Institutional members access full text with Ovid®

Share this article on:

00004714-201002000-0000900004714_2010_30_48_maguire_developmental_1article< 125_0_19_5 >Journal of Clinical Psychopharmacology© 2010 Lippincott Williams & Wilkins, Inc.Volume 30(1)February 2010pp 48-56Exploratory Randomized Clinical Study of Pagoclone in Persistent Developmental Stuttering: The EXamining Pagoclone for peRsistent dEvelopmental Stuttering Study[Original Contributions]Maguire, Gerald MD*; Franklin, David PsyD, MHA*; Vatakis, Nick G. MD†; Morgenshtern, Elena†; Denko, Timothey MD‡; Yaruss, J. Scott PhD, CCC-SLP, ASHA§; Spotts, Crystal MD‡; Davis, Larry MD∥; Davis, Aaron ND∥; Fox, Peter MD¶; Soni, Poonam MD#; Blomgren, Michael PhD, CCC-SLP**; Silverman, Andrew PhD††; Riley, Glyndon PhD*From the *University of California, Irvine School of Medicine, Orange, CA; †Social Psychiatry Research Institute, New York, NY; ‡University of Pittsburgh Medical Center; §University of Pittsburgh, Pittsburgh, PA; ∥Davis Clinic PC, Indianapolis, IN; ¶University of Texas, Health Science Center, San Antonio, TX; #University of Utah; **Department of Communication Sciences and Disorders, University of Utah, Salt Lake City, UT; ††Suite 300, Austin, TX.Received September 26, 2008; accepted after revision November 13, 2009.Reprints: Gerald Maguire, MD, University of California, Irvine School of Medicine, 101 The City Dr, Bldg 3, Rte No. 88, Orange, CA 92868 (e-mail: study was funded by Indevus Pharmaceuticals. Each author was a paid investigator in this study.AbstractIntroduction: Stuttering is a speech disorder in which the flow of speech is disrupted by repetitions, prolongation, and blocks of sounds, syllables, or words. No pharmacological treatments are approved for use in stuttering, and the most common form of treatment is speech therapy. This study was designed to assess the safety, tolerability, and effectiveness of pagoclone during 8 weeks of double-blind treatment followed by a 1-year open-label extension in patients who stutter.Methods: An 8-week, multicenter, parallel-group, 2-arm, randomized (ratio 2:1 pagoclone-placebo), double-blind study with a 1-year open-label extension conducted at 16 US centers, including men and women aged 18 to 65 years who developed stuttering before 8 years of age. Twice-daily dosing with pagoclone (n = 88 patients) or matching placebo (n = 44 patients), with primary and secondary efficacy variables defined a priori, including Stuttering Severity Instrument Version 3 outcomes, clinician global impressions of improvement, and the change in the percentage of syllables stuttered.Results: Pagoclone produced an average 19.4% reduction in percentage of syllables stuttered compared with 5.1% reduction for placebo. During open-label treatment, a 40% reduction in the percent syllables stuttered was observed after 1 year of treatment with pagoclone. The most commonly reported adverse event during double-blind treatment was headache (12.5% pagoclone patients, 6.8% placebo patients).Discussion: Pagoclone was effective in reducing symptoms of stuttering and was well tolerated. In light of its favorable tolerability profile, as well as consistency of effects across multiple efficacy variables, pagoclone may have potential as a pharmacological treatment of stuttering.Limitations: The main limitation of this study was the adequacy of the number of subjects who participated because this study was conducted as a pilot investigation. Furthermore, as this condition waxes and wanes, the assessment of stuttering within the clinic setting may not be an adequate reflection of the stuttering of the patients within the community.Stuttering is a chronic speech disorder that begins in childhood and is characterized by repetitions or prolongations of sounds or syllables resulting from a disruption in the flow and timing of speech.1 Stuttering, beginning in childhood, describes the developmental speech disorder defined by the International Classification of Diseases, 10th Revision, as consisting of frequent repetitions or prolongations of sounds or syllables or words, or frequent hesitations or pauses that disrupt the flow of speech. The Diagnostic and Statistical Manual of Mental Disorders (Text Revision 2000, Diagnostic and Statistical Manual of Mental Disorders-IV-TR, 307.0, pp 67-69) describes stuttering as an Axis I illness that is usually first diagnosed in infancy, childhood, or adolescence. The disturbances in normal fluency and time patterning of speech (that is inappropriate for the age of the individual) are characterized by frequent occurrences of one or more of the following: (1) sound and syllable repetitions; (2) sound prolongations; (3) interjections; (4) broken words (eg, pauses within a word); (5) audible or silent blocking (filled or unfilled pauses in speech); (6) circumlocution (word substitutions to avoid problematic words); (7) words produced with an excess of physical tension; and (8) monosyllabic whole-word repetitions.These observable speech behaviors are considered the "core" behaviors of stuttering and are generally thought to be involuntary. Teesson et al2 have categorized these core behaviors into 2 primary types: (1) repeated movements (whole syllable repetitions, incomplete syllable repetitions, or multisyllable unit repetitions) and (2) fixed articulatory postures (with or without audible airflow). Thus, stuttering also involves learned or "secondary" stuttering characteristics that include escape behaviors (such as facial grimaces, rapid eye blinking, jerking movements of head or body) and avoidance behaviors (such as avoiding specific speaking situations or problematic words).3,4 Affective components of stuttering such as fear, insecurity, shame, embarrassment, and anxiety may also play mediating roles for some people who stutter.5 In particular, anxiety during speaking situations may worsen the core stuttering symptoms.An estimated 3 million Americans (1% of the population) stutter, and up to 80% are men. Stuttering of a developmental (and not acquired) nature is most often first observed between the ages of 2 and 7 years, and onset is often insidious, occurring gradually for several months, although onset can also occur suddenly. The prevalence of stuttering in preschool children is between 5% and 15%; however, prevalence decreases with increasing age to approximately 1% by high school.6-10 Spontaneous recovery occurs in approximately 75% of those affected as children by or before age 16 years.11-13The primary etiology of stuttering is not well understood, but it seems to have a genetic predisposition, with pairwise concordance in identical twins (63%) more than same-sex or fraternal twins (19%).14,15 Studies on linkage analysis in highly inbred families suggest that a locus on chromosome 12q may possess a gene that contributes significantly to the complex inheritance.16 Recent reports indicate a differing linkage between males (chromosome 7) and females (chromosome 21).17Although psychotherapy has been used for adjustment problems secondary to the speech impediment, stuttering is likely to be of neurological origin with psychological consequences. Stuttering is thus considered to be a disorder of the central nervous system.18 Notable differences in brain function have been identified between individuals who stutter and those who do not; for example, healthy subjects and those who stutter may have differences in functional asymmetries of the cerebral hemispheres on electrophysiological assessments as well as functional magnetic resonance imaging and positron emission tomography.19-22 A direct link between stuttering and excessive release of dopamine may exist in brain regions (eg, ventral limbic cortical and subcortical regions).23,24 Various approaches to stuttering treatment exist and include nonpharmacological and pharmacological methods.25,26 Behavioral modification including desensitization, stuttering modification therapy, and fluency shaping are directed at helping patients gain control of their speech.27 Bothe et al25,26 indicated that the effect of these therapies has been reported primarily in small studies of poor research quality. However, despite these problems, many of these studies demonstrated reduced stuttering for many participants.Electronic antistuttering devices provide delayed or otherwise altered auditory feedback that may assist some individuals to speak more fluently. None of the available devices has received US Food and Drug Administration approval, and the maintenance of effect with these methods is not established.There are no US Food and Drug Administration-approved pharmacological treatments for stuttering. Experimental treatments (using medications approved for other indications) have included therapies such as stimulants (eg, methamphetamine), sedatives or tranquilizers (eg, reserpine), antipsychotics and atypical antipsychotics, and many others.25,26 The dopamine-blocking medications, such as haloperidol, risperidone, and most recently, olanzapine, seem to have a detectable beneficial effect in stuttering, although they are associated with considerable adverse effects28-30 that may inhibit use for long-term treatment of chronic stuttering. Olanzapine has been noted to induce a down-regulation of postsynaptic γ-aminobutyric acid A receptors (GABAA), suggesting that a direct-acting GABAA agonist or partial agonist (such as pagoclone) may have a beneficial effect in stuttering.31 As Bothe et al25,26 reported, the scope of clinical research for pharmacological therapy is limited both in the number of studies that have been conducted and the number of subjects studied.GABA, a widely distributed neurotransmitter and the primary mediator of neuronal inhibition, seems to modulate the dopaminergic system. Dopaminergic neurons and GABAergic synapses seem to be anatomically connected within areas of the brain such as the striatum and nucleus accumbens,32 and drug effects of potent antagonists of GABAA receptors and of adenosine have been shown to induce stuttering, likely by stimulating areas of the motor cortex connecting to Broca's area and to a greater degree, the basal ganglia, through inhibition of GABAergic neurotransmission. The subsequent drug-induced hyperexcitability of the motor cortex and a hyperdopaminergic activity may induce stuttering and is comparable to the brain dysfunction evident in developmental stuttering.10Pagoclone, a pure enantiomer, a novel nonbenzodiazepine GABAA modulator, and a member of the cyclopyrrolone class of compounds, is a partial agonist at GABAA receptors. Approximately 1000 subjects have received treatment with pagoclone in previous studies of generalized anxiety or panic disorder, and a small number of subjects in these clinical studies were reported to stutter during screening assessments. During treatment with pagoclone 0.30 to 0.60 mg/d, it was observed anecdotally that study subjects receiving pagoclone experienced a reduction in their stuttering symptoms.The EXamining Pagoclone for peRsistent dEvelopmental Stuttering Study (EXPRESS) was an exploratory trial, designed to test the hypothesis that pagoclone may be an effective treatment of individuals who stutter.MATERIALS AND METHODSTrial OrganizationThe EXPRESS was a randomized clinical trial (registered at , study identifier NCT00216255) conducted at 16 investigational centers in the United States between September 2005 and May 2006. Institutional review boards approved the protocol, and written informed consent was obtained from all patients. Data collection and analysis were coordinated by Indevus Pharmaceuticals, Inc (Lexington, Mass). These 16 different sites included both academic centers as well as private practices within the community. The patients in this study were recruited from multiple sources including referrals from speech pathologists, neurologists, and primary care physicians. Recruitment was also aided by internet awareness, radio advertisements, a project-specific stuttering Web site, and the posting of the study on criteria were designed to select patients with mild to severe persistent developmental stuttering and included male and female patients, aged 18 to 65 years, with current mild to severe stuttering (scores of 18-36 as assessed by the Stuttering Severity Instrument Version 3 [SSI-3]), and onset by 8 years old. Patients were excluded if they met diagnostic criteria for any DSM-IV-TR psychiatric disorder other than stuttering, as documented using the MINI International Neuropsychiatric Interview, or had symptoms of generalized anxiety, as documented by a screening Hamilton Anxiety Score greater than 14. Psychotropic or other medication(s) within 4 weeks before the screening were exclusionary. Previous speech and other therapies were not to have been performed within 5 months before screening and were not permitted during the study. After 8 weeks of participation in the double-blind study, subjects had the option of continuing into the open-label phase of this study. Patients must have completed the double-blind portion of the study for inclusion in the open-label phase.Study Design and Sample SizeThis was a multicenter, parallel-group, randomized, double-blind, placebo-controlled study of pagoclone 0.30 mg given twice daily. Five visits were conducted for the double-blind phase of this study: pretreatment (screening [week −4] and baseline) and on-treatment (weeks 2, 4, and 8). The screening visit included assessment of stuttering and medical history, physical examination, vital signs, clinical laboratory (including hematology, chemistry, and urinalysis panels), a neuropsychiatric and anxiety assessment, electrocardiography, a urine drug screen, and prior and current medications. Stuttering was assessed via the SSI-3,33 the Subjective Screening of Stuttering (SSS),34 the Stuttering Severity Scale (SEV), and the Speech Naturalness Scale (SNS).35 Patients for this study were evaluated at approximately the same time of day and within the same office space for the assessment of their speech.After a 4-week screening phase, patients returned for their baseline visit at which time their stuttering was reassessed (via the SSI-3, SSS, SEV, and SNS), and the Liebowitz Social Anxiety Scale (LSAS) was collected. Eligible patients were stratified by severity of SSI-3 score and randomized in a 2:1 ratio of pagoclone 0.15 mg BID, increasing after 2 weeks to pagoclone 0.30 mg BID for an additional 6 weeks, or to matching placebo, respectively. Efficacy and safety measures were collected during these on-treatment visits.This was an exploratory study in which we attempted to assess whether there was evidence of efficacy in the target population. The sample size of approximately 80 pagoclone and 40 placebo patients, although not sufficient to draw definitive conclusions, was based on expected changes in the key end points, which were assessed using 1-sided tests of significance. Consistent with the exploratory objectives in this study, no adjustments for overall experiment-wise α were planned or performed.After the completion of the double-blind phase, 119 subjects entered the open-label extension. The open-label phase was designed to allow for collection of additional longer-term efficacy and safety outcomes associated with treatment in the target population. During the open-label phase, each subject was seen at week 2 and months 3, 6, 9, and 12. Each subject was treated with pagoclone, with dosing starting at 0.30 mg/d and increasing to 0.60 mg/d. The SSI-3 was performed at months 3 and 12. Furthermore, the SSS, SEV, and Clinical Global Impression (CGI) assessments were performed at months 3, 6, 9, and 12, and the SNS and LSAS assessments were performed at months 3 and 12.Primary and Secondary Outcome ScalesAs this was an exploratory study, we predefined 3 primary variables and several secondary variables. The 3 primary efficacy outcomes were the SSI-3 Frequency and Duration subscore (centrally rated), the SSS Severity subscore (patient rated), and the SEV (site clinician rated). To allow for direct comparison of pagoclone treatment with placebo treatment, the primary analysis was performed at the completion of the double-blind phase.To assess the effect of pharmacological intervention on stuttering, an objective tool for the accurate and precise quantitative measurement of the severity of stuttering is required. To date, no surrogate end points to measure stuttering have been developed, although a multidimensional approach of measuring outcomes in trials of patients who stutter has been suggested.36 However, consistent with the DSM-IV diagnostic criteria, the goal of therapy is to enhance fluency via a reduction in the occurrence of stuttering events and also to reduce social avoidance by patients who stutter.1,10Using the SSI-3, stuttering events were defined according to the DSM-IV diagnostic criteria and were counted in this study as part of the assessment of the percentage of syllables stuttered. Stuttering events included 7 of the 8 DSM-IV diagnostic criteria: sound and syllable repetitions, sound prolongations, interjections, broken words, audible or silent blocking, words produced with an excess of physical tension, and monosyllabic whole-word repetitions. The DSM-IV criterion of circumlocutions was not assessed because this type of dysfluency cannot accurately or precisely be counted without knowledge of the patient's conscious (or unconscious) attempt to replace a more "difficult" word with an "easier" word.A key secondary outcome was the objective measurement of the percentage of syllables stuttered and was used as the core objective assessment of the DSM-IV criteria. This variable was defined as the number of stuttering events divided by the number of total syllables spoken and expressed as a percentage.37 From a treating clinician's perspective, reducing the percentage of syllables stuttered is the primary measurable objective of intervention for stuttering. From the patient's perspective, a reduction in the percentage of stuttering events is an important objective, and it is likely accompanied by an improvement in social communications and social avoidance. To obtain this variable, video recordings of a patient engaged in a prompted monologue and reading tasks were collected using the SSI-3. The SSI-3 also provides an assessment of stuttering symptoms via a procedural and scoring manual that uses stanines to ensure adequate psychometric properties. It includes reading samples and conversational tasks that were video-recorded for later scoring by trained central raters. To provide objective outcomes from the ratings, these videos were then batched, assigned random accession numbers, pseudorandomized to visit order, and only then were they analyzed in a blinded fashion by 2 independent trained central raters. During the scoring of the SSI-3, the raters computed the total number of syllables spoken and the total number of stuttering events. In addition to the fundamental stuttering measure of percentage of syllables stuttered, other characteristics of the patient's speech and behavior (physical concomitants, duration of stuttering blocks) were collected according to the SSI-3 procedures. A reconciliation algorithm was used to ensure agreement of the SSI-3 total scores by the raters to within 20% of one another. If agreement was not within 20%, the raters reassessed the videos, and in some cases, a third independent rater was included to mediate the differences. The average of the reconciled scores from the 2 raters was then used in the analysis. Assessment of other outcomes were performed in this exploratory study and included subscores of the SSI-3 and SSS; 4 questions on a CGI-Improvement of Stuttering (CGI-I) scale (including frequency and severity questions); a subset score of the LSAS comprising questions specific to the anxiety and avoidance brought on by speaking situations; and the SNS, a 9-point rating scale used to assess how natural the speech sounded to an observer and, in this study, was used to identify any potential adverse effects on naturalness from the study treatment.35 Although these outcomes were secondary to the key objectives previously outlined, evidence of the potential effect of pagoclone was expected to be observed for multiple measures. Furthermore, although rater reliability was not performed for these secondary outcomes, the trial was randomized and blinded, and thus patients in both treatment arms were assessed similarly.38 Each of the 4 CGI-I questions were measured on a 7-point scale, from 1 = "Very Much Improved" to 7 = "Very Much Worsened," with 4 = "No Change." For each CGI-I question, responses that indicated improvement (scores of 1, 2, or 3) were categorized as "Improved," whereas the remaining responses were categorized as "Not Improved," and this dichotomous response outcome was assessed for analysis of each question.Safety assessments and patient characteristics were summarized using all patients who were randomized and dispensed double-blind study medication. Efficacy assessments were performed using the Intent-to-Treat (ITT) sample (ie, all patients who were randomized, dispensed study medication, took at least 1 dose of study medication, and had at least 1 postbaseline evaluation). The ITT sample was determined on a by-variable basis, allowing for maximum use of collected data. All descriptive and inferential statistical analyses were performed using SAS statistical software (Version 9.1; SAS Institute Inc, Cary, NC).This study was designed with exploratory objectives, and thus there was no adjustment to the overall experiment-wise α level. Furthermore, P values for efficacy outcomes were predefined as one sided for efficacy outcomes (where values of P ≤ 0.05 were considered significant). A complete statistical analysis plan was written for this study, and documentation of the a priori methodologies applied was completed before unblinding the study treatment codes. Assumptions (eg, normality and homogeneity of variance) for parametric models were tested, and results for the appropriate statistical method (based on the assumption testing) are presented. For continuous variables that satisfied the parametric assumptions, analysis of variance (ANOVA) with effects for treatment and pooled center were used. For nonparametric continuous variables, the median test or a ranked ANOVA was applied. If there was a significant difference at baseline between the treatment groups, the baseline values were added to the model as a covariate, and the appropriate P value from the analysis based on assumption testing was presented (ie, either ANOVA or ranked ANOVA). For categorical variables, a Cochran-Mantel Haenszel (CMH) test was used, controlling for pooled center. For categorical variables with inherent ordering to the categories, a CMH row means score test was used. Efficacy analyses were performed using the last-observation-carried-forward (LOCF) data set. If data were missing at a given visit, the most recent prior data point collected was carried forward. However, pretreatment data were not carried forward into the treatment data. The primary end point was captured at the week 8 visit, however, week 4 and pooled time points were also assessed. Because of the inherent variability of stuttering severity, the average of the 2 pretreatment outcomes (the screening and baseline data) was compared with the average of the 2 on-treatment outcomes (the week 4 and week 8 data) for most variables.Open-label data for months 3 and 12 were also collected for this study. These included data from the SSI-3 and the CGI-I. Because all patients were treated with pagoclone during the open-label period, analysis of open-label outcomes was performed using descriptive methods only. Safety parameters were assessed in terms of change from pretreatment to on-treatment and the incidence of potentially clinically significant values using predefined criteria. The incidence of treatment-emergent adverse events (TEAEs) was tabulated by MedDRA system organ class and term and is presented for each preferred term occurring in at least 3 patients in the pagoclone group.To provide consistent descriptive values for outcomes, the adjusted means from the ANOVA models (ie, the LSMeans) are presented where appropriate. To demonstrate the inferential outcomes, both the median test and rank ANOVA P values are presented, where appropriate.RESULTSBaseline CharacteristicsOf the 232 patients screened by the study centers, 132 patients met eligibility requirements and were randomly assigned (in a 2:1 ratio) to treatment with either pagoclone (n = 88) or placebo (n = 44). The treatment groups were comparable with respect to the individual DSM-IV-TR diagnostic criteria for 307.0 Stuttering. All patients reported that their disturbance in fluency interfered with academic or occupational achievement. Patients were generally healthy with few comorbid conditions reported. Furthermore, the treatment groups were similar at screening and demonstrated low Hamilton Anxiety Rating Scale scores, indicative of a lack of significant symptoms of generalized anxiety. However, as expected for patients who stutter, significant symptoms of social anxiety were documented by the baseline LSAS. Baseline characteristics between the placebo and pagoclone groups did not differ significantly in sex, age, race, weight, or severity of the stuttering symptoms.Treatment After RandomizationEarly discontinuations were higher in the placebo group (6 patients, 14%) compared with the pagoclone group (6 patients, 7%); discontinuations caused by adverse events were comparable between the treatment groups. Ninety percent (119/132) of the randomized patients were eligible to continue into the open-label extension phase; 90 of these patients had at least 1 efficacy assessment during the 1-year open-label extension, and 56 of these patients completed the full 1-year treatment period.Key End PointsResults for the primary and key secondary outcomes are presented in Table 1. The pagoclone group demonstrated significance for most key variables at week 4, week 8, or at the average of weeks 4 and 8 time points. Placebo-corrected effects at week 8 were slightly less than those at week 4, perhaps reflecting the variability of stuttering; however, we cannot discount that it is possible that the drug's effect may have declined from week 4 to week 8. However, there seemed to be a consistency of the superiority of the effect of pagoclone on these outcomes when compared with placebo. Furthermore, this effect was observed as early as the week 2 and week 4 time points. We observed that pagoclone-treated patients demonstrated significantly greater reductions from pretreatment to on-treatment in the percent of syllables stuttered at week 4. Furthermore, although not significantly different from placebo at week 8, the pagoclone group clearly experienced a greater reduction in the percent of syllables stuttered compared with placebo at that time point. Clinically, patients in this study averaged 8.6% (placebo group) and 8.2% (pagoclone group) of syllables stuttered at pretreatment (a stutter every 11.6 syllables for the placebo group and every 12.2 syllables for the pagoclone group). The average reduction in the percentage of stuttering events in the pagoclone group was approximately 20% from pretreatment to on-treatment (the average of weeks 4 and 8), whereas the placebo group demonstrated a reduction of only 5% from pretreatment to on-treatment, reflecting an approximately 10% to 15% difference in effect between the active and the placebo-treated groups in this outcome (Fig. 1). The improvement in stuttering was observed throughout the open-label portion of this study. For those patients who completed the open-label phase, the percent syllables stuttered were reduced from baseline by 37.19% at month 3 and by 40.03% at month 12 of the open-label portion of this study (Fig. 2).TABLE 1. Primary End Points: Baseline and Mean Change From Pretreatment to On-Treatment (ITT) Double-Blind Treatment PeriodFIGURE 1. Average (±SE) percent change from pretreatment to on-treatment in percentage of syllables stuttered-ITT (LOCF) double-blind treatment period.FIGURE 2. Average (±SE) percent change from pretreatment to on-treatment in percentage of syllables stuttered-ITT (LOCF) double-blind and open-label treatment periods (open-label data include only patients who completed the 12-month open-label study visit).Other End PointsCompared with placebo, the pagoclone-treated patients demonstrated a reduction (improvement) in other end points, including SSI-3 measures, the SSS locus of control subscore, the SNS, the LSAS Total, and the LSAS Avoidance Subscore (Table 2). Note that a negative change (decrease from baseline) indicates improvement in the LSAS. Improvements on these scales were also seen during the open-label period. For instance, on the LSAS-Avoidance Scale, a decrease of 7.59 points at month 3 and a decrease of 7.49 points at the month 12 visit were observed. Compared with placebo, a higher proportion of pagoclone-treated patients were judged to have improved from pretreatment to on-treatment during the study for the clinician's assessment of treatment benefit using the CGI-I assessments of frequency, severity, and overall stuttering condition (Table 3). Improvements on these scales were also seen during the open-label period. For instance, the CGI-Frequency Scale demonstrated that 96.4% of the patients who completed the 12-month open-label phase improved, whereas 94.5% of the patients improved on the CGI-Severity Scale.TABLE 2. Other Secondary End Points: Baseline and Mean (±SE) Change From Pretreatment to On-Treatment (ITT) Double-Blind Treatment PeriodTABLE 3. CGI: No. (%) Patients Improved From Pretreatment to On-Treatment (ITT) Double-Blind Treatment PeriodSafetyDuring the double-blind phase, at least 1 TEAE was reported for 18 (40.9%) and 51 (58.0%) patients in the placebo and pagoclone groups, respectively. The TEAEs were most commonly reported for the nervous system, with the most frequent event being headache, which was reported in 3 (6.8%) and 11 (12.5%) patients in the placebo and pagoclone groups, respectively. Fatigue was reported in 7 (8%) patients in the pagoclone group, but none in the placebo group. Most other events were reported with a similar incidence rate between the treatment groups. There were no clinically important changes in any laboratory, electrocardiography, or vital sign assessments attributed to either treatment group.There were 2 serious adverse events reported in the pagoclone group, in which the same patient experienced both a tooth abscess and cellulitis, and both were judged by the investigator as definitely unrelated to the study treatment. There were 3 patients who discontinued the study prematurely because of adverse events: a placebo patient discontinued because of headache, lethargy, and somnolence (judged as possibly related to study treatment by the investigator), a pagoclone patient discontinued because of asthenia (judged as possibly related to study treatment), and a pagoclone patient discontinued because of palpitations (relationship to study treatment was considered remote). The AEs reported during the open-label phase were similar to those reported during the double-blind phase, with somnolence and headache the most frequently reported events.DISCUSSIONThe current study is believed to be the largest well-controlled clinical study yet reported in individuals who stutter and the first prospective study of pagoclone as a treatment of those who stutter. Our objectives were to determine the safety and effectiveness of pagoclone, as compared with placebo, as a treatment in patients with persistent developmental stuttering, to identify objective outcome measures sufficiently sensitive for use in future studies and to study the long-term effectiveness and safety of pagoclone as a treatment of stuttering.We examined the incidence of core stuttering events (frequency and duration of stuttering), affective and behavioral aspects (self-perception), and overall improvement (the clinician's global impression of improvement). Importantly, we documented the effects of pagoclone on the occurrence of stuttering events as defined by the DSM-IV diagnostic criteria, including sound and syllable repetitions, sound prolongations, interjections, broken words, audible or silent blocking, words produced with an excess of physical tension, and monosyllabic whole-word repetitions. We found that the percentage of syllables stuttered was the most objective assessment of the stuttering condition. Furthermore, given the high variability of the stuttering condition from day to day (where the occurrence of stuttering events can wax and wane considerably), the use of an average outcome from 2 pretreatment and 2 on-treatment time points likely provides an accurate measure of the true effect of pharmacological treatment.There was internal consistency of the efficacy analysis, with most treatment comparisons numerically favoring the pagoclone treatment group at each time point, and even with the limited sample size of the study, many assessments (including the 3 primary variables) attained statistical significance at one or more time points. Of considerable importance, pagoclone-treated patients demonstrated statistically significantly greater reductions from pretreatment to on-treatment in the percent of syllables stuttered, in addition to change in their perceptions of stuttering and the apparent impact of stuttering on their communication. During the open-label treatment period, this improvement in fluency was maintained up to 1 year and may have improved even further in some patients. In addition to the improvement observed in fluency and clinician assessments of treatment effect (CGI outcomes), the subjects' social anxiety also improved throughout the 1-year open-label extension. These outcomes demonstrate both the relatively rapid improvement in symptoms (within 2 to 4 weeks of initiation of therapy) as well as the persistence of the effect up to 1 year of treatment.Speech therapies are known to potentially alter the naturalness of speech because a common method suggested in therapy involves slowing down speech rate and drawing out particularly difficult syllables.39 Accordingly, measurement of the naturalness of speech is important in evaluating treatments for stuttering to be certain that a treatment that objectively reduces stuttered syllables does not impair naturalness to the point that the gains in fluency are offset by the development of a highly unnatural sounding speech.40 The SNS evaluations in this study showed that patients taking pagoclone did not demonstrate abnormally slowed or deliberate speech quality; in fact, pagoclone-treated patients actually improved in their speech naturalness.Treatment with pagoclone was well tolerated, and given that there is no currently approved alternative pharmacological treatment of stuttering, these data indicate that a larger trial with longer duration of pagoclone as a therapy for patients who stutter is warranted. Furthermore, the EXPRESS demonstrated that measurement of the percentage of syllables stuttered can be performed objectively (using video-recorded speech samples and multiple trained central raters), and that the effects of pharmacological treatment can be detected as early as 2 to 4 weeks after the initiation of therapy. The EXPRESS was a pilot trial to assess both the potential effects of pagoclone in patients who stutter and to assess the use of various end points for measuring meaningful changes in the stuttering condition over time.LimitationsDuring the design of this study, we were limited in our understanding of any potential placebo effect that might be observed in an 8-week double-blind pharmacological intervention in stuttering patients. As reported by Bothe et al,25,26 previous studies of pharmacological agents for the treatment of stuttering have included very small samples sizes generally insufficient for definitive conclusions to be drawn regarding effect sizes for placebo and active drug therapy. Thus, the sample size and duration of treatment were based on limited information, and this study was by necessity only exploratory. The a priori inferential testing plan (including the 1-sided testing) was clearly indicated as part of the study design, with the sample size based on this assumption. However, a new (and much larger) clinical study of pagoclone in patients who stutter is currently underway. That new study includes more than 300 patients and is highly powered using a more conservative 2-sided testing plan and an α = 0.05. Furthermore, the study includes a much longer treatment period (24 weeks of double-blind therapy) and 2 dose levels of pagoclone (including a higher dose arm than that used in the original study).The relatively high variability of the outcomes (as the natural course of stuttering waxes and wanes) may have limited the study to detecting significant differences in some efficacy outcomes, and the precision of point estimates for changes from pretreatment may not be as robust as would likely be observed with a more substantial sample size. The SSI-3, particularly the component related to the frequency of stuttered syllables, provided a useful tool for evaluating the effect of pagoclone upon the essential feature of the disorder. However, video recordings of an admittedly brief sample of in-clinic speech have limitations, and thus the more comprehensive evaluation of the severity of stuttering as encompassed in the CGI-I provided additional clinical confirmation of the efficacy of pagoclone. The CGI-I allows the treating physician to make an overall assessment of improvement in the frequency, severity, and overall stuttering condition based on a full clinical office visit (often 1-2 hours) and not just on a brief recording of the patient's speech (as is done for the SSI-3 and the SEV).Because stuttering is a highly variable condition even within a patient, assessment of the severity of stuttering using a tool with the ability to capture and average several patient recordings of speech might allow investigators to form a more comprehensive assessment of each individual's stuttering severity and, thus, might lead to more robust and precise estimates of treatment effect for future clinical studies.40 Furthermore, the speech samples collected in this study were composed of approximately 300 to 600 syllables, reflecting approximately 5 minutes of speech. There is considerable debate as to how large an adequate speech must be. Sawyer and Yairi41 have suggested, for instance, that regardless of the tool used, the key aspect in the measurement of stuttering is the collection of a speech sample that adequately represents the individual's "true" verbal output. A speech sample between 300 and 900 syllables in duration may be appropriate to characterize stuttering for most patients, however, multiple samples for each patient may add to the reliability of outcomes.CONCLUSIONSConclusions from the EXPRESS, although based on an admittedly limited sample size, have led to the design and initiation of a much larger clinical trial in the stuttering population. That large double-blind study (registered at the Web site as NCT00830154) is currently underway and was designed to both address some of the limitations of the completed exploratory study and to more rigorously test, with a larger patient sample and longer treatment interval, the efficacy of pagoclone as a therapy for patients who stutter.ACKNOWLEDGMENTSThe authors thank LuAnn Sabounjian, James Shipley, Ute Schwiderski, Mark Harnett, and Kathleen Harnett for their efforts supporting this clinical study. The authors also thank Roger Ingham for his insights into the treatment of stuttering.AUTHOR DISCLOSURE INFORMATIONDr Gerald Maguire has been a consultant for Endo Pharmaceuticals, Teva Neuroscience, Indevus Pharmaceuticals, Schering-Plough, Merck, and Eli Lilly and Company; he is on the speakers' bureau at Schering-Plough, Merck, Eli Lilly and Company and Pfizer; and has had research grants with Bristol-Myers Squibb, Otsuka, Eli Lilly and Company, Endo Pharmaceuticals, and Indevus Pharmaceuticals. Dr David Franklin has research grants with Eli Lilly and Company, Bristol-Myers Squibb, Endo Pharmaceuticals, Indevus Pharmaceuticals, and Otsuka.REFEERENCES1. Bloodstein O. A Handbook on Stuttering. 5th ed. San Diego, CA: Singular Publishing Group; 1995. [Context Link]2. Teesson K, Packman A, Onslow M. The Lidcombe behavioral data language of stuttering. J Speech Lang Hear Res. 2003;46:1009-1015. [CrossRef] [Medline Link] [Context Link]3. Conture EG. Stuttering: Its Nature, Diagnosis, and Treatment. Boston, MA: Allyn and Bacon; 2004. [Context Link]4. Guitar B. Stuttering: An Integrated Approach to Its Nature and Treatment. 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2006. [Context Link]5. Bennett EM. Working With People Who Stutter: A Lifespan Approach. Upper Saddle River, NJ: Pearson Merrill Prentice Hall; 2006. [Context Link]6. Andrews G, Craig A, Feyer A, et al. Stuttering: a review of research findings and theories circa 1982. J Speech Hear Disord. 1983;48:226-263. [Medline Link] [Context Link]7. World Health Organization. The International Classification of Functioning, Disability, & Health. Geneva, Switzerland: World Health Organization; 2001. [Context Link]8. Yaruss JS, Quesal RW. Stuttering and the International Classification of Functioning, Disability, and Health (ICF): an update. J Commun Disord. 2004;37:35-52. [CrossRef] [Medline Link] [Context Link]9. Ingham JC, Riley G. Guidelines for documentation of treatment efficacy for young children who stutter. J Speech Lang Hear Res. 1998;41:753-770. [Medline Link] [Context Link]10. Maguire G, Yu B, Franklin D, et al. Alleviating stuttering with pharmacological interventions. Exp Opin Pharmacother. 2004;5:1565-1571. [Context Link]11. Fritzell B. The prognosis of stuttering school children: a 10-year longitudinal study. In: Proceedings of the XVI Congress of the International Society of Logopedics and Phoniatrics. Basel, Switzerland: Karger; 1976:186-187. [Context Link]12. Howell P. Signs of developmental stuttering up to age eight and at 12 plus. Clin Psychol Rev. 2007;3:287-306. [Context Link]13. Yairi E, Ambrose N. Early childhood stuttering I: persistence and recovery rates. J Speech Lang Hear Res. 1999;42:1098-1112. [Context Link]14. Felsenfeld S, Kirk KM, Zhu G, et al. A study of the genetic and environmental etiology of stuttering in a selected twin sample. Behav Genet. 2000;30:359-366. [CrossRef] [Full Text] [Medline Link] [Context Link]15. Howie PM. Concordance for stuttering in monozygotic and dyzygotic twin pairs. J Speech Hear Res. 1981;24:317-321. [Medline Link] [Context Link]16. Riaz N, Steinberg S, Ahmad J, et al. Genomewide significant linkage to stuttering on chromosome 12. Am Hum Genet. 2005;76:647-651. [Context Link]17. Suresh R, Ambros N, Roe C, et al. New complexities in the genetics of stuttering: significant sex-specific linkage signals. Am J Hum Genet. 2006;78:554-563. [CrossRef] [Medline Link] [Context Link]18. Maguire G, Riley G, Yu B. A neurological basis of stuttering? Lancet Neurol. 2002;1:407. [CrossRef] [Medline Link] [Context Link]19. Blomgren M, Nagarajan S, Lee J, et al. Preliminary results of a functional MRI study of brain activation patterns in stuttering and nonstuttering speakers during a lexical access task. J Fluency Disord. 2003;28:337-356. [CrossRef] [Medline Link] [Context Link]20. Foundas AL, Bollich AM, Corey DM, et al. Anomalous anatomy of speech-language areas in adults with persistent developmental stuttering. Neurology. 2001;57:207-215. [CrossRef] [Full Text] [Medline Link] [Context Link]21. Fox PT, Ingham RJ, Ingham JC, et al. A PET study of the neural systems of stuttering. Nature. 1996;382:158-162. [CrossRef] [Full Text] [Medline Link] [Context Link]22. Neumann K, Euler H, von Gudenberg A, et al. The nature and treatment of stuttering as revealed by fMRI: a within- and between-group comparison. J Fluency Disord. 2003;28:381-410. [CrossRef] [Medline Link] [Context Link]23. Wu JC, Maguire G, Riley G, et al. Increased dopamine activity associated with stuttering. Clin Neurosci Neuropsychol. 1997;8:767-770. [Context Link]24. Goberman A, Blomgren M. Parkinsonian speech disfluencies: effects of l-dopa-related fluctuations. J Fluency Disord. 2003;28:55-70. [Context Link]25. Bothe AK, Davidow JH, Bramlett RE, et al. Stuttering treatment research, 1970-2005: I. Systematic review incorporating trial quality assessment of behavioral, cognitive, and related approaches. Am J Speech Lang Pathol. 2006;15:321-341. [CrossRef] [Medline Link] [Context Link]26. Bothe AK, Davidow JH, Bramlett RE, et al. Stuttering treatment research, 1970-2005: II. Systematic review incorporating trial quality assessment of pharmacological approaches. Am J Speech Lang Pathol. 2006;15:342-352. [CrossRef] [Medline Link] [Context Link]27. American Speech-Language-Hearing Association (ASHA). Guidelines for practice in stuttering treatment. Special Interest Division on Fluency and Fluency Disorders American Speech-Language-Hearing Association. ASHA Suppl. 1995;37(suppl 14):26-35. [Medline Link] [Context Link]28. Brady JP. The pharmacology of stuttering: a critical review. Am J Psychiatry. 1991;148:1309-1316. [Context Link]29. Brady JP, Rynn M. Stuttering: current pharmacological options. CNS Drugs. 1994;1:163-168. [Context Link]30. Maguire G, Riley G, Franklin B, et al. Olanzapine in the treatment of developmental stuttering: a double-blind, placebo-controlled trial. Ann Clin Psychiatry. 2004;16:63-67. [CrossRef] [Medline Link] [Context Link]31. Farnbach-Pralong D, Bradbury R, Copolov B, et al. Clozapine and olanzapine treatment decreases rat cortical and limbic GABAA receptors. Eur J Pharmacol. 1998;349:R7-R8. [CrossRef] [Medline Link] [Context Link]32. Sofuoglu M, Kosten T. Novel approaches to the treatment of cocaine addiction. CNS Drugs. 2005;19:13-25. [CrossRef] [Full Text] [Medline Link] [Context Link]33. Riley G. Stuttering Severity Instrument 3rd Edition (SSI-3). Austin, TX: Pro-Ed; 1994. [Context Link]34. Riley J, Riley G, Maguire G. Subjective screening of stuttering severity, locus of control and avoidance: research edition. J Fluency Disord. 2004;29:51-62. [CrossRef] [Medline Link] [Context Link]35. Martin RR, Haroldson SK, Triden KA. Stuttering and speech naturalness. J Speech Hear Disord. 1984;49:53-58. [Medline Link] [Context Link]36. Blomgren M, Roy N, Callister T, et al. Intensive stuttering modification therapy: a multidimensional assessment of treatment outcomes. J Speech Lang Hear Res. 2005;48:509-523. [CrossRef] [Medline Link] [Context Link]37. Andrews G, Ingham RJ. An approach to the evaluation of stuttering therapy. J Speech Hear Res. 1972;15:296-302. [Medline Link] [Context Link]38. Cordes AK, Ingham RJ. The reliability of observational data: II. Issues in the identification and measurement of stuttering events. J Speech Hear Res. 1994;37:279-294. [Medline Link] [Context Link]39. Ingham RJ. Stuttering and Behavior Therapy: Current Status and Empirical Foundations. San Diego, CA: College-Hill Press; 1984. [Context Link]40. Ingham RJ, Costello JM. Stuttering treatment outcome evaluation. In: Costello JM, ed. Speech Disorders in Adults: Recent Advances. San Diego, CA: College-Hill; 1985:189-223. [Context Link]41. Sawyer J, Yairi E. The effect of sample size on the assessment of stuttering severity. Am J Speech Lang Pathol. 2006;15:36-44. [CrossRef] [Medline Link] [Context Link] pagoclone; stuttering; percent syllables stuttered; clinical|00004714-201002000-00009#xpointer(id(R2-9))|11065213||ovftdb|SL00062905200346100911065213P78[CrossRef]|00004714-201002000-00009#xpointer(id(R2-9))|11065405||ovftdb|SL00062905200346100911065405P78[Medline Link]|00004714-201002000-00009#xpointer(id(R6-9))|11065405||ovftdb|SL0000533319834822611065405P82[Medline Link]|00004714-201002000-00009#xpointer(id(R8-9))|11065213||ovftdb|SL000047192004373511065213P84[CrossRef]|00004714-201002000-00009#xpointer(id(R8-9))|11065405||ovftdb|SL000047192004373511065405P84[Medline Link]|00004714-201002000-00009#xpointer(id(R9-9))|11065405||ovftdb|SL0006290519984175311065405P85[Medline Link]|00004714-201002000-00009#xpointer(id(R14-9))|11065213||ovftdb|00000958-200009000-00001SL0000095820003035911065213P90[CrossRef]|00004714-201002000-00009#xpointer(id(R14-9))|11065404||ovftdb|00000958-200009000-00001SL0000095820003035911065404P90[Full Text]|00004714-201002000-00009#xpointer(id(R14-9))|11065405||ovftdb|00000958-200009000-00001SL0000095820003035911065405P90[Medline Link]|00004714-201002000-00009#xpointer(id(R15-9))|11065405||ovftdb|SL0000533419812431711065405P91[Medline Link]|00004714-201002000-00009#xpointer(id(R17-9))|11065213||ovftdb|SL0000043620067855411065213P93[CrossRef]|00004714-201002000-00009#xpointer(id(R17-9))|11065405||ovftdb|SL0000043620067855411065405P93[Medline Link]|00004714-201002000-00009#xpointer(id(R18-9))|11065213||ovftdb|SL001366892002140711065213P94[CrossRef]|00004714-201002000-00009#xpointer(id(R18-9))|11065405||ovftdb|SL001366892002140711065405P94[Medline Link]|00004714-201002000-00009#xpointer(id(R19-9))|11065213||ovftdb|SL0001147520032833711065213P95[CrossRef]|00004714-201002000-00009#xpointer(id(R19-9))|11065405||ovftdb|SL0001147520032833711065405P95[Medline Link]|00004714-201002000-00009#xpointer(id(R20-9))|11065213||ovftdb|00006114-200107240-00013SL0000611420015720711065213P96[CrossRef]|00004714-201002000-00009#xpointer(id(R20-9))|11065404||ovftdb|00006114-200107240-00013SL0000611420015720711065404P96[Full Text]|00004714-201002000-00009#xpointer(id(R20-9))|11065405||ovftdb|00006114-200107240-00013SL0000611420015720711065405P96[Medline Link]|00004714-201002000-00009#xpointer(id(R21-9))|11065213||ovftdb|00006056-199607110-00023SL00006056199638215811065213P97[CrossRef]|00004714-201002000-00009#xpointer(id(R21-9))|11065404||ovftdb|00006056-199607110-00023SL00006056199638215811065404P97[Full Text]|00004714-201002000-00009#xpointer(id(R21-9))|11065405||ovftdb|00006056-199607110-00023SL00006056199638215811065405P97[Medline Link]|00004714-201002000-00009#xpointer(id(R22-9))|11065213||ovftdb|SL0001147520032838111065213P98[CrossRef]|00004714-201002000-00009#xpointer(id(R22-9))|11065405||ovftdb|SL0001147520032838111065405P98[Medline Link]|00004714-201002000-00009#xpointer(id(R25-9))|11065213||ovftdb|SL0004682420061532111065213P101[CrossRef]|00004714-201002000-00009#xpointer(id(R25-9))|11065405||ovftdb|SL0004682420061532111065405P101[Medline Link]|00004714-201002000-00009#xpointer(id(R26-9))|11065213||ovftdb|SL0004682420061534211065213P102[CrossRef]|00004714-201002000-00009#xpointer(id(R26-9))|11065405||ovftdb|SL0004682420061534211065405P102[Medline Link]|00004714-201002000-00009#xpointer(id(R27-9))|11065405||ovftdb|SL000083931995372611065405P103[Medline Link]|00004714-201002000-00009#xpointer(id(R30-9))|11065213||ovftdb|SL000121052004166311065213P106[CrossRef]|00004714-201002000-00009#xpointer(id(R30-9))|11065405||ovftdb|SL000121052004166311065405P106[Medline Link]|00004714-201002000-00009#xpointer(id(R31-9))|11065213||ovftdb|SL000036561998349r711065213P107[CrossRef]|00004714-201002000-00009#xpointer(id(R31-9))|11065405||ovftdb|SL000036561998349r711065405P107[Medline Link]|00004714-201002000-00009#xpointer(id(R32-9))|11065213||ovftdb|00023210-200519010-00002SL000232102005191311065213P108[CrossRef]|00004714-201002000-00009#xpointer(id(R32-9))|11065404||ovftdb|00023210-200519010-00002SL000232102005191311065404P108[Full Text]|00004714-201002000-00009#xpointer(id(R32-9))|11065405||ovftdb|00023210-200519010-00002SL000232102005191311065405P108[Medline Link]|00004714-201002000-00009#xpointer(id(R34-9))|11065213||ovftdb|SL000114752004295111065213P110[CrossRef]|00004714-201002000-00009#xpointer(id(R34-9))|11065405||ovftdb|SL000114752004295111065405P110[Medline Link]|00004714-201002000-00009#xpointer(id(R35-9))|11065405||ovftdb|SL000053331984495311065405P111[Medline Link]|00004714-201002000-00009#xpointer(id(R36-9))|11065213||ovftdb|SL0006290520054850911065213P112[CrossRef]|00004714-201002000-00009#xpointer(id(R36-9))|11065405||ovftdb|SL0006290520054850911065405P112[Medline Link]|00004714-201002000-00009#xpointer(id(R37-9))|11065405||ovftdb|SL0000533419721529611065405P113[Medline Link]|00004714-201002000-00009#xpointer(id(R38-9))|11065405||ovftdb|SL0000533419943727911065405P114[Medline Link]|00004714-201002000-00009#xpointer(id(R41-9))|11065213||ovftdb|SL000468242006153611065213P117[CrossRef]|00004714-201002000-00009#xpointer(id(R41-9))|11065405||ovftdb|SL000468242006153611065405P117[Medline Link]16533091Exploratory Randomized Clinical Study of Pagoclone in Persistent Developmental Stuttering: The EXamining Pagoclone for peRsistent dEvelopmental Stuttering StudyMaguire, Gerald MD; Franklin, David PsyD, MHA; Vatakis, Nick G. MD; Morgenshtern, Elena; Denko, Timothey MD; Yaruss, J. Scott PhD, CCC-SLP, ASHA; Spotts, Crystal MD; Davis, Larry MD; Davis, Aaron ND; Fox, Peter MD; Soni, Poonam MD; Blomgren, Michael PhD, CCC-SLP; Silverman, Andrew PhD; Riley, Glyndon PhDOriginal Contributions130