REVIEW OF: Moline M et al, J Clin Sleep Med 2021. Epub ahead of print.

TYPE OF STUDY: Randomized, double-blind, parallel-group study

As we age, sleep architecture worsens in ways that reduce sleep quality, particularly after age 55. Unfortunately, most hypnotics either do not improve sleep quality (eg, the z-hypnotics) or slightly worsen it (eg, the benzodiazepines). Specifically, benzodiazepines in higher doses can reduce stages of sleep that are critical for memory consolidation. Lemborexant (Dayvigo) was approved in 2019 for the treatment of insomnia in adults. Its effects are mediated by dual orexin receptor antagonism. Recently, researchers looked at how sleep architecture changed when older adults took this orexin antagonist.

The study was a secondary analysis of an industry-sponsored, randomized, double-blind trial of 1006 subjects known as SUNRISE 1. All subjects were over age 55 (average 63) and had primary insomnia; the majority were female (86%) and white (72%). The presence of sleep-maintenance insomnia was required; some subjects had sleep-onset insomnia as well. Patients with significant depression, anxiety, or substance use (including caffeine use after 6 pm) were excluded. Also excluded were patients with medical problems that could make hypnotic use unsafe, and those with comorbid sleep disorders (eg, sleep apnea, periodic limb movement disorder, restless legs syndrome, circadian rhythm sleep disorder, and narcolepsy).

Patients were randomly assigned to one of four conditions: lemborexant 5 mg, lemborexant 10 mg, zolpidem ER 6.25 mg, or placebo. Participants were treated for 30 nights followed by a follow-up period of 14 days before the end-of-study visit. The goal of this secondary analysis was to compare changes in sleep architecture (eg, total sleep time, non-REM sleep, and REM sleep) in older adults with insomnia disorder receiving lemborexant (5 or 10 mg), zolpidem, or placebo.

After a month of treatment, both doses of lemborexant outperformed placebo and zolpidem on measures of sleep architecture. Specific sleep architecture changes included increased total sleep time, non-REM sleep, and REM sleep, as well as reduced REM latency after the first 2 nights of treatment and again after a month of treatment. After 30 days, both doses of lemborexant increased total sleep time by about twice as much as placebo (total sleep increased by 30 minutes on placebo vs 64 minutes on lemborexant 5 mg and 69 minutes on 10 mg). Zolpidem, on the other hand, outperformed placebo in terms of total sleep time and non-REM sleep, but failed to reduce REM latency.

This study is weakened by being a secondary analysis, which makes it more prone to statistical error. Also, while lemborexant did improve sleep architecture, the changes were small and their clinical relevance is unclear. On the other hand, the changes in REM sleep seen here are relevant to age-related memory decline, and separate studies have found that lemborexant protects sleep-dependent memory consolidation in older adults with insomnia (Harand C et al, Front Neurol 2012;3:8).

The benefits seen in this study are consistent with prior research on another orexin antagonist, suvorexant, suggesting the orexin antagonists may have a beneficial class effect on sleep architecture (Snyder E et al, Sleep Med 2016;19:93–100).

TCPR’s Take
Quality is as important as quantity when it comes to sleep, particularly with older adults. Though this is only one study, it’s encouraging to see a hypnotic that improves sleep architecture and may have a positive impact on memory. Lemborexant also belongs to a small group of hypnotics that are relatively safe in older adults (along with suvorexant, ramelteon, and melatonin). Unfortunately, these benefits come at a cost of $10 a pill.