Rufinamide for the treatment of epileptic spasms

Methods We retrospectively reviewed patients treated with rufinamide for epileptic spasms from January 2009 to March 2010. Age, presence of hypsarrhythmia, change in seizure frequency following rufinamide initiation, and side effects were assessed. Patients who had a ≥ 50% reduction in spasm frequency were considered responders. Results Of all 107 children treated with rufinamide during the study period, 38 (36%) had epileptic spasms. Median patient age was 7 years (range: 17 months to 23). One patient had hypsarrhythmia at the time of treatment with rufinamide, and 9 other patients had a history of hypsarrhythmia. Median starting dose of rufinamide was 9 mg/kg/day (range: 2–18) and median final treatment dose was 39 mg/kg/day (range: 8–92). All patients were receiving concurrent antiepileptic drug therapy, with the median number of antiepileptic drugs being 3 (range: 2–6). Median duration of follow-up since starting rufinamide was 171 days (range: 10–408). Responder rate was 53%. Median reduction in spasm frequency was 50% (interquartile range = –56 to 85%, P < 0.05). Two patients (5%) achieved a > 99% reduction in spasms. Rufinamide was discontinued in 7 of 38 patients (18%) because of lack of efficacy, worsening seizures, or other side effects. Minor side effects were reported in 14 of 38 patients (37%). Conclusions Rufinamide appears to be a well-tolerated and efficacious adjunctive therapeutic option for children with epileptic spasms. A prospective study is warranted to validate our observations. Research Highlights ► Rufinamide elicited a 53% responder rate in patients with epileptic spasms. ► Median reduction in spasms was 50%. ► Rufinamide was well tolerated. ► Rufinamide is an effective adjunctive therapeutic option for epileptic spasms. Keywords Epileptic spasms Rufinamide West syndrome Epilepsy 1 Introduction The revised International League Against Epilepsy (ILAE) classification recognizes epileptic spasms as a separate seizure entity [1] . Associated EEG patterns may be generalized, focal, or of unclear onset [1] . Epileptic spasms include both infantile spasms and spasms in older patients [1–4] . Treatment options for epileptic spasms are limited [5–8] . Effective treatment of spasms remains an important unmet medical need, as long-term developmental and cognitive outcome of patients with spasms is likely improved with effective control of spasms [6,7,9–12] . Given its favorable safety profile and effectiveness for other difficult-to-control seizure disorders [13–22] , rufinamide may be an additional treatment option for refractory epileptic spasms. Rufinamide is a relatively new antiepileptic drug (AED), approved by the Food and Drug Administration (FDA) in November 2008 for use as adjunctive therapy of seizures associated with Lennox–Gastaut syndrome in children older than age 4 and in adults [17,18,23] . It has good oral absorption, is not metabolized by the cytochrome P450 enzyme system, and has a favorable side effect profile [14,20,22–25] . Studies have shown that rufinamide is effective for refractory partial seizures in adolescent and adults, in Lennox–Gastaut syndrome, and, most recently, in childhood epileptic encephalopathies other than Lennox–Gastaut syndrome [13–19,21] . Preliminary data on the efficacy of rufinamide in epileptic spasms have also recently been published by our center [26] . Accordingly, the goal of the present study was to more closely assess the effectiveness and safety of rufinamide for treatment of epileptic spasms in children of all ages. 2 Methods After approval by the institutional review board, we retrospectively identified patients with epileptic spasms treated with rufinamide from January 2009 to March 2010 at Children's Hospital Boston. Based on the 2010 ILAE classification [1] and other supporting literature [3,4,27,28] , we defined epileptic spasms clinically as brief contractions of axial and/or proximal limb muscles that may include flexion, extension, or mixed extension–flexion. They are more sustained than a myoclonic movement but less sustained than a tonic seizure, and typically last 1 to 15 seconds [3,27,28] . Spasms often but not always occur in clusters [3,27,28] . We also looked for a confirmatory EEG signature of a diffuse slow wave followed by a fast rhythm or brief polyspike burst of low amplitude and electrodecrement [3,4,28] . Not all patients had events captured on EEG, but those who did had consistent EEG patterns. We did not exclude patients for lack of a confirmatory EEG signature of spasms. Charts were reviewed for baseline information including age at time of study, age at seizure onset, gender, seizure types, epilepsy syndrome, MRI results, EEG findings, etiology, and number and types of antiepileptic medications used at the time of review and in the past. Additional collected information included rufinamide dosing and titration schedule, side effects, discontinuation rates and reason for discontinuation if applicable, spasm frequency before and after treatment with rufinamide, and duration of treatment with final dose of rufinamide at the time of last documented clinic visit. Spasm frequency was collected by retrospective review of neurology clinic notes and discharge summaries and from EEG monitoring reports when available. Spasm frequency was documented as number of individual spasms per day or, in patients with clusters of spasms, as number of clusters per day. In summarizing the clinical characteristics of our study population, we divided etiology of epilepsy into genetic, structural/metabolic, and unknown, using the 2010 ILAE guidelines [1] . We also further subcategorized the patients in an attempt to more precisely characterize the underlying cause of their epilepsy. We adapted the classification scheme to include a subcategory of cases of malformations of cortical development with known genetic etiology. Patients who had a ≥ 50% reduction in spasm frequency were considered responders. We calculated median reduction in spasm frequency as well as interquartile ranges. We also determined how many patients had a > 99% reduction in spasm frequency (seizure freedom). Statistical analysis was conducted for nonparametric measures with the Wilcoxon signed rank test (with a two-tail alternative). Calculations and graphics were performed with SPSS 13 for Macintosh computers (Apple, Cupertino, CA, USA). 3 Results 3.1 Demographics Of 107 children treated with rufinamide from January 2009 to March 2010 at our center, 38 (36%) had drug-resistant spasms. Median patient age was 7 years (range: 17 months to 23 years). The demographic and baseline clinical characteristics of the study patients are listed in Table 1 . In applying the new ILAE etiology classification, we listed several patients in the structural/metabolic group in more than one subcategory. For example, two patients with Aicardi syndrome had multiple structural abnormalities. In addition, one patient had polymicrogyria and heterotopias, one patient had polymicrogyria and schizencephaly, and one patient had heterotopias and a stroke (occipital infarct). In our study population, 18 of 38 patients had a history of infantile spasms with persistence of spasms or recurrence of drug-resistant spasms at an older age, and 20 of 38 had de novo onset of spasms at an older age. Only one patient had hypsarrhythmia at the time of treatment with rufinamide, and < 30% of our study population had hypsarrhythmia at any point. All patients were receiving concurrent AED therapy with a median of 3 (2–6) other AEDs. Other antiepileptic medications with which patients were treated at the time of the study included levetiracetam (23); benzodiazepines including diazepam, lorazepam, clonazepam, and clobazam (20); lamotrigine (11); valproic acid (10); zonisamide (8); topiramate (5); phenobarbital (4); vigabatrin (2); ACTH (1); phenytoin (1); oxcarbazepine (1); felbamate (1); methsuximide(1); gabapentin (1); and primidone (1). 3.2 Dosing and follow-up Median starting dose of rufinamide was 9 mg/kg/day (range: 2–18) and median final treatment dose was 39 mg/kg/day (range: 8–92). Median time to goal dose was 42 days (range: 7–408 days). Median duration of follow-up since starting rufinamide was 171 days (range: 10–408), and median duration of follow-up on final dose of rufinamide was 66 days (range: 0–277). For those with zero days of follow-up on the final dose (one patient), rufinamide dose was increased at the last visit and further follow-up was not yet documented. At the time of review, 31 of 38 patients (82%) were continuing on rufinamide, 2 of 38 (5%) were in the process of weaning off of rufinamide, and 5 of 38 (13%) had discontinued rufinamide. 3.3 Responder rate The responder rate, defined as ≥ 50% reduction in spasms, was 53%. All patients in the study were included in this calculation, including those who had discontinued rufinamide. Median reduction in spasm frequency was 50% (interquartile range: IQR = –56 to 85%, P < 0.05) (Wilcoxon test) ( Fig. 1 ). Median spasm frequency before rufinamide was 5.3 per day (IQR = 2.4–13.1), and after rufinamide, 3 per day (IQR = 1–5.5). Nine patients (24%) achieved ≥ 90% reduction in spasms, and two patients (5%) achieved > 99% reduction in spasms. Duration of follow-up for the two patients with > 99% reduction in spasms was 3 months for one and > 1 year for the other. In Fig. 1 , there are three outliers, children with significantly increased spasm frequency on rufinamide. We could not identify a common pattern in patient clinical characteristics or treatment among these three outliers. The outliers were included in the analysis. Specifically for the group of 10 patients with a history of hypsarrhythmia, the responder rate was 50% and the median reduction in spasm frequency was 42%. 3.4 EEG changes We evaluated, when available, if there were changes in the EEG with respect to interictal activity, ictal activity, or sleep architecture, before and after starting rufinamide. For 11 patients (29%) there was no significant change in the EEG. For 3 patients (8%) there was an improvement in the EEG. In one, there was an improvement in both sleep architecture and interictal activity. In one, there was a mild improvement in background slowing. In the third, there was a significant decrease in seizure frequency on LTM after starting rufinamide, but with interictal activity and sleep architecture unchanged. In three patients (8%), there was worsening of the EEG. One patient had worsening of spike–wave activity and developed burst suppression in sleep. Another patient developed generalized epileptiform activity, but sleep architecture was unchanged. The third patient had increased electrographic seizures on LTM after starting rufinamide. This patient had a 233% increase in spasm frequency on rufinamide. In 21 patients (55%), EEGs before and after starting rufinamide were not available. 3.5 Side effects Rufinamide was discontinued or was being weaned in 7 of 38 patients (15%), in 2 for lack of efficacy, in 1 for worsening seizures, in 1 for lack of efficacy and development of a new seizure type, in 2 for other side effects, and in 1 for lack of efficacy and side effects. Minor side effects were reported in 14 of 38 patients (37%). Side effects included increased seizure frequency at high dose (5), decreased appetite (3), sedation (3), eyes crossing (1), vomiting (1), behavioral changes (1), dizziness (1), falls (2), rash (1), difficulties with daily functions (1), visual blurriness (1), and increased duration of seizures (1). 4 Discussion 4.1 Summary Overall, this retrospective review of the safety and efficacy of rufinamide for epileptic spasms showed a > 50% responder rate and a favorable side effect profile. 4.2 Comparison with available literature Rufinamide is increasingly used in the treatment of a variety of seizure types. A recent postmarketing study from our institution reported that rufinamide is a useful adjunctive treatment option in children with refractory epilepsy of multiple types, with maximal responder rate for tonic/atonic and partial seizures [26] . The median percentage of seizure reduction was 58% for tonic/atonic seizures, and 50% for partial seizures, whereas the responder rate was 49% for tonic/atonic seizures and 47% for partial seizures [26] . Our data indicate that efficacy for treatment of epileptic spasms is similar, with a median reduction in spasm frequency of 50% and a responder rate of 53%. These numbers exceed and expand on our recent study, which showed a median percentage of seizure reduction for infantile spasms of 37.5% and a responder rate of 31% in a small ( n = 13) sample size (included in this report). Previous work, including a double-blind, randomized, placebo-controlled trial, has demonstrated that rufinamide reduces drop attacks in patients with Lennox–Gastaut syndrome [17,18] . Video/EEG analysis of patients with drop attacks has identified four seizure types: flexor spasms, tonic seizures, myoclonic–atonic seizures, and atonic seizures [29,30] . Therefore, the drop attacks that responded to rufinamide in the double-blind study were likely a mixed group of seizure types that included epileptic spasms. Our present study suggests that rufinamide is effective specifically in patients with clinically defined spasms, many with video/EEG confirmation. In addition, our population was diverse and included patients with Lennox–Gastaut syndrome, West syndrome, and a variety of other epilepsy syndromes. 4.3 Rufinamide levels and interactions Serum rufinamide levels were not available. Given the concomitant use of multiple other AEDs in all cases, it is important to consider drug–drug interactions. Rufinamide does have identified drug–drug interactions. For example, valproic acid may increase plasma rufinamide concentrations [1,14,23–25] . In contrast, carbamazepine, vigabatrin, phenytoin, phenobarbital, and primidone may lead to a minor reduction in serum rufinamide levels [1,14,23–25] . Rufinamide may decrease plasma concentrations of carbamazepine and lamotrigine and increase plasma concentrations of phenobarbital [24] . Rufinamide slightly decreases phenytoin clearance, and thus, plasma concentrations may increase, but the effect on plasma levels is difficult to predict because of the nonlinear elimination kinetics of phenytoin [23,24] . Rufinamide does not significantly change the plasma concentration of topiramate or valproic acid [24] . In our study, we could not identify differences in number or type of concomitant AEDs between patients who had side effects from rufinamide and those who did not, or between the responders and the nonresponders. The effect of rufinamide on concomitant AED plasma concentrations is also not known. 4.4 Safety In our current study, side effects other than increased seizure frequency were minor and led to discontinuation of the medication in only 3 of 38 (8%) patients. It is notable that there was an increase in seizure frequency in some patients, including the three outliers with a ≥ 400% increase in seizure frequency. It is difficult to determine, however, if the increased seizure frequency was related to rufinamide treatment or other factors. The recently published study by Coppola et al. on rufinamide in refractory childhood epileptic enphalopathies reported significant increases in seizure frequency in 13% of patients [15] . The side effect profile for rufinamide thus far has been favorable [14,20,22–25,31] . A pooled analysis of seven clinical studies suggested a favorable safety and tolerability profile, with only 9% of patients discontinuing rufinamide because of adverse events [22,25] . The majority of side effects were mild to moderate, most frequently headache, dizziness, fatigue, somnolence, and nausea [22,25] . In the double-blind population, serious adverse events occurred in 7.5% of rufinamide-treatment patients compared with 5.6% of placebo-treated patients, and no deaths were attributed to rufinamide [22,25] . In another study of 45 children and 15 adults, adverse events occurring in > 10% of patients included fatigue, vomiting, and loss of appetite [20] . There were no serious adverse events [20] . Some of the other AEDs used for epileptic spasms include ACTH, vigabatrin, valproic acid, clonazepam, levetiracetam, and topiramate. Many of these AEDs have more concerning side effect profiles. Rufinamide may provide a safe option for treatment of epileptic spasms. 4.5 Limitations Data have to be interpreted in the context of their acquisition and may be limited by the retrospective nature, the setting at a tertiary epilepsy center, and intent-to-treat bias. The retrospective nature of the study limits the amount of information available, which could either falsely increase or decrease the response rate. Quantification of spasm frequency before and after treatment was based on clinical notes and EEG data when available. Seizure frequency in clinical notes is primarily from parent report and, at times, seizure logs. Thus, the quantification is an approximation. In a prospective study, seizure logs and ambulatory EEGs may provide more detailed analysis of seizure frequency. Intervals between pre- and posttreatment assessments were variable, and retrospective data acquisition did not permit the study of the more homogeneous groups that prospective inclusion/exclusion criteria would provide. Also, the rate of dose escalation and final dose varied considerably. The study setting of our tertiary epilepsy center may have lowered the observed responder rate because of the patients with highly refractory seizures referred there. Likewise, the intent-to-treat analysis we used may have also lowered the apparent responder rate as the analysis included patients no longer receiving the medication. Because rufinamide was a relatively new drug at the onset of our study, there may have been a selection bias toward children who had already been treated with several other antiepileptic drugs, reducing the responder rate. The relatively small number of patients in our study did not allow for assessment of which antiepileptic drugs were most efficacious when used in combination with rufinamide. 4.6 Study population and future studies The majority of children in this study are older patients with refractory epileptic spasms, and all of the children have significant developmental delays. The population of patients in our study is representative of the population of older children with spasms described in the literature [2–4] , having persistence of spasms or recurrence of drug-resistant spasms at an older age or de novo onset of spasms at an older age. We had only one patient with West syndrome who had hypsarrhythmia at the time of treatment with rufinamide, and this is why we did not divide the patients into two groups, one with hypsarrhythmia and one without. Less than 30% of our study population had hypsarrhythmia at any point. Older patients with spasms do not typically have hypsarrythmia on the EEG [3] . Rufinamide has not yet been used extensively in the acute period of infantile spasms with hypsarrhythmia. A recent U.S. concensus report on infantile spasms addressed evaluation and treatment of infantile spasms [8] , concluding that ACTH and vigabatrin are the only drugs with proven efficacy to suppress clinical spasms and abolish the hypsarrhythmic EEG. The report emphasized, however, the need for further research to develop new therapies that may improve outcomes. Our data suggest the possibility that rufinamide may be a reasonable second-line agent when either ACTH (the gold standard treatment for infantile spasms) or the recently FDA-approved vigabatrin is not effective, though a prospective study is needed. 5 Conclusion Rufinamide appears to be a well-tolerated and efficacious alternative adjunctive therapeutic option for children with epileptic spasms, resulting in a median reduction in spasm frequency of 50% and a responder rate of 53% in this retrospective study. A prospective study is warranted to validate our observations. Ethical approval This study was approved by the institutional review board at Children's Hospital Boston, MA, USA. Acknowledgment This study was funded in part by an Investigator Initiated Grant from Eisai Pharma, Inc., to S.V.K. and T.L. References [1] A.T. Berg S.F. Berkovic M.J. Brodie Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE Commission on Classification and Terminology, 2005–2009 Epilepsia 51 2010 676 685 [2] P. Camfield C. Camfield A. Lortie H. Darwish Infantile spasms in remission may reemerge as intractable epileptic spasms Epilepsia 44 2003 1592 1595 [3] J. Goldstein J. Slomski Epileptic spasms: a variety of etiologies and associated syndromes J Child Neurol 23 2008 407 414 [4] D. Talwar M.A. Baldwin R. Hutzler D.A. Griesemer Epileptic spasms in older children: persistence beyond infancy Epilepsia 36 1995 151 155 [5] E.C. Hancock J.P. Osborne S.W. Edwards Treatment of infantile spasms Cochrane Database Syst Rev 2008 CD001770 [6] R.D. Elterman W.D. Shields K.A. Mansfield J. Nakagawa Randomized trial of vigabatrin in patients with infantile spasms Neurology 57 2001 1416 1421 [7] R.D. Elterman W.D. Shields R.M. Bittman S.A. Torri S.M. Sagar S.D. Collins Vigabatrin for the treatment of infantile spasms: final report of a randomized trial J Child Neurol 25 2010 1340 1347 [8] J.M. Pellock R. Hrachovy S. Shinnar Infantile spasms: a U.S. consensus report Epilepsia 51 2010 2175 2189 [9] A. Fois Infantile spasms: review of the literature and personal experience It J Pediatr 36 2010 15 [10] S. Hamano S. Yoshinari N. Higurashi M. Tanaka M. Minamitani Y. Eto Developmental outcomes of cryptogenic West syndrome J Pediatr 150 2007 295 299 [11] R. Riikonen Infantile spasms: therapy and outcome J Child Neurol 19 2004 401 404 [12] R.S. Riikonen Favourable prognostic factors with infantile spasms Eur J Paediatr Neurol 14 2010 13 18 [13] M.J. Brodie W.E. Rosenfeld B. Vazquez Rufinamide for the adjunctive treatment of partial seizures in adults and adolescents: a randomized placebo-controlled trial Epilepsia 50 2009 1899 1909 [14] A. Cheng-Hakimian G.D. Anderson J.W. Miller Rufinamide: pharmacology, clinical trials, and role in clinical practice Int J Clin Pract 60 2006 1497 1501 [15] G. Coppola S. Grosso E. Franzoni Rufinamide in refractory childhood epileptic encephalopathies other than Lennox–Gastaut syndrome Eur J Neurol 2010 10.1111/j.1468-1331.2010.03113.x [16] C.E. Elger H. Stefan A. Mann M. Narurkar Y. Sun C. Perdomo A 24-week multicenter, randomized, double-blind, parallel-group, dose-ranging study of rufinamide in adults and adolescents with inadequately controlled partial seizures Epilepsy Res 88 2010 255 263 [17] T. Glauser G. Kluger R. Sachdeo G. Krauss C. Perdomo S. Arroyo Rufinamide for generalized seizures associated with Lennox–Gastaut syndrome Neurology 70 2008 1950 1958 [18] G. Kluger T. Glauser G. Krauss R. Seeruthun C. Perdomo S. Arroyo Adjunctive rufinamide in Lennox–Gastaut syndrome: a long-term, open-label extension study Acta Neurol Scand 122 2010 202 208 [19] G. Kluger E. Haberlandt G. Kurlemann First European long-term experience with the orphan drug rufinamide in childhood-onset refractory epilepsy Epilepsy Behav 17 2010 546 548 [20] G. Kluger G. Kurlemann E. Haberlandt Effectiveness and tolerability of rufinamide in children and adults with refractory epilepsy: first European experience Epilepsy Behav 14 2009 491 495 [21] S. Palhagen R. Canger O. Henriksen J.A. van Parys M.E. Riviere M.A. Karolchyk Rufinamide: a double-blind, placebo-controlled proof of principle trial in patients with epilepsy Epilepsy Res 43 2001 115 124 [22] J.W. Wheless J. Conry G. Krauss A. Mann A. LoPresti M. Narurkar Safety and tolerability of rufinamide in children with epilepsy: a pooled analysis of 7 clinical studies J Child Neurol 24 2009 1520 1525 [23] C.S. Wisniewski Rufinamide: a new antiepileptic medication for the treatment of seizures associated with Lennox–Gastaut syndrome Ann Pharmacother 44 2010 658 667 [24] E. Perucca J. Cloyd D. Critchley E. Fuseau Rufinamide: clinical pharmacokinetics and concentration–response relationships in patients with epilepsy Epilepsia 49 2008 1123 1141 [25] J.W. Wheless B. Vazquez Rufinamide: a novel broad-spectrum antiepileptic drug Epilepsy Curr 10 2010 1 6 [26] M. Vendrame T. Loddenkemper V.D. Gooty Experience with rufinamide in a pediatric population: a single center's experience Pediatr Neurol 43 2010 155 158 [27] W.T. Blume H.O. Luders E. Mizrahi C. Tassinari van Emde Boas W, Engel J Jr. Glossary of descriptive terminology for ictal semiology: report of the ILAE task force on classification and terminology Epilepsia 42 2001 1212 1218 [28] P. Jeavons West syndrome: infantile spasms J. Roger C. Dravet M. Bureau F.E. Dreifuss P. Wolf Epileptic syndromes in infancy, childhood and adolescence 1985 John Libbey Eurotext 42 50 [29] M. Egli I. Mothersill M. O'Kane F. O'Kane The axial spasm-the predominant type of drop seizure in patients with secondary generalized epilepsy Epilepsia 26 1985 401 415 [30] T. Ikeno H. Shigematsu M. Miyakoshi A. Ohba K. Yagi M. Seino An analytic study of epileptic falls Epilepsia 26 1985 612 621 [31] S. Arroyo Rufinamide Neurotherapeutics 4 2007 155 162