Semax comes up in peptide catalogs as a nootropic, and unlike most of that shelf it has genuinely been given to patients. That makes it worth taking seriously and easy to overstate. The studies are real; they are also three decades old in places, carried out almost entirely in one country, and mostly built without the control arm that would let you trust the result. This page reads them one at a time. It sits under the longevity goal, which is where the cognitive claims are made.
What semax is
Semax is a synthetic fragment of adrenocorticotropic hormone with a short tail added to slow it down in the body: the ACTH(4-7) sequence joined to proline, glycine and proline. It was developed in Russia and is given intranasally, usually as a 1% solution. It carries none of the hormonal activity the full ACTH molecule has, which is the point of using a fragment.
The mechanism most often quoted for it is neurotrophic. A 2006 paper in the Journal of Neurochemistry reported that semax binds specifically in rat basal forebrain and raises levels of brain-derived neurotrophic factor protein there [4]. That is a rat study, and it is the kind of finding that explains why a compound is interesting rather than evidence that it works.
The human studies, one at a time
The earliest and most quoted is a 1997 stroke study [1]. Thirty patients in the acute period of hemispheric ischemic stroke were given semax alongside intensive therapy (n = 30). They were compared with a separately assembled control group on conventional therapy alone (n = 80), matched for severity and lesion location. Recovery was tracked on clinical rating scales, with EEG mapping and repeated somatosensory evoked potentials. The authors reported faster regression of general cerebral and focal deficits, motor deficits especially.
A 2000 paper in a Russian ophthalmology journal studied optic nerve disease [3]. Patients were split three ways by how the drug reached them: intranasal drops, endonasal electrophoresis, and a control group. The authors reported faster recovery of visual function where semax was added to standard treatment.
A 2007 study looked at motor neuron disease (n = 27) [2]. Needle electromyography was performed three times, sixty days before enrollment and again on day 1 and day 48, in three muscles on the less affected side. This is the study to read carefully, for a reason that has nothing to do with its result.
The newest study is also the most carefully built, and the least clinical [5]. In 2018, 24 healthy volunteers, 11 men and 13 women aged 43.9 ± 9.5 years, were scanned with resting-state functional MRI before and at 5 and 20 minutes after a single intranasal dose of semax (n = 14) or placebo (n = 10). The semax group showed a larger rostral subcomponent of the default mode network, in medial frontal cortex. That is a real placebo-controlled measurement. It is also a measurement of brain network topography in healthy people, not of anyone getting better.
What that adds up to
Three patient studies and one imaging study, across three decades, none of them large. The stroke study is the biggest and its control group was assembled separately rather than randomized, which is the design most likely to flatter a treatment. The motor neuron study was open-label, so everyone involved knew who was getting what. The only placebo comparison was in healthy volunteers.
None of this makes semax a fraud. It makes it a compound with early, mostly uncontrolled human data from one research tradition, which is a fair distance from the phrase “clinically studied” that a product page will reach for. No semax study appears on ClinicalTrials.gov, so no protocol, endpoint or result has been lodged in the registry Western regulators read.
Where it stands with the FDA
Semax appears on neither FDA bulk drug substance list: not in the 503A categories, updated May 14, 2026, and not in the 503B categories, updated March 21, 2025. There is no lawful route for a pharmacy to compound it in the United States, whatever the Russian literature found. A seller advertising it is not offering a prescription product you can lawfully have made.
For how that framework decides what a pharmacy may prepare, see what makes a peptide legal to prescribe. For the other Russian peptide with real human trials behind it, see selank, and for one with laboratory work and no human trial at all, KPV.