14.1 Controlled Clinical Studies

DAYVIGO was evaluated in two clinical trials in patients with insomnia characterized by difficulties with sleep onset and/or sleep maintenance (Study 1, NCT02952820 and Study 2, NCT02783729).

Study 1 was a 6-month, randomized, double-blind, placebo-controlled, multi-center trial in adult patients age 18 or older who met DSM-5 criteria for insomnia disorder. Patients were randomized to placebo (n=325), DAYVIGO 5 mg (n=323), or DAYVIGO 10 mg (n=323) once nightly. The primary efficacy endpoint was the mean change from baseline to end of treatment at 6 months for log-transformed patient-reported (subjective) sleep onset latency (sSOL), defined as the estimated minutes from the time that the patient attempted to sleep until sleep onset. Pre-specified secondary efficacy endpoints for sleep maintenance were change from baseline to end of treatment at 6 months for patient-reported sleep efficiency (sSEF) and wake after sleep onset (sWASO). sSEF is defined as the proportion of time spent asleep per time in bed. sWASO is defined as the minutes of wake from the onset of sleep until wake time. The primary and pre-specified secondary efficacy endpoints were measured by sleep diary.

The demographic characteristics of patients in Study 1 were similar across the treatment arms. Patients had a median age of 55 years (range 18 to 88) and were 68% female, 72% White, 8% Black or African American, 17% Japanese, and 3.5% other; 28% were elderly (≥65 years).

Examination of subgroups by age, race, and sex did not suggest differences in response to DAYVIGO. In Study 1, DAYVIGO 5 mg and 10 mg demonstrated statistically significant superiority on the primary efficacy measure, sSOL, compared to placebo (Table 3). DAYVIGO 5 mg and 10 mg also showed statistically significant superiority in sSEF and sWASO.

Table 3: Primary and Secondary Efficacy Results for Change from Baseline in Sleep Onset and Sleep Maintenance at 6 Months in Patients with Insomnia (Study 1)

Endpoint	Treatment Group	Number of Patients ITT	Baseline Meana (SD)	Month 6 LS Meana (SE)	Treatment Effect (95% CI)
Sleep Onset sSOL (minutes)	DAYVIGO 5 mg*	316	43.0 (31.5)	20.0 (1.1)	0.7 (0.6, 0.8)
	DAYVIGO 10 mg*	315	45.0 (33.4)	19.2 (1.1)	0.7 (0.6, 0.8)
	Placebo	318	45.0 (31.8)	27.3 (1.4)	(ratio vs placebo)b
Sleep Maintenance sSEF (%)	DAYVIGO 5 mg*	316	63.1 (18.2)	75.9 (0.9)	4.5 (2.2, 6.9)
	DAYVIGO 10 mg*	315	62.0 (17.2)	75.9 (0.9)	4.7 (2.4, 7.0)
	Placebo	318	61.3 (17.8)	71.4 (0.8)	(%)c
Sleep Maintenance sWASO (minutes)	DAYVIGO 5 mg*	316	132.8 (82.5)	87.9 (3.7)	-17.5 (-27.3, -7.6)
	DAYVIGO 10 mg*	315	136.8 (87.4)	92.7 (3.7)	-12.7 (-22.4, -3.0)
	Placebo	318	132.5 (80.2)	105.3 (3.6)	(minutes)c
ITT (intention to treat); sSOL (subjective sleep onset latency); SD (standard deviation); LS (least squares); SE (standard error); CI (unadjusted confidence interval); sSEF (subjective sleep efficiency); sWASO (subjective wake after sleep onset) a For the sleep onset sSOL endpoint, the mean refers to geometric mean, which was used due to the approximately log normal distribution of the outcomes; SD for the geometric mean is calculated as GMSD (log transformed sSOL); SE for the least squares geometric mean is calculated in the same way as the SD. b For the sleep onset sSOL endpoint, treatment effect refers to the ratio of [Month 6 sSOL / Baseline sSOL] for DAYVIGO versus placebo, such that a smaller ratio corresponds to a greater improvement. c Treatment effect refers to the treatment difference between DAYVIGO versus placebo, such that a larger value for sSEF and smaller value for sWASO correspond to greater improvement. Doses that were statistically significantly superior (p<0.05) to placebo after multiplicity adjustment.

Study 2 was a 1-month, randomized, double-blind, placebo- and active-controlled, multi-center, parallel-group clinical trial in adult female patients age 55 and older and male patients 65 years and older who met DSM-5 criteria for insomnia disorder. Patients were randomized to placebo (n=208), DAYVIGO 5 mg (n=266) or 10 mg (n=269), or active comparator (n=263) once nightly.

The primary efficacy endpoint was the mean change in log-transformed latency to persistent sleep (LPS) from baseline to end of treatment (Days 29/30), as measured by overnight polysomnography (PSG) monitoring. LPS was defined as the number of minutes from lights off to the first 10 consecutive minutes of non-wakefulness. The pre-specified secondary efficacy endpoints in Study 2 were the mean change from baseline to end of treatment (Days 29/30) in sleep efficiency (SEF) and wake after sleep onset (WASO) measured by PSG.

The demographic and baseline characteristics of patients in Study 2 were similar across the treatment arms. Patients had a median age of 63 years (range 55 to 88) and were 86% female, 72% White, 25% Black or African American, and 2% other; 45% were elderly (≥65 years).

In Study 2, DAYVIGO 5 mg and 10 mg demonstrated statistically significant superiority on the primary efficacy measure, LPS, compared to placebo (Table 4). DAYVIGO 5 mg and 10 mg demonstrated statistically significant improvement in SEF and WASO compared to placebo.

Table 4: Primary and Secondary Efficacy Results for Change from Baseline in Sleep Onset and Sleep Maintenance at 1 Month in Patients with Insomnia (Study 2)

Endpoint	Treatment Group	Number of Patients ITT	Baseline Meana (SD)	Day 29/30 LS Meana (SE)	Treatment Effect (95% CI)
Sleep Onset LPS (minutes)	DAYVIGO 5 mg*	266	33.0 (27.2)	15.5 (0.8)	0.8 (0.7, 0.9)
	DAYVIGO 10 mg*	269	33.3 (27.2)	14.5 (0.7)	0.7 (0.6, 0.8)
	Placebo	208	33.6 (25.9)	20.0 (1.1)	(ratio vs. placebo)b
Sleep Maintenance SEF (%)	DAYVIGO 5 mg*	266	68.4 (11.3)	80.7 (0.5)	7.1 (5.6, 8.5)
	DAYVIGO 10 mg*	269	67.8 (10.8)	82.7 (0.5)	8.0 (6.6, 9.5)
	Placebo	208	68.9 (9.6)	74.6 (0.6)	(%)c
Sleep Maintenance WASO (minutes)	DAYVIGO 5 mg*	266	113.4 (39.0)	68.3 (2.2)	-24.0 (-30.0, -18.0)
	DAYVIGO 10 mg*	269	114.8 (40.0)	66.9 (2.2)	-25.3 (-31.4, -19.3)
	Placebo	208	111.7 (37.2)	92.2 (2.5)	(minutes)c
ITT (intention to treat); LPS (latency to persistent sleep); SD (standard deviation); LS (least squares); SE (standard error); CI (unadjusted confidence interval); SEF (sleep efficiency); WASO (wake after sleep onset) a For the sleep onset LPS endpoint, the mean refers to geometric mean, which was used due to the approximately log normal distribution of the outcomes; SD for the geometric mean is calculated as GMSD (log transformed LPS); SE for the least squares geometric mean is calculated in the same way as the SD. b For the LPS endpoint, treatment effect refers to the ratio of [Day 29/30 LPS / Baseline LPS] for DAYVIGO versus placebo, such that a smaller ratio corresponds to a greater improvement. c Treatment effect refers to the treatment difference between DAYVIGO versus placebo, such that a larger value for SEF and smaller value for WASO correspond to greater improvement. Doses that were statistically significantly superior (p<0.05) to placebo after multiplicity adjustment.

The effects of DAYVIGO at the beginning of treatment were generally consistent with later timepoints.