Thymosin alpha-1
Thymalfasin, Tα1, TA1, Thymosin α1, Thymosin alfa 1
Thymosin alpha-1 is a 28-amino-acid peptide originally purified from calf thymus that acts as an immune modulator, mainly by engaging Toll-like receptors on dendritic cells. It is a licensed medicine for chronic hepatitis B in roughly 30 countries but has never been approved in the US, UK or EU. It has more human trial data than any other peptide in this class, and the pattern is small positive studies followed by large negative ones.
Mechanism
Thymosin alpha-1 is the N-terminally acetylated 28-residue fragment corresponding to residues 2-29 of prothymosin alpha, first isolated from the calf thymus preparation known as thymosin fraction 5. It is not a classical hormone acting on a dedicated high-affinity receptor. Its best-characterised activity is as a ligand for Toll-like receptors (principally TLR9, and to a lesser extent TLR2) on plasmacytoid and myeloid dendritic cells. Engagement drives MyD88-dependent signalling through IRF7 and NF-κB, producing type I interferon, IL-12 and dendritic cell maturation. Downstream, naive CD4+ T cells are pushed towards a Th1 phenotype, and CD8+ cytotoxic and natural killer cell activity increases.
The same pathway simultaneously induces indoleamine 2,3-dioxygenase (IDO1) in dendritic cells, a tolerogenic enzyme that restrains excessive T-cell responses. This is why thymosin alpha-1 is described as a bidirectional immune modulator rather than a straightforward stimulant: in a lymphopenic or immunoparalysed host it tends to restore effector function, while in an inflamed host the IDO arm damps it. It also influences thymocyte maturation markers and helps restore circulating lymphocyte counts in lymphopenic states. Two honest caveats belong here. First, almost all of this receptor-level mapping comes from mouse models and human cell culture, not from human pharmacodynamic studies. Second, the concentrations at which TLR engagement is demonstrated in vitro have never been clearly reconciled with the plasma exposures actually achieved by the subcutaneous doses used in clinical trials, a gap that matters when interpreting why the large trials failed.
What the research shows
In chronic hepatitis B, a phase 3 multicentre double-blind placebo-controlled trial (Mutchnick, n=97, 1999) enrolled HBeAg-positive, HBV DNA-positive patients and found complete response in 14% on thymosin alpha-1 versus 4% on placebo, a difference that was not statistically significant. Sustained loss of HBV DNA overall was 25% versus 13%. The authors stated plainly that the findings did not confirm the treatment efficacy reported in other clinical studies. A 2008 meta-analysis of four small trials (199 patients total) comparing thymosin alpha-1 with interferon alfa reported odds ratios at six months after treatment of 3.71 for virological response, 3.12 for biochemical response and 2.69 for complete response, with the pattern that its advantage accumulated after therapy rather than during it; the authors framed the conclusion around HBeAg-negative disease. Those trials are small and old. This body of work is the basis for its approvals in China, Italy and elsewhere; it would not clear a modern FDA or EMA review.
In sepsis the trajectory is instructive. The ETASS trial (n=361, six Chinese centres, 2013) reported 28-day all-cause mortality of 26.0% versus 35.0% for control, relative risk 0.74 (95% CI 0.54-1.02), with borderline statistics (p=0.062 unstratified, log-rank p=0.049). That promising signal prompted a definitive test. TESTS (BMJ 2025) randomised 1,106 adults at 22 Chinese centres to subcutaneous thymosin alpha-1 (n=552) or placebo (n=554) every 12 hours for seven days, double-blinded. It was flatly null: 28-day mortality 23.4% versus 24.1%, hazard ratio 0.99 (95% CI 0.77-1.27, p=0.93), with no secondary or safety outcome differing significantly. A 2025 systematic review of 11 RCTs (1,927 patients) found a pooled 28-day mortality benefit (OR 0.73, 95% CI 0.59-0.90), but that benefit vanished when analysis was restricted to higher-quality trials (OR 0.82, 95% CI 0.65-1.03) and to multicentre trials (OR 0.86, 95% CI 0.68-1.08), and trial sequential analysis indicated the evidence base remains underpowered for a definitive conclusion. Benefit that evaporates as trial quality rises is the classic signature of small-study bias.
COVID-19 evidence is weaker still. A 2023 systematic review and meta-analysis of eight studies reported lower mortality (RR 0.59, 95% CI 0.37-0.93) but found no difference in need for mechanical ventilation or hospital length of stay, and the pool was dominated by retrospective observational studies rather than randomised trials, a design highly vulnerable to confounding by indication, since sicker or less-sick patients are selected for treatment in ways that no statistical adjustment fully removes. The authors themselves called for further randomised trials.
Evidence assessment
Mixed evidence
Multiple randomised human trials exist and it holds marketing authorisations for chronic hepatitis B in around 30 countries, but the two largest and most rigorous placebo-controlled trials (the 1,106-patient TESTS sepsis trial and the 97-patient phase 3 hepatitis B trial) were both null, and meta-analytic benefit disappears when analysis is restricted to high-quality multicentre studies.
Tiers are applied consistently across the library and re-checked when new trials read out. Read the grading method.
Key studies
The efficacy and safety of thymosin α1 for sepsis (TESTS): multicentre, double blinded, randomised, placebo controlled, phase 3 trial High-quality evidence
No difference in 28-day all-cause mortality: 23.4% with thymosin α1 versus 24.1% with placebo (hazard ratio 0.99, 95% CI 0.77-1.27, p=0.93 by log-rank test). No secondary or safety outcome differed significantly.
The efficacy of thymosin alpha 1 for severe sepsis (ETASS): a multicenter, single-blind, randomized and controlled trial Mixed evidence
28-day all-cause mortality 26.0% versus 35.0% for control (relative risk 0.74, 95% CI 0.54-1.02); statistically borderline (p=0.062 unstratified, log-rank p=0.049), an absolute reduction of 9.0 percentage points.
Efficacy of thymosin α1 for sepsis: a systematic review and meta-analysis of randomized controlled trials Mixed evidence
Pooled 28-day mortality reduction (OR 0.73, 95% CI 0.59-0.90, p=0.003), but no significant benefit in the higher-quality-trial subgroup (OR 0.82, 95% CI 0.65-1.03) or the multicentre subgroup (OR 0.86, 95% CI 0.68-1.08); trial sequential analysis found the accumulated sample size inadequate for a definitive conclusion.
Thymosin alpha1 treatment of chronic hepatitis B: results of a phase III multicentre, randomized, double-blind and placebo-controlled study High-quality evidence
Complete response in 14% versus 4% on placebo, not statistically significant; sustained loss of HBV DNA 25% versus 13%. The authors concluded the findings did not confirm treatment efficacy reported in other clinical studies.
The efficacy of thymosin alpha-1 therapy in moderate to critical COVID-19 patients: a systematic review, meta-analysis, and meta-regression Limited evidence
Lower pooled mortality with thymosin α1 (RR 0.59, 95% CI 0.37-0.93, p=0.02); no significant difference in need for mechanical ventilation or hospital length of stay. The authors called for further randomised trials to confirm the result.
Comparison of the efficacy of thymosin alpha-1 and interferon alpha in the treatment of chronic hepatitis B: a meta-analysis Limited evidence
At six months after the end of treatment, odds ratios favouring thymosin α1 over interferon alfa were 3.71 (95% CI 2.05-6.71) for virological response, 3.12 (95% CI 1.74-5.62) for biochemical response and 2.69 (95% CI 1.47-4.91) for complete response; the advantage was absent at end of therapy and accumulated during follow-up.
Safety
The tolerability record across trials is genuinely good, and this is worth stating clearly because it is one of the few peptides in this class with real safety data. The most consistent finding is local injection-site reaction: erythema, discomfort, occasional transient rash. In TESTS, with over 500 patients receiving twice-daily subcutaneous injections for a week in critical illness, no safety outcome differed significantly from placebo.
The limits of that record matter. Essentially all controlled safety data come from short courses (days to a few months) in acutely unwell patients or people with chronic viral hepatitis. There are no controlled long-term safety data for the repeated-course, healthy-adult use pattern promoted in the wellness market. Because the mechanism involves TLR-driven immune activation alongside IDO induction, concerns about use in active autoimmune disease, in transplant recipients, or alongside immunosuppressive therapy are theoretically reasonable and have never been formally tested. It is also worth noting that TESTS found a prespecified subgroup signal suggesting a possible differential effect by age, with the point estimate in participants under 60 lying on the side of harm, a subgroup result that is hypothesis-generating in both directions, but not reassuring. Separately, material sold online as 'research grade' sits entirely outside pharmaceutical quality control: identity, purity, sterility, endotoxin content and actual peptide mass are unverified, and these are the failure modes that cause harm with injected products regardless of what the molecule itself does.
Regulatory status
| Jurisdiction | Status |
|---|---|
| United Kingdom | No MHRA marketing authorisation. Not a licensed medicine in the UK; any clinical use would be as an unlicensed import at prescriber liability. |
| United States | Not approved by the FDA for any indication. US phase 3 hepatitis B trials in the 1990s did not lead to approval. It is not a licensed medicine and is sold online only under the 'research use only' framing, which is a legal device rather than a regulatory category. |
| WADA (sport) | Not named on the WADA Prohibited List. This is a point of frequent confusion: thymosin beta-4 and its derivatives (including TB-500) are prohibited at all times under the growth factors section (S2), but thymosin alpha-1 is a chemically and functionally unrelated molecule and is not listed. Athletes should nonetheless verify against the current year's List, as status can change. |
Questions
No. Thymosin alpha-1 has never been approved by the FDA for any indication, and there is no MHRA marketing authorisation in the UK. It is a licensed medicine for chronic hepatitis B in roughly 30 other countries, including China and Italy. In the US and UK it is sold online only under 'research use only' labelling, which is a legal shield rather than a regulatory approval.
The best available evidence says no. The definitive test was the TESTS trial published in the BMJ in 2025: 1,106 patients, 22 centres, double-blind and placebo-controlled. 28-day mortality was 23.4% with thymosin alpha-1 and 24.1% with placebo, a hazard ratio of 0.99. Earlier smaller trials such as ETASS suggested a benefit, and a 2025 meta-analysis found one overall, but that benefit disappeared when the analysis was limited to high-quality trials and to multicentre trials.
They are unrelated molecules despite the shared 'thymosin' name, which is a historical artefact of both being isolated from the same thymus preparation. Thymosin alpha-1 is a 28-residue immune modulator acting on Toll-like receptors. Thymosin beta-4 is a 43-residue actin-sequestering protein studied for tissue repair. This matters practically: thymosin beta-4 and its derivatives including TB-500 are banned at all times by WADA, while thymosin alpha-1 is not named on the Prohibited List.
In clinical trials it has a good tolerability record, with injection-site reactions being the most consistent adverse effect and no safety outcome differing from placebo in over 500 critically ill patients in the TESTS trial. But that record covers short courses of days to months in unwell patients under medical supervision. There are no controlled long-term safety data for repeated courses in healthy adults, no data on use alongside immunosuppression or in active autoimmune disease, and material bought online has no verified identity, purity, sterility or endotoxin testing.