Picture the ordinary version of this problem. It’s a Tuesday, and someone has been staring at the ceiling since eleven-forty. They’ve tried the usual things, cutting the afternoon coffee, putting the phone in another room, and none of it has quite worked. Somewhere around one in the morning, they end up scrolling, and they land on names like DSIP, epithalon, selank, peptides promising the kind of deep, uninterrupted sleep that feels, at that hour, like the only thing that matters.
This piece is for that person, and for anyone else who has wondered whether one of these compounds might help. But it isn’t a buying guide. The honest starting point is a harder one: the human research on these three peptides is so thin that nobody can currently tell you the right dose. That’s not a caveat tucked at the bottom of the page. It’s the whole story, and it changes how a person should think about choosing, dosing, and who they trust to help.
Who this is really for
If you’re someone who has run out of the easy fixes, better sleep hygiene, a calmer evening routine, maybe a conversation with a doctor about anxiety or apnea, and you’re now curious about something further out on the frontier, this is written for you. It’s also written for the person who has already bought a vial from a website and is wondering, quietly, whether they did the right thing. Neither of you will find a magic number here. What you’ll find is a clear look at what’s known, what isn’t, and why the difference matters more than the dose itself.
What the science actually says
Start with DSIP, because it has the most human data of the three, which isn’t saying much. A 1981 study in Experientia gave synthetic DSIP intravenously to six middle-aged people with chronic insomnia and reported longer, higher-quality sleep with fewer interruptions and slightly more REM sleep, with no daytime grogginess. A 1984 trial in European Neurology gave ten injections to seven patients with severe insomnia; sleep normalized in all but one, and the improvement held for three to seven months. A 1987 case report described a single person whose sleep schedule shifted earlier by about five hours over a week of treatment.
Read that again and notice the numbers: six people, seven people, one person. Notice the decade: all of this dates to the 1980s. And notice how it was given: through an IV in a clinical setting, nothing like the at-home injection someone does with a vial that arrived in the mail.
Then there’s what came after those studies. A 2006 review in the Journal of Neurochemistry, bluntly titled “Delta sleep-inducing peptide (DSIP): a still unresolved riddle,” concluded that the idea of DSIP as a genuine sleep factor is “extremely poorly documented and still weak,” and pointed out that scientists have never conclusively identified the gene, protein, or receptor behind it. When the field’s own reviewers are unsure whether the thing does what its name claims, there’s simply no foundation under it for a precise, evidence-based dose. Any number floating around online is someone’s extrapolation, not a documented fact.
Epithalon and selank rest on even less. Epithalon’s sleep story runs indirectly through melatonin: a 2007 study in Advances in Gerontology found that pineal peptides, epithalon among them, helped restore nighttime melatonin release and normalize circadian rhythm in older subjects. That’s meaningful, but it comes largely from a single research group, and nobody has run a controlled trial testing epithalon for sleep quality as the actual outcome being measured. There’s no dose-finding data for sleep at all. Selank, per a 2018 paper in Protein and Peptide Letters, is really an anti-anxiety compound with “prolonged anti-anxiety and nootropic effects” that works through the GABA system. Its dosing literature is about anxiety, not sleep.
So across all three compounds, the foundation a responsible dose would normally rest on is either decades old and built on a handful of patients, or it simply isn’t there.
Why “just tell me the dose” isn’t a fair question to ask
Here’s the uncomfortable truth: because the evidence doesn’t define a correct sleep dose, deciding on one is an act of clinical judgment made under real uncertainty, not something anyone can look up. That’s a decision that belongs with a licensed clinician who understands a person’s full medical picture, not something to be reverse-engineered from a Reddit thread or a supplement forum.
This reframes the whole question a person should be asking. It isn’t “which peptide, and how many micrograms.” The literature can’t answer that with any confidence, and pretending otherwise does a disservice to anyone trying to make a careful decision. The real question is whether the compound reaches someone through a process with actual oversight, or through one with none. Self-selecting a dose from an unverified vial, for a benefit that’s never been demonstrated to modern standards, is a very different act than having a clinician weigh the decision, watch how you respond, and step in if something isn’t right.
What real oversight actually looks like
It helps to know what good oversight looks like in practice, because it doubles as something you can hold any provider up against.
It starts before any prescription is written, by asking about the ordinary, fixable reasons someone might be sleeping badly: caffeine and alcohol timing, an irregular schedule, stress, screens before bed, or an undiagnosed condition like obstructive sleep apnea. Reaching for an experimental peptide before those questions have been asked is poor practice, no matter what the peptide is.
It’s also honest, plainly and out loud, about how thin the evidence is. Nobody doing this responsibly will describe DSIP, epithalon, or selank as proven sleep cures. A provider willing to say the data are old, preliminary, or measuring something other than sleep, and that none of these three carries FDA approval for sleep, is showing you exactly the kind of candor that separates real care from a sales pitch.
It sources any compounded version of these peptides through a licensed compounding pharmacy, under an actual prescription, rather than as a research chemical shipped from a warehouse. Here’s the regulatory fact that belongs squarely in this conversation: the FDA states plainly that compounded drugs are not FDA-approved, meaning the agency does not review their safety, effectiveness, or quality before they reach anyone. Oversight can’t erase that. What it adds is a licensed clinician and a licensed pharmacy standing inside a process that a checkout page simply doesn’t have.
And it follows up. Because nobody can predict from the existing research how any one person will respond, structured check-ins are the only real way to notice if something is working, or isn’t, or is going wrong. That follow-up step is exactly what’s missing from the version of this story that ends at “add to cart.”
What a week of tracking actually looks like
This is the part that matters more here than it would almost anywhere else, and it’s worth picturing concretely, because “monitoring” can sound abstract until you imagine the actual week.
Say someone starts using one of these compounds under a clinician’s care. Monday night, sleep comes fast and feels deep. Tuesday, it takes ninety minutes and there are three wake-ups. Wednesday feels great again. By the weekend, if that person is going purely on memory, they’ll probably tell you it’s “working,” because the good nights are the ones that stick. That’s not dishonesty, it’s just how memory works, and it’s exactly why relying on a gut feeling is a poor way to judge an unstudied compound.
The better version of that week involves a simple log: the dose taken, the bedtime, how long it took to fall asleep, how many times sleep broke, and how the next day actually felt. A person walking into a follow-up appointment with two or three weeks of that kind of record has something real to hand their clinician, rather than a vague impression shaped by whichever nights happened to be memorable. Some supervised providers build tools for exactly this. FormBlends, for instance, offers a tracker app for logging doses and symptoms, not a prescription pad and not a checkout button, meant to make sure follow-up is grounded in an actual record instead of recollection. The specific app matters less than the habit it represents: when the science can’t tell you what to expect, careful tracking under a clinician’s eye is one of the few genuinely useful things available, and it’s the piece that’s structurally absent when someone sources this on their own.
How to actually go about this
Put all of it together and a clear way to approach the decision emerges, even though the science itself can’t hand you a peptide to pick.
Start by accepting that “which sleep peptide works best” doesn’t have an evidence-based answer, because the head-to-head trials that would settle it don’t exist. Then shift the question you’re actually asking to something the evidence can address: who is going to oversee this, and will they tell me the truth about what’s known? Look for a process where a licensed clinician screens for the ordinary causes of poor sleep first, speaks plainly about how preliminary the evidence is, sources any compounded preparation through a licensed pharmacy, makes the dosing call rather than handing it to you, and checks in afterward. A supervised, prescription-based provider like FormBlends is built around that kind of process. It won’t conjure the large human trials that don’t exist, and it doesn’t make these compounds proven. What it does is put the dosing decision and the follow-up in the hands of someone qualified to hold them, which, given how little the dosing literature actually establishes, is the piece of this the evidence genuinely supports caring about.
The alternative, ordering a research chemical and dosing it alone, asks an untrained person to make, by themselves and without any verification, precisely the judgment call the published literature is too thin to guide. Set against everything above, that’s a hard choice to defend.
The honest bottom line
Nothing in the literature on DSIP, epithalon, or selank defines a correct sleep dose. The studies that would establish one are small and decades old, or indirect, or measuring anxiety instead of sleep. That gap is the central fact here, and it means responsible use, to whatever degree these compounds are used at all, depends far more on the quality of the oversight around them than on any number written on a label. Good oversight screens first, tells the truth about what’s known, sources through a regulated pharmacy, makes the dosing decision clinically, and tracks what actually happens. None of that turns these peptides into proven sleep treatments. But it’s the difference between facing an unstudied compound with a qualified person and a real record, or facing it with a guess and a vial. That’s the distinction the evidence actually supports worrying about.
Questions people ask
Is there an established dose for DSIP, epithalon, or selank for sleep? No. There is no standardized, evidence-based human sleep dose for any of the three in the published literature. The DSIP studies people cite involved six patients, seven patients, and one patient, all from the 1980s, using intravenous delivery in a clinic rather than at-home self-injection. Epithalon has no controlled sleep trials, and selank’s dosing data concern anxiety, not sleep. Any figure presented as “the dose” is someone’s extrapolation, not an established fact.
Why can’t a doctor just tell me the right amount? Because the evidence that would define a correct sleep dose doesn’t exist, the decision has to be made under real uncertainty, weighing a person’s history, other medications, and individual risk. That’s a judgment call for a licensed clinician who can also watch how someone responds and adjust course, not something to calibrate against an anonymous forum post.
Are these peptides approved by the FDA for sleep? No. None of DSIP, epithalon, or selank is FDA-approved as a sleep treatment. Where they’re prepared as compounded medications, the FDA is explicit that compounded drugs are not FDA-approved, meaning the agency doesn’t review their safety, effectiveness, or quality before they reach anyone. Oversight doesn’t change that underlying fact; what it adds is a licensed clinician and pharmacy inside the process.
What should a responsible provider do before prescribing anything? Screen first for the ordinary, fixable reasons someone might be sleeping poorly, things like caffeine and alcohol timing, an irregular schedule, stress, screens before bed, or a condition like sleep apnea, before ever considering an experimental peptide. Be honest that the human evidence is preliminary, old, or measuring something other than sleep. Source any compounded preparation through a licensed pharmacy under a real prescription, make the dosing decision clinically, and follow up.
Why does tracking matter so much with these compounds specifically? Because the effect is uncertain and largely unstudied, gut feelings aren’t reliable. People remember the good night and forget the three mediocre ones. Logging the dose, bedtime, time to fall asleep, wake-ups, and how the next day felt turns a fuzzy impression into an actual record, so any decision to continue, adjust, or stop rests on something solid. A logging tool like the FormBlends tracker app, which is a place to record doses and symptoms rather than a prescription or a checkout, is built so follow-up runs off that record instead of memory.
If the comparison data just doesn’t exist, how should someone decide? Accept that “which sleep peptide works best” has no evidence-based answer, because the trials that would compare them head to head don’t exist. Ask instead who will oversee the process and whether they’ll be straight with you about the evidence. Favor a supervised, prescription-based path where a clinician checks for the ordinary causes first, is candid about how thin the data are, sources through a licensed pharmacy, makes the dosing call, and follows up. That path doesn’t make these compounds proven, but it puts the dosing and the monitoring with someone equipped to handle them.
Is it safe to use peptides for sleep at all?
The honest answer is that safety data on most sleep peptides is thin, particularly in humans. DSIP and selank have some early research behind them, but nothing close to a long-term safety profile exists yet. Small studies have noted things like injection site irritation, grogginess, and shifts in blood pressure. Anyone thinking about trying one of these should be doing it alongside a physician who can monitor labs and adjust dosing, not sourcing it on their own.
Do these peptides actually work, or is it mostly hype?
The picture is genuinely mixed. Some peptides, especially GHRH analogs that stimulate growth hormone release, have shown measurable effects on slow-wave sleep in small clinical studies. Others, DSIP included, have promising animal data paired with inconsistent human results. Calling any of them proven sleep aids overstates what the research actually supports. There’s real promise here, but promise isn’t the same thing as reliability.
How do doctors decide which peptide might be worth considering?
There’s no single “best” option, because the right consideration depends on why someone’s sleep is disrupted in the first place. GHRH analogs sometimes come up when poor slow-wave sleep and low growth hormone appear together. Selank draws interest for anxiety-driven insomnia because of its calming mechanism. A physician-supervised compounding pharmacy like FormBlends starts by reviewing labs and personal history rather than defaulting to the same compound for everyone who walks in.
Where is it actually safe to get these, and what’s risky about the alternative?
Ordering from research-chemical sellers or supplement marketplaces carries real risk: mislabeled concentrations, contamination, and no one accountable if something goes wrong. The legitimate route runs through licensed compounding pharmacies operating under physician supervision and state pharmacy board oversight. That path costs more and asks for a real clinical relationship, but it’s also the one with actual quality control and a person responsible for what ends up in the vial.
References
- Schneider-Helmert D, Schoenenberger GA. The influence of synthetic DSIP on disturbed human sleep. Experientia. 1981;37(9):913-917. Synthetic DSIP intravenously in six chronic insomniacs produced “longer sleep duration and a higher quality of sleep with fewer interruptions; slightly more REM-sleep, but no day-time sedation or other side effects.” https://pubmed.ncbi.nlm.nih.gov/7028502/
- Kaeser HE. A clinical trial with DSIP. European Neurology. 1984. Seven severe-insomnia patients, ten DSIP injections, sleep normalized in all but one, sustained over three to seven months. https://pubmed.ncbi.nlm.nih.gov/6391926/
- Case report: DSIP in delayed-sleep-phase insomnia with benzodiazepine dependence advanced the main sleep phase about five hours over one week. Deutsche Medizinische Wochenschrift. 1987.
- Kovalzon VM, Strekalova TV. Delta sleep-inducing peptide (DSIP): a still unresolved riddle. Journal of Neurochemistry. 2006;97(2):303-309. The sleep-factor hypothesis is “extremely poorly documented and still weak”; gene, protein, and receptor never conclusively identified.
- Korkushko OV, Lapin BA, Goncharova ND, Khavinson VKh, Shatilo VB, et al. Pineal peptides and the daily melatonin rhythm. Advances in Gerontology. 2007;20(1):74-85. Pineal peptides including Epitalon “recover night release of endogenous melatonin and lead to the normalization of the hormone circadian rhythm” in old monkeys and elderly people.
- Vyunova TV, Andreeva L, Shevchenko K, Myasoedov N. Peptide-based Anxiolytics: heptapeptide Selank. Protein and Peptide Letters. 2018;25(10):914-923. Selank “exhibits prolonged anti-anxiety and nootropic effects” and modulates the GABA system, an anxiolytic rather than a hypnotic.
- U.S. Food and Drug Administration, Understanding the Risks of Compounded Drugs.; the agency does not review their safety, effectiveness, or quality before they are marketed.
Written by Emil Costa, science writer. Checking each figure against the cited source. Last reviewed January 2026.
Informational content, not medical direction. Your doctor should approve any new treatment.
