Lisdexamfetamine Dimesylate and Mixed Amphetamine Salts Extended-Release in Children with ADHD: A Double-Blind, Placebo-Controlled, Crossover Analog Classroom Study

Lisdexamfetamine Dimesylate and Mixed Amphetamine Salts Extended-Release in Children with ADHD: A Double-Blind, Placebo-Controlled, Crossover Analog Classroom Study Joseph Biederman, Samuel W. Boellner, Ann Childress, Frank A. Lopez, Suma Krishnan, and Yuxin Zhang Background: Lisdexamfetamine dimesylate is a therapeutically inactive prodrug in which d-amphetamine is covalently bound to l-lysine, a naturally occurring amino acid. Pharmacologically active d-amphetamine is released from lisdexamfetamine following oral ingestion. Methods: This phase 2, randomized, double-blind, placebo- and active-controlled crossover study compared the efficacy and safety of lisdexamfetamine (LDX: 30, 50, or 70 mg) with placebo, with mixed amphetamine salts extended-release (MAS XR: 10, 20, or 30 mg) included as a reference arm of the study, in 52 children aged 6 to 12 years with attention-deficit/hyperactivity disorder (ADHD) in an analog classroom setting. The primary efficacy measure was the Swanson, Kotkin, Agler, M-Flynn, and Pelham (SKAMP) Rating Scale; secondary efficacy measures included the Permanent Product Measure of Performance (PERMP) Derived Measures, and the Clinical Global Impression (CGI) Scale. Results: LDX treatment significantly improved scores on SKAMP-deportment, SKAMP-attention, PERMP-attempted, PERMP-correct, and CGI-improvement from baseline. Adverse events were similar for both active treatments. Conclusions: In a laboratory classroom environment, LDX significantly improved ADHD symptoms versus placebo in school-age children with ADHD. Key Words: ADHD, amphetamine, analog classroom, double blind, lisdexamfetamine, mixed amphetamine salts

A

ttention-deficit/hyperactivity disorder (ADHD) is one of the most common psychiatric disorders of childhood and is estimated to affect as many as 8%–12% of children worldwide (1). According to the DSM-IV-TR, ADHD is characterized by inattention, hyperactivity, and impulsivity emergent before age 7 that are both more severe and more frequent than commonly observed in children at a comparable developmental level (2). Psychostimulants are considered first-choice treatment for patients with ADHD, and their short- and long-term efficacy in improving ADHD symptoms and academic performance is well established (3,4). Nonetheless, despite the availability of multiple options, several therapeutic needs remain unmet. These include consistent medication delivery, adequate duration of action, and reduced potential for abuse. Lisdexamfetamine dimesylate (LDX) is a therapeutically inactive prodrug in which d-amphetamine is covalently bonded to l-lysine. Following ingestion, therapeutically active d-amphetamine is released via biotransformation (5). The oral availability

From the Clinical and Research Program in Pediatric Psychopharmacology (JB), Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts; Clinical Study Centers (SWB), Little Rock, Arkansas; Center for Psychiatry and Medicine, Inc., (AC) Las Vegas, Nevada; Children’s Developmental Center (FAL), Maitland, Florida; New River Pharmaceuticals Inc., (SK) Blacksburg, Virginia; Xtiers Consulting (YZ), Potomac, Maryland. Address reprint requests to Joseph Biederman, M.D., Pediatric Psychopharmacology Unit, (ACC 725) Massachusetts General Hospital, Boston, MA 02114; E-mail: [email protected]. Received August 14, 2006; revised March 16, 2007; accepted April 12, 2007.

0006-3223/07/$32.00 doi:10.1016/j.biopsych.2007.04.015

of d-amphetamine is similar to the availability of intact LDX; however, intact LDX is cleared within 4 hours, and measurable levels of d-amphetamine persist for longer than 12 hours (5). The rate of LDX-associated d-amphetamine absorption, as indexed by the time to maximum drug concentration (Tmax), is approximately midway between the absorption rates of immediate- and extended-release formulations of amphetamine products (e.g., Tmax⫽ 3.7 hours for a 70-mg oral dose of LDX) (5). It appears that LDX undergoes rate-limited hydrolysis. A recently completed double-blind, placebo-controlled, threedose, parallel group naturalistic study with children aged 6 to 12 years demonstrated that symptoms of ADHD were significantly reduced after treatment with LDX compared with placebotreated children, and LDX was well tolerated (6). Significant effects were observed following 1 week of exposure to LDX, and parents reported positive treatment effects as measured at approximately 10 AM, 2 PM, and 6 PM using the Conners’ Parent Rating Scale. This study was designed to evaluate the efficacy and safety of LDX in children aged 6 to 12 years with ADHD. In addition, it was designed to evaluate the duration of therapeutic response of LDX. We hypothesized that LDX would provide control of ADHD symptoms superior to placebo over the course of the school day and into the evening.

Materials and Methods Study Participants The efficacy and safety of LDX compared with placebo in the treatment of ADHD were examined in this phase 2, randomized, multicenter, double-blind, placebo- and active-controlled, crossover study with mixed amphetamine salts extended-release (MAS XR) included as a reference arm. This study, conducted in an analog classroom environment, enrolled children aged 6 to 12 years who satisfied DSM-IV-TR criteria for a primary diagnosis of BIOL PSYCHIATRY 2007;62:970 –976 © 2007 Society of Biological Psychiatry

J. Biederman et al. combined or predominantly hyperactive–impulsive subtype of ADHD. Key inclusion criteria included a history of treatment with a stable regimen of stimulant medication, ability to follow classroom instructions, and functioning at age-appropriate academic levels. Exclusion criteria included presence of comorbid illness that could interfere with study participation or impact the efficacy and tolerability of LDX or MAS XR, documented allergy or intolerance to MAS XR, history of drug abuse, and concomitant medications with central nervous system effects. Other exclusion criteria included a current comorbid psychiatric diagnosis that would contraindicate treatment with MAS XR or LDX or confound efficacy or safety assessments, history of seizures within the last 2 years, tic disorders, hyperthyroidism, cardiac disorders, and significant laboratory abnormalities. This study was conducted at four study sites, each operating under the direction of its principal investigator. All activities were conducted in accordance with the Declaration of Helsinki and Good Clinical Practice according to the International Conference on Harmonisation guidelines. The appropriate institutional review boards at each research site approved the study protocol, all amendments, and informed consent forms. Written informed consent was obtained at study initiation from the subject’s parent/guardian along with assent from the subject. Study Design The study design comprised a 1) screening period (Visit 1), 2) dose-titration period (Visits 2 to 5), 3) double-blind crossover period (Visits 6 to 8), 4) final study visit (Visit 9), and 5) 30-day telephone follow-up. The school laboratory environment included an analog classroom and lasted for 13 hours. After screening and a washout period of at least 3 days, eligible subjects entered the dose-titration period, which involved open-label administration of MAS XR for 3 weeks. MAS XR was initiated at 10 mg/day. Depending on therapeutic response, the dose either was titrated upward at increments of 10 mg or remained the same during each subsequent weekly visit. If needed for tolerability, the dose could be decreased by 10-mg increments to a minimum dose of 10 mg/d MAS XR. The final dose of MAS XR at the end of the third titration week was considered the optimal daily dose and was used in the subsequent double-blind phase. On the last day of the dose-titration period, subjects participated in a practice assessment in the laboratory school environment. Subjects were familiarized with the staff and procedures associated with the double-blind period of the study. At the time of the practice assessment, subjects were receiving their optimal daily dose of MAS XR only. Following the dose-titration period, subjects entered the double-blind crossover portion of the study. Subjects received all three treatments during this phase. The order of treatments was randomized. The dose of LDX subjects received was determined by the optimized dose of MAS XR. At the end of each week, subjects were crossed over so that at the end of the double-blind period, each subject received 1 week of placebo, 1 week of MAS XR (at the individually optimized dose), and 1 week of LDX (in an approximately equivalent dose by amphetamine base content to MAS XR). At the end of each week, subjects participated in a laboratory classroom session with evaluation of behavioral and safety parameters. The schedule and procedures used in laboratory classroom sites were well-validated protocols used in previous timeresponse medication studies (7–9). On each classroom day, subjects arrived at approximately 6:30 AM and departed at

BIOL PSYCHIATRY 2007;62:970 –976 971 approximately 7:45 PM. During the double-blind period, subjects took the randomized treatment each morning at home for the first 6 days, and the Day 7 dose was administered during the analog classroom visit. The laboratory school assessment day (including the practice visit) comprised classroom sessions arranged at approximately 1, 2, 3, 4.5, 6, 8, 10, and 12 hours after the morning dose, with each session lasting about 30 minutes. During the last classroom visit, each participant had an indwelling catheter placed in a vein of the forearm for repeated plasma sampling. Blood samples were taken at 1, 2, 3, 4.5, 6, 8, 10, and 12 hours postdose. Immediately following collection, samples were centrifuged for 10 min at 4°C and 3000 rpm, and stored at –20°C. Plasma samples were assayed by CEDRA (Austin, Texas), using liquid chromatography/mass spectrometry/mass spectrometry method with a lower limit of quantification (LOQ) of 2.0 ng/mL for d-amphetamine and 1.0 ng/mL for intact LDX following LDX administration, and an LOQ of .5 ng/mL for d-amphetamine and .2 ng/mL for l-amphetamine following MAS XR administration. Pharmacokinetic Analysis Pharmacokinetic parameters included AUClast, Cmax, and Tmax based on blood samples collected on the last school assessment day (Visit 8) as well as the plasma drug concentrations at each time point (predose and at 1, 2, 3, 4.5, 6, 8, 10, and 12 hours postdose). Pharmacokinetic parameters were computed for d-amphetamine, l-amphetamine, and intact LDX using noncompartmental methods with WinNonlin (Enterprise Version 4.0, Pharsight, Cary, North Carolina). AUClast was defined as area under the drug concentration-time curve from time 0 to the last quantifiable concentration and calculated using the linear trapezoidal rule. Cmax was defined as maximum observed drug concentration at steady state over the time interval of 0 –12 hours on Day 8, and Tmax was the time to maximum drug concentration. All subjects receiving study medication underwent a safety evaluation during a final study visit no longer than 3 days after the last laboratory school visit or at the time of early study withdrawal. Follow-up interviews were conducted by telephone 30 days after the last dose of study medication to collect information about any serious adverse events (AEs) occurring during that period. Outcome Measures Primary Efficacy Measures. The primary efficacy measure was the least squares (LS) mean of the average scores from the Swanson, Kotkin, Agler, M-Flynn, and Pelham (SKAMP) Deportment Rating Scale (10) across a treatment day. The SKAMP rating scale measures the classroom manifestations of ADHD using an independent observer rating of subject impairment in classroom behavior (10). Comprising two subscales (deportment [SKAMPDS] and attention [SKAMP-AS]), SKAMP is scored on a 7-point impairment scale ranging from 0 to 6, with higher scores indicating more severe symptoms. The SKAMP rating scale was completed by classroom raters for each subject during each classroom session over the course of each laboratory school assessment day. Secondary Efficacy Measures. Secondary efficacy measures included SKAMP-AS (10) described earlier, Permanent Product Measure of Performance—Attempted (PERMP-A) and Correct (PERMP-C) scores (7), and Clinical Global Impressions (CGI) Scale scores (11). The PERMP is a validated 10-min math test developed to evaluate response to stimulant medication. Conwww.sobp.org/journal

972 BIOL PSYCHIATRY 2007;62:970 –976

J. Biederman et al.

taining 400 age-appropriate math problems, the test is scored to obtain an objective measure of academic performance by grading the number of attempted and completed problems (9). Subjects were given different levels of the math test based on their ability, as determined by a math pretest completed during the practice visit. Different versions of the math tests for a given level were used across the multiple classroom sessions so that subjects did not repeat the same test more than once during the classroom day. Both SKAMP and PERMP have been shown to be sensitive to dosage and time effects of stimulant medications (7). The CGI rating scale was used to evaluate global improvement in symptoms over time (11). The CGI-S (severity scale) was assessed at baseline, rating the severity of symptoms on a 7-point scale ranging from 1 (no symptoms) to 7 (very severe symptoms). At each postbaseline visit, symptom improvement for each subject was assessed by the clinician on the CGI-I (improvement scale) using a 7-point scale ranging from 1 (very much improved) to 7 (very much worse). Safety Measures. Information on AEs, vital signs, laboratory parameters, electrocardiograms (ECGs), and physical examinations was collected on all subjects. At each study visit, AEs and concomitant medications were recorded, as were blood pressure and heart rate. The ECGs were conducted at screening, practice visit (Visit 5), double-blind treatment visits (Visits 6 to 8), and at the final study visit. Physical examination and laboratory evaluations were conducted at screening and at study completion.

crossover study for a power of greater than .80 to detect a difference between LDX and placebo at the significance level of .05 (two sided) using a related t test. Considering that 10%–15% of subjects were predicted to drop out of the study during drug titration, this study planned to enroll 56 subjects into the dose titration period and to randomize approximately 48 subjects into double-blind treatment, using a balanced 3 ⫻ 3 Latin Square with six treatment sequences. The actual treatment received by an individual subject was determined by identical block-randomization schedules for all three study arms to maintain balance. Prepackaged drug kits each with a unique randomization number were shipped in advance to the four study sites with site personnel blind to sequence of study medication. Upon randomization, study site personnel assigned the next available randomization number to an eligible subject and selected the drug kit that corresponded to that subject’s optimal dose of MAS XR as determined in the dose-titration period. Statistical Analysis Analysis of efficacy was performed on the intent-to-treat (ITT) population that included all subjects randomized to treatment who had at least one SKAMP-DS treatment average score postrandomization. The treatment average score was defined as the mean daily average across the eight measurements (approximately 1, 2, 3, 4.5, 6, 8, 10, and 12 hours after the morning dose). A mixed-effects model of analysis of variance (ANOVA) was conducted on averages of the primary and all secondary efficacy measures across the treatment day, as well as CGI-I for the ITT population. The fixed effects were defined as three levels of treatment and three levels of period, and random effect was the subject-within-site. The three levels of treatment were LDX (30, 50, and 70 mg combined), MAS XR (10, 20, and 30 mg com-

Sample Size Determination and Randomization Previous studies of amphetamine products using the SKAMP-DS to assess efficacy in children with ADHD have disclosed an effect size of greater than .50. Assuming a similar effect size for LDX, 40 subjects would need to complete this

Table 1. Demographics and Baseline Characteristics of Study Subjects Randomized Population Characteristic Ethnicity, n (%) White Black Hispanic Other Gender, n (%) Male Female Mean Age (years ⫾ SD) ADHD Type, n (%) Combined Mean Age of ADHD Onset (years ⫾ SD) Mean Time Since Diagnosis (years ⫾ SD) Prior Treatment, n (%) Amphetamine Methylphenidate Stimulant NOS Stimulants with Atomoxetine Other Not Listed CGI Severity, n (%) Moderately Ill Markedly Ill Severely Ill

Cohort A (n ⫽ 10)

Cohort B (n ⫽ 17)

Cohort C (n ⫽ 25)

Safety Population (n ⫽ 52)

7 (70.0) 2 (20.0) 0 1 (10.0)

10 (58.8) 3 (17.6) 4 (23.5) 0 (0)

12 (48.0) 7 (28.0) 4 (16.0) 2 (8.0)

29 (55.8) 12 (23.1) 8 (15.4) 3 (5.8)

7 (70.0) 3 (30.0) 7.8 ⫾ 1.3

11 (64.7) 6 (35.3) 9.6 ⫾ 1.7

15 (60.0) 10 (40.0) 9.2 ⫾ 1.7

33 (63.5) 19 (36.5) 9.1 ⫾ 1.7

10 (100) 5.8 ⫾ 1.8 2.1 ⫾ 2.2

17 (100) 6.3 ⫾ 2.8 3.3 ⫾ 2.0

25 (100) 5.5 ⫾ 2.2 3.8 ⫾ 2.4

52 (100) 5.8 ⫾ 2.3 3.3 ⫾ 2.3

3 (30.0) 4 (40.0) 0 (0) 0 (0) 1 (10.0) 2 (20.0)

7 (41.2) 6 (35.3) 0 (0) 2 (11.8) 1 (5.9) 1 (5.9)

13 (52.0) 3 (12.0) 6 (24.0) 3 (12.0) 0 (0) 0 (0)

23 (44.2) 14 (26.9) 6 (11.5) 5 (9.6) 1 (1.9) 3 (5.8)

6 (60.0) 3 (30.0) 1 (10.0)

11 (64.7) 3 (17.6) 3 (17.6)

15 (60.0) 5 (20.0) 5 (20.0)

32 (61.5) 11 (21.2) 9 (17.3)

ADHD, attention-deficit/hyperactivity disorder; CGI, Clinical Global Impression Scale; NOS, not otherwise specified.

www.sobp.org/journal

J. Biederman et al.

BIOL PSYCHIATRY 2007;62:970 –976 973

Figure 1. Subject disposition. LDX, lisdexamfetamine; MAS XR, mixed amphetamine salts extended-release.

bined), and placebo. Given a significant p value (p ⬍ .05) of overall treatment effect, pairwise comparisons of LS means between individual treatment were conducted using a t test. The primary efficacy pairwise comparison in this study was LDX (30, 50, and 70 mg combined) versus placebo. The profile of therapeutic response over the day was evaluated by analyzing the primary outcome of SKAMP-DS and the secondary outcomes of SKAMP-AS, PERMP-A, and PERMP-C, observed at each classroom session on the treatment assessment day, using a mixed-effects ANOVA model followed by a t test pairwise comparison of active treatment versus placebo, given a significant overall effect (p ⬍ .05). The same ANOVA model was used to assess CGI, examining LS mean values. Descriptive results by CGI rating were also conducted a priori. Categorical analysis of CGI was also performed post hoc. Improvement was defined by CGI-I scores of 1 or 2 and nonimprovement defined as CGI-I scores of 3, 4, 5, 6, or 7 was assessed using McNemar’s test. QT intervals corrected (QTc) for heart rate using Bazett’s and Fridericia’s formulae were calculated. Outlier analyses of QT interval, QTc (Fridericia) interval, and QTc (Bazett) interval had cutoff values defined as the obtained value ⱖ 500 msec, and a change-from-baseline value ⱖ 60 msec and 30 –59 msec (12). Safety results were summarized descriptively. The outlier analyses of systolic and diastolic blood pressure and pulse rate with the cutoff values were defined as systolic blood pressure ⱖ 120 mm Hg, diastolic blood pressure ⱖ 80 mm Hg, and pulse rate ⱖ (mean ⫹ 2 · SD) or ⱕ (mean – 2 · SD).

Results Demographic Characteristics Of the 52 subjects enrolled, 50 completed the study, which was conducted from September 30, 2004, through December 23, 2004. Demographics and baseline characteristics of subjects are listed in Table 1. The majority of subjects were male (64%), and the mean age was 9.1 (⫾1.7) years. The average time since ADHD diagnosis was 3.3 (⫾ 2.3) years, and all had received amphetamine or methylphenidate treatment in the past. All subjects were diagnosed with ADHD combined subtype (both hyperactive–impulsive and inattentive subtypes). Of the 52 subjects titrated, 3 cohorts were identified based on the optimal MAS XR dose at the time of dose titration. Cohort A

(n ⫽ 10) was optimized on 10 mg/day, Cohort B (n ⫽ 17) was optimized on 20 mg/day, and Cohort C (n ⫽ 25) was optimized on 30 mg/d MAS XR. Two subjects in cohort A terminated the study while on placebo treatment during the first double-blind treatment week after randomization. The primary reason for discontinuation was viral gastroenteritis in the first subject; the second subject was lost to follow-up (Figure 1). The study observed protocol deviations on inclusion and exclusion criteria for 16 subjects with all being granted exemptions to enter the study. No subject was terminated during the study because of a protocol violation. The ITT population included 50 subjects. Pharmacokinetics At last visit, pharmacokinetics were assessed. The largest evaluable group for pharmacokinetic measures was Cohort C. In this group, d-amphetamine from LDX reached median peak plasma levels in 4.5 hours and MAS XR at 6 hours. The mean Cmax value for d-amphetamine following MAS XR (30 mg) administration was 119 ⫾ 52.5 ng/mL. The mean Cmax value for d-amphetamine following LDX (70 mg) was 155 ⫾ 31.4 ng/mL (Table 2). Table 2. Pharmacokinetic Parameters of Lisdexamfetamine Dimesylate (LDX) and Mixed Amphetamine Salts Extended-Release (MAS XR) Cmax (ng/mL)

AUClast (h·ng/mL)

8 4.50 4.50–6.0 15.33

8 155 ⫾ 31.4 99.9–187 20.34

8 1326 ⫾ 285.8 851.8–1618 21.56

9 6.00 3.00–12.00 52.77

9 119 ⫾ 52.5 49.8–218 43.96

9 1019 ⫾ 436.2 492.7–1899 42.83

Tmax (h) 70 mg LDX (d-Amphetamine) n Mean (⫾ SD)a Range %CV 30 mg MAS XR (d-Amphetamine) n Mean (⫾ SD)a Range %CV

AUClast, area under the drug concentration-time curve from time 0 to the last quantifiable concentration; Cmax, maximum observed drug concentration; %CV, coefficient of variance; Tmax, time to maximum drug concentration. a Data presented as mean (⫾ SD) for Cmax and AUClast. Data presented as median for Tmax.

www.sobp.org/journal

974 BIOL PSYCHIATRY 2007;62:970 –976

J. Biederman et al.

Figure 2. (A) Time-course in change in Swanson, Kotkin, Agler, M-Flynn, and Pelham (SKAMP) Rating Scale least squares (LS) mean scores across assessment day (intentto-treat [ITT] population). (B) Time-course in change in Permanent Product Measure of Performance (PERMP) LS mean scores across assessment day (ITT population). LDX, lisdexamfetamine; MAS XR, mixed amphetamine salts extended-release.

Low interpatient variability of pharmacokinetic parameters, as measured by coefficient of variance (%CV), may indicate that LDX can provide more consistent drug delivery among patients. This may be a result of the LDX prodrug formulation, which is not dependent on typical drug-release mechanisms. Additional study would be helpful to examine these findings further. Efficacy Outcomes Primary Efficacy Measures. LDX treatment significantly improved SKAMP-DS scores as compared with placebo. At endpoint, subjects treated with LDX had a mean SKAMP-DS score of .8 ⫾ .1 and placebo-treated subjects had a mean SKAMP-DS score of 1.7 ⫾ .1 (p ⬍ .0001). Subjects treated with MAS XR also produced significant decreases in mean SKAMP-DS scores at endpoint as compared with placebo (.8 ⫾ .1 vs. 1.7 ⫾ .1; p ⬍ .0001). Secondary Efficacy Measures. The LS mean SKAMP-AS score for both LDX and MAS XR was 1.2, a significant decrease relative to the placebo LS mean (1.8; p ⬍ .0001 for both active treatments vs. placebo). A post hoc analysis of SKAMP-AS, examining duration as measured by the change in score at each hour from the first measurement at 1 hour postdose, indicated significant improvement in LDX-treated subjects beginning at 2 hours, and with MAS XR-treated subjects beginning at 3 hours (Figure 2A). In addition, LS means of PERMP average scores for combined doses of active treatments across the treatment day were highly significant compared with placebo, with both associated with robust increases in the number of attempted and www.sobp.org/journal

correct math problems. The LS means PERMP-A scores were 133.3 for LDX, 133.6 for MAS XR, and 88.2 for placebo (p ⬍ .0001 for both active treatments vs. placebo), and the LS means PERMP-C scores were 129.6 for LDX, 129.4 for MAS XR, and 84.1 for placebo (p ⬍ .0001 for both active treatments vs. placebo). In a post hoc analysis of PERMP-A and PERMP-C ratings, the duration as measured by the change in score at each hour from first measurement at 1 hour postdose, favored both active treatments in the ITT population at all time points starting 2 hours postdose (Figure 2B). On the CGI-I scale, both LDX (2.2) and MAS XR (2.3) scores indicated significant improvement compared with placebo (4.2) (p ⬍ .0001). Categorically, investigators rated 74% (32% very much improved and 42% much improved) of LDX-treated and 72% (16% very much improved and 56% much improved) of MAS XR-treated subjects improved on the CGI-I compared with 18% of subjects on placebo (Figure 3). Safety Assessment. Eighty-nine AEs were reported by 29 of the 52 subjects (56%) during the study; 52 were reported during dose titration with MAS XR and 37 during the double-blind treatment period. Of these 89 events, 56 were mild and 33 were considered moderate. One subject in the placebo group was terminated in the first postrandomization week of the study due to viral gastroenteritis. The most common AEs reported during open-label dose titration with MAS XR were headache (15%), decreased appetite (14%), and insomnia (10%). During the double-blind part of the study, 16% of subjects reported AEs on LDX, 18% on MAS XR,

BIOL PSYCHIATRY 2007;62:970 –976 975

J. Biederman et al.

Table 4. Mean (Standard Deviation) of Vital Signs Parameters Recorded on Assessment Day During Double-Blind Treatment for Which a Treatment Effect of p ⬍ .05 Was Observed Vital Signs Diastolic BP (mm Hg): 2.5 hours postdose 5.0 hours postdose Pulse (bpm): 2.5 hours postdose

LDX (n ⫽ 50)

MAS XR (n ⫽ 50)

Placebo (n ⫽ 50)

66.6 (6.4)a 66.6 (7.3)a

64.7 (6.8)a 64.2 (7.4)a

62.0 (7.1) 61.8 (6.7)

91.3 (14.5)a

89.9 (11.6)a

84.6 (11.8)

BP, blood pressure; LDX, lisdexamfetamine dimesylate; MAS XR, mixed amphetamine salts extended-release. a p ⬍ .05, compared with placebo. Figure 3. Clinical Global Impression Scale—Improvement at assessment day from baseline (intent-to-treat population). LDX, lisdexamfetamine dimesylate; MAS XR, mixed amphetamine salts extended-release.

and 15% on placebo. AEs occurring at an incidence of ⬎ 2% during the double-blind treatment period were insomnia (8%), decreased appetite (6%), and anorexia (4%) with LDX; decreased appetite (4%), upper abdominal pain (4%), abdominal pain (4%), vomiting (2%), and insomnia (2%) with MAS XR; and vomiting (4%), insomnia (2%), and upper abdominal pain (2%) with placebo (Table 3). No serious AEs or deaths were reported during the study. Changes in vital signs during the double-blind phase of the trial were small and showed no clinically meaningful trends (Table 4). Furthermore, there was no apparent trend in the observed results of the outlier analyses of vital signs. Measurements of ECG parameters were taken three times a day on Visits 6, 7, or 8 during the double-blind treatment period and revealed small increases in the QT and QTc interval compared with placebo that were not found to be clinically significant (Table 5). No prolonged QT or QTc changes, defined as ⱖ 60 msec from baseline, were observed under any of the treatments. For the QTcF interval, a change-from-baseline value of 30 to 59 msec was not observed in any subject at 2.5 and 5 hours postdose but was seen in three subjects at 10.5 hours postdose in the LDX group. The corresponding numbers for MAS XR were one subject at 2.5 hours postdose, one subject at 5 hours postdose, and two subjects at 10.5 hours postdose. For placebo, the corresponding numbers were none at 2.5 and 5

hours postdose and one subject at 10.5 hours postdose. There was no apparent trend in the observed results of these outlier analyses of ECG QT parameters. By voltage criteria, a clinically significant abnormal ECG, indicating left ventricular hypertrophy, was observed for one subject receiving LDX. The investigator determined this to be unrelated to study medication, however, and the subject’s ECG was normal at the conclusion of the study.

Discussion In this study, LDX was significantly more effective than placebo in treating the symptoms of ADHD in school-age children. Significant improvements were made in both SKAMP-DS and SKAMP-AS scores with LDX compared with placebo, and treatment effects were seen at 12 hours postdose, the last time point measured. In addition, the numbers of math problems attempted and correct (as measured by PERMP) were significantly greater in LDXtreated subjects. Importantly, clinician ratings also indicated improvement with LDX, as measured by CGI, with the majority of subjects showing improvement and nearly a third being very much improved. LDX was generally well tolerated, and no meaningful abnormalities were observed in clinical laboratory data. The safety profile was similar to that seen in other studies with stimulants. The strengths of this study include the evaluation of schoolage children (the population most often treated for ADHD) (1,13), the use of validated efficacy measures in a sophisticated analog classroom setting previously described in ADHD treat-

Table 3. Adverse Events (⬎ 2%) During Either Titration Period or Double-Blind Treatment Period in Any Group Number (%) of Subjects Reporting Titration Period

Double-Blind Period

Adverse Events

MAS XR (n ⫽ 52)

LDX (n ⫽ 50)

MAS XR (n ⫽ 50)

Placebo (n ⫽ 52)

Any Events Abdominal Pain Upper Abdominal Pain Upper Respiratory Tract Infection Decreased Appetite Headache Affect Lability Insomnia Vomiting Anorexia

24 (46%) 3 (6%) 3 (6%) 2 (4%) 7 (14%) 8 (15%) 2 (4%) 5 (10%) 1 (2%) 1 (2%)

8 (16%) 0 0 1 (2%) 3 (6%) 0 0 4 (8%) 0 2 (4%)

9 (18%) 0 2 (4%) 1 (2%) 2 (4%) 0 0 1 (2%) 1 (2%) 0

8 (15%) 0 1 (2%) 0 0 0 0 1 (2%) 2 (4%) 0

LDX, lisdexamfetamine dimesylate; MAS XR, mixed amphetamine salts extended-release.

www.sobp.org/journal

976 BIOL PSYCHIATRY 2007;62:970 –976

J. Biederman et al.

Table 5. Mean (Standard Deviation) of Electrocardiogram Parameters Recorded on Assessment Day During Double-Blind Treatment for Which a Treatment Effect of p ⬍ .05 Was Observed ECG Parameters Heart rate (bpm): 5.0 hours postdose 10.5 hours postdose QRS interval (msec): 10.5 hours postdose QTc (Fridericia) interval (msec): 2.5 hours postdose 10.5 hours postdose

LDX (n ⫽ 50)

MAS XR (n ⫽ 50)

Placebo (n ⫽ 50)

84.2 (11.9)a 87.3 (14.6)a

85.6 (14.3) 88.6 (13.2)a

88.2 (13.5) 92.5 (12.5)

81.6 (7.8)a

82.3 (8.8)a

79.2 (7.8)

383.2 (15.4)a 383.0 (16.8)

382.2 (13.7)a 380.6 (15.4)a

377.1 (17.4) 375.3 (17.5)

ECG, electrocardiogram; LDX, lisdexamfetamine dimesylate; MAS XR, mixed amphetamine salts extended-release; QTC, QT intervals corrected. a p ⬍ .05, compared with placebo.

ment research (7,10,11), and the use of evaluations in both the classroom and nonclassroom setting. Limitations include that the objective of the study was not designed to compare efficacy or safety of LDX and MAS XR, as well as the short duration of the study, which did not provide the ability to extrapolate efficacy and safety findings to the long-term treatment generally required in management of ADHD symptoms. Because subjects with some comorbid psychiatric diagnoses were excluded from the current study, it may not reflect the real-world population of ADHD patients. Also, because participating children had been previously responsive to psychostimulants, our results do not generalize to treatment-naïve subjects. The open-label dose optimization of MAS XR may have resulted in improved tolerability in the double-blind period in this study, thus limiting the conclusions regarding tolerability of LDX from this part of the study. Finally, despite the diversity of the study population, the subgroups were of insufficient size to determine relative benefit concerning gender, age, or ethnic background. Despite these considerations, our study indicates that LDX appears to be a well-tolerated and effective treatment for schoolage children with ADHD, having a duration of action that extends through the school day and into the early evening. This finding is consistent with the results from the previously reported study of LDX. Longer-term clinical studies are needed to confirm these therapeutic benefits. This study was supported by New River Pharmaceuticals Inc. and Shire Development Inc. Joseph Biederman, MD, receives or received research support from, is or has been a speaker for, or is or has been on the advisory board for the following pharmaceutical companies: Shire, Eli Lilly, Pfizer, McNeil, Abbott, Bristol-Myers Squibb, New River Pharmaceuticals, Cephalon, Janssen, Novartis, UCB Pharma, AstraZeneca, Forest Laboratories, GlaxoSmithKline, and Neurosearch. Dr. Biederman has also received research support from Stanley Medical Institute, Inc.; Lilly Foundation; Prechter Foundation; National Institute of Mental Health; National Institute of Child Health and Human Development, and National Institute on Drug Abuse. Samuel W. Boellner, MD, is a consultant for

www.sobp.org/journal

Cephalon and has served as an investigator for CellTech, Cephalon, New River Pharmaceuticals, Novartis, and Shire. Ann Childress, MD, is a consultant and on the speakers bureaus for Novartis and Shire. Frank A. López, MD, is a consultant for Eli Lilly and New River Pharmaceuticals; receives research support from Cephalon, Eli Lilly, New River Pharmaceuticals, Novartis Noven, and Shire; and is on the speakers bureaus for Cephalon, Novartis, Shire, and Shire-Biochem Canada. Suma Krishnan, MS, is a full-time employee of New River Pharmaceuticals. Yuxin Zhang, PhD, is a consultant for New River Pharmaceuticals. Supplementary material cited in this article is available online. 1. Biederman J, Faraone SV (2005): Attention-deficit hyperactivity disorder. Lancet 366:237–248. 2. American Psychiatric Association (2000): Disorders usually first diagnosed in infancy, childhood, or adolescence. In: Diagnostic and Statistical Manual of Mental Disorders DSM-IV-TR. 4th ed. Text revision. Washington, DC: American Psychiatric Association. 3. Pliszka SR, Crismon ML, Hughes CW, Corners CK, Emslie GJ, Jensen PS, et al. (2006): The Texas Children’s Medication Algorithm Project: Revision of the algorithm for pharmacotherapy of attention-deficit/hyperactivity disorder. J Am Acad Child Adolesc Psychiatry 45:642– 657. 4. Dulcan M (1997): Practice parameters for the assessment and treatment of children, adolescents, and adults with attention-deficit/hyperactivity disorder. American Academy of Child and Adolescent Psychiatry. J Am Acad Child Adolesc Psychiatry 36:85S–121S. 5. Krishnan S (2006): A multiple-dose single-arm pharmacokinetics study of oral lisdexamfetamine (LDX; NRP104) in healthy adult volunteers. Abstract presented at the New Clinical Drug Evaluation Unit 46th Annual Meeting; June 12–15, 2006; Boca Raton, Florida. 6. Biederman J, Krishnan S, Zhang Y, McGough JJ, Findling RL (2007): Efficacy and tolerability of lisdexamfetamine dimesylate (NRP-104) in children with attention-deficit/hyperactivity disorder: A phase III, randomized, multicenter, double-blind, parallel-group study. Clin Ther 29: 450 – 463. 7. Swanson J, Wigal S, Greenhill L, Browne R, Waslick B, Lerner M, et al. (1998): Objective and subjective measures of the pharmacodynamic effects of Adderall in the treatment of children with ADHD in a controlled laboratory classroom setting. Psychopharmacol Bull 34: 55– 60. 8. McCracken JT, Biederman J, Greenhill LL, Swanson JM, McGough JJ, Spencer TJ, et al. (2003): Analog classroom assessment of a once-daily mixed amphetamine formulation, SLI381 (Adderall XR), in children with ADHD. J Am Acad Child Adolesc Psychiatry 42:673– 683. 9. Swanson JM, Agler D, Fineberg E, Wigal S, Flynn D, Fineberg K (2000): University of California Irvine, Laboratory School protocol for pharmacokinetic and pharmacodynamic studies. In: Greenhill LL, Osman B, editors. Ritalin: Theory and Practice. Larchmont, NY: Mary Ann Liebert; 405– 432. 10. Wigal SB, Gupta S, Guinta D, Swanson JM (1998): Reliability and validity of the SKAMP rating scale in a laboratory school setting. Psychopharmacol Bull 34:47–53. 11. Guy W (1976): Clinical global impression (CGI). In: ECDEU Assessment Manual for Psychopharmacology. Rockville, MD: US Department of Health and Human Services, Public Health Service, Alcohol Drug Abuse and Mental Health Administration, HIMH Psychopharmacology Research Branch, 218 –222. 12. International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use: The clinical evaluation of QT/QTC interval prolongation and proarrhythmic potential for non-antiarrhythmic drugs: E14. May 12, 2005. 13. Scahill L, Schwab-Stone M (2000): Epidemiology of ADHD in school-age children. Child Adolesc Psychiatr Clin N Am 9:541–555, vii.