BRAK LABS Peptide research, plainly written

compound

Thymosin alpha-1

The deepest clinical record of any unapproved compound on this site, read in the order the trials happened: a promising finding across 361 patients, then the same question asked again in 1,106 with tighter masking, and a very different answer.

Last Reviewed Editorial policy Methodology

Thymosin alpha-1 is unusual on this site in almost every way. It is a real medicine somewhere: approved and marketed outside the United States, holding an international nonproprietary name, and the subject of a pharmacopoeial reference standard. It has 62 registered trials, nine of them at phase 3. It has a measured half-life in people. Most compounds covered here have none of that.

So the question this page answers is not whether anyone has studied it. It is what the studies found, and the answer turns on one trial.

In January 2025 the BMJ published a phase 3 trial of this peptide in 1,106 adults with sepsis: 22 centres, randomised, placebo-controlled, and masked from participants, carers, investigators and outcome assessors alike. Twelve years earlier the same lead institution had run a smaller trial in the same condition and reported a nine percentage point mortality difference with a borderline p-value. The large trial reported a hazard ratio of 0.99.

That sequence is the most instructive thing in this corpus, and it is why this page is written in chronological order rather than by indication.

At a glance Last Reviewed
Summary properties of this compound, each with its source
PropertyValueSource
Category A 28 amino acid peptide originally isolated from a bovine thymus extract, now made synthetically, used as an immune modulator PMID 15482167
Also known as Thymalfasin, which is the international nonproprietary name rather than a brand, and Zadaxin, which is a trade name PMID 39814420
Registered human trials 62 whose intervention names the molecule, of which 9 are phase 3, 32 phase 2 and 7 phase 4. Only 2 of the 62 have posted results ClinicalTrials.gov, swept 23 August 2026
The largest completed trial 1,106 adults with sepsis, quadruple-masked against placebo. 28-day mortality 23.4 per cent against 24.1 per cent, hazard ratio 0.99, P equal to 0.93 PMID 39814420, BMJ 2025
US approval status No approved application. An openFDA query on 23 August 2026 for thymalfasin, thymosin and the trade name returned no record for any of the three openFDA drugsfda, 23 August 2026
FDA compounding position On the category 2 list of bulk substances that may present significant safety risks, for immunogenicity by certain routes and complexities of peptide-related impurities and characterisation FDA, read 23 August 2026

Not medical advice

This page reports what has been documented about a compound. It is not medical advice, not a diagnosis, not a dosing protocol, and not a recommendation to obtain or use anything described here. Talk to a licensed clinician about anything concerning your health. Read the full disclaimer.

Reviewed against editorial standards Updated

Key takeaways

  • The largest completed trial randomised 1,106 adults with sepsis across 22 centres, masked from participants, carers, investigators and outcome assessors, against placebo. Death from any cause at 28 days occurred in 23.4 per cent of the treated group and 24.1 per cent of placebo: hazard ratio 0.99, 95 per cent confidence interval 0.77 to 1.27, P equal to 0.93. The paper reports that no secondary or safety outcome differed statistically significantly. [Human RCT] [1] [17]
  • Reporting from the same lead institution went the other way in 2013, across 361 patients: 28-day mortality of 26.0 per cent against 35.0 per cent, which the paper itself described as carrying a marginal p-value. The compound did not change between the two trials. The size and the masking did. [Human RCT] [2] [18]
  • 62 registered trials name this molecule in their interventions, 9 of them at phase 3, and only 2 of the 62 have posted results to the registry. Two 500-participant phase 3 trials in hepatitis C started in 2002, completed, and no results or publication were located for either.
  • The published hepatitis C phase 3, in 552 non-responders, reported a sustained virological response of 12.7 per cent against 10.5 per cent on placebo in the intention-to-treat population, P equal to 0.407. The authors' own conclusion is that the peptide seems to play no role in the primary therapy of that disease. [Human RCT] [3] [19]
  • FDA lists it among bulk substances that may present significant safety risks for compounding, citing immunogenicity by certain routes and complexities of peptide-related impurities and characterisation. A separate analytical study found 23 identified related impurities above 0.1 mg per gram in a pharmacopoeial reference standard for the molecule.
  • The positive literature is real and it is mostly immune-marker endpoints in many small trials. Meta-analyses in acute exacerbations of chronic obstructive pulmonary disease (39 trials, 3,329 patients) and in severe acute pancreatitis (5 trials, 706 patients) both report higher CD4+ counts and a higher CD4 to CD8 ratio with treatment. [Human RCT] [9] [10]
REGISTERED HUMAN TRIALS
62
HAZARD RATIO IN THE 1,106 PATIENT TRIAL
0.99
NUMBERED SOURCES
26
EVIDENCE GRADE
B

Who researches Thymosin alpha-1?

Three readers arrive here and the record serves them differently.

Someone comparing this against the other thymus peptides wants to know that thymosin alpha-1 and thymosin beta-4 are different molecules that came out of the same bovine thymus extract in the 1970s, which is the entire reason the names get confused. Each has its own page.

Someone who has read that this is an approved medicine wants the qualifier: approved elsewhere, not in the United States, and FDA has separately named it as a compounding substance that may present significant safety risks.

Someone who has read a summary of its clinical record wants to know when that summary was written. A great deal of the material in circulation predates January 2025, and the trial published that month is the largest ever run on the molecule.

What is Thymosin alpha-1?

Plain-English version: a 28 amino acid piece of a larger protein the thymus gland makes, given by injection to change how the immune system behaves.

The thymus is a gland behind the breastbone where T cells, one of the immune system's main cell types, are trained before they go to work. It shrinks throughout adult life, and that shrinkage is one of the reasons immune function declines with age.

In the 1970s a bovine thymus extract called thymosin fraction 5 was found to restore immune function in animals whose thymus had been removed. That extract was a mixture of many peptides, and researchers separated out the ones that seemed to be doing the work. Thymosin alpha-1 is one of them. Thymosin beta-4 is another, which is why the two names sit next to each other everywhere and describe unrelated molecules with unrelated jobs [5] [14].

What is sold and studied today is made synthetically rather than extracted, and it is 28 amino acids long [5]. It has an international nonproprietary name, thymalfasin, which is the generic name a regulator uses rather than a company's trade name.

It is given by injection under the skin. Unlike almost everything else on this site, its behaviour in a person has been measured: absorption is rapid, peak concentration in blood arrives within about two hours, levels return to baseline within 24 hours, and the serum half-life is roughly two hours [13].

28 amino acids. This site did not verify the residue sequence against a primary chemical reference and does not print one · Not stated here. No primary chemical reference was read for this page

How strong is the evidence, claim by claim?

Every claim area this compound is discussed for gets a row, including the ones with the weakest support. The tier records what the row rests on, and the last column states what that work can and cannot show. Each tier links to its definition on the methodology page.

Claim Strength Matrix Last Reviewed
Evidence supporting each claim area, with the tier of the underlying studies
Evidence Area What Has Been Studied Evidence Level What It Can and Cannot Show
T cell counts and the ratio of CD4 to CD8 cells, which is what the compound is sold to change The claim the compound is sold for, and the one endpoint its positive literature agrees on. A 2024 meta-analysis of 39 randomised trials in 3,329 patients with acute exacerbations of chronic obstructive pulmonary disease reports higher CD4+ counts, a higher CD4 to CD8 ratio and lower CD8+ counts with treatment, all at p below 0.001. A 2025 meta-analysis of 5 randomised trials in 706 patients with severe acute pancreatitis reports the same direction on CD4+ percentage and the CD4 to CD8 ratio, and no significant change in CD8+ cells [9] [10] [Human RCT] Both pool many small trials drawn substantially from Chinese-language databases, and neither was examined trial by trial for overlap or risk of bias for this page. A change in a cell count is a laboratory measurement rather than a clinical outcome, and the pancreatitis analysis is candid about where its clinical outcomes did not move: hospital stay was not shortened, lung infections were not reduced, and C-reactive protein fell at the lower amount but not the higher one
Death from any cause in sepsis The most rigorously tested claim about this molecule and the only one asked twice at scale. A 2013 trial in 361 patients with severe sepsis reported 28-day mortality of 26.0 per cent against 35.0 per cent. A 2025 phase 3 trial in 1,106 adults, quadruple-masked against placebo across 22 centres, reported 23.4 per cent against 24.1 per cent, hazard ratio 0.99, 95 per cent confidence interval 0.77 to 1.27, P equal to 0.93 [1] [2] [17] [18] [Human RCT] The two results do not agree, and the 2025 trial is roughly three times the size, placebo-controlled, and masked from participants, carers, investigators and outcome assessors. Its authors state that no secondary or safety outcome differed statistically significantly. The 2013 paper described its own primary result as carrying a marginal p-value, and its publication title says single-blind while the registry record says double. Prespecified subgroups in the larger trial point in opposite directions by age and by diabetes, in a trial whose primary outcome was null
Viral and biochemical response in chronic hepatitis B, the indication it is registered for outside the United States A 2008 meta-analysis of four randomised trials in 199 patients compared the peptide against interferon alpha. At the end of six months of treatment the odds ratios favoured interferon: 0.62 for virological response, 0.60 for biochemical and 0.54 for complete response. Six months after treatment ended the direction reversed, to 3.71, 3.12 and 2.69 in favour of the peptide [4] [5] [Human RCT] 199 patients across four trials is a small base for a conclusion this widely repeated, and the comparison is against interferon rather than against placebo, so neither arm establishes an effect against no treatment. The reversal between timepoints is the actual finding and is why the compound is described as slower-acting rather than stronger. A 2004 review states that seven randomised monotherapy studies reported a significantly higher sustained response; that review was read as an abstract whose sentence is truncated in PubMed, so what it reported beyond that is not stated here
Sustained virological response in chronic hepatitis C A 552 patient phase 3 trial in people who had not responded to standard therapy, randomised to peginterferon and ribavirin plus either the peptide or placebo for 48 weeks. Sustained virological response in the intention-to-treat population was 12.7 per cent against 10.5 per cent, P equal to 0.407 [3] [19] [20] [Human RCT] The authors' own conclusion is that the peptide seems to play no role in the primary therapy of the disease. Among the subset who completed all 48 weeks the response was 41.0 per cent against 26.3 per cent, P equal to 0.048, which the same authors label a hypothesis rather than a finding: 83 and 99 patients of 275 and 277 randomised finished. Two further phase 3 trials of 500 participants each, both started in 2002, are recorded as completed with no results posted and no publication located
Clinical outcomes in COVID-19 Two small randomised trials and one pooled analysis. An open-label trial in 49 hospitalised patients with low oxygen and low lymphocyte counts reported a faster rise in CD4+ T cells on treatment, 3.84 times more on day 5 than day 1 relative to controls at P equal to .01. An open-label prevention trial in 189 dialysis patients posted 5 infections against 7, 27 hospitalisations against 26, and 3 deaths against 7. A 2023 pooled analysis of eight studies reported lower mortality, relative risk 0.59 [6] [7] [8] [22] [Human RCT] Neither trial was masked. The treatment trial's own recovery outcomes did not reach significance in either oxygen group, and the registry records it as terminated for slow enrolment and the departure of its investigator. The prevention trial posted no analyses and its record states that vaccine rollout made infection detection unreliable. The pooled analysis reports 84 per cent heterogeneity, found the mortality effect varied significantly with study size, and says randomised trials are still required to confirm it
Tumour response and survival when added to chemotherapy A 488 participant randomised trial in melanoma, five arms of 97 to 99, testing dacarbazine and interferon alpha with the peptide at three different amounts, plus one arm without interferon, against dacarbazine and interferon alone. Posted results record tumour responses of 7, 10, 6 and 12 in the four peptide arms against 4 in the control arm, and median overall survival of 9.3, 8.6, 10.3 and 9.3 months against 6.6 months [21] [Human RCT] The trial was open-label and no statistical analyses are posted for any outcome, so nothing about significance can be stated in either direction. Median progression-free survival is 1.8 to 2.0 months in every arm including the control, which is the same within the width of the reporting. There is no relationship between amount and response: the highest amount produced the fewest responses. A survival difference with no progression difference, no dose relationship and no posted statistics is a result to record rather than to interpret
The reasoning behind the immune-support claim The thymus shrinks with age, T cell production falls, and this peptide is one of the molecules that thymus makes. A 2025 review of ageing and this peptide sets out the mechanism case: stimulating T cell differentiation, increasing output from the thymus, and modulating dendritic cell and macrophage activity [11] [14] [In-vitro] The mechanism case is a description of what a molecule does in laboratory systems, and a review is not a study. No trial found for this page tested immune support in healthy adults, which is what the compound is sold for. The 2025 review's own statement is that further research is needed to validate long-term efficacy and safety, and a 2015 review names dose, schedule, combination treatments and choice of endpoint as questions the field has not properly addressed

Rolled up, that puts Thymosin alpha-1 at evidence grade B Human data, not settled . Human evidence for the marketed use has read out, and it has not settled the question: the record here does not clear the bar grade A sets, which is two phase-3-scale randomized trials of the marketed use, each cited and each stating whether it met its primary endpoint, agreeing with each other. A grade describes the quality of the research, not whether something works and not whether anyone should use it. The derivation is published.

How Thymosin alpha-1 is thought to work

The mechanism here is better characterised than for anything else on this site, and it is worth understanding precisely because it explains why the compound was expected to work in so many different conditions.

1. Where the molecule comes from, the gland that trains immune cells, and the extract it was first pulled out of (established as the origin)

T cells are the immune cells that recognise a specific threat and coordinate a response to it. They are trained in the thymus, a gland behind the breastbone that is largest in childhood and shrinks steadily through adult life.

In the 1970s researchers took a bovine thymus extract, called thymosin fraction 5, and found it restored immune function in animals whose thymus had been removed. Fraction 5 is a mixture, and the work that followed was an attempt to find which of its peptides did what. Thymosin alpha-1 came out of that search. So did thymosin beta-4, which turned out to do something entirely unrelated inside cells, and the shared name is a historical accident rather than a family resemblance.

What is used today is synthetic, 28 amino acids long, and identical in sequence to the natural peptide. [In-vitro] [5] [14] [15]

2. What it is proposed to do, push immature immune cells toward becoming working T cells, and change the signals they send (characterised in laboratory systems)

The proposal is that the peptide acts where the thymus normally acts: pushing immature cells toward becoming mature T cells, and shifting the balance of the signalling chemicals immune cells release toward the pattern used against viruses and tumours. Laboratory work describes it stimulating production of interferon gamma and interleukin-2 and activating natural killer cells [5].

This is why the compound was taken into so many unrelated conditions. Hepatitis, sepsis, cancer and COVID-19 have almost nothing in common except that in each of them the immune response is doing the wrong thing, either too little or too much. A molecule that nudges the immune system toward a particular pattern is a plausible candidate in all of them, and the trials followed that reasoning rather than any specific finding.

It is also why the clinical results are worth reading closely. A mechanism broad enough to justify trials in six unrelated diseases is broad enough to be hard to disprove in any single one. [In-vitro] [5] [11] [14]

3. The gap the trial record keeps landing in, the immune measurements move and the clinical outcomes often do not (consistent across the larger trials)

The same pattern appears in condition after condition. A measurement of immune-cell numbers or activity shifts in the treated group, and the outcome the patient would notice does not follow it.

The 2013 sepsis trial reported a clear improvement in monocyte HLA-DR, a marker of immune competence, at days 3 and 7. The 2025 trial in three times as many patients found mortality unchanged. The COVID treatment pilot reported 3.84 times more CD4+ T cells on day 5 among treated patients on low-flow oxygen, and its clinical recovery outcomes did not reach significance in either oxygen group. The pancreatitis meta-analysis reports CD4+ percentage and the CD4 to CD8 ratio both improved, and no reduction in hospital stay or in lung infections.

That gap is not evidence that the peptide does nothing. It is the reason a page about this compound has to distinguish between the two kinds of result, because a great deal of the summary material in circulation reports the first kind as though it settled the second. [Human RCT] [1] [2] [6] [10]

What we do not know

Why the 2013 and 2025 sepsis trials disagree. The larger trial is the better designed one, and nothing on the record explains the earlier result.

Whether the compound does anything for immune support in a healthy adult, which is the use it is sold for. No trial found for this page tested that.

What the two 500-participant hepatitis C phase 3 trials from 2002 found. Both completed, neither posted results, and no publication was located for either.

Whether the melanoma survival difference is real. The trial was open-label, posted no analyses, and showed no dose relationship and no progression-free survival difference.

How many countries register it as a medicine. The figure commonly given is more than thirty, and this site did not verify it against any regulator's register.

Its position on the WADA Prohibited List. The S0 class turns on a substance having no current approval by any governmental regulatory health authority, and this compound has approvals, so the placement is genuinely unsettled rather than resolved in either direction.

What this evidence can and cannot show

These results come from Human participants for the clinical work, cell and animal systems for the target work. in Randomised trials in sepsis, hepatitis, cancer and COVID-19; T cell and dendritic cell systems for the mechanism., not from people. An animal model is a deliberate simplification: the injury is created on purpose, the animal is young and healthy, the dose is scaled to body weight in a way that does not translate directly, and the outcome is measured at a fixed point rather than lived with. Cell and tissue work is a further step removed, because the concentrations that produce an effect in a dish are often far above anything a body reaches. Mechanism is a reason to run a trial. It is not a substitute for one, and compounds that looked mechanistically convincing have failed in people many times.

Has Thymosin alpha-1 been tested in humans?

62 registered trials name this molecule in their interventions, and 2 of them have posted results. The table below is the completed record that can actually be read, in the order the studies reported.

Where a trial posted results to the registry, the figures come from that posted results section read field by field. Where it was published, they come from the paper.

Human studies naming Thymosin alpha-1, and what each one measured
StudyPeopleWhat was doneResultEvidence level
Sepsis, 2025 (NCT02867267)1,106Phase 3, 22 centres, randomised 1:1, quadruple-masked against placebo, injection every 12 hours for seven days. 28-day all-cause mortality 23.4 per cent against 24.1 per cent, hazard ratio 0.99 (0.77 to 1.27), P equal to 0.93The largest trial of this molecule. No secondary or safety outcome differed statistically significantly. Published in the BMJ[Human RCT]
Severe sepsis, 2013 (NCT00711620)361Randomised, six teaching hospitals. 28-day mortality 26.0 per cent against 35.0 per cent, relative risk 0.74 (0.54 to 1.02). Monocyte HLA-DR improved more on treatment at days 3 and 7The paper describes its own primary result as carrying a marginal p-value. Its title says single-blind and the registry record says double[Human RCT]
Chronic hepatitis C, 2012 (NCT01178996)552Phase 3, double-masked, 48 weeks, added to peginterferon and ribavirin in prior non-responders. Sustained virological response 12.7 per cent against 10.5 per cent, P equal to 0.407The authors conclude the peptide seems to play no role in the primary therapy of the disease. A completer subset showed 41.0 against 26.3 per cent, which they label a hypothesis[Human RCT]
Melanoma, completed 2007 (NCT00911443)488Open-label, randomised, five arms of 97 to 99. Tumour responses 7, 10, 6 and 12 in the peptide arms against 4 in the control. Median overall survival 9.3, 8.6, 10.3 and 9.3 months against 6.6. Median progression-free survival 1.8 to 2.0 months in every armResults posted, no analyses posted, no dose relationship. Nothing about significance can be stated in either direction[Human RCT]
COVID-19 treatment, 2023 (NCT04487444)49Open-label, randomised against standard care in hospitalised patients with low oxygen and low lymphocytes. Clinical recovery hazard ratios 1.48 and 1.28, neither significant. CD4+ T cells rose 3.84 times more by day 5 on low-flow oxygen, P equal to .01Recorded as terminated, for slow enrolment and the departure of the investigator. Nine serious adverse events among treated patients were judged unrelated[Human RCT]
COVID-19 prevention, 2023 (NCT04428008)189Open-label, randomised, dialysis patients. Infections 5 against 7, hospitalisations 27 against 26, non-COVID infections 8 against 6, deaths 3 against 7Results posted with no analyses. The record's own limitations note that vaccine rollout made infection detection unreliable[Human RCT]
Chronic hepatitis B, pooled 2008199 across 4 trialsMeta-analysis against interferon alpha. Odds ratios favoured interferon at the end of six months of treatment and favoured the peptide six months afterwards, on virological, biochemical and complete responseSmall, and a comparison against interferon rather than against placebo. The reversal between timepoints is the finding[Human RCT]
Two phase 3 trials, hepatitis C (NCT00039962, NCT00040027)500 eachBoth randomised, both double-masked, both started 2002, both recorded as completedNo results posted and no publication located for either. A thousand participants of phase 3 evidence the public record does not contain[Anecdote]

Why we are not calling this proof

A meta-analysis of many small trials and one large trial are not the same kind of evidence, and where they disagree the large masked trial is the better answer. That is the position this page takes, and it is the reason the 1,106 patient trial leads every section it appears in.

An immune-marker result is not a clinical outcome. CD4+ counts, the CD4 to CD8 ratio and monocyte HLA-DR are measurements of the immune system, and in the larger trials on this page they moved while mortality, recovery and length of stay did not.

A registered trial that completed and posted nothing has told the record only that it happened. Two 500-participant phase 3 trials are in that position, and neither counts as evidence in either direction.

Trials in sepsis, hepatitis, melanoma and COVID-19 are trials in those conditions. None of them tested the use this compound is sold for, which is immune support in a person who is not acutely ill.

Benefits: what the research shows

The matrix above holds each claim with what sits behind it. This section is the shape of the record as a whole.

The honest bottom line on benefits

There are two literatures here and they point in different directions. One is large in trial count and small in individual trial size, reports immune-marker endpoints, and is drawn substantially from databases outside the English-language literature. The other is a handful of large, well-designed, internationally published trials, and its results are 0.99, 0.407 and a survival curve with no progression difference behind it.

Both are facts about the evidence base and this page reports both. Where they disagree, the position taken here is that a 1,106 patient placebo-controlled trial masked from everyone involved is a better answer than a pooled estimate over 39 small ones, and the reason is not that the small trials are dishonest. It is that mortality in sepsis is the kind of question small unmasked trials get wrong in a predictable direction, and this molecule now has a direct demonstration of it: the same institution, the same condition, three times the patients, and the effect went away.

None of that establishes that the compound does nothing. It has measurable effects on immune-cell counts in people, a real pharmacokinetic profile, a pharmacopoeial standard and registration as a medicine in other countries. What the record does not contain is a trial showing that any of it changes an outcome a person would notice, in the use the compound is sold for.

How long Thymosin alpha-1 stays in the body

This is the only compound on this site sold as a research chemical for which real human pharmacokinetic figures exist. A 2001 review in a pharmacy journal reports rapid absorption after injection, peak serum concentration within about two hours, return to baseline within 24 hours, and a serum half-life of approximately two hours.

Those figures were read from that review's abstract rather than from a primary pharmacokinetic study, and they are reported here as what that review states.

The short half-life is worth holding alongside the trial designs above. The 2025 sepsis trial gave an injection every 12 hours for seven days, and the hepatitis trials gave twice-weekly injections over months. Those are very different exposure patterns for a molecule that clears within a day, and the record does not establish which if either is the right one: a 2015 review names dose and schedule among the questions the field has not properly settled [14].

Rapid absorption, peak within about two hours, serum half-life about two hours, back to baseline within 24 hours. [Human RCT] [13]

What is documented about dosing

Nothing here is a dosing recommendation. Statements about dosing reflect what published studies administered.

Commonly cited protocols (extrapolated, not validated)

Nothing here is a dosing recommendation. Statements about dosing reflect community and research reports only.

What published studies gave, in which species and by which route, and what circulates elsewhere. The source class is named on every row, and none of it is a clinical protocol
StudyWhat was givenFrequencyDurationNotes
Sepsis, 2025 phase 3As stated in the trial reportEvery 12 hours7 daysSubcutaneous injection. The largest trial of this molecule, primary outcome null
Chronic hepatitis C phase 3As stated in the trial reportTwice weekly48 weeksAdded to peginterferon and ribavirin. Intention-to-treat result null
Chronic hepatitis B monotherapyThe amount reported by a 2004 review of that literatureTwice weekly6 monthsThe schedule the older hepatitis B literature used
MelanomaThree different amounts in separate armsAs stated in the registry recordAs stated in the registry recordNo relationship appeared between amount and tumour response
Any use for immune support in a healthy adultNothing documentedNothing documentedNothing documentedNo trial found for this page tested that use

The amounts and schedules across this record differ by more than an order of magnitude in exposure, from an injection every 12 hours for a week to twice weekly for a year. That is not a detail. A 2015 review names dose, schedule, combination treatment and choice of endpoint as questions the field has not properly addressed, and says that the more recent trials used higher amounts than the older ones.

The melanoma trial is the only one on this record that tested more than one amount head to head, and its posted results show no relationship between the amount given and the number of tumour responses. The highest of the three amounts produced the fewest.

Commonly cited protocols (extrapolated, not validated) circulate for this compound. They do not come from a trial in the population buying it, because no such trial was found for this page.

Nothing here is a dosing recommendation. Talk to a licensed clinician about anything concerning your health.

Main routes people compare, and what each source class actually reported, are set out on the documented protocols page.

Reported effects and what has been measured

What the largest trial reported

The 2025 phase 3 trial in 1,106 adults states that no secondary or safety outcome differed statistically significantly between the treated group and placebo.

That is the single most useful safety observation on this page, because it comes from the largest, most tightly masked study of the molecule. It is a statement about what that trial measured over its treatment period in critically ill adults, and it is not a general statement about the compound. [Human RCT] [1]

What FDA said, and why

FDA lists thymosin alpha-1 among bulk drug substances for use in compounding that may present significant safety risks. Its stated reasoning, quoted from that page as read on 23 August 2026, is that compounded drugs containing it may pose significant risk for immunogenicity for certain routes of administration and may have complexities with regard to peptide-related impurities and characterisation of the active ingredient, and that the safety-related information is inadequate for the agency to sufficiently understand the extent of any safety issues raised.

Read that sentence carefully, because it is a statement about missing information rather than a finding of harm. The agency is not reporting that the substance injured anyone.

The impurity half of it has an independent measurement behind it. A 2022 analytical study applied high-resolution mass spectrometry to a pharmacopoeial reference standard for this molecule and to three commercial materials, and identified and quantified 23 related impurities above 0.1 mg per gram in the standard itself. More than half of them arise directly or indirectly from the asparagine residue at one end of the sequence, which is a property of the molecule rather than of any manufacturer. [In-vitro] [25] [16]

What the trial literature reports

The 2013 sepsis trial records no serious drug-related adverse event. The COVID-19 treatment pilot records nine serious adverse events among treated patients, all judged by the investigators not to be related to the drug.

A 2015 review describes the safety profile as excellent and virtually devoid of toxicity. That review is written by authors reviewing a field they work in, and it is cited here as their stated assessment rather than as an independent finding.

This site does not describe any compound as safe. What is on the record is what these studies measured, over the periods they measured it, in the populations they enrolled. [Human RCT] [2] [6] [14]

What has been measured about Thymosin alpha-1, and what has never been studied

Talk to a licensed clinician about anything concerning your health.

Sourcing and quality

What a credible product should show

This molecule has something almost nothing else on this site has: a pharmacopoeial reference standard. That means an identity and purity specification exists against which a batch can be measured, which is the thing a certificate of analysis for most research chemicals has no external anchor for.

It also has an international nonproprietary name, thymalfasin, which regulators and the literature use. A material sold under that name is at least claiming to be a defined substance.

The impurity picture is documented rather than assumed. A 2022 study identified 23 related impurities above 0.1 mg per gram in the pharmacopoeial standard itself, in classes including deamination, amination, succinimide formation, amino acid insertion and deletion, dimers and isomers. A reader looking at a certificate of analysis for this compound now knows what the analysis has to be capable of separating.

Red flags

An evidence summary written before January 2025. The largest trial of this molecule published that month and changed what the record says.

The 2013 sepsis result quoted without the 2025 one. The two are the same question asked twice, and the larger, better-masked answer is the second.

The hepatitis C completer analysis quoted as the trial result. The intention-to-treat result was 12.7 per cent against 10.5 per cent at P equal to 0.407, and the completer figures are labelled a hypothesis by the authors themselves.

A trial count for this compound presented without its posted-results count. 62 registrations, 2 with posted results.

Prespecified subgroups from the 2025 trial presented as findings. Its primary outcome was null and its subgroups point in opposite directions.

Approval abroad presented as US approval. An openFDA query on 23 August 2026 found no application under any of this compound's names, and FDA separately lists it as a compounding substance that may present significant safety risks.

Will it show on a drug test?

The position is genuinely unsettled and this site states it as unsettled rather than resolving it. The WADA S0 class covers substances with no current approval by any governmental regulatory health authority, and this compound holds approvals as a medicine outside the United States, so S0's own definition appears not to reach it [26].

This site did not establish whether any other class on the 2026 list names it or reaches it. An athlete should take that question to the relevant anti-doping authority rather than to a page, and should treat an absence from a list as an open question rather than an answer.

Storage

No storage guidance is given here. The impurity chemistry documented for this molecule is dominated by reactions at one end of the sequence, which makes handling a real question rather than a formality, and it is one a characterised product's manufacturer answers rather than this page.

Regulatory status, as of 23 August 2026

Agency positions on Thymosin alpha-1, with the document each one comes from
BodyPositionDateDocument
FDANo approved application. A query for thymalfasin, thymosin and the trade name across generic and brand name fields returned no record for any of the threeQueried 23 August 2026openFDA drugsfda endpoint
FDAListed among bulk drug substances for use in compounding that may present significant safety risks, citing immunogenicity by certain routes and complexities of peptide-related impurities and characterisationPage content current as of 22 April 2026, read 23 August 2026FDA human drug compounding, category 2 list
World Health OrganizationHolds an international nonproprietary name, thymalfasinChecked 23 August 2026Use of the INN throughout the registry and the literature
Other national regulatorsRegistered as a medicine outside the United States. The figure commonly given is more than thirty countries, and this site has not verified it against any regulator's registerNot verifiedStated here as unverified rather than repeated as fact
ClinicalTrials.gov62 registrations whose interventions name the molecule, 9 at phase 3, and 2 with posted resultsSwept 23 August 2026ClinicalTrials.gov v2 API

The regulatory picture for this compound is the most interesting on the site, because it is genuinely split. It is a medicine somewhere and an unapproved substance in the United States, and both of those are true at once.

The split matters for two things a reader may care about. It is why an anti-doping rule that turns on whether any regulator has approved a substance does not obviously reach it. And it is why the FDA compounding listing is worth reading closely rather than as a verdict: the agency's stated position is that it lacks the information to judge the risks, which is a different statement from a finding that the substance is harmful.

This section is dated and re-checked on review. It records documents, their dates and their contents, not evidence about an effect, and carries no evidence tier for that reason.

Thymosin alpha-1 compared with other immune peptides

How Thymosin alpha-1 compares with the compounds it is most often set against
CompoundEvidence gradeWhat the human record coversRead more
Thymosin beta-4Graded on its own pageThe other peptide separated out of the same bovine thymus extract in the 1970s. A different molecule doing a different job inside cells, with a much thinner human record/peptides/thymosin-beta-4/
TB-500Graded on its own pageA seven residue fragment of thymosin beta-4, and the compound most often confused with everything else in this family/peptides/tb-500/
ThymulinGraded on the immune group pageA nine amino acid thymus hormone that needs zinc to work. One registered study, a zinc supplementation trial/peptides/groups/immune/#thymulin
Interferon alphaNot covered as a compound on this siteThe approved medicine this compound was compared against in the hepatitis B literature, rather than against placeboDescribed in the claims table above

Almost every confusion around this compound is a naming confusion, and it has two layers.

The first is that thymosin alpha-1 and thymosin beta-4 share a first name because they were separated out of the same extract, and share nothing else. The second is that TB-500 is a fragment of the second one, not of this one. Each has its own page and its own grade.

The comparison worth understanding is with interferon alpha, because the hepatitis B literature everyone cites is a comparison against that medicine rather than against no treatment. A result that reverses direction between the end of treatment and six months later says something about relative timing between two active drugs. It does not establish an effect against placebo.

What Thymosin alpha-1 typically costs

See the price comparison.

Frequently asked questions

Is thymosin alpha-1 the same as thymalfasin?

Yes. Thymalfasin is the international nonproprietary name, the generic name regulators and the literature use for the substance. Zadaxin is a trade name for a product containing it.

Is it approved by the FDA?

No. An openFDA query on 23 August 2026 for thymalfasin, thymosin and the trade name across generic and brand name fields returned no approved application for any of the three. It is registered as a medicine in other countries.

What did the largest trial find?

A 2025 phase 3 trial randomised 1,106 adults with sepsis across 22 centres, masked from participants, carers, investigators and outcome assessors, against placebo. 28-day all-cause mortality was 23.4 per cent on treatment and 24.1 per cent on placebo: hazard ratio 0.99, P equal to 0.93. No secondary or safety outcome differed statistically significantly.

Why do older summaries say it works for sepsis?

Because a 2013 trial in 361 patients reported 28-day mortality of 26.0 per cent against 35.0 per cent, which the paper itself described as carrying a marginal p-value. The 2025 trial asked the same question in three times as many patients with tighter masking and found no difference.

Did it work for hepatitis C?

The published phase 3 in 552 non-responders reported a sustained virological response of 12.7 per cent against 10.5 per cent, P equal to 0.407, and its authors concluded that the peptide seems to play no role in the primary therapy of that disease.

What about the hepatitis B evidence?

A 2008 meta-analysis of four randomised trials in 199 patients compared it against interferon alpha rather than placebo. The odds ratios favoured interferon at the end of six months of treatment and favoured the peptide six months after treatment ended. 199 patients is a small base for how widely that result is repeated.

How many trials have been run on it?

62 registrations name it in their interventions, 9 of them at phase 3. Only 2 of the 62 have posted results to the registry, and two 500-participant phase 3 trials in hepatitis C from 2002 completed without posting results or being published as far as this site could find.

What has FDA said about it?

FDA lists it among bulk drug substances for use in compounding that may present significant safety risks, citing immunogenicity risk for certain routes of administration, complexities regarding peptide-related impurities and characterisation, and that the safety-related information available is inadequate for the agency to understand the extent of any safety issues. That is a statement about missing information rather than a finding of harm.

Is it the same as thymosin beta-4 or TB-500?

No. Thymosin alpha-1 and thymosin beta-4 are different molecules separated out of the same bovine thymus extract in the 1970s, and TB-500 is a seven residue fragment of thymosin beta-4. Each has its own page and its own grade on this site.

Is it banned in sport?

The position is genuinely unsettled and this site does not resolve it. The WADA S0 class covers substances with no current approval by any governmental regulatory health authority, and this compound holds approvals outside the United States, so that class's own definition appears not to reach it. Whether any other class does was not established here. Take the question to the relevant anti-doping authority.

How long does it last in the body?

A 2001 review reports rapid absorption after injection, peak serum concentration within about two hours, a serum half-life of roughly two hours, and levels back to baseline within 24 hours.

Does anything support using it for general immune support?

No trial found for this page tested that use. Every trial on this record enrolled people with a specific illness: sepsis, hepatitis, cancer, COVID-19, pancreatitis or COPD. The mechanism case for immune support is set out in reviews rather than tested in a study.

References

  1. Wu J, Pei F, Zhou L, Li W, Sun R, Li Y, Wang Z, He Z, Zhang X, Jin X, Long Y, Cui W, Wang C, Chen E, Zeng J, Yan J, Lin Q, Zhou F, Huang L, Shang Y, Duan M, Zheng W, Zhu D, Kou Q, Zhang S, Liu Y, Yao C, Shang M, Peng S, Zhou Q, Cheng KK, Guan X, TESTS study collaborator group. The efficacy and safety of thymosin α1 for sepsis (TESTS): multicentre, double blinded, randomised, placebo controlled, phase 3 trial. BMJ. 2025. PMID 39814420 DOI 10.1136/bmj-2024-082583
  2. Wu J, Zhou L, Liu J, Ma G, Kou Q, He Z, Chen J, Ou-Yang B, Chen M, Li Y, Wu X, Gu B, Chen L, Zou Z, Qiang X, Chen Y, Lin A, Zhang G, Guan X. The efficacy of thymosin alpha 1 for severe sepsis (ETASS): a multicenter, single-blind, randomized and controlled trial. Crit Care. 2013. PMID 23327199 DOI 10.1186/cc11932
  3. Ciancio A, Andreone P, Kaiser S, Mangia A, Milella M, Solà R, Pol S, Tsianos E, De Rosa A, Camerini R, McBeath R, Rizzetto M. Thymosin alpha-1 with peginterferon alfa-2a/ribavirin for chronic hepatitis C not responsive to IFN/ribavirin: an adjuvant role? J Viral Hepat. 2012. PMID 22233415 DOI 10.1111/j.1365-2893.2011.01524.x
  4. Yang YF, Zhao W, Zhong YD, Yang YJ, Shen L, Zhang N, Huang P. Comparison of the efficacy of thymosin alpha-1 and interferon alpha in the treatment of chronic hepatitis B: a meta-analysis. Antiviral Res. 2008. PMID 18078676 DOI 10.1016/j.antiviral.2007.10.014
  5. Chien RN, Liaw YF. Thymalfasin for the treatment of chronic hepatitis B. Expert Rev Anti Infect Ther. 2004. PMID 15482167 DOI 10.1586/14787210.2.1.9
  6. Shehadeh F, Benitez G, Mylona EK, Tran QL, Tsikala-Vafea M, Atalla E, Kaczynski M, Mylonakis E. A Pilot Trial of Thymalfasin (Thymosin-α-1) to Treat Hospitalized Patients With Hypoxemia and Lymphocytopenia Due to Coronavirus Disease 2019 Infection. J Infect Dis. 2023. PMID 36056913 DOI 10.1093/infdis/jiac362
  7. Tuthill CW, Awad A, Parrigon M, Ershler WB. A pilot trial of Thymalfasin (Ta1) to prevent covid-19 infection and morbidities in renal dialysis patients: Preliminary report. Int Immunopharmacol. 2023. PMID 36881981 DOI 10.1016/j.intimp.2023.109950
  8. Soeroto AY, Suryadinata H, Yanto TA, Hariyanto TI. The efficacy of thymosin alpha-1 therapy in moderate to critical COVID-19 patients: a systematic review, meta-analysis, and meta-regression. Inflammopharmacology. 2023. PMID 37845598 DOI 10.1007/s10787-023-01354-2
  9. Cao A, Feng F, Zhou X. Thymosin Alpha 1 Plus Routine Treatment for the Acute Exacerbation of Chronic Obstructive Pulmonary Disease: A Systematic Review and Meta-Analysis. J Coll Physicians Surg Pak. 2024. PMID 39648386 DOI 10.29271/jcpsp.2024.12.1497
  10. Tian Y, Yao J, Ma Y, Zhang P, Zhou X, Xie W, Tang W. Thymosin alpha 1 alleviates inflammation and prevents infection in patients with severe acute pancreatitis through immune regulation: a systematic review and meta-analysis. Front Immunol. 2025. PMID 40599771 DOI 10.3389/fimmu.2025.1571456
  11. Simonova MA, Ivanov I, Shoshina NS, Komyakova AM, Makarov DA, Baranovskii DS, Klabukov ID, Telepenina KP, Atiakshin DA, Shegay PV, Kaprin AD, Stepanenko VN. Aging and Thymosin Alpha-1. Int J Mol Sci. 2025. PMID 41373628 DOI 10.3390/ijms262311470
  12. Dinetz E, Lee E. Comprehensive Review of the Safety and Efficacy of Thymosin Alpha 1 in Human Clinical Trials. Altern Ther Health Med. 2024. PMID 38308608
  13. Ancell CD, Phipps J, Young L. Thymosin alpha-1. Am J Health Syst Pharm. 2001. PMID 11381492 DOI 10.1093/ajhp/58.10.886
  14. Camerini R, Garaci E. Historical review of thymosin α 1 in infectious diseases. Expert Opin Biol Ther. 2015. PMID 26098768 DOI 10.1517/14712598.2015.1033393
  15. Tuthill C, Rios I, McBeath R. Thymosin alpha 1: past clinical experience and future promise. Ann N Y Acad Sci. 2010. PMID 20536460 DOI 10.1111/j.1749-6632.2010.05482.x
  16. Cheng Y, Wu P, Kan Y, Li M, Li H. Identification and determination of structurally related peptide impurities in thymalfasin by liquid chromatography-high-resolution mass spectrometry. Anal Bioanal Chem. 2022. PMID 36207535 DOI 10.1007/s00216-022-04336-5
  17. ClinicalTrials.gov. NCT02867267, the TESTS trial: a phase 3 multicentre randomized quadruple-masked placebo-controlled trial of thymosin alpha 1 in 1,106 adults with sepsis, sponsored by Sun Yat-sen University with SciClone Pharmaceuticals as collaborator, running September 2016 to March 2021, primary outcome 28-day all-cause mortality. No results are posted to the registry; the trial is published. Read through the v2 API on 23 August 2026. ClinicalTrials.gov. 2026. NCT02867267
  18. ClinicalTrials.gov. NCT00711620, the ETASS trial: a multicentre randomized controlled trial of thymosin alpha 1 in 361 patients with severe sepsis, sponsored by Sun Yat-sen University, completed. The registry records masking as DOUBLE, naming participant and outcomes assessor, while the publication's title describes the trial as single-blind. Both are recorded here and neither is resolved. Read through the v2 API on 23 August 2026. ClinicalTrials.gov. 2026. NCT00711620
  19. ClinicalTrials.gov. NCT01178996, a phase 3 multicentre double-masked trial in 552 patients with chronic hepatitis C who did not respond to prior peginterferon and ribavirin, sponsored by sigma-tau i.f.r. S.p.A. with SciClone Pharmaceuticals as collaborator, completed July 2009, primary outcome sustained virological response at week 72. No results are posted to the registry; the trial is published. Read through the v2 API on 23 August 2026. ClinicalTrials.gov. 2026. NCT01178996
  20. ClinicalTrials.gov. NCT00039962 and NCT00040027, two phase 3 randomized double-masked trials of thymalfasin plus peginterferon alfa-2a in hepatitis C non-responders, 500 participants each, sponsored by SciClone Pharmaceuticals, both started in 2002 and both recorded as completed. Neither has results posted, and no publication was located for either. Read through the v2 API on 23 August 2026. ClinicalTrials.gov. 2026. NCT00039962
  21. ClinicalTrials.gov. NCT00911443, a phase 2 open randomized dose-ranging trial in 488 participants with malignant melanoma, sponsored by sigma-tau i.f.r. S.p.A., completed September 2007, five arms of 97 to 99. Posted results: overall tumour response (complete plus partial) 7, 10, 6 and 12 in the four thymosin-alpha-1 arms against 4 in the dacarbazine plus interferon control arm; median overall survival 9.3, 8.6, 10.3 and 9.3 months against 6.6; median progression-free survival 1.9, 1.8, 1.8 and 2.0 months against 1.8. No statistical analyses are posted for any outcome. Read through the v2 API on 23 August 2026. ClinicalTrials.gov. 2026. NCT00911443
  22. ClinicalTrials.gov. NCT04428008, a randomized open-label trial of thymalfasin to prevent COVID-19 in 189 renal dialysis patients, completed January 2023. Posted results: COVID-19 infection in 5 of 91 on treatment against 7 of 98 in the control arm; hospitalisation 27 against 26; non-COVID infections 8 against 6; deaths 3 against 7. No analyses are posted. The record's limitations field states that the introduction of vaccine during the study made detection of infection by seroconversion unreliable. Read through the v2 API on 23 August 2026. ClinicalTrials.gov. 2026. NCT04428008
  23. ClinicalTrials.gov. Registry sweep for thymalfasin, thymosin alpha 1, thymosin alpha-1, Zadaxin and the bare term thymosin, run through the v2 API on 23 August 2026. The five queries return 91 distinct records, of which 62 name thymosin alpha-1 in their interventions: 9 at phase 3 including one phase 2/3, 32 at phase 2 including phase 1/2 and 2/3, 7 at phase 4, and 2 observational. Two of the 62 have posted results. The remaining 29 records in the union are almost entirely thymosin beta-4, a different molecule covered separately on this site. ClinicalTrials.gov. 2026. ClinicalTrials.gov
  24. US Food and Drug Administration. Drugs@FDA approved drug products, queried through the openFDA drugsfda endpoint on 23 August 2026 for thymalfasin, thymosin and Zadaxin across the generic name and brand name fields. All three queries returned NOT_FOUND: there is no approved US application under any of these names. openFDA. 2026. openFDA drugsfda
  25. US Food and Drug Administration. Certain Bulk Drug Substances for Use in Compounding that May Present Significant Safety Risks. The category 2 table carries the row Thymosin-alpha 1 (Ta1), with the stated reasoning that compounded drugs containing it may pose significant risk for immunogenicity for certain routes of administration and may have complexities with regard to peptide-related impurities and API characterization, and that the safety-related information is inadequate for the agency to sufficiently understand the extent of any safety issues raised. Page content current as of 22 April 2026, fetched with a browser user agent and read on 23 August 2026. FDA human drug compounding. 2026. FDA category 2 list
  26. World Anti-Doping Agency. The 2026 Prohibited List. Cited here only for the definition of class S0, which covers substances with no current approval by any governmental regulatory health authority for human therapeutic use. Thymosin alpha-1 holds approvals as a medicine outside the United States, so that definition appears not to reach it. This site did not establish whether any other class on the list names or reaches this compound, and states the placement as unsettled rather than resolved. The URL serves zero bytes to automation and is ledgered in tools/unverified-sources.json. World Anti-Doping Agency. 2026. WADA 2026 Prohibited List