Thymosin Alpha-1

What it is. Thymosin alpha-1 is a 28-amino-acid peptide first isolated from thymosin fraction 5, a bovine thymus extract, and now made synthetically. It is licensed in around thirty countries - not the United States, European Union or United Kingdom - principally for chronic hepatitis B and as an adjunct in sepsis and some cancers [5].

Sepsis, and how the story changed. ETASS, published in 2013, was a multicentre randomised single-blind trial in 361 patients with severe sepsis across six tertiary ICUs in China. 28-day all-cause mortality was 26.0% on thymosin alpha-1 against 35.0% on control - a hazard ratio of 0.74 whose confidence interval (0.54-1.02) touched 1. The trial also found greater improvement in monocyte HLA-DR expression at days 3 and 7, which is a genuine immunological readout rather than a symptom score [1].

That result, plus several trials of thymosin alpha-1 combined with ulinastatin [6][7][8], generated a decade of favourable meta-analyses [2].

Then TESTS. A multicentre, double-blinded, randomised, placebo-controlled phase 3 trial enrolled 1106 adults with sepsis, of whom 1089 were analysed. 28-day all-cause mortality was 127 of 542 (23.4%) on thymosin alpha-1 and 132 of 547 (24.1%) on placebo, hazard ratio 0.99. The authors concluded there was no clear evidence that it decreases 28-day mortality in sepsis [3]. A 2025 meta-analysis followed [9].

This is the same pattern as aviptadil in COVID-19 and exenatide in Parkinson's disease: an encouraging trial with a confidence interval brushing significance, followed by a larger, better-blinded trial that finds nothing. It is the most common way a promising drug fails, and it is why the size and blinding of a trial matter more than the direction of its point estimate.

Hepatitis B. Here the picture is longer-standing and still unresolved. Meta-analyses have compared it with interferon alpha [10][11] and trials have tested combinations with interferon and lamivudine [12][13][14]. The 2026 Cochrane review is the most rigorous assessment: it found thymosin alpha-1 may reduce all-cause mortality (RR 0.53, 95% CI 0.29-0.96), serious adverse events (RR 0.72, 0.53-0.99) and non-serious adverse events, with little or no effect on quality of life or histological improvement - and graded the certainty of the evidence as very low for every outcome, concluding that the effects are very uncertain [4].

Thymosin alpha-1 acts on the innate and adaptive interface rather than on a single receptor pathway. It signals through Toll-like receptors on dendritic cells and monocytes, promoting dendritic cell maturation, T cell differentiation and natural killer cell activity, and modulating cytokine production [5].

The sepsis rationale is specific and biologically sound. Sepsis involves an initial inflammatory surge followed by immunoparalysis, in which monocytes downregulate HLA-DR and lose antigen-presenting capacity - a state associated with secondary infection and death. Thymosin alpha-1 is intended to reverse that, and ETASS showed it measurably does: mHLA-DR improved more on treatment at days 3 and 7 [1].

The instructive part is that this did not translate. TESTS confirmed the drug can be given safely to septic patients at scale and found no mortality difference [3]. Correcting a biomarker of immunoparalysis is not the same as correcting the outcome it marks - which is a general caution worth carrying to every other compound on this site whose case rests on moving a marker.

A synthetic 28-residue acetylated peptide, supplied as lyophilised powder for subcutaneous injection and marketed as Zadaxin where licensed [5]. It is a defined synthetic molecule, which distinguishes it from Thymalin - a thymic extract of undefined composition used in the Russian peptide literature.

Consistently well tolerated, which is notable given the populations studied. In TESTS, a placebo-controlled trial in 1106 septic patients, no safety concern emerged that changed the conclusion [3]. In chronic hepatitis B, the Cochrane review found thymosin alpha-1 associated with fewer serious adverse events than control (RR 0.72, 95% CI 0.53-0.99) and fewer non-serious adverse events - though at low and very low certainty respectively [4].

That tolerability is the strongest and most consistent finding across its literature, and it is presumably why the drug has remained licensed in many countries despite equivocal efficacy.

Two caveats. Most of the trial evidence comes from China and other Asian countries, and the Cochrane review notes that among trials reporting age, none included certain groups - so generalisability is limited. And an immunomodulator's long-term consequences - autoimmunity, altered infection or malignancy risk over years - are not something any of these trials were designed or sized to detect.

The combinations that have been formally studied are therapeutic rather than problematic. With interferon alpha in chronic hepatitis B [10][12], with lamivudine [13], and with ulinastatin in sepsis, where the combination has its own meta-analyses [6][7][8].

No adverse drug interactions of note have been reported.