Semax Guide: Evidence & Sourcing

Semax at a glance

What it is

Semax, the synthetic heptapeptide Met-Glu-His-Phe-Pro-Gly-Pro — an ACTH(4-7) fragment extended with a Pro-Gly-Pro tail that slows enzymatic breakdown. It carries no corticotropic activity.

Researched for

Ischaemic stroke and cognitive/attention research, almost entirely within a Russian research tradition; the English-indexed literature is dominated by rat brain-ischaemia gene-expression work.

Mechanism

An ACTH fragment lacking corticotropic activity; associated in animal work with BDNF/neurotrophin signalling and with broad changes in ischaemia-related gene expression.

Plasma half-life

Minutes. The Pro-Gly-Pro tail slows degradation relative to the bare ACTH fragment but does not make it long-lived.

Regulatory status

Registered as a medicine in Russia; not FDA-approved and not approved anywhere in the EU or US. Research/educational use only.

Educational reference. This page does not publish dose ranges or reconstitution arithmetic.

What it is / mechanism

Semax is a synthetic analogue of the ACTH(4-10) region — specifically the ACTH(4-7) fragment Met-Glu-His-Phe, extended with Pro-Gly-Pro. The tail is the design feature: the bare fragment is degraded almost immediately by plasma and tissue peptidases, and the added tripeptide slows that enough for the molecule to be administered practically. It was developed at the Institute of Molecular Genetics of the Russian Academy of Sciences, and the whole compound is best understood as a stabilised piece of a hormone rather than a novel molecule. The important pharmacological point is what it does not do. ACTH's defining action is stimulating cortisol release from the adrenal cortex, and that activity lives in a different part of the molecule. The 4-7 fragment retains the behavioural and neurotrophic activity historically attributed to ACTH fragments while carrying no corticotropic effect, which is precisely why the fragment was chosen. Semax is not a corticosteroid, does not raise cortisol as its mode of action, and should not be reasoned about as though it were an ACTH substitute. What it does at a molecular level is less cleanly established than the marketing suggests. The most frequently cited association is with brain-derived neurotrophic factor and its receptor TrkB, with animal work reporting increased BDNF expression in relevant brain regions after administration. Beyond that, the strongest and most reproducible body of evidence is transcriptomic rather than receptor-level: in rat models of focal cerebral ischaemia, Semax produces broad, measurable changes in gene expression — Medvedeva et al. (BMC Genomics, 2014; PMID 24661604) reported genome-wide transcriptional changes affecting immune and vascular system genes, and Dergunova et al. (2021; PMID 34097675) reported suppression of transcripts encoding proinflammatory mediators after reversible brain ischaemia. Sudarkina et al. (2021; PMID 34201112) reported a corresponding protein-expression profile. That is a real mechanistic literature, and it is worth reading it for what it is: consistent evidence that the compound changes ischaemia-related gene expression in rat brain. It is not a demonstration of a receptor, and it is not a human efficacy finding. A separate strand of interest concerns copper — Semax has been characterised as a copper-chelating peptide affecting copper-catalysed reactive oxygen species production in amyloid model systems — which is chemistry in artificial membrane models rather than a clinical mechanism.

Researched effects

The reported effects and the indexed evidence for them are two different things, and separating them is the most useful thing this section can do. What is reported: in Russian clinical and preclinical research Semax has been associated with improvements in attention, memory consolidation and mental fatigue, and it is registered in Russia for use in ischaemic stroke and transient ischaemic attack, as well as in ophthalmological indications involving the optic nerve. Those registrations are genuine — a national regulator did approve the product — and they rest on clinical work conducted largely in Russian-language journals over several decades. What is indexed in English: overwhelmingly animal work. A PubMed search for Semax returns rat brain-ischaemia transcriptomics, Alzheimer's-model animal studies, artificial-membrane copper chemistry, and a 2025 mouse spinal-cord-injury study reporting an effect on the mu opioid receptor gene Oprm1. Human work is sparse; Panikratova et al. (Doklady Biological Sciences, 2020; PMID 32342318) applied a functional connectomic approach to Selank and Semax effects, which is a human neuroimaging study but not an efficacy trial. The gap between those two paragraphs is the honest summary of Semax's evidence base. It is not that the clinical research does not exist; it is that most of it is not accessible to, or indexed by, the databases an English-speaking reader can check, and that the portion which is indexed is preclinical. A reader outside Russia is largely being asked to take the clinical case on trust. That is a materially different position from a compound whose trials can be looked up, and it is a reasonable thing to weigh — in either direction — rather than something to be quietly omitted. One further asymmetry is worth naming. The nootropic and mental-fatigue claims that dominate discussion of Semax outside Russia are not what the registered product is registered for. The registrations concern stroke and optic-nerve indications. Attention and cognition in healthy people is the use most people are actually interested in, and it is the use with the least behind it.

Research timeline

Dated milestones taken from the primary records: the molecule's discovery, its first registered human study, the trials that shaped the evidence and the regulatory decisions. Each entry links to the record it comes from.

1. 1997Semax's Russian developers review 15 years of work on the peptide, a modified fragment (4-10) of the hormone ACTH, and state that it is already used to treat patients with diseases of the central nervous system. PMID 9173745
2. 1997A Russian clinical study gives Semax to 30 patients in the acute phase of ischaemic stroke and compares them with 80 patients on standard treatment, not with randomised controls, reporting some effect on recovery of neurological, especially motor, function. PMID 11517472
3. Apr 2006A rat study in the Journal of Neurochemistry reports specific Semax binding sites in the basal forebrain and higher levels of the growth factor BDNF there after Semax was given into the nose. PMID 16635254
4. Apr 22, 2026FDA's page on compounding safety risks, as updated on this date, lists Semax among substances that had been in Category 2 and whose nominations were withdrawn, citing possible immunogenicity and impurities and no, or limited, safety information for the proposed routes. FDA · compounding safety risks, Apr 22, 2026
5. Jul 24, 2026Semax goes before FDA's Pharmacy Compounding Advisory Committee for the pharmacy-compounding (503A) list, evaluated for cerebral ischaemia, migraine and trigeminal neuralgia; FDA's briefing document concludes that the balance of criteria weighs against listing it. No minutes or vote were posted by 4 Oct 2026. FDA · PCAC briefing document, Jul 23–24, 2026

Records checked Oct 4, 2026

Claims at a glance

The claims most often made about this compound, each with the strongest kind of evidence we found for it. The grade says what was studied, not whether the claim is true; the line beside it says what that study showed.

• Improves focus and memory in healthy peopleAnimal studies only. Learning effects come from rodent work; no retrievable human report contains a cognitive test result. PMID 9173745
• Raises BDNF in the brainAnimal studies only. Measured in the brain only in rats, where it rose in some regions and at some time points; in people only blood levels have been measured. PMID 16635254
• Aids recovery after strokeHuman, non-randomised. A study of 30 patients in acute stroke, compared with 80 on standard treatment rather than randomised controls, reported some recovery of motor function. PMID 11517472
• Helps recovery from brain injuryHuman, non-randomised. In 40 patients with moderate head injury a blood marker (Bcl-2) rose, but the study did not link it to better function. PMID 24049550
• Reduces anxietyAnimal studies only. Animal results point in two directions; the one placebo-controlled human study imaged the brain and did not measure anxiety. PMID 32342318

The reasoning behind each grade, study by study, is in this guide's Benefit & Outcome Review.

Grades checked Oct 5, 2026

Half-life & pharmacokinetics

Semax is cleared from plasma very rapidly, on the order of minutes, despite the Pro-Gly-Pro stabilisation that protects it relative to the unmodified ACTH(4-7) fragment. The stabilising tail buys enough time for the molecule to be administered practically; it does not produce a long-lived compound, and no reading of the pharmacokinetics supports the idea of a sustained blood level between administrations. That creates the same interpretive problem as several short-lived peptides: the reported effects, whatever their size, are described as outlasting the compound's presence in circulation by a wide margin. The usual explanation offered is that the effect is downstream — a change in gene expression or neurotrophin signalling initiated during a brief exposure and persisting afterwards — and in Semax's case that explanation has more support than most, since the strongest preclinical evidence is precisely transcriptomic. A measurable change in gene expression in rat brain after ischaemia is exactly the kind of finding that could outlast a peptide cleared in minutes. It is still an inference rather than a demonstrated human mechanism, and it is worth holding at that level of confidence. The short half-life is also part of why the intranasal route matters in the original research: direct nose-to-brain delivery is a recurring rationale in this literature, whether or not it is fully established.

Storage & handling

• Lyophilized (unreconstituted) vials are typically stored refrigerated and protected from light; the sealed powder is generally described as stable for extended periods, and long-term storage is usually at -20 C.
• Reconstituted solution: refrigerate at ~2-8 C, protect from light, and do not freeze once mixed; commonly reported usable window is ~2-4 weeks.
• The registered Russian product is a ready-made intranasal solution, so its pack storage conditions are a label matter; the notes above concern lyophilised research material.

Evidence File

The Semax Evidence File: The Registry Blank, the Dev-Name Trial, and the Dose Nobody Studied

The free guide says the clinical work behind Semax is not indexed in English. That is half the story, and the smaller half. This file separates two very different absences, recovers a randomised human trial that no search for the word Semax will ever return, finds the single registry record that the same naming trick hides, reads each of the four indexed Russian trials for what it actually reports rather than what it is cited for, and traces the circulating dose, half-life and neurotrophin figures back to the papers that generated them. It closes by naming, search string by search string, what the literature does and does not contain.

• 1 registry record worldwide, under its development name only, and it enrolled rats
• 40 patients in the only randomised human trial, which a Semax search never returns
• 10x to 60x: how far every human trial dose sits above the circulating community figure
• 89.5% vs 30.8%: the only clean responder effect size anywhere in the human record

Sourcing File

Sourcing Semax: What the Sequence MEHFPGP Decides About Every Certificate

Seven residues, and four of them settle questions a generic certificate-reading guide cannot reach. This file works the Semax numbers rather than describing them: the verified identity record and the one field vendors reuse as a regulatory claim, a computed mass table covering every plausible substitute and degradation product, the arithmetic trap that lets three different correct numbers all look like the molecular weight, and salt-form maths carried all the way through to the micrograms of peptide actually in the vial. It closes with a line-by-line read of a certificate and the two questions any Semax document has to answer.

• 813.348 neutral, 813.9 average, 814.355 protonated: three correct molecular weights
• +15.995 Da, the one-methionine oxidation peak a purity percentage hides
• 10,000 vs 8,771 vs 7,811 mcg in the same 10 mg vial, on label ambiguity alone
• 0 usable absorbance at 280 nm: the sequence has no tryptophan and no tyrosine

Benefit & Outcome Review

Benefit and Outcome Review: Semax

Fourteen claims circulate for this compound across research summaries and community protocols, and they were not selected by the strength of the evidence behind them. Each is graded here on the same four questions: what was measured, in what species and system, against what comparator, and whether any human function improved rather than merely a number changing. Responder splits and time courses are given where they were published and named as missing where they were not. The verdict vocabulary is defined first, so a reader can disagree with a grade rather than with a summary of it.

• 14 circulating claims graded on the same four questions
• 2 placebo-controlled human studies exist, both imaging, neither testing performance
• 0 retrievable human reports contain a cognitive test result
• Post-course durability has never been measured for the peptide on its own

Guide FAQ

Quick answers about guide scope, access, and educational use context.