Compounds with a graded row

Ordered by rung, strongest first. The rung rates the study design; the chip beside it says what the study measured. Both are needed.

1. DSIP

   human RCT Direct outcome Not approved for this condition

   Who was studied

   patients with chronic insomnia, n=16 in a matched-pairs parallel-groups design (8 DSIP, 8 placebo), 5 consecutive laboratory nights; sex and age not reported in the source

   A double-blind trial gave 25 nmol/kg intravenously in the afternoon before each of three consecutive nights and measured polysomnographic sleep structure, subjective sleep quality and subjective tiredness; sleep efficiency and sleep latency favoured DSIP, but the authors judged the significant effects weak and partly attributable to an incidental change in the placebo group, and concluded that short-term treatment of chronic insomnia with DSIP is unlikely to be of major therapeutic benefit. A separate double-blind crossover study at the same 25 nmol/kg intravenous dose found the same differences already present at baseline and described the sleep improvement as of little clinical significance, while a 1987 seven-night study in 14 middle-aged insomniacs reported a substantial improvement. Every controlled human trial on file is intravenous and dates from 1984 to 1992; no controlled trial has been published since, and no human trial of subcutaneous DSIP, the route it is sold in, exists on file.

   Source PubMed 4 primary sources read for this row

2. Progesterone

   human RCT Direct outcome Not approved for this condition

   Who was studied

   n=388 pooled across 9 randomised controlled trials, 8 of which enrolled postmenopausal women; oral micronised progesterone; participants were not required to carry an insomnia diagnosis

   A systematic review and meta-analysis of nine randomised controlled trials measured sleep by polysomnography and by self-report; pooling four of them favoured oral micronised progesterone for sleep onset latency (effect size 7.10, CI 1.30 to 12.91) but showed no difference in total sleep time or sleep efficiency, and the reviewers noted that concomitant oestradiol and improvement in vasomotor symptoms limit what several trials can be read to show. In the most detailed single trial, eight postmenopausal women aged 48 to 74 without sleep complaints had 300 mg nightly for three weeks: sleep was the same as placebo on an undisturbed night, and differed only on a night deliberately fragmented by 15-minute blood sampling. No trial in a population with diagnosed chronic insomnia is on file, and medroxyprogesterone acetate, a different molecule, must not be read across to this record.

   Source PubMed 2 primary sources read for this row

3. Sermorelin

   human RCT Direct outcome Not approved for this condition

   Who was studied

   19 healthy older adults (10 women, 9 men), aged 55 to 71, no sleep disorder reported. Single-blind randomised placebo-controlled trial of five months: four weeks placebo followed by 16 weeks of [Nle27]GHRH(1-29)NH2, 10 mcg/kg, self-administered subcutaneously each night. Sleep quality was a secondary self-reported measure; no polysomnography.

   The only trial on file that administered a named GHRH(1-29) molecule and recorded a sleep endpoint reported that sleep quality was unaffected in both sexes after 16 weeks of nightly subcutaneous dosing at 10 mcg/kg. The molecule was [Nle27]GHRH(1-29)NH2, an analogue of sermorelin rather than sermorelin itself, and the endpoint was self-report rather than polysomnography. The acute sleep-EEG literature people cite here is a different body of work: repeated intravenous boluses, typically 4 x 50 mcg between 22.00 and 01.00, raised slow-wave and non-REM sleep in healthy men, and intranasal 300 mcg raised slow-wave and REM sleep in young and aged men. Those reports name only "GHRH" and do not state which fragment was given, so whether they were sermorelin cannot be established from the published abstracts, and an intravenous bolus is not the nightly subcutaneous route people use. The direction is also not uniform by sex: with 4 x 50 mcg intravenously, non-REM sleep rose in men and fell in women, in both depressed patients and controls. No trial in an insomnia population is on file for any GHRH preparation.

   Source PubMed 1 primary source read for this row

4. GHRH (Growth Hormone-Releasing Hormone)

   human RCT Surrogate marker Not approved for this condition

   Who was studied

   48 normal women and men aged 19 to 67 years, studied across a baseline night and a recovery night after 40 hours of sleep deprivation; GHRH, corticotropin-releasing hormone or placebo injected repetitively during the recovery night

   Sleep was measured directly by electroencephalography. After GHRH the rise in non-rapid-eye-movement sleep and the fall in wakefulness were larger than after placebo, and growth hormone rose while cortisol fell. Two things stop this being a result about insomnia. The participants were normal sleepers deprived of sleep for 40 hours, not patients with a sleep disorder, and recovery sleep is described by the authors as differing markedly from spontaneous sleep. The direction is also not the same in both sexes: earlier work in this line reports that GHRH promotes sleep in men and impairs it in women, though no sex difference appeared in this particular study. The abstract does not state which GHRH preparation was used. A separate experiment cuts the other way — blocking endogenous GHRH receptors overnight in healthy men suppressed the growth hormone response by 93 per cent yet left the percentage of slow-wave sleep unchanged, from which those authors concluded that endogenous GHRH is unlikely to generate slow-wave sleep. No trial has given GHRH to people with insomnia and measured whether their insomnia improved. Rung set from the study's own design wording, "Randomized Controlled".

   Source PubMed 2 primary sources read for this row

5. hGH

   human RCT Surrogate marker Not approved for this condition

   Who was studied

   Two crossover sleep-laboratory studies in healthy men, neither in an insomnia population. Kern 1993: double-blind within-subject crossover, 12 men then 3 men then 10 men, 5 IU intramuscularly and 5 and 48 IU intravenously, sleep stages scored across the night. Mendelson 1980: crossover against saline in normal volunteers, 2 units and 5 units intramuscularly 15 minutes before bed.

   Exogenous growth hormone has been given to healthy men in a sleep laboratory twice, and neither result supports it as a sleep aid. In a double-blind crossover, 5 IU intramuscularly and 5 and 48 IU intravenously changed neither total sleep time nor time in any sleep stage. In an earlier crossover, 2 units had no effect on the sleep EEG and 5 units cut slow-wave sleep by 19 per cent and raised REM sleep by 13 per cent. The direction of the larger dose is the opposite of the effect attributed to the GH axis in marketing copy, which rests on GHRH studies rather than on hGH studies. Both studies scored sleep architecture in healthy men; nobody with insomnia was studied, and no trial of hGH with an insomnia outcome is on file.

   Source PubMed 2 primary sources read for this row

6. Galanin

   human RCT Mechanistic only Not approved for this condition

   Who was studied

   patients with depression on a stable dose of trimipramine, n=10, sex not reported; four hourly intravenous boluses of 50 mcg between 09:00 and 12:00 on each of two days, crossed over with placebo on two further days

   Recorded, and not evidence for this condition

   The primary measure was the change in the Hamilton depression rating scale, not sleep; the sleep electroencephalogram was recorded once after placebo and once after galanin, and the overall change was driven mainly by a longer rapid-eye-movement latency that did not itself reach significance. An earlier study in healthy young men, at the same dose and at three times it, found changes resembling rapid-eye-movement sleep deprivation rather than sedation. No study measured insomnia in any population, so the reviews that list galanin among the sleep-promoting peptides are summarising these electroencephalogram changes and animal work, not an insomnia result.

   Source PubMed 2 primary sources read for this row

7. Neuropeptide Y

   human case series Surrogate marker Not approved for this condition

   Who was studied

   healthy young men, n=9 given 4 x 50 mcg and n=11 given 4 x 100 mcg as hourly intravenous boluses from 22:00 to 01:00 on a single night, against saline; no participant had insomnia

   Overnight sleep electroencephalography showed longer sleep period time and stage 2 sleep and shorter sleep latency and time awake after the peptide, alongside lower overnight ACTH and cortisol. The report does not describe randomisation or blinding, which is why this sits on the lowest human rung: the same laboratory's later study — a fixed adaptation night, then a placebo night, then a peptide night, in patients with depression and in controls — reproduced the shorter sleep onset latency but attributed part of it to an adaptation effect. Neither study enrolled anyone with insomnia, and no trial of neuropeptide Y in an insomnia population is on file.

   Source PubMed 2 primary sources read for this row

Recorded absences

10 compounds are listed here without a rung. Each row exists to mark a distinction a reader would otherwise draw wrongly, and each says what is missing rather than leaving the compound off the page.

1. CJC-1295 DAC

   No rung — nothing on file to grade Not approved for this condition

   No sleep measure of any kind has been recorded in a human given CJC-1295 with DAC. The published human trials measured growth hormone and IGF-I concentrations only. The sleep-EEG work sometimes cited alongside this compound used repeated short intravenous pulses of GHRH, and the one determinant that work identified was pulsatility: episodic dosing raised slow-wave sleep where a continuous infusion of the same 200 mcg did not. CJC-1295 with DAC is built to do the opposite of pulsing, binding albumin to hold the growth hormone axis stimulated for days, so the acute pulse findings are not evidence for it. Nobody with insomnia has been given it in a published study.

2. CJC-1295 no-DAC

   No rung — nothing on file to grade Not approved for this condition

   Nothing is recorded for CJC-1295 without DAC in humans against any endpoint, sleep included. Searches of the published literature return the DAC form only, plus doping-control assay development. The two forms are different molecules with different durations of action, and a result for one is not a result for the other. Where a source cites GHRH sleep-EEG studies here, note that those reports name only "GHRH" and do not state the fragment given, so they cannot be attributed to this compound either.

3. CRH (Corticotropin-Releasing Hormone)

   No rung — nothing on file to grade Not approved for this condition

   CRH infusion is used experimentally to disrupt sleep architecture as a model of depressive sleep. It is a provocation, and a harmful one; no trial gives it to improve sleep.

4. GHRP-1

   No rung — nothing on file to grade Not approved for this condition

   No trial has used this compound against this condition; no polysomnographic study of GHRP-1 was retrieved.

5. GHRP-2

   No rung — nothing on file to grade Not approved for this condition

   No trial has used this compound against this condition; no polysomnographic study of GHRP-2 was retrieved. The sleep work in this family used ghrelin and growth hormone-releasing hormone.

6. GHRP-6

   No rung — nothing on file to grade Not approved for this condition

   No trial has used this compound against this condition; no polysomnographic study of GHRP-6 was retrieved.

7. Hexarelin

   No rung — nothing on file to grade Not approved for this condition

   No trial has used this compound against this condition; no polysomnographic study of hexarelin was retrieved.

8. Ipamorelin

   No rung — nothing on file to grade Not approved for this condition

   No sleep measure has been recorded in a human given ipamorelin. The one published human trial studied postoperative ileus and did not record sleep. Ipamorelin is a ghrelin-receptor agonist rather than a GHRH analogue, so GHRH sleep findings do not carry across, and within its own class the results run in both directions: ghrelin raised slow-wave sleep in healthy men, while hexarelin, another ghrelin-receptor agonist, decreased it. Which of those ipamorelin would resemble is not on file, and nobody with insomnia has been given it in a published study.

9. Orexin-A

   No rung — nothing on file to grade Not approved for this condition

   Orexin-A is an agonist at the orexin receptors and promotes wakefulness. The licensed insomnia medicines that act on this system are orexin receptor antagonists — different molecules with the opposite action — and no result about them transfers to this peptide. The only controlled human study of administered orexin-A on sleep that could be located enrolled eight people with narcolepsy with cataplexy, the opposite complaint, and gave a single 435 nmol intranasal dose before one night of polysomnography; it found no significant change in nocturnal wakefulness and reduced rapid-eye-movement sleep. No trial of orexin-A in insomnia exists on file.

10. Vasopressin

    No rung — nothing on file to grade Not approved for this condition

    Endogenous vasopressin varies with the sleep-wake cycle; no study administers it as a sleep intervention.

Checked, and nothing recorded

13 compounds were checked against insomnia and produced no gradeable row.

• Agouti-Related Protein (AgRP) — checked, no row on file
• ARA-290 — checked, no row on file
• C-Max — checked, no row on file
• CART (Cocaine- and Amphetamine-Regulated Transcript) — checked, no row on file
• Epitalon — checked, no row on file
• Melanotan II — checked, no row on file
• Orexin-B — checked, no row on file
• Sauvagine — checked, no row on file
• Selank — checked, no row on file
• Semax — checked, no row on file
• Simonson Alpha 1 — checked, no row on file
• Tirzepatide — checked, no row on file
• Urotensin II — checked, no row on file

A compound is here because someone looked and the sources did not support a row — not because nobody looked. Under the rubric, uncertainty resolves to no row, and a row resting on a vendor page or a clinic blog is worse than no row at all.