DSIP (Delta Sleep Inducing Peptide): What It Is, Where to Get It Prescribed, and What It Costs

DSIP is a nonapeptide — nine amino acids, sequence Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu. In the early 1960s Marcel Monnier's group in Basel reported that blood from the brain of a sleeping rabbit could induce sleep-like EEG patterns in a second rabbit. The factor was isolated, sequenced and synthesised by 1977, and the Proceedings of the National Academy of Sciences paper that year named it: infused into rabbit brains, the synthetic peptide was the only one of nine related sequences that increased delta-wave and spindle activity.

It was proposed as a circulating "sleep factor." It is not a hormone, not a steroid, and not related to melatonin. Research describes it as a modulator of sleep architecture and of the stress response, not a substance that forces sleep.

DSIP is a laboratory name. The World Health Organization assigned the International Nonproprietary Name emideltide in 1993, and that is the name the FDA uses. Its 2026 review covered two forms, the free base and the acetate salt, because the nominations did not make clear which was proposed.

It is not a sedative. In the first healthy-volunteer study, subjects felt a "pressure to sleep" during the infusion but two hours later showed no sleepiness and no impaired performance. It is not a treatment for sleep apnoea or any sleep problem with a breathing cause, and it is not a growth hormone secretagogue. And it is not a proven treatment for insomnia.

The defining observation, repeated in rabbits, rats and cats from 1977 onward, is more slow delta-wave and spindle power — the EEG hallmarks of deep non-REM sleep — after DSIP is given into the brain or systemically. It was not universal: one 1980 rat study found no change, and the original group described bell-shaped dose curves. The Neuroscience and Biobehavioral Reviews review of 1984 lays this out, inconsistencies included.

In rats, the opioid antagonist naloxone blocked DSIP's ability to prolong sleep. Cell work found DSIP does not bind opioid receptors directly; it was instead reported to trigger calcium-dependent release of enkephalins from neurons in several brain regions. This is the mechanism the 1980s withdrawal studies were built on. It is an indirect action, not a direct opioid effect.

A 1985 paper in Neuroendocrinology reported that intravenous DSIP reduced corticosterone release triggered by corticotropin-releasing factor in rats, pointing to an action at the pituitary. A 2001 paper in Stress found DSIP reduced stress-induced Fos activation in the hypothalamus and septum of rats. This is where the "stress response" language comes from. It is animal data.

Animal studies describe DSIP crossing the blood–brain barrier. It also disappears fast: a half-life of minutes in animals and about eight minutes in human plasma, because blood peptidases clip it apart. That is why the human researchers found the infusion itself mattered — a one-minute infusion did nothing, and the sleep effect appeared in the second hour. Any effect is a signalling event, not a drug level held through the night.

Nearly fifty years on, no gene encoding DSIP and no DSIP receptor has been identified. A 2006 review in the Journal of Neurochemistry called the sleep-factor hypothesis "extremely poorly documented and still weak." The mechanism is a working hypothesis, not a settled pathway.

This is where the human data sits, and it is unusual for a compounded peptide: there are randomised, double-blind, placebo-controlled studies in people. They are small, old, and split.

The first appeared in the Lancet in 1981, with a fuller report that year in Experientia: six middle-aged chronic insomniacs received an intravenous infusion or placebo before bed in a crossover design, and the authors reported longer sleep with fewer interruptions and no daytime sedation. A 1986 paper in European Neurology summarised further double-blind work, and a 1987 study in the same journal treated fourteen chronic insomniacs on seven consecutive nights and reported better night sleep and daytime alertness — but against matched normal sleepers rather than a placebo arm, which the FDA singled out as a weakness.

Two independent groups did not agree. A 1987 crossover study in the International Journal of Clinical Pharmacology Research found no significant difference from placebo nights and called the improvement "of little clinical significance." A 1992 parallel-group study of sixteen chronic insomniacs in Neuropsychobiology found higher sleep efficiency but judged the effects weak and concluded short-term treatment "is not likely to be of major therapeutic benefit." Neither found slow-wave sleep itself increased.

The FDA counted roughly 209 people across all this work, all treated intravenously, none after 1992.

One single-patient case report from 1984 described a man with narcolepsy whose sleep attacks became less frequent during weeks of intravenous dosing. A 1984 pilot study in European Neurology treated seven chronic pain patients and reported lower pain ratings in six, with no control group. A single case and an uncontrolled pilot are hypotheses, not evidence.

A 1984 European Neurology paper from Geneva gave intravenous DSIP to 107 inpatients in alcohol or opiate withdrawal and reported symptoms resolving or improving in the large majority, with insomnia often returning afterwards. A 1998 open trial in the Journal of Clinical Psychopharmacology treated seven opioid-dependent patients and reported withdrawal scores falling after the first injection, but only two completed the schedule. Both were open-label with no control group, and both sets of authors said so. This is not a use Pepti prescribes for.

A 2005 paper in Seizure reported that DSIP and an analogue reduced chemically provoked seizures in rats. A 2003 study in Mechanisms of Ageing and Development gave female mice lifelong monthly courses of a DSIP-plus-glycine preparation and reported fewer chromosome aberrations, a longer maximum lifespan and fewer spontaneous tumours; the FDA's reviewers noted glycine alone has reported anti-tumour properties, so it cannot be attributed to DSIP.

This is the honest part. There are no large randomised controlled trials of DSIP in humans, and there are no randomised human trials of subcutaneous DSIP at all. What exists is a cluster of small, mostly double-blind, placebo-controlled intravenous studies in chronic insomniacs from 1981 to 1992, in which the originating group reported benefit and two independent groups reported effects too weak to matter clinically, plus uncontrolled series in withdrawal and pain.

That is more human data than many compounded peptides have. But the populations are tiny, the route differs from how it is prescribed today, and nobody studied months of use. The accurate framing is "studied in small human trials with mixed results, and used clinically," not "clinically proven." Any source that says DSIP is proven to improve sleep is overstating it; any source that says it has never been tested in people is wrong the other way.

• It does not establish that subcutaneous DSIP at bedtime does what intravenous infusion did in the 1980s. Nobody has run that trial.
• It does not establish a clinically meaningful effect size for sleep quality. The two independent placebo-controlled studies concluded there was not one.
• It does not establish that DSIP increases slow-wave sleep in humans. The controlled studies that measured it did not find that.
• It does not establish long-term safety. The longest human course on record is fifteen days.
• It does not establish efficacy for any condition as an FDA-approved treatment. It is not approved for anything.

What the 1980s work described is a change in the shape of sleep — fewer awakenings, less light sleep — rather than sedation. People expecting the heaviness of a sleeping tablet are usually surprised by how little they notice going in.

Prescribers generally treat the first vial as the assessment window: a sleep diary against the Monday-to-Friday schedule. Some people report steadier sleep and clearer mornings within two weeks; some report nothing. Both are normal, and "nothing" is an answer for your physician, not a reason to change the dose yourself.

The animal work on the stress axis leans toward DSIP being described as most relevant when poor sleep sits on top of a stress response. Long-standing insomnia is a different problem, and the human studies in exactly that population were the ones that split. Be honest with your physician about which one you have.

No withdrawal and no rebound insomnia was reported in the controlled studies that measured post-treatment nights. Whether any benefit persists depends on whether the thing driving the poor sleep has changed.

The directions are written in millilitres and in insulin-syringe units because that is what you can read off the barrel. 0.25 mL is 25 units on a U-100 syringe. You are not calculating anything — the number is printed on your medication.

Subcutaneous means into the fat layer, not the muscle. Abdomen and thigh are the usual sites, rotated. The reviewed directions are Monday through Friday nights at bedtime — five on, two off. That is a prescribing convention, not a schedule taken from the trials, which used intravenous infusion. Your directions say what your physician has decided; follow them.

20 doses at five nights a week is four weeks, which is why refills run on a 28-day cycle. One vial per fill, always — not a stockpile.

Refrigerated, and it ships that way in insulated packaging with ice packs. Beyond-use dating runs from first puncture and is printed on the label. Do not freeze it.

Both were reviewed at the same FDA meeting, and they are described around different things. DSIP is described around sleep architecture and the stress axis. Epithalon is a tetrapeptide described around the pineal gland and melatonin rhythm. If the problem reads as timing, a physician may reasonably start with epithalon; if it reads as fragmented sleep under stress, DSIP is the more usual conversation.

Selank is described around anxiety signalling and taken during the day. DSIP is taken at bedtime. They are not competing for the same job, which is why the SERENITY blend puts them in one vial with CJC-1295 and Ipamorelin.

Approved hypnotics have large randomised trials, FDA labels, and known dependence and next-day profiles. DSIP has small 1980s trials with mixed results and no dependence signal in the human record. One is a sedative with proven short-term effect; the other is described as a modulator of sleep architecture without sedation. Neither substitutes for the other; do not stop a prescription hypnotic without your prescriber.