No published Semax dose has been established as standard. The number that circulates most, 18 milligrams a day for severe stroke, comes from a single 1997 Russian trial that was not described as randomized or placebo-controlled.[1] That distinction matters more than the milligram figure does. A reader hunting for “the right dose” is asking a question this literature was never built to answer, and dosing guides that hand out a number without saying so are borrowing a precision the source study doesn’t have.
The numbers behind the dosing claims
One study accounts for most of the specific dosing figures attached to Semax online. In it, researchers added intranasal Semax to standard therapy in patients with acute hemispheric ischemic stroke and tracked recovery against a separate group treated with conventional therapy alone.[1]
The doses used, by stroke severity:
Moderate-severity stroke: 12 mg per day, given as a 5-day course.
Severe stroke: 18 mg per day, given as a 10-day course.
18 mg daily dose used for severe stroke in the source trial behind most Semax dosing claims PMID 11517472
Both regimens were intranasal, added on top of conventional stroke therapy. Neither was tested against a placebo arm. The 30 patients who received Semax were compared with 80 patients who didn’t, and the write-up reports faster regression of neurological deficits in the Semax group.[1] That’s a real finding, though nothing in the published record describes it as a randomized, blinded trial.
What the record doesn’t say: the study most often cited for Semax’s stroke dosing never describes randomized allocation or a placebo arm, only a controlled add-on comparison against conventional care.
Why the missing word matters
An open-label add-on design can still show something real. It just can’t rule out the ordinary confounders that randomization exists to handle: clinicians deciding who gets the extra treatment, patients who received it also getting more attention, or the two groups differing in ways the write-up doesn’t report. None of that means the 12 and 18 milligram figures are wrong. It means calling them “the dose” overstates what a single, decades-old, non-blinded comparison can support.
This is worth saying plainly because so much online material treats Semax dosing as settled, when the number behind it comes from one study, of a design that wouldn’t clear a modern journal’s bar for a causal claim.
What Semax is
Semax is a synthetic heptapeptide, Met-Glu-His-Phe-Pro-Gly-Pro, built from the N-terminal fragment of adrenocorticotropic hormone, ACTH(4-10).[2] It is one of several short synthetic neuropeptides, including kortagen and pinealon, used clinically in Russia as neuroprotective agents, with reported use in older adults, distinct from larger polypeptide-complex drugs such as cortexin and cerebrolysin.[3]
Structurally, that ACTH lineage matters. Semax is described in the literature as a noncorticotropic analog of the ACTH(4-10) fragment, meaning it lacks the adrenal-stimulating hormonal activity of full-length ACTH.[4] Whatever Semax is doing in the brain, it isn’t acting as a stress hormone.
The mechanism: solid in rats, unconfirmed in people
The best-supported biological claim about Semax comes from rodents; nobody has confirmed it in humans. A single intranasal dose of Semax (50 micrograms per kilogram of body weight) raised BDNF protein levels in the rat hippocampus by up to 1.4-fold and BDNF mRNA by up to 3-fold, alongside increased trkB receptor phosphorylation.[2] Brain-derived neurotrophic factor supports neuron survival and synaptic plasticity, which is the mechanistic story behind Semax’s reputation as a cognitive peptide.
A broader 2025 review of neuroprotective peptides groups Semax with BDNF, GDNF and exendin-4 among compounds whose preclinical effects include lowering oxidative stress and neuroinflammation.[5] That review frames these as preclinical, early-phase findings across the peptide class generally, not confirmed mechanisms in Semax-treated humans specifically. The gap between “does this in a rat’s hippocampus” and “does this in a person’s brain” hasn’t been closed.
The BDNF signal is real. It has only been measured in rats.
What nobody has published
Some marketing copy for Semax claims it has been tested in children with attention-deficit/hyperactivity disorder. That specific claim doesn’t hold up: no indexed study links Semax to a pediatric ADHD trial. It may exist somewhere outside the indexed literature, but it isn’t sitting in the record this article’s sourcing could locate, so it isn’t included here.
Adverse-event data specific to Semax is similarly missing from what a literature search turns up. Absence of reported harm in a handful of short trials is not the same as a demonstrated safety record, and the two shouldn’t be confused.
Long-term safety is unstudied. The published Semax trials run for short courses, from days to a couple of weeks. None extends to months or years, so nothing establishes what chronic or repeated use does.
The ACTH connection is untested at other doses. Because Semax derives from an ACTH fragment, its effect on the hypothalamic-pituitary-adrenal axis outside the specific doses and durations already trialed has not been characterized.
No pharmacokinetic study sets a human standard. Nothing in the literature establishes a bioavailability figure or half-life that applies to the formulations sold outside a trial setting.
Bottom line on safety: absence of reported harm in a handful of short Russian trials is not the same as a demonstrated safety profile, and nobody should read it that way.
Retail Semax versus trial Semax
The nasal-spray products sold under the Semax name aren’t manufactured or dosed to the specifications used in the clinical studies above. A trial that measures 12 or 18 milligrams delivered under controlled conditions, with a defined concentration and a defined course length, describes a different product from a bottle bought online with no independent verification of what’s in it. Treating the trial numbers as a shopping guide skips over that gap entirely. For a broader look at how thin the human-trial base runs across the peptide category Semax sits in, see Natural Peptides: Which Claims Have Human Trials.
Semax against better-documented alternatives
Modafinil, a wakefulness-promoting drug prescribed for narcolepsy and shift-work sleep disorder, sits on a substantially larger base of randomized human trials for its cognitive effects than Semax does. That comparison alone should recalibrate expectations: Semax is competing, in the popular imagination, with a compound that has decades of blinded human data behind it, using a fraction of the evidence.
Selank, a related heptapeptide from the same Russian research tradition, offers a sharper comparison. Selank is developed specifically for anxiolytic use rather than cognition. A randomized trial of 62 patients with generalized anxiety disorder and neurasthenia compared it directly with the benzodiazepine medazepam, finding similar anti-anxiety effects along with added antiasthenic and psychostimulant effects attributed to Selank.[6] That trial is explicitly randomized. The Semax stroke data cited earlier in this article is not. On the specific axis of study design, Semax’s better-known cousin currently has the stronger record. For the full comparison, see Semax vs Selank: two Russian peptides, one shared evidence problem.
What would settle the dosing question
Two things are missing from the record, and either one would change how confidently anyone could answer “what’s the Semax dose.”
No large, internationally registered randomized controlled trial of Semax currently appears on major trial registries such as ClinicalTrials.gov. And of the primary sources behind this article, most come from Russian-language journals sharing overlapping authors: Kamensky, Grivennikov, Dolotov and Inozemtseva appear on both of the mechanism papers, and Miasoedov appears on both the 1997 stroke trial and the 2008 Selank trial. A handful of connected labs, publishing mostly in their own language, is not the same evidentiary base as an independently replicated finding.
None of that means Semax does nothing. It means the specific numbers attached to it, the 12 and 18 milligrams, the BDNF fold-increases, rest on a narrower foundation than their confident circulation online would suggest.
The bottom line on Semax dosage
There is no standard, validated human dose for Semax. The figures that exist, 12 milligrams for moderate stroke and 18 for severe stroke, come from one open-label-appearing comparison in a single 1997 trial, added on top of conventional therapy rather than tested against placebo.[1] The mechanism people cite to justify taking it, increased BDNF expression, has only been measured in a rat hippocampus.[2] Anyone treating a retail nasal spray’s suggested dose as clinically validated is extending a single, non-blinded, decades-old Russian trial further than its own authors did.
This article is for research and informational purposes only and is not intended to diagnose, treat, cure, or prevent any disease. The peptides discussed here are sold for research use only and are not for human consumption. Nothing in this article constitutes medical advice. Consult a qualified clinician before making changes to a health, training, or supplementation protocol.
References
[Effectiveness of semax in acute period of hemispheric ischemic stroke (a clinical and electrophysiological study)].. Zhurnal nevrologii i psikhiatrii imeni S.S. Korsakova, 1997.
Semax, an analog of ACTH(4-10) with cognitive effects, regulates BDNF and trkB expression in the rat hippocampus.. Brain research, 2006.
[Neuroprotective effects of peptides bioregulators in people of various age].. Advances in gerontology = Uspekhi gerontologii, 2013.
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.. European journal of pharmacology, 2024.
Modulation of neuropathological pathways by bioactive peptides and proteins/polypeptides: Targeting oxidative stress in neurodegenerative diseases.. Neuropeptides, 2025.
[Efficacy and possible mechanisms of action of a new peptide anxiolytic selank in the therapy of generalized anxiety disorders and neurasthenia].. Zhurnal nevrologii i psikhiatrii imeni S.S. Korsakova, 2008.
