Thymosin alpha-1

A synthetic copy of a 28-amino-acid peptide the thymus gland makes, sold under the official drug name thymalfasin and the brand name Zadaxin. The chain is the stretch of the human prothymosin alpha protein that runs from the second residue to the twenty-ninth, with the front end capped by an acetyl group. It is a real prescription medicine in Italy — approved there since the 1990s as a helper for the flu jab in people whose immune system is weak — and in the United States it is sold as a compounded or research-labelled online peptide for immunity, inflammation, long COVID and ageing, uses that its one European approval does not cover.

Sold as

Injection under the skin or into a muscle. Nasal sprays are sold online; no published human measurement used one.

Legal status in the EU

An approved prescription medicine in Italy and nowhere else in Europe. ZADAXIN 1,6 mg/ml, AIC 028364026, holder SciClone Pharmaceuticals Italy S.r.l., is listed as authorised, not revoked and not suspended in AIFA's Banca Dati Farmaci; the approved product information served by AIFA on 26 February 2026 gives a first authorisation of 27 April 1993 and a most recent renewal of May 2008, while the register's own authorisation date field says 15 May 1996 and AIFA's public assessment report says the Ministry of Health granted the authorisation in 1996 — the three dates are not reconciled in any public document. The sole approved indication is as an adjuvant to influenza vaccination in immunocompromised adults; it is prescription-only and reimbursement class C, meaning the patient pays. There is no centrally authorised EU product: the European Commission granted thymalfasin an orphan designation for liver cancer on 30 July 2002 (EU/3/02/110), which is a research incentive and never became a marketing authorisation. The WHO index assigns ATC code L03AX25. In the United States nothing containing it is approved; FDA's product databases hold only an unfinished bulk-ingredient registration by a Chinese supplier, and on 4 December 2024 the Pharmacy Compounding Advisory Committee voted 4 to 17 against adding thymosin alpha-1 to the list of substances American pharmacies may compound from, after the original nomination had already been withdrawn. It does not appear anywhere in the WADA 2026 Prohibited List, which was searched in full; its namesake thymosin beta-4 (TB-500) does appear, in section S2.3.

It attaches to a sensor on the scout cells that show the immune system what an infection looks like — TLR9 on dendritic cells and on the immature cells they grow from — and is meant to push young T cells to finish maturing, which is why it is described as an immune modulator rather than an antiviral or a chemotherapy. The chain SDAAVDTSSEITTKDLKEKKEVVEEAEN matches residues 2–29 of human prothymosin alpha in UniProt P06454 letter for letter; every residue is a standard L-amino acid and the only chemical difference from that stretch of the natural protein is an acetyl cap at the front end, which is why the one-letter code can honestly be printed here. FDA's own 2024 chemical description states the same structure as Ac-Ser-Asp-Ala-Ala-Val-Asp-Thr-Ser-Ser-Glu-Ile-Thr-Thr-Lys-Asp-Leu-Lys-Glu-Lys-Lys-Glu-Val-Val-Glu-Glu-Ala-Glu-Asn-OH, with formula C129H215N33O55 and a mass of 3108.3, and PubChem's chemical name for CID 16130571 opens with an acetylated serine.

Part of the bodyWhat happensHow it is knownKind of evidence

The immune system, taken as a whole

The idea is that it wakes up the cells that decide how the immune system responds. FDA's own reading of the science is that its effects are carried at least in part through one sensor, TLR9, on the scout cells that show the immune system what an infection looks like and on the immature cells those grow from. The one place this was measured properly against a dummy injection, it did not happen. In 1,089 adults with sepsis, a marker of how awake the immune system's front-line cells are ended at almost exactly the same level in both groups.

Measured in 1,089 adults with sepsis in a double-blind trial against dummy injections: after seven days, the monocyte HLA-DR marker was 7,494 antibodies per cell on the drug and 7,518 on the dummy (P = 0.89), and lymphocyte counts were 0.96 against 0.91 billion per litre (P = 0.07).

Measured in people

Source [04]Source [06]

How well a flu jab takes, in people whose immune system is weak

This is the only thing it is approved for anywhere in Europe. The claim is that giving it around the time of a flu vaccination makes the body produce more antibodies than the vaccine alone would. Five studies have looked, in older men and in people on dialysis, and they are the weakest part of the file rather than the strongest: four of the five measured antibodies with a laboratory test rather than counting who got flu, two exist only as conference abstracts, and none of them compared against the higher-dose flu vaccines now recommended for older people.

Measured across five studies totalling roughly 645 people: 9 and 330 elderly people in abstracts only, 90 elderly men in a double-blind placebo-controlled trial (85 analysed), 97 people on haemodialysis, and 120 people on dialysis in an open-label three-arm trial run by the manufacturer's licensee. FDA's 2024 review of all five concluded there is not enough evidence to reach any conclusion about it as a vaccine helper.

Measured in people

Source [04]Source [14]Source [15]

Hepatitis B virus in the blood

This is what it is best known for, and it is the use that has been tested longest. Given by itself for six months, it does not reliably clear the virus. The one trial designed properly to answer the question — everyone blinded, half given a dummy injection — did not reach the mark it set for itself, and its own authors wrote that the result did not confirm the earlier positive reports.

Measured in 97 adults with active hepatitis B in a double-blind placebo-controlled phase 3 trial, 1.6 mg twice a week for six months with six months of follow-up. Complete response was 7 of 49 (14%) on the drug against 2 of 48 (4%) on the dummy — a gap that did not reach statistical significance (P = 0.084).

Measured in people

Source [07]

Hepatitis C virus in the blood

Added on top of the standard hepatitis C treatment of its day, it made no difference to whether the virus was cleared. Modern hepatitis C tablets cure more than nine people in ten in eight to twelve weeks, so even the treatment it was added to has been superseded.

Measured in 552 adults whose hepatitis C had not responded to previous treatment, randomised to standard therapy plus the drug or standard therapy plus dummy injections for 48 weeks. Cure rates were 12.7% against 10.5% (P = 0.407).

Measured in people

Source [08]

The skin where the needle goes in

The one thing every source agrees actually happens. The Italian approved leaflet says the injection can occasionally cause temporary pain at the injection site and that no other clinically relevant reactions are known. FDA's review of the whole published literature reaches the same short list: local irritation, redness, discomfort where it was injected.

Stated in the approved Italian product information without a frequency figure, and confirmed as the commonest reported effect across published studies in FDA's 2024 review of doses from roughly 1 to 16 mg given for up to 12 months.

Measured in people

Source [02]Source [04]

How long it stays in the blood

It comes and goes fast. Injected under the skin it peaks within hours and is back to the starting level inside a day, which is why every schedule ever studied repeats the injection — twice a week for the approved use, twice a day in the sepsis trials.

The approved Italian product information states a blood peak six hours after an injection under the skin and a return to baseline within 24 hours. The published human study FDA relies on, in healthy men given 900 micrograms per square metre of body surface, puts the peak nearer two hours with a blood half-life of about two hours, and found no build-up after five days of daily dosing. The two figures for the peak are not reconciled anywhere public.

Measured in people

Source [02]Source [04]

Graded as an approved medicine because a current national marketing authorisation was verified directly against Italy's own live medicines register, not because of the widely repeated claim that it is approved in around thirty countries. AIFA's Banca Dati Farmaci returns ZADAXIN, active substance timosina alfa 1, AIC 028364026, holder SciClone Pharmaceuticals Italy S.r.l., administrative status AUTORIZZATA, revoked 0, suspended 0, marketed 1, prescription-only, reimbursement class C. FDA states independently that the substance is not approved in the United States, Japan or Europe except Italy. Two things have to be printed next to that grade. First, the Italian approval covers one indication only — adjuvant to influenza vaccination in immunocompromised adults — and AIFA's own public assessment report answers "how was Zadaxin studied" with laboratory and animal work alone. Second, what the compound is famous for has been tested and mostly failed. TESTS (BMJ 2025, 1,106 adults with sepsis enrolled at 22 Chinese centres, double-blind, 1,089 analysed) found 28-day mortality of 23.4 per cent against 24.1 per cent on dummy injections, hazard ratio 0.97 (0.76–1.24), with no difference in 90-day mortality, organ failure scores, secondary infections, lymphocyte counts or monocyte HLA-DR — the immune marker the drug is supposed to raise. The phase 3 hepatitis B trial in 97 adults reached a complete response in 14 per cent against 4 per cent on placebo (P = 0.084, not significant) and its authors wrote that the result did not confirm earlier positive reports. Adding it to hepatitis C treatment in 552 adults gave a cure rate of 12.7 per cent against 10.5 per cent (P = 0.407). The 488-patient melanoma trial missed its primary endpoint. Sixty-five studies are registered on ClinicalTrials.gov and only two have posted results.

Fast in the blood, and then nothing measurable at any point after that. A single injection under the skin peaks within two to six hours depending on which source you read, and is back to the starting level within 24 hours; giving it daily for five days produced no build-up. That is why every schedule ever studied repeats the dose. What happens after that is the part with no answer. In the sepsis trial, seven days of injections every 12 hours left the immune marker, the organ-failure score, the lymphocyte count and the death rate all indistinguishable from dummy injections at day 7, day 28 and day 90. In hepatitis B, six months of twice-weekly injections plus six months of follow-up did not reach the trial's own definition of success. Across the published literature FDA reviewed, treatment periods ran from a single day to twelve months, and the agency's summary of that entire range is that the most common reaction was discomfort where the needle went in.

What was measuredWhat happenedIn whomHow it is known

Dying within 28 days of severe infection

127 of 542 people died on thymosin alpha-1 (23.4%) against 132 of 547 on dummy injections (24.1%) — a hazard ratio of 0.97, with a range from 0.76 to 1.24 (P = 0.82). The range includes benefit, no difference and harm, which is what a trial says when the answer is no.

1,089 adults with sepsis across 22 centres in nine Chinese provinces, double-blind, injections every 12 hours for seven days

Measured in people

Source [06]

Dying within 90 days of severe infection

168 of 542 (31.0%) on thymosin alpha-1 against 177 of 547 (32.4%) on dummy injections — hazard ratio 0.95, range 0.77 to 1.17 (P = 0.61).

The same 1,089 adults with sepsis

Measured in people

Source [06]

The immune marker the drug is supposed to raise

After seven days, the monocyte HLA-DR reading — how switched-on the immune system's front-line cells are — was 7,494 antibodies per cell on the drug against 7,518 on the dummy injection (P = 0.89). Lymphocyte counts rose by the same amount in both groups (0.96 against 0.91 billion per litre, P = 0.07). Organ failure scores fell by a median of four points in both groups (P = 0.50).

The same 1,089 adults with sepsis

Measured in people

Source [06]

Picking up a second infection while in intensive care

137 of 542 (25.3%) on thymosin alpha-1 against 143 of 547 (26.1%) on dummy injections within 28 days (P = 0.74). Time in intensive care was a median of 15 days in both groups (P = 0.28).

The same 1,089 adults with sepsis

Measured in people

Source [06]

Dying within 28 days of severe sepsis, in the earlier and smaller trial

47 of 181 died on thymosin alpha-1 (26.0%) against 63 of 180 on saline dummy injections (35.0%). This is the number people quote, and it needs two caveats printed beside it: the answer depends on which test is read — the straightforward comparison of the two percentages did not reach statistical significance (P = 0.062) while the survival-curve test did (P = 0.049), the relative risk being 0.74 with a range from 0.54 to 1.02 — and only the patients and the statisticians were blinded, not the doctors treating them. Intensive-care deaths were 19.3% against 26.7% (P = 0.098), and time in intensive care and time on a ventilator were the same in both groups.

361 adults with severe sepsis in China, single-blind, 1.6 mg twice a day for five days then once a day for two

Measured in people

Source [16]

Clearing hepatitis B virus from the blood

7 of 49 people (14%) reached the trial's definition of a complete response against 2 of 48 (4%) given dummy injections. The gap did not reach statistical significance (P = 0.084). Counting more loosely, 25% against 13% eventually lost detectable virus (P < 0.11, again not significant). The authors' own summary: these results do not confirm the effect reported in earlier studies.

97 adults with active hepatitis B, double-blind, 1.6 mg twice weekly for six months, followed for a further six

Measured in people

Source [07]

Being cured of hepatitis C

12.7% cured against 10.5% on dummy injections, counting everyone who was randomised (P = 0.407). Among the minority who finished all 48 weeks the figures were 41.0% (34 of 83) against 26.3% (26 of 99), P = 0.048 — but that comparison drops two-thirds of the people who started, so it is a hypothesis the authors themselves labelled as one, not a result.

552 adults whose hepatitis C had already failed to respond to standard treatment, double-blind, 48 weeks

Measured in people

Source [08]

Tumours shrinking in advanced melanoma

The trial's main question was how many tumours responded, and the answer was no different: 7.2%, 10.3%, 6.1% and 12.1% in the four thymosin arms against 4.1% in the control arm, with no significant difference for any of them. Time before the cancer grew was 1.8 to 2.0 months on thymosin against 1.8 months on the control. Median survival posted to the trial registry was 9.3, 8.6, 10.3 and 9.3 months against 6.6 months — a difference that looks large, but it came from an open-label exploratory trial whose main endpoint had already failed, and no significance test for it was posted.

488 adults with advanced spreading melanoma, open-label, randomised to five groups, sponsored by sigma-tau

Measured in people

Source [04]Source [18]

Catching COVID-19 while on kidney dialysis

5 of 91 people on thymosin alpha-1 caught COVID-19 against 7 of 98 on standard care over six months. Hospital admissions were 27 of 91 against 26 of 98, non-COVID infections 8 of 91 against 6 of 98, and deaths 3 of 91 against 7 of 98. All of these are single-digit differences in small groups, in a trial with no dummy injection and no blinding, so none of them separates the drug from chance.

189 adults with end-stage kidney disease on dialysis, randomised open-label, 1.6 mg twice weekly for eight weeks, results posted to the registry in April 2026

Measured in people

Source [21]

Death in sepsis, when every trial is pooled together

The pooled figure across 11 randomised trials favours the drug — an odds ratio of 0.73, range 0.59 to 0.90 — but it falls apart the moment the weaker trials are set aside. Restricted to the trials the reviewers rated high quality, the odds ratio is 0.82 with a range from 0.65 to 1.03, which crosses no-difference. Restricted to multi-centre trials it is 0.86, range 0.68 to 1.08, which also crosses no-difference. The reviewers' own trial-sequential analysis says the total number of patients studied is still not enough to settle it.

1,927 adults with sepsis across 11 randomised trials, 967 given the drug and 960 controls, including the 1,106-patient TESTS trial

Studies in people disagree

Source [17]

Dying of COVID-19 in hospital

Two published pooled analyses of the same literature reach opposite answers. One, across 8 studies, reports a risk ratio of 0.59 with a range from 0.37 to 0.93, meaning fewer deaths. The other, across 9 studies and 5,352 patients, reports 1.03 with a range from 0.60 to 1.75, meaning no difference at all. FDA reviewed four such analyses and reported that three of the four found no reduction in deaths.

Adults hospitalised with COVID-19, almost all in retrospective records from Chinese hospitals early in the pandemic rather than in randomised trials

Studies in people disagree

Source [04]Source [19]Source [20]

The approved Italian leaflet names one side effect — occasional temporary pain where the needle goes in — and states that no other clinically relevant reactions are known, giving no frequency for even that. The most informative safety figure comes from the sepsis trial, where side effects were reported by 66.4 per cent of the 542 people on the drug and 67.6 per cent of the 547 on dummy injections, and serious events by 26.8 against 29.3 per cent: nothing separated. Specific harms are on record without rates. Liver-enzyme flares in people with hepatitis B, written into the label used in Indonesia. Fatal immune destruction of red cells and graft failure in stem-cell transplant recipients, where pushing the immune system runs directly against the deliberate suppression the transplant needs. Thyroid blood tests going out of range in a hepatitis C study. The label bars it in pregnancy and breastfeeding for lack of data, bars it under 18, asks for caution in anyone with a history of allergic reactions, and says use in autoimmune disease has to be decided case by case. FDA's central objection was that nobody has measured whether repeated injection teaches the body to make antibodies against it, and that there is no pharmacopoeia monograph, no impurity testing data and no aggregation data for the raw substance — so what is in an unapproved compounded or online vial is verified by nobody.

EffectWhat it feels likeHow often

A dangerous immune reaction in people having a stem-cell transplantSerious

This is the one place FDA names a specific mechanism for harm rather than a general worry. People having a bone-marrow or stem-cell transplant are deliberately having their immune system suppressed so the graft takes, and a drug that pushes the immune system in the opposite direction can cause or worsen graft-versus-host disease and can stop the graft from taking at all. Fatal immune destruction of red blood cells and graft failure have both been reported in that setting.

Not measured as a rate. Reported in a study of 14 transplant recipients; a separate case series of 8 recipients found infections went up rather than down in the four given the drug.

Source [04]

Anything at all, compared against a dummy injection

This is the most useful safety number in the whole file, and it cuts both ways. Two-thirds of people in a sepsis trial have side effects whether or not they get the drug, because they are critically ill. Nothing in that trial separated the drug from the dummy — which is the honest way to say "well tolerated", and it is a much narrower claim than the marketing version.

In the largest trial, 360 of 542 people on the drug reported some side effect (66.4%) against 370 of 547 on the dummy injection (67.6%), P = 0.70. Serious events were 145 of 542 (26.8%) against 160 of 547 (29.3%), P = 0.38.

Source [06]

Injection-site pain

The only side effect the approved European product information names, and it names it without a number, which tells you how old and how small the dossier behind the approval is. FDA's read of the whole published literature lands in the same place: local irritation, redness and discomfort at the injection site are the commonest thing reported.

No frequency exists. The approved Italian leaflet says only that it can "occasionally" cause temporary pain where the needle went in.

Source [04]Source [22]

Liver enzymes jumping in people with hepatitis B

A sudden rise in the blood tests that track liver inflammation, called a flare. In hepatitis B a flare can be the immune system finally attacking infected liver cells, which is sometimes the point of the treatment and sometimes dangerous — the drug is being asked to provoke exactly the reaction that can also damage a liver. No trial has reported how often it happens or how often it matters.

Not measured as a rate. Reported in one Japanese hepatitis B study and written into the product label used in Indonesia.

Source [04]

Thyroid blood tests going out of range

The one report in FDA's adverse-event system describes a 46-year-old man in a trial who was hospitalised with anxiety, an irregular heartbeat and a temporary drop in his thyroid signal. He was also taking pegylated interferon, which causes all three of those on its own label, so nobody can say which drug did it.

Not measured as a rate. Reported in a hepatitis C study, and in the single adverse-event report FDA found in its own database.

Source [04]

An immune reaction against the drug itself

A 28-building-block peptide injected under the skin can teach the body to make antibodies against it, and clumping in the vial makes that more likely. FDA's central objection to letting American pharmacies make this was that nobody has looked: there is no impurity testing data, no clumping data, and no measurement of whether people who inject it for years develop antibodies. That is a gap in the record, not a reassurance.

Never measured. No study has looked.

Source [04]

These are the amounts the studies below gave their participants, and they are here as facts about those studies. They are not a recommendation, not a starting point and not a range to pick from. What a person should take, if anything, is a question for a doctor who knows them.

The only approved use anywhere in Europe: helper for the flu jab in immunocompromised adults, as written in Italy's approved product information

One 1.6 mg vial injected into muscle or under the skin, twice a week — a weekly total worked out as 900 micrograms per square metre of body surface

Four weeks from the day of vaccination, then the same again from week 8 to week 12

Source [02]

TESTS, the 1,106-adult sepsis trial that found nothing

1.6 mg injected under the skin every 12 hours

Seven days, or until leaving intensive care

Source [06]

ETASS, the earlier 361-adult severe sepsis trial

1.6 mg injected under the skin twice a day, then once a day

Five days at twice daily, then two days at once daily

Source [16]

The 97-adult double-blind hepatitis B trial

1.6 mg twice a week

Six months, with six months of follow-up

Source [07]

The 552-adult hepatitis C trial, on top of standard treatment

1.6 mg injected under the skin twice a week

48 weeks

Source [08]

The 488-adult melanoma trial, which tested three different amounts against a control arm

1.6 mg, 3.2 mg or 6.4 mg alongside chemotherapy, repeated every four weeks

Up to six cycles

Source [23]

The 120-patient dialysis trial of it as a flu-vaccine helper, run by the manufacturer's licensee

3.2 mg or 6.4 mg, given seven days before the vaccination and again on the day of it

Two doses

Source [15]

The 189-patient dialysis trial of it as COVID-19 prevention

1.6 mg injected under the skin twice a week after dialysis

Eight weeks, with six months of follow-up

Source [24]

What people report

• Whether it actually reduces influenza in the people Italy approves it for — every study of that use measured antibodies in a laboratory, and none counted who got flu
• How often the one side effect its approved label names actually happens — the label says "occasionally" and gives no number
• Whether injecting it for months or years teaches the body to make antibodies against it — no study has looked, and FDA named this as its central unanswered safety question
• What is in it after long-term use, chemically — there is no United States Pharmacopeia monograph for the substance, and FDA found no impurity or clumping data in the material submitted to it
• Whether it does anything for early-stage autoimmune disease, which is what clinicians told FDA they mostly prescribe it for — that use has never been studied
• Whether it helps chronic fatigue syndrome or long COVID — FDA searched and found no clinical study in people at all
• Why its country-by-country approvals differ so widely in what they cover — the only European approval verified here is for one narrow vaccination use, while the same product is described elsewhere as a hepatitis treatment
• Whether the sepsis signal in the smaller, single-blind 2013 trial was real — the 1,106-patient blinded trial that followed found nothing, and the pooled analysis loses its effect as soon as weaker trials are excluded
• Whether it is safe in pregnancy or breastfeeding — the approved label bars both because there are no data either way, not because harm was seen
• Whether it does anything in children — the approved label bars it under 18, and AIFA's public assessment report says effectiveness and safety were never established there
• What its long-term effects are beyond twelve months — that is the longest treatment period in any published study FDA could identify
• Whether the drug reaches the blood at all from the nasal sprays sold online, since every published human measurement used an injection
• How much of it is in an online vial, and whether it is the right molecule — FDA found compounded and research-labelled versions sold online in the United States, including one website advertising its product as FDA-approved when no product containing this peptide is approved there at all

Does thymosin alpha-1 actually work?

For the one thing it is approved for in Europe — making a flu vaccination take better in people whose immune system is weak — the evidence is five small studies, four of which measured antibodies in a laboratory rather than counting who got flu, and FDA's 2024 review of all five concluded there was not enough there to reach any conclusion. For everything it is famous for, the answer is closer to no. The largest trial ever run gave it to 542 adults with sepsis and dummy injections to 547: 23.4% died within 28 days against 24.1%, and even the immune marker the drug is meant to raise ended identical in both groups. In hepatitis B and hepatitis C, the properly blinded trials did not beat their dummy injections either.

Source [04]Source [06]Source [07]

Is thymosin alpha-1 FDA approved?

No. FDA has approved no medicine containing thymosin alpha-1 for any use, and said so in its own words in December 2024: it is not approved in the United States, Japan, or Europe except Italy. The only trace of it in FDA's product databases is a raw ingredient registration by a Chinese supplier, which is a listing of a chemical, not an approval of a medicine.

Source [04]Source [11]Source [25]

Is thymosin alpha-1 legal, and can you buy it?

In Italy it is a normal prescription medicine, sold as Zadaxin, prescription-only and paid for by the patient. In the rest of Europe and in the United States there is no approved product, so nothing sold to you is an approved medicine. In the United States it is also not on the list of substances American pharmacies may legally compound from: an FDA advisory committee voted 4 to 17 on 4 December 2024 against adding it, after the original request to add it had already been withdrawn. What is sold online under research labelling is made by nobody a regulator inspects, and FDA found one website advertising a compounded version as FDA-approved when no such approval exists.

Source [01]Source [04]Source [05]

What are the side effects of thymosin alpha-1?

The approved European leaflet names exactly one: occasional temporary pain where the needle goes in, with no other clinically relevant reactions known — and it gives no frequency for even that. The largest randomised trial is the better guide, and it found side effects in 66.4% of people given the drug and 67.6% of people given a dummy injection, meaning nothing in it was traceable to the drug. Two specific worries are on record without a rate: liver-enzyme flares in people with hepatitis B, and graft-versus-host disease or graft failure in people having a stem-cell transplant. Whether long-term use teaches the body to make antibodies against it has never been measured.

Source [04]Source [06]Source [22]

Thymosin alpha-1 vs thymosin beta-4 (TB-500) — what is the difference?

They share a name and nothing else. Thymosin alpha-1 is a 28-building-block peptide aimed at the immune system, approved as a medicine in Italy, and not banned in sport. Thymosin beta-4 is a different molecule aimed at tissue repair, sold online as TB-500, approved nowhere — and it is explicitly banned in sport at all times, named as "Thymosin-ß4 and its derivatives e.g. TB-500" in the growth-factor section of the 2026 World Anti-Doping Agency list. Anyone who buys one thinking it is the other has bought the wrong thing, with different legal consequences.

Source [04]Source [12]

Is thymosin alpha-1 banned in sport?

No. It does not appear anywhere in the 2026 World Anti-Doping Agency Prohibited List, which was searched in full for this entry. Its namesake thymosin beta-4, sold as TB-500, is banned. Two cautions still apply: the WADA list is revised every year, and an unapproved product bought online can contain something that is banned even when the peptide named on the label is not.

Source [12]

Does thymosin alpha-1 help long COVID, chronic fatigue or a weak immune system?

For chronic fatigue syndrome, FDA searched the literature in 2024 and could not find a single clinical study in people. For COVID-19 itself, two published pooled analyses of the same hospital records reach opposite answers — one finds fewer deaths, the other finds no difference — and three of the four analyses FDA reviewed found no reduction in deaths. As for a generally weak immune system, the drug did not move the immune marker it is supposed to move in the one large trial that measured it against a dummy injection. The most honest summary is that this is the use it is most sold for and the use with the least evidence behind it.

Source [04]Source [06]Source [20]

Each letter represents one amino acid.

1. [01]AIFA Banca Dati Farmaci — ZADAXIN (timosina alfa 1), AIC 028364026, status AUTORIZZATA, holder SciClone Pharmaceuticals Italy S.r.l., prescription-only, reimbursement class C (2026)
2. [02]ZADAXIN 1,6 mg/ml — Riassunto delle Caratteristiche del Prodotto (approved Italian summary of product characteristics), served by AIFA 26 February 2026 (2026)
3. [03]AIFA — Riassunto della Relazione Pubblica di Valutazione: ZADAXIN (timalfasina), AIC 028364, dated 18 October 2016 (2016)
4. [04]FDA Pharmacy Compounding Advisory Committee briefing slides — Thymosin Alpha-1 (Ta1) Related Bulk Drug Substances, 4 December 2024 (chemistry, safety, effectiveness for 12 evaluated uses; "Ta1 is not approved in the United States, Japan, or Europe (except Italy)") (2024)
5. [05]FDA Pharmacy Compounding Advisory Committee, Topic 3, 4 December 2024 — official transcript, vote recorded as 4 yeses, 17 noes, 0 abstentions (2024)
6. [06]TESTS: the efficacy and safety of thymosin α1 for sepsis — multicentre, double blinded, randomised, placebo controlled, phase 3 trial. BMJ 2025;388:e082583 (1,106 adults; 28-day mortality 23.4% vs 24.1%) (2025)
7. [07]Mutchnick MG et al. Thymosin alpha1 treatment of chronic hepatitis B: phase III multicentre, randomized, double-blind, placebo-controlled study. J Viral Hepat 1999 (97 adults; 14% vs 4%, P = 0.084) (1999)
8. [08]Ciancio A et al. Thymosin alpha-1 with peginterferon alfa-2a/ribavirin for chronic hepatitis C not responsive to IFN/ribavirin. J Viral Hepat 2012 (552 adults; 12.7% vs 10.5%, P = 0.407) (2012)
9. [09]UniProt P06454 — human prothymosin alpha; residues 2–29 give SDAAVDTSSEITTKDLKEKKEVVEEAEN
10. [10]PubChem CID 16130571 — thymalfasin, C129H215N33O55, 3108.3 g/mol; chemical name begins with an acetylated serine
11. [11]European Medicines Agency — EU/3/02/110, orphan designation of thymalfasin for hepatocellular carcinoma, granted 30 July 2002; never followed by a marketing authorisation (2002)
12. [12]WADA World Anti-Doping Code International Standard — Prohibited List 2026 (searched in full; thymosin alpha-1 and thymalfasin do not appear, thymosin beta-4 and TB-500 appear in S2.3) (2026)
13. [13]WHO Collaborating Centre for Drug Statistics Methodology — ATC index, thymalfasin = L03AX25 (last updated 20 January 2026) (2026)
14. [14]Gravenstein S et al. Augmentation of influenza antibody response in elderly men by thymosin alpha one — a double-blind placebo-controlled clinical study. J Am Geriatr Soc 1989 (1989)
15. [15]ClinicalTrials.gov NCT01031966 — pilot, randomised, open-label study of two doses of thymosin alpha 1 on the immunogenicity of H1N1 vaccine in 120 dialysis patients, sponsor sigma-tau i.f.r. S.p.A., completed July 2010, no results posted (2010)
16. [16]ETASS: the efficacy of thymosin alpha 1 for severe sepsis — a multicenter, single-blind, randomized and controlled trial. Crit Care 2013 (2013)
17. [17]Efficacy of thymosin α1 for sepsis: a systematic review and meta-analysis of randomized controlled trials. Front Cell Infect Microbiol 2025 (2025)
18. [18]ClinicalTrials.gov NCT00911443 — posted results: overall survival and progression-free survival by arm, 488 participants, sponsor sigma-tau i.f.r. S.p.A. (2007)
19. [19]Soeroto AY et al. The efficacy of thymosin alpha-1 therapy in moderate to critical COVID-19 patients: systematic review, meta-analysis and meta-regression. Inflammopharmacology 2023 (2023)
20. [20]Wang Y et al. Thymosin alpha1 use in adult COVID-19 patients: a systematic review and meta-analysis on clinical outcomes. Int Immunopharmacol 2023 (2023)
21. [21]ClinicalTrials.gov NCT04428008 — a pilot trial of thymalfasin to prevent COVID-19 infection in renal dialysis patients, posted results, 189 participants (2026)
22. [22]ZADAXIN 1,6 mg/ml — Foglio illustrativo, section 4, and Riassunto delle Caratteristiche del Prodotto, section 4.8, served by AIFA 26 February 2026 (2026)
23. [23]ClinicalTrials.gov NCT00911443 — arm descriptions and posted results (2007)
24. [24]ClinicalTrials.gov NCT04428008 — arm descriptions and posted results (2026)
25. [25]openFDA Drugs@FDA query for thymalfasin and Zadaxin — no matches; the NDC directory holds one unfinished bulk-ingredient listing (2026)