ARTICLE |
a Pediatric Epilepsy Center, Division of Pediatric and Developmental Neurology, Departments of Neurology and Pediatrics, Washington University in St Louis School of Medicine, St Louis, Missouri
b GlaxoSmithKline, Research Triangle Park, North Carolina
| ABSTRACT |
|---|
|
|
|---|
2 years of age; we report the data from children and adolescents 2 to 20 years of age from this randomized clinical trial. This is the first published analysis of data from a randomized, double-blind, controlled clinical trial of primary generalized tonic-clonic seizures focusing on children and adolescents.
PATIENTS AND METHODS. We randomly assigned (1:1) 117 patients, aged 2 to 55 years, with primary generalized tonic-clonic seizures inadequately controlled on 1 to 2 current antiepileptic drugs and with evidence of primary generalized tonic-clonic seizures on electroencephalogram and no historical or electroencephalogram evidence of partial seizures to either lamotrigine or placebo in a double-blind parallel group clinical trial from 2001 through 2004. We analyzed the subgroup of children and adolescents, aged 2 to 20 years (n = 45), from this randomized clinical trial. Patients having
3 primary generalized tonic-clonic seizures over an 8-week baseline were randomly assigned (1:1) to receive either lamotrigine or placebo. The treatment period consisted of an escalation phase (12 weeks for patients 212 years; 7 weeks for patients >12 years) and a maintenance phase (12 weeks). The study had 4 phases: screening phase, baseline phase, escalation phase, and maintenance phase. During the screening phase, baseline medical examinations and seizure type and seizure frequency assessments were performed. During the 8-week baseline phase, the number and dosages of concomitant antiepileptic drugs were maintained while seizure frequency was assessed. The assessment of primary generalized tonic-clonic seizure frequency was determined during the 8-week baseline phase. Patients eligible for random assignment experienced
3 primary generalized tonic-clonic seizures during the baseline phase and
1 primary generalized tonic-clonic seizure in the 8 weeks before the baseline phase. Lamotrigine was introduced and titrated using a schedule based on the patients' age and concurrent antiepileptic drug regimen. During the escalation phase, the number and doses of concomitant antiepileptic drugs were not changed. The escalation phase was followed by a 12-week maintenance phase during which time the lamotrigine dose was maintained at a specific dose defined by the patients' age and concomitant antiepileptic drugs, whereas the doses of concurrent antiepileptic drugs were maintained at a constant dose. Concurrent antiepileptic drugs could not be discontinued or added during the maintenance phase. The primary efficacy end point measure was the median reduction in the frequency of primary generalized tonic-clonic seizures from baseline; seizure counts were recorded prospectively in standardized daily seizure diaries. Other efficacy end point data for analysis were as follows: the median seizure counts, the median percentage change from the baseline phase in average monthly seizure frequency for other generalized seizure types, and the percentage of patients with a reduction of
25%,
50%,
75%, or 100% in frequencies of primary generalized tonic-clonic seizures and all generalized seizures during the escalation phase and/or maintenance phase relative to the baseline phase. Accurate counts of absence seizure frequency require electroencephalogram-video monitoring; absence seizure frequency was not an outcome measure for this analysis.
RESULTS. Forty-five (21 lamotrigine and 24 placebo) patients 2 to 19 years of age were randomly assigned and received study drug. Eight patients (3 lamotrigine and 5 placebo) had a combination of clinical (myoclonus and/or absence seizures) and electroencephalogram findings that were consistent with juvenile myoclonic epilepsy. Among the 45 children randomly assigned, 74% had generalized spike, polyspike, and/or generalized spike and wave discharges on routine electroencephalogram recordings; the remaining 26% of children had no electroencephalogram findings suggestive of partial epilepsy and a clear history consistent with primary generalized tonic-clonic seizures. Electroencephalogram findings were not significantly different between the lamotrigine and the placebo treatment groups. The median percentage decrease from baseline in primary generalized tonic-clonic seizures during the entire treatment period was 77% in the lamotrigine group and 40% in the placebo group (P = .044). The median primary generalized tonic-clonic seizure counts per month were 0.7 in the lamotrigine group and 3.6 in the placebo group during escalation (P = .008), 0.3 in the lamotrigine group and 2.0 in the placebo group during maintenance (P = .005), and 0.4 in the lamotrigine group and 2.5 in the placebo group during the entire treatment period (P = .007). Trends were noted during escalation and maintenance with a median percentage decrease in primary generalized tonic-clonic seizures during escalation of 72% in the lamotrigine group and 30% in the placebo group (P = .059), and 83% in the lamotrigine group and 42% in the placebo group during maintenance (P = .058). During the maintenance phase, 48% of lamotrigine patients were seizure free compared with 17% treated with placebo (P = .051). One patient from each treatment group discontinued from the study because of an adverse event; 1 patient who received lamotrigine experienced "disorientation"; and 1 patient who received placebo had a convulsion with apnea. No rashes occurred among patients taking lamotrigine or placebo. No patient experienced worsening of the intensity or frequency of myoclonus.
CONCLUSIONS. Adjunctive lamotrigine therapy seems effective in controlling primary generalized tonic-clonic seizures among patients 2 to 20 years of age.
Key Words: generalized tonic-clonic seizures childhood epilepsy clinical trial lamotrigine juvenile myoclonic epilepsy
Abbreviations: PGTCprimary generalized tonic-clonic RCTrandomized clinical trial EEGelectroencephalogram AEDantiepileptic drug
Primary generalized tonic-clonic (PGTC) seizures are relatively more common among children than among adults.1 PGTC seizures are associated with idiopathic generalized epilepsy and several generalized epilepsy syndromes including juvenile myoclonic epilepsy and juvenile absence epilepsy. Onset of PGTC seizures is age related and typically starts in older children, adolescents, and young adults commensurate with the onset of idiopathic generalized epilepsies.2 PGTC seizures are associated with increased risk of injury3 and death.4 Therefore, effective control of PGTC seizures is necessary to reduce epilepsy-related morbidity and mortality. This is the first published analysis of data from a randomized, double-blind, controlled clinical trial of PGTC seizures focusing on children and adolescents.
Lamotrigine has demonstrated efficacy from published randomized clinical trials (RCTs) for childhood partial seizures,5 absence seizures,6,7 and for the generalized seizures associated with Lennox-Gastaut syndrome.8 Lamotrigine has been reported in open-label studies9,10 and in double-blind, active-comparator RCTs to be effective in patients with newly diagnosed, previously untreated primary or secondary generalized seizures.1113 Most of these studies did not differentiate between adults and children in the analysis or did not specifically exclude patients whose generalized tonic-clonic seizures were actually partial seizures with secondary generalization.
The efficacy and tolerability of adjunctive lamotrigine in patients with generalized seizures were investigated in a multicenter, double-blind, placebo-controlled, add-on RCT that included both children and adults from the United States and from Latin America.14 Patients who experienced partial seizures or who had electroencephalogram (EEG) evidence of partial onset seizures were excluded. The relative paucity of published RCTs of childhood PGTC seizures led us to do a posthoc analysis to evaluate the efficacy and safety data among children and adolescents (aged 220 years) from a double-blind, placebo-controlled RCT of lamotrigine in the treatment of PGTC seizures among children and adults.
| METHODS |
|---|
|
|
|---|
3 PGTC seizures over an 8-week baseline were randomly assigned (1:1) to receive either lamotrigine or placebo. The treatment period consisted of an escalation phase (12 weeks for patients 212 years and 7 weeks for patients >12 years) and a maintenance phase (12 weeks; see Fig 1).
|
Patients were also excluded from the study if they were breastfeeding, pregnant, attempting to become pregnant, capable of bearing children without using acceptable forms of contraceptives, following the ketogenic diet, had a coexisting disease that the investigators believed would interfere with the completion of the study, abused alcohol or other illicit drugs, received chronic treatment with a medication that could interfere with seizure treatment, and/or planned vagal nerve stimulator implantation or epilepsy surgery during the study period. Informed consent was obtained, and the study protocol was approved by an institutional review board for each of the 38 sites that enrolled adults and/or children.
The study had 4 phases: screening phase, baseline phase, escalation phase, and maintenance phase (Fig 1). During the screening phase, baseline medical examinations and seizure type and seizure frequency assessments were performed, allowing for determination of the patients' eligibility for the study. During the 8-week baseline phase, the number and dosages of concomitant AEDs were maintained while seizure frequency was assessed. The assessment of PGTC seizures was prospectively determined during the 8-week baseline phase. For patients with reliable documentation of prestudy seizure type and seizure frequency, the baseline assessment of PGTC seizures could be obtained from either a historical prestudy baseline or a combination of the historical and prospective baselines (total: 8 weeks) before random assignment. Patients eligible for random assignment experienced
3 PGTC seizures during the baseline phase and
1 PGTC seizure in the 8 weeks before the baseline phase. Patients meeting exclusion-inclusion criteria were randomly assigned 1:1 to either lamotrigine or placebo during the escalation and maintenance phases.
Lamotrigine was introduced and titrated using a schedule based on the patients' age and concurrent AED regimen (Table 1). During the escalation phase, which lasted 7 weeks for patients >12 to 20 years of age and 12 weeks for patients 2 to 12 years of age, the number and doses of concomitant AEDs were not changed. The escalation phase was followed by a 12-week maintenance phase during which time the lamotrigine dose was maintained at a specific dose defined by the patients' age and concomitant AEDs, while the doses of concurrent AEDs were maintained at a constant dose. Concurrent AEDs could not be discontinued or added during the maintenance phase. After completion of the maintenance phase, patients were offered the opportunity to receive open-label lamotrigine for up to 1 year during a continuation phase of the study. Adjustment of concomitant AEDs was allowed during the continuation phase.
|
72 hours) was permitted
3 times during both the escalation and the maintenance phases to control clinically significant increases in seizure frequency, duration, and/or intensity. Patients who discontinued the study medication during the study or at the end of the study were advised to reduce the dose in a gradual stepwise fashion over a minimum of 2 weeks, unless safety concerns (eg, rash) required a more rapid dose reduction.
Outcome Measure
Patients and/or their caregivers (parents or guardians) recorded PGTC seizure counts prospectively in standardized daily seizure diaries throughout the study. These seizure diaries were reviewed by the investigators at clinic visits during each study phase. Accurate counts of absence seizure frequency require EEG-video monitoring; absence seizure frequency was not an outcome measure for this analysis. Adverse events were recorded regardless of cause and defined as any untoward medical occurrences reported by patients or noted by investigators at any time during the study. Investigators recorded whether they thought the adverse event was likely caused by the study medication. Weight, vital signs, and physical examination findings were recorded at each clinic visit. Serum lamotrigine concentrations were measured by standard liquid chromatography/mass spectrometry at the screening visit and at treatment weeks 7, 11, and 19 for patients >12 to 20 years of age and at treatment weeks 7, 11, and 24 for patients 2 to 12 years of age.
Data Analysis
The median percentage of change in average monthly PGTC seizure frequency from the baseline phase was the primary efficacy end point. Other efficacy end point data for analysis were as follows: the median seizure counts, the median percentage of change from the baseline phase in average monthly seizure frequency for other generalized seizure types, and the percentage of patients with a reduction of
25%,
50%,
75%, or 100% in frequencies of PGTC seizures and all generalized seizures during the escalation phase and/or maintenance phase relative to the baseline phase. In addition, the number of adverse events that was reported during the double-blind treatment considered by the investigator to be possibly, probably, or definitely drug related was also summarized. Seizure frequency end points were calculated for both the escalation phase and the maintenance phases separately and for these phases combined.
Differences between the placebo group and the lamotrigine group were determined using a 2-way analysis of variance based on ranks with treatment group and age categories as predictors of percent changes in average seizure frequency, with Fisher's exact test for percentages of patients with specific percent reductions in seizure frequency, and with a 2-way analysis of variance based on ranks with treatment group and age category as predictors for seizure counts.
Power calculations for the overall clinical trial (adults and children
2 years of age) were performed based on an 80% power to detect a difference of 25% in the median percent reduction in PGTC seizures between baseline and completion of the double-blind treatment phase, with an SD of 45% at a significance level of .05. It was estimated that 150 patients (adults and children
2 years) would need to be enrolled to randomly assign 104 patients required to satisfy the power calculations. The clinical trial was not specifically powered for this posthoc subanalysis of the children and adolescents (220 years).
| RESULTS |
|---|
|
|
|---|
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
This report is a posthoc analysis of the pediatric and adolescent data from a larger RCT that included both adults and children14 and, as such, has limitations. Specifically, this RCT was not powered for this subgroup analysis. However, the magnitude of the difference between the lamotrigine and the placebo groups is very similar between the overall clinical trial and this subgroup analysis.14 The significant percentage reduction in PGTC seizures during the escalation plus maintenance phase among patients who received lamotrigine is not likely due to chance; we believe that it is unlikely that a larger sample size would have altered our overall conclusions. Despite the underpowered nature of this subgroup analysis, this is the first report of pediatric-only data from a placebo-controlled clinical trial of PGTC seizures that demonstrates efficacy.
The relatively high placebo response rate in the subgroup of children and adolescents was very similar to the placebo response rate in the overall clinical trial.14 Therefore, the high placebo response in our subanalysis is not likely because of inadequate sample size. High placebo response rates have been reported in other clinical trials of AEDs.8 Placebo-controlled trials in children have previously corrected misperceptions of drug efficacy from open-label trials. For example, in the 1980s, open-label studies of cinromide showed impressive reduction in seizure frequency among children with multiple seizure types, including generalized seizures. Yet the cinromide placebo-controlled clinical trial demonstrated both a high placebo response rate and no evidence of cinromide efficacy19; cinromide development as an AED was terminated. Pediatricians should view the results from open-label reports of treatment effectiveness with a degree of skepticism and, whenever possible, base clinical treatment decisions on the results of RCTs.
This is the first RCT of PGTC seizures that has used prospective EEG data with clinical data to exclude patients with partial seizures. A study of topiramate in the treatment of PGTC seizures required that historical EEG data not show evidence of partial seizures, but prospective EEG data were not required for entry into the study.20 In a study of gabapentin in PGTC seizures, the only other randomized, controlled clinical trial among patients with PGTC seizures, efficacy was not demonstrated, and patients with partial seizures were not excluded.21 Failure to properly exclude patients with partial seizures in studies of PGTC seizures may impact the efficacy data, especially when the study drug has known efficacy in partial seizures as well.
The majority of children with PGTC seizures respond to the first AED prescribed.22 By choosing children who had failed to respond to their first or second AED, this study selected for children whose seizures were more refractory to therapy than the typical child in the general population with PGTC seizures. The majority of the children who enrolled in this study failed previous treatment with valproate, and all of the children enrolled in this study failed previous treatment with marketed AEDs that were believed to be effective in the treatment of PGTC seizures.
There are very few proven options for treating PGTC seizures among children. Valproate, which also has an intravenous formulation, is commonly used for treatment of PGTC seizures. Valproate is associated with rare but serious idiosyncratic adverse events (ie, hepatic failure, aplastic anemia, and pancreatitis) and more common adverse events (eg, weight gain and polycystic ovarian syndrome) that make its use problematic for some children and adolescents.2325 Phenytoin and carbamazepine are frequently used to treat "tonic-clonic" seizures, yet these drugs are known to exacerbate PGTC seizures among some patients.1618 Topiramate seems effective in the treatment of tonic-clonic seizures, but cognitive slowing associated with topiramate has limited its use as well.26
Lamotrigine has been associated with an increased risk of potentially serious rash (including Stevens-Johnson syndrome and toxic epidermal necrolysis). The risk of lamotrigine-associated rash is increased by using rapid dose-escalation schedules, and the risk is reduced by the dose-escalation schedules recommended in the lamotrigine (Lamictal [GlaxoSmithKline, Research Triangle Park, NC]) package insert.27,28 There are rare case reports of lamotrigine-associated exacerbation of myoclonus,29 an adverse event that was not seen among the patients in this study. Unlike valproate, lamotrigine is not associated with significant weight gain.24
Lamotrigine seems effective in the treatment of PGTC seizures and has demonstrated a good safety profile among children and adolescents. Pediatricians continue to have limited options for treating PGTC seizures in children and adolescents. New drug development for primary generalized seizures is needed, as are additional clinical trials of PGTC seizures in children. The population of children with PGTC seizures is relatively heterogeneous, as demonstrated in our study. Future clinical trials of generalized epilepsy should ideally be performed among more homogeneous patient populations with specific epilepsy syndromes.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Address correspondence to Edwin Trevathan, MD, MPH, Division of Pediatric and Developmental Neurology, Washington University School of Medicine, Campus Box 8111, 660 S Euclid Ave, St Louis, MO 63110-1093. E-mail: trevathan{at}wustl.edu
Financial Disclosure: This study was funded by GlaxoSmithKline, the maker of Lamictal (lamotrigine). Dr Trevathan has served as a paid consultant to GlaxoSmithKline but did not receive compensation for writing this article or for assisting with the analysis of these data. Ms Kerls, Ms Hammer, Mr Voung, and Dr Messenheimer are full-time employees of GlaxoSmithKline.
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||