0354 Effect of Lemborexant on Sleep Architecture in Subjects with Comorbid Insomnia and Mild Obstructive Sleep Apnea from a Ph 3 Trial

Abstract

Abstract Introduction Many patients with insomnia have co-morbid obstructive sleep apnea (COMISA). Patients with obstructive sleep apnea (OSA) may also have decreased REM-sleep. Lemborexant (LEM) is a dual-orexin-receptor-antagonist approved to treat adults with insomnia. LEM has been demonstrated to increase total sleep, with prominent increases in rapid eye movement (REM). Thus, sleep architecture was analyzed in subjects with insomnia/mild OSA receiving LEM in a phase 3 trial. Methods Study 304 (E2006-G000-304; NCT02783729) was a 1-month, randomized, double-blind, placebo (PBO)-controlled and active-comparator study in adults ≥55y with insomnia disorder. Subjects received PBO, LEM 5mg (LEM5), LEM 10mg (LEM10) or zolpidem tartrate extended-release 6.25mg (ZOL). Least-squares-mean (LSM) duration of each sleep stage (minutes) was compared across treatment groups from pooled data on Nights (NT)1/2 and NT29/30 for ­subjects with mild OSA (apnea hypopnea index ≥5 to < 15). Treatment-emergent adverse events (TEAEs) were recorded. Results The analysis set comprised 409 subjects (LEM5=114/LEM10=105/ZOL=112/PBO=78) with mild OSA. Mean (SD) change from baseline (CFB) in total sleep time was significantly larger (more sleep) in both LEM groups than ZOL or PBO at the post-randomization assessments. CFB for REM sleep was also significantly larger (P< 0.0001) for both LEM vs ZOL/PBO at both assessments, NT1/2: LEM5 21.08 (21.79), LEM10 32.57 (24.34), ZOL 3.06 (23.51), PBO 2.42 (17.51); NT29/20: LEM5 14.49 (18.12), LEM10 24.26 (24.50), ZOL 0.75 (22.32), PBO 2.14 (22.62). Compared with PBO, CFB for total nonREM sleep significantly increased (P< 0.0001) in the LEM groups at both assessments, NT1/2: LEM5 39.64 (34.30), LEM10 44.63 (37.85), ZOL 45.80 (37.43); PBO 7.80 (37.88); NT29/30: LEM5 44.66 (37.77), LEM10 44.08 (39.68), ZOL 36.22 (41.95), PBO 15.33 (36.52). Compared to ZOL, total nonREM sleep in either LEM group was not significantly different on NT1/2, and only LEM5 was significantly different vs ZOL at NT29/30 (P< 0.05). LEM was well-tolerated; most TEAEs were mild/moderate. Conclusion LEM significantly increased total sleep time in subjects with insomnia and mild OSA. The additional sleep was comprised importantly with REM sleep, a sleep stage that may be decreased in patients with OSA. These data support the use of LEM in a population with COMISA. Support (if any) Eisai Inc.