Semax: benefits, evidence and legal status

Semax is a synthetic seven-amino-acid peptide (Met-Glu-His-Phe-Pro-Gly-Pro) derived from a fragment of adrenocorticotropic hormone (ACTH), sold as nasal drops in Russia but not approved by the FDA; its best human evidence is small, mostly unblinded studies.

In rats, semax raises brain-derived neurotrophic factor (BDNF), a protein that supports nerve-cell survival [Animal] (Dolotov 2006), and it reduced infarct size after experimental stroke [Animal] (Romanova 2006). Those findings are consistent, but they come from animals. In people, the FDA’s July 2026 review found insufficient evidence of effectiveness and “insufficient clinical information to characterize the safety profile”; ClinicalTrials.gov lists no registered semax trial.

Item	Status
US FDA	Not approved; not a component of any FDA-approved drug (FDA briefing, 2026)
Compounding (503A/503B)	Withdrawn from Category 2 by nominators (FDA, page of April 22, 2026); on July 24, 2026 FDA’s advisory committee voted 8–5 with one abstention in favor of adding it to the 503A Bulks List, as reported by legal and trade press; the vote is advisory, not binding, and semax is not on the 503A Bulks List
WADA 2026 / 2027	Not named on either List; status unclear, see the legal section (WADA 2027 List)
Russia	Registered drug, 0.1% and 1% nasal drops (FDA briefing)
Strongest evidence	Human, small and mostly unblinded studies (n=12 to 187), most in Russian; no blinded efficacy trial found
Human trials	0 registered on ClinicalTrials.gov
Last checked	2026-10-02

Identity	Value
Sequence	Met-Glu-His-Phe-Pro-Gly-Pro (7 amino acids)
Molecular formula	C37H51N9O10S (free base)
Molecular weight	813.9 g/mol (free base); acetate salt 874.0 g/mol
CAS	80714-61-0
PubChem CID	9811102
Half-life	Not measured in humans; rapidly broken down in rat blood, main fragment Pro-Gly-Pro (Shevchenko 2006)
Routes studied	Intranasal (humans); intranasal, intraperitoneal, intravenous (animals)
Class	Synthetic ACTH(4-10) analogue without hormonal activity (nootropic / neuropeptide)

What is Semax?

Semax comes from Moscow research groups; study material described in the FDA review was made at the Institute of Molecular Genetics of the Russian Academy of Sciences. The peptide joins the ACTH(4-7) fragment, Met-Glu-His-Phe, to a Pro-Gly-Pro tail that is thought to slow breakdown by enzymes (FDA briefing). It keeps the neurobehavioral effects attributed to ACTH fragments but lacks ACTH’s hormonal effects. Its developers summarized fifteen years of work in 1997, reporting effects on attention, memory and tolerance of low oxygen in animals and people (Ashmarin 1997). Semax is a registered medicine in Russia, and Russian clinical studies tested it mainly in stroke and other brain disorders. Outside Russia it is unapproved and is sold online as a “nootropic” (a substance claimed to improve cognition), usually as a nasal spray. The PubChem record lists the sequence and the CAS number 80714-61-0.

How does Semax work?

No single receptor explains semax. The proposed mechanisms below come from animal and lab work; none has been confirmed in people.

• BDNF and TrkB signalling: one dose raised hippocampal BDNF protein up to 1.4-fold and activation of its receptor TrkB 1.6-fold in rats [Animal] (Dolotov 2006).
• Specific binding sites: labelled semax bound rat basal forebrain membranes with high affinity (KD 2.4 nM), and nasal doses raised BDNF there within 3 hours [Animal] (Dolotov 2006b).
• Serotonin and dopamine: in rodents the serotonin metabolite 5-HIAA rose to 180% in the striatum, and semax amplified amphetamine-driven dopamine release [Animal] (Eremin 2005).
• Enkephalinase inhibition: semax blocked enzymes in human serum that break down enkephalins, the body’s own opioid peptides (IC50 10 µM) [Lab] (Kost 2001).
• Immune and vascular genes: after experimental stroke in rats, over half of the genes semax changed were immune-response genes [Animal] (Medvedeva 2014).

What the research shows

The table grades each common claim by the strongest evidence behind it. Human rows are small and mostly unblinded; animal rows describe rodents, not people. The grading follows the site’s editorial policy and methodology.

Claim	Evidence level	Best study and result
Raises brain BDNF	Animal	Dolotov 2006, rats: BDNF protein up to 1.4-fold
Limits stroke damage	Animal	Romanova 2006, rats: smaller infarcts, better memory retention
Speeds stroke recovery	Human, controlled (n=30 vs 80)	Gusev 1997: faster regression of motor deficits
Raises blood BDNF after stroke	Human, unblinded (n=110)	Gusev 2018: higher BDNF, better Barthel index
Changes brain networks	Human, placebo-controlled (n=24)	Lebedeva 2018: fMRI change; no cognitive outcome
Relieves migraine pain	Human, uncontrolled (n=12)	Koroleva 1996, per FDA: pain stopped in 4 of 12
Slows motor neuron disease	Human, open-label (n=27)	Serdiuk 2007: no effect on course; quality of life improved
Antidepressant effect	Animal	Inozemtseva 2024, rats: reversed stress-induced anhedonia
Alzheimer’s disease	Animal	Radchenko 2025, transgenic mice: fewer amyloid deposits
Spinal cord repair	Animal	Liu 2025, mice: better functional recovery
Protects cells from copper	Lab (in vitro)	Tabbì 2015: less copper toxicity in cell lines
Treats ADHD	Hypothesis only	Tsai 2007: proposal, no study

Stroke and brain ischemia

This is the best-developed research area. In rats with a photochemically induced stroke, six days of nasal semax reduced infarct volume and preserved memory [Animal] (Romanova 2006), and a second rat study restored water-maze learning [Animal] (Silachev 2009). In a Russian clinical study, adding semax to standard care in 30 patients with acute ischemic stroke had “some influence” on the speed of recovery, mainly of motor deficits, compared with 80 patients on standard care alone [Human, controlled, n=110] (Gusev 1997). A 2018 study of 110 rehabilitation patients linked semax to higher plasma BDNF and better daily-living scores, without reported randomization or blinding [Human, n=110] (Gusev 2018). The FDA did not weigh these papers because no full English translation was available (FDA briefing).

Cognition and mood

The nootropic reputation rests mainly on rodent learning tests and on BDNF findings. The only placebo-controlled human data located are brain-imaging studies: in 24 healthy adults, nasal semax changed the default mode network on fMRI within 20 minutes [Human, placebo-controlled, n=24] (Lebedeva 2018), and a 52-person study found changed amygdala connectivity [Human, n=52] (Panikratova 2020). Neither measured memory or attention. In stressed rats, daily semax reversed anhedonia, the loss of interest in reward [Animal] (Inozemtseva 2024).

Pain, eyes and neurodegeneration

In 12 migraine patients, one nasal dose stopped headache in 4; in typical trigeminal neuralgia, pain did not change, and the authors concluded semax has no analgesic activity by itself [Human, uncontrolled, n=37] (FDA briefing). A Russian ophthalmology study reports better visual function in optic nerve disease when semax was added to standard treatment, but the abstract does not give patient numbers [Human] (Polunin 2000). In motor neuron disease, semax did not alter the disease course in 27 patients but improved quality-of-life scores [Human, open-label, n=27] (Serdiuk 2007). In Alzheimer’s model mice, semax improved maze performance and reduced amyloid deposits [Animal] (Radchenko 2025).

Human studies and trials

No semax trial is registered on ClinicalTrials.gov (searched 2026-10-02). The published human record is listed below; most papers are in Russian with English abstracts.

Study	Design, n	Result
Gusev 1997, acute stroke	30 semax vs 80 controls	Faster motor recovery
Gusev 2005, cerebrovascular insufficiency	187 patients; design not stated	Fewer strokes and TIAs reported; few side effects
Serdiuk 2007, motor neuron disease	Open-label, 27	No effect on course; better quality of life
Gusev 2018, stroke rehabilitation	Subgroups, 110	Higher BDNF, better Barthel index
Alekseeva 1999, post-hypoxic brain injury	Case series, 73	Helped memory problems; EEG paroxysms in some
Lebedeva 2018, healthy adults	Placebo-controlled, 24	fMRI network change
Panikratova 2020, healthy adults	Placebo-controlled, 52 (with Selank arm)	Amygdala connectivity change
Koroleva 1996, migraine and neuralgia	Uncontrolled, 37	Migraine pain stopped in 4 of 12; no effect in typical neuralgia

Status: 0 registered trials and 0 published double-blind efficacy trials found. The FDA briefing notes that semax has also been given to 451 children with depression or tic disorders in studies known only from meeting abstracts, where adverse events were not discussed.

Doses used in studies

These amounts are reported for scientific context only. They are not recommendations, and the human studies were small and mostly uncontrolled.

Route and amount	Species	Study
Intranasal, 50 µg/kg once	Rat	Dolotov 2006
Intranasal, 250 µg/kg daily for 6 days	Rat	Silachev 2009
Intraperitoneal, 0.15 mg/kg once	Rodents	Eremin 2005
Intraperitoneal, 60 nmol/kg daily	Rat	Inozemtseva 2024
Intranasal, 12–18 mg per day	Human, acute stroke	Gusev 1997
Intranasal, 6,000 µg per day	Human, stroke rehabilitation	Gusev 2018
Intranasal, 12 mg per day	Human, motor neuron disease	Serdiuk 2007
Intranasal, 0.5 mg/kg once	Human, migraine and neuralgia	Koroleva 1996, per FDA

No established human dose exists outside approved labels.

Side effects and safety

Semax’s safety in people is largely unmeasured rather than shown to be good or bad. The Russian clinical papers describe it as well tolerated, with a “minor percent of side-effects” in 187 patients [Human, n=187] (Gusev 2005). The FDA, reviewing the same literature, found that most papers did not report adverse events at all and that no human pharmacokinetic study exists by any route (FDA briefing).

• Reported effects: in 73 patients with post-hypoxic brain injury, EEG showed episodes of paroxysmal (seizure-like) activity in some cases after a dose, and the authors advised EEG monitoring at the first administration [Human case series, n=73] (Alekseeva 1999). FDA’s adverse-event database held one report: eye pain and burning after online-bought nasal drops, with hospitalization.
• Bleeding (theoretical): nasal semax increased anticoagulant and clot-dissolving activity in rats and cut platelet aggregation by about 35% [Animal]; FDA says this raises concern for people at risk of bleeding or taking drugs that increase bleeding risk.
• Dopamine and abuse potential (theoretical): semax amplified amphetamine-induced dopamine release in rodents [Animal] (Eremin 2005). FDA calls this “concerning” and found no study of abuse potential.
• Missing toxicology: FDA identified no acute, repeat-dose, genotoxicity, reproductive or standard two-year cancer studies. One mouse study reported smaller existing tumors, but FDA judged it “not adequately designed” to assess cancer risk.
• Immunogenicity: FDA lists a risk of immune reactions for some routes because of aggregation and peptide-related impurities (FDA Category 2 page).

Product quality. Products sold online are unregulated. FDA notes that “semax” is not an official drug name, that different salts and derivatives are sold under it, and that no website it searched said which form it contained. A Belgian state laboratory identified Semax and Selank in seized unknown preparations and found them sold online as powders and nasal sprays [Lab analysis] (Vanhee 2020).

Who should avoid it. No US label exists. Based on FDA’s concerns, people with bleeding disorders or on anticoagulant or antiplatelet drugs, people with a seizure history, and pregnant people (no reproductive studies) face risks that the evidence cannot rule out.

Legal and regulatory status

Jurisdiction	Status (checked 2026-10-02)
US FDA	Not approved; no USP monograph (FDA briefing)
US compounding	Withdrawn from Category 2; not on the 503A Bulks List (FDA 503A page)
WADA	Not named in the 2026 or 2027 List; S0 or S2 may apply
Russia	Registered, 0.1% and 1% nasal drops
Europe, Japan, international	No pharmacopoeia monograph (FDA briefing)

FDA advisory committee, July 24, 2026. FDA reviewed semax free base and semax acetate for cerebral ischemia, migraine and trigeminal neuralgia; the meeting page notes that advisory committees make non-binding recommendations (FDA meeting page). FDA staff concluded the criteria “weigh against” adding either form to the 503A Bulks List (FDA briefing). The committee voted on each form (FDA questions); legal and trade press reported an 8–5 vote with one abstention in favor of inclusion. FDA had not posted minutes on the meeting page when this entry was checked. FDA states it will not issue a final determination until the committee’s input has been considered, and FDA adds substances to the 503A list through notice-and-comment rulemaking (FDA 503A page). Until FDA acts, the vote does not change semax’s status.

Earlier events. Semax was placed in Category 2, the FDA list of substances that may pose significant safety risks in compounding, and was moved to the “nominated but withdrawn” list after nominators withdrew (FDA Category 2 page). Withdrawal is not approval. The FDA briefing also notes that no outsourcing facility has reported compounding semax since 2019.

Sport. Semax is not named in the WADA 2026 or 2027 Prohibited List. S0 covers substances with “no current approval by any governmental regulatory health authority for human therapeutic use”, and the 2027 List adds peptides to its examples (WADA 2027 List). Because semax is registered in Russia, whether S0 applies is not settled. A 2025 anti-doping review classed semax as “unclear” status and linked it to tetracosactide, an ACTH analogue prohibited under S2 (Pokrywka 2025). Athletes should not assume it is permitted; USADA points athletes to Global DRO and its Drug Reference Team for substance questions. The site’s WADA banned peptides 2027 page explains the classes.

Related peptides

Semax was reviewed at the same FDA advisory meeting as BPC-157, which has a similar regulatory history but is studied for tissue repair. Selank, a related heptapeptide (Thr-Lys-Pro-Arg-Pro-Gly-Pro), ends in the same Pro-Gly-Pro sequence and inhibits the same serum enzymes (Kost 2001). Melanotan II, another synthetic ACTH(4-10) analogue and a potent melanocortin-receptor agonist, was tested alongside semax in a rat stress model (Inozemtseva 2024).