Adaptive phase I study of OROS methylphenidate treatment of attention deficit hyperactivity disorder with epilepsy

Abstract

Objective

The goal of this study was to pilot a randomized controlled trial of OROS methylphenidate (OROS-MPH) to treat attention deficit hyperactivity disorder (ADHD) plus epilepsy.

Methods

Thirty-three patients, 6–18 years of age, taking antiepileptic drugs and with a last seizure 1–60 months prior were assigned to a maximum daily dose of 18, 36, or 54 mg of OROS-MPH in a double-blind placebo-controlled crossover trial.

Results

There were no serious adverse events and no carryover effects in the crossover trial. OROS-MPH reduced ADHD symptoms more than did placebo treatment. There were too few seizures during the active (5) and placebo arms (3) to confidently assess seizure risk; however, considering exposure time, we observed an increased daily risk of seizures with increasing dose of OROS-MPH, suggesting that potential safety concerns require further study.

Conclusion

A larger study to assess the effect of OROS-MPH on seizure risk is needed. A crossover design including subjects with frequent seizures could maximize power and address high patient heterogeneity and recruitment difficulties.

Introduction

Epilepsy is a highly prevalent neurological disorder estimated to affect 0.5% of children in the developed world [1], [2], [3], [4], [5]. Attention deficit hyperactivity disorder (ADHD) affects 12–39% of patients with epilepsy and is a major source of additional impairment in this population [6], [7], [8], [9]. The Physician's Desk Reference contains warnings stating that stimulants can lower the threshold for having a seizure and that in the face of seizures, these agents should be discontinued [10], yet the evidence on which this warning is based is unclear. Patients with epilepsy were excluded from the clinical trials that established the safety and efficacy of stimulant medications, including methylphenidate (MPH), for the treatment of ADHD [11], [12], [13]. Consequently, most patients who have ADHD comorbid with epilepsy are not treated for ADHD because of concerns that stimulants might worsen seizures [7].

Studies examining the potential risk of seizures from stimulants in patients with epilepsy are relatively few and inconclusive. Although there are a limited number of case series of amphetamine treatment for symptoms of ADHD in children with epilepsy [14], [15], [16], we have found no prospective or controlled trials of amphetamine compounds in this population. To our knowledge, only immediate-release MPH has been studied prospectively, with three studies reported. In the only double-blind placebo-controlled study, 10 children, seizure free at baseline were given MPH 0.6 mg/kg/day in a crossover design, with no seizures occurring on active MPH or on placebo [17]. In another study, 30 children were observed for 8 weeks and then given 0.3 mg/kg/day MPH for 8 weeks. Twenty-five of the 30 patients were seizure free at baseline and remained so; 5 patients with an average of 1.8 seizures/week during the observation period experienced an average of 3 seizures/week during MPH treatment [18]. This difference in seizure rate was not statistically significant, possibly because of the low power. In both studies, 70% of the children significantly improved on MPH. Finally, Gucuyener et al. followed 57 children with epilepsy for 1 year while they took MPH open label. No increase in seizure frequency was detected [19]. Thus, a total of 97 children with epilepsy have been studied prospectively while taking MPH. Their combined average seizure frequency was 0.65/week prior to and 0.93/week during MPH treatment [16].

This literature provides some evidence of the short-term efficacy of short-acting MPH in decreasing ADHD symptoms in children with epilepsy. Because of the low baseline seizure rate and small number of patients studied, however, the conclusion that MPH does not lower the seizure threshold to a clinically significant degree remains problematic. Given the prevalence and clinical significance of this dilemma, there is a critical need for prospective, double-blind, placebo-controlled trials of stimulants in children with comorbid epilepsy and ADHD to develop an evidence base that can guide clinical practice.

Implementing such trials, however, is challenging precisely because of the dilemma itself: trials need to be designed in such a way that they maximize the possibility of detecting benefit while, importantly, not putting children at risk. Moreover, heterogeneity among the epilepsy population, in terms of type and frequency of seizures, further complicates the design of studies and the interpretation of findings. The primary aim of this study, therefore, was to provide pilot data to evaluate the feasibility of clinical trial methods that might be used for a larger study of OROS-MPH in children with comorbid epilepsy and ADHD while limiting risks to participants. To this end, the feasibility of recruiting for a double-blind placebo controlled crossover trial and the behavior of efficacy and safety measures in the crossover trial were explored. The secondary aim was to obtain preliminary estimates of the efficacy and safety of OROS-MPH to inform power calculations for a larger study.

We chose to examine OROS-MPH because of its popularity and ability to provide consistent MPH plasma levels daily [20], [21], [22]. Patients with epilepsy had been excluded from all previous OROS-MPH clinical trials. Also, the sustained-release preparation raises a theoretical concern that any adverse effect on seizures might be prolonged relative to immediate-release MPH. Thus, prior to exposing study patients with epilepsy to the full range of OROS-MPH doses, the Children's Hospital Boston institutional review board (IRB) required sufficient exposure to lower doses of OROS-MPH without significant toxicity before any patient was exposed to higher doses. For this reason, the adaptive phase I dosing escalation strategy was incorporated into the pilot study. An additional objective of the phase I design is to look for any very large safety problem in a small number of patients so as to establish a dose range of acceptable risk for subsequent larger clinical trials [23]. Adaptive phase I designs are commonly used in initial oncology trials where dose-limiting toxic effects are expected, and finding the maximum acceptable dose within the potential therapeutic dose range is required before undertaking efficacy trials [23], [24]. We hypothesized that we would have no serious adverse effects or increase in seizure risk at any of the tested doses so that the maximum dose, the lesser of 54 mg/day or 2 mg/kg/day, could be designated the target dose for subsequent studies.