SCIENTIFIC INVESTIGATIONS

Tiagabine Increases Slow-Wave Sleep in a Dose-Dependent Fashion Without Affecting Traditional Efcacy Measures in Adults With Primary Insomnia
James K. Walsh, Ph.D.1; Michael Perlis, Ph.D.2; Murray Rosenthal, D.O.3; Andrew Krystal, M.D.4; John Jiang, Ph.D.5; Thomas Roth, Ph.D.6
1 Sleep Medicine and Research Center, St. Johns/St. Lukes Hospitals, St. Louis, MO; 2Sleep and Neurophysiology Laboratory, Department of Psychiatry, University of Rochester, Rochester, NY; 3BMR HealthQuest, San Diego, CA; 4Department of Psychiatry and Behavioral Sciences, Duke University Medical Center, Durham, NC; 5Biometrics Department, Cephalon, Frazer, PA; 6Sleep Disorders and Research Center, Henry Ford Hospital, Detroit, MI

Introduction: This study evaluated dose-response effects of tiagabine on sleep in adults with primary insomnia. Methods: Men and women with primary insomnia (DSM-IV-TR) were randomly assigned to receive tiagabine 4, 6, 8, 10 mg or placebo in a randomized, double-blind, parallel-group study. Efcacy was assessed using polysomnography and self-report measures. Safety analyses included measures of residual sedation and adverse events. Results: A total of 232 patients (31% men; mean age 44.3 years) received study drug. No signicant differences were observed between tiagabine and placebo in wake after sleep onset, latency to persistent sleep, or total sleep time. Signicantly greater increases from baseline in slow-wave sleep (stages 3 and 4) were found with the 3 highest doses of tiagabine compared with placebo (p < .01). Stage 1 sleep showed a signicantly greater decrease from baseline for all doses of tiagabine than for placebo (p < .01). Self-report measures of sleep and daytime function did not differ from placebo, except for poorer ratings on the 10-mg dose.
Disclosure Statement This was an industry supported study supported by Cephalon, Inc. The data were analyzed by Dr. Jiang. The paper was written by the authots. Dr. Jiang is an employee of Cephalon, Inc. Dr. Walsh has received research support from Pzer, Inc., Merck & Co., Neurocrine Biosciences, and Cephalon, Inc.; and has provided consulting services to Abbott Labs, Pzer, Inc., SanoAventis, Cephalon, Inc., Organon, Neurocrine Biosciences, Takeda America, Actelion, Sepracor, Elan, Guilford, Respironics, Merck KgaA- Darmstadt, King, TransOral, GlaxoSmithKline, Sleep Tech, Eli Lilly, and Merck & Co. Dr. Krystal has received support for consulting, research, or being a member of the advisory board for Sano-Aventis, GlaxoSmithKline, King Pharmaceuticals, Inc., Wyeth, Merck & Co., Inc., Cephalon, Inc., Neurocrine Biosciences, Inc., Sepracor Inc., Pzer Inc., Johnson & Johnson, Organon, and Takeda Pharmaceuticals North America, Inc. Dr. Roth has received research support from Aventis, Cephalon, GlaxoSmithKline, Neurocrine, Pzer, Sano, Sepracor, Somaxon, Syrex, and Takeda; has participated in speaking engagements supported by Sano; and is a consultant for AstraZenca, Aventis, Cephalon, Cypres, Eli Lilly, GlaxoSmithKline, Hypnion, King, Lundbeck, McNeil, Merck, Neurocrine, Organon, Orginer, Pzer, Roche, Sano, Sepracor, Somaxon, Syrex, Takeda, Transoral, Vivometrics, and Wyeth. Dr. Rosenthal is the CEO of California Clinical Trials, and has received research support and participated in speaking engagements supported by Cephalon. Dr. Perlis has indicated no nancial conict of interest. Submitted for publication April 19, 2005 Accepted for publication June 30, 2005 Address correspondence to: James K. Walsh, Ph.D., Sleep Medicine and Research Center, 232 S. Woods Mill Road, Chestereld, MO 63017; Tel: (314) 205-6030; Fax: (314) 205-6025; E-mail: jwalsh@stlo.mercy.net
Journal of Clinical Sleep Medicine, Vol. 2, No. 1, 2006 35

Similarly, psychomotor performance on the 10-mg dose was worsened compared with placebo. Tiagabine was generally well tolerated; dizziness and nausea were the most common adverse events, particularly at the 2 higher doses. Conclusions: In adults with primary insomnia, tiagabine signicantly increased slow-wave sleep in a dose-dependent manner with a corresponding signicant decrease in Stage 1 sleep, whereas no signicant differences were observed in wake after sleep onset, latency to persistent sleep, or total sleep time compared with placebo. Keywords: Primary insomnia, slow-wave sleep, Stage 3 and 4 sleep, sleep disturbance, tiagabine Citation: Walsh JK; Perlis M; Rosenthal M et al. Tiagabine increases slow-wave sleep in a dose-dependent fashion, without affecting traditional efcacy measures, in adults with primary insomnia. J Clin Sleep Med 2006;2(1):35-41.

t the present time, all medications with an indication for the treatment of insomnia are benzodiazepine receptor agonists (BzRA); however, a variety of other drugs are commonly used to promote sleep. In fact, in 2 studies of medications used for insomnia, nonhypnotics were found to be used more commonly than were BzRA hypnotics.1,2 Regulatory issues rather than scientific evidence are suspected to contribute significantly to this practice.3 Insufficient research has been performed for most of the non-BzRA drugs used for insomnia to conclude that they are effective and safe. It is important, therefore, to systematically evaluate drugs with potential benefit in the treatment of insomnia. Tiagabine has recently received attention in the management of sleep disturbance associated with posttraumatic stress disorder4 and generalized anxiety disorder.5 The possibility that tiagabine may promote sleep is consistent with its mechanism of action. As a selective gamma aminobutyric acid (GABA) reuptake inhibitor, tiagabine increases synaptic GABA availability via selective inhibition of the GAT-1 GABA transporter.6,7 Since GABA is the primary inhibitory neurotransmitter, dose-related sedative effects would seem likely. Indeed, dizziness and somnolence, commonly noted with hypnotic medications, are 2 of the more common adverse events in epilepsy trials with tiagabine. Another reason to investigate the potential role of tiagabine and other non-BzRAs for the management of insomnia involves

JK Walsh, M Perlis, M Rosenthal et al

differences from BzRAs, which potentially have clinical benefit. The efficacy of all BzRAs in reducing sleep latency and most BzRAs in increasing total sleep time (TST) is well established, as is their safety. Efficacy for specific sleep maintenance measures is less well studied, but shorter-acting BzRAs have not been shown to consistently reduce wake after sleep onset (WASO) or number of awakenings. Moreover, some BzRAs reduce slow-wave sleep (SWS), a finding with no certain clinical consequence, but SWS has been associated with decreased arousability from sleep. This suggests that increased SWS may have positive effects on overall sleep quality, and drugs that increase SWS may have clinical utility. In preliminary studies, tiagabine has been shown to improve sleep maintenance and to significantly increase SWS in healthy elderly subjects and adult patients with primary insomnia.8-10 The present study evaluated dose-response effects of tiagabine on sleep in a large sample of adult patients with primary insomnia. METHODS Study Design and General Methods A double-blind, parallel-group, dose-response study of tiagabine was conducted in 30 centers in the United States, utilizing an identical research protocol. Four doses of tiagabine (4, 6, 8, and 10 mg) were compared with placebo in individuals with primary insomnia. Patient selection included a clinical assessment visit and 2 consecutive nights of polysomnography (PSG) with singleblind administration of placebo to ensure all study entry criteria were met. Patients who were randomly assigned to treatment or placebo groups returned for 2 consecutive nights of PSG on the assigned dose of study drug, followed by safety monitoring, prior to study discharge. The protocol was approved by an institutional review board for each center, and the study was performed in accordance with the Declaration of Helsinki and the Good Clinical Practice guidelines. Study Drug Tiagabine is rapidly absorbed, with a tmax of about 45 minutes in the fasting state.17 The rate, but not the extent, of absorption is reduced when ingested with food. Tiagabine has an elimination half-life of 7 to 9 hours. In the United States, tiagabine is approved as adjunctive therapy for partial seizures, with dose to be titrated starting at 4 mg per day.17 The maximum recommended daily dose is 56 mg per day for adults taking hepatic enzyme-inducing drugs. Patient Screening Procedures Men and women, aged 18 to 64 years, with chronic primary insomnia were eligible for participation in the study. Initial screening was performed via telephone. Prospective subjects, who appeared to meet entry criteria and continued to express interest in participation, were scheduled for a clinical evaluation at the study site. After giving written informed consent, each subject underwent medical, sleep, and psychiatric histories; a physical examination; laboratory tests; and an electrocardiogram. Patients were required to meet the following inclusion and exclusion criteria:
Journal of Clinical Sleep Medicine, Vol. 2, No. 1, 2006

INCLUSION (1) Diagnosis of primary insomnia according to Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition-criteria11, text revision (DSM-IV-TR) (2) Self-reported time to fall asleep of at least 15 minutes, wake time during the night of 60 minutes or longer, and total sleep duration of less than 6 hours, all on at least 3 nights each week during the preceding month (3) Self-report of 6.5 to 9 hours in bed per night, at least 5 nights per week during the preceding month (4) A habitual bedtime at least 5 nights per week of between 9:00 pm and midnight, varying no more than 1 hour. EXCLUSION (1) A diagnosis of any DSM-IV-TR axis-1 psychiatric disorder other than primary insomnia (2) Symptoms or signs of any primary sleep disorder other than primary insomnia (3) A history of substance abuse (4) A diagnosis of any disorder that may interfere with study drug pharmacokinetics (5) Performing shift work (6) Traveling across more than 3 time zones in the past 2 weeks (7) Taking any psychotropic medication, any cytochrome P3A4inducer/inhibitor, or any medication that may affect sleepwake function in the past 2 weeks (8) Napping 3 or more times each week over the preceding month (9) Consuming 300 mg of xanthines per day (10) Typically consuming more than 14 alcoholic units per week (11) Smoking more than 1 pack of cigarettes per day (12) A body mass index of 34 kg/m2 or greater (13) Known sensitivity to sedative hypnotics (14) Previously received tiagabine or had received an investigational drug within the last month PSG Screening Individuals meeting all clinical entry criteria were scheduled for 2 consecutive nights of PSG with single-blind placebo, administered 30 minutes prior to lights out. The scheduled lightsout time for each patient was within 30 minutes of his or her usual bedtime and was held constant for both screening and treatment PSGs. Time in bed was 480 minutes for all PSG nights. On the first screening night, patients were evaluated for sleep apnea and periodic leg movements and were excluded if the apnea-hypopnea index was greater than 10 or the periodic limb movement arousal index was 10 or more events per hour. Additional PSG entry criteria were (1) latency to persistent sleep (LPS) 10 minutes for both screening nights; (2) mean WASO of 45 minutes, with neither night < 30 minutes; and (3) TST 240 and 390 minutes on both nights. Experimental Procedures and Study Assessments After meeting all study entry criteria, randomly assigned patients were scheduled for PSGs on 2 consecutive nights within 2 to 10 days of the second screening PSG. On each night, the study drug was administered 30 minutes prior to PSG start time.
36

Tiagabine in Adults With Primary Insomnia

Patients randomized to treatment N=232

Tiagabine n=185

Safety analysis n=185

Placebo n=47

Safety analysis n=47

Efficacy analysis n=183

Efficacy analysis n=47

Completed study n=179 (97%)

Discontinued Adverse event (n=6)

Completed study n=46 (98%)

Discontinued Non-compliance (n=1)

Figure 1Patient disposition for those randomized to treatment.

Standard PSG techniques12 were used, and scoring of the PSG recording was performed at a centralized site, according to standard criteria.13 Self-reported sleep variables were assessed using a post-sleep questionnaire, completed 15 to 25 minutes after PSG termination. The Digit Symbol Substitution Test14 and sleepiness/alertness visual analog scale (0 mm = very sleepy, 100 mm = very alert) were both administered approximately 1 hour following each PSG. Patients recorded their ability to function, alertness, and physical well-being during the day by completing the Assessment of Daily Function questionnaire, comprising 3 visual analog scale ratings, at home on the evening prior to the second treatment night and on the following evening. Adverse events and vital signs were assessed throughout the study. Clinical laboratory tests, physical examination, and electrocardiograms were performed the day following the final PSG. Statistical Analysis For each patient, mean values for all variables from the 2 screening PSG visits were calculated and used as baseline values; means of variables from the 2 treatment PSG visits were also calculated for each patient. Change from baseline values were used in all analyses for PSG, self-report, Digit Symbol Substitution Test, and Assessment of Daily Function variables. If 1 of 2 measures was missing, the remaining value was used as the mean. Randomly assigned patients who had at least 1 baseline and 1 treatment PSG were included in the efficacy analysis. The primary efficacy variable was the change from baseline in PSG WASO following tiagabine or placebo. Other PSG measurements analyzed were LPS, TST, number of 30-second awakenings, minutes of sleep stages 1, 2, and 3+4 (i.e., SWS), and rapid eye movement (REM) sleep. Two additional variables were (1) sleep efficiency during sleep period time, which was calculated as TST divided by the sleep period, where sleep period time was time in bed minus the LPS; and (2) the ratio Stage 3+4 divided by Stage 1 + WASO. The latter ratio has previously been reported as a potential measure of sleep disruption, where an increase in value is indicative of less sleep disruption.15 In addition, Pearson correlation coefficients were calculated between the SWS (PSG) and self-reported ratings
Journal of Clinical Sleep Medicine, Vol. 2, No. 1, 2006 37

of sleep depth and of sleep quality, using both absolute numbers and mean changes, for each patient group and across all patient groups. Because study hypotheses addressed each dose of tiagabine compared with placebo, paired comparisons were the primary analyses and were performed using an analysis of covariance model, with treatment and center as factors, and baseline value as covariate. All paired comparison tests were 2-tailed, with set at .05. Parametric analyses findings were corroborated with nonparametric analyses. Safety analyses included all patients who received at least 1 dose of study medication. Adverse events were defined by Medical Dictionary for Regulatory Activities-preferred term.16 The Cochran-Armitage trend test was used to determine if adverse events were dose-related. RESULTS Patients A total of 822 patients were evaluated, and 590 were excluded primarily because of failure to meet entry criteria; 232 patients were randomly assigned (tiagabine: 4 mg, n = 46; 6 mg, n = 45; 8 mg, n = 45; 10 mg, n = 49; placebo, n = 47; Figure 1). All randomly assigned patients received at least 1 dose of study medica-

Table 1Demographic Characteristics for Each Treatment Group at Baseline Placebo Tiagabine Variable 4 mg 6 mg 8 mg 10 mg n=47 n=46 n=45 n=45 n=49 Mean age 44.3 (12.5) 44.9 (12.8) 44.6 (12.9) 40.7 (11.3) 47.2 (11.8) (SD), years Sex, n (%) Men 11 (23) 14 (30) 9 (20) 20 (44) 18 (37) Women 36 (77) 32 (70) 36 (80) 25 (56) 31 (63) Mean BMI 25.6 (4.1) 25.6 (4.0) 26.5 (4.4) 25.5 (3.4) 25.9 (3.1) (SD), kg/m2 BMI refers to body mass index.

JK Walsh, M Perlis, M Rosenthal et al Table 2Demographic Characteristics for Each Treatment Group at Baseline Placebo Variable 4 mg n=47 n=46 WASO, min Baseline 100.1 (31.4) 102.3 (29.7) Treatment 68.5 (27.8) 76.1 (34.0) LPS, min Baseline 44.3 (26.1) 56.2 (24.8) Treatment 34.5 (29.2) 41.9 (25.2) Baseline 342.1 (30.4) 328.2 (33.5) TST, min Treatment 384.0 (41.8) 367.8 (43.6) Number of awakenings Baseline 32.5 (11.4) 30.5 (12.4) Treatment 32.1 (9.5) 31.1 (12.0) Baseline 71.3 (6.3) 68.4 (7.0) Sleep efficiency, % Treatment 80.0 (8.7) 76.6 (9.1) Stage 1, min Baseline 35.8 (16.4) 39.3 (22.5) Treatment 37.6 (17.4) 33.6** (19.0) Stage 2, min Baseline 207.1 (29.0) 193.3 (28.7) Treatment 225.1 (38.4) 216.7 (39.0) Stage 3+4 sleep, min Baseline 36.8 (29.0) 33.7 (27.0) Treatment 47.2 (34.2) 52.0 (29.5) REM sleep, min Baseline 62.4 (23.0) 61.9 (19.1) Treatment 74.1 (24.3) 65.5* (19.9) 0.3 (0.3) 0.3 (0.3) Stage 3+4/(Stage 1 +WASO), % Baseline Treatment 0.7 (1.2) 0.7 (0.6)

Tiagabine 6 mg 8 mg n=45 n=45 95.3 (33.0) 104.0 (32.4) 68.7 (37.0) 68.8 (40.8) 54.3 (32.4) 45.3 (22.3) 35.7 (25.1) 31.1 (23.0) 338.0 (35.9) 336.0 (32.3) 381.8 (47.9) 384.1 (47.9) 33.5 (12.5) 30.6 (12.9) 30.5 (8.1) 29.1 (9.8) 70.4 (7.5) 70.0 (6.7) 79.5 (10.0) 80.0 (10.0) 32.7 (18.0) 34.5 (19.1) 25.6**** (16.8) 27.2*** (18.2) 198.7 (40.1) 205.5 (37.9) 216.3 (54.5) 223.4 (51.0) 41.4 (32.1) 30.4 (25.5) 73.1*** (47.1) 70.0**** (44.8) 65.1 (23.9) 65.7 (17.4) 66.8* (24.0) 63.6** (24.2) 0.4 (0.3) 0.2 (0.2) 1.0 (1.0) 1.1*** (1.1)

10 mg n=47 105.7 (34.6) 74.8 (49.6) 47.5 (24.3) 39.4 (36.5) 333.7 (35.5) 372.3 (51.7) 29.9 (9.9) 27.8* (8.9) 69.5 (7.4) 77.6 (10.8) 42.4 (24.1) 26.9**** (17.7) 199.2 (32.3) 210.3 (47.6) 25.4 (22.9) 78.3**** (44.3) 66.6 (24.9) 56.7**** (26.3) 0.2 (0.2) 1.2**** (1.2)

Data are presented as mean (SD). * p < .05; **p < .01; ***p < .001; ****p < .0001 change from baseline vs placebo. LPS refers to latency to persistent sleep; REM, rapid eye movement; TST, total sleep time; WASO, wake after sleep onset.

tion and are included in the safety analyses. Table 1 contains demographic characteristics for each treatment group. Two patients receiving tiagabine 10 mg discontinued prior to the completion of the first randomization PSG because of adverse events (discussed below). The efficacy analysis, therefore, was performed on 230 patients with at least 1 postrandomization efficacy assessment. There were no significant differences among groups in any of the demographic characteristics. PSG Variables PSG variable means for each treatment group following tiagabine or placebo are presented in Table 2. Change from baseline in minutes of WASO did not differ significantly between any tiagabine dose and placebo. The mean reduction from baseline in minutes of WASO with tiagabine ranged from 26.3 to 35.2, as compared with 31.7 with placebo. There also were no significant differences between any tiagabine dose and placebo in LPS and TST. Significantly greater increases from baseline in minutes of SWS were found with the 3 higher doses of tiagabine, compared with placebo (p < .01 for all). The mean change from baseline in minutes of SWS for placebo was 10.5; whereas, for the 3 higher tiagabine doses, SWS increased by 31.7, 39.6, and 52.9 minutes, respectively. When analyzed by hour of the night (Figure 2), tiagabine 10 mg was associated with greater time spent in SWS, with all but the first hour being significantly greater than placebo (p < .05). Tiagabine 6 mg and 8 mg had significantly greater increases in Stage 3+4 sleep at 5 of the 8 hours of the night, in comparison with placebo. The ratio Stage 3+4/(Stage 1+WASO) was increased significantly more with tiagabine 8 mg and 10 mg than with placebo (p < .01). All doses of tiagabine significantly reduced minutes of Stage
Journal of Clinical Sleep Medicine, Vol. 2, No. 1, 2006 38

1 sleep relative to placebo (p < .01 for all). Mean decreases from baseline for ascending doses of tiagabine were -5.7, -7.1, -7.3, and -15.5 minutes, respectively; minutes of Stage 1 sleep increased 1.8 minutes in the placebo group. The change from baseline in minutes of REM sleep differed from placebo for all tiagabine doses (p < .05 for each). The mean change for placebo was 11.6 and, for ascending doses of tiagabine, was 3.6, 1.7, -2.1, and -9.9 minutes, respectively. REM latency did not differ from placebo for any tiagabine dose.

Mean (SD) Stage 3+4 Sleep During 8-h PSG (min)

20

Tiagabine 10 mg Tiagabine 8 mg Tiagabine 6 mg Tiagabine 4 mg

15

Placebo

10

0 0 1 2 3 4 5 6 7 8

Hour of night Figure 2Minutes of slow-wave sleep (Stage 3+4) by hour of the night for each treatment group. Symbols highlighted by a shaded box indicate p < .05 for change from baseline vs placebo.

Tiagabine in Adults With Primary Insomnia Table 3Self-Report Measures From the Post-sleep Questionnaire for Each Treatment Group Placebo Tiagabine Variable 4 mg 6 mg 8 mg n=47 n=46 n=45 n=45 Sleep latency, min Baseline 52.2 (32.8) 65.7 (31.4) 62.3 (38.8) 58.1 (33.5) Treatment 45.6 (36.8) 52.7 (29.0) 50.5 (37.9) 45.4 (29.4) Number of awakenings Baseline 5.1 (5.6) 4.3 (2.5) 3.8 (2.0) 4.3 (1.8) Treatment 4.2 (3.3) 3.4 (1.9) 2.9 (2.0) 3.9 (2.1) Wake after sleep onset, min Baseline 97.4 (70.3) 88.3 (56.9) 79.8 (41.0) 89.8 (60.2) Treatment 74.9 (66.3) 75.4 (62.8) 48.7 (34.7) 73.6 (56.2) Total sleep time, min Baseline 325.8 (64.1) 298.0 (66.2) 326.3 (55.0) 319.9 (58.4) Treatment 355.5 (65.6) 331.4 (77.9) 363.3 (61.6) 351.4 (67.1) Sleep depth* Baseline 2.8 (0.5) 2.9 (0.5) 2.7 (0.5) 2.8 (0.5) Treatment 2.6 (0.5) 2.7 (0.6) 2.4 (0.7) 2.6 (0.7) Sleep quality* Baseline 2.9 (0.6) 2.9 (0.5) 2.7 (0.5) 3.0 (0.5) Treatment 2.7 (0.7) 2.6 (0.6) 2.4 (0.7) 2.7 (0.8) Refreshing quality of sleep Baseline 43.7 (16.4) 44.5 (22.7) 50.7 (17.8) 44.9 (19.6) Treatment 57.1 (20.7) 56.9 (21.8) 57.7 (25.1) 53.8 (24.6) *p < .05 change from baseline vs placebo. *Lower score represents patient improvement.

10 mg n=47 58.5 (43.3) 60.6 (48.2) 4.7 (3.0) 4.6 (4.6) 95.4 (56.0) 92.4 (65.7) 308.4 (63.6) 312.2* (71.4) 2.8 (0.6) 2.9* (0.7) 2.9 (0.7) 2.9* (0.7) 48.0 (18.4) 47.8* (27.5)

Self-Report Variables There were no consistent, drug-related improvements in selfreported ratings of sleep, compared with placebo (Table 3). Estimates of TST, sleep depth and quality, and refreshing quality of sleep were found to be significantly less favorable with tiagabine 10 mg compared with placebo (p < .05). Notably, SWS changes were not correlated with either sleep depth (change from baseline ranging from -0.14 to 0.25) or sleep quality (change from baseline ranging from -0.12 to 0.13). Ability to concentrate or think clearly, level of alertness, and sense of physical well-being, as assessed by the Assessment of Daily Function, also did not differ between placebo and tiagabine 4, 6, or 8 mg. All 3 Assessment of Daily Function variables were significantly reduced with the 10mg dose, as compared with placebo (p < .05). Tolerability and Safety Tiagabine 4 mg and 6 mg had a tolerability profile similar to that of placebo, with the exception of the incidence of headache, which was numerically greater with the 4-mg dose (9%) compared with placebo (4%; Table 4). In patients receiving tiagabine 8 mg and 10 mg, the most commonly reported adverse events were dizziness (11% and 18%, respectively, versus placebo, 2%) and nausea (7% and 14%, respectively, versus placebo, 0%). The Cochran-Armitage trend test showed a significant dose-related increase in the rate of anxiety (p = .024), dizziness (p = .002), and nausea (p = .002). No other adverse events were significantly associated with dose. A total of 6 patients receiving tiagabine discontinued because of 1 or more adverse events (10 mg, n = 5; 4 mg, n = 1). Dizziness (n = 4), anxiety (n = 3), and nausea (n = 2) were the most common adverse events leading to discontinuation. There were no clinically meaningful changes in laboratory values and vital signs for any patient in the study. Mean number of correct responses on the Digit Symbol Substitution Test for each group were 4 mg, 59.1; 6 mg, 57.8; 8 mg, 60.7; 10 mg, 55.7; and placebo, 61.6. The 10-mg group, but not other doses, had significantly fewer mean number of correct responses on the Digit Symbol Substitution Test, compared with
Journal of Clinical Sleep Medicine, Vol. 2, No. 1, 2006 39

placebo. Mean visual analog scale values (lower = more sleepiness) for each group were 4 mg, 59.2; 6 mg, 64.2; 8 mg, 57.1; 10 mg, 52.2; placebo, 60.2. The 10-mg group, but not other doses, reported significantly more sleepiness on the visual analog scale in comparison with placebo. DISCUSSION The results of this randomized, double-blind, placebo-controlled study show that, relative to placebo, tiagabine 6, 8, and 10 mg significantly increased SWS, with a corresponding significant decrease in Stage 1 sleep at all tiagabine doses. The traditional hypnotic-efficacy variables of WASO, LPS, and TST were not affected by any tiagabine dose. Tiagabine 4 and 6 mg had a tolerability profile similar to that of placebo. The most commonly reported adverse events for tiagabine 8 and 10 mg were nausea and dizziness, which are commonly observed with tiagabine.17 Tiagabine 10 mg differed from placebo on morning sedation. The findings of the present study are reasonably consistent with another study of primary insomniacs,9 in which tiagabine significantly increased SWS and decreased Stage 1 sleep in a dose-dependent manner compared with placebo. In that investigation, doses of tiagabine up to 8 mg were found to be similar to placebo in tolerability profiles and psychomotor performance, again largely consistent with
Table 4Adverse Events Placebo Variable n=47 Dizziness* 1 (2) Nausea* 0 (0) Somnolence 2 (4) Headache 2 (4) Anxiety* 0 (0)

4 mg n=46 1 (2) 1 (2) 2 (4) 4 (9) 0 (0)

Tiagabine 6 mg 8 mg n=45 n=45 2 (4) 5 (11) 1 (2) 3 (7) 0 (0) 2 (4) 1 (2) 3 (7) 0 (0) 1 (2)

10 mg n=49 9 (18) 7 (14) 4 (8) 3 (6) 3 (6)

Data are presented as number (%) of adverse events (as defined by the Medical Dictionary for Regulatory Activities-preferred term) that occurred in 5% of patients in any treatment condition. *p < .05 vs placebo for dose-related increase in rate of incidence.

JK Walsh, M Perlis, M Rosenthal et al

the current findings. In a small double-blind, placebo-controlled study of healthy elderly volunteers, tiagabine 5 mg significantly increased SWS, as well as significantly improved sleep efficiency compared with placebo.8 Significant improvement in self-report measures of sleep have not been observed in any of these studies. At present, the physiologic and/or clinical correlates of changes in SWS have not been established. However, it is interesting to consider the possibility that drugs that increase SWS, without strongly affecting traditional hypnotic efficacy measures (i.e., PSG and self-reported TST, LPS, etc.), may have a clinical or physiologic benefit. For example, individuals with paradoxical insomnia may benefit from changes in sleep other than increased sleep duration and more-rapid sleep onset. Individuals with chronic pain as a contributor to sleep disturbance may have fewer arousals and awakenings with enhanced SWS. Reduced SWS is a consistent finding in major depressive disorder, an illness for which insomnia is a very common comorbid condition. Clearly, studies in these types of populations would be interesting and of scientific value. On the other hand, the lack of benefit on selfreport measures in primary insomniacs may limit the usefulness of tiagabine, at least in primary insomnia. The lack of correlation between SWS and sleep depth and sleep quality highlights the lack of understanding of the physiologic processes underlying self-reported estimates of sleep. There is a well-documented relation between time spent in SWS and arousability from sleep. The putative importance of SWS for restoration18,19 raises the hypothesis that tiagabine, or other drugs that significantly increase SWS, may positively affect sleep in some uncharacterized way. Further research should determine whether the effects on SWS persist beyond a few nights and if a qualitative difference in sleep can be shown beyond that attributable to increased sleep time seen with BzRA. In conclusion, the results of this large-sample investigation show that, in adults with primary insomnia, tiagabine significantly increased SWS, with a corresponding significant decrease in Stage 1 sleep. However, the traditional hypnotic-efficacy variables of WASO, LPS, and TST were not affected by any dose of tiagabine. Moreover, the Food and Drug Administration has recently issued a statement emphasizing that the efficacy and safety of tiagabine have not yet been established for uses other than as an adjunctive anticonvulsant. Tiagabine was generally well tolerated, with the 4-mg and 6-mg doses having tolerability profiles similar to placebo, and was associated with more adverse events, psychomotor impairment, and negative self-reports about daytime function at 10 mg. Investigation into the clinical correlates of the effect of tiagabine on SWS is warranted at doses below 10 mg in adults. ACKNOWLEDGEMENTS The authors gratefully acknowledge the work of the following investigators and their staff: Daniel Aeschbach, Ph.D., Harvard Medical School, Boston, MA; Richard Bogan, M.D., SleepMed of South Carolina, Columbia, SC; Gerry Burns, M.D., Neurotrials Research, Inc., Metairie, LA; Martin A. Cohn, M.D., Sleep Disorders Center of Southwest Florida, Naples, FL; Bruce Corser, M.D., Community Research, Cincinnati, OH; Karl Doghramji, M.D., Center for Sleep Disorders, Inc., Pottstown, PA; Sean Drummond, Ph.D., San Diego VA Medical Center, San Diego,
Journal of Clinical Sleep Medicine, Vol. 2, No. 1, 2006 40

CA; Helene Emsellem, M.D., Center for Sleep and Wake Disorders, Chevy Chase, MD; Neil T. Feldman, M.D., Clinical Research Group, St Petersburg, FL; James Ferguson, M.D., Radiant Research, Salt Lake City, UT; Andrew O. Jamieson, M.D., Sleep Medicine Associates of Texas, Plano, TX; David W. Mayleben, Ph.D., Community Research, Crestview Hills, KY; Dennis J. Munjack, M.D., Southwest Research, Inc., Beverly Hills, CA; John F. Pinto, M.D., Clinical Research Center of Nevada, Las Vegas, NV; Angela C. Randazzo, Ph.D., St. Johns/St. Lukes Hospitals, St. Louis; Kathleen Rice, Ph.D., Clinilabs Inc., New York, NY; Bart Sangal, M.D., Clinical Neurophysiology Services, Troy, MI; Martin B. Scharf, Ph.D., Tri-State Sleep Disorders Center, Cincinnati, OH; Markus H. Schmidt, M.D., Ohio Sleep Medicine and Neuroscience Institute, Inc., Dublin, OH; Jonathan Schwartz, M.D., Clinical Research Studies, Oklahoma City, OK; David Seiden, M.D., Broward Research Group, Pembrook Pines, FL; Eric Sheldon, M.D., Miami Research Associates, Miami, FL; Edward Stepanski, Ph.D., Rush University Medical Center, Chicago, IL; Kenneth Wright, Ph.D., University of Colorado, Boulder, CO. This research was sponsored and study medication was supplied by Cephalon, Inc., Frazer, PA. REFERENCES
1. 2. 3. 4. 5. 6. 7. Roehrs T, Roth T. Hypnotic prescription patterns in a large managed-care population. Sleep Med 2004;5:463-6. Walsh JK, Schweitzer PK. Ten-year trends in the pharmacological treatment of insomnia. Sleep 1999;22:371-5. Walsh JK. Drugs used to treat insomnia in 2002: regulatory-based rather than evidence-based medicine. Sleep 2004;27:1441-2. Taylor FB. Tiagabine for posttraumatic stress disorder: a case series of 7 women. J Clin Psychiatry 2003;64:1421-5. Rosenthal M. Tiagabine for the treatment of generalized anxiety disorder: A randomized, open-label, clinical trial with paroxetine as a positive control. J Clin Psychiatry 2003;64:1245-9. Borden LA, Murali Dhar TG, Smith KE, et al. Tiagabine, SK&F 89976-A, CI-966, and NNC-711 are selective for the cloned GABA transporter GAT-1. Eur J Pharmacol 1994;269:219-24. Fink-Jensen A, Suzdak PD, Swedberg M.D., et al. The gammaaminobutyric acid (GABA) uptake inhibitor, tiagabine, increases extracellular brain levels of GABA in awake rats. Eur J Pharmacol 1992;220:197-201. Mathias S, Wetter TC, Steiger A, Lancel M. The GABA uptake inhibitor tiagabine promotes slow wave sleep in normal elderly subjects. Neurobiol Aging 2001;22:247-53. Walsh JK, Zammit GK, Schweitzer PK, Ondrasik J, Roth T. Tiagabine enhances slow wave sleep and sleep maintenance in primary insomnia. Sleep Med 2005:[epub ahead of print]. Walsh JK, Randazzo AC, Frankowski S, Shannon K, Schweitzer PK, Roth T. Dose-response effects of tiagabine on the sleep of older adults. Sleep 2005;28:673-6. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington: American Psychiatric Association; 2000. Carskadon MA, Rechtschaffen A. Monitoring and staging human sleep. In: Kryger MR, Roth T, Dement WC, eds. Principles and Practice of Sleep Medicine. 2nd ed. Philadelphia: W.B. Saunders Company; 1994:943-60. Rechtschaffen A, Kales A. A manual of standardized terminology, techniques and scoring system for sleep stages in human subjects. Washington: US Government Printing Office; 1968. Stone BM. Pencil and paper testssensitivity to psychotropic drugs. Br J Clin Pharmacol 1984;18 Suppl 1:15S-20S. Roehrs T, Bonahoom A, Pedrosi B, Rosenthal L, Roth T. Disturbed sleep predicts hypnotic self-administration. Sleep Med 2002;3:61-6.

8. 9. 10. 11. 12.

13. 14. 15.

Tiagabine in Adults With Primary Insomnia 16. Northrop Grumman Corporation. Medical Dictionary for Regulatory Activities; 2004. Available at http://www.meddramsso.com/NewWeb2003/meddra_overview/index.htm; Accessed July 16, 2005. 17. GABITRIL [package insert]. Frazer, PA: Cephalon, Inc.; 2005. 18. Ambrosini MV, Giuditta A. Learning and sleep: the sequential hypothesis. Sleep Med Rev 2001;5:477-90. 19. Edinger JD, Glenn DM, Bastian LA, Marsh GR. Slow-wave sleep and waking cognitive performance II: Findings among middleaged adults with and without insomnia complaints. Physiol Behav 2000;70:127-34.

Journal of Clinical Sleep Medicine, Vol. 2, No. 1, 2006

41