Selank: Three Russian Trials, and Not One of Them Says How Much

What it is

Selank is a seven amino acid peptide, a short chain of amino acids linked end to end. The first four are tuftsin, a peptide fragment found in antibodies, to which three more amino acids are added at the tail to slow its breakdown by enzymes. Russia registered it with its Ministry of Health in 2009 and it is marketed there as nasal drops for anxiety. It has not been approved anywhere else.

What the trials found

Go to PubMed, the US database for medical abstracts, filter for clinical trials, and search for Selank.

Three results, all in Russian, all in the Korsakov Journal of Neurology and Psychiatry. All three include N.F. Miasoedov of the Institute of Molecular Genetics, who created Selank.

Zozulya 2008: 62 patients diagnosed with generalized anxiety and neurasthenia, an older word for persistent exhaustion, received either Selank or the benzodiazepine medazepam, 30 on Selank and 32 on the sedative. Both treatments reduced anxiety scores by about the same amount. Selank additionally showed antiasthenic and psychostimulant effects, meaning it reduced the fatigue rather than adding to it, which is the direction that separates it from a sedative. (PMID 18454096)

Medvedev 2014: 60 patients with phobias and anxiety disorders plus somatoform diagnoses, meaning physical symptoms with no physical cause found, received either Selank or another benzodiazepine, phenazepam. The effect on anxiety was strong, with minor impairments of thinking, and still noticeable a week after treatment ended. (PMID 25176261)

Medvedev 2015: 30 patients received phenazepam and 40 received it together with Selank. Here is where it gets interesting: the claim is not that adding Selank reduced the dose of phenazepam, but that it reduced phenazepam's side effects, the loss of memory and attention, the sedation, the flattened mood and the sexual dysfunction. (PMID 26356395)

192 patients across the three trials, of whom about half received Selank.

None of these trials used an untreated control group. A control or placebo group is where patients receive an inert substance without their knowledge, and it exists to separate the treatment's effect from everything else that could produce an improvement, like anxiety settling on its own. All three trials compared Selank only to an established drug, which means the researchers cannot say whether two separate interventions each worked, or whether both arms improved for a reason that had nothing to do with either. No abstract mentions a control group, and the Russian full texts were not openly accessible for cross checking.

None of the trials report a dose or a treatment length. Imagine a bread recipe listing flour, water, salt and yeast, and claiming the loaf came out as good as the bakery's, with no amount of anything and no baking time. None of the three trials can be reproduced from what is published, and none of them can show the signature of a real drug effect, which is a response that grows with the dose. The 1350 mcg per day that vendors attribute to the 2008 trial is absent from every primary source I could find.

The combination has not been tested. Selank is usually sold as one half of a pair with Semax, another peptide with its own entry in this reference. Search PubMed for both names and you get 11 articles. I read all of them, and in none of them did researchers give both peptides to the same subjects.

The only study containing both names gives the opposite of what a stack implies. Panikratova 2020 used a brain scan to show how connections between the right amygdala and the rest of the brain change when a person is given Semax, or Selank, or nothing. (PMID 32342318) Three separate arms in 52 healthy subjects, scanned at rest. Not the stack. This is also the only Selank study I found that includes a control group.

The two peptides are chemically related, which is worth knowing and is not an argument for combining them. Selank is Thr-Lys-Pro-Arg-Pro-Gly-Pro and Semax is Met-Glu-His-Phe-Pro-Gly-Pro. They share the last three amino acids. Peptides carrying that ending, given to rats after a stroke, changed the expression of 29 genes at three hours and 57 at 24 hours. (PMID 25850296) That shared tail is a plausible route to a shared mechanism, and it is also a reason two compounds that act alike may not add up to more when taken together. Nobody has checked.

The only combinations that appear in the published literature are with benzodiazepines: diazepam in a rat stress model (PMID 28280289) and the 2015 phenazepam trial above. The effect in both is a reduction of the other drug's side effects.

GABA is the neurotransmitter responsible for damping activity down, and benzodiazepines work by enhancing it, so it is an obvious place to look. Two gene expression studies went there. One used the frontal cortex of rats (PMID 26924987) and the other used human tumour cells in culture, where Selank had no detectable effect on the genes expressed. (PMID 28293190) The only study that looked directly at GABA receptors also used cells in a dish. (PMID 30255741) The single mention of BDNF, a protein that helps neurons survive and connect, is a rat gene expression study from 2008 that PubMed carries no abstract for.

The most cited mechanism has an exposure problem. Selank inhibits the breakdown of enkephalins, the body's own opioid-like signals, halving the activity of the degrading enzyme in a test tube. (PMID 11443939) That was measured at 20 micromolar, which is a concentration in the fluid around the cells and not a dose you can put in a syringe. Converting one to the other needs a volume of distribution that nobody has published for this peptide, so I am not going to pretend the two numbers can be compared. What can be said is that the Russian label describes nasal drops at 1.5 mg per mL and blood clearance within about five minutes, and a test tube experiment lets the researcher choose the incubation time.

Two notes on the registry. First, I searched ClinicalTrials.gov three ways, by intervention, by term and by exact quoted term, and found no Selank trials at all. Second, every record on that registry carries an identifier beginning with the letters NCT, so any NCT number you are shown for a Selank trial is fabricated, and you can check that yourself in about ten seconds. The cluster of fraudulent registrations under the sponsor name Hudson Biotech, one of which carries "(Mock Study)" in its own title, covers several peptides and does not reach this one.

The FDA has never reviewed Selank. It approved Semax for its compounding list in July 2026 on an 8 to 5 vote with one abstention, but Selank was not on the agenda, because the request to review it had been withdrawn in 2024.

The source of doses

Every documented use is intranasal. The injectable solution that is actually sold has never been compared to it in a clinical study.

The Russian instructions are the only official source I could read in the original. Nasal drops at a concentration of 1.5 mg per mL, two drops per nostril, three times a day, for two weeks, with courses repeated every one to three months on a physician's advice. That is 12 drops a day, with no milligram figure stated anywhere. The 900 mcg per day number in circulation comes from assuming 20 drops to a millilitre, which gives 0.075 mg per drop. It is a derivation from an assumed dropper, not a number on the label.

The Russian instructions go further than the marketing does, stating the drug is contraindicated in pregnancy and breastfeeding and not indicated under 18, with no efficacy or safety data in those groups.

The only head to head route comparison I found gave the peptide to two different mouse strains by two different routes and found it effective in only one. (PMID 29787664)

What we do not know

Whether it beats a fake treatment, and at what dose. No placebo controlled trial exists, and no trial reports a dose.

What it does when injected, since every human study used the nose. No safety or blood level data for injection, in any species.

What happens past two weeks. That is where the treatment course ends and where the published record stops.

Whether it interacts with Semax. No study in any species has given both.

My conclusion

The three trials are probably not misleading. Somebody enrolled real patients, gave a real treatment, and used real anxiety scales, and dismissing that work because it is in Russian would be its own kind of error. My problem is with the product being sold, which is not the one that was tested: a different route, a dose with no source I can find, and a second peptide that has never been given alongside it in any study of anything. Translating and publishing those three Russian papers tomorrow would change this entry more than any other source cited in it.

If you have taken Selank and Semax together, where did you first see the claim that the two belong in the same protocol? Because it is not in the studies.

---

This content is for educational and informational purposes only and is not medical advice. Peptides discussed are research compounds and may not be approved for human use. Nothing here should be used to diagnose, treat, cure, or prevent any disease. Always consult a qualified healthcare professional before starting any peptide, supplement, or protocol. Individual responses vary. Do not self-administer compounds without proper medical supervision.