Peptides for sleep: DSIP, epitalon and what actually has data

Why peptides are marketed for sleep

Sleep is regulated partly by peptides, which makes the idea of a sleep peptide plausible. The clearest example is orexin, also called hypocretin, a brain peptide that keeps us awake: people with narcolepsy type 1 have lost most of the neurons that make it, and over 90% have very low orexin levels in their spinal fluid (review) ¹. Our orexin-A entry covers that system.

The peptides sold for sleep, though, are not orexin. They are delta sleep-inducing peptide, a nine-amino-acid molecule isolated from rabbit cerebral blood in Basel in 1977 (review) ², and epitalon, a four-amino-acid synthetic peptide from the St Petersburg group of Vladimir Khavinson. Both have small literatures that are decades old or come largely from one research group, and neither is approved anywhere as a sleep medicine. That combination, a plausible story and very little modern testing, is common among peptides sold online, and sleep is a field where it is especially easy to mistake a good night for a treatment effect.

DSIP: what the insomnia trials found

DSIP was tested in people with chronic insomnia in a cluster of small sleep-laboratory trials in the 1980s and early 1990s, usually by intravenous injection. Their results disagree, and the disagreement mostly tracks study quality.

The most careful trial, published in 1992, gave 16 chronic insomniacs either DSIP at 25 nmol/kg or a glucose placebo intravenously before three nights in the sleep lab, in a double-blind parallel design. Sleep efficiency and time to fall asleep looked better with DSIP, but the authors found the effects weak and partly due to a chance change in the placebo group; subjective sleep quality did not change, and they concluded short-term DSIP was not likely to be of major benefit (randomised trial) ³. A double-blind crossover trial over four nights found total sleep and light stage-2 sleep increased, but the same differences were already present before treatment, and judged the improvement of little clinical significance (randomised trial) ⁴.

Other trials from the same period were more positive. In 14 middle-aged insomniacs given DSIP for seven nights under placebo-controlled, double-blind conditions, night sleep improved and daytime alertness rose (randomised trial) ⁵. An 18-patient study of one week's injections reported sleep normalising in middle-aged and older insomniacs, with the larger effects in those with worse sleep, though it compared patients with themselves rather than with a placebo group (uncontrolled study) ⁶.

Taken together, these are small, short trials of an injected peptide, run over 30 years ago, with no large confirmatory trial since, and the positive and negative results came from studies of similar size and design. Our DSIP entry lists the full set.

Why DSIP never became a drug

DSIP's history explains much of its weak evidence. A 2006 review called it a still unresolved riddle: isolated in 1977 as a candidate sleep factor, its link to sleep was never properly characterised, partly because no gene for it has been identified and no receptor confirmed (review) ². Without a known gene or receptor, it is hard to say how, or whether, it acts in the body.

The early human results were striking enough to keep interest alive. In the first human study, six healthy volunteers given a slow intravenous infusion reported sleep pressure soon afterwards (randomised crossover study) ⁷. Other findings pointed the other way: in cats, DSIP reduced sleep over the following hours (animal study) ⁸, and in 24 women having surgery, intravenous DSIP made anaesthesia lighter rather than deeper (randomised trial) ⁹. Claims that DSIP rebalances stress hormones also failed when tested: in healthy young men it left ACTH and cortisol responses almost unchanged (randomised study) ¹⁰, and in eight healthy women it had no effect on growth hormone or prolactin (controlled study) ¹¹. A molecule that seems to promote sleep in some settings and reduce it in others, with no confirmed mechanism and none of its proposed hormone effects confirmed, is a difficult basis for a medicine, and it was never developed as one.

What approved sleep medicines achieve

For scale, a network meta-analysis of 153 randomised trials with 46,412 adults compared insomnia drug classes against placebo. Orexin receptor antagonists added about 32 minutes of objectively measured total sleep time, non-benzodiazepine 'Z-drugs' about 22 minutes, and certain antidepressants about 36 minutes; Z-drugs and melatonin receptor agonists shortened the time to fall asleep. The authors highlighted doxepin, suvorexant and lemborexant among drugs that were relatively effective with relatively good tolerability (meta-analysis) ¹².

Those effects are modest, and every class has side effects, which is part of why guidelines put CBT-I first. But they are measured in thousands of patients against placebo, which is the standard no sleep peptide has met.

Epitalon: a melatonin finding, not a sleep trial

Epitalon's link to sleep comes through the pineal gland, which makes melatonin. A 2007 Russian report described lower night-time melatonin and flatter daily rhythms in old monkeys and elderly people, and said that epithalamin, a pineal extract, and epitalon, its synthetic tetrapeptide counterpart, restored night-time melatonin release (human and animal study) ¹³. The abstract does not report sleep outcomes, numbers of patients or a placebo group.

That is the extent of the sleep-relevant human evidence we could find. Raising melatonin is not the same as improving sleep, and a blood-hormone change in a small group says nothing about how people slept. Our epitalon entry covers the rest of its literature, which centres on telomeres and ageing in cell cultures and comes mostly from one group.

What FDA's 2026 review did with both

Both peptides reached FDA's Pharmacy Compounding Advisory Committee in July 2026. FDA's meeting page lists the uses its staff reviewed: for emideltide, which is DSIP, opioid withdrawal, chronic insomnia and narcolepsy; for epitalon, insomnia (FDA meeting page) ¹⁴. The committee voted against adding emideltide to the 503A list and narrowly in favour of epitalon, against the advice of FDA staff, who had recommended neither. Both margins were narrow: our report on the votes records emideltide rejected 6 to 7 with one abstention, and epitalon recommended 7 to 4 or 7 to 5, with reports differing on the count against.

As of October 4, 2026, FDA has not acted on either recommendation, so no pharmacy has a lawful basis to compound epitalon or DSIP. Our guide to what the vote means explains why a committee recommendation is not a green light.

What works for insomnia

The contrast with established treatments is stark. The American Academy of Sleep Medicine's 2021 guideline strongly recommends multicomponent cognitive behavioural therapy for insomnia (CBT-I) for chronic insomnia in adults, and conditionally recommends brief behavioural therapies, stimulus control, sleep restriction and relaxation therapy (clinical guideline) ¹⁵. CBT-I combines techniques such as stimulus control and sleep restriction to reset the link between bed and sleep, and addresses the thoughts that keep people awake.

The Academy's 2026 guideline on combining treatments suggests CBT-I plus an insomnia medicine over a medicine alone, while rating the certainty of that evidence as low (clinical guideline) ¹⁶. Medicines are an option, not the first line.

Among medicines, the newest class works through the same orexin system that peptides cannot reach: orexin receptor antagonists such as suvorexant, lemborexant and daridorexant block orexin's wake signal. A network meta-analysis of 10 trials with 7,806 patients found they improved time awake after falling asleep, time to fall asleep and total sleep time compared with placebo (meta-analysis) ¹⁷. They are small molecules, not peptides, but they show what targeting a sleep peptide's receptor can do when the evidence is built properly.

Side by side

• DSIP: a cluster of small insomnia trials from 1981 to 1992, mixed results, injected; rejected by FDA's advisers in 2026; not lawfully compoundable.
• Epitalon: a report of raised night-time melatonin in elderly people; no sleep trial found; narrowly backed by FDA's advisers for insomnia in 2026, with FDA yet to act.
• CBT-I: strong recommendation from the American Academy of Sleep Medicine as first-line treatment for chronic insomnia.
• Orexin receptor antagonists: approved medicines with randomised trials in thousands of patients.

Growth hormone boosters and 'deeper sleep'

Another group of products is sold partly on sleep claims: growth hormone secretagogues such as ipamorelin and MK-677. Deep sleep and growth hormone release are linked, and there is one small trial behind the claim. In 1997, eight young men took MK-677 at bedtime for a week in a double-blind, placebo-controlled crossover; at 25 mg, stage IV deep sleep lasted about 50% longer and REM sleep more than 20% longer than with placebo, and six older adults given it for two weeks had a near 50% increase in REM sleep (randomised trial) ¹⁸.

Two caveats matter. MK-677 is not a peptide but an orally active small molecule, and it was never approved; its development ended without an approval, and our MK-677 entry covers the side effects seen in longer trials. And 14 people in a sleep laboratory nearly 30 years ago is a long way from evidence that it improves insomnia.

How to read a 'better sleep' claim

Sleep claims for supplements and peptides usually rest on one of four kinds of evidence, in rising order of weight: a mechanism (the peptide acts on something related to sleep), a hormone or brain-wave change in a laboratory, a small trial measuring sleep, and large placebo-controlled trials in people with insomnia measuring how they sleep and function. DSIP reached the third level decades ago with mixed results; epitalon has evidence at the second; approved medicines and CBT-I are at the fourth (clinical guideline) ¹⁵.

A product page that cites a mechanism or a melatonin change as proof of better sleep is skipping two levels. Our methodology page explains how we grade the evidence on every entry.

The risks of trying sleep peptides anyway

DSIP and epitalon sold online are research-grade products, outside any quality control; testing of comparable gray-market peptides has found missing, wrong or contaminated contents, as our article on group buys and testing describes. The DSIP trials used intravenous injection in sleep laboratories, which says little about home subcutaneous use of a product of unknown purity. No dose of either peptide has been established for sleep in a controlled trial.

Poor sleep can also be a symptom of something treatable, such as sleep apnoea, depression, restless legs or a medicine's side effect. Treating it with an unproven peptide can delay finding the cause, and an injection routine at bedtime adds its own risks of infection and reactions at the injection site, with no trial evidence that the benefit outweighs them.

A practical path if you cannot sleep

1. See a clinician if poor sleep lasts more than a few weeks, especially with loud snoring, gasping at night, restless legs, low mood or daytime sleepiness that affects driving. These point to causes that need specific treatment.
2. Ask about CBT-I. It is the first-line treatment sleep specialists recommend for chronic insomnia (clinical guideline) ¹⁵, and it is offered in person, by telehealth and through structured digital programmes.
3. Discuss medicine if needed, ideally alongside CBT-I rather than instead of it, as the 2026 guideline suggests (clinical guideline) ¹⁶. Ask about each option's side effects, including next-day drowsiness and falls in older adults.
4. Be wary of products sold online for sleep that rest on mechanisms or hormone changes rather than sleep trials, including peptides sold for research use.

None of these steps needs a peptide or a special supplement, and each has more evidence behind it than any peptide sold for sleep.

A note on narcolepsy

Narcolepsy is the one sleep disorder where a peptide is central: type 1 is caused by loss of orexin-producing neurons, with postmortem studies showing up to 95% lost, and a strong genetic and immune association (review) ¹. That has made orexin-receptor agonists, drugs that would replace the missing signal, a major research target, with trials reported in recent years. DSIP was among the uses FDA reviewed for narcolepsy in 2026 (FDA meeting page) ¹⁴, and its advisers voted against it. Anyone with excessive daytime sleepiness or sudden muscle weakness with emotion should see a sleep specialist rather than try a peptide.

Questions to ask before trying a sleep peptide

• Has a recent, placebo-controlled trial shown this peptide improves sleep, at the dose and route I would use?
• Have I tried CBT-I, the treatment sleep specialists recommend first?
• Could my poor sleep have a cause, such as apnoea or a medicine, that needs treating?
• Where would the product come from, and who has checked what is in it?

How it adds up (as of October 4, 2026)

The peptides sold for sleep have thin evidence. DSIP was tested in small insomnia trials decades ago, and the most careful found little clinically meaningful benefit; FDA's advisers rejected it in 2026. Epitalon's sleep claim rests on a melatonin measurement in older people, not on sleep trials; FDA's advisers narrowly backed it for insomnia, but FDA has not acted. Cognitive behavioural therapy for insomnia is strongly recommended as first-line treatment, and approved medicines, including orexin receptor antagonists, have trials in thousands of patients. For better sleep, those are the options with data. If a sleep peptide ever earns a place, it will be through the kind of large, placebo-controlled trials that neither DSIP nor epitalon has had, and until then the honest answer to 'which peptide helps sleep?' is that none has been shown to.

Frequently asked questions

What is the best peptide for sleep?

None has good evidence. DSIP had mixed results in small 1980s trials and epitalon has no sleep trials; cognitive behavioural therapy for insomnia and approved medicines have far stronger data.

Does DSIP work for insomnia?

Small trials disagreed; the most careful, a 1992 double-blind study in 16 patients, found weak effects and concluded DSIP was unlikely to be of major benefit.

Does epitalon help sleep?

A 2007 report said pineal peptides including epitalon raised night-time melatonin in elderly people, but no trial measuring sleep was found.

Can a pharmacy compound DSIP or epitalon?

No. FDA's advisers rejected DSIP and narrowly backed epitalon in July 2026, but FDA has not acted, so neither can lawfully be compounded as of October 4, 2026.

What works best for chronic insomnia?

The American Academy of Sleep Medicine strongly recommends multicomponent cognitive behavioural therapy for insomnia (CBT-I) as treatment for chronic insomnia in adults.

Related on Grey Peptides

• DSIP
• Epitalon
• The July 2026 PCAC votes
• What the PCAC vote means

Sources

1. Rauf, R., et al. (2025). Orexin Deficiency in Narcolepsy: Molecular Mechanisms, Clinical Phenotypes, and Emerging Therapeutic Frontiers. Brain Behav, 15(10), e70984. PMID: 41076550
2. Kovalzon, V. M., et al. (2006). Delta sleep-inducing peptide (DSIP): a still unresolved riddle. J Neurochem, 97(2), 303-9. PMID: 16539679
3. Bes, F., et al. (1992). Effects of delta sleep-inducing peptide on sleep of chronic insomniac patients. A double-blind study. Neuropsychobiology, 26(4), 193-7. PMID: 1299794
4. Monti, J. M., et al. (1987). Study of delta sleep-inducing peptide efficacy in improving sleep on short-term administration to chronic insomniacs. Int J Clin Pharmacol Res, 7(2), 105-10. PMID: 3583493
5. Schneider-Helmert, D. (1987). Effects of delta-sleep-inducing peptide on 24-hour sleep-wake behaviour in severe chronic insomnia. Eur Neurol, 27(2), 120-9. PMID: 3622582
6. Schneider-Helmert, D. (1986). Efficacy of DSIP to normalize sleep in middle-aged and elderly chronic insomniacs. Eur Neurol, 25(6), 448-53. PMID: 3792404
7. Schneider-Helmert, D., et al. (1981). Acute and delayed effects of DSIP (delta sleep-inducing peptide) on human sleep behavior. Int J Clin Pharmacol Ther Toxicol, 19(8), 341-5. PMID: 6895513
8. Sommerfelt, L. (1985). Reduced sleep in cats after intraperitoneal injection of delta-sleep-inducing peptide (DSIP). Neurosci Lett, 58(1), 73-7. PMID: 3840239
9. Pomfrett, C. J., et al. (2009). Delta sleep-inducing peptide alters bispectral index, the electroencephalogram and heart rate variability when used as an adjunct to isoflurane anaesthesia. Eur J Anaesthesiol, 26(2), 128-34. PMID: 19142086
10. Späth-Schwalbe, E., et al. (1995). Delta-sleep-inducing peptide does not affect CRH and meal-induced ACTH and cortisol secretion. Psychoneuroendocrinology, 20(3), 231-7. PMID: 7777652
11. Giusti, M., et al. (1993). Delta sleep-inducing peptide administration does not influence growth hormone and prolactin secretion in normal women. Psychoneuroendocrinology, 18(1), 79-84. PMID: 8475226
12. Pan, B., et al. (2023). The Comparative Effectiveness and Safety of Insomnia Drugs: A Systematic Review and Network Meta-Analysis of 153 Randomized Trials. Drugs, 83(7), 587-619. PMID: 36947394
13. Korkushko, O. V., et al. (2007). [Normalizing effect of the pineal gland peptides on the daily melatonin rhythm in old monkeys and elderly people]. Adv Gerontol, 20(1), 74-85. PMID: 17969590
14. Food and Drug Administration. July 23-24, 2026 Meeting of the Pharmacy Compounding Advisory Committee: substances and uses evaluated; read October 4, 2026. Source
15. Edinger, J. D., et al. (2021). Behavioral and psychological treatments for chronic insomnia disorder in adults: an American Academy of Sleep Medicine clinical practice guideline. J Clin Sleep Med, 17(2), 255-262. PMID: 33164742
16. Buysse, D. J., et al. (2026). Combination treatment for chronic insomnia disorder in adults: an American Academy of Sleep Medicine clinical practice guideline. J Clin Sleep Med, 22(1). PMID: 41975142
17. Rocha, R. B., et al. (2023). Dual orexin receptor antagonists for the treatment of insomnia: systematic review and network meta-analysis. Arq Neuropsiquiatr, 81(5), 475-483. PMID: 37257468
18. Copinschi, G., et al. (1997). Prolonged oral treatment with MK-677, a novel growth hormone secretagogue, improves sleep quality in man. Neuroendocrinology, 66(4), 278-86. PMID: 9349662

Educational information, not medical advice. Each dose in this guide names its source, an approved label or a published study. None is a recommendation for you. An unapproved compound has no established safe or effective human dose, and products sold for “research use only” are not made or tested for people. Talk to a doctor before acting on anything on this site, including before you start, stop or change any medicine or dose.