- Email: [email protected]
An evaluation of the efficacy and safety of eszopiclone over 12 months in patients with chronic primary insomnia
Sleep Medicine 6 (2005) 487–495 www.elsevier.com/locate/sleep
Original article
An evaluation of the efficacy and safety of eszopiclone over 12 months in patients with chronic primary insomnia Thomas Rotha,*, James K. Walshb, Andrew Krystalc, Thomas Wesseld, Timothy A. Roehrsa a
Henry Ford Hospital Sleep Center, 2799 West Grand Blvd, CFP-3, Detroit, MI 48202, USA b Sleep Medicine and Research Center, St Louis, MO, USA c Duke University Medical Center, Durham, NC, USA d Sepracor Inc., Marlborough, MA, USA Received 8 March 2005; received in revised form 8 June 2005; accepted 8 June 2005
Abstract Background and purpose: A double-blind placebo-controlled study of eszopiclone found significant, sustained improvement in sleep and daytime function. The 6-month open-label extension phase is described herein. Patients and methods: Adults (21–64) with primary insomnia who reported sleep duration !6.5 h/night or sleep latency O30 min/night were included. Patient-reported endpoints included sleep and daytime function. Safety and compliance were assessed at monthly clinic visits. The final double-blind month was used as the baseline for efficacy analyses of the open-label period. Results: Patients who were initially randomized to double-blind placebo and then switched to open-label eszopiclone (nZ111) significantly reported the following: (1) decreased sleep latency, wake time after sleep onset, and number of awakenings; (2) increased total sleep time and sleep quality; and (3) improved ratings of daytime ability to function, alertness and sense of physical well-being compared to baseline (P%0.0001 all monthly endpoints). There was no evidence of tolerance on any measure in either group. These subjects (nZ360) sustained the double-blind treatment gains for all sleep and daytime parameters, with further significant improvement in a number of measures. Eszopiclone was well tolerated in both groups; unpleasant taste was the only undesirable effect reported by O5% of patients. Conclusions: The significant improvements in sleep and daytime function were evident in those switched from double-blind placebo to 6 months of open-label eszopiclone therapy and were sustained during the 6 months of open-label treatment for those receiving prior doubleblind eszopiclone. During 12 months of nightly treatment, eszopiclone 3 mg was well tolerated; tolerance was not observed. q 2005 Elsevier B.V. All rights reserved. Keywords: Chronic insomnia; Sleep initiation and maintenance disorders; tolerance; Long-term; Non-benzodiazepine; Eszopiclone; Sleep latency; WASO; daytime function
1. Introduction The prevalence of chronic insomnia (i.e. 1 month or more) in population surveys is approximately 10–15% [1–5]. Although limited data regarding the natural course of chronic insomnia exist, available data suggest that untreated insomnia persists at least 2 years in more than half of the respondents [6–9]. Approximately 2.5% of the US population receive hypnotics for the treatment of insomnia each year, with * Corresponding author. Tel.: C1 313 916 5171; fax: C1 313 916 5167. E-mail address: [email protected] (T. Roth).
1389-9457/$ - see front matter q 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.sleep.2005.06.004
almost 25% of these people continuing nightly treatment for 4 months or longer [10]. However, the median duration of clinical trials evaluating hypnotic agents is only 1 week [11]; there are relatively few studies assessing the efficacy and safety of nightly hypnotic treatment for longer periods. Two large double-blind placebo-controlled studies of 5 weeks’ duration [12,13], 1 of 8 weeks’ duration [14], and one smaller study of 6 months’ duration [15] have been reported prior to our recently published 6-month placebocontrolled study [16] of nightly treatment with the nonbenzodiazepine eszopiclone (3 mg). The present report evaluates the results of a 6-month open-label extension of this study. Although none of these previous studies show evidence of pharmacologic tolerance, tolerance is persistently
488
T. Roth et al. / Sleep Medicine 6 (2005) 487–495
perceived to be characteristic of hypnotic insomnia treatment [17,18]. Given these perceptions, as well as the evidence that primary insomnia commonly lasts for a year or more and the paucity of well controlled long-term investigations, further scientific attention is needed. While the 6-month placebo-controlled data indicated that eszopiclone 3 mg was well tolerated and associated with sustained efficacy for virtually all measures of sleep and daytime function [16], the additional 6 months of open-label nightly use data provided a valuable opportunity to further assess the efficacy and safety of eszopiclone in adults with primary insomnia in up to a year of continuous nightly treatment. The present report evaluates the results of the 6-month open-label extension of the 6-month double-blind eszopiclone study. Analysis of these data allowed us to address several specific issues: (1) the continued (i.e. months 7–12) treatment response of patients randomized to eszopiclone during months 1–6; (2) the treatment response of patients switching from double-blind placebo to open-label eszopiclone; (3) the therapeutic effect of open-label therapy, which more closely approximates clinical practice; and (4) adverse effects during 12 months of nightly eszopiclone treatment.
2. Methods The methods of the 6-month double-blind phase of this investigation have been fully described in the original publication [16]. A synopsis of that information is presented below with details specific to the open-label extension phase. 2.1. Study patients and procedures Eligible patients were 21–64 years old (inclusive) with a Diagnostic and Statistical Manual of Mental Disorders—Fourth Edition (DSM-IV) diagnosis of primary insomnia. These patients were additionally required to report sleeping !6.5 h per night or having sleep latency O30 min each night for at least 1 month prior to screening. Individuals consuming more than two alcoholic beverages per day or who were using a hypnotic or any other medication known to affect sleep were excluded. All patients gave written informed consent, the institutional review board for each study site approved the protocol, and the study was carried out in accordance with the Declaration of Helsinki (1989). A total of 788 patients were randomized and received double-blind treatment, with 471 completing the doubleblind phase and entering the open-label extension. Fig. 1 provides details for attrition from the double-blind cohort. All patients who completed the double-blind phase were eligible to enter the 6-month open-label extension of nightly treatment with eszopiclone 3 mg. During the open-label period, patients continued to report to the study site at monthly intervals for assessments of safety, adverse events,
compliance, and for medication refills. A termination visit occurred within 7 days following the last dose of study medication. Patient-reported data were captured weekly using an interactive voice response system (IVRS). On the same day each week (G1 day) between 8 pm and midnight, patients called in to the IVRS and reported weekly averages of their sleep and daytime function. IVRS provides for predetermined and verifiable data collection intervals. 2.2. Study assessments Variables assessed weekly with IVRS included the following: (1) sleep latency; (2) wake time after sleep onset (WASO; patients were not explicitly instructed to report the final morning awakening, i.e. terminal insomnia, as WASO); (3) total sleep time (TST); (4) number of nightly awakenings; (5) number of nights awakened/week; (6) sleep quality; (7) daytime ability to function; (8) daytime alertness; and (9) sense of physical well-being. The last four of these variables were rated on a discrete Likert scale ranging from 0 to 10 with two semantic anchors. Daytime ability to function captured the patients’ ability to concentrate or think clearly over the past week (Likert anchors: poor, excellent); daytime alertness anchors were ‘very sleepy’ and ‘wide-awake and alert;’ and sense of anchors for physical well-being were ‘poor’ and ‘excellent’. All new or worsening signs and symptoms of illness, regardless of relation to study drug, were recorded as adverse events (AEs) at the monthly visits. Adherence was assessed by calculating the number of tablets taken during the study, divided by the number of days on the study multiplied by 100. The number of tablets taken was derived from the number of tablets dispensed minus the number of tablets returned to the investigator or reported as lost/stolen. 2.3. Statistical analysis The last assessment point (month 6) of the doubleblind phase was used as the baseline for all efficacy endpoints for all patients entering the open-label phase. Monthly values reflecting the average of 4 weeks per subject were calculated, as were monthly change from these baseline values. A last observation carried forward (LOCF) approach was used to impute missing values. This method ensured that subjects who discontinued early, including treatment failures, were included in the analysis. The analyses were conducted using ranked data, based on an a priori assumption that some of the dependent variables would not be normally distributed. The Signed Rank test was used to test the within-group change from baseline to each subsequent month of open-label treatment (e.g. month 7 vs 6, month 8 vs 6, month 9 vs 6, etc.). Between-groups comparisons were conducted using a ranked analysis of variance (ANOVA) with site as an explanatory variable to
T. Roth et al. / Sleep Medicine 6 (2005) 487–495
489
Randomized N=788
Placebo N=195
Discontinued AE Protocol violation Voluntary withdrawal Lost to follow-up Did not meet entry criteria Other
Eszopiclone N=593
85 (43.4%) 14 (7.1%) 7 (3.6%) 51 (26.0%) 8 (4.1%) 1 (0.5%) 4 (2.0%)
Completed 6 months of treatment N=111 (56.6%)
Entered Open-Label Period (PBO-ESZ) N=111
Discontinued AE Protocol violation Voluntary withdrawal Lost to follow-up Other
25 (22.5%) 7 (6.3%) 3 (2.7%) 6 (5.4%) 5 (4.5%) 4 (3.6%)
Completed 12 Months of Treatment (PBO-ESZ) N=86
Discontinued AE Protocol violation Voluntary withdrawal Lost to follow-up Did not meet entry criteria Other
235 (39.5%) 76 (12.8%) 17 (2.9%) 82 (13.8%) 52 (8.7%) 0 (0.0%) 8 (1.3%)
Completed 6 months of treatment N=360 (60.5%)
Entered Open-Label Period (ESZ-ESZ) N=360
Discontinued AE Protocol violation Voluntary withdrawal Lost to follow-up Other
64 (17.8%) 11 (3.1%) 6 (1.7%) 29 (8.1%) 10 (2.8%) 8 (2.2%)
Completed 12 Months of Treatment (ESZ-ESZ) N=296
Fig. 1. Patient disposition over the 12-month clinical trial.
compare the two groups at each open-label month. All tests were two-tailed at the 5% significance level. Analyses were performed on all patients entering the open-label phase. For reference, baseline values prior to entry into the doubleblind phase of the study are provided and referred to as ‘prestudy baseline.’ The open-label baseline is referred to as ‘month 6 baseline.’ The decision to analyze monthly averages (seven time-points) instead of the overall average across the nine endpoints increased the odds that a Type I error may have occurred. Although generally acknowledged as a conservative correction strategy, the Bonferroni method is often applied in situations such as this. Applying this correction to the present study would yield a P-value of 0.0008 (0.05/63).
3. Results 3.1. Patient disposition Of the 471 patients entering the open-label phase, 360 previously received eszopiclone (ESZ-ESZ group), and 111 previously received placebo (PBO-ESZ group). A total of 382 (81.0%) patients completed the open-label phase: 86 (77.5%) in the PBO-ESZ group and 296 (82.2%) in the ESZ-ESZ group. The most common reason for discontinuation was voluntary withdrawal (7.4%), followed by adverse events (3.8%) and loss to follow-up (3.2%). The proportion of patients who discontinued due to an adverse event was lower for patients in the ESZ-ESZ group (3.1%) than in
490
T. Roth et al. / Sleep Medicine 6 (2005) 487–495
Table 1 Pre-study baseline characteristics of patients in the 6-month open-label extension
Age in years Mean (SD) Median Range Race (%) Caucasian Black Asian Hispanic Gender (%) Female Male Body mass index (kg/m2) Mean (SD) Median Range
ESZ-ESZ (nZ360)
PBO-ESZ (nZ111)
45.7 (11.1) 46.5 21–64
45.1 (11.0) 45.0 21–65
80.8 12.5 1.1 5.0
79.3 12.6 2.7 5.4
61.4 38.6
66.7 33.3
29.4 (7.2) 28.1 17–58
27.4 (6.5) 26.0 16–48
No demographic parameters were significantly different; ESZ-ESZ: double-blind eszopiclone, open-label eszopiclone group; PBO-ESZ: double-blind placebo, open-label eszopiclone group.
the PBO-ESZ group (6.3%). The most common reasons for discontinuation due to AEs were unpleasant taste (2 patients, considered treatment-related) and anxiety (2 patients, considered unrelated to treatment). Adherence to the nightly treatment regimen for the open-label phase (96.0%) was similar to adherence during the double-blind period (94.4% for the eszopiclone group and 90.6% for the placebo group). Patient demographics were similar between groups (Table 1). 3.2. Sleep variables When compared to their baseline at month 6, PBO-ESZ patients had significantly decreased sleep latency (P! 0.0001; Fig. 2(a)), WASO (P!0.0005; Fig. 2(b)), number of awakenings (P!0.0001), and number of nights with an awakening (P!0.0001) at all monthly time-points. Additionally, there were significantly increased ratings for TST (P!0.0001, Fig. 2(c)) and sleep quality (P!0.0001) at all monthly time-points (Table 2). The median latency at month 12 represented a 47% decrease from month 6 and a 68% from pre-study. The median WASO at month 12 represented a 52% decrease from month 6 and a 68% decrease from pre-study. The median number of awakenings at month 12 represented a 25% decrease from month 6 and a 50% decrease from pre-study. The median TST at month 12 represented a 15% increase from month 6 and a 33% increase from pre-study. Similar findings were noted for number of nights awakened per week and sleep quality (Table 2). Compared to their baseline at month 6, ESZ-ESZ patients had decreased sleep latency (P!0.05 for months 8, 9, 10, and 12; Fig. 2(a)), WASO (P!0.05 for months 8–12;
Fig. 2(b)), number of awakenings (P!0.05 for months 9–12), and number of nights with an awakening (P!0.05 for months 8 and 10). This group also had an increase in TST (P!0.02; Fig. 2(c)) and sleep quality (P!0.01; Table 2) at all monthly time-points. The median latency at month 12 represented an 8% decrease from month 6 and a 55% from pre-study. The median WASO at month 12 represented a 29% decrease from month 6 and a 75% decrease from pre-study. The median number of awakenings at month 12 represented a 19% decrease from month 6 and a 57% decrease from pre-study. The median TST at month 12 represented a 4% increase from month 6 and a 33% increase from pre-study. There were no consistent differences between the ESZ-ESZ and PBO-ESZ groups during the open-label period (Table 2). 3.3. Daytime variables At all months relative to their month 6 baseline, the PBOESZ group had significantly improved daytime alertness (P!0.0001), ability to function (P!0.0001), and sense of physical well-being (P!0.0001; Table 3). The median daytime alertness rating at month 12 represented a 40% increase from pre-study. The median ability to function/ concentrate rating at month 12 represented a 22% increase from pre-study. The median physical well-being rating at month 12 represented a 17% increase from pre-study. Patients in the ESZ-ESZ group had significantly improved monthly ratings of daytime alertness (P!0.03 for all months), ability to function ratings (P!0.02 for all months except month 7), and sense of physical well-being (P!0.03 for all months) compared with their month 6 baseline (Table 3). The median daytime alertness rating at month 12 represented a 46% increase from pre-study. The median rating for ability to function/concentrate at month 12 represented a 30% increase from pre-study. The median physical well-being rating at month 12 represented a 17% increase from pre-study. There were no significant group differences for the parameters for daytime function at any monthly time-point (Table 3). 3.4. Safety Analysis of end-of-study hematology, blood chemistry, urinalysis, vital signs, electrocardiograms, and physical examination findings revealed no systematic differences during the open-label period. The incidence of all-causality AEs was 75% overall, with minimal between-group variation (76.6% for PBO-ESZ, 74.7% for ESZ-ESZ). Mild, moderate, and severe AEs occurred in 23.6, 41.2, and 10.4% of patients, respectively; with minimal group differences. The overall incidence of potentially treatment-related AEs was 31.4%, with higher rates in the PBO-ESZ group (44.1%) than in the ESZ-ESZ group (27.5%), due primarily to the incidence of unpleasant taste in the PBO-ESZ group (19.8 vs 2.8%). Among all 471 patients in the open-label phase, the most
T. Roth et al. / Sleep Medicine 6 (2005) 487–495
(a)
(b)
60 45 30 †
15
*
*
†
†
* †
†
†
PBO-ESZ
75
ESZ-ESZ
Minutes (median)
PBO-ESZ
75
Minutes (median)
491
ESZ-ESZ
60 45 30
*
*
*
*
*
†
†
†
†
*
†
7
8
9
10
11
12
15
0
0 Base 1
2
3
4
5
6
7
8
9
10
11
12
Base 1
2
3
4
Month of Study Open-Label Period
Double-Blind Period
Double-Blind Period
(c) 450
PBO-ESZ †
420
Minutes (median)
5
6
Month of Study Open-Label Period
ESZ-ESZ †
† †
† †
390 *
*
*
†
*
*
7
8
9
10
11
12
360 330 300 270 Base 1
2
3
4
5
6
Month of Study Double-Blind Period
Open-Label Period
Fig. 2. (a) Median sleep latency over the 12-month clinical trial; during the initial double-blind period (months 1–6), differences between eszopiclone and placebo were significant (P!0.0001); *P!0.05; †P%0.0001 for open-label period vs baseline (month 6) within treatment groups; base: pre-study baseline. (b) Median WASO over the 12-month clinical trial; during the initial double-blind period (months 1–6), differences between eszopiclone and placebo were significant (P%0.0032); *P!0.05; †P%0.0001 for open-label period vs baseline (month 6) within treatment groups; base: pre-study baseline. (c) Median TST over 12-month clinical trial; during the initial double-blind period (months 1–6), differences between eszopiclone and placebo were significant (P!0.0001); *P!0.05; †P%0.0001 for open-label period vs baseline (month 6) within treatment groups; base: pre-study baseline.
common treatment-related AEs were unpleasant taste (6.8%), headache (4.7%), somnolence (3.8%), abnormal dreams (3.0%), and dizziness (2.5%). Eleven patients reported a serious AE during the openlabel period (1.9%), with 10 considered unrelated to treatment and one determined to be potentially related (diabetes mellitus). The 10 unrelated events were chest pain (2 patients), accidental injury (2 patients), enlarged uterine fibroids (2 patients), anemia (1 patient), atrial fibrillation (1 patient), diabetes (1 patient), joint disorder (1 patient), and skin disorder (1 patient). Two of these events resulted in treatment discontinuation.
4. Discussion When considered with our prior publication [16], results of this study suggest that eszopiclone 3 mg was an effective and well tolerated hypnotic agent for up to 12 months of
continuous nightly usage in this cohort of adults with primary insomnia. Across the 12-month clinical trial, patients receiving eszopiclone 3 mg in both the doubleblind and open-label phases demonstrated consistent improvement in sleep and daytime variables. These results support the previous findings and extend the duration of sustained benefit without tolerance from 6 [16] to 12 months. PBO-ESZ patients reported significantly improved sleep and daytime function at all monthly time-points compared with the final month of double-blind treatment. Although the initial response to active treatment appeared to be slightly larger in the PBO-ESZ group vs the ESZ-ESZ group, these differences were not statistically significant. The continued improvement of the ESZ-ESZ group upon entering the open-label phase of the study was, however, not expected. These improvements occurred without any increase in study dosage or change in assessment methodology, and may have been related, at least in part, to the confirmation of receiving an active
492
T. Roth et al. / Sleep Medicine 6 (2005) 487–495
Table 2 Summary of sleep parameters Parameter
Time-point
ESZ-ESZ Mean (SD)
Sleep latency (min)
WASO (min)
# Awakenings/night
# Nts awakened/week
Total sleep time (min)
Sleep qualityc
PBO-ESZ Median
P vs OL baselinea
Mean (SD)
Median
P vs OL baselinea
P vs ESZESZb
Pre-study baseline
90.6 (79.6)
60.0
–
96.1 (94.7)
75.0
–
–
Month 6, OL baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12 Pre-study baseline Month 6, OL baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12 Pre-study baseline
47.0 (50.6) 44.8 (51.5) 40.8 (41.8) 39.8 (39.3) 42.5 (54.7) 44.3 (52.8) 45.9 (42.7) 83.2 (120.7) 44.2 (74.2) 35.2 (58.3) 34.2 (67.1) 36.7 (67.4) 39.4 (78.6) 34.6 (65.3) 40.1 (74.2) 3.2 (2.3)
30.0 28.8 27.5 27.0 27.3 28.8 27.5 60.0 21.0 17.0 15.0 15.0 14.0 13.9 15.0 3.0
– 0.2638 0.0315 0.0262 0.0570 0.2371 0.0057 – – 0.0853 0.0014 0.0260 0.0061 0.0061 0.0014 –
63.1 (57.9) 35.1 (38.4) 32.5 (30.0) 36.6 (44.5) 35.2 (34.2) 38.4 (45.8) 44.9 (39.5) 70.7 (72.8) 48.2 (59.4) 27.9 (43.9) 21.1 (24.9) 26.9 (42.7) 32.9 (55.7) 22.6 (31.5) 27.4 (35.1) 3.5 (2.8)
45.0 21.3 21.8 21.7 23.3 22.5 23.9 45.0 30.0 15.0 15.0 15.0 16.3 13.8 14.5 3.0
– %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 – – %0.0001 %0.0001 %0.0001 0.0003 %0.0001 %0.0001 –
– 0.0625 0.0874 0.2904 0.3381 0.1335 0.4273 – – 0.3719 0.5718 0.7437 0.5787 0.7194 0.9932 –
1.6 1.3 1.0 1.0 1.0 1.0 1.3 6.0
– 0.7213 0.1060 0.0085 0.0015 0.0074 0.0012 –
2.0 1.3 1.3 1.3 1.5 1.3 1.5 7.0
– %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 –
– 0.7009 0.2126 0.0994 0.0195 0.2295 0.0638 –
Month 6, OL baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12 Pre-study baseline
1.9 1.7 1.6 1.5 1.5 1.5 1.6 5.3
(1.5) (1.6) (1.5) (1.5) (1.9) (1.5) (1.4) (2.0)
2.6 1.6 1.8 1.7 1.7 1.6 1.8 5.6
(2.7) (1.3) (1.9) (1.8) (1.4) (1.5) (1.5) (1.8)
Month 6, OL baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12 Pre-study baseline
3.9 (2.5) 3.7 (2.6) 3.6 (2.7) 3.6 (2.7) 3.6 (2.7) 3.6 (2.7) 3.7 (2.6) 302.4 (123.2)
4.0 3.8 3.5 3.3 3.5 3.5 3.8 300.0
– 0.4322 0.0380 0.2540 0.0329 0.1480 0.3721 –
4.7 (2.4) 3.6 (2.5) 3.7 (2.6) 3.8 (2.6) 3.9 (2.6) 3.8 (2.7) 4.0 (2.5) 303.6 (78.3)
5.2 3.5 3.0 3.8 3.8 3.5 4.3 300.0
– %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 –
– 0.9943 0.3819 0.3658 0.1288 0.3646 0.3303 –
Month 6, OL baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12 Pre-study baseline
378.3 (72.3) 389.5 (72.0) 392.4 (71.6) 390.8 (73.4) 395.0 (73.1) 393.0 (71.4) 382.8 (73.9) 3.5 (2.0)
382.5 393.8 397.5 397.5 398.9 397.5 397.5 4.0
– 0.0123 0.0003 0.0047 !0.0001 0.0051 0.0008 –
339.3 (77.1) 400.6 (74.0) 406.0 (69.8) 412.4 (64.5) 399.1 (72.5) 397.8 (72.6) 381.8 (70.8) 3.5 (2.0)
345.0 410.0 416.5 419.3 408.0 408.3 398.0 4.0
– %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 –
– 0.1437 0.0368 0.0099 0.5822 0.4681 0.8524 –
6.5 7.3 7.0 7.3 7.0 7.0 7.0
– 0.0069 !0.0001 0.0003 0.0004 0.0003 !0.0001
5.5 7.0 7.0 7.2 7.0 7.0 7.0
– %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001
– 0.1223 0.8547 0.6647 0.6058 0.6448 0.3293
Month 6, OL baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12
5.5 6.8 6.9 6.9 6.9 6.9 6.6
(1.8) (1.7) (1.7) (1.7) (1.7) (1.7) (1.8)
6.4 7.2 7.0 7.1 6.9 7.0 6.5
(1.8) (1.4) (1.5) (1.4) (1.5) (1.6) (1.5)
Pre-study baseline parameters were not significantly different; OL, open-label; nts, nights. a Statistical tests compare mean ranked values at each monthly open-label time-point to the month 6, open-label baseline value within each group. b Statistical tests compare the mean ranks between the ESZ-ESZ and PBO-ESZ groups; 1-point. c 11-point Likert scale from 0Z poor to 10Zexcellent.
T. Roth et al. / Sleep Medicine 6 (2005) 487–495
493
Table 3 Summary of daytime function parameters Parameter
Daytime alertnessc
Ability to functiond
Sense of physical wellbeingd
Time-point
Pre-study baseline Month 6, OL Baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12 Pre-study baseline Month 6, OL Baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12 Pre-study baseline Month 6, OL baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12
ESZ-ESZ
PBO-ESZ
Mean (SD)
Median
4.6 (2.1)
5.0
6.5 (1.7)
6.8
6.8 (1.7) 7.0 (1.7) 7.0 (1.7) 6.9 (1.7) 7.0 (1.7) 6.8 (1.7) 5.6 (2.1)
7.0 7.3 7.3 7.3 7.3 7.3 5.0
6.8 (1.7)
7.0
7.0 (1.7) 7.2 (1.6) 7.1 (1.6) 7.1 (1.6) 7.2 (1.7) 7.0 (1.7) 5.9 (2.1)
P vs OL baselinea
P vs ESZESZb
Mean (SD)
Median
P vs OL baselineb
–
4.7 (2.0)
5.0
–
–
–
5.9 (1.7)
6.0
–
–
7.0 6.9 7.2 6.9 7.1 6.7 5.6
(1.4) (1.4) (1.3) (1.4) (1.5) (1.4) (1.8)
7.0 7.1 7.3 7.0 7.3 7.0 6.0
%0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 –
0.9308 0.3456 0.7490 0.4750 0.9900 0.2293 –
–
6.2 (1.8)
6.3
–
–
7.3 7.4 7.3 7.5 7.5 7.3 6.0
0.1172 0.0001 0.0201 0.0047 0.0002 0.0011 –
7.3 7.2 7.3 7.2 7.2 7.0 5.9
(1.3) (1.3) (1.3) (1.4) (1.5) (1.4) (2.0)
7.5 7.4 7.5 7.3 7.5 7.3 6.0
%0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 –
0.6114 0.3587 0.9345 0.5883 0.7325 0.3528 –
6.7 (1.7)
6.9
–
6.1 (1.8)
6.3
–
–
7.0 (1.5) 7.0 (1.5) 7.2 (1.4) 7.0 (1.4) 7.1 (1.4) 6.9 (1.5)
7.0 7.3 7.3 7.3 7.3 7.3
0.2617 0.0041 0.0449 0.0011 0.0013 0.0012
6.9 7.0 7.0 7.1 7.1 7.0
7.3 7.0 7.3 7.0 7.3 7.0
%0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001
0.8921 0.3745 0.8603 0.2472 0.6204 0.3043
0.0244 !0.0001 0.0002 0.0006 !0.0001 !0.0001 –
(1.7) (1.7) (1.7) (1.7) (1.7) (1.7)
Pre-study baseline parameters were not significantly different; OL, open-label. a Statistical tests compare mean ranked values at each monthly open-label time-point to the month 6, open-label baseline value within each group. b Statistical tests compare the mean ranks between the ESZ-ESZ and PBO-ESZ groups; 1-point. c 11-point Likert scale from 0Zvery sleepy to 10Zwide awake and alert. d 11-point Likert scale from 0Zpoor to 10Zexcellent.
treatment. Alternately, long-term hypnotic treatment may result in continuous improvement in a subset of patients. Although it is possible that selection or attrition-related factors could also have contributed to these improvements, the absence of differential dropout rates in the double-blind phase and the consistent pattern of results regardless of what population was analyzed based on the previously reported results (analyzed LOCF, observed case, and completer populations) suggest that these groups were similar. However, because of the absence of a control group it is difficult to determine the effect that these non-specific factors (including confirmation of an active agent) had on these results. The consistent efficacy of eszopiclone over time in adults with chronic primary insomnia is noteworthy, particularly given the evidence suggesting that many cases of insomnia persist long-term. The evidence for sustained efficacy seen in this study contrasts with the commonly held view that
hypnotics are unlikely to remain effective in the longer-term treatment of insomnia. There appeared to be no evidence of pharmacologic tolerance in the present trial, even though over half of the patients randomized to receive eszopiclone during the double-bind period received 12 months of continuous nightly eszopiclone. The primary aim of this 12-month clinical trial was to determine the long-term efficacy and safety of eszopiclone in the management of chronic insomnia. However, some issues related to long-term therapy were not addressed in this 12-month trial. For example, while these data support the sustained efficacy of eszopiclone without dose escalation, additional research is needed to evaluate drug discontinuation effects. The use of patient self-report and the absence of objective polysomnography (PSG) may be considered a limitation of the present study. Although insomnia sufferers often over-report their symptomatology compared with
494
T. Roth et al. / Sleep Medicine 6 (2005) 487–495
PSG, correlations between the two metrics are generally high [19,20], and poor sleepers appear to consistently overreport their sleep [21]. In addition, we would expect similar long-term efficacy results using PSG based on the data from a similarly designed 6-week trial (study included PSG and self-report) [22]. In summary, complementary PSG data in this study would have been valuable, but the value associated with long-term assessment would not likely outweigh the significant additional expense. Nightly administration of eszopiclone 3 mg for up to 12 months in adults with primary insomnia appeared to be well tolerated; there was no evidence of any systematic adverse effects that led to discontinuation or which represented a safety risk unique to long-term treatment. The absence of worsening rates of adverse effects across time, coupled with the corresponding absence of pharmacologic tolerance, provides intriguing preliminary evidence which contradicts the commonly held view that longer-term hypnotic medication use is inevitably dangerous and without sustained benefit. There is a growing consensus that a broader view of insomnia treatment outcome is needed. Although the randomized, double-blind, placebo-controlled design is necessary to determine efficacy and safety, it is also desirable to more closely mimic the clinical setting and assess clinical significance or effectiveness. Efficacy generally refers to the extent to which an intervention is superior on selected dependent variables in a controlled setting in a well-defined population. Effectiveness, on the other hand, is generally defined as how well a treatment works in a more naturalistic setting, including a variety of patient types, often utilizing measures that are more clinically relevant. Therefore, effectiveness trials assess how well a treatment produces a desired health outcome (cure, alleviation of pain, return of functional abilities) under conditions that closely approximate the clinical setting. The open-label phase of the present study more closely approximates the clinic environment than the standard double-blind clinical trial methodology and thus may provide insight into the magnitude of the therapeutic effect in actual clinical practice. In the clinical setting, patients expect to receive an active and effective treatment. Stated differently, in this setting, patients typically experience both placebo and pharmacologic effects. The continued improvement in the ESZ-ESZ group during the open-label phase and the numerically larger initial treatment response in the PBOESZ group (compared with the initial response in the ESZESZ group) suggest that efficacy results may underestimate treatment effectiveness. Although establishing treatment efficacy is an essential step in evaluating the value of any pharmacologic agent, determining the therapeutic benefits experienced by patients in the clinical environment is an equally important goal. The present data suggest that many patients report continually improved sleep over time and that these improvements
appear to result in improved daytime function. These results support the clinical utility of eszopiclone in the management of insomnia.
Acknowledgements Support for this study is provided by Sepracor Inc., Marlborough, MA. The authors wish to acknowledge Kendyl Schaefer and H. Heith Durrence of Sepracor Inc. (Marlborough, MA) for their assistance in preparing this manuscript.
References [1] Lichstein KL, Durrence HH, Riedel BW, et al. Epidemiology of sleep: age, gender, and ethnicity. Mahway, NJ: Lawrence Erlbaum Associates; 2004. [2] Ancoli-lsrael S, Roth T. Characteristics of insomnia in the United States: results of the 1991 National Sleep Foundation Survey. Sleep 1999;22:S347–S53. [3] Breslau N, Roth T, Rosenthal L, Andreski P. Sleep disturbance and psychiatric disorders: a longitudinal epidemiological study of young adults. Biol Psychiatry 1996;39:411–8. [4] Ford DE, Chamber DB. Epidemiologic study of sleep disturbance and psychiatric disorders: an opportunity for prevention? J Am Med Assoc 1989;262:1479–84. [5] Mellinger GD, Baiter MB, Uhlenhuth EH. Insomnia and its treatment. Arch Gen Psychiatry 1985;42:225–32. [6] Morgan K, Healey DW, Healey PJ. Factors influencing persistent subjective insomnia in old age: a followup study of good and poor sleepers aged 65–74. Age Ageing 1989;18:117–22. [7] Ganguli M, Reynolds CF, Gilby JE. Prevalence and persistence of sleep complaints in a real older community sample: the MOVIES project. J Am Geriatr Soc 1996;44:778–84. [8] Morgan K, Clarke D. Longitudinal trends in late-life insomnia: implications for prescribing. Age Ageing 1997;26:179–84. [9] Katz DA, McHorney CA. Clinical correlates of insomnia in patients with chronic illness. Arch Int Med 1998;158:1099–107. [10] Balter MB, Uhlenhuth EH. New epidemiologic findings about insomnia and its treatment. J Clin Psychiatry 1992;53(Suppl): 34–9. [11] Nowell PD, Mazumdar S, Buysse DJ, et al. Benzodiazepines and zolpidem for chronic insomnia: a meta-analysis of treatment efficacy. J Am Med Assoc 1997;278:2170–7. [12] Scharf MB, Roth T, Vogel GW, Walsh JK. A multicenter, placebocontrolled study evaluating zolpidem in the treatment of chronic insomnia. J Clin Psychiatry 1994;55:192–9. [13] Walsh JK, Vogel GW, Scharf M, et al. A five week, polysomnographic assessment of zaleplon 10 mg for the treatment of primary insomnia. Sleep Med 2000;1:41–9. [14] Morin CM, Colecchi C, Stone J, et al. Behavioral and pharmacological therapies for late-life insomnia: a randomized controlled trial. J Am Med Assoc 1999;281:991–9. [15] Oswald I, French C, Adam K, Gilham J. Benzodiazepine hypnotics remain effective for 24 weeks. Br Med J (Clin Res Ed) 1982;284: 860–3 [Mar 20]. [16] Krystal AD, Walsh JK, Laska E, et al. Sustained efficacy of eszopiclone over 6 months of nightly treatment: results of a randomized, double-blind, placebo-controlled study in adults with chronic insomnia. Sleep 2003;26:793–9.
T. Roth et al. / Sleep Medicine 6 (2005) 487–495 [17] Gillin JC, Byerley WF. The diagnosis and management of insomnia. New Eng J Med 1990;322:239–48. [18] Rickels K, Schweizer E, Case WG, Greenblatt DJ. Long-term therapeutic use of benzodiazepines. I. Effects of abrupt discontinuation. Arch Gen Psychiatry 1990;47:899–907. [19] Frankel BL, Coursey RD, Buchbinder R, Snyder F. Recorded and reported sleep in chronic primary insomnia. Arch Gen Psychiatry 1976;33:615–23.
495
[20] Morin CM, Kowatch RA, Barry T, Walton E. Cognitive-behavior therapy for late-life insomnia. J Consult Clin Psychol 1993;61:137–46. [21] Espie CA, Lindsay WR, Espie LC. Use of the sleep assessment device to validate insomniacs’ self-report of sleep pattern. J Psychopathol Behav Assess 1989;11:71–9. [22] Zammit G, Gillin JC, McNabb L, et al. Efficacy and safety of eszopiclone across 6-weeks of treatment for primary insomnia. Curr Med Res Opin 2004;20:1979–91.
Original article
An evaluation of the efficacy and safety of eszopiclone over 12 months in patients with chronic primary insomnia Thomas Rotha,*, James K. Walshb, Andrew Krystalc, Thomas Wesseld, Timothy A. Roehrsa a
Henry Ford Hospital Sleep Center, 2799 West Grand Blvd, CFP-3, Detroit, MI 48202, USA b Sleep Medicine and Research Center, St Louis, MO, USA c Duke University Medical Center, Durham, NC, USA d Sepracor Inc., Marlborough, MA, USA Received 8 March 2005; received in revised form 8 June 2005; accepted 8 June 2005
Abstract Background and purpose: A double-blind placebo-controlled study of eszopiclone found significant, sustained improvement in sleep and daytime function. The 6-month open-label extension phase is described herein. Patients and methods: Adults (21–64) with primary insomnia who reported sleep duration !6.5 h/night or sleep latency O30 min/night were included. Patient-reported endpoints included sleep and daytime function. Safety and compliance were assessed at monthly clinic visits. The final double-blind month was used as the baseline for efficacy analyses of the open-label period. Results: Patients who were initially randomized to double-blind placebo and then switched to open-label eszopiclone (nZ111) significantly reported the following: (1) decreased sleep latency, wake time after sleep onset, and number of awakenings; (2) increased total sleep time and sleep quality; and (3) improved ratings of daytime ability to function, alertness and sense of physical well-being compared to baseline (P%0.0001 all monthly endpoints). There was no evidence of tolerance on any measure in either group. These subjects (nZ360) sustained the double-blind treatment gains for all sleep and daytime parameters, with further significant improvement in a number of measures. Eszopiclone was well tolerated in both groups; unpleasant taste was the only undesirable effect reported by O5% of patients. Conclusions: The significant improvements in sleep and daytime function were evident in those switched from double-blind placebo to 6 months of open-label eszopiclone therapy and were sustained during the 6 months of open-label treatment for those receiving prior doubleblind eszopiclone. During 12 months of nightly treatment, eszopiclone 3 mg was well tolerated; tolerance was not observed. q 2005 Elsevier B.V. All rights reserved. Keywords: Chronic insomnia; Sleep initiation and maintenance disorders; tolerance; Long-term; Non-benzodiazepine; Eszopiclone; Sleep latency; WASO; daytime function
1. Introduction The prevalence of chronic insomnia (i.e. 1 month or more) in population surveys is approximately 10–15% [1–5]. Although limited data regarding the natural course of chronic insomnia exist, available data suggest that untreated insomnia persists at least 2 years in more than half of the respondents [6–9]. Approximately 2.5% of the US population receive hypnotics for the treatment of insomnia each year, with * Corresponding author. Tel.: C1 313 916 5171; fax: C1 313 916 5167. E-mail address: [email protected] (T. Roth).
1389-9457/$ - see front matter q 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.sleep.2005.06.004
almost 25% of these people continuing nightly treatment for 4 months or longer [10]. However, the median duration of clinical trials evaluating hypnotic agents is only 1 week [11]; there are relatively few studies assessing the efficacy and safety of nightly hypnotic treatment for longer periods. Two large double-blind placebo-controlled studies of 5 weeks’ duration [12,13], 1 of 8 weeks’ duration [14], and one smaller study of 6 months’ duration [15] have been reported prior to our recently published 6-month placebocontrolled study [16] of nightly treatment with the nonbenzodiazepine eszopiclone (3 mg). The present report evaluates the results of a 6-month open-label extension of this study. Although none of these previous studies show evidence of pharmacologic tolerance, tolerance is persistently
488
T. Roth et al. / Sleep Medicine 6 (2005) 487–495
perceived to be characteristic of hypnotic insomnia treatment [17,18]. Given these perceptions, as well as the evidence that primary insomnia commonly lasts for a year or more and the paucity of well controlled long-term investigations, further scientific attention is needed. While the 6-month placebo-controlled data indicated that eszopiclone 3 mg was well tolerated and associated with sustained efficacy for virtually all measures of sleep and daytime function [16], the additional 6 months of open-label nightly use data provided a valuable opportunity to further assess the efficacy and safety of eszopiclone in adults with primary insomnia in up to a year of continuous nightly treatment. The present report evaluates the results of the 6-month open-label extension of the 6-month double-blind eszopiclone study. Analysis of these data allowed us to address several specific issues: (1) the continued (i.e. months 7–12) treatment response of patients randomized to eszopiclone during months 1–6; (2) the treatment response of patients switching from double-blind placebo to open-label eszopiclone; (3) the therapeutic effect of open-label therapy, which more closely approximates clinical practice; and (4) adverse effects during 12 months of nightly eszopiclone treatment.
2. Methods The methods of the 6-month double-blind phase of this investigation have been fully described in the original publication [16]. A synopsis of that information is presented below with details specific to the open-label extension phase. 2.1. Study patients and procedures Eligible patients were 21–64 years old (inclusive) with a Diagnostic and Statistical Manual of Mental Disorders—Fourth Edition (DSM-IV) diagnosis of primary insomnia. These patients were additionally required to report sleeping !6.5 h per night or having sleep latency O30 min each night for at least 1 month prior to screening. Individuals consuming more than two alcoholic beverages per day or who were using a hypnotic or any other medication known to affect sleep were excluded. All patients gave written informed consent, the institutional review board for each study site approved the protocol, and the study was carried out in accordance with the Declaration of Helsinki (1989). A total of 788 patients were randomized and received double-blind treatment, with 471 completing the doubleblind phase and entering the open-label extension. Fig. 1 provides details for attrition from the double-blind cohort. All patients who completed the double-blind phase were eligible to enter the 6-month open-label extension of nightly treatment with eszopiclone 3 mg. During the open-label period, patients continued to report to the study site at monthly intervals for assessments of safety, adverse events,
compliance, and for medication refills. A termination visit occurred within 7 days following the last dose of study medication. Patient-reported data were captured weekly using an interactive voice response system (IVRS). On the same day each week (G1 day) between 8 pm and midnight, patients called in to the IVRS and reported weekly averages of their sleep and daytime function. IVRS provides for predetermined and verifiable data collection intervals. 2.2. Study assessments Variables assessed weekly with IVRS included the following: (1) sleep latency; (2) wake time after sleep onset (WASO; patients were not explicitly instructed to report the final morning awakening, i.e. terminal insomnia, as WASO); (3) total sleep time (TST); (4) number of nightly awakenings; (5) number of nights awakened/week; (6) sleep quality; (7) daytime ability to function; (8) daytime alertness; and (9) sense of physical well-being. The last four of these variables were rated on a discrete Likert scale ranging from 0 to 10 with two semantic anchors. Daytime ability to function captured the patients’ ability to concentrate or think clearly over the past week (Likert anchors: poor, excellent); daytime alertness anchors were ‘very sleepy’ and ‘wide-awake and alert;’ and sense of anchors for physical well-being were ‘poor’ and ‘excellent’. All new or worsening signs and symptoms of illness, regardless of relation to study drug, were recorded as adverse events (AEs) at the monthly visits. Adherence was assessed by calculating the number of tablets taken during the study, divided by the number of days on the study multiplied by 100. The number of tablets taken was derived from the number of tablets dispensed minus the number of tablets returned to the investigator or reported as lost/stolen. 2.3. Statistical analysis The last assessment point (month 6) of the doubleblind phase was used as the baseline for all efficacy endpoints for all patients entering the open-label phase. Monthly values reflecting the average of 4 weeks per subject were calculated, as were monthly change from these baseline values. A last observation carried forward (LOCF) approach was used to impute missing values. This method ensured that subjects who discontinued early, including treatment failures, were included in the analysis. The analyses were conducted using ranked data, based on an a priori assumption that some of the dependent variables would not be normally distributed. The Signed Rank test was used to test the within-group change from baseline to each subsequent month of open-label treatment (e.g. month 7 vs 6, month 8 vs 6, month 9 vs 6, etc.). Between-groups comparisons were conducted using a ranked analysis of variance (ANOVA) with site as an explanatory variable to
T. Roth et al. / Sleep Medicine 6 (2005) 487–495
489
Randomized N=788
Placebo N=195
Discontinued AE Protocol violation Voluntary withdrawal Lost to follow-up Did not meet entry criteria Other
Eszopiclone N=593
85 (43.4%) 14 (7.1%) 7 (3.6%) 51 (26.0%) 8 (4.1%) 1 (0.5%) 4 (2.0%)
Completed 6 months of treatment N=111 (56.6%)
Entered Open-Label Period (PBO-ESZ) N=111
Discontinued AE Protocol violation Voluntary withdrawal Lost to follow-up Other
25 (22.5%) 7 (6.3%) 3 (2.7%) 6 (5.4%) 5 (4.5%) 4 (3.6%)
Completed 12 Months of Treatment (PBO-ESZ) N=86
Discontinued AE Protocol violation Voluntary withdrawal Lost to follow-up Did not meet entry criteria Other
235 (39.5%) 76 (12.8%) 17 (2.9%) 82 (13.8%) 52 (8.7%) 0 (0.0%) 8 (1.3%)
Completed 6 months of treatment N=360 (60.5%)
Entered Open-Label Period (ESZ-ESZ) N=360
Discontinued AE Protocol violation Voluntary withdrawal Lost to follow-up Other
64 (17.8%) 11 (3.1%) 6 (1.7%) 29 (8.1%) 10 (2.8%) 8 (2.2%)
Completed 12 Months of Treatment (ESZ-ESZ) N=296
Fig. 1. Patient disposition over the 12-month clinical trial.
compare the two groups at each open-label month. All tests were two-tailed at the 5% significance level. Analyses were performed on all patients entering the open-label phase. For reference, baseline values prior to entry into the doubleblind phase of the study are provided and referred to as ‘prestudy baseline.’ The open-label baseline is referred to as ‘month 6 baseline.’ The decision to analyze monthly averages (seven time-points) instead of the overall average across the nine endpoints increased the odds that a Type I error may have occurred. Although generally acknowledged as a conservative correction strategy, the Bonferroni method is often applied in situations such as this. Applying this correction to the present study would yield a P-value of 0.0008 (0.05/63).
3. Results 3.1. Patient disposition Of the 471 patients entering the open-label phase, 360 previously received eszopiclone (ESZ-ESZ group), and 111 previously received placebo (PBO-ESZ group). A total of 382 (81.0%) patients completed the open-label phase: 86 (77.5%) in the PBO-ESZ group and 296 (82.2%) in the ESZ-ESZ group. The most common reason for discontinuation was voluntary withdrawal (7.4%), followed by adverse events (3.8%) and loss to follow-up (3.2%). The proportion of patients who discontinued due to an adverse event was lower for patients in the ESZ-ESZ group (3.1%) than in
490
T. Roth et al. / Sleep Medicine 6 (2005) 487–495
Table 1 Pre-study baseline characteristics of patients in the 6-month open-label extension
Age in years Mean (SD) Median Range Race (%) Caucasian Black Asian Hispanic Gender (%) Female Male Body mass index (kg/m2) Mean (SD) Median Range
ESZ-ESZ (nZ360)
PBO-ESZ (nZ111)
45.7 (11.1) 46.5 21–64
45.1 (11.0) 45.0 21–65
80.8 12.5 1.1 5.0
79.3 12.6 2.7 5.4
61.4 38.6
66.7 33.3
29.4 (7.2) 28.1 17–58
27.4 (6.5) 26.0 16–48
No demographic parameters were significantly different; ESZ-ESZ: double-blind eszopiclone, open-label eszopiclone group; PBO-ESZ: double-blind placebo, open-label eszopiclone group.
the PBO-ESZ group (6.3%). The most common reasons for discontinuation due to AEs were unpleasant taste (2 patients, considered treatment-related) and anxiety (2 patients, considered unrelated to treatment). Adherence to the nightly treatment regimen for the open-label phase (96.0%) was similar to adherence during the double-blind period (94.4% for the eszopiclone group and 90.6% for the placebo group). Patient demographics were similar between groups (Table 1). 3.2. Sleep variables When compared to their baseline at month 6, PBO-ESZ patients had significantly decreased sleep latency (P! 0.0001; Fig. 2(a)), WASO (P!0.0005; Fig. 2(b)), number of awakenings (P!0.0001), and number of nights with an awakening (P!0.0001) at all monthly time-points. Additionally, there were significantly increased ratings for TST (P!0.0001, Fig. 2(c)) and sleep quality (P!0.0001) at all monthly time-points (Table 2). The median latency at month 12 represented a 47% decrease from month 6 and a 68% from pre-study. The median WASO at month 12 represented a 52% decrease from month 6 and a 68% decrease from pre-study. The median number of awakenings at month 12 represented a 25% decrease from month 6 and a 50% decrease from pre-study. The median TST at month 12 represented a 15% increase from month 6 and a 33% increase from pre-study. Similar findings were noted for number of nights awakened per week and sleep quality (Table 2). Compared to their baseline at month 6, ESZ-ESZ patients had decreased sleep latency (P!0.05 for months 8, 9, 10, and 12; Fig. 2(a)), WASO (P!0.05 for months 8–12;
Fig. 2(b)), number of awakenings (P!0.05 for months 9–12), and number of nights with an awakening (P!0.05 for months 8 and 10). This group also had an increase in TST (P!0.02; Fig. 2(c)) and sleep quality (P!0.01; Table 2) at all monthly time-points. The median latency at month 12 represented an 8% decrease from month 6 and a 55% from pre-study. The median WASO at month 12 represented a 29% decrease from month 6 and a 75% decrease from pre-study. The median number of awakenings at month 12 represented a 19% decrease from month 6 and a 57% decrease from pre-study. The median TST at month 12 represented a 4% increase from month 6 and a 33% increase from pre-study. There were no consistent differences between the ESZ-ESZ and PBO-ESZ groups during the open-label period (Table 2). 3.3. Daytime variables At all months relative to their month 6 baseline, the PBOESZ group had significantly improved daytime alertness (P!0.0001), ability to function (P!0.0001), and sense of physical well-being (P!0.0001; Table 3). The median daytime alertness rating at month 12 represented a 40% increase from pre-study. The median ability to function/ concentrate rating at month 12 represented a 22% increase from pre-study. The median physical well-being rating at month 12 represented a 17% increase from pre-study. Patients in the ESZ-ESZ group had significantly improved monthly ratings of daytime alertness (P!0.03 for all months), ability to function ratings (P!0.02 for all months except month 7), and sense of physical well-being (P!0.03 for all months) compared with their month 6 baseline (Table 3). The median daytime alertness rating at month 12 represented a 46% increase from pre-study. The median rating for ability to function/concentrate at month 12 represented a 30% increase from pre-study. The median physical well-being rating at month 12 represented a 17% increase from pre-study. There were no significant group differences for the parameters for daytime function at any monthly time-point (Table 3). 3.4. Safety Analysis of end-of-study hematology, blood chemistry, urinalysis, vital signs, electrocardiograms, and physical examination findings revealed no systematic differences during the open-label period. The incidence of all-causality AEs was 75% overall, with minimal between-group variation (76.6% for PBO-ESZ, 74.7% for ESZ-ESZ). Mild, moderate, and severe AEs occurred in 23.6, 41.2, and 10.4% of patients, respectively; with minimal group differences. The overall incidence of potentially treatment-related AEs was 31.4%, with higher rates in the PBO-ESZ group (44.1%) than in the ESZ-ESZ group (27.5%), due primarily to the incidence of unpleasant taste in the PBO-ESZ group (19.8 vs 2.8%). Among all 471 patients in the open-label phase, the most
T. Roth et al. / Sleep Medicine 6 (2005) 487–495
(a)
(b)
60 45 30 †
15
*
*
†
†
* †
†
†
PBO-ESZ
75
ESZ-ESZ
Minutes (median)
PBO-ESZ
75
Minutes (median)
491
ESZ-ESZ
60 45 30
*
*
*
*
*
†
†
†
†
*
†
7
8
9
10
11
12
15
0
0 Base 1
2
3
4
5
6
7
8
9
10
11
12
Base 1
2
3
4
Month of Study Open-Label Period
Double-Blind Period
Double-Blind Period
(c) 450
PBO-ESZ †
420
Minutes (median)
5
6
Month of Study Open-Label Period
ESZ-ESZ †
† †
† †
390 *
*
*
†
*
*
7
8
9
10
11
12
360 330 300 270 Base 1
2
3
4
5
6
Month of Study Double-Blind Period
Open-Label Period
Fig. 2. (a) Median sleep latency over the 12-month clinical trial; during the initial double-blind period (months 1–6), differences between eszopiclone and placebo were significant (P!0.0001); *P!0.05; †P%0.0001 for open-label period vs baseline (month 6) within treatment groups; base: pre-study baseline. (b) Median WASO over the 12-month clinical trial; during the initial double-blind period (months 1–6), differences between eszopiclone and placebo were significant (P%0.0032); *P!0.05; †P%0.0001 for open-label period vs baseline (month 6) within treatment groups; base: pre-study baseline. (c) Median TST over 12-month clinical trial; during the initial double-blind period (months 1–6), differences between eszopiclone and placebo were significant (P!0.0001); *P!0.05; †P%0.0001 for open-label period vs baseline (month 6) within treatment groups; base: pre-study baseline.
common treatment-related AEs were unpleasant taste (6.8%), headache (4.7%), somnolence (3.8%), abnormal dreams (3.0%), and dizziness (2.5%). Eleven patients reported a serious AE during the openlabel period (1.9%), with 10 considered unrelated to treatment and one determined to be potentially related (diabetes mellitus). The 10 unrelated events were chest pain (2 patients), accidental injury (2 patients), enlarged uterine fibroids (2 patients), anemia (1 patient), atrial fibrillation (1 patient), diabetes (1 patient), joint disorder (1 patient), and skin disorder (1 patient). Two of these events resulted in treatment discontinuation.
4. Discussion When considered with our prior publication [16], results of this study suggest that eszopiclone 3 mg was an effective and well tolerated hypnotic agent for up to 12 months of
continuous nightly usage in this cohort of adults with primary insomnia. Across the 12-month clinical trial, patients receiving eszopiclone 3 mg in both the doubleblind and open-label phases demonstrated consistent improvement in sleep and daytime variables. These results support the previous findings and extend the duration of sustained benefit without tolerance from 6 [16] to 12 months. PBO-ESZ patients reported significantly improved sleep and daytime function at all monthly time-points compared with the final month of double-blind treatment. Although the initial response to active treatment appeared to be slightly larger in the PBO-ESZ group vs the ESZ-ESZ group, these differences were not statistically significant. The continued improvement of the ESZ-ESZ group upon entering the open-label phase of the study was, however, not expected. These improvements occurred without any increase in study dosage or change in assessment methodology, and may have been related, at least in part, to the confirmation of receiving an active
492
T. Roth et al. / Sleep Medicine 6 (2005) 487–495
Table 2 Summary of sleep parameters Parameter
Time-point
ESZ-ESZ Mean (SD)
Sleep latency (min)
WASO (min)
# Awakenings/night
# Nts awakened/week
Total sleep time (min)
Sleep qualityc
PBO-ESZ Median
P vs OL baselinea
Mean (SD)
Median
P vs OL baselinea
P vs ESZESZb
Pre-study baseline
90.6 (79.6)
60.0
–
96.1 (94.7)
75.0
–
–
Month 6, OL baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12 Pre-study baseline Month 6, OL baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12 Pre-study baseline
47.0 (50.6) 44.8 (51.5) 40.8 (41.8) 39.8 (39.3) 42.5 (54.7) 44.3 (52.8) 45.9 (42.7) 83.2 (120.7) 44.2 (74.2) 35.2 (58.3) 34.2 (67.1) 36.7 (67.4) 39.4 (78.6) 34.6 (65.3) 40.1 (74.2) 3.2 (2.3)
30.0 28.8 27.5 27.0 27.3 28.8 27.5 60.0 21.0 17.0 15.0 15.0 14.0 13.9 15.0 3.0
– 0.2638 0.0315 0.0262 0.0570 0.2371 0.0057 – – 0.0853 0.0014 0.0260 0.0061 0.0061 0.0014 –
63.1 (57.9) 35.1 (38.4) 32.5 (30.0) 36.6 (44.5) 35.2 (34.2) 38.4 (45.8) 44.9 (39.5) 70.7 (72.8) 48.2 (59.4) 27.9 (43.9) 21.1 (24.9) 26.9 (42.7) 32.9 (55.7) 22.6 (31.5) 27.4 (35.1) 3.5 (2.8)
45.0 21.3 21.8 21.7 23.3 22.5 23.9 45.0 30.0 15.0 15.0 15.0 16.3 13.8 14.5 3.0
– %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 – – %0.0001 %0.0001 %0.0001 0.0003 %0.0001 %0.0001 –
– 0.0625 0.0874 0.2904 0.3381 0.1335 0.4273 – – 0.3719 0.5718 0.7437 0.5787 0.7194 0.9932 –
1.6 1.3 1.0 1.0 1.0 1.0 1.3 6.0
– 0.7213 0.1060 0.0085 0.0015 0.0074 0.0012 –
2.0 1.3 1.3 1.3 1.5 1.3 1.5 7.0
– %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 –
– 0.7009 0.2126 0.0994 0.0195 0.2295 0.0638 –
Month 6, OL baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12 Pre-study baseline
1.9 1.7 1.6 1.5 1.5 1.5 1.6 5.3
(1.5) (1.6) (1.5) (1.5) (1.9) (1.5) (1.4) (2.0)
2.6 1.6 1.8 1.7 1.7 1.6 1.8 5.6
(2.7) (1.3) (1.9) (1.8) (1.4) (1.5) (1.5) (1.8)
Month 6, OL baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12 Pre-study baseline
3.9 (2.5) 3.7 (2.6) 3.6 (2.7) 3.6 (2.7) 3.6 (2.7) 3.6 (2.7) 3.7 (2.6) 302.4 (123.2)
4.0 3.8 3.5 3.3 3.5 3.5 3.8 300.0
– 0.4322 0.0380 0.2540 0.0329 0.1480 0.3721 –
4.7 (2.4) 3.6 (2.5) 3.7 (2.6) 3.8 (2.6) 3.9 (2.6) 3.8 (2.7) 4.0 (2.5) 303.6 (78.3)
5.2 3.5 3.0 3.8 3.8 3.5 4.3 300.0
– %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 –
– 0.9943 0.3819 0.3658 0.1288 0.3646 0.3303 –
Month 6, OL baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12 Pre-study baseline
378.3 (72.3) 389.5 (72.0) 392.4 (71.6) 390.8 (73.4) 395.0 (73.1) 393.0 (71.4) 382.8 (73.9) 3.5 (2.0)
382.5 393.8 397.5 397.5 398.9 397.5 397.5 4.0
– 0.0123 0.0003 0.0047 !0.0001 0.0051 0.0008 –
339.3 (77.1) 400.6 (74.0) 406.0 (69.8) 412.4 (64.5) 399.1 (72.5) 397.8 (72.6) 381.8 (70.8) 3.5 (2.0)
345.0 410.0 416.5 419.3 408.0 408.3 398.0 4.0
– %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 –
– 0.1437 0.0368 0.0099 0.5822 0.4681 0.8524 –
6.5 7.3 7.0 7.3 7.0 7.0 7.0
– 0.0069 !0.0001 0.0003 0.0004 0.0003 !0.0001
5.5 7.0 7.0 7.2 7.0 7.0 7.0
– %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001
– 0.1223 0.8547 0.6647 0.6058 0.6448 0.3293
Month 6, OL baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12
5.5 6.8 6.9 6.9 6.9 6.9 6.6
(1.8) (1.7) (1.7) (1.7) (1.7) (1.7) (1.8)
6.4 7.2 7.0 7.1 6.9 7.0 6.5
(1.8) (1.4) (1.5) (1.4) (1.5) (1.6) (1.5)
Pre-study baseline parameters were not significantly different; OL, open-label; nts, nights. a Statistical tests compare mean ranked values at each monthly open-label time-point to the month 6, open-label baseline value within each group. b Statistical tests compare the mean ranks between the ESZ-ESZ and PBO-ESZ groups; 1-point. c 11-point Likert scale from 0Z poor to 10Zexcellent.
T. Roth et al. / Sleep Medicine 6 (2005) 487–495
493
Table 3 Summary of daytime function parameters Parameter
Daytime alertnessc
Ability to functiond
Sense of physical wellbeingd
Time-point
Pre-study baseline Month 6, OL Baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12 Pre-study baseline Month 6, OL Baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12 Pre-study baseline Month 6, OL baseline Month 7 Month 8 Month 9 Month 10 Month 11 Month 12
ESZ-ESZ
PBO-ESZ
Mean (SD)
Median
4.6 (2.1)
5.0
6.5 (1.7)
6.8
6.8 (1.7) 7.0 (1.7) 7.0 (1.7) 6.9 (1.7) 7.0 (1.7) 6.8 (1.7) 5.6 (2.1)
7.0 7.3 7.3 7.3 7.3 7.3 5.0
6.8 (1.7)
7.0
7.0 (1.7) 7.2 (1.6) 7.1 (1.6) 7.1 (1.6) 7.2 (1.7) 7.0 (1.7) 5.9 (2.1)
P vs OL baselinea
P vs ESZESZb
Mean (SD)
Median
P vs OL baselineb
–
4.7 (2.0)
5.0
–
–
–
5.9 (1.7)
6.0
–
–
7.0 6.9 7.2 6.9 7.1 6.7 5.6
(1.4) (1.4) (1.3) (1.4) (1.5) (1.4) (1.8)
7.0 7.1 7.3 7.0 7.3 7.0 6.0
%0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 –
0.9308 0.3456 0.7490 0.4750 0.9900 0.2293 –
–
6.2 (1.8)
6.3
–
–
7.3 7.4 7.3 7.5 7.5 7.3 6.0
0.1172 0.0001 0.0201 0.0047 0.0002 0.0011 –
7.3 7.2 7.3 7.2 7.2 7.0 5.9
(1.3) (1.3) (1.3) (1.4) (1.5) (1.4) (2.0)
7.5 7.4 7.5 7.3 7.5 7.3 6.0
%0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001 –
0.6114 0.3587 0.9345 0.5883 0.7325 0.3528 –
6.7 (1.7)
6.9
–
6.1 (1.8)
6.3
–
–
7.0 (1.5) 7.0 (1.5) 7.2 (1.4) 7.0 (1.4) 7.1 (1.4) 6.9 (1.5)
7.0 7.3 7.3 7.3 7.3 7.3
0.2617 0.0041 0.0449 0.0011 0.0013 0.0012
6.9 7.0 7.0 7.1 7.1 7.0
7.3 7.0 7.3 7.0 7.3 7.0
%0.0001 %0.0001 %0.0001 %0.0001 %0.0001 %0.0001
0.8921 0.3745 0.8603 0.2472 0.6204 0.3043
0.0244 !0.0001 0.0002 0.0006 !0.0001 !0.0001 –
(1.7) (1.7) (1.7) (1.7) (1.7) (1.7)
Pre-study baseline parameters were not significantly different; OL, open-label. a Statistical tests compare mean ranked values at each monthly open-label time-point to the month 6, open-label baseline value within each group. b Statistical tests compare the mean ranks between the ESZ-ESZ and PBO-ESZ groups; 1-point. c 11-point Likert scale from 0Zvery sleepy to 10Zwide awake and alert. d 11-point Likert scale from 0Zpoor to 10Zexcellent.
treatment. Alternately, long-term hypnotic treatment may result in continuous improvement in a subset of patients. Although it is possible that selection or attrition-related factors could also have contributed to these improvements, the absence of differential dropout rates in the double-blind phase and the consistent pattern of results regardless of what population was analyzed based on the previously reported results (analyzed LOCF, observed case, and completer populations) suggest that these groups were similar. However, because of the absence of a control group it is difficult to determine the effect that these non-specific factors (including confirmation of an active agent) had on these results. The consistent efficacy of eszopiclone over time in adults with chronic primary insomnia is noteworthy, particularly given the evidence suggesting that many cases of insomnia persist long-term. The evidence for sustained efficacy seen in this study contrasts with the commonly held view that
hypnotics are unlikely to remain effective in the longer-term treatment of insomnia. There appeared to be no evidence of pharmacologic tolerance in the present trial, even though over half of the patients randomized to receive eszopiclone during the double-bind period received 12 months of continuous nightly eszopiclone. The primary aim of this 12-month clinical trial was to determine the long-term efficacy and safety of eszopiclone in the management of chronic insomnia. However, some issues related to long-term therapy were not addressed in this 12-month trial. For example, while these data support the sustained efficacy of eszopiclone without dose escalation, additional research is needed to evaluate drug discontinuation effects. The use of patient self-report and the absence of objective polysomnography (PSG) may be considered a limitation of the present study. Although insomnia sufferers often over-report their symptomatology compared with
494
T. Roth et al. / Sleep Medicine 6 (2005) 487–495
PSG, correlations between the two metrics are generally high [19,20], and poor sleepers appear to consistently overreport their sleep [21]. In addition, we would expect similar long-term efficacy results using PSG based on the data from a similarly designed 6-week trial (study included PSG and self-report) [22]. In summary, complementary PSG data in this study would have been valuable, but the value associated with long-term assessment would not likely outweigh the significant additional expense. Nightly administration of eszopiclone 3 mg for up to 12 months in adults with primary insomnia appeared to be well tolerated; there was no evidence of any systematic adverse effects that led to discontinuation or which represented a safety risk unique to long-term treatment. The absence of worsening rates of adverse effects across time, coupled with the corresponding absence of pharmacologic tolerance, provides intriguing preliminary evidence which contradicts the commonly held view that longer-term hypnotic medication use is inevitably dangerous and without sustained benefit. There is a growing consensus that a broader view of insomnia treatment outcome is needed. Although the randomized, double-blind, placebo-controlled design is necessary to determine efficacy and safety, it is also desirable to more closely mimic the clinical setting and assess clinical significance or effectiveness. Efficacy generally refers to the extent to which an intervention is superior on selected dependent variables in a controlled setting in a well-defined population. Effectiveness, on the other hand, is generally defined as how well a treatment works in a more naturalistic setting, including a variety of patient types, often utilizing measures that are more clinically relevant. Therefore, effectiveness trials assess how well a treatment produces a desired health outcome (cure, alleviation of pain, return of functional abilities) under conditions that closely approximate the clinical setting. The open-label phase of the present study more closely approximates the clinic environment than the standard double-blind clinical trial methodology and thus may provide insight into the magnitude of the therapeutic effect in actual clinical practice. In the clinical setting, patients expect to receive an active and effective treatment. Stated differently, in this setting, patients typically experience both placebo and pharmacologic effects. The continued improvement in the ESZ-ESZ group during the open-label phase and the numerically larger initial treatment response in the PBOESZ group (compared with the initial response in the ESZESZ group) suggest that efficacy results may underestimate treatment effectiveness. Although establishing treatment efficacy is an essential step in evaluating the value of any pharmacologic agent, determining the therapeutic benefits experienced by patients in the clinical environment is an equally important goal. The present data suggest that many patients report continually improved sleep over time and that these improvements
appear to result in improved daytime function. These results support the clinical utility of eszopiclone in the management of insomnia.
Acknowledgements Support for this study is provided by Sepracor Inc., Marlborough, MA. The authors wish to acknowledge Kendyl Schaefer and H. Heith Durrence of Sepracor Inc. (Marlborough, MA) for their assistance in preparing this manuscript.
References [1] Lichstein KL, Durrence HH, Riedel BW, et al. Epidemiology of sleep: age, gender, and ethnicity. Mahway, NJ: Lawrence Erlbaum Associates; 2004. [2] Ancoli-lsrael S, Roth T. Characteristics of insomnia in the United States: results of the 1991 National Sleep Foundation Survey. Sleep 1999;22:S347–S53. [3] Breslau N, Roth T, Rosenthal L, Andreski P. Sleep disturbance and psychiatric disorders: a longitudinal epidemiological study of young adults. Biol Psychiatry 1996;39:411–8. [4] Ford DE, Chamber DB. Epidemiologic study of sleep disturbance and psychiatric disorders: an opportunity for prevention? J Am Med Assoc 1989;262:1479–84. [5] Mellinger GD, Baiter MB, Uhlenhuth EH. Insomnia and its treatment. Arch Gen Psychiatry 1985;42:225–32. [6] Morgan K, Healey DW, Healey PJ. Factors influencing persistent subjective insomnia in old age: a followup study of good and poor sleepers aged 65–74. Age Ageing 1989;18:117–22. [7] Ganguli M, Reynolds CF, Gilby JE. Prevalence and persistence of sleep complaints in a real older community sample: the MOVIES project. J Am Geriatr Soc 1996;44:778–84. [8] Morgan K, Clarke D. Longitudinal trends in late-life insomnia: implications for prescribing. Age Ageing 1997;26:179–84. [9] Katz DA, McHorney CA. Clinical correlates of insomnia in patients with chronic illness. Arch Int Med 1998;158:1099–107. [10] Balter MB, Uhlenhuth EH. New epidemiologic findings about insomnia and its treatment. J Clin Psychiatry 1992;53(Suppl): 34–9. [11] Nowell PD, Mazumdar S, Buysse DJ, et al. Benzodiazepines and zolpidem for chronic insomnia: a meta-analysis of treatment efficacy. J Am Med Assoc 1997;278:2170–7. [12] Scharf MB, Roth T, Vogel GW, Walsh JK. A multicenter, placebocontrolled study evaluating zolpidem in the treatment of chronic insomnia. J Clin Psychiatry 1994;55:192–9. [13] Walsh JK, Vogel GW, Scharf M, et al. A five week, polysomnographic assessment of zaleplon 10 mg for the treatment of primary insomnia. Sleep Med 2000;1:41–9. [14] Morin CM, Colecchi C, Stone J, et al. Behavioral and pharmacological therapies for late-life insomnia: a randomized controlled trial. J Am Med Assoc 1999;281:991–9. [15] Oswald I, French C, Adam K, Gilham J. Benzodiazepine hypnotics remain effective for 24 weeks. Br Med J (Clin Res Ed) 1982;284: 860–3 [Mar 20]. [16] Krystal AD, Walsh JK, Laska E, et al. Sustained efficacy of eszopiclone over 6 months of nightly treatment: results of a randomized, double-blind, placebo-controlled study in adults with chronic insomnia. Sleep 2003;26:793–9.
T. Roth et al. / Sleep Medicine 6 (2005) 487–495 [17] Gillin JC, Byerley WF. The diagnosis and management of insomnia. New Eng J Med 1990;322:239–48. [18] Rickels K, Schweizer E, Case WG, Greenblatt DJ. Long-term therapeutic use of benzodiazepines. I. Effects of abrupt discontinuation. Arch Gen Psychiatry 1990;47:899–907. [19] Frankel BL, Coursey RD, Buchbinder R, Snyder F. Recorded and reported sleep in chronic primary insomnia. Arch Gen Psychiatry 1976;33:615–23.
495
[20] Morin CM, Kowatch RA, Barry T, Walton E. Cognitive-behavior therapy for late-life insomnia. J Consult Clin Psychol 1993;61:137–46. [21] Espie CA, Lindsay WR, Espie LC. Use of the sleep assessment device to validate insomniacs’ self-report of sleep pattern. J Psychopathol Behav Assess 1989;11:71–9. [22] Zammit G, Gillin JC, McNabb L, et al. Efficacy and safety of eszopiclone across 6-weeks of treatment for primary insomnia. Curr Med Res Opin 2004;20:1979–91.