Semax: the complete research guide.

Among Russian-developed peptides, Semax has the deepest research record we've found. We read 13 published studies on it: stroke transcriptomics, BDNF and NGF effects, fMRI brain imaging, and independent Italian replication of the copper-binding mechanism.

If you were going to design the most-credentialed grey-market peptide possible, you'd probably end up with something close to Semax. The molecule is a structural analog of a known hormone fragment (ACTH 4-7). It's been approved in Russia for over 20 years for a serious indication (acute ischemic stroke). It has the most active and most consistent Russian research program of any peptide we cover. And, crucially, independent non-Russian academic groups have replicated key mechanistic findings without any authorship overlap.

That last point matters. The most common methodological critique of Russian-school peptides (BPC-157 has the same problem from its Zagreb lab) is that the literature is concentrated in one group, with limited outside replication. Semax has actually been replicated. Italian groups have confirmed its copper-binding and amyloid-β effects. That replication doesn't validate the full Russian clinical claim. But it does show that some of the mechanism survives when other labs run the experiments.

And yet: no Western Phase III, no PubMed-indexed Western RCT in any indication, no FDA or EMA approval. The same evidentiary wall as Selank. Here's what the published Western literature actually documents.

What is Semax, structurally?

Semax is a 7-residue peptide. The sequence is Met-Glu-His-Phe-Pro-Gly-Pro. The first 4 residues match a section of ACTH (adrenocorticotropic hormone), specifically the 4–7 fragment that has neuro-active properties on its own. The C-terminal Pro-Gly-Pro tail is the same trick used in Selank. It makes the molecule resistant to enzymatic breakdown, while keeping the active core's behavior intact.

Here's the structural twist that matters. Semax is described as "non-corticotropic". Despite its ACTH lineage, it doesn't release cortisol the way full-length ACTH does. The 4–7 sequence carries the brain-active part of ACTH without the steroid-releasing part. That's the basis for the "nootropic without HPA-axis activation" pitch that the Russian clinical use rests on.

The development team (N.F. Myasoedov and L.A. Andreeva, at what's now the Kurchatov Institute) is the same group that made Selank. The two molecules are commercial and academic siblings, designed by the same Russian research complex. Their evidentiary structure is parallel. Selank is the anxiety drug. Semax is the nootropic and stroke-recovery drug.

What does Semax actually do?

The mechanism picture for Semax is unusually well-developed for a Russian-school peptide. The Kurchatov Institute group has published a series of studies over the past 5 years that map the molecule's effects at three levels: gene expression, protein expression, and behavior. The list of effects includes BDNF and NGF upregulation, anti-inflammatory cytokine changes in stroke models, modulation of CREB/MMP-9/c-Fos/JNK signaling, GABA receptor allosteric modulation, copper binding, and dopamine and serotonin effects.

BDNF and NGF are the two best-studied growth factors for nerve cells (BDNF = brain-derived neurotrophic factor; NGF = nerve growth factor). Both are central to learning and memory. Compounds that raise their levels have been the focus of cognitive-enhancement research for decades. Semax raises both in rodents. That's the proposed substrate for its cognitive effects.

The Kurchatov stroke transcriptomics work is the deepest single part of the evidence base. Filippenkov and colleagues published a 2024 study showing Semax restored normal expression of 1,171 genes that ischemia had disrupted, measured 24 hours after a rat stroke model (transient middle cerebral artery occlusion). Sudarkina's 2021 paper documented Semax's effects on four key proteins (CREB, MMP-9, c-Fos, JNK) at the same 24-hour time point. Filippenkov's 2023 paper added immune-gene data from the early post-stroke window. And Dergunova's Russian-language 2021 paper showed Semax suppresses pro-inflammatory signals (IL-1α, IL-1β, IL-6, CCL3, CXCL2) triggered by stroke.

This is unusually rigorous preclinical work. Multi-time-point RNA-seq studies in a standardized stroke model, with consistent methodology across papers, isn't what a typical "research peptide" evidence base looks like.

Semax

ACTH(4-10) analog 7 aa N-acetylated

The same compound cited across the 13 studies in this review. Lab-verified identity and purity.

View Semax

What does independent replication look like?

The Italian copper work is the most important non-Russian piece of the Semax evidence base. Sciacca and colleagues at the University of Catania published a 2022 study showing Semax binds copper ions (Cu²⁺), blocks Cu(II)-Aβ complex formation, reduces copper-driven reactive oxygen species production, and protects neuroblastoma cells from oxidative stress. Tomasello and colleagues from the same Italian environment extended the finding in 2025.

Why this matters: copper imbalance is one of several proposed contributors to Alzheimer's disease pathology. It interacts with amyloid-β aggregation and oxidative-stress generation. A compound that binds Cu(II), prevents Cu-Aβ complexes, and lowers oxidative stress has a plausible mechanistic relationship to Alzheimer's. Not as a treatment. As a research tool with a defined biochemical effect.

For our purposes, what's important is who did this work. The Italian group has no authorial overlap with the Russian developers. They work in a Western academic institution. Their methods are published and can be independently checked. The finding was replicated by an overlapping but distinct Italian team in a follow-up paper in 2025. That's exactly the kind of corroboration that the broader Russian-school peptide literature usually lacks.

What does the 2020 fMRI study show?

The same 2020 Panikratova study that gives Selank its best Western human evidence also covered Semax. The placebo-controlled three-way design (Selank vs Semax vs placebo) enrolled 52 healthy adults. Brain scans were taken before, 5 minutes after, and 20 minutes after intranasal dosing (Panikratova et al., 2020).

For Semax specifically, the team documented changes in the right amygdala's connectivity to several right temporal cortical regions (fusiform, inferior temporal, middle temporal, parahippocampal gyri). It's broadly the same fear-and-emotion network engaged by Selank, with some quantitative differences. The interpretation is the same: real, measurable CNS engagement in humans within a short timeframe. It isn't the same as proof of clinical efficacy in patients with stroke or cognitive impairment.

For a healthy-volunteer biomarker study, the methodology is solid (placebo control, three arms, repeated measures, MRI). For a Phase III efficacy substitute, it isn't what it would need to be. Both things can be true.

What about the other preclinical findings?

Beyond the stroke transcriptomics and the Italian copper work, the recent Semax literature has a handful of other findings worth knowing.

Spinal cord injury (2025). Liu and colleagues published in the British Journal of Pharmacology: Semax improved recovery in a mouse spinal cord injury model by suppressing a pro-death cell-signaling pathway (pyroptosis) through μ-opioid receptor engagement. This is a novel mechanism that implicates the opioid receptor system. BJP is a major peer-reviewed pharmacology journal. That publication venue alone is a meaningful evidentiary signal.

Antidepressant-like effects (2024). Inozemtseva and the Kurchatov team showed Semax at 60 nmol/kg/day reversed depression-like changes in chronically stressed rats: anhedonia, body weight loss, adrenal gland enlargement, and dropped hippocampal BDNF.

Neonatal SSRI exposure (2020). Glazova and colleagues reported that Semax given to rats neonatally exposed to an SSRI reduced anxiety behaviors, improved learning, and normalized brain amine levels.

Diabetes lipid effects (2020). Elagina and colleagues documented Semax at 200 mcg/kg correcting lipid metabolism problems in a rat diabetes model. It reduced total cholesterol, triglycerides, LDL, and atherogenicity. It raised HDL.

Restraint stress and the gut (2021). Svishcheva and colleagues showed Semax at 5–450 mcg/kg reduced corticosterone, eased stress-driven colon damage, and helped rats adapt under restraint. The pattern parallels the Selank gut-protection data.

GABA receptor modulation (2023). Vyunova and colleagues characterized Semax's direct and delayed effects on the GABA receptor system, adding mechanism detail to its anxiety-adjacent activity.

Does delivery route matter? (Nasal vs subcutaneous)

Semax is approved in Russia as a nasal-spray formulation. Intranasal delivery is the primary administration route in both Russian clinical practice and Russian preclinical work. The nasal route has two practical advantages. It bypasses first-pass metabolism. And some of the dose reaches the brain directly via the olfactory tract and trigeminal nerve, sidestepping the blood-brain-barrier problem that hampers most peptide drugs.

Western grey-market use is mostly subcutaneous injection. That's the standard route for research peptides. The injection route presumably produces a different drug behavior in the body: higher peak blood levels, lower direct brain delivery, longer half-life from slower absorption. Whether subcutaneous Semax produces the same effects as intranasal Semax is unanswered by any published study.

The practical implication for researchers: subcutaneous administration, the most common route in Western preclinical work, represents a delivery route that was not validated by the Russian clinical trials underpinning the molecule's approval. The mechanism is the same. The pharmacokinetic profile is fundamentally different.

Why isn't there a Western RCT?

The same structural answer as Selank applies, with one twist. Russian institutions hold the intellectual property. No Western sponsor has a commercial incentive. The competitive landscape for stroke neuroprotection is dominated by tPA and thrombectomy, with limited room for an add-on peptide. And Phase III failure rates in CNS drug development are high enough to deter even well-evidenced compounds.

A secondary factor is the research context most frequently studied in preclinical literature. The largest body of mechanistic work covers ischemic stroke and specific cognitive deficit models. Studies in neurologically intact adult subjects are limited to the 2020 Panikratova fMRI biomarker study. The mechanistic work on BDNF, NGF, copper chelation, and stroke gene expression is substantial. It does not constitute validated efficacy data for cognitive enhancement in healthy study participants.

Semax

11 mg ≥99% pure Lyophilized

ACTH(4-10) analog · 7 aa, N-acetylated. The same reference compound used across the cited preclinical studies. COA available with each lot.

Learn more

What about side effects?

Published Russian clinical literature describes Semax as well-tolerated at nasal-spray doses used in the approved short-course indications. The most commonly noted issue in clinical reports is mild nasal irritation. Systemic adverse events are described as uncommon in Russian clinical literature. Because Semax is "non-corticotropic," it does not trigger cortisol release despite its ACTH lineage — HPA-axis activation has not been documented in published studies.

Limitations in the safety evidence base are notable. Russian approval covers short-course use (days to weeks in stroke or specific cognitive deficit contexts). Long-term safety data are essentially absent in the published literature. Drug interactions with neuroactive agents (SSRIs, opioids, GABA-active compounds) are theoretically plausible given Semax's broad receptor-level effects documented in preclinical studies, but formal pharmacokinetic interaction studies have not been published. The 2025 Liu et al. finding that Semax engages the μ-opioid receptor axis in spinal cord injury introduces a specific open question regarding co-administration with opioid compounds that warrants investigation.

Open research questions

Several questions remain unresolved in the published literature. For cognitive endpoints in neurologically intact study participants, the mechanistic evidence is suggestive but clinical efficacy has not been established in a Western randomized controlled trial. For post-stroke cognitive recovery and specific deficit contexts, Russian clinical experience exists but has not been replicated to Western Phase III standards.

• Kurchatov transcriptomics: multi-time-point RNA-seq data in a standardized stroke model represents the deepest mechanistic work published for any Russian-school peptide.
• Delivery route equivalence: the validated Russian route is intranasal. Subcutaneous administration, used in most Western preclinical protocols, has not been compared head-to-head with intranasal in a published pharmacokinetic study.
• Interaction studies: the μ-opioid receptor engagement documented by Liu et al. (2025) raises a specific open question for co-administration protocols with opioid compounds — not yet characterized in a published interaction study.
• Independent replication: the Italian copper / Aβ work and the Liu et al. spinal cord injury finding represent the most credible non-Russian replications; further independent replication of the full Kurchatov stroke transcriptome claims would strengthen the evidentiary base.
• Western RCT: no PubMed-indexed Phase II or Phase III trial has been registered or published for any Western indication as of 2026.