Semax

A synthetic chain of seven amino acids, made in the 1980s at the Institute of Molecular Genetics of the Russian Academy of Sciences. Its first four building blocks, Met-Glu-His-Phe, copy positions 4 to 7 of ACTH, the hormone the pituitary gland uses to tell the adrenal glands to make cortisol; the last three, Pro-Gly-Pro, are a tail added to slow the body's enzymes down. It is not a hormone: the literature describes it as having none of ACTH's effects on the adrenal glands or on skin pigment. Semax is a registered medicine in Russia, sold there as 0.1% and 1% nasal drops and named on the government's list of vital and essential medicines for 2019. It is not approved in the EU or the United States, where it is bought online as a nasal spray and as a freeze-dried powder labelled for research use only. PubChem gives C37H51N9O10S and 813.9 g/mol (CID 9811102) and its chemical name shows every residue in the standard L form and a plain, uncapped end to the chain, which is why the sequence can be printed here. That sequence describes plain semax and its acetate salt. It does not describe the version widely sold online as N-Acetyl Semax Amidate, which carries a cap at each end of the chain and is a different molecule.

Sold as

Nasal drops or nasal spray in every published human study, and the only form registered in Russia. Also sold online as a nasal spray and as a freeze-dried powder for injection under the skin; no published study has ever given semax to a person by injection, and the FDA reported in 2026 that it could find no safety or pharmacokinetic data for that route in any species.

Legal status in the EU

Not approved as a medicine in the EU or the United States, and no trial of it is registered on ClinicalTrials.gov. When the FDA searched the Russian state medicines register, the European Pharmacopoeia (11.8), the Japanese Pharmacopoeia (18th) and the International Pharmacopoeia (12th) in December 2025, the only registration it reported was Russia's: semax is a registered drug there, available as 0.1% and 1% nasal drops, and the nominators' own answer on foreign recognition named the Russian Pharmacopoeia and nothing else. Russia also named it, as methionyl-glutamyl-histidyl-phenylalanyl-prolyl-glycyl-proline in nasal-drop form, on the government's list of vital and essential medicines for 2019; that order was superseded in October 2019 and this register could not read the current list. In the United States it cannot legally be made up by a pharmacy either: two nominations asked for semax to be added to the section 503A list of bulk substances, both were withdrawn, and the FDA carried on evaluating it anyway — its assessment of 11 May 2026, put to an advisory committee meeting held 23–24 July 2026, proposed adding neither semax free base nor semax acetate, on the grounds that neither is well characterised, that there is insufficient evidence of effectiveness for cerebral ischaemia, migraine or trigeminal neuralgia, that the safety profile cannot be characterised at all, and that approved medicines for those conditions already exist. No outsourcing facility has reported compounding semax since 2019.

Nobody knows how it works. In rats it switches on the genes for two nerve-growth proteins, BDNF and NGF, within an hour of a dose, and after an experimental stroke it changes the activity of dozens of genes — 96 at three hours and 68 at a day, most of them genes of the immune system and the blood vessels — but the US regulator's own summary in 2026 is that the molecular mechanism by which semax regulates gene expression is unknown. Radiolabelled semax sticks to something in rat brain tissue tightly, and one proposal is that the target is the enzymes that break down the body's own painkilling peptides, which semax blocks in a test tube. That does not explain much: when the pain-relief effect was tested in rats, blocking opioid receptors did not abolish it while blocking two serotonin receptors did. It is also not selective — in mice it made a dose of amphetamine release more dopamine in the brain's reward area than the amphetamine did alone, and in rats it slows blood clotting. Some of what is credited to semax may belong to its leftovers: sprayed into the nose it is chopped up fast, and the main fragment found in blood and brain, Pro-Gly-Pro, has effects of its own.

Part of the bodyWhat happensHow it is knownKind of evidence

Face and head pain

Trigeminal neuralgia is sudden, stabbing pain down one side of the face, set off by eating or talking. In the one published study, 16 people with it were given a single dose of semax up the nose. Nothing changed. Their pain was the same, the amount of electrical stimulation of the face they could take before it hurt was the same, and the electrical recordings of the face nerve were the same. In nine other people with a related jaw and gum pain, six said the pain went and three said it eased — but the rating scale showed no change in how often the attacks came, how long they lasted or what they felt like. In twelve people with migraine, the headache stopped in four and did not stop in the other eight. The authors' own conclusion was that semax does not have a painkilling effect by itself.

One study in 37 adults in 1996, one dose each, given up the nose. There was no dummy spray, nobody was blinded, and the paper does not print the actual scores or how much they varied. The rating scale it used is not described anywhere and the US regulator could not find any account of it. The regulator read this paper in detail in 2026 and concluded it showed a lack of effect.

One small study

Source [01]Source [13]

Motor neurone disease

Motor neurone disease kills off the nerves that drive the muscles, so the muscles waste and stop working. Twenty-seven people with it were given semax up the nose for two ten-day courses. It did not change the nerve and muscle recordings, and it did not change how fast the disease took away their speech, arm or leg function. What did change was a questionnaire score about how they were living with the illness, which the authors put down to better mood and motivation, strongest on day 10.

An open study in 27 patients in 2007, meaning everybody knew what they were getting and there was no comparison group. Three muscles were recorded before, during and after treatment. The authors state plainly that semax did not affect the course of the nerve damage or the clinical scores.

One small study

Source [09]

Stomach ulcers

An ulcer is a raw patch in the lining of the stomach or the gut. In one study, people whose ulcers had not healed on the usual medicines were given semax up the nose on top of those medicines, and compared with people who got the usual medicines alone. After two weeks, far more of the semax group had healed. This is one of the very few human results on this compound that has a comparison group at all, and it comes from the same institute that invented the peptide.

32 adults with ulcers that had resisted standard treatment, 1% semax as nose drops three times a day for 10 days on top of omeprazole, bismuth and a tissue-repair injection, against the same treatment without semax. The report does not say how people were put into the two groups, does not say whether anyone was blinded, does not print the group sizes, and does not discuss side effects. Its own authors wrote that clinical studies of this were still needed.

One small study

Source [01]Source [10]

Recovery after a stroke

This is the use semax is registered for in Russia. Two studies from the same Moscow neurology group are the core of it. In the first, from 1997, 30 people in the first days after a stroke were given semax on top of normal intensive treatment and compared with 80 people who got normal treatment alone; the authors report that the movement and general brain symptoms came back faster. In the second, from 2018, 110 people in rehabilitation after a stroke were compared according to whether or not they were given semax; the authors report a faster and better recovery of everyday function, and a rise in a blood marker called BDNF, a protein that helps nerve cells survive and rewire.

Neither study used a dummy spray, and neither abstract reports any blinding. In the 1997 study the comparison group was picked to match on stroke severity and location, not randomised. In the 2018 study the comparison was between people in the same trial who did and did not get the drug. Both papers are in Russian; only their English abstracts could be read here, and neither abstract prints the actual scores. The US regulator set both aside in 2026 because it had no verified English translation, so its conclusion of insufficient evidence is partly a conclusion about language.

One small study

Source [01]Source [08]Source [14]

Attention and memory, in healthy people

This is what almost everyone outside Russia actually buys it for. The published support in healthy people is one study from 1996 in 19 healthy men, given one or two doses up the nose. It appeared in a small journal that is not indexed in the main medical database, and it could not be opened for this entry. The US regulator did not treat it as evidence of effect — it noted the study was in healthy people rather than patients, and listed it under safety information. When the same regulator looked for literature supporting the attention-disorder use the substance had been nominated for, it found none, and it noted that there is no medical definition or treatment guideline for a nootropic at all.

One 1996 study in 19 healthy adult men, 1 mg up the nose daily for two days or a single 0.25 mg dose. No later controlled study of memory or attention in people has been published. The 2026 regulatory review searched PubMed, Embase, the Cochrane database and the trials register and found nothing further.

One small study

Source [01]

Background brain activity, seen on a scanner

The default mode network is the pattern of activity the brain falls into when a person is lying still and not doing anything in particular. In one experiment, healthy volunteers were scanned before a nasal dose and again five and twenty minutes after. In the people given semax, one part of that network at the front of the brain came out larger than in the people given a dummy spray. A second experiment from the same group looked at how strongly the fear-processing part of the brain talked to the temporal lobe, and reported changes there too. Nobody has shown that either of these corresponds to anything a person would notice.

24 healthy adults, 14 given 1% semax up the nose and 10 given a dummy spray, scanned three times in half an hour. A second study reported 52 healthy participants split between semax, a related peptide called selank and a dummy. The US regulator noted the two groups of 14 look very similar and it is not clear whether they were the same people. These are pictures of the brain, not measurements of memory, mood or behaviour.

One small study

Source [01]Source [11]Source [15]

Blood clotting

In rats, semax makes the blood slower to clot. Given up the nose for five days it raised the activity of the system that dissolves clots, and the animals formed smaller clots when a vein was clamped. In rats put under stress, which normally makes blood clot more readily, semax pushed them the other way — and it pushed them past normal, into a state where blood clotted less than in untreated animals. The US regulator picked this out in 2026 as a bleeding risk, particularly for anyone already on a blood thinner, and pointed out that a product bought online comes with no warning label saying so.

Rats only. One fixed amount was used in each experiment, so nobody knows whether a smaller amount would nudge clotting and a larger one cause bleeding. No study has measured clotting in a person given semax. The tripeptide left over when the body breaks semax down, Pro-Gly-Pro, has its own anticlotting effect, so part of what is seen may not be semax at all.

Only seen in animals

Source [01]Source [16]

The brain after a stroke, in rats

This is where the neuroprotection claim actually comes from. In rats given a clot in one part of the brain, semax up the nose left a smaller damaged area than water did, and the animals kept a memory task the untreated ones lost. In rats whose brain blood supply was cut off on both sides, semax into the abdomen reduced the visible damage and left the small blood vessels flowing rather than stalled. Semax also switches on the genes for two nerve-growth proteins, BDNF and NGF, within an hour of a dose, and after a stroke it changes the activity of dozens of genes — 96 at three hours and 68 at a day — most of them genes of the immune system and the blood vessels.

Rats, at one fixed amount per experiment: 250 micrograms per kilogram up the nose or 100 micrograms per kilogram into the abdomen. The damaged areas were about 20 per cent smaller than in the water-treated animals eight days after the clot. None of these studies measured how much semax was in the blood, so there is no way to link an amount in an animal to an amount in a person. How semax changes gene activity at all is unknown.

Only seen in animals

Source [01]Source [17]

Mood and anxiety

Anxiety and depression are among the things semax is advertised for online, and the animal work is real: in rats put under weeks of unpredictable stress, daily injections of semax reduced the loss of interest in sweet water that the stress normally causes. In people, there is one report, and it is a conference abstract: 120 children aged 12 to 14 with depression were given semax alongside talking therapy and a physiotherapy machine, all at once, so nothing in it can be pinned on the peptide. A second abstract covers 331 children with tics and Tourette syndrome given semax alongside two other drugs. Those 451 children are the largest group of people ever reported to have taken semax, and neither report is a full paper.

Rats, one fixed amount daily. In people, two conference abstracts in which semax was never given on its own and no full study could be located. Neither abstract discusses side effects.

Only seen in animals

Source [01]Source [18]

Dopamine, the brain's reward signal

In mice, semax made a dose of amphetamine release more dopamine in the reward part of the brain than amphetamine did on its own, and made the animals more active on it. Dopamine going up in that part of the brain is the same response street stimulants produce, and the US regulator flagged it in 2026 as concerning. Whether semax itself is habit-forming has never been tested, in any animal or any person, and neither has what it does alongside a stimulant in a human being.

Mice, two amounts, measured by sampling fluid from the brain. No study of abuse potential exists. The regulator states this in its own words: it did not identify nonclinical studies to demonstrate whether semax has reinforcing and addictive properties.

Only seen in animals

Source [01]

The one patient study the US regulator was able to read in English failed, and it had no comparison group of any kind. A single dose up the nose changed nothing at all in 16 adults with trigeminal neuralgia — not their pain, not how much electrical stimulation of the face they could take before it hurt, not the electrical recordings of the face nerve — and the authors' own conclusion was that semax does not have a painkilling effect by itself; in 12 adults with migraine the headache stopped in 4 and was still there in the other 8 (Koroleva et al. 1996, read through the FDA briefing document of 11 May 2026). That document also counted the entire readable clinical record for this compound: 33 to 47 healthy adults, 69 adults with a medical condition, and 451 children — and the children only ever received semax alongside other treatments, in conference abstracts that were never published as full papers. The rest of the human evidence is in Russian and the FDA set it aside untranslated under 21 CFR 10.20(c)(2), so its finding of insufficient evidence is partly a finding about language. What that Russian literature contains, read here through English abstracts: two non-randomised stroke studies from one Moscow group (30 treated against 80 matched controls in 1997; 110 patients compared by whether or not they received semax in 2018, with no figures printed in the abstract); an open study in 27 people with motor neurone disease where semax did not affect the nerve recordings or the disease scores at all and only a quality-of-life questionnaire moved; and an add-on ulcer study in 32 adults where 89.5 per cent healed by day 14 against 30.8 per cent on the standard medicines alone, with no group sizes and no allocation method printed. The only placebo-controlled work in people is two brain-scan studies from one group, in which 14 volunteers on semax showed a larger front section of the brain's resting network than 10 on a dummy spray, with nothing measured about how anyone felt or performed. No published study has given semax to a person by injection, which is one of the two ways it is sold. A ClinicalTrials.gov API v2 search on 3 August 2026 returned totalCount 0 for semax as a general term and as an intervention, against a control query returning 757 studies for semaglutide.

Unknown in people. The US regulator reported in May 2026 that it could find no study of what happens to semax in a human body — not how much gets in, not where it goes, not how fast it is broken down — by any route. What exists is in rats. Sprayed up the nose, the amount in the brain peaks about two minutes later, and it is tiny: roughly 0.07 per cent of the dose given. Injected into a vein it is smaller still, about 0.005 per cent, which is why the nose is used. It is then chopped up quickly from one end, and the main thing left over in blood and brain is a three-amino-acid fragment, Pro-Gly-Pro, which has effects of its own — so some of what is credited to semax may belong to its leftovers. As for how long people took it: a single dose in the pain and brain-scan studies, 10 days in the ulcer study, and two 10-day courses with a gap in the stroke and motor neurone studies. Nobody has published anything about taking it for longer than that.

What was measuredWhat happenedIn whomHow it is known

Whether an ulcer had healed after two weeks

89.5 per cent healed on semax added to the standard ulcer medicines, against 30.8 per cent on the standard medicines alone. The paper does not print the number of people in each group, does not describe how they were allocated and does not say whether anyone was blinded.

32 adults whose ulcers had not healed on standard treatment, given 1% semax as nose drops three times a day for 10 days on top of that treatment, or the treatment alone

One small study

Source [01]Source [10]

Facial nerve pain, and the nerve recordings behind it

Nothing changed. The pain scale, the amount of electrical stimulation of the face people could take before it hurt, and the electrical recordings of the face nerve were all the same before and after. There was no comparison group at all, so there is not even a dummy figure to set against it. The authors' conclusion was that semax does not have a painkilling effect by itself.

16 adults with trigeminal neuralgia, aged 45 to 65, given one dose of 0.5 mg per kilogram of body weight up the nose

One small study

Source [01]

Migraine headache after a single dose

The headache stopped in 4 of the 12 people, 90 to 120 minutes after the dose. In the other 8 it was still there, described as less severe. The overall pain score fell from 78 per cent to 32 per cent. There was no dummy spray and nobody was blinded, so there is no comparison figure for a headache that would have eased on its own — which most migraines do.

12 adults with migraine, 10 women and 2 men aged 19 to 56, one dose of 0.5 mg per kilogram of body weight up the nose

One small study

Source [01]

Jaw and gum pain, where people said one thing and the scale said another

Six of the nine people said their pain had gone and three said it was easier. The rating scale used in the same study showed no reduction in how often the attacks came, how long they lasted, or what the pain felt like. There was no comparison group and nobody was blinded, and this is the clearest example in the whole record of what people report and what is measured pointing in different directions.

9 adults with dental plexalgia, 3 women and 6 men aged 15 to 52, one dose of 0.5 mg per kilogram of body weight up the nose

One small study

Source [01]

Nerve and muscle function in motor neurone disease

Unchanged. The needle recordings of the muscles and the disease rating scales moved the same way they would have anyway. A separate questionnaire about living with the illness improved, peaking on day 10, which the authors attributed to mood and motivation. There was no comparison group, so there is no figure to set the questionnaire change against.

27 adults with motor neurone disease, 1% semax up the nose at 12 mg a day, in two 10-day courses two weeks apart, open label

One small study

Source [09]

The size of one part of the brain's resting network

The front part of the resting network came out larger in the 14 people given semax than in the 10 given a dummy spray. The paper reports no measurement of memory, attention, mood or anything else the volunteers did or felt.

24 healthy adults, 11 men and 13 women, average age 44, scanned before and 5 and 20 minutes after 1% semax up the nose or a dummy spray

One small study

Source [11]

Everyday function and a nerve-growth protein in the blood after a stroke

The authors report that the everyday-function score improved faster and ended higher in the people given semax than in the people in the same study who were not, and that BDNF in the blood rose and stayed up. The English abstract prints no actual figures for either, and reports no dummy spray.

110 adults after an ischaemic stroke, average age 58, split into early and late rehabilitation and then by whether or not they received semax, 6,000 micrograms a day for 10 days in two courses 20 days apart

One small study

Source [08]

Blood fats in people with psoriasis and metabolic syndrome

Total cholesterol, triglycerides and the harmful LDL cholesterol came down and the protective HDL went up in the group given semax on top of their usual treatment, while the group on usual treatment alone showed no statistically significant change in any of the four. The English abstract prints no numbers at all, only directions.

118 adults with psoriasis and metabolic syndrome: 60 given 0.1% semax up the nose for 10 days on top of conventional treatment, 58 given conventional treatment alone

One small study

Source [19]

Breast tumour size and lifespan in mice that already had tumours

Tumours measured 7.4 cubic centimetres at the end of life in the semax mice against 15.0 in the salt-water mice, and the semax mice lived about 68 per cent longer. The regulator's own warning is attached: this was one fixed amount, female mice only, and the treatment lasted about 74 days in an animal that lives around a year, which is far too short to say anything about cancer risk either way.

Female SHK mice with established breast tumours, 16 given 3 mg per kilogram into the abdomen five days a week and 27 given salt water, until natural death

Only seen in animals

Source [01]

The size of the damaged area after a stroke in rats

About 20 per cent smaller in the rats given semax than in the rats given water, eight days after the clot was made. The semax rats also kept a memory task that the water rats lost. Both groups moved around less than healthy rats.

Adult rats with a clot induced in the front of the brain, 250 micrograms per kilogram up the nose twice on the first day and then daily for six days, against water

Only seen in animals

Source [01]

There is no clinical safety data worth the name. One study in 37 adults reported that no side effects were seen; every other published human study of semax did not discuss side effects at all, and several gave it alongside other drugs. There is no human pharmacokinetic study by any route, no single-dose toxicity study, no repeat-dose toxicity study, no genotoxicity study, no study of effects on fertility or a developing pregnancy, no two-year cancer study, no immunogenicity study and no study of whether it is habit-forming. The one risk the FDA names on the basis of animal work is bleeding: semax reliably slows clotting in rats and pushed stressed rats past normal into the too-slow range, which matters most for anyone already prone to bleeding or taking a blood thinner — and a product bought online or made up by a pharmacy carries no label warning about it. The agency also could not rule out an immune reaction, because it had no information on what impurities the powders contain or whether the peptide clumps; the certificates of analysis it examined allowed a single unnamed impurity at up to 1 or 2 per cent and carried no result for bacterial contamination or bacterial toxins, both of which are critical for anything injected. Semax is a nickname rather than an official drug name and at least two different substances are sold under it, which the FDA called a safety risk in its own words. Its adverse-event database holds one report through 3 December 2025: a member of the public who reported eye pain and burning after Semax 0.1% nasal drops bought online, a hospital admission, and pain that had not settled a year later.

EffectWhat it feels likeHow often

Bleeding, or blood that clots too slowlySerious

In rats semax reliably slows clotting and shrinks experimental clots, and in stressed rats it pushed clotting past normal into the too-slow range. The US regulator raised this in 2026 as a concern about bleeding, and named the two situations that make it worse: people who are already prone to bleeding, and people taking other medicines that thin the blood. Its second point matters as much: a product made up by a pharmacy or bought from a website carries no label warning anyone about this. Nobody has established how much semax it would take to move clotting in a person, because no study has ever tested more than one amount in an animal.

Never measured in a person. No study of semax has taken a clotting test.

Source [01]

The body starting to attack the peptide, or the injection itselfSerious

Peptides can clump together in the vial, and clumps make the immune system more likely to react. The regulator wrote in 2026 that the result can range from antibodies with nothing noticeable happening to a life-threatening reaction, and that injecting or spraying into the nose makes this more likely than other routes. It could not rule the risk out for semax because it did not know what impurities the powders contain and had no information at all about clumping. Both certificates of analysis it examined allowed a single unnamed impurity at up to 1 or 2 per cent.

Never tested. No study of semax has looked for antibodies against it.

Source [01]

What is actually in a vial that is going up a nose or into a veinSerious

Anything injected has to be sterile and free of the bacterial fragments that cause fever. None of the certificates of analysis the regulator examined in 2026 carried a result for bacterial contamination or for those fragments, and it could not find one in the published literature either. It also flagged a naming problem: semax is a nickname, not an official drug name, and at least two different substances — the plain peptide and its acetate salt — are sold under it, which the regulator called a safety risk because someone can be given a different substance from the one that was ordered. For the nasal sprays there was no information about the spray device at all, which decides how much of a dose actually goes where.

Never measured for the products people buy.

Source [01]

Eye pain after nasal drops bought from a websiteSerious

A member of the public reported that in May 2024 she had eye pain and burning after using Semax 0.1% nasal drops bought online. She reported being taken to hospital, and reported a year later that the pain had not gone. One report proves nothing about cause, and almost nobody files a report about something bought from a website, so an almost-empty database means nobody is watching rather than that nothing happens. The only other record the same search returns is a different medicine entirely — a citalopram antidepressant sold under the brand name Semax — which is its own reason to be careful with a nickname.

One report in the whole of the US regulator's side-effect database, searched through 3 December 2025.

Source [01]Source [20]

What the published studies recorded

This is an absence of reporting rather than an absence of harm. None of the studies describes how side effects were looked for, over how long, or by whom. Several of them gave semax at the same time as other drugs, so anything that happened could not have been pinned on it anyway. The regulator's summary in 2026 was that there is insufficient clinical information to characterise the safety profile of semax at all.

One study in 37 adults reported that no side effects were seen. Every other published human study of semax did not discuss side effects at all.

Source [01]

Whether it is habit-forming

In mice semax made a dose of amphetamine release more dopamine in the reward part of the brain than the amphetamine did alone, and made the animals more active on it. Dopamine rising in that part of the brain is the signature of drugs people get hooked on, and the regulator said so in 2026. It then said that no study exists of whether semax itself is reinforcing or addictive. That is a question mark, not a warning — but it is a question mark on the exact mechanism people use to argue that a nootropic is safe.

Never studied, in any animal or any person.

Source [01]

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.

Kaplan et al. 1996 — the only published study of semax in healthy people that measured anything about thinking

1 mg a day up the nose, about 16 micrograms per kilogram of body weight, or a single 0.25 mg dose

2 days, or one dose

Source [01]

Koroleva et al. 1996 — the pain study, and the only one FDA read in full that measured a hard outcome

0.5 mg per kilogram of body weight, up the nose

One dose

Source [01]

Ivanikov et al. 2002 — stomach ulcers that had not healed, semax added on top of the standard medicines

1% solution, 2 to 4 drops in each nostril three times a day, against the same standard medicines without semax

10 days

Source [01]Source [10]

Lebedeva et al. 2018 and Panikratova et al. 2020 — the brain-scan studies in healthy volunteers, against a dummy spray

1.2 mg up the nose

One dose, scanned over the following 20 minutes

Source [01]

Gusev et al. 2018 — rehabilitation after a stroke, the largest adult study

6,000 micrograms a day, up the nose

10 days, twice, with 20 days in between

Source [08]

Serdiuk et al. 2007 — motor neurone disease, open label, no comparison group

1% solution up the nose, 12 mg a day

10 days, twice, with 2 weeks in between

Source [09]

Romanova et al. 2006 and Stavchansky et al. 2011, in rats, not people — the stroke experiments behind the neuroprotection claim

250 micrograms per kilogram of body weight up the nose, or 100 micrograms per kilogram into the abdomen, against water or salt water

Seven days in the first study, four doses over eight hours in the second

Source [01]

What people report

• Whether it does anything to memory, attention or focus in a healthy person — the use nearly everyone outside Russia buys it for has one 1996 study behind it and no controlled follow-up in thirty years.
• What happens to it in a human body: how much is absorbed, where it goes, how fast it is broken down. No study of any kind exists, by any route, in a person.
• What it does when injected under the skin, which is one of the two ways it is sold — every published human study of semax used the nose.
• Whether the capped version sold online as N-Acetyl Semax Amidate behaves like plain semax at all; it has never been given to a person or an animal in a published study.
• Whether it slows blood clotting in people the way it does in rats, and by how much — no human clotting measurement of any kind has been published.
• Whether it is safe: there is no single-dose toxicity study, no repeated-dose toxicity study, no genotoxicity study, no study of effects on fertility or a developing pregnancy, and no long-term cancer study.
• Whether it is habit-forming, which nobody has tested, even after mice given semax released more dopamine in response to a stimulant.
• Whether it provokes antibodies against itself, which has never been tested in any study.
• What is actually in the vials sold online and whether they are sterile — no certificate of analysis the US regulator examined tested for bacterial contamination or for the bacterial fragments that cause fever.
• Whether the two Russian stroke studies hold up: neither used a dummy spray, both come from the same Moscow group, and neither has been repeated by an independent team.
• What the Russian regulator's own approved label says about who must not take it — Russia's state medicines register is behind an image test that this entry could not get past.
• Whether it is banned in sport: it is named nowhere on the 2026 doping list, and the two sections that might catch it both arguably do not.
• What happens after two ten-day courses, the longest anyone has been given it in a published study.
• Whether it does anything for most of the fifteen conditions US clinics and online shops advertise it for — Parkinson's disease, Alzheimer's disease, opioid withdrawal, diabetic nerve pain, anxiety and nerve regeneration have never been studied with semax in a person at all.

Does Semax actually work?

For the things it is sold for online — focus, memory, mood — nobody has shown that it does. The only human study of thinking is one from 1996 in 19 healthy men, in a journal that is not indexed in the main medical database. Where semax has been put to a hard test in patients, it failed: it changed nothing at all in 16 people with facial nerve pain, and it left the headache in 8 of 12 people with migraine. Its home use, recovery after a stroke, rests on two Russian studies from one Moscow group, neither of which used a dummy spray.

Source [01]Source [08]

Is Semax legal?

It depends entirely on where you are. In Russia it is an ordinary registered medicine, sold as 0.1% and 1% nasal drops, and it was named on the government's list of vital and essential medicines for 2019. In the EU and the United States it is not an approved medicine at all, so selling it for people to take is not lawful, and in May 2026 the US regulator proposed that American pharmacies should not be allowed to make it up either. Buying it online for yourself is a different question from selling it, and it varies by country.

Source [01]Source [04]

What are the side effects of Semax?

Honestly, nobody knows. One study in 37 people reported that no side effects were seen; every other published human study of semax simply did not discuss the question. There is one report in the US regulator's side-effect database, from a woman who said she had eye pain and burning after using nasal drops bought online, was taken to hospital, and still had the pain a year later. The one risk the regulator names on the basis of animal work is bleeding, because semax slows blood clotting in rats.

Source [01]Source [20]

Semax vs Selank — what is the difference?

They are two different seven-amino-acid peptides developed at the same Moscow institute, and they end in the same three amino acids, Pro-Gly-Pro. Semax starts with a fragment of the stress hormone ACTH and is sold as a focus and stroke-recovery drug; selank has a different front end and is sold as a calming one. The only study that gave both to the same kind of volunteer compared what they did to brain scans in 52 healthy people, not to how anyone felt or performed. Neither is a component of any medicine approved in the United States, and semax is not approved in the EU either.

Source [12]Source [15]Source [21]

Is N-Acetyl Semax Amidate the same as Semax?

No. Plain semax has a bare amino acid at each end of the chain. The version sold online as N-Acetyl Semax Amidate has a cap added at both ends, which changes the molecule and is meant to make it last longer. There is no published study of that capped version in a person, or in an animal, for any condition — the only published work on an acetylated semax is laboratory chemistry about how it binds copper and zinc, plus a test on cells in a dish. Everything on this page describes plain semax.

Source [01]Source [22]

Is Semax banned in sport?

It is not settled, which is its own kind of answer. Semax is not named anywhere on the World Anti-Doping Agency's 2026 list. The catch-all section that bans anything unapproved does not obviously apply, because Russia has approved it, and the section covering ACTH-type hormones does not obviously apply either, because semax is repeatedly described as having none of the hormone effects. A 2025 review in Biology of Sport put it in a table of substances of unclear status, flagged for resembling one that is banned. Anyone competing should ask their anti-doping body rather than reading anything reassuring into this.

Source [05]Source [06]

Does Semax help ADHD?

There is no evidence that it does. Attention deficit disorder was one of the uses put to the US regulator, and the regulator dropped it from its evaluation because it could not find any published literature supporting it at all. The idea traces back to a 2007 article that was explicitly a hypothesis rather than a study. No trial of semax in anyone with the condition has ever been published or registered.

Source [01]Source [23]

Are there any Semax trials running now?

None are registered. The United States clinical trials register returns no studies at all for semax — not completed, not recruiting, not planned — whether it is searched as a treatment or as a general term, and the same is true for the chain written out as amino acids. For comparison, the same register returns 757 studies for semaglutide. The regulator searched the same register in 2026 and reported the same result.

Source [01]Source [07]

Each letter represents one amino acid.

1. [01]FDA briefing document, Pharmacy Compounding Advisory Committee, semax-related bulk drug substances (semax free base and semax acetate), dated 11 May 2026 for the meeting of 23–24 July 2026 — sequence H-Met-Glu-His-Phe-Pro-Gly-Pro-OH, Russian registration as 0.1% and 1% nasal drops, the whole readable clinical record, the exclusion of Russian-language references, and the recommendation not to add either substance to the 503A list (2026)
2. [02]PubChem: Semax (CID 9811102) — C37H51N9O10S, 813.9 g/mol, IUPAC name showing all seven residues in the L form and a free C-terminal carboxylic acid, so no amide cap
3. [03]UniProt P01189 (proopiomelanocortin, human) FASTA — the ACTH sequence SYSMEHFRWG…, showing Met-Glu-His-Phe at ACTH positions 4 to 7
4. [04]Government of the Russian Federation, order no. 2738-r of 10 December 2018, list of vital and essential medicines for 2019 — methionyl-glutamyl-histidyl-phenylalanyl-prolyl-glycyl-proline, nasal drops (in Russian; superseded by order no. 2406-r of 12 October 2019, whose list this register could not read) (2018)
5. [05]WADA: World Anti-Doping Code International Standard – Prohibited List 2026 — S0 non-approved substances and S2.2.2 corticotrophins; semax is not named anywhere in the document (2026)
6. [06]Pokrywka A, Surała O, Grabowska K, Przybyła M, Granda D, Małecki A, Faiss R, Nowacka-Chmielewska M. 'Brain doping' substances: prohibited or not in sports? Biol Sport 2025;42(4):189-201 — Table 6 lists semax among substances of unclear status, flagged against tetracosactide (S2) (2025)
7. [07]ClinicalTrials.gov API v2 search for semax: totalCount 0 as a general term and as an intervention, checked 3 August 2026; a control query for semaglutide returns 757 (2026)
8. [08]Gusev EI, Martynov MY, Kostenko EV, Petrova LV, Bobyreva SN. The efficacy of semax in the treatment of patients at different stages of ischemic stroke. Zh Nevrol Psikhiatr Im S S Korsakova 2018;118(3 Pt 2):61-68 (110 patients, no placebo group, in Russian with English abstract) (2018)
9. [09]Serdiuk AV, Levitskiĭ GN, Miasoedov NF, Skvortsova VI. The study of chronic partial denervation and quality of life in patients with motor neuron disease treated with semax. Zh Nevrol Psikhiatr Im S S Korsakova 2007;107(4):29-39 — open label, 27 patients, no effect on the nerve recordings or the clinical scores (2007)
10. [10]Ivanikov IO, Brekhova ME, Samonina GE, Myasoedov NF, Ashmarin IP. Therapy of peptic ulcer with semax peptide. Bull Exp Biol Med 2002;134(1):73-74 — 89.5% healed at day 14 on semax added to standard therapy against 30.8% on standard therapy alone (2002)
11. [11]Lebedeva IS, Panikratova YR, Sokolov OY, Kupriyanov DA, Rumshiskaya AD, et al. Effects of Semax on the default mode network of the brain. Bull Exp Biol Med 2018;165(5):653-656 — 14 on semax against 10 on a dummy spray, the only placebo-controlled human work (2018)
12. [12]Deigin VI, Poluektova EA, Beniashvili AG, Kozin SA, Poluektov YM. Development of peptide biopharmaceuticals in Russia. Pharmaceutics 2022;14(4):716 — semax as a heptapeptide synthesised at the Institute of Molecular Genetics of the Russian Academy of Sciences (2022)
13. [13]Koroleva MV, Meizerov EE, Nezavibat'ko VN, Kamenskii AA, Dubynin VA, Yakovlev YB. Analgesic action of the new drug Semax. Bull Exp Biol Med 1996;122:1107-1109 (1996)
14. [14]Gusev EI, Skvortsova VI, Miasoedov NF, Nezavibat'ko VN, Zhuravleva EIu, et al. Effectiveness of semax in acute period of hemispheric ischemic stroke. Zh Nevrol Psikhiatr Im S S Korsakova 1997;97(6):26-34 (30 treated against 80 matched controls, in Russian with English abstract) (1997)
15. [15]Panikratova YR, Lebedeva IS, Sokolov OY, Rumshiskaya AD, Kupriyanov DA, et al. Functional connectomic approach to studying Selank and Semax effects. Dokl Biol Sci 2020;490(1):9-11 (2020)
16. [16]Cherkasova KA, Lyapina LA, Ashmarin IP. Comparative study of modulatory effects of Semax and primary proline-containing peptides on hemostatic reactions. Bull Exp Biol Med 2001;132:625-626 (2001)
17. [17]Medvedeva EV, Dmitrieva VG, Povarova OV, Limborska SA, Skvortsova VI, Myasoedov NF, Dergunova LV. The peptide semax affects the expression of genes related to the immune and vascular systems in rat brain focal ischemia: genome-wide transcriptional analysis. BMC Genomics 2014;15:228 (2014)
18. [18]Inozemtseva LS, Yatsenko KA, Glazova NY, Kamensky AA, Myasoedov NF, et al. Antidepressant-like and antistress effects of the ACTH(4-10) synthetic analogs Semax and Melanotan II on male rats in a model of chronic unpredictable stress. Eur J Pharmacol 2024;984:177068 (2024)
19. [19]Dontsova EV. Possible drug correction of lipid metabolism disturbances associated with metabolic syndrome in patients with psoriasis. Eksp Klin Farmakol 2015;78(12):30-33 (in Russian with English abstract) (2015)
20. [20]openFDA drug adverse event API, query for medicinal product semax — two records: 25343150 (eye pain, consumer report, received 20 May 2025) and 20185724, in which the product named Semax is a citalopram medicine (2026)
21. [21]openFDA Drugs@FDA API, queries for substance name semax and for selank — no approved product for either (checked 3 August 2026) (2026)
22. [22]Magrì A, Tabbì G, Giuffrida A, Pappalardo G, Satriano C, Naletova I, Nicoletti VG, Attanasio F. Influence of the N-terminus acetylation of Semax, a synthetic analog of ACTH(4-10), on copper(II) and zinc(II) coordination and biological properties. J Inorg Biochem 2016;164:59-69 (metal-binding chemistry only) (2016)
23. [23]Tsai SJ. Semax, an analogue of adrenocorticotropin (4-10), is a potential agent for the treatment of attention-deficit hyperactivity disorder and Rett syndrome. Med Hypotheses 2007;68(5):1144-1146 (a hypothesis paper, not a study) (2007)