0451 Effect of Lemborexant Treatment on Polysomnographic Sleep Measures in Older Adults with Insomnia and Objective Short Sleep

In Phase 3 Study 304 (NCT02783729), lemborexant (LEM) provided significant benefit versus placebo (PBO) on polysomnographic (PSG) and sleep diary-based sleep onset and maintenance outcomes over 1mo in subjects with insomnia disorder. Based on evidence that patients with insomnia and objective short sleep (ISS [total sleep time; TST] <6hrs) may respond less well to interventions such as cognitive behavioral therapy for insomnia (CBT-I) than patients with insomnia and objective long sleep (TST ≥6hrs), we conducted post-hoc analyses of LEM efficacy in the ISS subgroup (subjects with PSG TST<6hrs).

Study 304 was a 1mo, randomized, double-blind, PBO- and active-controlled, parallel-group study in female (age ≥55y) and male (age ≥65y) subjects (n=1006). Subjects received PBO, LEM 5mg (LEM5), LEM 10mg (LEM10), or zolpidem tartrate extended-release 6.25mg (ZOL). Latency to persistent sleep (LPS) and wake after sleep onset (WASO) were assessed at Nights (NT) 1/2 and NT29/30 using PSG and averaged; change from baseline (paired PSGs during single-blind PBO run-in) were analyzed using mixed-effect model repeated measurement analysis.

The ISS subgroup comprised 710/1006 (70.58%) subjects. Mean (SD) baseline LPS was similar across treatments: PBO=52.80(35.73); ZOL=54.77(40.93); LEM5=54.28(39.30); LEM10=53.31(34.45). On NT1/2, LEM5/10 led to statistically significantly greater (P<0.05) decreases from baseline (PBO=−9.65[36.52]; ZOL=−17.78[36.34]; LEM5=−22.02[31.62]; LEM10=−25.42[34.73]) versus PBO and ZOL. On NT29/30, LEM5/10 led to statistically significantly greater (P<0.0005) decreases from baseline (PBO=−11.88[35.09]; ZOL=−12.57[38.50]; LEM5=−25.37[37.06]; LEM10=−28.20[34.75] versus PBO. ZOL was not different from PBO. Mean (SD) baseline WASO was similar: PBO=123.79(37.21); ZOL=128.37(38.94); LEM5=128.14(37.52); LEM10=129.07(37.98). On NT1/2, LEM5/10 led to statistically significantly greater (P<0.0001) decreases from baseline (LSM[SE]: PBO=−23.52[2.74]; ZOL=−54.74[2.47]; LEM5=−60.58[2.45]; LEM10=−69.35[2.41] versus PBO. LEM10 was significantly different than ZOL (P<0.0001). On NT29/30, LEM5/10 led to statistically significantly greater (P<0.0001) decreases from baseline (PBO=−29.62[3.09]; ZOL=−46.86[2.80]; LEM5=−52.67[2.74]; LEM10=−55.37[2.70] versus PBO. LEM10 was significantly different than ZOL (P<0.05).

The data support LEM as an effective therapy for older adult patients with ISS and suggest LEM may be more beneficial than ZOL, particularly for patients with ISS and sleep onset difficulties. These findings suggest that LEM may be a reasonable therapy to consider for treating older patients with ISS where CBT-I may have relatively limited efficacy.

Eisai Inc.