Most classes in this field are limited by how little research exists. The cognitive peptides have the opposite problem: a substantial body of research exists, and almost none of it can be evaluated the way Western medicine evaluates evidence.
Semax and Selank were developed at the Institute of Molecular Genetics in Moscow and have been used clinically in Russia for decades — Semax is a registered medicine there, indicated for cerebrovascular and cognitive conditions. Selank, an analogue of the immune peptide fragment tuftsin, has been studied as an anxiolytic.
Both have plausible pharmacology. Semax derives from a fragment of ACTH (residues 4–10) modified for stability, and the published mechanism concerns effects on BDNF and monoaminergic signalling. Selank's proposed mechanism runs through GABAergic and enkephalin systems. These are not invented stories; they are documented research programs with decades behind them.
Why the evidence is hard to weigh
Four features of this literature make it resistant to the standard reading.
Language and venue. Much of the primary work is published in Russian-language journals with limited indexing in Western databases. What reaches PubMed in English is a subset, often review-level rather than primary trial data.
Trial design and reporting conventions. Studies in this tradition frequently report smaller samples, and details Western readers expect — randomization procedure, blinding, pre-registered primary endpoints, effect sizes with confidence intervals — are inconsistently reported. That does not make the results wrong. It makes them difficult to appraise, which is a different problem with the same practical consequence.
Almost no independent replication. The compounds have not been through Western regulatory programs, so there is no FDA or EMA review file, no phase 3 dataset, and very little work by groups outside the originating research tradition.
Endpoint variety. Cognitive and anxiolytic endpoints are among the hardest to measure well, most sensitive to placebo effects, and most dependent on blinding — exactly the methodological features that are hardest to verify here.
What this does and doesn't mean
It would be lazy to write these compounds off as pseudoscience. They are prescribed medicines in a country with a real pharmaceutical research establishment, with published mechanistic work behind them. Absence of Western validation is not evidence of absence.
It would be equally lazy to treat "used clinically in Russia for thirty years" as equivalent to "demonstrated in randomized controlled trials." It isn't. Regulatory traditions differ in what they require, and the standard that produced the incretin dataset — thousands of participants, blinded, pre-registered, independently scrutinised — is simply not the standard behind these compounds.
The honest position sits in neither corner: a real research program whose evidence has never been tested by the system most readers implicitly assume when they hear "clinically used."
That's an uncomfortable summary. It's also the accurate one, and this class is a good reminder that "there's research on it" and "it's been demonstrated" are two different sentences.
Research use only. No compound discussed here is FDA-approved for any indication, and nothing on this page is medical advice.
For research use only · Not for human or veterinary use · No compound discussed here is FDA-approved for any indication