Thymosin alpha 1 is a 28-amino-acid peptide first isolated from calf thymus tissue. It carries a far longer clinical record than most peptides sold for recovery or immune support.
That record is the useful part, because it is large enough to answer the question properly. Trials have run for three decades, in hepatitis, sepsis and cancer, and several of them are big.
What they show is a peptide that is well tolerated and whose benefit stays uncertain when the trials are pooled honestly. The same pattern shows up in TB-500, the other thymosin peptide sold for recovery, which has far less human data behind it.
What it is, and where it is a drug
The peptide corresponds to the first 28 residues of prothymosin alpha, a protein made in the thymus. It is given by subcutaneous injection in every trial below.
It is marketed as a prescription drug in a number of countries under the name Zadaxin. The United States is not one of them: a Drugs@FDA search on its ingredient name, thymalfasin, and on that brand returns no approved product [6].
Chronic hepatitis B: the longest record
A Cochrane review published in September 2026 gathered every randomized trial of thymosin alpha 1 in chronic hepatitis B [1]. It found 10 trials with 1,349 randomized participants, published between 1991 and 2018, in six countries.
The pooled results all favored the peptide. All-cause mortality was lower: RR 0.53, 95% CI 0.29 to 0.96, across 3 studies and 907 participants. Serious adverse events were lower too, at RR 0.72, 95% CI 0.53 to 0.99, across 5 studies and 1,056 participants.
The review then rated how much those numbers can be trusted. It graded the certainty of the evidence as very low for every outcome except serious adverse events, which it graded low.
Its reasons were risk of bias in the trials, imprecision with wide intervals, and inconsistency between studies. The authors' own conclusion is that they are not sure whether the peptide reduces mortality at all.
Two details matter for reading that pooled figure. Only two of the 10 trials used a placebo, and four were funded by industry.
Sepsis: a positive small trial, then a large null one
The sepsis literature is the clearest illustration of why trial size matters here.
| Trial | Participants | Design | 28-day mortality |
|---|---|---|---|
| ETASS, 2013 | 361 adults in six Chinese hospitals | Single-blind, control group | 26.0% vs 35.0% (RR 0.74, 95% CI 0.54 to 1.02) |
| TESTS, 2025 | 1,106 adults at 22 Chinese centers | Double-blind, placebo-controlled, phase 3 | 23.4% vs 24.1% (HR 0.99, 95% CI 0.77 to 1.27) |
ETASS randomized 361 patients with severe sepsis and reported 28-day mortality of 26.0% against 35.0% in the control group [3]. The confidence interval crossed one, and the trial was single-blind with no placebo.
TESTS was built to settle it. It gave 1,106 adults with sepsis a subcutaneous injection every 12 hours for seven days, or a matching placebo, at 22 centers [2].
Twenty-eight-day mortality was 23.4% on the peptide and 24.1% on placebo, a hazard ratio of 0.99 (95% CI 0.77 to 1.27, P = 0.93). No secondary or safety outcome differed either.
The trial reported two subgroup signals, in people under 60 and in people with diabetes. Subgroup results from a null trial generate hypotheses; they do not establish an effect.
A 2025 meta-analysis of 11 sepsis trials found a pooled 28-day mortality benefit (OR 0.73, 95% CI 0.59 to 0.90) [4]. That benefit disappeared in the high-quality subgroup, at OR 0.82, 95% CI 0.65 to 1.03. It disappeared again in the multicenter subgroup, at OR 0.86, 95% CI 0.68 to 1.08.
Their own trial sequential analysis concluded the accumulated sample is still not large enough to settle the question.
What the safety record looks like
Tolerability is the most consistent finding across all of this work. ETASS recorded no serious drug-related adverse event [3], and TESTS found no safety outcome that differed from placebo [2].
A 2024 review counted more than 11,000 people across more than 30 trials and described the peptide as well tolerated [5]. That review also argues that the FDA should permit 503A pharmacies to compound it, so its conclusions about efficacy read as advocacy rather than a neutral synthesis.
Good tolerability is not the same as benefit. The Cochrane authors found the evidence on serious adverse events low-certainty, which is the highest grade they gave anything [1].
What the trials gave
No human dose of this peptide is established for any use in the United States, because no US product exists to carry a label.
The trials themselves used subcutaneous injection. TESTS gave its dose every 12 hours for seven days, in an intensive care setting [2]. The hepatitis trials ran for months, with follow-up from six months to five years [1].
Those schedules belong to the conditions they were tested in. They are not a regimen for a healthy adult, and no trial has studied one.
Putting it together
Thymosin alpha 1 is unusual among peptides sold for immune support in having real trial data, including one properly powered phase 3 trial. That trial was negative.
The hepatitis evidence looks more favorable but is rated very low certainty by the people who pooled it. Both readings come from the same place: the trials are mostly small, mostly unblinded, and mostly from one region.
For how this compares with peptides that have almost no human evidence, see BPC-157 human evidence. For the broader safety question across the class, see are peptides safe, and for the rest of the research on this goal, the recovery hub.