Effects of suvorexant on sleep architecture and power spectral profile in patients with insomnia: analysis of pooled phase 3 data

Highlights

• Suvorexant is a first-in-class orexin receptor antagonist for treating insomnia.

• Sleep architecture is preserved in insomnia patients taking suvorexant.

• The power spectral profile of suvorexant is similar to placebo.

Abstract

Background

The orexin receptor antagonist, suvorexant, is approved for treating insomnia at a maximum dose of 20 mg. We evaluated its effects on sleep architecture.

Methods

The analyses included pooled polysomnography data from two similar randomized, double-blind, placebo-controlled, 3-month trials evaluating two age-adjusted (non-elderly/elderly) dose regimes of 20/15 mg and 40/30 mg in 1482 patients with insomnia. Polysomnography was recorded at baseline and on three nights during the treatment: Night-1, Month-1, and Month-3. Effects on non-REM sleep stages 1 (N1), 2 (N2), 3 (N3)/slow wave sleep (SWS), and REM sleep were evaluated. A power spectral analysis of non-REM sleep was also performed.

Results

Suvorexant increased the time (in minutes) spent in all sleep stages compared with placebo. When suvorexant and placebo were compared in terms of changes in percentage of total sleep time spent in each stage, there were small decreases of ≤1%, ≤2.2%, and ≤0.8% for N1, N2, and N3/SWS on average, respectively, and an average increase of ≤3.9% in REM. The largest differences from placebo were observed at Night-1 and generally diminished over time. Suvorexant reduced REM latency (number of non-REM 30-s epochs from lights-off to the first REM epoch) compared with placebo; the reduction was greater at Night-1 (~40–50 non-REM epochs) in comparison to later time points (~12–25 non-REM epochs at Month-3). The spectral analysis of non-REM showed a small decrease in power of 3–6% in the gamma and beta bands, and a small increase of 4–8% in the delta band, at Night-1 for suvorexant relative to placebo; these effects were not apparent at the later Month-1 and Month-3 time points.

Conclusion

Overall sleep architecture appears to be preserved in insomnia patients taking suvorexant. The power spectral profile of suvorexant is generally similar to placebo.

Introduction

Most of the present insomnia drugs act through increasing the action of the inhibitory neurotransmitter gamma-aminobutyric acid (GABA) in the brain [1]. However, this mechanism of sleep induction differs from normal physiological sleep-inducing mechanisms by causing general central nervous system (CNS) suppression. By contrast, the orexin (hypocretin) system is a normal physiological mechanism for inducing wakefulness, acting via the action of orexin peptides that bind to orexin excitatory receptors present in key parts of the brain involved in wake regulation [2]. High endogenous levels of orexins are found in mammals during the active phase of the circadian cycle, and low levels are found during the inactive phase [3], [4]. Orexin receptor antagonists cause somnolence in rodents, dogs, and humans and have been evaluated as novel treatments for insomnia [5], [6].

In 2014, suvorexant became the first orexin receptor antagonist to be approved for treating insomnia [7], [8]. The maximum dose is 20 mg. Suvorexant has been shown to promote sleep in both healthy subjects at doses of 10–100 mg [9] and insomnia patients at doses of 10–80 mg [10], [11], [12]. In two pivotal phase 3 trials, suvorexant was effective at improving sleep onset and maintenance in elderly (15 or 30 mg) or non-elderly (20 or 40 mg) insomnia patients, and these benefits were maintained over three months of nightly dosing [12], and up to one year in a long-term phase 3 trial [11].

Studies of insomnia therapies frequently use polysomnography (PSG) recordings to objectively quantitate sleep and effects on sleep architecture. The usual method of analysis is to review the electroencephalograpy (EEG) signals in the PSG recordings and assign a sleep–wake stage score to each 30-s epoch using standardized sleep scoring rules [13], [14]. Evaluation of sleep architecture involves an assessment of the amount of time spent in non-rapid eye movement (NREM) sleep stages 1, 2, and 3/slow-wave sleep (SWS), and REM sleep. Previous studies in healthy subjects and insomnia patients have suggested that present z-drug (non-benzodiazepine) benzodiazepine receptor agonist treatments (eg, zolpidem) may reduce the amount of time spent in REM [15], [16], [17]. Studies of orexin receptor antagonists in healthy subjects and insomnia patients suggest that they increase the time spent in both REM and NREM sleep, and that they shorten REM latency [10], [17], [18].

EEG activity during sleep can also be quantitated by objectively calculating the power spectral density in various frequency bands (eg, delta, theta, alpha, sigma, beta, and gamma) directly from the EEG signal [19]. Previous studies have suggested that the current benzodiazepine receptor agonist treatments (eg, zolpidem) and other sleep treatments (eg, the antidepressant trazodone) change power spectral profiles in characteristic ways, whereas orexin receptor antagonists generally do not [17], [20]. For example, in a recent study which evaluated the power spectral profiles of various sleep agents in healthy subjects [20], zolpidem reduced activity in the lower-frequency ranges and increased activity in the middle-frequency ranges; trazodone reduced activity in the middle-frequency ranges and increased activity in the higher ranges; and suvorexant had no effects in any of the frequency ranges. The findings with suvorexant in healthy subjects were confirmed in insomnia patients using data from a phase 2b trial with sample sizes of approximately 60 patients per dose group [20].

The two pivotal phase 3 trials of suvorexant included PSG recording in a prespecified proportion of patients [12]. Suvorexant improved the primary PSG end points of sleep onset and sleep maintenance. The aim of this analysis was to describe the sleep architecture and power spectral effects of suvorexant over 3 months in patients with insomnia, using a large data set comprising pooled data from the two pivotal phase 3 trials.