What is DSIP?

A peptide pulled from rabbit blood in 1977

DSIP, short for delta sleep-inducing peptide, comes from one of the stranger experiments in sleep science. In 1977, a research group led by Schoenenberger and Monnier in Basel electrically stimulated the thalamus of a rabbit until it fell into deep sleep, then collected the blood draining from its brain. When they injected that blood into the brain ventricles of an awake rabbit, the animal produced the slow, rolling delta waves that mark deep, non-REM sleep. The active ingredient turned out to be a nine-amino-acid peptide: Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu. They named it for what it appeared to do.

Nearly fifty years later, DSIP is still sold today, marketed as a sleep and stress peptide to the same off-label research-chemical market that supplies compounds like MOTS-c and Semax. What has not changed since 1977 is more notable than what has: nobody has ever isolated a DSIP gene or identified a confirmed receptor for it in the human body.

What DSIP is supposed to do

Without a confirmed receptor, DSIP's mechanism stays a working theory rather than a settled fact. The leading explanations point to the GABAergic and opioidergic systems, the same signaling pathways involved in sedation and pain relief, and to a possible role damping down the hypothalamic-pituitary-adrenal axis, the hormonal cascade that drives the body's stress response and its daily cortisol rhythm. That would explain both of the effects DSIP has been studied for: better sleep, and softer withdrawal symptoms when the body is under acute physiological stress.

It is a plausible story built from real, decades-old animal and cell work. It has also stayed a hypothesis for almost half a century, which is unusual even by the standards of the peptides covered in our peptide stacks guide.

The sleep evidence: promising at first, thinner under scrutiny

The strongest human data comes from a series of small trials Schneider-Helmert and colleagues ran between 1981 and 1987. A 1981 study gave six chronic insomniacs a single intravenous dose of synthetic DSIP and found longer sleep, fewer interruptions, and slightly more REM sleep, with no daytime sedation. A 1987 placebo-controlled, double-blind trial of 14 middle-aged insomniacs found sleep improved after the first dose and further with repeated doses over seven nights.

That is where the trail gets weaker. A larger, more tightly controlled double-blind study in the early 1990s, designed to correct for the small sample sizes and looser methodology of the earlier work, found the sleep benefit shrank close to nothing once proper blinding was in place. That arc, a striking result in a small open trial that fades under stricter controls, is a pattern the FDA's own reviewers would flag directly when they took up DSIP in 2026.

The withdrawal studies: the numbers everyone quotes

DSIP's most-cited data point has nothing to do with sleep. In a 1984 study, Dick and colleagues gave intravenous DSIP to 107 hospitalized patients, 47 withdrawing from alcohol and 60 from opiates, and reported that withdrawal symptoms disappeared or markedly improved in 97% of the opiate group and 87% of the alcohol group. Those figures show up constantly in marketing copy for DSIP as a research chemical.

What that copy usually leaves out: the study was open-label, with no placebo arm and no blinding, run at a time when withdrawal management options were far more limited than they are today. No one has reproduced it under modern trial standards in the four decades since. A dramatic effect size from an unblinded 1984 hospital study is a real data point, not proof the compound does what the number suggests.

What FDA staff and the panel actually found in 2026

DSIP, under its INN name emideltide, was one of seven peptides the FDA's Pharmacy Compounding Advisory Committee took up on July 23-24, 2026, the same hearing that reviewed BPC-157 and TB-500. There was an unusual wrinkle in emideltide's file that none of the other six shared: the original nominations had actually been withdrawn, and the FDA proceeded to evaluate it anyway, at its own discretion. Staff reviewed both the free-base and acetate-salt forms, since it was not clear which one the withdrawn nomination had even meant, and flagged inadequate characterization and the potential for peptide-related impurities on top of the thin efficacy record. The proposed uses on the table were chronic insomnia and opioid withdrawal, given subcutaneously.

The committee had voted to approve every other peptide on the agenda, in several cases against its own staff's recommendation. Emideltide broke that streak, failing 6-7 with one abstention, the only no vote of the two-day hearing. STAT News reported that David Pope, chief pharmacy officer at XiFin Pharmacy Solutions, had voted yes on every peptide the day before and switched to no specifically on emideltide, citing concern about potentially dangerous downstream consequences. That single flipped vote was the margin.

Where this leaves DSIP

A committee vote is a recommendation, not a rule, and the FDA does not have to follow it either way. But a no vote from a panel that said yes to six other peptides with comparably thin human data, including some staff had flagged as having zero human studies at all, is a meaningful signal. We covered the full two-day breakdown, including how BPC-157, TB-500, KPV, MOTS-c, Semax, and Epitalon all cleared the same bar DSIP did not, in our writeup of the July 24 results.

None of this changes how DSIP is actually sold or bought today. It was never FDA-approved, it is not a controlled substance, and it continues to circulate as a research-labeled vial regardless of what the 503A rulemaking eventually decides. What the 2026 review adds is a clearer public record of the gap between DSIP's marketing, built on a 1970s discovery story and a dramatic 1984 withdrawal statistic, and the modern evidence bar the FDA actually applies. If you are comparing DSIP against other compounds people stack for sleep and recovery, see our peptide stacks guide for how the evidence lines up across the field.