The impact of lemborexant treatment on insomnia disease severity: results from a pooled analysis of two phase 3 studies

Document Type

Conference Proceeding

Publication Date


Publication Title

Sleep Med


Introduction: In clinical trials, polysomnographic and patient assessments of sleep onset and/or sleep maintenance relative to placebo (PBO) are generally used to evaluate the effectiveness of treatments for insomnia. An efficacious treatment should decrease insomnia disease severity, as measurable by the Insomnia Severity Index (ISI). Treatment with lemborexant (LEM) has demonstrated statistically significant benefits on sleep diary-based sleep onset latency and sleep maintenance variables at two doses (LEM 5 mg [LEM5]; LEM 10 mg [LEM10]) versus PBO in two Phase 3 studies. Here we present the results of a pooled analysis of the ISI from these 2 clinical studies after one month of treatment. Materials and methods: SUNRISE-1 was a 1-month, blinded, PBO- and active-controlled (zolpidem extended release [ZOL]; results not reported here), parallel-group study that randomized 1006 female and male subjects aged ≥55y with insomnia disorder, with baseline ISI total score (TS) ≥13 after PBO run-in. Subjects were randomized to PBO, ZOL, LEM5, or LEM10. SUNRISE-2 was a 12-month PBO-controlled (first 6 months), double-blind, parallel-group study that randomized 949 female and male subjects (full analysis set) aged ≥18y with insomnia disorder, with baseline ISI TS ≥15 after PBO run-in. Here we present pooled ISI TS data from both studies collected at baseline and the end of Month 1. Changes from baseline were analyzed by an analysis of covariance model, with baseline ISI TS as a covariate, and region, age group, and study as factors. Responder rates were analyzed by Cochran-Mantel-Haenszel test stratified by study, region and age group. Results: Mean baseline ISI TS for the pooled PBO (n=527), LEM5 (n=582), and LEM10 (n=584) treatment groups was 19.2, 19.3 and 19.0, respectively. For all treatment groups, mean ISI TS decreased at Month 1 relative to baseline. Decreases were significantly larger for the LEM5 and LEM10 groups versus the PBO group (LSM treatment difference: LEM5 -1.67, LEM10 -1.94; both P< 0.0001). The percentage of subjects whose ISI TS decreased by ≥7 points (considered a clinically meaningful change) was 33.6% in the PBO group versus 47.3% and 47.8% in the LEM5 and LEM10 groups, respectively. The percentage of subjects whose ISI TS was < 10 points at Month 1 (defined as the threshold for clinical insomnia) was 20.3% in the PBO group versus 33.0% and 33.4% in the LEM5 and LEM10 groups, respectively. These differences versus PBO were statistically significant (P< 0.0001). Lemborexant was well tolerated in both studies. Most treatment-emergent adverse events were mild to moderate. Conclusions: The severity of insomnia symptoms was significantly decreased in subjects treated with LEM. Additionally, in approximately 1/3 of LEM-treated subjects, ISI TS was reduced below the threshold for clinically important insomnia. Acknowledgements: Supported by Eisai Inc. and Purdue.





First Page


Last Page


This document is currently not available here.