Does Semax work? What published studies of Semax found

Evidence limits

Does Semax actually work?

Read those lines together. Russia registered Semax as a drug, and Russian groups published clinical reports. No FDA-approved drug contains Semax, and the FDA's briefing document says it did not consider Russian-language references that lacked a full English version, a list that includes Gusev et al. 1997, 2005 and 2018 (FDA briefing document, July 2026). A drug can be registered in one country and have evidence that a foreign regulator did not weigh. The ARTG, Australia's register of approved therapeutic goods, was not searched for this page.

A PubMed search on 14 October 2026 returned 232 records for 'semax' and none tagged as a systematic review or meta-analysis, so that search did not find a pooled rating of the human studies.

What has Semax been studied for?

Semax has been studied in stroke, chronic cerebrovascular disease, optic nerve disease, glaucoma, motor neuron disease, peptic ulcer, migraine, trigeminal neuralgia, depression and tics in children, and brain damage after oxygen loss, mostly in small studies (FDA briefing document, July 2026; PubMed records, 14 October 2026). The same briefing document, citing the Russian State Register of Medicines, says Semax is a registered drug in Russia (FDA briefing document, July 2026). The FDA says it did not evaluate ADHD, which was nominated as a use, because it did not find supporting literature (FDA briefing document, July 2026).

A 2022 review lists Semax among original Russian peptide drugs in the nootropic group, a label for compounds studied for effects on memory and attention (Deigin et al., Pharmaceutics, 2022). The Russian State Register of Medicines, which is in Russian, was not searched for this page, so the list above is what studies tested, not an approved-uses list.

The catalogue lists Semax reference material for laboratory research, and a catalogue listing is not a statement about any use.

What have clinical studies of Semax found?

In 110 stroke patients, plasma BDNF stayed raised and recovery scores improved more in the Semax subgroups, the abstract does not describe random allocation, and the authors' conclusions rest on correlations (Gusev et al., Zhurnal Nevrologii i Psikhiatrii, 2018). In 30 stroke patients given Semax and 80 comparison patients, the authors reported some influence on the rate of neurological recovery, in a study whose abstract does not describe randomisation or blinding (Gusev et al., 1997). In an open-label study of 27 patients with motor neuron disease and no placebo group, the authors reported no effect on the course of chronic partial denervation and a higher quality-of-life score (Serdiuk et al., 2007).

Results differ by condition. In a study of patients with optic nerve disease split into two Semax groups and a control group, the authors reported better visual acuity, visual field and colour vision, and the abstract does not give the number of patients (Polunin et al., Vestnik Oftalmologii, 2000). In 32 adults with refractory peptic ulcer given Semax with standard ulcer medicines, 89.5% of the Semax group were healed at day 14 against 30.8% of controls, in a combined-treatment comparison whose authors called for further clinical studies (Ivanikov et al., Bulletin of Experimental Biology and Medicine, 2002; patient count from the FDA briefing document, July 2026). In 42 patients after retinal detachment surgery, both groups received Semax and group 1 also received a physiotherapy course, and group 1 showed higher visual-function recovery by month 6, in a study with no group without Semax that cannot show what Semax added (Sheludchenko et al., Vestnik Oftalmologii, 2026). The table below lists each study with its limits.

In 24 healthy adults, 14 given Semax and 10 given placebo, resting brain scans showed a larger volume of one part of the default mode network in the Semax group, in a small study that measured scan patterns and not outcomes in illness (Lebedeva et al., Bulletin of Experimental Biology and Medicine, 2018). In 52 healthy participants scanned under Semax, Selank or placebo, the study reports differences in connectivity between the right amygdala and a region of right temporal cortex, a scan measure and not a health outcome (Panikratova et al., Doklady Biological Sciences, 2020).

How does Semax affect the brain?

In rats, hippocampal BDNF protein rose up to 1.4-fold and tyrosine phosphorylation of its receptor, TrkB, rose 1.6-fold, in a rat study whose abstract gives no rat count and which does not establish a human result (Dolotov et al., Brain Research, 2006). In rats, NGF and BDNF gene expression rose in the hippocampus one hour after Semax and Ngf expression fell in the frontal cortex, a measure of gene activity and not of function, in a rat study whose abstract gives no rat count and which does not establish a human result (Agapova et al., Neuroscience Letters, 2007).

In rats with a stroke-like injury, the expression of immune- and vessel-related genes changed, a result from an abstract with no rat count that does not establish a human result (Medvedeva et al., BMC Genomics, 2014).

The default mode network is a set of brain regions that is active when you are not focused on a task. In 14 healthy adults given Semax and 10 given placebo, scans taken 5 and 20 minutes later showed a larger front part of that network in the Semax group, in a small study that measured network size and not any change in thinking or health (Lebedeva et al., Bulletin of Experimental Biology and Medicine, 2018).

The FDA's briefing document says the molecular mechanisms by which Semax regulates gene expression are unknown (FDA briefing document, July 2026). In a test-tube experiment, labelled Semax bound specifically and reversibly to plasma membranes from rat forebrain, a rat-membrane result from an abstract with no replicate count that does not establish a human result (Dolotov et al., Russian Journal of Bioorganic Chemistry, 2004).

What does research show Semax does in the body?

In rodents, Semax increased the extracellular level of a serotonin breakdown product and enhanced amphetamine-induced dopamine release and movement; the abstract gives no group sizes, and this does not establish a human result (Eremin et al., Neurochemical Research, 2005). In 32 adults with refractory peptic ulcer given Semax with standard ulcer medicines, 89.5% of the Semax group were healed at day 14 against 30.8% of controls, in a combined-treatment comparison whose authors called for further clinical studies (Ivanikov et al., Bulletin of Experimental Biology and Medicine, 2002; patient count from the FDA briefing document, July 2026).

Rat studies of the heart and stress hormones are covered on the Semax side effects page, where the recorded adverse events also sit. For how peptides signal in the body in general, see what peptides do.

How quickly do effects of Semax show up in studies, and do they last?

No study read for this page defines when effects of Semax begin, or how long they last, in a person. In rats given radiolabelled Semax, 0.093% of the total radioactivity per gram was found in brain at the first sample, 2 minutes later, and 80% of it was intact Semax, a measure of uptake and not of an effect, from an abstract with no rat count that does not establish a human result (Shevchenko et al., Russian Journal of Bioorganic Chemistry, 2006).

In 14 healthy adults given Semax and 10 given placebo, scans taken 5 and 20 minutes later showed a larger volume of one default mode network component in the Semax group, in a small study that measured network size and not a health outcome (Lebedeva et al., Bulletin of Experimental Biology and Medicine, 2018).

In cultured rat glial cells, BDNF messenger RNA rose eight-fold and NGF five-fold at 30 minutes, the largest rise reported, a cell-culture measure of gene activity from an abstract with no replicate count that does not establish a human result (Shadrina et al., Neuroscience Letters, 2001). In rats, hippocampal gene expression of both factors had risen at the 1-hour measurement, a gene-activity result and not a functional one, in a rat study whose abstract gives no rat count and which does not establish a human result (Agapova et al., 2007).

The one duration statement among the sources read is in a 1997 review by Semax's developers, whose abstract gives no individual measurements and no participant numbers, so this page does not repeat its figure (Asmarin et al., Zhurnal Vysshei Nervnoi Deiatelnosti, 1997). In 110 stroke patients, plasma BDNF stayed raised across a 5-month study that included treatment, which does not measure how long Semax persists (Gusev et al., 2018).

Each time is when researchers looked, not when an effect began or ended. The Semax half-life page tabulates every time point and explains what is and is not known about how long Semax stays in tissue.

What clinical trials and reviews of Semax are in PubMed?

A PubMed search on 14 October 2026 returned 232 records matching 'semax'. Four are tagged as clinical trials, on stroke in 1997 and 2018, optic nerve disease in 2000 and motor neuron disease in 2007. Fifty-two carry the MeSH 'Humans' tag, and none is tagged as a systematic review or meta-analysis. A tag depends on how a record was indexed, so these counts describe the search and not the whole literature. The FDA's July 2026 briefing document is the most detailed review of the Semax literature among the sources read.

Two review articles among the sources read name Semax: a 2022 review of Russian peptide drugs and a 2025 review of brain-doping substances (Deigin et al., Pharmaceutics, 2022; Pokrywka et al., Biology of Sport, 2025). The FDA's briefing document says it did not consider Russian-language references that lacked a full English version, so a title-level reading of the record can mislead.