Semax: what the stroke studies found, and which ACTH fragment it really is

Semax is a seven-amino-acid peptide made in Moscow in the 1980s from a fragment of the stress hormone ACTH, and it is sold in Russia as a prescription nasal medicine for stroke and other brain conditions [1]. Its human record runs from 1997 to 2018 and points the same way each time — faster motor recovery, higher blood BDNF, better independence scores — sitting on top of the most consistent neurotrophic animal data of any peptide we cover; the honest limit is that every human study came from the developers’ own group and none used a placebo.

What semax is, and the ACTH(4-7) versus ACTH(4-10) muddle

Adrenocorticotropic hormone (ACTH) is a 39-amino-acid pituitary hormone. Short pieces of its front end, particularly positions 4 to 10, were known to affect learning and attention in animals, and the Moscow team set out to build a longer-lasting version of that fragment [1]. Positions 4 to 10 of ACTH read Met-Glu-His-Phe-Arg-Trp-Gly.

Semax keeps the first four of those (Met-Glu-His-Phe, which is ACTH 4 to 7) and swaps the last three for Pro-Gly-Pro. The finished molecule is Met-Glu-His-Phe-Pro-Gly-Pro [2]. The Pro-Gly-Pro tail was added to slow down breakdown by enzymes, so the peptide lasts longer after a nasal dose.

That explains why you’ll see two descriptions in the literature, often from the same Moscow institute:

Both are correct. Semax contains ACTH(4-7) exactly and is a modified analogue of ACTH(4-10). What it is not is the full ACTH(4-10) sequence, so studies on the natural fragment don’t automatically apply to it. Because it lacks the later part of ACTH that acts on the adrenal glands, it isn’t expected to raise cortisol, though we found no human study that measured this.

As for mechanism, the leading theory comes from rat and cell studies: semax raises levels of brain-derived neurotrophic factor (BDNF), a protein involved in the growth and survival of nerve cells. In rat glial cells in culture, BDNF messenger RNA rose eight-fold and nerve growth factor messenger RNA five-fold, peaking 30 minutes after treatment [7], and in live rats a nasal dose raised BDNF protein in the basal forebrain after 3 hours [2]. The same rat study found specific binding sites for semax in brain membranes, though the receptor itself hasn’t been identified [2].

The human evidence

A 1997 overview by the development team says semax improved memory and attention in healthy men “under extreme conditions of activities” and was used for a range of brain disorders without side effects [1]. That overview doesn’t give the underlying study designs, so we can’t assess those claims. The studies we could examine are about stroke and chronic reduced blood flow to the brain.

The 1997 stroke study is the one most often cited [4]. It had a control group, but those 80 patients are described only as having strokes of similar severity and location, with no mention of random assignment or blinding, and the abstract says semax had “some influence” on the rate of recovery. That’s modest language from the authors themselves.

A 2005 study followed 187 people with what Russian neurology calls chronic cerebrovascular insufficiency, a diagnosis without a close equivalent in UK or US practice [5]. The authors report clinical improvement and fewer strokes and transient ischaemic attacks with semax. The abstract gives no event numbers and no description of a control arm, so the stroke-reduction claim can’t be checked from what’s published.

The 2018 study is the most recent [6]. It enrolled 110 people after ischaemic stroke, split them by when rehabilitation started (around 3 months or around 7 months after the stroke), and then into semax and no-semax subgroups. The semax regimen was two 10-day courses of 6,000 micrograms a day, 20 days apart. Blood BDNF went up in the semax groups and stayed high, and the authors report faster, better recovery on the Barthel index, a scale of everyday independence [6]. Again, there was no placebo, and blood BDNF is not a reliable window into brain BDNF.

The one placebo-controlled human study we found measured brain wiring rather than any clinical outcome. In 52 healthy volunteers, resting-state MRI showed that semax and selank both altered connectivity between the right amygdala and part of the temporal lobe within 20 minutes [10]. It says nothing about memory, focus or recovery.

Semax is promoted online mainly as a cognitive enhancer, yet cognitive enhancement in healthy people is the one use with no human outcome data behind it at all.

Doses and routes in the studies

The doses in the clinical papers vary a lot, and none of them was set by a formal dose-finding trial of the kind regulators expect. The 1997 overview describes nasal doses of 0.015 to 0.050 mg per kg of body weight, with effects said to last 20 to 24 hours [1]. The 1997 stroke study tried several daily doses and settled on 12 mg for moderate and 18 mg for severe strokes, given for 5 or 10 days [4]. The 2018 rehabilitation study used 6,000 micrograms (6 mg) a day in two 10-day courses [6]. So the stroke regimens were several times higher than the doses discussed for memory and attention, and there’s no published work on how blood or brain levels relate to either.

We report these figures so you can see what was tested, not as guidance. None of these regimens has been checked against a placebo.

What the animal and lab work suggests

The rat research is the strongest part of the semax literature. In rats with experimentally blocked brain arteries, semax changed the activity of a large set of genes, more than half of them tied to the immune response, plus a smaller group linked to blood-vessel growth [8]. It also switched on genes for nerve growth factors and their receptors in the damaged cortex [9]. A 2017 study in the same model found semax turned up immune signalling pathways further, while its Pro-Gly-Pro tail on its own did the opposite [3]. Whether that immune activity helps or harms recovery is unclear.

These are careful molecular studies, and several were published in international journals. Gene expression changes are an early step, and the animal work has not yet been extended to long-term functional recovery.

Safety and side effects reported

The Russian studies describe semax as well tolerated, with the 2005 paper reporting a “minor percent” of side effects [5] and the 1997 overview saying no complications were seen [1]. Those statements come from small studies without placebo groups and without systematic adverse-event reporting of the kind a regulator would ask for. We found no independent safety data, no long-term data, and nothing on pregnancy or children.

The BDNF mechanism also raises an unanswered question. BDNF supports nerve cells, but growth-factor signalling isn’t automatically harmless when pushed for long periods, and nobody has studied that for semax.

Legal and regulatory status

Semax is used as a prescription nasal medicine in Russia. It is not approved as a medicine by the UK’s MHRA, the European Medicines Agency or the US FDA. Outside Russia and a few neighbouring countries it is sold only as a research chemical, without the manufacturing and quality checks that apply to medicines.

What is still open

• Whether semax improves stroke recovery compared with a placebo in a randomised, blinded trial.
• Whether it does anything for memory or attention in healthy people.
• Whether nasal semax meaningfully raises BDNF in the human brain, rather than in the blood.
• What its receptor is.
• Long-term safety of repeated use.
• Whether groups without links to its developers get the same results.

What this means

Semax is the most mis-sold peptide we cover. It is not a research chemical with a shaky story; it is a licensed Russian medicine with a defined structure, a plausible mechanism and two decades of clinical papers pointing in the same direction. What it is not is a cognitive enhancer for healthy people, which is how almost everyone outside Russia meets it. The one placebo-controlled human study we found measured amygdala connectivity on a scanner, not memory or focus, and no trial anywhere tests semax against placebo for attention, learning or recall in healthy adults. Buy it as a nootropic and you are paying for a drug whose only human outcome data comes from stroke wards.

The reason the definitive stroke trial has never happened is not that semax is unownable. It is ownable, and has been owned for forty years by its Russian developers and the institutes around them. A placebo-controlled multicentre trial is expensive, and the only party with the commercial interest in running one already sells the drug in its home market, where the health service funded small single-centre studies instead. A Western sponsor will not pay for a registration-grade trial of a compound it does not control and cannot sell. That is the gap, and it is commercial rather than scientific.

So the honest direction is narrow but real. Semax belongs in ischaemic brain injury as an adjunct worth testing properly, and it does not yet belong in anyone’s daily stack. If you are weighing it, the useful question is not whether the rat BDNF data are real — they are unusually consistent — but whether a nasal peptide that raises growth-factor signalling for years has been checked for long-term safety in people. It has not. Until someone funds the trial that answers both questions, the licence it holds in Russia is the strongest thing that can be said for it, and the absence of that trial anywhere else is the strongest thing against.

Educational content only — not medical advice. Many peptides discussed on HPR are not approved for human use. Talk to a qualified clinician before making any decision about your health.