Oxytocin vs Selank
Oxytocin is a nine-amino-acid neuropeptide synthesised in the hypothalamus, used intranasally since the mid-2000s as a research tool for probing social cognition, threat processing and affiliative behaviour, and now the subject of one of the largest and most contested replication debates in human neuroscience. Selank is a synthetic heptapeptide, Thr-Lys-Pro-Arg-Pro-Gly-Pro, built from the immunomodulatory tetrapeptide tuftsin and developed at the Institute of Molecular Genetics in Moscow as an anxiolytic. The two are set side by side because both are given intranasally and both are promoted for mood, anxiety and social comfort. One published experiment applied both compounds within a single assay, an unbiased screen of 109 neuropeptides against acid-sensing ion channels, and neither showed activity. No study has compared their behavioural or clinical effects in any species, and the two evidence bases are separated by roughly an order of magnitude in size, scrutiny and geography.
They have been compared directly.
One published experiment tested both compounds under identical conditions. Vyvers and colleagues screened 109 short neuropeptides, oxytocin and Selank among them, against five homomeric acid-sensing ion channels expressed heterologously and assayed by electrophysiology. Neither peptide was identified as a direct agonist of any channel, and neither appeared among the three modulators the screen did detect, which were dynorphin A, YFMRFamide and endomorphin-1. That is a shared negative result on one ion-channel family rather than a comparison of anything either compound is used for, and it says nothing about anxiety, social cognition or mood. Beyond that screen, the only publication engaging both is a 2022 review of Russian peptide drug development, which listed Selank with its full sequence as an anxiolytic and named oxytocin among the peptide hormones synthesised as prohormones, placing them in separate categories without comparison. No study has administered both to an animal or a human, and no behavioural, imaging or clinical endpoint has ever been measured for the two under one protocol.
The 2 studies that tested both
- 01In vitro2018
Screening of 109 neuropeptides on ASICs reveals no direct agonists and dynorphin A, YFMRFamide and endomorphin-1 as modulators
Vyvers A, et al. · Scientific Reports · unbiased electrophysiological screen of 109 short neuropeptides, including oxytocin and Selank, applied to five homomeric acid-sensing ion channels: ASIC1a, ASIC1b, ASIC3, ASIC4 and BASIC
Both oxytocin and Selank were among the peptides applied to the channel panel. Neither was identified as a direct agonist of any channel tested, and neither was among the three modulators the screen detected. The only modulatory effects found were dynorphin A on ASIC1a, YFMRFamide on ASIC1b and ASIC3, and endomorphin-1 on ASIC3. The authors concluded that most short peptides acting as ligands of acid-sensing ion channels are already known.
- 02Review2022
Development of Peptide Biopharmaceuticals in Russia
Deigin VI, et al. · Pharmaceutics · review of the history and current state of peptide drug development in the Russian Federation, with a catalogue of original Russian peptide preparations
The review listed Selank among original Russian peptide preparations, giving its full sequence and classifying it as an anxiolytic. Oxytocin was named separately, among the peptide hormones synthesised as prohormones and among the endogenous peptides involved in regulating emotional state, sexual behaviour and sleep. The two compounds appeared in different sections and were not compared with each other.
Side by side.
| Oxytocin | Selank | |
|---|---|---|
| Evidence maturity | Clinical | Early clinical |
| Studies cited here | 21 | 15 |
| Published 2023 or later | 16 | 0 |
| Newest paper | 2026 | 2022 |
| Sequence and origin | A nine-residue endogenous neuropeptide with a disulfide bridge, synthesised in the hypothalamus and released by the posterior pituitary, with established physiological roles in parturition and lactation independent of any research use. | A seven-residue synthetic peptide, Thr-Lys-Pro-Arg-Pro-Gly-Pro, derived from the endogenous tetrapeptide tuftsin with a Pro-Gly-Pro extension added to resist enzymatic degradation. It has no endogenous counterpart of its own. |
| Mechanism as described in the literature | Acts through a characterised G-protein-coupled oxytocin receptor. Neuroimaging syntheses describe sex-dependent amygdala modulation and a thalamus-striatum-insula circuit enriched for CD38, OXT and OXTR expression as the most reproducible neural signature. | No specific receptor has been identified. Proposed mechanisms are inhibition of enkephalin-degrading enzymes, allosteric interaction with the GABAergic system inferred from gene-expression panels, and regulation of BDNF and inflammation-related genes. |
| Size of the evidence base | Twenty-one verified publications in the research entry here, of which ten are systematic reviews or meta-analyses and nine are human clinical trials. The underlying corpus those syntheses draw on runs to hundreds of administration studies. | Fifteen verified publications: five human clinical studies, five rodent in vivo studies, three in vitro studies and two reviews. There is no meta-analysis of Selank at all. |
| Human data | Extensive and frequently null. A 24-week placebo-controlled phase 2 trial in 290 children and adolescents with autism found no difference from placebo, a multisite alcohol use disorder trial found no difference in heavy drinking days, and a 2026 meta-analysis of 42 randomised trials pooling 1922 participants found a small non-significant overall effect with a small significant signal only in schizophrenia spectrum disorders. | Five studies, all Russian: comparative trials against medazepam and phenazepam in generalised anxiety disorder, neurasthenia and somatoform disorders, an immunological study in anxiety-asthenic patients, and one placebo-controlled resting-state fMRI study in 52 healthy participants. |
| Evidence maturity | Clinical. Large registered randomised trials and preregistered meta-analyses exist, and the field's central problem is now the size and reproducibility of effects rather than their absence of study. | Early clinical. Controlled trials exist but used active comparators rather than placebo, were conducted in one country, and were never followed by a phase III programme. |
| Recency of primary research | Highly active. Sixteen of the twenty-one publications in the entry are dated 2023 or later, including meta-analyses and randomised trials published in 2026. | Dormant. No primary Selank research article has been indexed in PubMed since 2022, and nothing in the entry is dated 2023 or later. |
| Delivery to the brain | Unresolved and openly debated. The proportion of an intranasal dose reaching central oxytocin receptors has never been directly quantified in humans, and a widely cited critical review argued the quantity is very small relative to endogenous release. | Also unquantified. Intranasal administration produced measurable central changes in rats, including altered hippocampal BDNF expression, and altered resting-state connectivity in humans, but no pharmacokinetic characterisation of central exposure exists. |
| Regulatory status | An approved drug for obstetric indications in most jurisdictions. There is no approved indication anywhere for cognitive, social or emotional enhancement, and the largest trial in autism was fully null. | Registered in Russia as an anxiolytic and not approved in any Western jurisdiction. A 2020 analytical study identified Selank among cognitive-enhancing research peptides present in seized pharmaceutical preparations. |
And what it does not.
The single experiment that tested both compounds screened them against acid-sensing ion channels and found nothing for either, which rules out one narrow mechanism and settles nothing else. Everything beyond that is inference across two literatures that are not comparable in scale. Oxytocin has been administered in hundreds of human studies, synthesised in preregistered meta-analyses, and tested in large registered randomised trials, and the resulting picture is one of small, heterogeneous and frequently null effects, with a definitive negative trial in autism and a null multisite trial in alcohol use disorder. Selank has five human studies, all Russian, mostly against benzodiazepine comparators rather than placebo, with no meta-analysis, no independent replication and no primary research published since 2022. Neither compound has quantified central exposure after intranasal dosing, and neither has an approved indication for mood or cognitive enhancement. The asymmetry between them is one of scrutiny: oxytocin's limits are documented, and Selank's are simply unexamined.
Common questions.
- What was the one experiment that used both oxytocin and Selank?
It was an ion-channel screen rather than a behavioural study. Vyvers and colleagues applied 109 short neuropeptides, including both compounds, to five homomeric acid-sensing ion channels expressed in a heterologous system and recorded the currents. Neither oxytocin nor Selank acted as a direct agonist or as a modulator of any channel tested. The screen identified only dynorphin A, YFMRFamide and endomorphin-1 as modulators. The shared result is a negative one on a single receptor family and carries no implication for anxiety or social behaviour.
- Which compound has stronger human evidence for an anxiety-related effect?
Neither has strong evidence, and the weakness is different in each case. Selank has Russian comparative trials in diagnosed anxiety disorders reporting effects broadly similar to medazepam and phenazepam, but those used active comparators rather than placebo, came from a small group of laboratories, and were never replicated independently. Oxytocin has been examined far more rigorously and the results have been largely disappointing: a 2026 meta-analysis of 42 randomised trials found a small non-significant pooled effect across mental disorders, with a small significant signal only in schizophrenia spectrum disorders.
- Do the two peptides share a receptor or pathway?
No shared target has been demonstrated. Oxytocin acts through a characterised G-protein-coupled oxytocin receptor, and neuroimaging syntheses locate its most reproducible effects in a thalamus-striatum-insula circuit enriched for CD38, OXT and OXTR expression. Selank has no identified receptor; its proposed actions are inhibition of enkephalin-degrading enzymes, allosteric interaction with the GABAergic system and regulation of BDNF-related gene expression. The one experiment that applied both to the same target family, acid-sensing ion channels, found activity for neither.
- Why does the oxytocin literature look so much less certain than it did a decade ago?
Because the field applied larger samples and preregistration to findings originally reported in small studies. The foundational trust and emotion-recognition experiments have not been consistently reproduced, a multiverse meta-analysis of 530 effect sizes from 185 studies produced summary estimates ranging widely depending on analytic choices, and a preregistered meta-analysis of non-social executive function found no overall significant effect. Publication-bias assessments have indicated that some remaining positive effect sizes, including those in autism, may be inflated.
- Is the intranasal route better characterised for one compound than the other?
It is better argued for oxytocin, not better established. The quantity of oxytocin reaching central receptors after intranasal dosing has never been directly measured in humans, and a critical review in Biological Psychiatry contended that it is very small relative to endogenous release. For Selank there is no equivalent pharmacokinetic debate because there is no pharmacokinetic dataset; central effects have been inferred from changes in rat hippocampal BDNF expression and from one human resting-state connectivity study.
For research purposes only. Not for human consumption, diagnosis, treatment, or prevention of any condition. Nothing on this page is medical advice, and no comparison here should be read as a recommendation of either compound.