Semax

Semax has been given to people under study conditions, and the packet holds one such trial. Neurologists at the Pirogov Russian National Research Medical University in Moscow followed 110 stroke survivors for five months, randomly splitting them into a group that received Semax on top of standard rehabilitation and a group that received rehabilitation alone. There was no placebo, and the report does not describe any blinding.

The Semax groups showed a much larger rise in plasma BDNF, a growth factor the body makes itself, and they gained points on the Barthel index of daily independence faster in the first month. Recovery of limb weakness did not differ with Semax. The rest of the packet is rat stroke work from a Moscow genetics institute, measuring which genes and proteins shift in the injured brain.

In the United States it has no approval, and it dropped off FDA's Category 2 compounding list when its nominators withdrew, leaving it on neither list. An FDA advisory committee took it up in July 2026; no minutes, transcript or vote from that session have been released.

Built in the lab rather than taken from tissue, Semax is a heptapeptide, Met-Glu-His-Phe-Pro-Gly-Pro. Its first four residues are the 4 to 7 stretch of ACTH (adrenocorticotropic hormone), and the last three are a Pro-Gly-Pro tail added to the fragment. Its developers group it with the melanocortins, the hormone family built from the same precursor as ACTH and the melanocyte-stimulating hormones.

Almost all of its literature comes from Russian institutions. The Moscow genetics group behind the rat studies describes it as a drug used in treating ischemic stroke, and the 2018 trial authors describe it as one of the agents given in the earliest period after a stroke. FDA's lists name it Semax (heptapeptide).

What the Rat Stroke Studies Measured

Both animal papers in the packet come from the Institute of Molecular Genetics in Moscow, and both block a cerebral artery in rats to mimic a stroke, then look at what Semax changes in brain tissue.

The 2017 study read gene activity across the rat cerebral cortex after focal ischemia. The biggest shift attributed to Semax was in immune genes: stronger antigen presentation signalling, amplified interferon signalling and higher expression of an immunoglobulin heavy chain gene, along with changes in cytokine, stress response and ribosomal protein genes. Given alone, the Pro-Gly-Pro tail did something different, damping immune activity and suppressing neurotransmission. The authors propose that Semax protects through crosstalk between the nervous and immune systems, which is a hypothesis drawn from gene lists, not a demonstrated pathway.

The 2021 study measured proteins 24 hours after a temporary artery blockage in male Wistar rats, five to seven animals per group. Semax raised active CREB, a protein the authors link to neuroprotection and recovery, by more than 1.5-fold in the deep subcortical tissue where the damage centred. It lowered MMP-9 and c-Fos in the neighbouring cortex and lowered active JNK, which they link to inflammation and cell death, in both regions. A drop in MMP-9 in the damaged subcortex itself did not reach significance.

From Rat Tissue to a Blood Test

The human trial did not measure any of those brain proteins. It measured BDNF in blood, which is a marker of the body's own growth-factor activity, not a measure of repair. The trial authors themselves note that high BDNF did not always track with recovery of movement, and that patients who started rehabilitation early recovered limb strength better even at equal or lower BDNF levels.

Most of the Semax overreach comes from carrying a stroke rehabilitation result into healthy brains.

Treating a BDNF rise as a brain upgrade. The Moscow trial showed plasma BDNF roughly doubled within a month on Semax. BDNF in blood is a marker, and the same authors report it did not reliably track recovery of movement. A higher number on a blood test is not evidence of better thinking, learning or memory.

Moving from stroke recovery to everyday focus. The patients were people relearning daily tasks months after a confirmed stroke, and the measured gain was in Barthel scores such as walking, transfers and self-care. Nothing in the packet tests attention, memory or productivity in people without brain injury.

Reading clinical use in Russia as proof of effect or safety. One randomised but open, placebo-free trial of 110 patients is the entire controlled human record here. Widespread prescribing in one country does not substitute for blinded trials, and it does not establish Western regulatory acceptance.

Calling rat gene changes neuroprotection in people. The immune gene and protein shifts were measured in rat brain tissue within a day of an induced stroke. They suggest where to look. They do not show that Semax limits brain damage in humans.

The human safety record in the packet is one trial: 110 stroke patients, each Semax recipient given two ten-day courses within a five-month follow-up. The authors write that Semax caused no side effects and no worsening of general health, and that 95 percent of patients rated its tolerability good or very good. The paper gives no adverse event table and no comparison against placebo, so that statement cannot be checked against rates.

FDA has not approved Semax, and it is on neither the Category 1 nor the Category 2 compounding list. It left Category 2 on or about 22 April 2026 because the nominators withdrew their nomination; FDA did not resolve its safety concerns, and those concerns remain published. FDA's Pharmacy Compounding Advisory Committee reviewed Semax at its meeting of 23 and 24 July 2026. FDA has posted the agenda, roster, voting questions and its own slides from that meeting but no minutes, transcript or written vote tally, so no outcome is known. Whatever the committee recommended, a committee recommendation is not an FDA decision and changes no legal status; rulemaking on the full record does that.

WADA status is unknown. Semax is not named on the Prohibited List. Whether the S0 class for non-approved substances covers it depends on whether any national regulator currently approves it for human therapeutic use, reportedly including Russia, and that has not been confirmed against a primary registry.

Semax has something many compounds in its category lack: a randomised human trial with a named population and published numbers. That trial was in stroke rehabilitation, and its clearest results were a marker, BDNF, and a faster climb on a daily-function scale, not a change in muscle strength.

What the packet supports: in 110 Russian stroke survivors, adding Semax to rehabilitation was associated with higher plasma BDNF and faster early Barthel gains, with good reported tolerability over two short courses. In rats, it shifts immune and stress-signalling genes and proteins in injured brain tissue.

What it does not support: any effect on cognition, focus or mood in healthy people, any blinded confirmation of the stroke result, or any safety picture for repeated or long-term use. The trial gave a defined preparation to patients under neurological supervision, which is a different situation from an unregulated product.

Not a tested nootropic. The only human outcomes in the packet are blood BDNF, limb strength and daily-function scores in stroke survivors. Focus, memory and productivity in healthy adults were never measured.

Not FDA approved. It sits on neither 503A list after its nomination was withdrawn, FDA's published safety concerns stand, and the July 2026 advisory committee review has no published outcome.

Not proven by its Russian clinical history. The controlled human evidence here is a single open-label trial without placebo, from one Moscow university.

Not settled for tested sport. WADA does not name it, and whether the non-approved substances class applies turns on a national approval that has not been confirmed.

I think Semax deserves more credit than the usual research-peptide treatment and a lot less than the nootropic marketing gives it. There is a real randomised trial behind it, run by a stroke group in Moscow, and the BDNF and Barthel results are not nothing.

But I keep coming back to what that trial actually was: stroke patients, no placebo, no blinding, and the outcome that would have been easiest to measure objectively, muscle strength, did not move with Semax. When the strongest human data is that thin and that specific, I am not going to stretch it to cover sharper focus on a Tuesday afternoon.

I will not publish doses or suggest anyone try it. The numbers on this page came from a supervised neurological setting, and the honest summary is that a blinded trial outside Russia is the missing piece.