Abstract
Study Objectives:
Patients with chronic insomnia may respond differently to therapeutic modalities. This study examined differences in response of individuals with 2 insomnia phenotypes—short sleep duration (I-SSD; < 6 hours) and normal sleep duration (I-NSD; ≥ 6 hours) determined by polysomnography—to treatment with lemborexant and zolpidem tartrate extended-release 6.25 mg (zolpidem ER), compared with placebo.
Methods:
Study E2006-G000-304 (Study 304; SUNRISE-1; NCT02783729) was a global, randomized, double-blind, placebo, and active comparator-controlled, parallel-group study comparing lemborexant 5 and 10 mg in individuals aged ≥ 55 years with insomnia disorder. In this analysis, changes in subjective (self-reported) variables based on sleep diaries and objective variables based on polysomnographs were assessed after 1-month administration of study drugs. Data from participants with I-SSD and I-NSD were compared.
Results:
In the I-SSD subgroup, both lemborexant doses provided significant benefit for sleep-onset latency (SOL), total sleep time (TST), and wake after sleep onset (WASO) vs placebo; zolpidem ER also provided significant benefit for TST and WASO, but not SOL, on both measures vs placebo. In the I-NSD subgroup, lemborexant and zolpidem ER provided significant benefit for TST and WASO vs placebo objectively but not subjectively; both doses of lemborexant provided significant benefit for SOL vs placebo subjectively, but not objectively.
Conclusions:
Both drugs, but lemborexant more consistently, showed subjective and objective benefits compared with placebo in participants with insomnia with objective short sleep duration. However, neither lemborexant nor zolpidem provided consistent benefits for participants with normal sleep duration on sleep-onset and sleep maintenance variables.
Clinical Trial Registration:
Registry: ClinicalTrials.gov; Name: Study of the Efficacy and Safety of Lemborexant in Subjects 55 Years and Older With Insomnia Disorder (SUNRISE 1); URL: https://clinicaltrials.gov/ct2/show/record/NCT02783729; Identifier: NCT02783729.
Citation:
Inoue Y, Nishida M, Kubota N, et al. Comparison of the treatment effectiveness between lemborexant and zolpidem tartrate extended-release for insomnia disorder subtypes defined based on polysomnographic findings. J Clin Sleep Med. 2023;19(3):519–528.
Keywords: lemborexant, dual orexin receptor antagonist, zolpidem tartrate extended-release, insomnia with normal sleep duration, insomnia with short sleep duration
BRIEF SUMMARY
Current Knowledge/Study Rationale: Patients with chronic insomnia may present with short or normal sleep duration. This post hoc analysis examined whether lemborexant, a dual orexin receptor antagonist, and zolpidem extended-release, a benzodiazepine receptor agonist, provide favorable effects for individuals having insomnia with short sleep duration and/or those with normal sleep duration.
Study Impact: Both drugs, but more consistently with lemborexant, showed subjective (self-reported) and objective benefits compared with placebo in insomnia subjects with objective short sleep duration < 6 hours. However, neither of these drugs consistently conferred a benefit for patients with normal sleep duration ≥ 6 hours.
INTRODUCTION
Chronic insomnia is defined as a complaint of trouble initiating or maintaining sleep at least 3 nights per week for at least 3 months with impairment or daytime functioning.1 The prevalence of the disorder has been estimated to be 5–10%2 in industrialized nations and has been shown to give rise to work absenteeism and increased risk of psychiatric disorders or cardiovascular diseases, leading to increased health care costs.2 The diagnosis of insomnia disorder is made based only on subjective (self-reported) complaints and does not require objective sleep measures such as polysomnographic (PSG) findings.1
According to objective and subjective evaluations of sleep status, individuals with insomnia can be classified into 2 phenotypes: insomnia with short sleep duration (I-SSD) and insomnia with normal sleep duration (I-NSD).3–6 I-SSD refers to individuals who sleep less than 6 hours nightly, as evidenced by objective measures, whereas I-NSD refers to those who meet Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5), criteria for insomnia disorder, even though they have a relatively sufficient amount of nocturnal sleep (ie, objective sleep duration of ≥ 6 hours, even though a considerable number of individuals in this group underestimate their nocturnal sleep duration).3–5 Several previous studies have shown the association of I-SSD with medical conditions such as hypertension,7–9 diabetes,10 and cognitive impairment,11,12 as well as higher mortality risk.13,14 On the other hand, patients with I-NSD are reportedly associated with certain psychological profiles.15,16 To date, a specific treatment strategy based on classification of I-SSD and I-NSD has not been established.
The clinical practice guideline developed by the American College of Physicians indicates that cognitive behavioral therapies for insomnia (CBT-I) have been identified as a standard treatment for insomnia disorder17; however, pharmacotherapy is still considered an important treatment option.2 To date, drugs that exert their effect by acting as benzodiazepine receptor agonists (BzRAs) have been widely accepted as the mainstay drugs for the treatment of insomnia. However, hypnotics with novel mechanisms of action different from BzRAs, such as melatonin receptor agonists and dual orexin receptor antagonists, have been developed recently. Among these, lemborexant is a dual orexin receptor antagonist approved in multiple countries including the United States, Japan, Canada, Australia, and several Asian countries for the treatment of adults with insomnia.
In Study E2006-G000-304 (Study 304; SUNRISE 1; NCT02783729),18 one of the pivotal phase 3 studies of lemborexant, subjective (self-reported sleep diary) and objective (PSG) measures of sleep were evaluated concurrently in older patients with insomnia disorder to determine the efficacy and tolerability of 5-mg and 10-mg daily doses of the drug compared with placebo and zolpidem tartrate extended-release (zolpidem ER), a common BzRA. The study showed significant benefit after 1 month of treatment with either dose of lemborexant or zolpidem ER compared with placebo for both subjective and objective sleep parameters, except for objective sleep onset with zolpidem ER compared with placebo. Both therapies were well tolerated. To our best knowledge, there have been no randomized clinical trials other than Study 304 comparing the effects of insomnia treatments with different mechanisms of action using both sleep diaries and PSG. Therefore, we set out in this post hoc analysis of Study 304 to clarify whether both types of insomnia (I-SSD and I-NSD) responded differently to pharmacological treatments with dual orexin receptor antagonists and BzRAs, and whether improvements in objective PSG data and subjective sleep diary data are concordant in each group.
METHODS
Study population and design
The inclusion criteria of Study 304 have been previously described.18 Briefly, women aged ≥ 55 years and men aged ≥ 65 years were eligible for participation in the study if they met the DSM-5 criteria for insomnia disorder.1 The study was conducted at 67 sites in North America and Europe. Patients with mild depression (Beck Depression Inventory–II scores ≤ 19) and anxiety (Beck Anxiety Inventory scores ≤ 15) were eligible to participate in the study. Participants were required to have an Insomnia Severity Index score of 13 or higher and a history of subjective wake after sleep onset (sWASO) of ≥ 60 minutes at least 3 nights per week in the previous 4 weeks. Eligibility was confirmed by sleep history, sleep diary, and PSG findings (average value of wake after sleep onset [WASO] ≥ 60 minutes on averaged values of 2 consecutive PSGs, with neither night < 45 minutes).
The study was a global, multicenter, randomized, double-blind, placebo and zolpidem ER (6.25 mg; the generally recommended dose for women and elderly patients19) controlled, parallel-group study involving administration of 5 mg or 10 mg of lemborexant for 30 nights. Following an initial screening period, eligible participants received placebo for approximately 2 weeks to rule out placebo responders. All study medication was taken within 5 minutes of the time the participant intended to try to sleep. All participants were instructed to spend at least 7 hours in bed for sleep; PSGs were standardized at 8 hours.
The study protocol was approved by relevant institutional review boards and independent ethics committees. All protocol amendments were approved. The study adhered to Good Clinical Practice guidelines, the Declaration of Helsinki, and local regulations. All study participants provided written informed consent before taking part in this study.
Efficacy assessments
Efficacy assessments for these analyses were made based on the full analysis set as long as participants had sleep diary entries corresponding to 2 nights of PSG in order to align with the PSG assessment period. The full analysis set comprised all patients who received at least 1 dose of the study drug and had at least 1 postdose primary efficacy measurement. Participants were divided into 2 subgroups classified by total sleep time (TST) on PSGs based on the average of the 2 nights of baseline PSGs. We defined individuals with TST < 6 hours as the I-SSD group and those with TST ≥ 6 hours as the I-NSD group based on previous studies.5,7 Efficacy assessments included the averaged changes in subjective sleep-onset latency (sSOL), subjective TST (sTST), and sWASO on sleep diaries as well as changes in latency to persistent sleep (LPS), TST, and WASO on PSGs from night 29 and night 30 (nights 29/30) from baseline. In this study we compared the outcome variables among the participants who received lemborexant, those who received placebo, and those who received zolpidem ER in the I-SSD and I-NSD groups.
Statistical analysis
Details of the sample-size calculations and statistical analyses have been previously reported.18 Efficacy analyses were conducted on the full analysis set. For endpoints derived from sleep diaries (sSOL, sTST, and sWASO) and PSGs (LPS, TST, and WASO), reported values were calculated as the average of night 29 and night 30. The statistical comparisons of sTST, sWASO, TST, and WASO were made based on the least squares mean (LSM) of these variables. Because sSOL and LPS were non–normally distributed, a log-transformation was used for the analyses of these endpoints and statistical comparisons were made based on the least squares geometric means (LSGM) of these variables. In addition, the LSM with no log-transformation was also calculated for sSOL and LPS. The LSM and its 95% confidential interval (CI) were compared with the clinical significance threshold for favorable treatment outcome according to the American Academy of Sleep Medicine Clinical Practice Guideline for the Pharmacologic Treatment of Chronic Insomnia in Adults—that is, sSOL, 20 minutes; LPS, 10 minutes; sWASO and sTST, 30 minutes; WASO and TST, 20 minutes.2 The efficacy analyses of the subgroups followed the statistical methods previously reported.18 With regard to demographic characteristics, the variables were compared between the 2 subgroups using Wilcoxon test for continuous variables and chi-square test for categorical variables. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Inc, Cary, NC, USA).
RESULTS
Patient baseline demographics and characteristics
Of 3,537 participants screened, 1,006 were randomized to receive placebo (n = 208), zolpidem ER (n = 263), lemborexant 5 mg (n = 266), and lemborexant 10 mg (n = 269); 710 participants were categorized into the I-SSD group and 295 into the I-NSD group. The patients’ characteristics at baseline are shown in Table 1. The age and sex distributions between the subgroups were significantly different, although the absolute differences were small (age, P = .0028, Cohen’s d = 0.234789; sex, P < .0001, φ coefficient = 0.1351). There was no statistically significant difference by race. With regard to sleep parameters, not only objective measures (LPS, TST, WASO, and sleep efficiency) but also subjective measures (sSOL, sTST, sWASO, and subjective sleep efficiency) were significantly worse in the I-SSD group than in the I-NSD group (P < .0001 for all these sleep variables) (Table 2).
Table 1.
Baseline characteristics of participants in the two groups.
| I-SSD (TST < 6 Hours) (n = 710) | I-NSD (TST ≥ 6 Hours) (n = 295) | P† | |
|---|---|---|---|
| Age,* mean (SD), y | 64.39 (7.027) | 62.84 (6.147) | .0028 |
| Sex,‡ n (%) | <.0001 | ||
| Male | 118 (16.6) | 19 (6.4) | |
| Female | 592 (83.4) | 276 (93.6) | |
| Race, n (%) | .9328 | ||
| White | 512 (72.1) | 214 (72.5) | |
| Black or African American | 183 (25.8) | 73 (24.7) | |
| Asian | 8 (1.1) | 6 (2.0) | |
| Other | 7 (0.9) | 2 (0.6) | |
| BMI, mean (SD), kg/m2 | 27.37 (4.694) | 27.45 (4.842) | .8701 |
Age is calculated at date of informed consent. †Chi-square test. ‡In Study 304, 86.4% of the original participants were female.18 BMI = body mass index, I-NSD = insomnia with normal sleep duration, I-SSD = insomnia with short sleep duration, SD = standard deviation, TST = total sleep time.
Table 2.
Sleep variables at baseline.
| I-SSD (n = 710) | I-NSD (n = 295) | P* | |
|---|---|---|---|
| Sleep diary variables, mean (SD) | |||
| sSOL,† minutes | 72.08 (53.641) | 45.46 (33.269) | <.0001 |
| sTST,‡ minutes | 241.05 (98.046) | 293.46 (87.301) | <.0001 |
| sWASO,§ minutes | 186.61 (88.946) | 145.53 (77.188) | <.0001 |
| sSE,‡ % | 50.09 (20.324) | 61.05 (18.189) | <.0001 |
| Polysomnography sleep variables, mean (SD) | |||
| LPS, minutes | 53.84 (37.690) | 22.03 (12.335) | <.0001 |
| TST, minutes | 304.67 (45.110) | 382.46 (13.698) | <.0001 |
| WASO, minutes | 127.55 (37.925) | 80.32 (12.976) | <.0001 |
| SE, % | 63.53 (9.334) | 79.69 (2.859) | <.0001 |
Wilcoxon test. †n = 709 for I-SSD. ‡n = 693 for I-SSD, n = 294 for I-NSD. §n = 707 for I-SSD. I-NSD = insomnia with normal sleep duration, I-SSD = insomnia with short sleep duration, LPS = latency to persistent sleep, SD = standard deviation, SE = sleep efficiency, sSE = subjective sleep efficiency, sSOL = subjective sleep-onset latency, sTST = subjective total sleep time, sWASO = subjective wake after sleep onset, TST = total sleep time, WASO = wake after sleep onset.
Sleep-onset latency
In the I-SSD, the mean changes from baseline in sSOL after administration of the study drugs were significantly greater in the participants taking either dose of lemborexant compared with those taking placebo, whereas zolpidem ER did not show a statistical difference over placebo: LSGM treatment ratio vs placebo for lemborexant 5 mg, 0.861 (95% CI, 0.746, 0.994; P = .0417); lemborexant 10 mg, 0.717 (95% CI, 0.621, 0.827; P < .0001); and zolpidem ER, 0.890 (95% CI, 0.771, 1.029; P = .1148) (Figure 1A). LSM treatment difference vs placebo was –9.44 minutes (95% CI, –17.49, –1.39 minutes) with lemborexant 5 mg; –13.62 minutes (95% CI, –21.59, –5.64 minutes) with lemborexant 10 mg; and –8.41 minutes (95% CI, –16.48, –0.35 minutes) with zolpidem ER. Only the 95% CI of lemborexant 10 mg was in the range of the clinically significant threshold of 20 minutes suggested by the American Academy of Sleep Medicine. There was no significant difference in the change from baseline in sSOL between lemborexant 5 mg and zolpidem ER in the I-SSD group (P = .6257). However, the change from baseline in sSOL with lemborexant 10 mg was significantly larger than that with zolpidem ER (P = .0015) (Figure 1B).
Figure 1. Sleep-onset outcomes at nights 29/30.
Mean change from baseline in sSOL (A, B) and LPS (C, D). Due to the nonnormal distribution of LPS, the values were log-transformed, and the geometric mean ratio (Active Drug to Placebo or LEM to ZOL-ER) was used for testing statistically significant treatment differences. (A, C) LEM 5, LEM 10, and ZOL-ER compared with placebo. (B, D) LEM 5 and LEM 10 compared with ZOL-ER. CI = confidence interval, I-NSD = insomnia with normal sleep duration, I-SSD = insomnia with short sleep duration, LEM 5 = lemborexant 5 mg, LEM 10 = lemborexant 10 mg, LPS = latency to persistent sleep, LSGM = least-squares geometric means, sSOL = subjective sleep-onset latency, ZOL-ER = zolpidem extended-release.
With regard to LPS, in the I-SSD group, the mean changes from baseline in LPS were also significantly larger in the participants taking either dose of lemborexant compared with those taking placebo, whereas zolpidem ER did not show a statistical difference over placebo: LSGM treatment ratio vs placebo for lemborexant 5 mg, 0.745 (95% CI, 0.632, 0.879; P = .0005); lemborexant 10 mg, 0.675 (95% CI, 0.573, 0.795; P < .0001); and zolpidem ER, 1.160 (95% CI, 0.984, 1.368; P = .0777) (Figure 1C). The 95% CI of LSM difference vs placebo included the clinically significant threshold of 10 minutes as suggested by the American Academy of Sleep Medicine guideline for lemborexant 5 mg and 10 mg, but not for zolpidem ER: LSM treatment difference vs placebo for lemborexant 5 mg, –11.63 minutes (95% CI, –17.19, –6.06 minutes); lemborexant 10 mg, –16.18 minutes (95% CI, –21.70, –10.67 minutes); and zolpidem ER, 0.95 minutes (95% CI, –4.62, 6.52 minutes). The changes from baseline in LPS in the participants taking either dose of lemborexant were significantly larger compared with those taking zolpidem ER in the I-SSD group (lemborexant 5 mg, P < .0001; lemborexant 10 mg, P < .0001) (Figure 1D).
In the I-NSD group, the mean changes from baseline in sSOL were significantly larger in the patients taking either dose of lemborexant compared with those taking placebo, whereas zolpidem ER did not show a statistically significant difference over placebo: LSGM treatment ratio vs placebo for lemborexant 5 mg, 0.764 (95% CI, 0.606, 0.964; P = .0232); lemborexant 10 mg, 0.736 (95% CI, 0.582, 0.931; P = .0108); and zolpidem ER, 0.977 (95% CI, 0.771, 1.236; P = .8431) (Figure 1A). The lower limit of the 95% CI of LSM difference vs placebo was below the clinically significant threshold of 20 minutes for lemborexant 5 mg and 10 mg: LSM treatment difference vs placebo for lemborexant 5 mg, –13.86 minutes (95% CI, –23.54, –4.17 minutes); lemborexant 10 mg, –14.61 minutes (95% CI, –24.44, –4.78 minutes); for zolpidem ER, –2.11 minutes (95% CI, –11.98, 7.75 minutes). There were significant differences in the changes from baseline in sSOL between both doses of lemborexant compared with zolpidem ER (P < .05 for both doses) (Figure 1B). However, there were no differences in the changes from baseline in LPS of participants taking either dose of lemborexant vs placebo in the I-NSD group (lemborexant 5 mg, P = .2488; lemborexant 10 mg, P = .2785). The changes from baseline in LPS in the patients taking zolpidem ER were significantly larger compared with those taking placebo (LSGM treatment ratio vs placebo for zolpidem ER, 1.350; 95% CI, 1.028, 1.773; P = .0307) (Figure 1C). The 95% CI of LSM difference vs placebo reached the clinically significant threshold of 10 minutes for lemborexant 5 mg and 10 mg: LSM treatment difference vs placebo for lemborexant 5 mg, –7.21 minutes (95% CI, –12.47, –1.96 minutes); lemborexant 10 mg, –5.71 minutes (95% CI, –11.06, –0.37 minutes); and zolpidem ER, 2.12 minutes (95% CI, –3.27, 7.50 minutes). The changes from baseline in LPS were significantly larger in the participants taking either dose of lemborexant compared with those taking zolpidem ER in I-NSD group (lemborexant 5 mg, P < .05; lemborexant 10 mg, P < .05) (Figure 1D).
Total sleep time
In the I-SSD group, mean changes from baseline in sTST were significantly larger in participants taking either dose of lemborexant or zolpidem ER vs placebo: LSM treatment difference vs placebo for lemborexant 5 mg, 24.26 minutes (95% CI, 4.62 to 43.91 minutes; P = .0156); for lemborexant 10 mg, 31.64 minutes (95% CI, 12.19 to 51.10 minutes; P = .0015); and for zolpidem ER, 39.97 minutes (95% CI, 20.17 to 59.76 minutes; P < .0001), with either of the upper limits of the 95% CIs being above the clinically significant threshold of 30 minutes (Figure 2A). Between the participants taking lemborexant and those taking zolpidem ER, there was no difference in the change from baseline in sTST in this group (lemborexant 5 mg, P = .0956; lemborexant 10 mg, P = .3714) (Figure 2A). In the I-SSD group, mean changes from baseline in TST were also significantly larger in the participants taking either dose of lemborexant or zolpidem ER compared with those taking placebo: LSM treatment difference vs placebo for lemborexant 5 mg, 35.35 minutes (95% CI, 26.01, 44.68 minutes; P < .0001); lemborexant 10 mg, 41.97 minutes (95% CI, 32.71, 51.23 minutes; P < .0001); and zolpidem ER, 16.27 minutes (95% CI, 6.92, 25.61 minutes; P = .0007), with either of the upper limits of the 95% CIs being above the clinically significant threshold of 20 minutes (Figure 2C). Compared with the participants taking zolpidem ER, the changes from baseline in TST were significantly larger in those taking either dose of lemborexant in this group (lemborexant 5 mg, P < .0001; lemborexant 10 mg, P < .0001) (Figure 2D).
Figure 2. Total sleep time outcomes at nights 29/30.
The LSM difference (Active Drug to Placebo or LEM to ZOL-ER) in change from baseline in sTST (A, B) and TST (C, D). (A, C) LEM 5, LEM 10, and ZOL-ER compared with placebo. (B, D) LEM 5 and LEM 10 compared with ZOL-ER. CI = confidence interval, I-NSD = insomnia with normal sleep duration, I-SSD = insomnia with short sleep duration, LEM 5 = lemborexant 5 mg, LEM 10 = lemborexant 10 mg, LSM = least squares mean, sTST = subjective total sleep time, TST = total sleep time, ZOL-ER = zolpidem extended-release.
In the I-NSD group, there was no significant difference in the changes from baseline of sTST with either dose of lemborexant or zolpidem ER compared with placebo: LSM treatment difference vs placebo for lemborexant 5 mg, 8.66 minutes (95% CI, –13.69, 31.01 minutes; P = .4461); lemborexant 10 mg, 21.42 minutes (95% CI, –1.35, 44.20 minutes; P = .0651); and zolpidem ER, 10.53 minutes (95% CI, –12.28, 33.35 minutes; P = .3642) (Figure 2A). Although all the upper limits of the 95% CIs exceeded the clinically significant threshold of 30 minutes, the LSM difference fell much below the threshold with either of the treatment drugs. There was also no difference in the change from baseline of sTST in this group between either dose of lemborexant and zolpidem ER (lemborexant 5 mg, P = .8625; lemborexant 10 mg, P = .3239) (Figure 2B). On the other hand, the mean changes from baseline in TST were significantly larger in those taking either dose of lemborexant or zolpidem ER compared with those taking placebo in the I-NSD group: LSM treatment difference vs placebo for lemborexant 5 mg, 29.98 minutes (95% CI, 19.82, 40.13 minutes; P < .0001); lemborexant 10 mg, 30.45 minutes (95% CI, 20.11, 40.79 minutes; P < .0001); and zolpidem ER, 10.73 minutes (95% CI, 0.32, 21.14 minutes; P = .0433), with the upper limit of the 95% CIs exceeding the clinically significant threshold of 20 minutes (Figure 2C). Compared with zolpidem ER, the changes from baseline in TST were significantly larger in participants taking either dose of lemborexant in this group (lemborexant 5 mg, P < .05; lemborexant 10 mg, P < .0001) (Figure 2D).
WASO
In the I-SSD group, the mean changes in sWASO from baseline were significantly larger in participants taking either dose of lemborexant or zolpidem ER compared with those taking placebo: LSM treatment difference vs placebo for lemborexant 5 mg, –23.98 minutes (95% CI, –40.79, –7.17 minutes; P = .0052); lemborexant 10 mg, –23.26 minutes (95% CI, –39.93, –6.60 minutes; P = .0063); and zolpidem ER, –35.78 minutes (95% CI, –52.67, –18.89 minutes; P < .0001); and the 95% CI was in the range of the clinically significant threshold of 30 minutes with either of the treatments (Figure 3A). There were no differences in the change from baseline in sWASO between either dose of lemborexant and zolpidem ER (lemborexant 5 mg, P = .1443; lemborexant 10 mg, P = .1178) (Figure 3B). The mean changes from baseline in WASO were significantly larger in the participants taking either dose of lemborexant or zolpidem ER compared with those taking placebo in the I-SSD group: LSM treatment difference vs placebo for lemborexant 5 mg, –23.81 minutes (95% CI, –31.65, –15.97 minutes; P < .0001); lemborexant 10 mg, –26.40 minutes (95% CI, –34.17, –18.64 minutes; P < .0001); and zolpidem ER, –17.42 minutes (95% CI, –25.26, –9.58 minutes; P < .0001), with the lower limits of the 95% CIs below the clinically significant threshold of 20 minutes (Figure 3C). Compared with participants taking zolpidem ER, the changes from baseline in WASO were significantly larger in those taking lemborexant 10 mg but not in those taking 5 mg in the I-SSD group (lemborexant 5 mg, P = .0885; lemborexant 10 mg, P < .05) (Figure 3D).
Figure 3. Wake after sleep onset outcomes at nights 29/30.
The LSM difference (Active Drug to Placebo or LEM to ZOL-ER) in change from baseline in sWASO (A, B) and WASO (C, D). (A, C) LEM 5, LEM 10, and ZOL-ER compared with placebo. (B, D) LEM 5 and LEM 10 compared with ZOL-ER. CI = confidence interval, I-NSD = insomnia with normal sleep duration, I-SSD = insomnia with short sleep duration, LEM 5 = lemborexant 5 mg, LEM 10 = lemborexant 10 mg, LSM = least-squares mean, sWASO = subjective wakefulness after sleep onset, WASO = wake after sleep onset, ZOL-ER = zolpidem extended-release.
In the I-NSD group, there was no significant change from baseline in sWASO with either dose of lemborexant or zolpidem ER compared with placebo: LSM treatment difference vs placebo for lemborexant 5 mg, –2.84 minutes (95% CI, –22.91, 17.23 minutes; P = .7806); lemborexant 10 mg, –10.68 minutes (95% CI, –31.05, 9.69 minutes; P = .3030); and zolpidem ER, –12.47 minutes (95% CI, –32.96, 8.02 minutes; P = .2321) (Figure 3A). The lower limits of the 95% CIs for lemborexant 10 mg and zolpidem ER were below the clinically significant threshold of 30 minutes; however, the LSM difference did not reach the threshold with either of the treatments. There was also no difference in the change from baseline of sWASO between either dose of lemborexant and zolpidem ER in this group (lemborexant 5 mg, P = .3212; lemborexant 10 mg, P = .8564) (Figure 3B). The mean changes from baseline in WASO were significantly larger in the patients taking either dose of lemborexant or zolpidem ER vs placebo in the I-NSD group: LSM treatment difference vs placebo for lemborexant 5 mg, –23.04 minutes (95% CI, –31.89, –14.18 minutes; P < .0001); lemborexant 10 mg, –23.00 minutes (95% CI, –32.02, –13.98 minutes; P < .0001); and zolpidem ER, –12.57 minutes (95% CI, –21.61, –3.53 minutes; P = .0066); and the 95% CI of either treatment included the clinically significant threshold of 20 minutes (Figure 3C). Both doses of lemborexant showed a significantly larger change from baseline in the values compared with zolpidem ER (lemborexant 5 mg, P < .05; lemborexant 10 mg, P < .05) (Figure 3D).
DISCUSSION
This is the first analysis to compare the effects of lemborexant, a dual orexin receptor antagonist, and zolpidem ER, a BzRA, in a large sample of individuals with either I-SSD or I-NSD who were classified based on the cutoff TST value of 6 hours on PSG used in previous studies.5,20
In the present study, the phenomenon that most of the participants belonged to the I-SSD group might reflect the relatively strict inclusion criteria for Study 304. However, 6 hours of cutoff TST on PSG to define I-SSD in this study could be appropriate because the average TST of the general elderly population is reported to be 6.5 hours.21 The participants in the I-NSD group were found to be slightly but significantly younger than those in the I-SSD group. This might partly reflect the phenomenon that sleep structure becomes shallow or fragmented with aging.22 Moreover, our results appear to support previous reports showing that younger patients with insomnia have a greater discrepancy between sTST and TST (ie, longer TST).21,23 However, the effect size was small despite the significant difference between the 2 groups. Additionally, because the age criterion for inclusion in the current study was limited to ≥ 55 years, further studies with a wider age range would be needed to clarify this issue.
There was also a slightly higher percentage of women in the I-NSD group, which is consistent with previous studies showing that female patients with insomnia had shorter sTST relative to TST compared with male patients.24 This might indicate that women are more prone to have I-NSD. However, it should be noted that over 80% of the total participants were female in Study 304,18 similar to a larger proportion of female participants in previous clinical trials of other hypnotics.25,26 Moreover, in this study, the small effect size for this difference also made it difficult to confirm the difference between the 2 groups based on sex.
Interestingly, not only objective measures but also subjective measures in the I-SSD group were worse than those in the I-NSD group, suggesting that this group had substantial abnormality in sleep structure together with a certain sleep state misperception. On the other hand, the I-NSD group had a low severity of objective measures and larger differences between PSG and sleep diaries for all sleep parameters. In particular, the objective LPS of the I-NSD group was 22 minutes, on average, which was almost within the normal range,27 despite the presence of subjectively impaired sleep initiation.
The present study revealed that participants with I-SSD or I-NSD responded differently to the pharmacological treatments (Figure 4). Most important, in the I-SSD group, both doses of lemborexant and zolpidem ER 6.25 mg showed consistent benefit compared with placebo between PSG and self-reported sleep diary measures, except for sleep onset with zolpidem ER, which did not differ from placebo. Both treatments largely yielded effects satisfying the clinically meaningful differences of sleep measures2 in comparison with placebo, as the 95% CIs of the LSM differences vs placebo was in the range of the clinically significant threshold in most of the sleep measures with either of the treatments in the I-SSD group, except for sSOL with lemborexant 5 mg and that with zolpidem ER and LPS with zolpidem ER. This finding corroborates the idea that the effectiveness of pharmacological treatment on subjective measures is closely associated with the improvement in objective measures in the I-SSD group. Furthermore, the overall effect on objective measures was statistically stronger with lemborexant than with zolpidem ER (Figure 1D, Figure 2D, and Figure 3D), with the difference in LPS being much greater than the clinically meaningful difference, thus possibly indicating that lemborexant has a robust action for promoting sleep.28
Figure 4. Summary of the treatment outcomes with lemborexant and ZOL-ER in I-SSD and I-NSD groups.
I-NSD = insomnia with normal sleep duration, I-SSD = insomnia with short sleep duration, LEM5 = lemborexant 5 mg, LEM10 = lemborexant 10 mg, LPS = latency to persistent sleep, PBO = placebo, sSOL = subjective sleep-onset latency, sTST = subjective total sleep time, sWASO = subjective wakefulness after sleep onset, TST = total sleep time, WASO = wake after sleep onset, ZOL-ER = zolpidem extended-release.
In contrast, the I-NSD group showed considerable discrepancies between objective and subjective treatment responses to the study drugs; significant benefits compared with placebo on PSG variables do not seem to reflect the changes in subjective parameters. In particular, although both lemborexant and zolpidem ER showed significantly larger changes from baseline in both WASO and TST compared with placebo objectively, and the effects on these measures exceeded the above-indicated clinical meaningful threshold, these improvements did not reflect the changes in sWASO and sTST. With regard to sleep-onset measures, LPS did not show statistically significant changes with either drug, partly because of the nearly normal values at baseline in this group, as indicated above. On the other hand, changes in sSOL were significantly larger in individuals receiving either dose of lemborexant compared with those receiving placebo. This incongruence can be possibly explained by results of a previous study that suggested that patients with insomnia are likely to misperceive short and fragmented sleep during the early phase of nocturnal sleep as wakefulness—that is, increased WASO during this phase may lead to longer estimation of sSOL.29 Given this, the present results in which lemborexant showed statistically significant improvement in sSOL in the I-NSD group might come from a decrease in WASO during the early phase of nocturnal sleep.
The inconsistency between objective and subjective measurement in the I-NSD group may be explained by neurophysiological characteristics such as cortical hyperarousal during non–rapid eye movement sleep30 and alteration of sensory/cognitive processes,31 both of which may be involved in the mechanism of sleep state misperception. Therefore, it may be unlikely that subjective insomnia symptoms in this patient group would be improved by medications that target the sleep/wake regulation system per se.
In this study, neither lemborexant nor zolpidem ER could provide a certain benefit to subjective parameters in the I-NSD group. Only a limited number of studies have examined the effects of pharmacotherapy on paradoxical insomnia or I-NSD, and no clearly positive evidence has yet been established.32,33 However, it has been reported that CBT-I reduces the discrepancies between objective and subjective assessments for sleep onset and WASO in patients with insomnia.24 In addition, patients with I-NSD were reportedly more likely to respond to CBT-I compared with those with I-SSD.5,34 Given these observations, CBT-I can be expected to become an important treatment option for I-NSD. The present study showed that lemborexant provided benefit on subjective sleep onset; however, more studies in a larger number of patients would be needed to draw a conclusion.
Nevertheless, our results should be considered with the following potential limitations. First, the characteristics of participants in our randomized controlled trial might differ from those in clinical settings, and it is uncertain whether they are representative of the general population with insomnia. In addition, participants of Study 304 were predominately elderly females. Second, because these post hoc analyses were not part of the original design of Study 304, the numbers of patients in respective subgroups lacked power to show statistical significance. Thus, a forest plot was used to evaluate the consistency of treatment effect across subgroups, and P values assessing difference in treatment response among study drugs in the 2 subgroups were nominal. Third, this study comprised a 1-month trial period; therefore, a longer-term analysis would be needed. Finally, the dose of zolpidem ER was 6.25 mg in this study but can be increased to 12.5 mg if patients do not show subjective improvement in clinical settings. Therefore, an additional benefit should also be considered by increasing the dose of zolpidem ER to 12.5 mg.
In conclusion, the results of the present study suggest that patients with these 2 insomnia types respond differently to pharmacotherapy. In the I-SSD group, pharmacological treatments provided substantial benefits on subjective parameters that were mostly consistent with those of objective parameters, and the effectiveness was greater with lemborexant treatment. On the other hand, in participants with I-NSD, the changes in subjective assessments mostly did not reflect the changes in objective assessments with lemborexant or zolpidem ER administration, suggesting the necessity of alternative approaches including nonpharmacological therapy in individuals of this type.
DISCLOSURE STATEMENT
All authors have seen and approved the manuscript. This study was funded by Eisai, Inc. Editorial assistance was provided by Envision Pharma Group and was funded by Eisai, Inc, in compliance with international guidelines for Good Publication Practice (GPP3). Work for this study was performed at Eisai, Inc. Y.I. receives daily allowances (lecture fees, etc) from Eisai, MSD, and Takeda, along with research expenses from Astellas, Eisai, and Takeda. He is also affiliated with the donated courses provided by Astellas, Eisai, Merck, Sharp & Dohme, Otsuka, Koike Medical, Takeda, and Philips Japan. M.N., N.K., M.K., T.T., and K.I. are current employees, and K.M. a former employee, of Eisai Co, Ltd, Tokyo. M.M. is an employee of Eisai, Inc, Nutley, New Jersey.
ACKNOWLEDGMENTS
The authors thank all the patients, researchers, and facilities that participated in Study 304. Authors’ contributions: Y.I. contributed to interpretation of clinical data, conception and design of the study, analysis of data, and drafting the manuscript or figures. M.N., M.K., T.T. and K.M. contributed to interpretation of clinical data and drafting the manuscript or figures. N.K. contributed to the conception and design of the study. I.K. provided substantial contributions to data analysis. M.M. contributed to the conception and design of the study, interpretation of clinical data, and drafting the manuscript or figures. All authors have contributed to and approved this manuscript. Data availability: The datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request. Deidentified data will not be shared.
ABBREVIATIONS
- BzRA
benzodiazepine receptor agonist
- CI
confidence interval
- I-NSD
insomnia with normal sleep duration
- I-SSD
insomnia with short sleep duration
- LPS
latency to persistent sleep
- LSGM
least-squares geometric means
- LSM
least squares mean
- PSG
polysomnography/polysomnograhic
- SOL
sleep-onset latency
- sSOL
subjective sleep-onset latency
- sTST
subjective total sleep time
- sWASO
subjective wake after sleep onset
- TST
total sleep time
- WASO
wake after sleep onset
- zolpidem ER
zolpidem extended-release
REFERENCES
- 1. American Psychiatric Association . Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Arlington, VA: : American Psychiatric Association; ; 2013. . [Google Scholar]
- 2. Sateia MJ , Buysse DJ , Krystal AD , Neubauer DN , Heald JL . Clinical practice guideline for the pharmacologic treatment of chronic insomnia in adults: an American Academy of Sleep Medicine clinical practice guideline . J Clin Sleep Med. 2017. ; 13 ( 2 ): 307 – 349 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Lichstein KL . Insomnia identity . Behav Res Ther. 2017. ; 97 : 230 – 241 . [DOI] [PubMed] [Google Scholar]
- 4. Vgontzas AN , Fernandez-Mendoza J , Liao D , Bixler EO . Insomnia with objective short sleep duration: the most biologically severe phenotype of the disorder . Sleep Med Rev. 2013. ; 17 ( 4 ): 241 – 254 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Miller CB , Espie CA , Bartlett DJ , Marshall NS , Gordon CJ , Grunstein RR . Acceptability, tolerability, and potential efficacy of cognitive behavioural therapy for insomnia disorder subtypes defined by polysomnography: a retrospective cohort study . Sci Rep. 2018. ; 8 ( 1 ): 6664 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Trimmel K , Eder HG , Böck M , Stefanic-Kejik A , Klösch G , Seidel S . The (mis)perception of sleep: factors influencing the discrepancy between self-reported and objective sleep parameters . J Clin Sleep Med. 2021. ; 17 ( 5 ): 917 – 924 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Vgontzas AN , Liao D , Bixler EO , Chrousos GP , Vela-Bueno A . Insomnia with objective short sleep duration is associated with a high risk for hypertension . Sleep. 2009. ; 32 ( 4 ): 491 – 497 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Fernandez-Mendoza J , Vgontzas AN , Liao D , et al . Insomnia with objective short sleep duration and incident hypertension: the Penn State cohort . Hypertension. 2012. ; 60 ( 4 ): 929 – 935 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Bathgate CJ , Edinger JD , Wyatt JK , Krystal AD . Objective but not subjective short sleep duration associated with increased risk for hypertension in individuals with insomnia . Sleep. 2016. ; 39 ( 5 ): 1037 – 1045 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Vgontzas AN , Liao D , Pejovic S , Calhoun S , Karataraki M , Bixler EO . Insomnia with objective short sleep duration is associated with type 2 diabetes: a population-based study . Diabetes Care. 2009. ; 32 ( 11 ): 1980 – 1985 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Fernandez-Mendoza J , Calhoun S , Bixler EO , et al . Insomnia with objective short sleep duration is associated with deficits in neuropsychological performance: a general population study . Sleep. 2010. ; 33 ( 4 ): 459 – 465 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Fernandez-Mendoza J , He F , Puzino K , et al . Insomnia with objective short sleep duration is associated with cognitive impairment: a first look at cardiometabolic contributors to brain health . Sleep. 2021. ; 44 ( 1 ): zsaa150 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Sivertsen B , Pallesen S , Glozier N , et al . Midlife insomnia and subsequent mortality: the Hordaland health study . BMC Public Health. 2014. ; 14 ( 1 ): 720 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Fernandez-Mendoza J , He F , Calhoun SL , Vgontzas AN , Liao D , Bixler EO . Objective short sleep duration increases the risk of all-cause mortality associated with possible vascular cognitive impairment . Sleep Health. 2020. ; 6 ( 1 ): 71 – 78 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Fernandez-Mendoza J , Calhoun SL , Bixler EO , et al . Sleep misperception and chronic insomnia in the general population: role of objective sleep duration and psychological profiles . Psychosom Med. 2011. ; 73 ( 1 ): 88 – 97 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Edinger JD , Fins AI , Glenn DM , et al . Insomnia and the eye of the beholder: are there clinical markers of objective sleep disturbances among adults with and without insomnia complaints? J Consult Clin Psychol. 2000. ; 68 ( 4 ): 586 – 593 . [PubMed] [Google Scholar]
- 17. Qaseem A , Kansagara D , Forciea MA , Cooke M , Denberg TD ; Clinical Guidelines Committee of the American College of Physicians . Management of chronic insomnia disorder in adults: a clinical practice guideline from the American College of Physicians . Ann Intern Med. 2016. ; 165 ( 2 ): 125 – 133 . [DOI] [PubMed] [Google Scholar]
- 18. Rosenberg R , Murphy P , Zammit G , et al . Comparison of lemborexant with placebo and zolpidem tartrate extended release for the treatment of older adults with insomnia disorder: a phase 3 randomized clinical trial . JAMA Netw Open. 2019. ; 2 ( 12 ): e1918254 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Sanofi-Aventis US . Ambien CR (zolpidem tartrate) extended-release prescribing information. 2020. [last update: September 2020]. https://products.sanofi.us/ambien_cr/ambien_CR.pdf . Accessed June 19, 2022.
- 20. Castelnovo A , Ferri R , Punjabi NM , et al . The paradox of paradoxical insomnia: a theoretical review towards a unifying evidence-based definition . Sleep Med Rev. 2019. ; 44 : 70 – 82 . [DOI] [PubMed] [Google Scholar]
- 21. Van Den Berg JF , Van Rooij FJA , Vos H , et al . Disagreement between subjective and actigraphic measures of sleep duration in a population-based study of elderly persons . J Sleep Res. 2008. ; 17 ( 3 ): 295 – 302 . [DOI] [PubMed] [Google Scholar]
- 22. Stanley N . The physiology of sleep and the impact of ageing . Eur Urol Suppl. 2005. ; 3 ( 6 ): 17 – 23 . [Google Scholar]
- 23. Moon HJ , Song ML , Cho YW . Clinical characteristics of primary insomniacs with sleep-state misperception . J Clin Neurol. 2015. ; 11 ( 4 ): 358 – 363 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Kay DB , Buysse DJ , Germain A , Hall M , Monk TH . Subjective-objective sleep discrepancy among older adults: associations with insomnia diagnosis and insomnia treatment . J Sleep Res. 2015. ; 24 ( 1 ): 32 – 39 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Herring WJ , Connor KM , Ivgy-May N , et al . Suvorexant in patients with insomnia results from two 3-month randomized controlled clinical trials . Biol Psychiatry. 2016. ; 79 ( 2 ): 136 – 148 . [DOI] [PubMed] [Google Scholar]
- 26. Walsh JK , Soubrane C , Roth T . Efficacy and safety of zolpidem extended release in elderly primary insomnia patients . Am J Geriatr Psychiatry. 2008. ; 16 ( 1 ): 44 – 57 . [DOI] [PubMed] [Google Scholar]
- 27. Schutte-Rodin S , Broch L , Buysse D , Dorsey C , Sateia M . Clinical guideline for the evaluation and management of chronic insomnia in adults . J Clin Sleep Med. 2008. ; 4 ( 5 ): 487 – 504 . [PMC free article] [PubMed] [Google Scholar]
- 28. Beuckmann CT , Suzuki M , Ueno T , Nagaoka K , Arai T , Higashiyama H . In vitro and in silico characterization of lemborexant (E2006), a novel dual orexin receptor antagonist . J Pharmacol Exp Ther. 2017. ; 362 ( 2 ): 287 – 295 . [DOI] [PubMed] [Google Scholar]
- 29. Hermans LWA , Leufkens TR , van Gilst MM , et al . Sleep EEG characteristics associated with sleep onset misperception . Sleep Med. 2019. ; 57 : 70 – 79 . [DOI] [PubMed] [Google Scholar]
- 30. Krystal AD , Edinger JD , Wohlgemuth WK , Marsh GR . NREM sleep EEG frequency spectral correlates of sleep complaints in primary insomnia subtypes . Sleep. 2002. ; 25 ( 6 ): 630 – 640 . [PubMed] [Google Scholar]
- 31. Perlis ML , Giles DE , Mendelson WB , Bootzin RR , Wyatt JK . Psychophysiological insomnia: the behavioural model and a neurocognitive perspective . J Sleep Res. 1997. ; 6 ( 3 ): 179 – 188 . [DOI] [PubMed] [Google Scholar]
- 32. Khazaie H , Rezaie L , Darvishi F , Najafi F , Avis K . Treatment of paradoxical insomnia with atypical antipsychotic drugs. A comparison of olanzapine and risperidone [abstract]. Neurosciences (Riyadh). 2013. ; 18 ( 1 ): 64 – 69 . [PubMed] [Google Scholar]
- 33. Rezaie L , Fobian AD , McCall WV , Khazaie H . Paradoxical insomnia and subjective-objective sleep discrepancy: a review . Sleep Med Rev. 2018. ; 40 : 196 – 202 . [DOI] [PubMed] [Google Scholar]
- 34. Bathgate CJ , Edinger JD , Krystal AD . Insomnia patients with objective short sleep duration have a blunted response to cognitive behavioral therapy for insomnia . Sleep. 2017. ; 40 ( 1 ): zsw012 . [DOI] [PMC free article] [PubMed] [Google Scholar]




