Semax: what the human studies measured, and where they stop

Semax, available here in 10 mg and 5 mg presentations, is a synthetic ACTH(4-10) analogue, and its human literature is unusual in shape: a handful of clinical studies from one research tradition, mostly published in Russian, mostly not randomised. That is worth describing exactly, because the compound is often described as if it had a Western trial record.

The acute stroke study, with its control group named

The 1997 study examined 30 patients in the acute period of hemispheric ischaemic stroke. Its control group, in the authors’ own words, “consisted of 80 patients with the strokes analogous in severity and location of the damages and which were treated by conventional therapy” — that is a historical control, not a randomised one, drawn from a comparable population rather than assigned at the same time.

Outcome measurement was both clinical and electrophysiological: rating scales for severity and neurological deficit, plus EEG with mapping and repeated somatosensory evoked potentials with mapping. The conclusion is carefully hedged: including Semax in combined intensive therapy “had some influence on the rate of restoration of the damaged neurological functions”, in terms of greater regression of general cerebral and focal, especially motor disorders.

The abstract also records the regimen the study used: most effective daily doses were 12 mg for moderate strokes and 18 mg for severe strokes, over courses of 5 and 10 days. Those are figures from that trial, quoted here as what the study ran.

The cerebrovascular insufficiency study

The 2005 paper covered 187 patients with different stages of cerebrovascular insufficiency, diagnosed on neurological and neuropsychological assessment, dopplerography, rheo- and encephalography, electrocardiography, brain MRI and fundoscopy. Its reported results were significant clinical improvement, stabilisation of disease progression, and reduced risk of stroke and transient ischaemic attacks during the disease course, with a minor percentage of side effects.

What the abstract does not contain is a control arm. A 187-patient study with no comparator described in the indexed abstract is a description of an experience, not a measurement of an effect.

The fMRI study: a connectivity result, not a cognitive one

The most recent human work is from 2020, and it is the most interesting because of what it tried to do. Resting-state functional connectivity was assessed in 52 healthy participants, with fMRI performed three times — before, and 20 minutes after, an injection of Semax, Selank, or placebo. The regions of interest were the amygdala and dorsolateral prefrontal cortex.

The finding: between-group and between-condition differences in functional connectivity between the right amygdala and a region in the right temporal cortex, including fusiform, inferior and middle temporal and parahippocampal gyri, with post-hoc analysis defining general and specific effects of each peptide “for the first time”.

Read this precisely. It is a change in the coupling of two brain regions on an imaging measure, in healthy people, twenty minutes after a single injection, with three formulations including placebo. It is not a cognitive test, it is not a mood scale, and the authors themselves frame the discovery of the effect as the contribution.

The animal work, for contrast

The 2008 rat study used photochemically induced thrombosis to create a defined infarct zone in prefrontal cortex, which impaired acquisition of the Morris water maze on postoperative days 20 to 24. Six days of intranasal Semax at 250 µg/kg restored spatial learning ability, and the authors attribute this to neuroprotective action and induction of neurotrophic factors.

That is a clean animal experiment with a behavioural endpoint. It is a rat.

Where the class stands in 2026

The 2026 orthopaedics review lists Semax among neuroactive peptides acting through neurotrophic and HGF/c-Met pathways, and states plainly: “Although preclinical studies are promising, there is a current lack of clinical trials.” The same judgement appeared in the sports-medicine scoping review of the peptide class this year.

What the data do not show

• No randomised, blinded, placebo-controlled trial of Semax in people is cited here. The two clinical studies used historical or absent controls.
• No cognitive test outcome exists in the human file. The healthy-volunteer study measured functional connectivity on imaging, not performance.
• No trial establishes an effect on mood, anxiety or attention in humans.
• The animal results do not transfer, and the review literature of 2026 says so in one sentence.
• Nothing on this page is a dose recommendation or a health claim. The milligram figures are what those studies administered; these are research materials sold as research materials and nothing here advises any person to take anything.