LAUSR

In Phase 3 Study 304 (SUNRISE-1; NCT02783729) and Study 303 (SUNRISE-2; NCT02952820), lemborexant (LEM) provided significant benefit versus placebo on sleep diary-based sleep onset/maintenance outcomes over 1mo and 6mo, respectively, in subjects with insomnia disorder. Both studies included the Insomnia Severity Index (ISI) and the Patient Global Impression–Insomnia version (PGI-I). On the PGI-I scale, subjects assess positive, neutral or negative treatment impact on falling asleep, overall benefit on sleep, and total sleep time. Using an anchor-based approach, ratings were compared with mean changes in ISI scores from baseline to evaluate what would be considered a responder definition on the ISI. Study 304 was a 1mo, randomized, double-blind, placebo- and active-controlled, parallel-group study in female (age ≥55y) and male (age ≥65y) subjects (n=1006); subjects received placebo, LEM 5mg, LEM 10mg, or zolpidem tartrate extended-release. Study 303 was a 12mo, randomized, double-blind study in subjects age ≥18y (n=950). Subjects received placebo, LEM 5mg, or LEM 10mg for 6mo. Data from both studies were pooled for the first month of treatment across all treatment groups. A modified ISI total score (ISI-ts) was used based on a confirmatory factor analysis that showed no incremental value to including Question 5 (How NOTICEABLE to others do you think your sleep problem is in terms of impairing the quality of your life?). The ‘Overall Sleep’ PGI-I item demonstrated that mean [SD] changes from baseline in ISI-ts decreased more in association with self-reported positive effects (−8.15 [4.98]) than for neutral (−3.59 [3.37]) or negative effects (−1.67 [2.91]) at Day 31/Month 1. For PGI-I Item ‘Time to Fall Asleep,’ mean (SD) changes from baseline in ISI-ts were −7.49 (5.10) for positive, −4.54 (4.13) for neutral, and −2.94 (4.04) for negative effects at Day 31/Month 1. For PGI-I Item ‘Total Sleep Time’ mean (SD) changes from baseline in ISI-ts were −8.15 (5.04) for positive, −3.74 (3.23) for neutral, and −2.40 (3.49) for negative effects at Day 31/Month 1. Results of this anchor-based approach using the PGI-I suggest that the responder definition for ISI-ts, using the modified ISI, should be approximately −8 points. Eisai Inc.