Abstract
Currently in the United States, an estimated 4% to 5% of adults have ADHD (Kessler et al., 2006). The core ADHD symptoms in adults include a frequent and persistent pattern of inattention/distractibility and/or hyperactivity-impulsivity (Adler, 2004; Adler & Chua, 2002; McGough & Barkley, 2004). These symptoms and their resulting clinical characteristics lead to significant functional deficits, such as poor executive functioning, poor time management, and motivational deficits in affected individuals (Barkley & Murphy, 2010). Such deficits have the potential to lead to marital separation and divorce, poor academic/occupational achievement, and substance abuse disorders (Adler & Chua, 2002; Adler & Cohen, 2004; Asherson, 2005; Biederman, 2004; McGough & Barkley, 2004; Montano, 2004; Wilens, 2004).
The most commonly used treatments for ADHD are stimulants such as methylphenidate and amphetamine preparations (Adler, 2004; Adler & Chua, 2002; Asherson, 2005; Wilens, Haight, et al., 2005, Wilens, Hammerness, et al., 2005). Response rates in adults have ranged from 25% to 95%, with the higher rates generally being seen in more recent studies (Adler & Chua, 2002; Adler, Spencer, McGough, Jiang, & Muniz, 2009; Kooij et al., 2004; Wilens, Spencer, & Biederman, 2002). Concerns about side effects may limit the utility of stimulants for adults with ADHD. These side effects include severe decreased appetite, insomnia, headache, and cardiac sequelae, including hypertension (Montano, 2004; Weiss & Murray, 2003). Potential adverse effects include drug-induced psychosis (Berman, Kuczenski, McCracken, & London, 2009) and stimulant substance use among high-risk patient subgroups (Wilens, 2004). Restricted use in patients with ADHD and psychiatric comorbidities (Berman et al., 2009; Montano, 2004; Wilens, 2004) is another limiting factor.
Modafinil is a wakefulness-promoting medication that is approved for the treatment of adult patients with excessive sleepiness associated with obstructive sleep apnea, shift work disorder, and narcolepsy (Minzenberg & Carter, 2008). Modafinil was shown to be effective and well tolerated at dosages up to 425 mg/day in a pooled analysis of children and adolescents with inattentive or combined (inattentive and hyperactive-impulsive) ADHD subtypes from three randomized, double-blind, placebo-controlled studies (Biederman & Pliszka, 2008). Specifically, modafinil significantly improved total scores on the ADHD Rating Scale–IV (ADHD-RS-IV) School Version—the primary efficacy measure—and also the Home Version, the Conners’ Parent Rating Scale–Revised: Short Form scores, and Clinical Global Impressions (CGI) of Improvement scores.
An earlier dose-ranging study conducted in adults with ADHD receiving 100, 200, or 400 mg of modafinil for 6 weeks, however, failed to demonstrate a benefit over placebo (Data on file, Teva Pharmaceutical Industries Ltd). In that study, modafinil was well tolerated in adults at dosages up to 400 mg/day. A few other studies suggest that modafinil may improve some prefrontal-dependent cognitive domains from the Cambridge Neuropsychological Test Automated Battery, including measures of working memory, spatial planning, and decision making in adult ADHD (Minzenberg & Carter, 2008; Taylor & Russo, 2000; Turner, Clark, Dowson, Robbins, & Sahakian, 2004). Based on the limited therapeutic options for ADHD treatment in adults, and the evidence of efficacy for modafinil in pediatric trials, the present study was conducted to evaluate the efficacy of modafinil versus placebo, as assessed by change from baseline in the Adult ADHD Investigator Symptom Rating Scale (AISRS) total score (Spencer et al., 2010), as well as the tolerability of modafinil in dosages of 255, 340, 425, or 510 mg/day in adults with ADHD.
Method
Study Design
This 9-week, multicenter, randomized, double-blind, placebo-controlled, parallel-group study evaluated the efficacy and tolerability of modafinil at a range of doses in adults with ADHD. The study was performed at 18 medical centers in the United States from May 2006 to January 2007. The study was conducted in full accordance with the Good Clinical Practice: Consolidated Guidance, approved by the International Conference on Harmonisation, and the applicable local laws and regulations. Written informed consent was obtained from each patient prior to starting the study.
The primary objective of the study was to determine whether different dosages of modafinil (255, 340, 425, or 510 mg/day) were more effective than placebo in alleviating the symptoms of ADHD in adults, as determined by the change from baseline in the total score from the investigator-completed AISRS at final visit. The study consisted of a 1- to 3-week screening period, followed by baseline assessment, and 1:1:1:1:1 randomization to five treatment groups for 9 weeks of treatment with modafinil 255, 340, 425, or 510 mg/day or placebo. After 222 patients had been enrolled in the study, however, the sponsor decided that it would not pursue the manufacturing of a 510-mg tablet formulation of modafinil, and so enrollment of patients in the 510-mg/day treatment group was stopped; all patients already enrolled in the 510-mg/day group were allowed to complete the study.
Patient Selection
Major Inclusion Criteria
Patients were included in the study if they met the full Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; American Psychiatric Association [APA], 2000) criteria for ADHD (combined type, predominantly inattentive subtype, or predominantly hyperactive-impulsive subtype), for which symptoms were present before the age of 7 years and persisted for at least the prior 6 months, according to a psychiatric/clinical evaluation using the Adult ADHD Clinical Diagnostic Scale (ACDS). The ACDS is an interview-based assessment, conducted by trained mental health care professionals, that reviews symptoms of childhood ADHD and recent (previous 6 months) symptoms of adult ADHD, including all 9 Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; APA, 1994) Criterion A symptoms of inattention and 9 hyperactivity/impulsivity symptoms plus an additional 14 non-DSM symptoms (e.g., trouble managing time and priorities, lack of self-discipline, moodiness, difficulty expressing anger; Adler & Cohen, 2004).
Patients who were taking medication for ADHD had to discontinue use of it, per physician instruction. The washout period was a minimum of 7 days after the last dose of an ADHD medication. No diagnostic tests were performed or rating scales administered until the washout phase was completed.
Eligible patients were also required to have a Hamilton Anxiety Scale (HAM-A) and Hamilton Depression Scale (HAM-D; Hamilton, 1960; Maier, Buller, Philipp, & Heuser, 1988) score <15, and an AISRS total score of >24 at the screening and baseline visits, with a difference in the AISRS total score from screening to baseline <25%. In addition, a CGI Severity of Illness (CGI-S; Guy, 1976) rating for ADHD of ≥4 at the baseline visit was required for study entry. Women of childbearing potential were required to use a medically accepted method of contraception during the study and for 30 days following their participation.
Major Exclusion Criteria
Exclusion criteria included a history or current diagnosis of schizophrenia, bipolar disorder, or other psychotic disorders; suicidal ideation, history of suicide attempt, or a clinical assessment of suicide risk; any acute psychiatric comorbidity (including but not limited to depression or other mood or anxiety disorder) that required pharmacotherapy, as determined by the Structured Clinical Interview for DSM-IV-TR (SCID) module assessment (Ekselius, Lindstrom, von Knorring, Bodlund, & Kullgren, 1994); a clinically significant sleep disorder; being intellectually challenged, as determined by the investigator; being satisfied with his or her current ADHD medication and having no unacceptable side effects; previous use of modafinil; use of other prescription medications for ADHD with psychoactive properties as of the baseline visit; drug or alcohol dependence within the prior 6 months; use of any antidepressant within 2 weeks before baseline; being pregnant or lactating; and presence of any clinically significant uncontrolled medical conditions.
Treatment
Patients were instructed to take six tablets by mouth, once daily in the morning. The study drug was titrated during double-blind treatment, starting with 85 mg of modafinil or matching placebo, and then increased by 85 mg/day during the first 3 weeks (on or about Days 3, 5, 8, 15, and 22) up to the assigned dosages of 255, 340, 425, or 510 mg/day. Patients and investigators were blinded to treatment assignment during the study until the database was locked for analysis and the treatment assignments unblinded.
Primary and Secondary Efficacy Outcomes
The primary study outcome was the mean change from baseline in the total score of the AISRS at final visit (Week 9 or last visit prior to early termination). AISRS is an interview-based, physician-reported instrument that assesses the severity of each of 18 individual criteria symptoms of ADHD in the DSM-IV-TR on a 4-point Likert-type scale (0 = none, 1 = mild, 2 = moderate, or 3 = severe).
Secondary outcome measures included the change from baseline in AISRS total and subscale scores, the Adult Self-Report Scale (ASRS; Adler et al., 2006; Kessler et al., 2007) v1.1 total and subscale scores, and the Time-Sensitive ADHD Symptom Scale (TASS; Adler et al., 2011) scores at 5 and 10 hr following study drug administration. The ASRS and TASS complement the AISRS in that they are patient-reported instruments, with the ASRS focusing on frequency (never, rarely, sometimes, often, very often) and the TASS on description (not at all, mild, moderate, severe, not applicable) of ADHD symptoms in a time-sensitive manner, relative to behavior in relation to 18 DSM-IV-TR criteria.
Other secondary outcomes were the percentage of responders (much or very much improved) according to the CGI of Change (CGI-C) clinical condition ratings (Guy, 1976) and the change from baseline in scores from the Behavior Rating Inventory of Executive Function–Adult Version (BRIEF-A™; cognitive function; Gioia, Isquith, Retzlaff, & Espy, 2002). In addition, quality of life was evaluated using the change from baseline in the Quality of Life Enjoyment and Satisfaction Questionnaire–Short Form (Q-LES-Q-SF; Endicott, Nee, Harrison, & Blumenthal, 1993; Mick, Faraone, Spencer, Zhang, & Biederman, 2008) and, for those in the study who were employed, the Endicott Work Productivity Scale (EWPS), which is a patient-rated instrument describing types of behavior and subjective feelings that may reduce productivity and efficiency in the work environment, with higher scores indicating greater symptom severity (0 = never to 4 = almost always; Endicott & Nee, 1997).
Tolerability Measurements
Tolerability was assessed by monitoring and evaluating reported adverse events throughout the study; clinical laboratory test results at Weeks 1, 5, and 9 and at final visit; vital signs at Weeks 1, 2, 3, 5, 7, and 9 and at final visit; electrocardiogram (ECG) results at Week 9 and at final visit; physical examination findings, including body weight measurement, at Week 9 and at final visit; the Pittsburgh Sleep Quality Index (PSQI; Buysse, Reynolds, Monk, Berman, & Kupfer, 1989) at Weeks 1, 5, 9, and at final visit; and concomitant medication usage throughout the study. Specific adverse events were predefined for expedited reporting and evaluation. These were termed protocol-defined adverse events and included severe psychiatric adverse events, suicidal ideation, and skin rashes. Skin rashes were monitored closely due to the potential for severe dermatologic reactions (e.g., Stevens–Johnson syndrome [SJS]), previously observed in pediatric patients receiving modafinil in clinical trials (Provigil [Package Insert], 2010). When these adverse events occurred, the study drug was immediately discontinued and the patient was brought in for an unscheduled visit within 24 hr, if possible, to evaluate the adverse event. Those patients evaluated for tolerability who had at least 1 postbaseline AISRS assessment were included in the efficacy analysis.
Statistical Analyses
The intent-to-treat (ITT) population included all patients who received the study drug and had at least one postbaseline AISRS assessment. The safety analysis set included all patients who received at least 1 dose of the study drug. Efficacy analyses were performed using the ITT population. Demographic and baseline characteristics were summarized by treatment group using descriptive statistics, as were the continuous variables of patient age, weight, height, body mass index, ACDS, HAM-A, and HAM-D, and the categorical variable of CGI-S.
The statistical hypothesis for the primary efficacy measure of change from baseline to final visit (Week 9 or last postbaseline observation) in the AISRS total score was tested using an ANCOVA model, with treatment as a factor and the corresponding baseline value as a covariate. The statistical model used for the primary efficacy analysis (i.e., ANCOVA) was also used for the analyses of all secondary variables. No separate treatment by covariate interaction was tested for any of the secondary analyses.
Adverse-event data were analyzed by frequency, severity, and attribution. No additional comparative analyses of adverse events were performed.
All statistical tests were two-tailed at the .05 level of significance, and all efficacy analyses used last observation carried forward (LOCF) in the event of patient withdrawal or discontinuation before Week 9. Due to the discontinued enrollment of the modafinil 510-mg/day treatment group, inferential statistics for pairwise comparisons of this group with placebo were not performed. Because this was a Phase 2 dose-finding study, p values were not adjusted for multiplicity.
Results
Patient Baseline Characteristics and Disposition
Of the 456 patients screened, 338 met entry criteria and were randomized (Figure 1). Of these, 330 received at least one dose of the study drug and were included in the safety analysis; 8 did not receive the study drug. Of the 330 patients evaluated for tolerability, 96% had at least one postbaseline AISRS assessment and were thus evaluable for efficacy. A total of 43% of patients withdrew from the study, including 22% of patients who withdrew due to adverse events, 26% in the modafinil groups, and 8% in the placebo group. A total of 57% of patients completed the 9-week study, including 54% of modafinil patients and 69% of placebo patients. Completer rates among the modafinil arms were similar, with 59% of patients in the 255-mg/day arm completing the study, compared with 51%, 53%, and 52% in the 340-, 425-, and 510-mg/day groups, respectively.
Patient disposition.
The treatment groups had similar demographic characteristics and baseline characteristics (Table 1). A similar percentage of patients in all modafinil groups combined and in the placebo group were either moderately or markedly ill according to the CGI-S rating (91% for both). In each treatment group, the mean HAM-A and HAM-D scores were below clinically significant levels (Table 1), and the mean ACDS scores indicated that patients met DSM-IV-TR criteria for adult ADHD. There were no clinically significant differences in baseline values across treatment arms.
Baseline Characteristics of Randomized Patients.
Note: BMI = body mass index; CGI-S = Clinical Global Impression of Severity of Illness; HAM-A = Hamilton Anxiety Scale; HAM-D = Hamilton Depression Scale; ACDS = Adult ADHD Clinical Diagnostic Scale.
Efficacy
The results for the primary efficacy measure, change from baseline to final visit in the adult AISRS total score, showed no statistically significant differences between any of the modafinil treatment groups and the placebo group (Figure 2). Among the secondary efficacy measures, none showed a statistically significant difference at final visit for patients treated with modafinil compared with placebo, nor were there any clinically relevant changes observed for any of the secondary variables at final visit (Table 2).
Change from baseline to final visit in the AISRS total score.
Change in Secondary Efficacy Measures at Final Visit.
Note: ASRS = Adult ADHD Self-Report Scale; TASS = Time-Sensitive ADHD Symptom Scale; CGI-C = Clinical Global Impression of Change; BRIEF-A™ = Behavior Rating Inventory of Executive Function-Adult Version; Q-LES-Q-SF = Quality of Life Enjoyment and Satisfaction Questionnaire–Short Form; EWPS = Endicott Work Productivity Scale.
Number of patients with observations at final visit: 255 mg/day (n = 60), 340 mg/day (n = 56), 425 mg/day (n = 60), 510 mg/day (n = 34), placebo (n = 58).
Number of patients with observations at final visit: 255 mg/day (n = 56), 340 mg/day (n = 57), 425 mg/day (n = 57), 510 mg/day (n = 34), placebo (n = 55).
Responders had CGI-C ratings of “much improved” or “very much improved.”
Number of patients with observations at final visit: 340 mg/day (n = 69).
Number of patients with observations at final visit: 255 mg/day (n = 64), 340 mg/day (n = 59), 425 mg/day (n = 66), 510 mg/day (n = 34), placebo (n = 64).
Number of patients with observations at final visit: 255 mg/day (n = 53), 340 mg/day (n = 48), 425 mg/day (n = 49), 510 mg/day (n = 26), placebo (n = 52).
Among patients who completed 9 weeks of the study, a few of the end points reached statistical significance (Table 3). Significantly greater improvements in AISRS score were observed for study completers treated with modafinil 340 mg compared with those given placebo, with a mean change of −18.4 (p = .044). Modafinil patients completing the study also showed significantly greater improvements in ASRS total scores in the 340-mg group (mean change, −18.6; p = .047) and the 425-mg group (mean change, −17.3; p = .027), mean TASS scores 10 hr postdose in the 425-mg group (mean change, −12.1; p = .015), change in mean score on the BRIEF-A™ in the 340-mg group (mean change, −14.9; p = .022) and the 425-mg group (mean change, −13.0; p = .026), and EWPS scores in the 340-mg group (mean change, −19.0; p = .045).
Change in Secondary Efficacy Measures at 9 Weeks.
Note: AISRS = Adult ADHD Investigator Symptom Rating Scale; ASRS = Adult ADHD Self-Report Scale; TASS = Time-Sensitive ADHD Symptom Scale; CGI-C = Clinical Global Impression of Change; BRIEF-A™ = Behavior Rating Inventory of Executive Function–Adult Version; Q-LES-Q-SF = Quality of Life Enjoyment and Satisfaction Questionnaire–Short Form; EWPS = Endicott Work Productivity Scale.
Number of patients with observations at 9 weeks: 255 mg/day (n = 35), 340 mg/day (n = 26), 425 mg/day (n = 29), 510 mg/day (n = 20), placebo (n = 37).
Number of patients with observations at 9 weeks: 255 mg/day (n = 32), 340 mg/day (n = 27), 425 mg/day (n = 30), 510 mg/day (n = 19), placebo (n = 44).
Responders had CGI-C ratings of “much improved” or “very much improved.”
Number of patients with observations at 9 weeks: 255 mg/day (n = 42).
Number of patients with observations at 9 weeks: 255 mg/day (n = 35), 340 mg/day (n = 29), 425 mg/day (n = 32), 510 mg/day (n = 17), placebo (n = 37).
p = .047. **p = .045. ***p = .044. ****p = .027. †p = .026. ††p = .022. †††p = .015.
Tolerability
The majority of patients (63%) received treatment with the study drug or placebo for ≥7 weeks, and exposure to the study drug was comparable across the treatment groups. The mean duration of exposure was 44.2 days for all the modafinil treatment groups combined, and 52 days for the placebo group. However, the median number of days of treatment in each treatment group was higher, 61 days for all modafinil treatment groups combined and 63 days for the placebo group.
During the double-blind period, 86% of all modafinil patients and 85% of the placebo patients experienced at least 1 adverse event. No p values were calculated for treatment differences in adverse-event rates.
The most frequently reported adverse event in all treatment groups was headache, which occurred in 78 (30%) patients who received modafinil and 15 (21%) of patients in the placebo group (Table 4). Other adverse events that occurred most frequently in the combined modafinil groups (≥10% and more often than placebo) were insomnia, nervousness, nausea, anorexia, anxiety, and dry mouth (Table 4). Most adverse events were mild to moderate in severity, with 30 patients reporting severe adverse events, including 11% in the combined modafinil groups (e.g., headache, anorexia, and insomnia) and 3% in the placebo group (e.g., tachycardia and palpitations).
Adverse Events Occurring in ≥10% of Patients Treated With Any Dosage of Modafinil and More Frequently Than Placebo.
Adverse events led to withdrawal from the study in 27% of patients in the modafinil groups and 8% of patients in the placebo group. The most frequent reasons for withdrawal due to adverse events among patients in the modafinil group were insomnia (8%), anxiety or nervousness (6% each), headache (5%), and nausea or depression (3% each). The most frequent reasons for withdrawal due to adverse events among patients receiving placebo were nervousness or depression (3% each). No adverse event resulting in withdrawal from the study was reported as serious, and no deaths occurred.
A total of 10 protocol-defined adverse events requiring expedited reporting (i.e., severe psychiatric adverse events, suicidal ideation, or skin rashes) occurred in nine patients (8 adverse events in seven patients receiving modafinil [0.4% receiving either 255 mg/day or 510 mg/day, 0.8% receiving 340 mg/day, and 1.7% receiving 425 mg/day] and 2 events in two patients receiving placebo [2.8%]). In the modafinil groups, there was 1 severe psychiatric event, 2 instances of suicidal ideation, and 5 skin rashes. The 2 adverse events in the placebo group were both skin rash. Although none of the 10 adverse events was judged to be serious by study investigators, two patients withdrew from the study as a result of these events, one in the modafinil 340-mg group because of suicidality, considered probably related to the study drug, and one in the modafinil 255-mg group due to severe psychosis, considered possibly related to the study drug, yet not judged to be a serious adverse event by the investigator.
Mean changes from baseline to final visit for pulse, systolic blood pressure, and diastolic blood pressure in patients treated with modafinil were slightly higher than in patients treated with placebo. Mean change in pulse rate was 3.0 (SD = 9.61) beats per minute (bpm) versus −0.03 (9.21) bpm in the placebo group, mean change in systolic blood pressure was 1.3 (10.24) mmHg versus −1.6 (12.88) mmHg, and mean change in diastolic blood pressure was 0.8 (7.34) mmHg versus 0.2 (8.66) mmHg, respectively. These changes in vital signs are consistent with the known profile of modafinil and were not considered clinically meaningful.
Of the six patients who had clinically significant abnormal vital sign measurements, two were receiving modafinil. One of these two experienced an increase in pulse rate (abnormal value, 120 bpm) and the other a decrease (abnormal value, 50 bpm). No patient had an adverse event in relation to a clinically significant vital sign abnormality.
A total of 33 patients (10%; 25 [9.7%] patients in the combined modafinil group and 8 [11.1%] in the placebo group) had shifts in ECG values, which were not considered to be clinically meaningful. Small mean decreases (the greatest being −4.5 ms in the 425-mg/day group, Fridericia’s formula) from baseline to end point in QTc intervals were observed in patients in the modafinil groups compared with small mean increases of 0.7 to 0.9 ms in patients in the placebo group.
Patients treated with modafinil had dose-related mean decreases in weight from baseline to final visit of −0.4, −1.1, −1.1, and −1.7 kg in the 255-, 340-, 425-, and 510-mg/day treatment groups, respectively, compared with a mean increase of 0.8 kg for patients in the placebo group. Clinically significant increases (≥7%) in body weight were observed in three patients (one patient each in the modafinil 255- and 425-mg/day treatment groups and the placebo treatment group); however, no adverse events associated with clinically significant increases in weight were reported in patients treated with modafinil.
Discussion
The current study was conducted as a follow-up to an earlier dose-ranging study of modafinil 100, 200, and 400 mg/day in adults with ADHD (Data on file, Teva Pharmaceutical Industries Ltd. Although no reduction in ADHD symptoms was seen in the earlier study, the present study was conducted to evaluate whether the benefit seen in children and adolescents with ADHD could be achieved in adults using an adult-specific efficacy measure. Consistent with the earlier study in adults, however, the current study demonstrated no benefit on the primary efficacy measure of modafinil compared with placebo for adults with ADHD, as no significant difference in the change from baseline in total AISRS score at final visit (LOCF) was observed.
An intriguing finding of the study was that statistically significant improvements were seen among the subgroups of patients who completed the full 9 weeks and received modafinil 340 and 425 mg/day. At 9 weeks, parameters showing improvement in association with the 340-mg/day dosage included patient self-assessment, cognition, and quality of life/productivity. Those associated with the 425-mg/day dosage included patient self-assessment, cognition, and time-sensitive effect on symptom severity.
A number of studies have shown that modafinil improves cognition related to the frontal cortex, which is implicated in the pathophysiology of ADHD (Konrad, Neufang, Hanisch, Fink, & Herpertz-Dahlmann, 2006; Minzenberg & Carter, 2008; Zang et al., 2005). A study of modafinil 200 mg/day in 20 adults with ADHD (Turner et al., 2004) showed it to have cognition-enhancing effects similar to those effects seen in healthy adult volunteers. The study authors noted, though, that longer and larger studies might explore modafinil’s effect on different subtypes of ADHD, because not all may respond equally, and also examine whether the cognitive improvements seen with modafinil actually result in clinical improvement.
Although it is thought to be structurally and pharmacologically different from amphetamine, with a distinct mechanism of action (Minzenberg & Carter, 2008; Turner et al., 2004), preclinical studies of modafinil in animals and a pilot study in male volunteers have demonstrated dopaminergic activity, with a mechanism of action similar to methylphenidate in its ability to bind to and block dopamine transporters (Volkow et al., 2009). Nevertheless, reports of modafinil abuse are uncommon compared with those of stimulant drugs (Volkow et al., 2009). This difference is reflected in the classifications of these drugs: Modafinil is a Schedule IV drug, whereas methylphenidate is Class B, Schedule II.
The efficacy of modafinil, as measured by the change in total scores on the ADHD-RS-IV School and Home Versions and improvement in CGI-C at final visit, was demonstrated in three studies in children and adolescents with ADHD (Biederman et al., 2005; Biederman & Pliszka, 2008; Greenhill et al., 2006; Swanson et al., 2006). In the present study in adults, a modest but statistically significant response in cognitive measures (BRIEF-A™) was seen only in those patients receiving ≥340 mg/day who completed the study. As did the authors of the pediatric studies (Biederman et al., 2005; Greenhill et al., 2006), which were also of 9 weeks duration, the authors of the present study question whether additional benefit might be seen in a study of longer duration. Indeed, during the 7-week double-blind portion of the other study (Swanson et al., 2006), the investigators noted increasing separation of efficacy scores between the modafinil and placebo groups over time. In light of our findings, it might be worth exploring the specific interplay between the cognitive effects of modafinil and their impact on specific ADHD symptoms and subtypes in adults.
The fact that the efficacy of modafinil in children differs from that observed in adults is particularly interesting. The reason for this dichotomy is unknown. Whether findings in the current study might be associated with the lower effect size of modafinil in adults compared with children or differences in disease presentation and target symptoms can only be speculated. The fact that those patients who completed this study and received higher doses of modafinil (particularly the 340-mg dose) showed significant improvement in AISRS, ASRS, TASS at 10-hr postdose, BRIEF-A™, and EWPS is intriguing and, as mentioned, may warrant further investigation.
The overall withdrawal rate in this study was high (43%), and this may have influenced outcomes. (Of interest, termination of the modafinil pediatric program occurred while the adult study was under way.) Although the overall withdrawal rates in the pediatric studies were similar, if somewhat lower on average (30%-48%; Biederman et al., 2005; Greenhill et al., 2006; Swanson et al., 2006), the withdrawal rates by treatment groups appear to differ between this adult study and the pediatric studies. Discontinuation by adults receiving modafinil was 46% in the present study compared with 25% to 41% in the three pediatric studies (Biederman et al., 2005; Greenhill et al., 2006; Swanson et al., 2006), whereas the withdrawal rate among adults receiving placebo was 31%, compared with 47% to 61% in the three pediatric studies. Again, reasons for these differences, such as the possibility of parental influence on reporting adverse events or prompting withdrawal for lack of efficacy, can only be hypothesized. In contrast to the above studies, the overall withdrawal rate in a study of methylphenidate in adult ADHD was 9% (Medori et al., 2008), and the withdrawal rate in two 10-week studies of the nonstimulant atomoxetine in adult ADHD was 24% (Michelson et al., 2003).
In the present study, a forced titration schema was used, and all patients took six tablets each morning to maintain dose-group blinding. By contrast, in two of the pediatric studies, patients took up to five tablets per day to titrate modafinil up to an individually optimized dose (Biederman et al., 2005; Greenhill et al., 2006). In the third study (Swanson et al., 2006), patients were titrated to a dosage (either 340 mg/day or 425 mg/day) relative to their weight. In the study that looked at methylphenidate in adult ADHD, only the highest dose studied, 72 mg, was titrated. Patients took the study drug either once daily (18 mg or 36 mg) or twice daily (72 mg divided as two 36-mg capsules). Whether the titration or dosing schedule affected outcome is unknown.
Modafinil was reasonably tolerated in these adult patients with ADHD. The most frequently reported adverse event was headache, which occurred in 30% of patients who received modafinil and 21% of patients who received placebo. The adverse-event profile was similar to that seen in other studies of modafinil in adults, in which insomnia, headache, nervousness, nausea, anorexia, and anxiety were the most common complaints (Biederman et al., 2005; Czeisler et al., 2005; Greenhill et al., 2006; Pack, Black, Schwartz, & Matheson for US Modafinil in Obstructive Sleep Apnea Study Group, 2001; Schwartz, Hirshkowitz, Erman, & Schmidt-Nowara, 2003; Swanson et al., 2006; Taylor & Russo, 2000; US Modafinil in Narcolepsy Multicenter Study Group, 1998, 2000). However, stimulant-like side effects (insomnia, nervousness, and anxiety) were among the most common reasons for withdrawal from this study.
No cases of SJS were seen, although the issue of dermatologic events is of importance in light of a small number (eight) of reports of serious adverse events related to skin rash in children and adolescents treated with modafinil in the context of clinical trials (Rugino, 2007). A postmarketing review of reported adverse events determined that only five reports of severe cutaneous adverse reactions have been made since 1999 among approximately 673,000 adults worldwide who have taken modafinil (Rugino, 2007).
There were a number of limitations associated with this study. First, the high dropout rate may have influenced the effect size in the efficacy analysis, and this could be a factor in the difference in significant improvement between the population completing 9 weeks and the final-visit population. In addition, although the intention was to evaluate modafinil dosages of up to 510 mg/day, the allocation of patients to the 510-mg/day treatment arm was discontinued while the study was well into enrollment because the manufacturer made an independent corporate decision not to pursue production of a 510-mg tablet formulation. Patients already enrolled and allocated to the 510-mg/day dosage continued in the study. Whether more robust results might have been seen if enrollment continued and inferential statistics for pairwise comparisons versus with placebo were performed is, again, a matter of speculation.
Conclusion
In summary, modafinil—even at modestly higher dosages than those previously studied in adults and dosages similar to those proved to be beneficial in children and adolescents—did not elicit a therapeutic response in adults with ADHD, as assessed by the AISRS at final visit. However, significant improvement in ADHD symptoms was observed in patients who completed 9 weeks of treatment with modafinil at 340 mg/day compared with patients treated with placebo.
Footnotes
Acknowledgements
The authors wish to acknowledge James Youakim, MD, formerly of Cephalon, Inc., which is now part of Teva Pharmaceutical Industries Ltd., for input on the design and conduct of the study, as well as Aji Nair, PhD, and Angela Kaya, PhD, Teva Pharmaceutical Industries Ltd., for overall editorial support during the development of this article. Funding for editorial support was provided by Teva Pharmaceutical Industries Ltd., to The Curry Rockefeller Group, LLC (Tarrytown, NY, USA), which included assistance in literature searches, editing and fact checking, graphic support, and, at the request of and with guidance from the authors, support for the drafting of this article.
Declaration of Conflicting Interests
Dr. Valerie K. Arnold and Dr. David Feifel each report that their institutions received research funding for this study. Dr. Craig Q. Earl is a former employee and Dr. Ronghua Yang is a current employee of Teva Pharmaceutical Industries Ltd. In the past 12 months, Dr. Lenard Adler has received grant/research support from the following sources: Shire Development, Inc, Chelsea Therapeutics, Eli Lilly & Company, the National Institute on Drug Abuse (NIDA), and Theravance, Inc. In the past 12 months, Dr. Adler has consulted with Alcobra Pharmaceuticals, Major League Baseball, Otsuka Pharmaceuticals, Shire Development, Inc, and Theravance, Inc. In the past 12 months, Dr. Adler has received honoraria from Alcobra Pharmaceuticals, Major League Baseball, Otsuka Pharmaceuticals, Shire Development, Inc, and Theravance, Inc. Dr. Adler has received royalty payments (as inventor) from the NYU School of Medicine for license of adult ADHD scales and training materials since 2004.
Funding
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: This study was sponsored by Cephalon, Inc. (NCT000315276), which is now a wholly owned subsidiary of Teva Pharmaceuticals Industries Ltd.