Anxiety and sleep peptides compared — Selank, DSIP, oxytocin, low-dose Sermorelin

You've tried sleep hygiene. You've tried the CBT-I protocol or at least parts of it. Maybe you've tried melatonin at various doses with diminishing results. You've been cautious about benzodiazepines or z-drugs because you understand the dependency risk and the architecture disruption, and you don't want to solve a short-term problem by creating a longer-term one. You've encountered four names — Selank, DSIP, oxytocin, low-dose Sermorelin — described as peptides that might work on the anxiety-sleep axis through upstream mechanisms. You want to understand what they actually do and where the evidence actually sits before you bring this to a prescribing provider.

Start with Selank, because it's the most directly anxiolytic of the four and has the most consistent mechanistic story. Selank is a synthetic analog of tuftsin, a naturally occurring tetrapeptide involved in immune regulation, with modifications that extend its stability and enhance its CNS activity. It was developed in Russia — specifically at the Institute of Molecular Genetics in Moscow — as part of a search for anxiolytics that didn't carry the dependency, sedation, and cognitive impairment profile of benzodiazepines. The mechanistic interest centers on GABA-A receptor modulation and neuropeptide Y. Selank appears to enhance GABA-A function — which is why it sits in anxiolytic territory — but via a different binding site than benzodiazepines, without direct GABA-A agonism, and without the associated sedation, muscle relaxation, and tolerance development that characterize benzos. It also appears to modulate neuropeptide Y expression, which is relevant because neuropeptide Y is a key regulator of stress resilience, fear extinction, and the HPA axis response to threat. Low neuropeptide Y tone is associated with anxiety, PTSD, and impaired recovery from acute stress.

What does the evidence look like? Selank has been used in Russian clinical practice for decades and has a published evidence base — small trials, mostly conducted in Soviet-era and post-Soviet institutions, examining effects on anxiety scores, cognitive performance under stress, and some neurophysiological measures. The studies are real and peer-reviewed in their contexts, but they haven't been replicated at scale in Western research settings with the methodological standards that FDA review would require. Independent large-sample RCTs are lacking. This is the same caveat that applies to Semax and a number of other Russian-developed peptides: the research tradition is legitimate, the mechanism is plausible, and the independent replication is not where you'd want it to be before drawing hard clinical conclusions. Selank is not FDA-approved. It exists in compounded forms in the United States.

The practical case for Selank in the anxiety-sleep context is that it addresses the upstream condition — the anxious arousal that prevents sleep onset and drives 3 a.m. waking — without sedating you directly. It doesn't put you to sleep the way a z-drug does. What users and clinicians who work with it describe is a reduction in the cognitive and physiological noise that makes sleep impossible: less rumination, less physiological hyperarousal, more capacity to actually settle. Whether that effect translates broadly and reliably in people outside the study populations is genuinely uncertain. But the mechanism addresses the right problem.

DSIP — Delta Sleep-Inducing Peptide — is older in the research literature and has a more complicated evidentiary story. It was identified in the 1970s from rabbit brain perfusate, named for its apparent ability to induce delta (slow-wave) sleep when administered to rabbits. The subsequent decades of research produced inconsistent results: some human studies found effects on sleep latency and sleep architecture; others found minimal effect. The molecule is small, has poor stability in plasma, and the question of whether it crosses the blood-brain barrier reliably in humans — rather than in specific experimental conditions — is unresolved. DSIP also appears to act on the HPA axis, with some evidence of cortisol-curve modulation: shifting the morning cortisol peak and reducing the HPA hyperreactivity that drives anxiety-type insomnia. This is the more interesting hypothesis for the anxiety-sleep overlap: not direct sedation but normalization of the stress-hormone rhythm that underlies disrupted sleep. The evidence for this specific mechanism in humans is limited and mostly older. DSIP has not been developed through formal clinical trials in ways that would clarify its clinical utility, and it's not FDA-approved. The honest position is that DSIP's research legacy is real but inconsistent, and the human data is thinner than its name and reputation suggest.

Oxytocin is the most discussed of the four in mainstream contexts, which creates its own confound: the popular framing as the "bonding hormone" or "love hormone" has outrun the actual clinical research considerably. Oxytocin does have genuine mechanistic relevance to the anxiety-sleep problem. It modulates amygdala reactivity — the brain region that generates fear and threat responses — and appears to reduce HPA axis activation, lower cortisol, and shift the nervous system toward parasympathetic tone. Intranasal oxytocin has been studied for social anxiety, PTSD, and broadly for anxiolytic effects. The early research produced promising results; subsequent meta-analyses have been more sobering. Effect sizes across studies are smaller and less consistent than the original reports suggested, particularly for general anxiety rather than socially-mediated anxiety specifically. Context effects are pronounced: oxytocin appears to modulate emotional salience rather than producing uniform anxiolysis, which means the same dose can produce different effects depending on social context, baseline attachment style, and other variables that are hard to control in research and even harder to predict in individual use. Intranasal oxytocin is investigational — it's not FDA-approved for anxiety or sleep indications. Some prescribing providers use compounded intranasal oxytocin; the clinical literature supporting this is preliminary, and the translation from research studies to real-world use is not straightforward.

The sleep-specific case for oxytocin is indirect. Better social-emotional regulation, reduced amygdala hyperreactivity, and lower evening cortisol are all conditions that support sleep. But oxytocin is not a sleep-promoting compound in the direct sense that DSIP was supposed to be or that a z-drug undeniably is. Its potential contribution is upstream — supporting the nervous system state that makes sleep possible — which means its utility depends heavily on whether anxiety-driven HPA activation is actually the proximate cause of the sleep disruption. For some people, it is. For others, the insomnia has architectural causes that social-emotional regulation won't fix.

Low-dose Sermorelin occupies a different lane entirely. Sermorelin is a GHRH analog — it stimulates the pituitary to release growth hormone by mimicking the hypothalamic signal. The relevance to sleep comes from two distinct mechanisms. First, GHRH itself has direct somnogenic properties: it promotes slow-wave sleep independently of its effect on GH release, acting on GHRH receptors in the brain. Second, the GH pulse that Sermorelin may support preferentially occurs during slow-wave sleep, and the GH pulse itself appears to deepen that slow-wave episode — creating a positive feedback loop between GH secretion and sleep depth. This is not an anxiolytic mechanism. Sermorelin doesn't address rumination or HPA hyperreactivity. What it may address is the architectural quality of sleep — specifically slow-wave depth — that has declined with age, independent of anxiety. If your sleep is architecturally shallow rather than anxiety-disrupted — if you're getting hours of sleep but not waking restored, if slow-wave seems to be the missing piece rather than arousal being the problem — the Sermorelin mechanism is pointing at the right target in a way that Selank, DSIP, and oxytocin are not. These are different problems. Knowing which one you have matters for knowing which compound belongs in the conversation.

None of these four compounds should be positioned as equivalent in efficacy to benzodiazepines or z-drugs for acute sleep disruption. Benzos and z-drugs are pharmacologically potent, fast-acting, and reliable in the short term. The question of whether to use them is a clinical conversation with real trade-offs — dependency risk, architectural disruption, rebound insomnia, cognitive effects — not a reason to avoid medication altogether. The compounds discussed here work on upstream systems, through mechanisms that produce subtler effects, over longer timeframes, with less robust human evidence for their efficacy. The ceiling is lower. The floor is also different: no dependency risk from Selank, no cognitive architectural disruption from Sermorelin, no hangover from DSIP. For people who need the acute efficacy of a z-drug, these aren't the answer. For people who've moved through the acute phase and are looking for something that might support the system that generates sleep, rather than forcing sleep pharmacologically, the conversation is different.

Before any of these compounds, the foundational interventions deserve honest emphasis — because the evidence base for them is better than for anything discussed above. CBT-I (cognitive behavioral therapy for insomnia) has larger and more consistent effect sizes than any pharmaceutical for chronic insomnia and doesn't carry dependency risk. Sleep consolidation protocols, stimulus control, consistent wake times, light management — the effect sizes in the clinical literature are real and not modest. These aren't consolation prizes for people who can't access medications. They are, by the evidence, the highest-value interventions available. The peptides discussed here are potential additions to a picture where those fundamentals are already in place, evaluated with a prescribing provider who understands your specific clinical situation — not substitutes for them, and not the first line of anything.

The anxiety-sleep loop is a real and specific clinical problem, not just a lifestyle complaint. Getting clarity on whether your presentation is primarily anxiety-driven, architecturally shallow, or something else entirely — ideally with objective sleep data — is the kind of evaluation that makes the peptide conversation meaningful rather than speculative. What you're looking for is a prescribing provider who can look at the whole picture, not just the symptom you came in with.