What is Semax?

So what is Semax? A Soviet-era stroke drug, not a new discovery

Semax is a synthetic heptapeptide, seven amino acids in the sequence Met-Glu-His-Phe-Pro-Gly-Pro. The first four amino acids are lifted from ACTH(4-7), a fragment of adrenocorticotropic hormone, the pituitary hormone that normally tells your adrenal glands to release cortisol. Researchers at the Institute of Molecular Genetics in Moscow clipped that fragment down so it no longer triggers a cortisol response, then welded on a three-amino-acid tail, Pro-Gly-Pro, to stop enzymes in the blood from breaking it down too fast. What is left is a peptide that keeps ACTH's effects on the brain without touching the adrenal system at all.

That work dates to the 1980s. By the 1990s, Semax was a licensed prescription drug in Russia, sold as a nasal spray under that exact brand name for ischemic stroke and cognitive impairment. It has decades of real-world use there. None of that history translates to US regulatory status. Semax has never been submitted for, let alone granted, FDA approval for any indication, in any form.

How Semax is supposed to work

The best-documented mechanism is a jump in brain-derived neurotrophic factor, or BDNF, a protein that supports the survival and growth of neurons and underlies a lot of what researchers call synaptic plasticity, the brain's ability to strengthen or rewire connections. A widely cited 2006 rat study found a single dose of Semax produced a roughly 1.4-fold rise in hippocampal BDNF protein and a threefold jump in BDNF gene expression within hours. Semax also appears to modulate dopamine and serotonin signaling and to dial down inflammatory cytokines during oxygen deprivation, which is the theoretical basis for using it after a stroke.

All of that is real, published, peer-reviewed animal and cell-culture science. It is the same category of preclinical evidence that underlies several of the peptides in our peptide stacks guide: a plausible, mechanistically interesting story that has not been confirmed in a large, controlled human trial.

The human evidence is real, and it is also thin

The strongest human data comes from Gusev and colleagues, published in 2018: 110 patients recovering from ischemic stroke, split by how far out from their stroke they were, with subgroups receiving intranasal Semax on top of standard rehabilitation. Patients on Semax showed higher plasma BDNF and better scores on stroke-recovery measures, including motor function and a general disability index, than those on standard care alone.

The catch is in the design. The Gusev trial was open-label: no placebo group, no blinding, everyone involved knew who was getting the drug. That is a much lower bar than a randomized, double-blind trial, and it is the kind of study design that tends to inflate reported benefit, especially on subjective measures. It was also run in Russia, the one country where Semax is an approved, marketed product, by researchers with an interest in confirming it works. None of that means the result is wrong. It means the evidence sits well below what the FDA typically requires to call a drug effective, which turned out to matter directly when the agency reviewed Semax in 2026.

What the FDA actually found in 2026

Semax was one of seven peptides the FDA's Pharmacy Compounding Advisory Committee reviewed for the 503A Bulk Drug Substances List, the list that lets a licensed compounding pharmacy fill an individual, physician-written prescription under formal FDA recognition. The nomination FDA staff evaluated proposed Semax for cerebral ischemia, migraine, and trigeminal neuralgia. In briefing documents posted ahead of the July 23-24, 2026 hearing, staff recommended against including Semax, in either its free-base or acetate-salt form.

Two findings stand out. First, a characterization problem: "Semax" is a common name, not a USAN-recognized designation, and FDA said it had encountered multiple different salts and derivatives sold commercially under that same name, exactly the kind of inconsistency that made epitalon's review complicated too. Second, and more striking: FDA staff said they could not identify a single human pharmacokinetic study of Semax, by any route of administration, in the entire record. Not injectable, not intranasal, not the route Semax is actually approved for in Russia. On top of that gap, staff rated the human evidence for Semax's proposed neurological uses as thin.

The panel voted yes anyway

On July 24, 2026, the Pharmacy Compounding Advisory Committee voted 8-5 to recommend Semax for the 503A list, against its own staff's written recommendation. We covered that full vote when it happened, alongside Epitalon's 7-4 approval and Emideltide's rejection, the only no vote across all seven peptides the committee reviewed. Semax fits a pattern that held for six of the seven substances in this hearing: FDA staff flagged real gaps in the evidence, and the outside panel recommended access anyway.

A committee vote is advice, not a rule. The FDA can adopt it, narrow it, or ignore it entirely, and it has already broken with its own staff's recommendation on this exact substance once. If the agency moves forward, the next step is a formal notice-and-comment rulemaking, with its own public comment period and no fixed deadline, before a compounding pharmacy could legally fill a Semax prescription under this pathway.

Is Semax legal, and should you use it?

Nothing about buying Semax today changes because of the July vote. It was never FDA-approved, and it is not a controlled substance, so it continues to be sold online as a research chemical regardless of how the 503A rulemaking eventually lands. That market sits entirely outside the process this article describes. A 503A decision only ever governs whether a licensed pharmacy can compound the substance for a documented patient under a physician's prescription, not whether a website can sell it to you directly.

The more useful takeaway is about the evidence gap itself. Semax has a real mechanism, real animal data, and decades of use in Russia. What it does not have, according to the agency that just reviewed it, is a single human study measuring how the drug actually behaves in the body, by any route, in the standard the FDA uses to evaluate anything else. That is a different and larger gap than "needs more research." If you are weighing Semax against other peptides people stack alongside a GLP-1, see our guide to peptides and GLP-1s for how the evidence compares across the field.