID 23327199. springer.com — the trial cited everywhere as positive. Read the primary endpoint.
3. Gu B, Zhou Y, Nie Y, et al. Efficacy of thymosin α1 for sepsis: a systematic review and meta-analysis of randomized controlled trials. Front Cell Infect Microbiol 2025;15:1673959. DOI 10.3389/fcimb.2025.1673959. PMID 40969554. — the subgroup analyses are the important part.
4. Ciancio A, Andreone P, Kaiser S, et al. Thymosin alpha-1 with peginterferon alfa-2a/ribavirin for chronic hepatitis C not responsive to IFN/ribavirin: an adjuvant role? J Viral Hepat 2012;19(Suppl 1):52–59. DOI 10.1111/j.1365-2893.2011.01524.x. PMID 22233415. — n = 552, and the source of both the 12.7% and the 41% figures.
5. Yang YF, Zhao W, Zhong YD, et al. Comparison of the efficacy of thymosin alpha-1 and interferon alpha in the treatment of chronic hepatitis B: a meta-analysis. Antiviral Res 2008;77(2):136–141. DOI 10.1016/j.antiviral.2007.10.014. PMID 18078676. — read via the CRD quality-assessed record, which carries the critique.
6. Shehadeh F, Benitez G, Mylona EK, et al. A pilot trial of thymalfasin to treat hospitalized patients with hypoxemia and lymphocytopenia due to COVID-19. J Infect Dis 2023;227(2):226–235. DOI 10.1093/infdis/jiac362. PMID 36056913. NCT04487444. — the only prospective randomised COVID trial. It missed.
7. Huang C, Fei L, Xu W, et al. Efficacy evaluation of thymosin alpha 1 in non-severe patients with COVID-19: a retrospective cohort study based on propensity score matching. Front Med 2021;8:664776. DOI 10.3389/fmed.2021.664776. PMID 33968969. — 1,388 patients, no difference.
8. Wang Z, Chen J, Zhu C, et al. Thymosin alpha-1 has no beneficial effect on restoring CD4+ and CD8+ T lymphocyte counts in COVID-19 patients. Front Immunol 2021;12:568789. DOI 10.3389/fimmu.2021.568789. PMID 34149679. — the title is the finding.
9. Liu Y, Pang Y, Hu Z, et al. Thymosin alpha 1 reduces the mortality of severe COVID-19. Clin Infect Dis 2020;ciaa630. DOI 10.1093/cid/ciaa630. PMID 32442287. — 76 patients, retrospective, unadjusted. The origin of the COVID claim.
10. Maio M, Mackiewicz A, Testori A, et al. Large randomized study of thymosin alpha 1, interferon alfa, or both in combination with dacarbazine in patients with metastatic melanoma. J Clin Oncol 2010;28(10):1780–1787. DOI 10.1200/JCO.2009.25.5208. PMID 20194853. — a dose-finding phase II, never confirmed in phase III.
11. US Food and Drug Administration. Thymosin alpha-1 related bulk drug substances — evaluation for the 503A Bulks List. Pharmacy Compounding Advisory Committee, 4 December 2024. fda.gov — the source of the solubility, characterisation and immunogenicity findings. Read in full.
12. US Food and Drug Administration. Certain bulk drug substances for use in compounding that may present significant safety risks. Category 2; thymosin-alpha 1 added 29 September 2023. fda.gov
13. Health Canada. Unauthorized injectable peptide drugs seized from Optimum Wellness Centre, Calgary, Alberta. Advisory RA-77218, 1 April 2025. recalls-rappels.canada.ca — names thymosin alpha-1 twice, with and without mannitol.
14. European Medicines Agency. EU/3/02/110 — orphan designation, thymalfasin, 30 July 2002. ema.europa.eu — the source of the Italian authorisation statement.
15. Elizondo-Riojas MA, Chamoun-Emanuelli AM, Gaubin M, et al. Phosphatidylserine binding and NMR structure of thymosin alpha-1. Biochem Biophys Res Commun 2011;416:356–361. PDB 2L9I. — solved in 40% trifluoroethanol, which is the point.

Listed rather than hidden, what we could not check. PubMed and its programmatic interface were unreachable throughout this work, so several identifiers come from institutional repositories, publisher pages and quality-assessed review records rather than from PubMed itself. We could not read the TESTS full text — its figures here come from the publishing institution's record of the paper plus the published correction — so per-arm adverse event counts and the funding source are unverified by us. The melanoma paper is behind a publisher block; we have its response rates from the corresponding patent filings but no survival figures, and we have therefore printed none. We could not open the two most-cited individual hepatitis B trials and have not repeated their numbers. The 2002 Cochrane review would not load, so we cannot say whether it is current. And on handling: there is no measured in-solution stability study, no solubility figure for bacteriostatic water, no compound-specific hygroscopicity or freeze-thaw data, and no defined acetate stoichiometry — those absences are stated in the text rather than filled with estimates.