Published online November 13, 2006
PEDIATRICS
Vol. 118
No. 6
December 2006, pp.
e1701-e1706
(doi:10.1542/peds.2005-2999)
Low-Dose Atomoxetine for Maintenance Treatment of Attention-Deficit/Hyperactivity Disorder
Jeffrey H. Newcorn, MDa,
David Michelson, MDb,
Christopher J. Kratochvil, MDc,
Albert J. Allen, MD, PhDb,
Dustin D. Ruff, PhDb,
Rodney J. Moore, PhDb for the Atomoxetine Low-dose Study Group
a Department of Psychiatry, Mount Sinai Medical Center, New York, New York
b Lilly Research Laboratories, Indianapolis, Indiana
c Department of Psychiatry, University of Nebraska Medical Center, Omaha, Nebraska
 |
ABSTRACT
|
|---|
OBJECTIVE. Data from acute studies of atomoxetine in patients with attention-deficit/hyperactivity disorder suggest that a dose of 
1.2 mg/kg per day is required to attain a maximal symptom response. However, lower doses could be effective during maintenance treatment, which would reduce drug exposure and potential problems related to tolerability during chronic treatment.
METHODS. Patients 6 to 16 years of age who had a robust response to an initial acute trial of atomoxetine were assigned randomly under double-blind conditions to continue treatment for up to 8 months with either the dose to which they had responded acutely (1.2–1.8 mg/kg per day, N = 116, continued same dose) or a lower dose (0.5 mg/kg per day, N = 113, low dose). The primary outcome measure was relapse, defined as an Attention-Deficit/Hyperactivity Disorder Rating Scale-IV-Parent Version: Investigator-Administered and Scored total score returned to 
90% of the original baseline value (before acute treatment) for 2 consecutive visits. Mean change in Attention-Deficit/Hyperactivity Disorder Rating Scale total score was assessed as a secondary outcome.
RESULTS. At randomization, symptom severity was low and similar in both groups. At end point, relapse rates did not differ between the groups. Mean change in Attention-Deficit/Hyperactivity Disorder Rating Scale total score from the conclusion of acute treatment to end point also was not different between groups. Reports of affective lability were higher in the patients in the low-dose group. Also, increases in heart rate (compared with when atomoxetine was started) were higher in the patients in the continued same-dose group than in the low-dose group. Finally, increases in weight over the course of the trial were greater for the low-dose group than the continued same-dose group.
CONCLUSIONS. For patients who experience a robust response to atomoxetine, it may be possible to retain the response during maintenance treatment with a reduced dose of atomoxetine.
Key Words: atomoxetine • ADHD • dose • maintenance treatment
Abbreviations: ADHD—attention-deficit/hyperactivity disorder • K-SADS-PL—Kiddie Schedule for Affective Disorders and Schizophrenia for School-Aged Children-Present and Lifetime Version • ADHD RS—Attention-Deficit/Hyperactivity Disorder Rating Scale-IV-Parent Version: Investigator-Administered and Scored • CGI-ADHD-S—Clinical Global Impressions-Attention-Deficit/Hyperactivity Disorder-Severity • CHQ—Child Health Questionnaire
The acute dose response to atomoxetine for the treatment of attention-deficit/hyperactivity disorder (ADHD) has been well characterized,1 with data suggesting that for most patients, maximal response is associated with a dose of 1.2 mg/kg per day. Placebo-controlled studies also have demonstrated that for patients who respond to acute treatment, continued drug treatment is superior to placebo in preventing symptom return and relapse.2 However, dose response during maintenance treatment has not been investigated in controlled studies. This question is of interest, because it is important to know whether it is possible to retain symptom control with a lower dose, which could provide flexibility to accommodate potential tolerability problems, as well as reduce drug exposure during long-term treatment.
We hypothesized that continued treatment with atomoxetine at the final acute-phase dose would be superior to continued treatment at a lower dose (ie, one which had previously not been associated with robust response in previous trials). To test this hypothesis, we assigned patients randomly who participated in an acute study of atomoxetine and demonstrated a robust response to continuation treatment with either the effective titrated dose of atomoxetine or with a reduced dose (0.5 mg/kg per day).
|
METHODS
|
|---|
Participants
Patients were initially enrolled in a double-blind, placebo-controlled comparator trial of atomoxetine and OROS methylphenidate that was previously reported.3 Children and adolescents 6 to 16 years of age entering this comparator trial (n = 516) met Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition,4 criteria for ADHD (any subtype) as determined by clinical history and confirmed by semistructured interview (Kiddie Schedule for Affective Disorders and Schizophrenia for School-Aged Children-Present and Lifetime Version [K-SADS-PL]).5 Symptom severity at entry was required to be at least 1.5 SD above age and gender norms as assessed by the Attention-Deficit/Hyperactivity Disorder Rating Scale-IV-Parent Version: Investigator-Administered and Scored (ADHD RS).6 Patients with bipolar disorder, a psychotic illness, seizures, pervasive developmental disorder, or who were taking concomitant psychoactive medications were excluded from the study. Comorbid anxiety and tic disorders are contraindicated in OROS methylphenidate product labeling and, therefore, were excluded. Other concurrent psychiatric diagnoses were permitted and were assessed by clinical interview and confirmed by K-SADS-PL.
Patients could have been treated previously with stimulants or could have been treatment-naive. However, for ethical reasons, patients were excluded if they had been treated previously with an adequate trial of methylphenidate or amphetamine and either did not experience at least some improvement in ADHD signs and symptoms (ie, nonresponders) or had intolerable adverse effects.
This study was conducted at 20 sites in the United States. After oral description of the study, written informed consent was obtained from a parent or guardian for each patient, and each youth provided written assent. The study was reviewed and approved by each site’s ethical review board and conducted in accordance with the ethical standards of the 1975 Declaration of Helsinki as revised in 2000.
Measures
The primary efficacy measure was the investigator-administered and investigator-scored version of the ADHD RS, a validated 18-item scale with well-described psychometric properties.6 The validity of clinician administration and scoring of the ADHD RS based on observer reports has previously been demonstrated.7 The Clinical Global Impressions-ADHD-Severity (CGI-ADHD-S)8 and the Child Health Questionnaire (CHQ)10 were included as secondary measures. At the outset of the trial, all raters were trained in the use of the study instruments using observed live and video interviews.
Safety Analyses
Weight and vital signs, including heart rate and blood pressure, were measured at each visit. CYP2D6 genotyping was performed at the screening visit. Laboratory tests (hematology, chemistry, urinalysis, urine drug screen, and serum pregnancy test for females), an electrocardiogram, and physical examination were performed at the screening visit and at end point. Adverse events and concomitant medications were collected by the physician at each visit via open-ended discussion with the parent or guardian.
Study Design
The randomized treatment group was drawn from an acute study of atomoxetine and OROS methylphenidate. All patients received 7 to 9 weeks of atomoxetine acutely before randomization. Approximately half of these were patients randomized to acute treatment initially with atomoxetine, and the other half were patients who first received OROS methylphenidate or placebo for 6 weeks, and then received an acute course of atomoxetine. However, regardless of the order, all subjects were treated with atomoxetine for the same period of time acutely before randomization into this study. After this acute treatment with atomoxetine (maximum dose: 1.8 mg/kg per day), those patients with a 40% reduction in ADHD RS total score were assigned randomly to continue therapy on their current dose (continued same dose) or to a decreased dose of atomoxetine (low dose: 0.5 mg/kg per day), for 8 months under double-blind conditions. The maximum dose for the atomoxetine titration (1.8 mg/kg) was the dose tested in the large majority of the premarketing trials. Although the product label indicates a maximum recommended dose of 1.4 mg/kg, there was no label at the time this study was initiated. The dose of 0.5 mg/kg per day in the low-dose arm was chosen because in an earlier dose-response study,1 this dose had been found to have a modest effect relative to placebo but was less efficacious than 1.2 mg/kg per day. Investigators and patients were blinded to the efficacy criteria, as well as the timing of rerandomization. Dose was kept constant on a weight-adjusted basis for the duration of the study.
Data Analyses
The hypothesis that continued treatment with atomoxetine at the final acute-phase dose was superior to continued treatment with a lower dose was tested by statistical comparison of Kaplan-Meier time-to-relapse curves using the log-rank test. The curves were tabulated for each treatment group (previous versus lower dose), pooling all patients regardless of the treatment they received during the acute comparator trial. Patients were defined as meeting the relapse criterion if symptoms worsened sufficiently so that the ADHD RS total score returned to 90% of the original baseline value (before acute treatment) for 2 consecutive visits. This is the same definition of relapse used in an earlier relapse prevention study.2 Secondary measures included mean change from randomization to end point on the ADHD RS, CGI-ADHD-S, and CHQ.
The ADHD RS total score, its inattentive and hyperactive/impulsive subscales, the CGI-ADHD-S scale, and the CHQ were analyzed using a last-observation-carried-forward analysis of change from baseline to end point by 1-way analysis of variance with an effect for treatment (continued same-dose versus low-dose). Safety analyses were conducted separately for each of the groups, with treatment-emergent adverse events summarized by treatment group. Baseline for all treatment-emergent adverse events was defined as the last visit before receiving atomoxetine. Analyses presented are based on all available data. However, because of missing values on some scales, the number of patients included within the different analyses varies slightly.
|
RESULTS
|
|---|
Of 459 patients who received atomoxetine acutely, 229 (51%) met response criteria and were assigned randomly to continue atomoxetine at their final acute dose (mean [SD] dose: 1.43 [0.28] mg/kg per day, n = 116) or to a low dose of 0.5 mg/kg per day (n = 113).
Patient demographics and characteristics (Table 1), as well as baseline symptom measures (Table 2) were similar for patients in both treatment groups. At randomization (ie, after acute treatment), symptom severity was low and similar in both groups of patients (mean [SD]: continued same-dose ADHD RS total score, 15.1 [7.7]; low-dose, 14.0 [7.2]). As shown in Fig 1, at end point, relapse rates did not differ between patients in the groups (continued same-dose group: 3 [2.6%] of 116; low-dose group: 3 [2.7%] of 113; P = .924). The proportion of patients who did not relapse but who no longer met response criterion (40% decrease in symptom severity from study entry) at end point was also similar between groups (continued same-dose group: 20 [17.4%] of 115; low-dose: 25 [22.9%] of 109; P = .321).
There was no difference in mean change (worsening) in ADHD RS total scores (mean [SD] change as a function of treatment group: continued same-dose group, 1.1 [10.8]; low-dose group, 3.1 [10.4]; P = .237). Differences also were not observed in analyses of change on the inattentive and hyperactivity/impulsivity subscales of the ADHD RS (Table 2) or CGI-ADHD-S scores (mean [SD] change in CGI-ADHD-S: continued same-dose group, –0.16 [1.39]; low-dose group, 0.15 [1.36]; P = .078). There also was no group difference in mean change on the CHQ psychosocial summary score (mean [SD] change: continued same-dose group, 1.1 [10.6]; low-dose group, –0.52 [11.8]; P = .205). However, the patients in the low-dose group did have a significantly lower score (worsening) than those in the continued same-dose group on the Role Emotion/Behavior subscale (mean change [SD]: continued same-dose group, 5.75 [29.9]; low-dose group, –1.13 [33.9]; P = .017). For the sake of brevity, data from the remaining CHQ subscales are not presented, and only the CHQ psychosocial summary score is presented in Table 2.
Adverse events are summarized in Table 3. The only statistically significant difference in adverse events between the 2 groups was in reports of affective lability, which was higher in the patients in the low-dose group (affect liability N [%]: continued same-dose group: 0 [0%] of 116; low-dose group: 5 [4.5%] of 112; P = .027). Discontinuations because of an adverse event did not differ between groups (discontinuations N [%]; continued same-dose group: 4 [3.4%]; low-dose group: 3 [2.7%]; P = 1.00). All reasons for patient discontinuation are summarized in Table 4.
Vital signs, weight, and electrocardiographic results are presented in Table 5. There were no statistically significant differences between groups in changes in blood pressure or QTc, but at end point, increases in heart rate (compared with when atomoxetine was started) were higher in patients in the continued same-dose group than in the low-dose group (heart rate, mean [SD]: continued same-dose group, 9.0 [12.5]; low-dose group, 5.2 [13.9]; P = .013). Increases in weight over the course of the trial were greater for patients in the low-dose group than in the continued same-dose group (weight, mean [SD]: continued same-dose group, 0.6 [3.1]; low-dose group, 2.3 [3.0]; P < .001).
Finally, to assess whether the patients originally assigned to atomoxetine, OROS methylphenidate, or placebo in the original comparator study had different outcomes, in this low-dose comparison study, an analysis assessing the effect of randomization timing as a covariate was performed for the last-observation-carried-forward analysis of the primary measure. No statistically significant interactions between randomization time and treatment were seen.
|
DISCUSSION
|
|---|
The dose-response curve for acute treatment with atomoxetine was characterized1 and data have shown that in patients who respond to an initial acute trial of atomoxetine, continued drug treatment is superior to placebo in preventing relapse.2 To date, however, dose-response relationships during chronic treatment have not been studied in a prospective, controlled trial. To our knowledge, this is one of the first studies of a psychiatric treatment to prospectively study dose-response during maintenance treatment using a randomized design similar to the withdrawal designs typically used to study relapse prevention. The issue of whether lower dose treatment during maintenance is of interest as potentially providing a means for guiding the management of adverse events, as well as examining the possibility of reducing drug exposure during long-term treatment.
The results shown here do not demonstrate a statistically significant difference in treatment response between the patients in the continued same-dose and low-dose groups and suggest that many patients who have been robust responders to atomoxetine are likely to maintain satisfactory symptom control, even if the dose is reduced during maintenance treatment. However, the interpretation of these data is limited by several factors. First, and perhaps most important, is that the study was not adequately powered to detect a difference between groups as small as the one observed. We cannot, therefore, rule out the possibility that the modest numerical differences in outcome (ie, slightly higher scores on the ADHD RS scale in the patients in the low-dose group) in this study would have achieved statistical significance had the sample been larger. Thus, although the data provide evidence that switching to a lower dose may be an effective maintenance strategy, they do not establish equivalence between the 2 doses. A second factor that also could have impacted the results is the absence of a placebo arm. Patients and physicians knew during the randomized continuation phase that everyone would receive atomoxetine, and expectation biases related to being on active medication could have worked against reporting an increase in symptoms in the low-dose arm. For these reasons, although these data suggest that, overall, treatment gains can be maintained at lower doses during continuation treatment, they do not provide definitive evidence that the dose of atomoxetine should be lowered during the maintenance phase of treatment.
Even with the caveats noted above, these data suggest that the dose-response curve for atomoxetine during maintenance treatment is not as steep as that associated with acute treatment. We cannot fully account for this difference but speculate that it may be related to a combination of factors. It could be that changes that occur during successful treatment are in some way internalized and become self-reinforcing, perhaps through social learning and positive feedback or new skills acquisition. If this were true, it would minimize any potential differences because of medication dose. It also could be that the amount of drug required to induce a change in baseline pathophysiology is less than that required to maintain changes, although admittedly we have no preclinical or in vivo evidence to support this hypothesis. Finally, it could be that subjects in this study who were previously robust responders to atomoxetine may have been sensitive enough to this medication to permit treatment at lower doses. Whether this strategy would hold for patients who have a good response but not one that approximates normalization (because the inclusionary criterion for being a responder and entering this trial was a 40% decline in ADHD RS rating with treatment) cannot be determined from these data.
One of the key rationales for reducing dose during chronic treatment is as a means of managing adverse events. Interestingly, in this study, tolerability did not seem to be markedly different between patients in the 2 groups. Nevertheless, several parameters including heart rate and weight change did differ between the patients in the 2 groups in ways consistent with a dose-related noradrenergic effect. However, the absence of differences in tolerability may, to some degree, be artifacts of the fact that all patients in this study had been treated with atomoxetine for 8 weeks before randomization. This is an important consideration, because adverse events are generally experienced early in treatment and tend to diminish over time.10
|
CONCLUSIONS
|
|---|
These data suggest that during long-term treatment, it may be possible to treat responders to atomoxetine with a lower dose than is required to achieve acute response. This may be an important strategy for those patients who respond very well to atomoxetine but continue to experience troublesome adverse effects that do not abate over the course of treatment.
|
ACKNOWLEDGMENTS
|
|---|
This study was supported by Eli Lilly and Company.
The Atomoxetine Low-dose Study Group consists of Charles E. Bailey, MD; Charles D. Casat, MD; David W. Dunn, MD; L. Matthew Frank, MD; Robert B. Lehman, MD; Peter D. Londborg, MD; Raun Melmed, MD; Humberto Quintana, MD; R. Bart Sangal, MD; Shannon K. Robinson, MD; Elias H. Sarkis, MD
|
FOOTNOTES
|
|---|
Accepted Jun 15, 2006.
Address correspondence to Jeffrey H. Newcorn, MD, Mount Sinai Medical Center, Department of Psychiatry, One Gustave L Levy Place, Box 1230, New York, NY 10029. E-mail: jeffrey.newcorn{at}mssm.edu
Financial Disclosure: Dr Newcorn is a recipient of grants for research support from Eli Lilly, McNeil, and Shire. He is a consultant and/or advisor for Eli Lilly, Novartis, Cephalon, McNeil, Novartis, and Shire, as well as a speaker for Eli Lilly, McNeil, and Novartis. Dr Kratochvil is a recipient of grants for research support from Eli Lilly and Cephalon. He is a consultant for Eli Lilly, Organon, AstraZeneca, and Pfizer and is on the speaker roster for Eli Lilly. Drs Allen, Michelson, Moore, and Ruff are employees and shareholders of Eli Lilly and Company.
|
REFERENCES
|
|---|
- Michelson D, Faries D, Wernicke J, et al; Atomoxetine ADHD Study Group. Atomoxetine in the treatment of children and adolescents with attention-deficit/hyperactivity disorder: a randomized, placebo-controlled, dose-response study.
Pediatrics. 2001;108
(5). Available at: www.pediatrics.org/cgi/content/full/108/5/e83
- Michelson D, Buitelaar J, Danckaerts MJ, et al. Relapse prevention in pediatric patients with ADHD treated with atomoxetine: a randomized, double-blind, placebo-controlled study.
J Am Acad Child Adolesc Psychiatry. 2004;43
:896
–904[CrossRef][ISI][Medline]
- Michelson D. Results from a double-blind study of atomoxetine, OROS methylphenidate, and placebo. Paper presented at: 51st Annual Meeting of the American Academy of Child and Adolescent Psychiatry; October 19–24, 2004; Washington, DC
- American Psychiatric Association.
Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Text rev. Washington, DC: American Psychiatric Association; 2000
- Kaufman J, Birmaher B, Brent D, Rao U, Ryan N.
Kiddie-SADS-Present and Lifetime Version (K-SADS-PL). Pittsburgh, PA: University of Pittsburgh Press; 1996
- DuPaul GJ, Power TJ, Anastopoulos AD, Reid R.
ADHD Rating Scale-IV: Checklists, Norms, and Clinical Interpretations. New York, NY: Guilford Press; 1998
- Faries DE, Yalcin I, Harder D, Heiligenstein JH. Validation of the ADHD Rating Scale as a clinician administered and scored instrument.
J Atten Disord. 2001;5
:39
–47
- National Institute of Mental Health. Clinical global Impressions.
Psychopharmacol Bull. 1985;21
:839
–843[ISI]
- Landgraf JM, Abetz L, Ware JE.
The CHQ User’s Manual. 1st ed. Boston, MA: The Health Institute, New England Medical Center; 1996
- Wernicke JF, Kratochvil CJ. Safety profile of atomoxetine in the treatment of children and adolescents with ADHD.
J Clin Psychiatry. 2002;63
:50
–55[ISI][Medline]
PEDIATRICS (ISSN 1098-4275). ©2006 by the American Academy of Pediatrics
TOP
ABSTRACT
METHODS
