Abstract
Keywords
Introduction
ADHD is a heterogeneous neurobehavioral disorder characterized by a persistent pattern of developmentally inappropriate inattentiveness, impulsivity, and hyperactivity. It is the most common pediatric neurobiological condition with a worldwide prevalence of 5.3% of children and adolescents (Polanczyk, deLima, Horta, Biederman, & Rohde, 2007). Although the core diagnostic criteria of ADHD are hyperactivity, impulsivity, and inattention (per the Diagnostic and Statistical Manual of Mental Disorders, 5th ed.; DSM-5; American Psychiatric Association [APA], 2013), many children with ADHD experience difficulties in executive function and emotional regulation (Mares, McLuckie, Schwartz, & Saini, 2007; Strine et al., 2006; van Stralen, 2016).
Executive function is an umbrella term for a group of cognitive processes that affect goal-directed behavior, play a role in self-regulation, and control emotional functioning (Barkley, 1997; Gioia, Isquith, Retzlaff, & Espy, 2002; Gioia, Isquith, Guy, & Kenworthy, 2000). Executive function includes inhibition, set-shifting, regulation of behavior, organizing, planning, working memory, and emotional regulation (Barkley, 1997). It has been postulated that executive function deficit is a core feature of ADHD (Barkley, 1997; Brown, 2006). Regardless of whether it is a core feature, clinically, it is evident that many children with ADHD have difficulty with executive function deficits and that such deficits affect their daily function (Biederman & Spencer, 2004).
Although some aspects of executive function can be assessed using neuropsychological tests such as the Stroop test and the Tower of London Task, such tests fail to assess real-world executive function behaviors. The Behavioural Rating Inventory of Executive Function (BRIEF) was developed to assess such real-world expressions of executive function in the home using the BRIEF–Parent (BRIEF-P) assessed by the parent (GA et al., 2002; Gioia et al., 2000). It has been more recently used to assess pharmacological effects on executive function in individuals with ADHD (Adler et al., 2013; Findling, Ginsberg, Jain, & Gao, 2009; Turgay et al., 2010).
Guanfacine extended release (GXR), marketed as INTUNIV XR®, is an approved pharmacotherapeutic treatment for ADHD. At the time of this study, GXR was approved in Canada as monotherapy for the treatment of ADHD in children aged 6 to 12 years and as adjunctive therapy to stimulants for the treatment of ADHD in children aged 6 to 12 years with a suboptimal response to stimulants. Since then, GXR has been approved, in Canada, for the above indications in the age group 6- to 17-year-olds. Guanfacine is a known antihypertensive compound, but its mechanism of action in ADHD is unknown. The efficacy and safety of INTUNIV XR in children was demonstrated in three placebo (PLB)-controlled monotherapy trials and one PLB-controlled adjunctive therapy trial on children and adolescents (Biederman, Melmed, Patel, McBurnett, Konow, et al., 2008; Hervas et al., 2014; Sallee, Mcgough, et al., 2009; Wilens et al., 2012). These trials were conducted on children and adolescents aged 6 to 17 years.
The efficacy of INTUNIV XR® was evaluated using the ADHD Rating Scale (ADHD-RS-IV) total score as the primary outcome measurement and was supported by the ADHD-RS-IV-Hyperactivity/Impulsivity and Inattentiveness subscales, the Clinical Global Impression of Improvement (CGI-I), and the Connors’ Parent and Teacher Rating Scales. INTUNIV XR® at doses of 1 to 4 mg was generally well tolerated. Most adverse events were mild or moderate in severity. In the short-term monotherapy studies, dose-related treatment-emergent adverse events (TEAEs; related to INTUNIV XR®) and with a higher incidence than in PLB patients included, in order of decreasing incidence, somnolence/sedation (38%), headache (23.8%), and fatigue (14%). Other frequently reported TEAEs included abdominal pain, dizziness, decreased appetite, irritability, lethargy, dry mouth, constipation, and hypotension (Biederman, Melmed, Patel, McBurnett, Konow, et al., 2008; Hervas et al., 2014).
Stimulant medication is widely recognized as first-line pharmacological therapy for treating ADHD (Floet, Scheiner, & Grossman, 2010; Hosenbocus & Chahal, 2012). It has been well documented that stimulant medication not only treats the core symptoms of ADHD but also results in improvement in executive function, cognitive performance, self-esteem, family functioning, academic focus, social adjustment, and emotional regulation (Biederman & Spencer, 2004; Group, 1999; Hinshaw, Buhrmester, & Heller, 1989; Robert et al., 2009; Vitiello, 2001). Stimulant medications, as well as atomoxetine, have been shown to have a positive improvement in executive function in children, adolescents, and adults with ADHD (Adler et al., 2013; Gau & Shang, 2010; Hale et al., 2011; Hosenbocus & Chahal, 2012). Despite their effectiveness, treatment with stimulants sometimes results in inadequate response in symptom control (Hervas et al., 2014). Such lingering ADHD symptoms clinically results in patients having ongoing functional impairment including executive function deficits. Furthermore, although stimulant medications have been shown to have a positive impact on executive function, little has been documented about the effect of GXR on executive function in children. Demonstrating that the addition of GXR to usual stimulant therapy is effective for symptom control as well as in improving executive function may influence clinical treatment algorithms and the need for health care resources to effectively manage patients.
The present phase IV study was conducted to evaluate the effect and safety of INTUNIV XR® (guanfacine extended-release [GXR]) compared with PLB as adjunct therapy to clinically optimized stimulants in improving executive function in children aged 6 to 12 years with a primary diagnosis of ADHD and ongoing clinically significant executive function deficits as measured by the BRIEF.
Method
A single-centre, randomized, double-blind, two-period crossover study that evaluated GXR in children patients with ADHD (ClinicalTrials.gov identifier: NCT01985581) was conducted in Canada. The study was performed in accordance with the current applicable regulations, the international Conference on Harmonisation of Good Clinical Practice, the principles of the Declaration of Helsinki, and local ethical and legal requirements. The study protocol was approved by an independent ethics committee, Institutional Review Board (IRB) Services. Written informed consent was obtained from each participant’s parent or legal guardian, and assent was obtained from each participant, as applicable, before commencing study-related procedures.
The study population consisted of pediatric patients, 6 to 12 years of age, with a primary diagnosis of predominantly inattentive, hyperactive/impulsive, or combined subtype based on the Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; APA, 2000) diagnosis of ADHD, who were treated with a stable stimulant (methylphenidate or amphetamine) regimen and presented with suboptimal executive function. ADHD was based on clinical assessment and ADHD-RS-IV. Suboptimal executive function was defined as having a t score of ≥65 on the BRIEF-P. Subjects were required to be receiving a stable dose of stimulant for a period of at least 30 days prior to enrollment, and stimulant was confirmed to be optimized by the investigator.
Additional important inclusion criteria included age-appropriate intellectual function, ability to swallow intact tablets, blood pressure measurements within the 95th percentile for age, sex, and height, and the willingness and the ability of the patient and parent/legally authorized representative (LAR) to comply with the study protocol. Exclusion criteria included comorbid psychiatric diagnosis (except Oppositional Defiant Disorder [ODD]), clinically significant medical illness, known personal history or presence of structural cardiac abnormalities, cardiovascular or cerebrovascular disease, serious heart rhythm abnormalities, syncope, tachycardia, cardiac conduction problems (such as QT-corrected [QTc] > 0.44 seconds or clinically significant heart block), and exercise-related cardiac events including syncope, presyncope, or clinically significant bradycardia. A known family history of sudden cardiac death, ventricular arrhythmia, or QT prolongation, as well as a known history of hypertension, glaucoma, a history of a seizures disorder (other than a simple childhood febrile seizure), and those with known renal or hepatic insufficiency were also excluded. Patients were not allowed to have taken another investigational product within 30 days prior to the enrollment visit (Visit 2). They must not have had a known or suspected allergy or clinically significant intolerance to guanfacine hydrochloride or any of its active ingredients. A history of adverse event or failure to respond to an adequate trial of an alpha2-agonist was prohibited. Pregnant and lactating patients and those considered a suicide risk were also excluded.
The study was composed of two treatment periods, where each patient was randomized in a double-blind fashion to either GXR or PLB in the first period and then received the other treatment in the second period of the study. Subjects were evaluated for eligibility at screening (Visit 1). At baseline (Visit 2), eligibility was confirmed, and subjects were randomized in a 1:1 ratio to receive either sequence treatment arm GXR/PLB or PLB/GXR. The actual treatment sequence was determined by randomization at the enrollment visit (Visit 2). Treatment assignment was automatically provided by the Interactive Web Randomization System (IWRS). Visits 3 through 6 were dose optimization visits where the dose could be changed by 1 mg upward titration each visit with the option for a one-time downward titration up to and including at Visit 6. Visits 2 through 6 occurred at 7 ± 3 day intervals. Visit 8 occurred 8 weeks ± 3 days after Visit 6, and at this visit, the weaning of the study drug was initiated. The wean occurred over 11 days with a further 10-day washout. The start of the next treatment arm was Visit 9, and Visits 9 through 15 replicated the Visits 1 to 8 for the remainder of the second treatment arm. A final, safety, follow-up visit occurred 21 ± 3 days after Visit 15.
Subjects received GXR or PLB at the start of both optimization phases (starting Visits 2 and 9), and the dose was optimized over 4 weeks. Subjects were then maintained at their optimal dose of GXR during the 8-week maintenance period, and tapered off GXR or PLB at the end of each phase. Stimulant doses were to remain fixed through the study. GXR was initiated at 1 mg/d and increased no sooner than weekly in 1 mg/d increments to a maximum of 4 mg/d. Subjects’ doses were optimized at the discretion of the investigator based on a significant reduction in ADHD symptoms and acceptable tolerability and safety. During the optimization phases (Visits 2 to 6 and 9 to 13), subjects could have a one-time 1 mg/d reduction in their dosing for tolerability reasons. Beginning at Visits 8 and 15, the dose of GXR or PLB was tapered over 11 days following a schedule based on the subjects’ specific dose prior to Visit 8 or 15.
The primary efficacy measure was the BRIEF-P, which is a 90-item questionnaire designed to measure real-world expression of executive function. It was completed by the parent. Each item is scored per the scoring instructions (computerized scoring used), and the results are reported as a t score. The BRIEF-P was completed by the parent or LAR at all visits except midway through the maintenance phase at Visits 7 and 14 (GA et al., 2002; Gioia et al., 2000).
Secondary efficacy measures included the ADHD-RS-IV, CGI-Severity of illness (CGI-S), and the CGI-I. The ADHD-RS-IV based on the DSM-IV-TR diagnostic criteria is an 18-item measuring ADHD symptomatology. Each item is scored from 0 (behavior occurring never or rarely) to 3 (behavior occurring very often) resulting in a total score from 0 to 54 with higher scores representing more severe ADHD symptoms (DuPaul, Power, Anastopoulos, & Reid, 1998). The ADHD-RS-IV was administered by the clinician at all study visits through dose tapering. The clinician-rated CGI-S scale, a global rating of disease severity, was performed at every study visit. The CGI-S rates the severity of a subject’s condition on a 7-point scale ranging from 1 (normal, not at all ill) to 7 (among the most extremely ill subjects; Guy, 1976). The CGI-I scale, a clinician-rated global rating of improvement relative to baseline as assessed by a seven-point scale ranging from 1 (very much improved) to 7 (very much worse), was also performed at all postbaseline study visits (Guy, 1976).
Safety assessment was based on reported adverse events, vital sign measurements, physical examinations (weight and height), and Columbia Suicide Severity Rating Scale (C-SSRS; “The Fourth Report on the Diagnosis,” 2004; Posner et al., 2010; Posner, Brown, & Stanley, 2011; Postner et al., 2010).
Data Analysis
A total of 40 patients were needed to complete this two-treatment crossover study. The probability was 80% that the study detected a treatment difference at a two-sided .05 significance level, if the true difference between treatments was 6.5 units. This was based on the assumption that the standard deviation of the difference in the response variables was 10. Assuming a 10% screen failure rate and 20% discontinuation rate, a minimum of 55 patients was required to obtain 40 completers. An ANOVA was used with treatment, period, and sequence as the fixed effects and patients nested within sequence as a random effect. Similar analyses were conducted for the secondary outcomes including that for ADHD-RS-IV total score and CGI-S score, and the exploratory endpoints of the differences in two composite scores of the BRIEF-P (metacognition and behavioral regulation).
All randomized patients received at least one dose of study medication and, therefore, were included in the safety population. Similarly, all randomized patients received at least one dose of study medication and completed at least one non-baseline BRIEF questionnaire during either Period 1 or 2 (i.e., INTUNIV XR® or PLB); they were included in the intent to treat (ITT) population.
Results
Of the 50 randomized patients, 39 patients (78.0%) completed the study (19 patients [76.0%] in the GXR-PLB sequence and 20 patients [80.0%] in the PLB-GXR sequence). Four patients out of 48 (8%) withdrew due to treatment-emergent side effects while on the PLB arm. There were no withdrawals due to adverse events during the active (GXR) arm. One patient withdrew from the study early due to insomnia, but this event started prior to first dose of study medication. One patient discontinued due to a protocol violation (changed the dose of stimulant they were taking), one withdrew consent, and four discontinued due to lack of efficacy. Patients were recruited from the study center’s clinical practice, advertising in medical offices and in print and radio advertisements.
Baseline characteristics were similar across treatment groups as shown in Table 1. The mean age (± SD) was 9.4 ± 1.6 years (range = 7.0-12.0 years) and 9.0 ± 1.4 years (range = 7.0-11.0 years) in the GXR-PLB sequence and PLB-GXR sequence, respectively. A majority of patients were male (22 males [88.0%] and 20 males [80.0%] in the GXR-PLB sequence and PLB-GXR sequence, respectively).
Patient Baseline Characteristics and Demographics Data.
Note. GXR-PLB = guanfacine extended-release–placebo; BRIEF-P = Behavioural Rating Inventory of Executive Function; ADHD-RS = dADHD-rating scale total score; CGI = Clinical Global Impressions of Severity of Illness.
Doses were optimized from Visit 2 to Visit 6 and Visit 15 to Visit 19, respectively. The mean optimized dose for GXR was 3.4 mg, whereas the mean dose of PLB was 3.9 mg. Compliance with GXR was 100%, while compliance with PLB was 95.8%.
The BRIEF-P global executive composite (GEC) score was significantly different between the GXR arm and the PLB arm (LS means = −3.0, 95% CI [−5.9, −0.2]; p value = .0392, ITT population). At the end of treatment and compared with baseline, a change in BRIEF-P GEC score of −9.2% (SE = 2.11%) versus −6.9% (SE = 1.94%) was reported in the GXR arm and the PLB arm, respectively (ITT population) as shown in Table 1 and illustrated in Figure 1.
Mean (SE) Parent-Reported BRIEF, GEC, and ADHD-RS at Baseline and End of Period.
In the secondary efficacy analyses, a significant difference between the GXR arm and the PLB arm was observed for these endpoints: ADHD-RS-IV total score (LS means = −6.9, 95% CI [−9.8, −4.0]; p value < .0001), CGI-S (LS means = −0.9, 95% CI[−1.4, −0.4]; p value = .0007), and CGI-I (LS means = −0.7, 95% CI [−1.2, −0.3]; p value = .0030; ITT population) as illustrated in Figure 1. A summary of the primary and secondary outcome measures are provided in Table 2.
Primary and Secondary Outcome Measures.
Note. GXR = guanfacine extended-release; PLB = placebo; BRIEF-P = Behavioural Rating Inventory of Executive Function; ADHD-RS = ADHD-rating scale; CGI-S = Clinical Global Impressions of Severity of Illness; CGI = Clinical Global Impressions of Improvement.
A significant correlation was detected for ADHD-RS-IV total score versus BRIEF-P GEC (Period 1: correlation coefficient: 0.78; p value < .0001 and Period 2: correlation coefficient 0.747; p value < .0001). Two exploratory analyses were performed. First, exploring the difference in the two composite scores (behavioral regulation index [BRI] and metacognition index [MI]) of BRIEF-P, only BRI of BRIEF-P was detected as significantly different between the GXR arm and PLB arm (LS means = −3.7, 95% CI [−7.1, −0.4]; p value = .0299). The BRIEF-P MI was approaching significance for a difference between the GXR arm and the PLB arm (LS means = −2.3, 95% CI [−5.0, −0.3]; p value =.0867). For the subpopulation of those patients who had ≥ 30% improvement of ADHD-RS-IV while on GXR, a significant difference was detected for the BRIEF-P GEC score (LS means = −7.6, 95% CI [−11.1, −4.1]; p value = .0002).
Forty-one patients (87%) reported TEAEs when taking GXR, and 41 patients (85%) reported TEAEs when taking PLB. The majority of TEAEs were considered mild, and only 6% of patients in both arms experienced moderate TEAEs. Moderate TEAEs were sleep disorder (2% vs. 2%), fatigue (2% vs. 0%), somnolence (2% vs. 0%), depressed mood (2% vs. 0%), vomiting (0% vs. 2%), and gastroenteritis (0% vs. 2%), respectively for the GXR arm and the PLB arm. No severe TEAEs were reported during the conduct of this study, and there were no deaths. No patients discontinued due to a TEAE in the GXR arm while 8% of patients (4/48) in the PLB arm discontinued due to a TEAE. The most commonly reported TEAEs (≥5% incidence) with the highest patient incidence rates in the GXR arm versus PLB were headaches, abdominal pain, fatigue, affect lability, gastroenteritis, somnolence, and sleep disorder as shown in Table 3. A total of 60% (28/47) of patients in the GXR arm and 27% (13/48) of patients in the PLB arm reported at least one TEAE considered to be related to treatment by the investigator. As a TEAE of interest, somnolence, reported for 11% of patients in the GXR arm and 4% of patients in the PLB arm, was experienced for a median duration of 7.0 days (range = 4-383 days) versus 8.5 days (range = 4-13 days), respectively.
Treatment-Emergent Adverse Events > 10% frequency in GXR group.
Note. GXR = guanfacine extended-release; PLB = placebo; AE = adverse events.
Overall, for lifetime history at the onset of the study, four patients had one or more “yes” responses on the C-SSRS with three patients having “yes” responses to the suicidal ideation category of the C-SSRS of “wish to be dead” and one patient to “nonspecific active suicidal thoughts.” Overall, during the study, three patients during the PLB phase responded “yes” to the C-SSRS, two responding “yes” to “wish to be dead” and one responding “yes” to “nonspecific active suicidal thoughts.” There were no “yes” responses to the C-SSRS during the GXR arms of the study.
Discussion
The results of this study show that adjunctive administration of the selective α2A-adrenoceptor agonist, GXR, to a psychostimulant in patients with suboptimal response to psychostimulants improves executive function, as measured by the parent in the home environment, using the BRIEF-P GEC score compared with psychostimulant with PLB. This adjunct therapy also improved the ADHD symptom control as assessed by the ADHD-RS-IV and the CGI-I and CGI-S scales. These findings corroborate the findings in two previous adjunctive therapy studies (Spencer, Greenbaum, Ginsberg, & Murphy, 2009; Wilens, Spencer, Biederman, Wozniak, & Connor, 1995). Assessing the congruency of these effects, a correlation was established between ADHD-RS-IV and BRIEF-P. In addition, BRIEF-P GEC score was significantly improved for patients experiencing 30% improvement on ADHD-RS-IV while on GXR (p value = .0002). Furthermore, twice as many patients on GXR experienced a 30% improvement on ADHD-RS-IV than those on PLB.
The majority (n = 39, 78%) of patients completed the study while nobody discontinued while on the GXR arm. GXR was generally well tolerated with all TEAEs being mild (94%) and moderate (6%) in severity; there were no serious TEAEs. The most common TEAEs reported were consistent with the known safety profiles of GXR (Biederman, Melmed, Patel, McBurnett, Donahue, & Lyne, 2008; Biederman, Melmed, Patel, McBurnett, Konow, et al., 2008; F. R. Sallee, Lyne, Wigal, & McGough, 2009; Floyd R. Sallee, Mcgough et al., 2009; Spencer et al., 2009; Wilens et al., 2012). Notably, headaches were reported more frequently for both the GXR (49%) and PLB (33%) arms as compared with that seen in the aforementioned studies. In this study, compared with GXR monotherapy studies, somnolence (11% vs. 43.9%) was seen less frequently. This is consistent with that seen in a previous adjunctive therapy study and may reflect the interaction between stimulant and GXR (Wilens et al., 2012).
The study results show that GXR was efficacious in augmenting treatment response in patients with a suboptimal response to a psychostimulant alone. It is important to note that the results presented in this study do not represent overall improvement with this treatment regimen; they show improvements beyond those already provided by psychostimulant monotherapy.
A reduction of ADHD-RS-IV symptoms of 30% is generally accepted to be clinically significant in monotherapy studies; however, the amount of improvement needed to be clinically significant for adjunctive therapy studies has not been established. For this reason, conservative estimates of clinically significant improvements were used for all measures. Clinically significant improvements were documented as the percentage of patients who encountered an improvement that would be considered clinically significant in monotherapy studies. For CGI-I, there was an improvement to a score of very much improved or improved in 57.4% (GXR) versus 27.7% (PLB). The CGI-S score of normal or borderline mentally ill was achieved in 29.8% (GXR) versus 10.6% (PLB). There was a 30% further reduction (these patients had already experienced reduction in ADHD symptoms with their stimulant usage) in ADHD-RS-IV scores in 51.1% (GXR) versus 25.5% (PLB). In all, 46.8% (GXR) versus 36.1% (PLB) patients went from a BRIEF-P GEC score in the clinically significant range (t score ≥ 65) to a normal range (t score < 65) as illustrated in Figure 2.
Clinically significant improvements.
It has been well established that GXR is effective in approximately 55% of patients (Biederman, Melmed, Patel, McBurnett, Konow, et al., 2008; Salle, Mcgough et al. 2009; Shire Pharmaceuticals, 2010; Wilens et al., 2012). The population of patients that had the most robust effect on their executive function were those who had a response to GXR based on improvement in ADHD-RS-IV; in fact, in that population, the improvement nearly doubled. The change in GEC on BRIEF-P was −12.3 and in the ANOVA model LS mean was −7.6 (p = .002). This would suggest that if a patient responds with the adjunctive therapy GXR to stimulant, then their executive function is also likely to improve.
There are a number of methodological limitations in the present study. To minimize the limitations of the crossover design, a washout period of 3 weeks, equivalent to 16.7 times the half-life, was included in the design (Boellner, Pennick, Fiske, Lyne, & Shojaei, 2007). Furthermore, a test for carryover was performed to ensure that there was no carryover effect when analyzing the primary outcome measure. The determination of whether a patient was optimized on medication was left up to the clinical judgment of the investigator based on a clinical assessment. As is frequently seen in ADHD studies, a fairly large PLB effect was seen. This may, in part, be due to the effect that compliance with psychostimulant prior to the study was not established. Given the frequency of study visits, it is probable that the compliance with psychostimulant improved during the study. The frequency of study visits and contact with physician also likely contribute to the PLB effect. Despite efforts to make this study as clinically relevant as possible, for safety and efficacy determinations, it was necessary to establish inclusion and exclusion criteria, which may limit the generalizability of the study results.
This single center, double-blind, PLB-controlled crossover trial in children with ADHD suggests that the addition of GXR to usual, optimized stimulant therapy is safe and beneficial in further improving executive function and ADHD symptom control. The level of illness as noted by the t scores on the BRIEF-P and the ADHD-RS-IV scores, despite optimized treatment with psychostimulants, suggests that there is a need for further options to treat ADHD and that GXR is useful to contribute to this treatment need.
Footnotes
Author’s Note
Dr. Erica Corsi, Center for Pediatric Excellence, Ottawa, Ontario, contributed to the clinical research as a subinvestigator. The statistician was Tina Haller, Ottawa, Ontario.
Declaration of Conflicting Interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr. van Stralen receives or has received grant support from Shire, Janssen, and Purdue Pharma. She acts or has acted on advisory board, for Janssen, Purdue Pharma and Shire. She has served as a speaker and a consultant for Janssen, Purdue Pharma and Shire. Dr. van Stralen has received research funding from Janssen, Purdue Pharma, and Shire.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by an investigator-initiated grant from Shire Canada Inc (Grant/Award IST-CAN-000549).