Semax vs Selank: Two Russian Peptides, Their Evidence and Lab Results

Semax and Selank are sibling peptides from the same Moscow institute, but they start from different molecules. Semax is built from a fragment of the hormone ACTH and is studied mainly for stroke recovery and cognition. Selank is built from tuftsin, an immune-system fragment, and is studied mainly for anxiety. Neither is FDA-approved (DailyMed lists no label for either as of September 29, 2026), though a 2022 peer-reviewed review lists both among 14 peptide products registered in the Russian Federation (Deigin et al., Pharmaceutics, 2022). The human evidence for each is small, mostly Russian, and unreplicated, and only one study we found included both peptides in the same design (participants received Semax, Selank or placebo).

At a glance

How they work differently

Semax's sequence is Met-Glu-His-Phe-Pro-Gly-Pro. It combines the ACTH(4-7) fragment with the Pro-Gly-Pro tail, which the developers added so enzymes break it down more slowly, and it shows no hormonal activity (Radchenko et al., Acta Naturae, 2025). Selank is a synthetic analog of tuftsin, a short piece of an antibody chain, extended at one end with the same three amino acids (Volkova et al., Frontiers in Pharmacology, 2016).

The mechanism work is mostly in rats. Semax raised BDNF, a brain growth factor, in the rat hippocampus along with better performance on a learning task (Dolotov et al., Brain Research, 2006), and activated serotonin and dopamine systems in rodents (Eremin et al., Neurochemical Research, 2005). Selank changed the activity of many neurotransmission-related genes in the rat frontal cortex, in a pattern that overlapped with GABA's own effect, and the authors suggest one possible mechanism is allosteric modulation of the GABA system (Volkova et al., 2016). The two share one laboratory finding: both inhibited enkephalin-degrading enzymes from human serum in test-tube experiments (Kost et al., Bioorganic Chemistry, 2001). These are rat and lab results. They describe plausible mechanisms, not what either does in a person.

What the evidence shows

Head to head: one study put both peptides in the same design. Panikratova and colleagues, from the same Moscow research network that developed the peptides, scanned 52 healthy adults with resting-state fMRI after they received Semax, Selank or placebo, and found differences between groups in connectivity between the right amygdala and regions of the right temporal cortex, with both shared and peptide-specific effects (Panikratova et al., Doklady Biological Sciences, 2020). That shows both peptides do something measurable to brain networks. The study did not report anxiety, memory or any clinical outcome, so it can't say which works better for anything. No trial compared them for anxiety, stroke or cognition.

Semax. Most of its human data is in stroke. A 1997 Russian study gave Semax alongside standard therapy to 30 people in the acute period of ischemic stroke and compared them with 80 similar patients on standard therapy alone. It reported faster recovery of general and motor deficits (Gusev et al., 1997). A 2018 study of 110 stroke patients, split by whether they received Semax and by the timing of rehabilitation, reported higher blood BDNF levels and better functional scores with Semax (Gusev et al., 2018). Neither abstract describes randomization or blinding, both come from a single research network, and we found no independent replication. A separate small imaging study in 24 healthy volunteers, with a placebo group, found a difference in one brain network (Lebedeva et al., 2018). Memory and focus claims rest on rat data plus these imaging studies; none reported cognitive test results. More in our Semax guide.

Selank. Its clinical case is one 2008 Russian study of 62 people with generalized anxiety disorder or neurasthenia: 30 received Selank and 32 received medazepam, an older benzodiazepine-type drug. The anxiolytic effects were reported as similar, with Selank also showing antiasthenic and psychostimulant effects (Zozulya et al., 2008). PubMed indexes it as a randomized comparative study, but the abstract doesn't describe blinding, and there was no placebo arm, so it can't separate drug effect from placebo response. Rat studies add GABA-related findings, and the imaging study above covers healthy adults. More in our Selank guide.

Which has stronger evidence? Semax has more published human studies that we found, and Selank's rests on one comparison trial. That is a difference in quantity, not quality: all of it is small, from the same network that developed the peptides, and none of it is the placebo-controlled, independently replicated work regulators usually want. The evidence is not strong for either, and it is for different conditions, so it can't rank them.

Side effects and risks compared

Both are sold as research chemicals, and that is part of the risk picture: there is no approved label to check a vial against.

Neither peptide has a dedicated human safety trial that we could find; the published human studies weren't designed to track adverse events. What FDA has said is on its page about bulk drug substances that may present significant safety risks (content current as of April 22, 2026), where both appear under "nominated but withdrawn." For Selank, FDA says compounded drugs may carry a risk of immunogenicity (an immune reaction to the peptide) from aggregation and peptide-related impurities, and that it lacks important information about safety in humans. For Semax, it gives the same immunogenicity concern and says it lacks sufficient information to know whether the drug would cause harm in humans (FDA).

Semax has since moved further in that process. FDA's Pharmacy Compounding Advisory Committee had Semax on its July 24, 2026 agenda (FDA meeting page), and the National Community Pharmacists Association reported that it voted to recommend adding Semax to the 503A bulks list. That vote is non-binding and needs approval from the Secretary of Health and Human Services (NCPA, July 31, 2026). Selank is not on that meeting's agenda.

What the lab tests show

These figures come from independent lab reports, mostly paid for by buyers, in the VendorVerify database as of September 30, 2026 (vendors we can't identify excluded; a vial counts as failed if it is empty, under 95% pure, more than 20% under label or more than 50% over).

Semax: 12 vials from 7 vendors, 8 reports. Median purity 99.55%. Median dose came in 17.5% over label, and 6 of the 12 vials were more than 20% over. As of September 30, 2026, every vial was 99% pure or higher and none had failed.

Selank: 9 vials from 5 vendors, 6 reports. Median purity 99.79%. Median dose came in 25.2% over label, and 6 of the 9 vials were more than 20% over. None had failed as of September 30, 2026.

One caution on the overfill: extra peptide isn't contamination, and neither set had a vial fail our definition, but it means the labeled amount is a rough starting point rather than a measurement. The pattern is the same for both: clean purity and heavy overfill, so the number on the label tells you little about how much peptide is in the vial. The samples are small, the Selank set is smaller, and one vendor's vials can move these figures, so they can't rank one compound against the other.

Here is who has the strongest independent lab record for Semax right now. A vendor needs five or more lab reports on a compound to be ranked, so the table may be short or empty:

And for Selank, where the table also flags vendors with more than 5% failed vials:

Buying smarter

Because both run over label in our data, a quantity result from your own vial says more than another purity number. The Semax results page and Selank results page show each vendor's record on that specific compound, including how many reports sit behind a score and how recent they are. Treat a posted COA as a starting point; how to read a peptide COA and fake peptide COAs cover what to check. To order your own test, see how to get a peptide tested. If a report comes back bad, stop using that batch, tell your group, take the report to the vendor, and submit it to VendorVerify.

Database figures without a date update automatically from VendorVerify's lab results; dated figures are snapshots. This isn't medical advice.