- Email: [email protected]
The long-term use of felbamate in children with severe refractory epilepsy
Epilepsy Research 47 (2001) 1 – 7 www.elsevier.com/locate/epilepsyres
The long-term use of felbamate in children with severe refractory epilepsy Maria Roberta Cilio a,*, Alex I. Kartashov b, Federico Vigevano a a
b
Di6ision of Neurology, Bambino Gesu´ Children’s Hospital, Rome, Italy Clinical Research Core Programs Office, Children’s Hospital, Har6ard Medical School, Boston, MA 02115, USA Received 14 March 2001; received in revised form 1 June 2001; accepted 18 June 2001
Abstract The aim of the study was to assess the efficacy and safety of felbamate (FBM) as add-on therapy in pediatric patients with severe uncontrolled seizures during a 3-year follow-up. Thirty-six patients were enrolled between February 1994 and February 1997. Patients suffered from partial epilepsy (n= 13), Lennox– Gastaut syndrome (LGS) (n =9), infantile spasms (IS) n= 8 or other forms of generalized epilepsy (n= 6). FBM was titrated weekly from 15 up to 45 mg/kg. By February 1995, all patients had hematological and biochemical monitoring prior to FBM therapy and every 15 days during the study. The results achieved at different treatment durations were analyzed. Overall efficacy measured as ]50% reduction in seizure frequency varied during follow-up: 69% at 3 months, progressively decreasing to 66% at 6 months, to 47% at 1 year and 41% of the initial cohort at the end of the study. Most frequent side effects were anorexia, weight loss, urinary retention, somnolence, nervousness and insomnia. FBM controlled a broad spectrum of otherwise refractory seizures. Best results were obtained against simple partial seizures with or without secondary generalization, tonic and atonic seizures. A substantial improvement in seizure control was maintained in one-third of the patients for at least 3 years. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Anticonvulsants; Felbamate; Add-on treatment; Intractable seizures; Open-label study; Long-term follow-up
1. Introduction Felbamate (FBM) (2-phenyl-1,3-propanediol dicarbamate) is one of the new generation antiepileptic drugs (AEDs). Its chemical structure differs from that of all categories of AEDs and, * Corresponding author. Present address: Department of Neurology, Children’s Hospital, EN 316, 300, Longwood Avenue, Boston, MA 02115, USA. Tel.: + 1-617-355-4468; fax: +1-617-734-1646. E-mail address: [email protected] (M.R. Cilio).
although the precise mechanism of action has not been elucidated, it can both inhibit N-methyl-Daspartate receptor current and potentiate GABAA receptor current (Rho et al., 1994). FBM has proven to be effective in the treatment of partial epilepsy and Lennox –Gastaut syndrome (LGS) in both adults and children (Leppik et al., 1991; Sachdeo et al., 1992; Bourgeois et al., 1993; Dodson, 1993; The Felbamate Study Group in LGS, 1993; Carmant et al., 1994; Canger et al., 1999). It has recently been proposed as an add-on drug for the treatment of refractory infantile spasms (IS)
0920-1211/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 0 - 1 2 1 1 ( 0 1 ) 0 0 2 9 0 - X
2
M.R. Cilio et al. / Epilepsy Research 47 (2001) 1–7
(Hosain et al., 1997). Nevertheless, FBM came very close to being withdrawn from the market in 1994 because of several cases of aplastic anemia and hepatic failure. Currently, FBM bears a black box warning, it is considered a third- or fourth-line therapy (Pellok, 1996) and it is indicated for use in children only with LGS. Thus, it is not surprising that since 1994, the number of new publications dealing with FBM has become very small. To date, no studies with long-term follow-up have been reported. This is the first systematic prospective study that reports on the efficacy and safety of the long-term use of FBM in a group of pediatric patients. Follow-up at 3, 6, 12 and 36 months is reported. Preliminary data from our study have been previously presented (Cilio and Vigevano, 1997).
2. Methods Patients were enrolled at the Bambino Gesu´ Children’s Hospital in Rome, Italy, between February 1994 and February 1997; recruitment was interrupted in August 1994 because of the reports of aplastic anemia and hepatic failure in patients on FBM and resumed in February 1995. Patients with severe refractory epilepsy, who had a minimum seizure frequency of four seizures per month and had not responded to adequate trials of at least two AEDs were eligible for inclusion. Patients who had evidence of progressive central nervous system lesion on magnetic resonance or computed tomography, a history of identifiable progressive neurologic disorder, poor compliance with past antiepileptic therapy, inadequate supervision by parents or guardians or had received other experimental drugs within the previous 30 days were excluded. All patients were classified according to the Commission on the Classification and Terminology of the ILAE, 1989 classification of epilepsies and epileptic syndromes. Prior to the beginning of the study, all patients were monitored for at least 4 h with video-EEG in order to confirm the types of
seizures and the parents were taught how to count and classify seizures on the basis of observations and video-EEG recordings. The frequency of the individual seizure types was determined, based on parental count, for the 3 months prior to FBM treatment and this period was considered a baseline. FBM was added to the concomitant AED regimen (average 2.2 AEDs per patient) at an initial dosage of 15 mg/kg per day, which was increased weekly by 15 mg/kg per day, up to a maintenance dosage of 45 mg/kg per day, but not exceeding 3600 mg per day. Clinical evaluation, hematology and hepatic function monitoring, concomitant AEDs plasma concentrations were assessed prior to FBM therapy onset, weekly during titration and then every month while FBM treatment continued. In order to avoid adverse events due to pharmacokinetics interactions, attempts were made to maintain concomitant AEDs plasma levels in steady state, equal to those obtained during the baseline period, by a 20% dose reduction. By February 1995, according to the manufacturer’s recommendations, blood tests (hepatic, hematological and concomitant AEDs plasma concentrations) were performed prior to FBM therapy onset, weekly during titration and then every 15 days throughout the study period. During the FBM treatment, the average seizure frequencies for the previous 3 months period were evaluated according to the parental reports at 3 and 6 months and 1 year during the FBM, and every 3 months thereafter. Responders were defined as patients who showed a reduction in seizure frequency of 50% or more. Patients were considered short-term responders when the improvement in seizure control lasted 1 year or less. Long-term responders were patients in whom the reduction of seizure frequency was maintained for more than 1 year and up to 3 years. Patients who dropped out because of treatment failure had at least a 3month therapy period. Patients who responded to FBM therapy continued to receive FBM after the end of the study, but for the purpose of the study, their final data were collected after 3 years of follow-up.
M.R. Cilio et al. / Epilepsy Research 47 (2001) 1–7 Table 1 Patients’ characteristics Age Age at onset
Range: 2.8–27 years; median: 9 years Range: 10 days–12 years; median: 9 months Etiology 19 Cryptogenic; 17 symptomatic Type of epilepsy 13: Partial epilepsy; 9: Lennox–Gastaut syndrome; 8: infantile spasms; 6: other forms of generalized epilepsy Duration of Range: 1–20 years; median: 7.5 years epilepsy
3. Results Thirty-six patients (24 males) presenting with intractable seizures were enrolled. Patients were classified as cryptogenic or symptomatic by history, physical examination, neuroimaging and comprehensive metabolic studies when indicated. Table 1 summarizes the patients’ characteristics. Three patients were older than 18 years (20, 22 and 27 years). These patients were treated for epilepsy in our hospital since childhood. The underlying etiologies were known in 14 out of 17 symptomatic patients: cerebral malformation (n= 8), cerebral hemorrhage (n = 2), CNS infection (n= 1), head trauma (n=1), hypoxic-ischemic encephalopathy (n =1) and white matter disease (n = 1). A specific diagnosis was lacking for the remaining three patients who presented with microcephaly and psychomotor retardation. Patients with partial epilepsy were demonstrated to have either simple or complex partial seizures, in ten cases followed by generalization. Of nine LGS patients presenting with both tonic seizures and atypical absence, four had also atonic seizures and one patient also had clonic generalized seizures. Patients with generalized epilepsy
3
had myoclonic-astatic seizures in two cases, clonic seizures in three cases (associated with myoclonic jerks in two and with atonic seizures in one) and tonic seizures associated with myoclonic seizures in one case. In 20 of 36 patients, seizures led to frequent, dangerous falls. Although the subjects were required to have a minimum of four seizures per month to be eligible, the average of seizure frequency at baseline was much higher (multiple weekly or daily seizures). All patients enrolled were initially found to be resistant to four to seven AEDs (median: five AEDs per patient), including phenobarbital, valproate (VPA), carbamazepine, phenytoin, benzodiazepines and steroids. Most of them also failed treatment with new AEDs, such as vigabatrin and lamotrigine. The overall efficacy measured as ] 50% reduction in seizure frequency varied during follow-up, as shown in Table 2. The highest response was observed in the first 3 months (69%) and later it gradually declined to a stable level of : 40%. In ten of 36 (27%) cases, the good response permitted the simplification of the polytherapy by stopping one or two concomitant AEDs. FBM was discontinued when the reduction in total seizure frequency was B 50%. During withdrawal, the study medication was tapered over 3 weeks. No withdrawals were due to adverse events. At a 1-year follow-up, 17 patients, that is 47% of the 36 treated patients, maintained substantial improvement in seizure. These 17 patients received FBM for more than 1 year. One patient was lost to follow-up after 15 months, one progressively reverted to baseline after 13 months and 15 (41%) out of the initial 36 patients who entered the study are still receiving FBM at the time of this report and are considered to be long-term
Table 2 Seizure reduction during follow-up Time points
3 months
6 months
12 months
3 years
Overall efficacy (%) Seizure-free (%) ]50% seizure reduction (%) B50% seizure reduction (%)
69 13 56 31
63 13 50 37
47 11 36 53
41 8 33 59
4
M.R. Cilio et al. / Epilepsy Research 47 (2001) 1–7
Fig. 1. Seizure per month at each time point for the 15 long-term responders. Top and bottom of shaded rectangles denote 75th and 25th percentiles, respectively and the vertical lines extend to the 90th and 10th percentiles. Horizontal lines represent the median number of seizures for the time point. Individual values are indicated by open triangles.
responders. Three of them (8%) are seizure-free: two have partial epilepsy, one cryptogenic and one symptomatic; one has symptomatic LGS. Twelve (33%) have experienced ] 50% reduction in seizure frequency: three have partial epilepsy, five have LGS, two have IS and two generalized epilepsy, six are cryptogenic and six symptomatic. Fig. 1 shows the individual numbers of seizures per month at each time point for the long-term responders. Table 3 illustrates how the different type of seizures responded to FBM therapy. Tonic seizures, simple partial seizures and tonic-clonic seizures responded better and longer compared to epileptic spasms, which tended to be more resistant and to recur after a few months. The proportion of patients with refractory partial seizures, simple or complex, showing a good response to FBM therapy was 38% (50% of responders among the patients with single partial seizures and 30% of responders among those with complex partial seizures). A positive response (40%) was also obtained regarding the reduction of the secondary generalization of partial seizures. It should be underlined that nine out of 20 patients (45%)
suffering from drop attacks stopped falling with remarkable reduction of injuries and improvement in quality of life. In seven of the long-term responders, one or more concomitant AEDs were progressively reduced and discontinued. Twenty-three of 36 patients (63%) had adverse events as listed in Table 4. In none of these were the side effects strong enough to warrant discontinuation of the drug. One-third of patients with loss of appetite experienced weight loss of \ 5% of baseline body weight. Patients who had urinary retention sometimes required intermittent bladder catheterization. All of these patients were responders. In these patients, decreasing FBM doses from 45 to 30 mg/kg per day did not change efficacy and markedly reduced urinary retention. One patient also experienced urinary retention in the past, when treated with vigabatrin. Insomnia resolved in all patients by giving the evening FBM dose 2 or 3 h earlier. Somnolence, in one case, was found to be related to an increased plasma level of phenobarbital and in another case to increased plasma level of carbamazepine-epoxide. One patient had acute tremor associated with diplopia, headache and ataxia, which was at-
M.R. Cilio et al. / Epilepsy Research 47 (2001) 1–7
5
Table 3 Felbamate and seizure type Seizure type
Number of patients
Seizure free (%)
]50% Seizures reduction (%)
B50% Seizure reduction (%)
Tonic Simple partial Tonic/clonic Epileptic spasms Atypical absence Atonic Myoclonic Clonic Complex partial
18 10 10 10 9 7 6 3 3
11 20 30 0 0 43 0 0 0
33 30 10 20 55 0 16 66 33
55 50 60 80 44 57 83 33 66
Patients may have more than one type of seizure.
tributed to a plasma level of phenytoin of 41 mg/ml; symptoms promptly disappeared after phenytoin was discontinued for 48 h and the plasma level dropped within the therapeutic range. In five of 36 patients (13%), all responders, parents reported improvement in social, intellectual and motor functioning, attention, concentration and alertness. Throughout the entire study, no increase in seizure frequency was reported with FBM and no changes in hepatic or hematological blood values were noted. No gender differences were observed.
4. Discussion We sought to assess the long-term efficacy and tolerability of what we still consider a valuable drug for children with otherwise intractable seizures. FBM proved to be remarkably effective in this group of pediatric patients who had been unable to achieve seizure control, failing prior trials of four to seven AEDs. The analysis of our data on the responders’ follow-up indicated that the best response for each patient occurred within 3 months of therapy and the main drop-off due to a decrease of the efficacy was observed between 3 and 6 months of therapy with no significant variation thereafter. Our study also demonstrates FBM’s unique broad spectrum of anticonvulsant activity in patients presenting with different types of seizures. Best results were obtained against simple partial
seizures, tonic, atonic and secondarily generalized seizures. Moreover, nine of 20 patients (45%) suffering from epileptic falls stopped falling, with remarkable reduction of injuries and improvement in quality of life. Epileptic falls due to both tonic and atonic seizures are considered the most debilitating as well as one of the most resistant to treatment. According to the data in the literature (The Felbamate Study Group in LGS, 1993; Dodson, 1993), FBM is effective against epileptic falls. FBM’s therapeutic effect on drop attacks has recently been suggested to be due to increased VPA plasma levels (Siegel et al., 1999). In our study, we did not notice any positive relationship between VPA plasma levels up to 110 mg/ml and decrease in seizure frequency. Thus, we conclude that FBM itself should be considered responsible for the positive effects on this type of seizure. Table 4 Felbamate and side effects Side effect
No. of patients
%
Any side effect Appetite loss Insomnia Nervousness Somnolence Urinary retention Vomiting Ataxia Headache Tremor Diplopia
23 16 4 4 4 4 3 2 1 1 1
63.8 44 11 11 11 11 8 5 3 3 3
Patients may have had more than one side effect.
6
M.R. Cilio et al. / Epilepsy Research 47 (2001) 1–7
FBM’s spectrum of activity may result from the drug’s apparent dual mechanism of action. Early studies (Rho et al., 1994) demonstrated that the excitatory responses NMDA-mediated were inhibited by clinically relevant concentrations of FBM. However, FBM has also been shown to enhance the inhibitory GABAA system, in common with phenobarbital and benzodiazepines. In some responders, we noticed a marked improvement in alertness and motor skills that may have been due to the reduction in the number of seizures or to a direct drug effect, or both. However, there was no direct correlation between the amount of overt seizure reduction and behavioral improvement. Although the incidence of adverse events in our patients was high (68%), we found FBM to have relatively low toxicity compared with other AEDs. Some of the adverse events were most likely related to FBM-induced changes in plasma concentration of other AEDs that might occur transiently despite adjustments of dosages. Indeed, as previously reported (Fuerst et al., 1988; Albani et al., 1991), we confirmed FBM to have a high potential for interaction: doses of co medications needed to be corrected several times during the first 3 months in order to maintain their plasma levels at the baseline values. After that period, blood levels of concomitant AEDs tended to stabilize. Other side effects, including appetite and weight loss as well as nervousness and insomnia, seemed to be related to FBM per se. Nonetheless, the intensity of these reactions was mild or moderate, most of them resolved with a slight decrease of FBM dose and none resulted in premature discontinuation of treatment. An unusual adverse event was urinary retention, sometimes important enough to require catheterization. Urinary retention has also been sporadically described as a side effect of other AEDs, such as phenytoin and carbamazepine (Steiner and Birmanns, 1993; Nussinovitch et al., 1995). There are limitations to our study design. Patients were receiving multiples AEDs, which made it more difficult to evaluate the effect of a single added drug, felbamate, in improving seizure control. Because the study was open label, it is possible that at least part of the observed effect on
seizure frequency was the result of a placebo effect. There was no control group and the effect of FBM was assessed by comparison to the baseline. Patients often enter studies at a relatively high point in their seizure frequency and part of the decrease observed could be due to a statistical regression towards the mean. Long-term follow-up can correct the phenomenon of the natural fluctuation in seizure frequency and allows a reliable assessment of both the efficacy and tolerability. At least for our responders, sustainable and consistent decrease in seizure frequency, as shown in Fig. 1, suggests that the effect of regression towards the mean is not very important. Tolerance is a common problem in the treatment of LGS patients and in most drug-resistant patients in whom development of tolerance encourages dosage increase, which in turn may increase the side effects. In agreement with two previous reports (Dodson, 1993; Carmant et al., 1994) we noted that tolerance did develop only in a small percentage of patients who dropped out for treatment failure after having responded to FBM therapy for a certain time. In conclusion, despite the need for frequent monitoring of hematological parameters and liver function, we believe that FBM can be a valuable resource where the potential benefit outweighs the possible risk. Although data from this prospective analysis of open, long-term experience with FBM are considered only suggestive, the overall rate of long-term response of 41%, the successful results maintained up to 3 years and the percentage of patients currently seizure-free indicate that this drug remains a viable treatment option for children who are unable to achieve seizure control with traditional AEDs, especially those with LGS, partial seizures and drop attacks. It was extremely beneficial for several patients who are still receiving FBM at the time of this report. References Albani, F., Theodore, W.H., Washington, P., Devinsky, O., Bronfield, E., Porter, R.J., Nice, F.J., 1991. Effect of felbamate on plasma levels of carbamazepine and its metabolites. Epilepsia 32, 130 – 132.
M.R. Cilio et al. / Epilepsy Research 47 (2001) 1–7 Bourgeois, B., Leppik, I.E., Sackellares, J.C., Laxer, K., Lesser, R., Messenheimer, J.A., Kramer, L.D., Kamin, M., Rosemberg, A., 1993. Felbamate: a double blind controlled trial in patients undergoing presurgical evaluation of partial seizures. Neurology 43, 693 –696. Canger, R., Vignoli, A., Bonardi, R., Guidolin, L., 1999. Felbamate in refractory partial epilepsy. Epilepsy Res. 34, 43 – 48. Carmant, L., Holmes, G.L., Sawyer, S., Rifai, N., Anderson, J., Mikati, M.A., et al., 1994. Efficacy of Felbamate in therapy for partial epilepsy in children. J. Pediatr. 125, 481 – 486. Cilio, M.R., Vigevano, F., 1997. Felbamate add-on in refractory epilepsy: an open label study with long-term follow-up. Ann. Neurol. 42, 506. Commission on the Classification and Terminology of the International League Against Epilepsy, 1989. Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 30, 389 – 399. Dodson, W.E., 1993. Felbamate in the treatment of Lennox – Gastaut syndrome: results of a 12-month open-label study following a randomized clinical trial. Epilepsia 34, S18 – S24. Fuerst, R.H., Graves, N.M., Leppik, I.E., Brundage, R.C., Holmes, G.B., Remmel, R.P., 1988. Felbamate increases phenytoin but decreases carbamazepine concentrations. Epilepsia 29, 488 – 491. Hosain, S., Nagarajan, L., Carson, D., Solomon, G., Mast, J., Labar, D., 1997. Felbamate for refractory infantile spasms. J. Child Neurol. 12, 466 –468.
7
Leppik, I.E., Dreifuss, F.E., Pledger, G.W., Graves, N.M., Santilli, N., Drury, I., Tsay, J.Y., Bertram, E., Cereghino, J.J., 1991. Felbamate for partial seizures: results of a controlled clinical trial. Neurology 41, 1785 – 1789. Nussinovitch, M., Soen, G., Voloviz, B., Varsano, I., Luttwak, P.Z., 1995. Urinary retention related to phenytoin therapy. Clin. Pediatr. (Phil.) 34, 382 – 383. Pellok, J.M., 1996. Utilization of new antiepileptic drugs in children. Epilepsia 37, S66 – S73. Rho, J.M., Donevain, S.D., Rogawski, M.A., 1994. Mechanism of action of the anti-convulsant felbamate: opposing effects on N-methyl-D-aspartate and GABA A receptors. Ann. Neurol. 35, 224 – 234. Sachdeo, R., Kramer, L.D., Rosemberg, A., Sachdeo, S., 1992. Felbamate monotherapy: controlled trial in patients with partial onset seizures. Ann. Neurol. 32, 386 – 392. Siegel, H., Kelley, K., Stertz, B., Reeves-Tyer, P., Flamini, R., Malow, B., Gaillard, W.D., Ko, D., Theodore, W.H., 1999. The efficacy of Felbamate as add-on therapy to valproic acid in the Lennox – Gastaut syndrome. Epilepsy Res. 34, 91 – 97. Steiner, I., Birmanns, B., 1993. Carbamazepine-induced urinary retention in long-standing diabetes mellitus. Neurology 43, 1855 – 1856. The Felbamate Study Group in Lennox – Gastaut Syndrome, 1993. Efficacy of Felbamate in childhood epileptic encephalopathy (Lennox – Gastaut syndrome). New Engl. J. Med. 328, 29 – 33.
The long-term use of felbamate in children with severe refractory epilepsy Maria Roberta Cilio a,*, Alex I. Kartashov b, Federico Vigevano a a
b
Di6ision of Neurology, Bambino Gesu´ Children’s Hospital, Rome, Italy Clinical Research Core Programs Office, Children’s Hospital, Har6ard Medical School, Boston, MA 02115, USA Received 14 March 2001; received in revised form 1 June 2001; accepted 18 June 2001
Abstract The aim of the study was to assess the efficacy and safety of felbamate (FBM) as add-on therapy in pediatric patients with severe uncontrolled seizures during a 3-year follow-up. Thirty-six patients were enrolled between February 1994 and February 1997. Patients suffered from partial epilepsy (n= 13), Lennox– Gastaut syndrome (LGS) (n =9), infantile spasms (IS) n= 8 or other forms of generalized epilepsy (n= 6). FBM was titrated weekly from 15 up to 45 mg/kg. By February 1995, all patients had hematological and biochemical monitoring prior to FBM therapy and every 15 days during the study. The results achieved at different treatment durations were analyzed. Overall efficacy measured as ]50% reduction in seizure frequency varied during follow-up: 69% at 3 months, progressively decreasing to 66% at 6 months, to 47% at 1 year and 41% of the initial cohort at the end of the study. Most frequent side effects were anorexia, weight loss, urinary retention, somnolence, nervousness and insomnia. FBM controlled a broad spectrum of otherwise refractory seizures. Best results were obtained against simple partial seizures with or without secondary generalization, tonic and atonic seizures. A substantial improvement in seizure control was maintained in one-third of the patients for at least 3 years. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Anticonvulsants; Felbamate; Add-on treatment; Intractable seizures; Open-label study; Long-term follow-up
1. Introduction Felbamate (FBM) (2-phenyl-1,3-propanediol dicarbamate) is one of the new generation antiepileptic drugs (AEDs). Its chemical structure differs from that of all categories of AEDs and, * Corresponding author. Present address: Department of Neurology, Children’s Hospital, EN 316, 300, Longwood Avenue, Boston, MA 02115, USA. Tel.: + 1-617-355-4468; fax: +1-617-734-1646. E-mail address: [email protected] (M.R. Cilio).
although the precise mechanism of action has not been elucidated, it can both inhibit N-methyl-Daspartate receptor current and potentiate GABAA receptor current (Rho et al., 1994). FBM has proven to be effective in the treatment of partial epilepsy and Lennox –Gastaut syndrome (LGS) in both adults and children (Leppik et al., 1991; Sachdeo et al., 1992; Bourgeois et al., 1993; Dodson, 1993; The Felbamate Study Group in LGS, 1993; Carmant et al., 1994; Canger et al., 1999). It has recently been proposed as an add-on drug for the treatment of refractory infantile spasms (IS)
0920-1211/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 0 - 1 2 1 1 ( 0 1 ) 0 0 2 9 0 - X
2
M.R. Cilio et al. / Epilepsy Research 47 (2001) 1–7
(Hosain et al., 1997). Nevertheless, FBM came very close to being withdrawn from the market in 1994 because of several cases of aplastic anemia and hepatic failure. Currently, FBM bears a black box warning, it is considered a third- or fourth-line therapy (Pellok, 1996) and it is indicated for use in children only with LGS. Thus, it is not surprising that since 1994, the number of new publications dealing with FBM has become very small. To date, no studies with long-term follow-up have been reported. This is the first systematic prospective study that reports on the efficacy and safety of the long-term use of FBM in a group of pediatric patients. Follow-up at 3, 6, 12 and 36 months is reported. Preliminary data from our study have been previously presented (Cilio and Vigevano, 1997).
2. Methods Patients were enrolled at the Bambino Gesu´ Children’s Hospital in Rome, Italy, between February 1994 and February 1997; recruitment was interrupted in August 1994 because of the reports of aplastic anemia and hepatic failure in patients on FBM and resumed in February 1995. Patients with severe refractory epilepsy, who had a minimum seizure frequency of four seizures per month and had not responded to adequate trials of at least two AEDs were eligible for inclusion. Patients who had evidence of progressive central nervous system lesion on magnetic resonance or computed tomography, a history of identifiable progressive neurologic disorder, poor compliance with past antiepileptic therapy, inadequate supervision by parents or guardians or had received other experimental drugs within the previous 30 days were excluded. All patients were classified according to the Commission on the Classification and Terminology of the ILAE, 1989 classification of epilepsies and epileptic syndromes. Prior to the beginning of the study, all patients were monitored for at least 4 h with video-EEG in order to confirm the types of
seizures and the parents were taught how to count and classify seizures on the basis of observations and video-EEG recordings. The frequency of the individual seizure types was determined, based on parental count, for the 3 months prior to FBM treatment and this period was considered a baseline. FBM was added to the concomitant AED regimen (average 2.2 AEDs per patient) at an initial dosage of 15 mg/kg per day, which was increased weekly by 15 mg/kg per day, up to a maintenance dosage of 45 mg/kg per day, but not exceeding 3600 mg per day. Clinical evaluation, hematology and hepatic function monitoring, concomitant AEDs plasma concentrations were assessed prior to FBM therapy onset, weekly during titration and then every month while FBM treatment continued. In order to avoid adverse events due to pharmacokinetics interactions, attempts were made to maintain concomitant AEDs plasma levels in steady state, equal to those obtained during the baseline period, by a 20% dose reduction. By February 1995, according to the manufacturer’s recommendations, blood tests (hepatic, hematological and concomitant AEDs plasma concentrations) were performed prior to FBM therapy onset, weekly during titration and then every 15 days throughout the study period. During the FBM treatment, the average seizure frequencies for the previous 3 months period were evaluated according to the parental reports at 3 and 6 months and 1 year during the FBM, and every 3 months thereafter. Responders were defined as patients who showed a reduction in seizure frequency of 50% or more. Patients were considered short-term responders when the improvement in seizure control lasted 1 year or less. Long-term responders were patients in whom the reduction of seizure frequency was maintained for more than 1 year and up to 3 years. Patients who dropped out because of treatment failure had at least a 3month therapy period. Patients who responded to FBM therapy continued to receive FBM after the end of the study, but for the purpose of the study, their final data were collected after 3 years of follow-up.
M.R. Cilio et al. / Epilepsy Research 47 (2001) 1–7 Table 1 Patients’ characteristics Age Age at onset
Range: 2.8–27 years; median: 9 years Range: 10 days–12 years; median: 9 months Etiology 19 Cryptogenic; 17 symptomatic Type of epilepsy 13: Partial epilepsy; 9: Lennox–Gastaut syndrome; 8: infantile spasms; 6: other forms of generalized epilepsy Duration of Range: 1–20 years; median: 7.5 years epilepsy
3. Results Thirty-six patients (24 males) presenting with intractable seizures were enrolled. Patients were classified as cryptogenic or symptomatic by history, physical examination, neuroimaging and comprehensive metabolic studies when indicated. Table 1 summarizes the patients’ characteristics. Three patients were older than 18 years (20, 22 and 27 years). These patients were treated for epilepsy in our hospital since childhood. The underlying etiologies were known in 14 out of 17 symptomatic patients: cerebral malformation (n= 8), cerebral hemorrhage (n = 2), CNS infection (n= 1), head trauma (n=1), hypoxic-ischemic encephalopathy (n =1) and white matter disease (n = 1). A specific diagnosis was lacking for the remaining three patients who presented with microcephaly and psychomotor retardation. Patients with partial epilepsy were demonstrated to have either simple or complex partial seizures, in ten cases followed by generalization. Of nine LGS patients presenting with both tonic seizures and atypical absence, four had also atonic seizures and one patient also had clonic generalized seizures. Patients with generalized epilepsy
3
had myoclonic-astatic seizures in two cases, clonic seizures in three cases (associated with myoclonic jerks in two and with atonic seizures in one) and tonic seizures associated with myoclonic seizures in one case. In 20 of 36 patients, seizures led to frequent, dangerous falls. Although the subjects were required to have a minimum of four seizures per month to be eligible, the average of seizure frequency at baseline was much higher (multiple weekly or daily seizures). All patients enrolled were initially found to be resistant to four to seven AEDs (median: five AEDs per patient), including phenobarbital, valproate (VPA), carbamazepine, phenytoin, benzodiazepines and steroids. Most of them also failed treatment with new AEDs, such as vigabatrin and lamotrigine. The overall efficacy measured as ] 50% reduction in seizure frequency varied during follow-up, as shown in Table 2. The highest response was observed in the first 3 months (69%) and later it gradually declined to a stable level of : 40%. In ten of 36 (27%) cases, the good response permitted the simplification of the polytherapy by stopping one or two concomitant AEDs. FBM was discontinued when the reduction in total seizure frequency was B 50%. During withdrawal, the study medication was tapered over 3 weeks. No withdrawals were due to adverse events. At a 1-year follow-up, 17 patients, that is 47% of the 36 treated patients, maintained substantial improvement in seizure. These 17 patients received FBM for more than 1 year. One patient was lost to follow-up after 15 months, one progressively reverted to baseline after 13 months and 15 (41%) out of the initial 36 patients who entered the study are still receiving FBM at the time of this report and are considered to be long-term
Table 2 Seizure reduction during follow-up Time points
3 months
6 months
12 months
3 years
Overall efficacy (%) Seizure-free (%) ]50% seizure reduction (%) B50% seizure reduction (%)
69 13 56 31
63 13 50 37
47 11 36 53
41 8 33 59
4
M.R. Cilio et al. / Epilepsy Research 47 (2001) 1–7
Fig. 1. Seizure per month at each time point for the 15 long-term responders. Top and bottom of shaded rectangles denote 75th and 25th percentiles, respectively and the vertical lines extend to the 90th and 10th percentiles. Horizontal lines represent the median number of seizures for the time point. Individual values are indicated by open triangles.
responders. Three of them (8%) are seizure-free: two have partial epilepsy, one cryptogenic and one symptomatic; one has symptomatic LGS. Twelve (33%) have experienced ] 50% reduction in seizure frequency: three have partial epilepsy, five have LGS, two have IS and two generalized epilepsy, six are cryptogenic and six symptomatic. Fig. 1 shows the individual numbers of seizures per month at each time point for the long-term responders. Table 3 illustrates how the different type of seizures responded to FBM therapy. Tonic seizures, simple partial seizures and tonic-clonic seizures responded better and longer compared to epileptic spasms, which tended to be more resistant and to recur after a few months. The proportion of patients with refractory partial seizures, simple or complex, showing a good response to FBM therapy was 38% (50% of responders among the patients with single partial seizures and 30% of responders among those with complex partial seizures). A positive response (40%) was also obtained regarding the reduction of the secondary generalization of partial seizures. It should be underlined that nine out of 20 patients (45%)
suffering from drop attacks stopped falling with remarkable reduction of injuries and improvement in quality of life. In seven of the long-term responders, one or more concomitant AEDs were progressively reduced and discontinued. Twenty-three of 36 patients (63%) had adverse events as listed in Table 4. In none of these were the side effects strong enough to warrant discontinuation of the drug. One-third of patients with loss of appetite experienced weight loss of \ 5% of baseline body weight. Patients who had urinary retention sometimes required intermittent bladder catheterization. All of these patients were responders. In these patients, decreasing FBM doses from 45 to 30 mg/kg per day did not change efficacy and markedly reduced urinary retention. One patient also experienced urinary retention in the past, when treated with vigabatrin. Insomnia resolved in all patients by giving the evening FBM dose 2 or 3 h earlier. Somnolence, in one case, was found to be related to an increased plasma level of phenobarbital and in another case to increased plasma level of carbamazepine-epoxide. One patient had acute tremor associated with diplopia, headache and ataxia, which was at-
M.R. Cilio et al. / Epilepsy Research 47 (2001) 1–7
5
Table 3 Felbamate and seizure type Seizure type
Number of patients
Seizure free (%)
]50% Seizures reduction (%)
B50% Seizure reduction (%)
Tonic Simple partial Tonic/clonic Epileptic spasms Atypical absence Atonic Myoclonic Clonic Complex partial
18 10 10 10 9 7 6 3 3
11 20 30 0 0 43 0 0 0
33 30 10 20 55 0 16 66 33
55 50 60 80 44 57 83 33 66
Patients may have more than one type of seizure.
tributed to a plasma level of phenytoin of 41 mg/ml; symptoms promptly disappeared after phenytoin was discontinued for 48 h and the plasma level dropped within the therapeutic range. In five of 36 patients (13%), all responders, parents reported improvement in social, intellectual and motor functioning, attention, concentration and alertness. Throughout the entire study, no increase in seizure frequency was reported with FBM and no changes in hepatic or hematological blood values were noted. No gender differences were observed.
4. Discussion We sought to assess the long-term efficacy and tolerability of what we still consider a valuable drug for children with otherwise intractable seizures. FBM proved to be remarkably effective in this group of pediatric patients who had been unable to achieve seizure control, failing prior trials of four to seven AEDs. The analysis of our data on the responders’ follow-up indicated that the best response for each patient occurred within 3 months of therapy and the main drop-off due to a decrease of the efficacy was observed between 3 and 6 months of therapy with no significant variation thereafter. Our study also demonstrates FBM’s unique broad spectrum of anticonvulsant activity in patients presenting with different types of seizures. Best results were obtained against simple partial
seizures, tonic, atonic and secondarily generalized seizures. Moreover, nine of 20 patients (45%) suffering from epileptic falls stopped falling, with remarkable reduction of injuries and improvement in quality of life. Epileptic falls due to both tonic and atonic seizures are considered the most debilitating as well as one of the most resistant to treatment. According to the data in the literature (The Felbamate Study Group in LGS, 1993; Dodson, 1993), FBM is effective against epileptic falls. FBM’s therapeutic effect on drop attacks has recently been suggested to be due to increased VPA plasma levels (Siegel et al., 1999). In our study, we did not notice any positive relationship between VPA plasma levels up to 110 mg/ml and decrease in seizure frequency. Thus, we conclude that FBM itself should be considered responsible for the positive effects on this type of seizure. Table 4 Felbamate and side effects Side effect
No. of patients
%
Any side effect Appetite loss Insomnia Nervousness Somnolence Urinary retention Vomiting Ataxia Headache Tremor Diplopia
23 16 4 4 4 4 3 2 1 1 1
63.8 44 11 11 11 11 8 5 3 3 3
Patients may have had more than one side effect.
6
M.R. Cilio et al. / Epilepsy Research 47 (2001) 1–7
FBM’s spectrum of activity may result from the drug’s apparent dual mechanism of action. Early studies (Rho et al., 1994) demonstrated that the excitatory responses NMDA-mediated were inhibited by clinically relevant concentrations of FBM. However, FBM has also been shown to enhance the inhibitory GABAA system, in common with phenobarbital and benzodiazepines. In some responders, we noticed a marked improvement in alertness and motor skills that may have been due to the reduction in the number of seizures or to a direct drug effect, or both. However, there was no direct correlation between the amount of overt seizure reduction and behavioral improvement. Although the incidence of adverse events in our patients was high (68%), we found FBM to have relatively low toxicity compared with other AEDs. Some of the adverse events were most likely related to FBM-induced changes in plasma concentration of other AEDs that might occur transiently despite adjustments of dosages. Indeed, as previously reported (Fuerst et al., 1988; Albani et al., 1991), we confirmed FBM to have a high potential for interaction: doses of co medications needed to be corrected several times during the first 3 months in order to maintain their plasma levels at the baseline values. After that period, blood levels of concomitant AEDs tended to stabilize. Other side effects, including appetite and weight loss as well as nervousness and insomnia, seemed to be related to FBM per se. Nonetheless, the intensity of these reactions was mild or moderate, most of them resolved with a slight decrease of FBM dose and none resulted in premature discontinuation of treatment. An unusual adverse event was urinary retention, sometimes important enough to require catheterization. Urinary retention has also been sporadically described as a side effect of other AEDs, such as phenytoin and carbamazepine (Steiner and Birmanns, 1993; Nussinovitch et al., 1995). There are limitations to our study design. Patients were receiving multiples AEDs, which made it more difficult to evaluate the effect of a single added drug, felbamate, in improving seizure control. Because the study was open label, it is possible that at least part of the observed effect on
seizure frequency was the result of a placebo effect. There was no control group and the effect of FBM was assessed by comparison to the baseline. Patients often enter studies at a relatively high point in their seizure frequency and part of the decrease observed could be due to a statistical regression towards the mean. Long-term follow-up can correct the phenomenon of the natural fluctuation in seizure frequency and allows a reliable assessment of both the efficacy and tolerability. At least for our responders, sustainable and consistent decrease in seizure frequency, as shown in Fig. 1, suggests that the effect of regression towards the mean is not very important. Tolerance is a common problem in the treatment of LGS patients and in most drug-resistant patients in whom development of tolerance encourages dosage increase, which in turn may increase the side effects. In agreement with two previous reports (Dodson, 1993; Carmant et al., 1994) we noted that tolerance did develop only in a small percentage of patients who dropped out for treatment failure after having responded to FBM therapy for a certain time. In conclusion, despite the need for frequent monitoring of hematological parameters and liver function, we believe that FBM can be a valuable resource where the potential benefit outweighs the possible risk. Although data from this prospective analysis of open, long-term experience with FBM are considered only suggestive, the overall rate of long-term response of 41%, the successful results maintained up to 3 years and the percentage of patients currently seizure-free indicate that this drug remains a viable treatment option for children who are unable to achieve seizure control with traditional AEDs, especially those with LGS, partial seizures and drop attacks. It was extremely beneficial for several patients who are still receiving FBM at the time of this report. References Albani, F., Theodore, W.H., Washington, P., Devinsky, O., Bronfield, E., Porter, R.J., Nice, F.J., 1991. Effect of felbamate on plasma levels of carbamazepine and its metabolites. Epilepsia 32, 130 – 132.
M.R. Cilio et al. / Epilepsy Research 47 (2001) 1–7 Bourgeois, B., Leppik, I.E., Sackellares, J.C., Laxer, K., Lesser, R., Messenheimer, J.A., Kramer, L.D., Kamin, M., Rosemberg, A., 1993. Felbamate: a double blind controlled trial in patients undergoing presurgical evaluation of partial seizures. Neurology 43, 693 –696. Canger, R., Vignoli, A., Bonardi, R., Guidolin, L., 1999. Felbamate in refractory partial epilepsy. Epilepsy Res. 34, 43 – 48. Carmant, L., Holmes, G.L., Sawyer, S., Rifai, N., Anderson, J., Mikati, M.A., et al., 1994. Efficacy of Felbamate in therapy for partial epilepsy in children. J. Pediatr. 125, 481 – 486. Cilio, M.R., Vigevano, F., 1997. Felbamate add-on in refractory epilepsy: an open label study with long-term follow-up. Ann. Neurol. 42, 506. Commission on the Classification and Terminology of the International League Against Epilepsy, 1989. Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 30, 389 – 399. Dodson, W.E., 1993. Felbamate in the treatment of Lennox – Gastaut syndrome: results of a 12-month open-label study following a randomized clinical trial. Epilepsia 34, S18 – S24. Fuerst, R.H., Graves, N.M., Leppik, I.E., Brundage, R.C., Holmes, G.B., Remmel, R.P., 1988. Felbamate increases phenytoin but decreases carbamazepine concentrations. Epilepsia 29, 488 – 491. Hosain, S., Nagarajan, L., Carson, D., Solomon, G., Mast, J., Labar, D., 1997. Felbamate for refractory infantile spasms. J. Child Neurol. 12, 466 –468.
7
Leppik, I.E., Dreifuss, F.E., Pledger, G.W., Graves, N.M., Santilli, N., Drury, I., Tsay, J.Y., Bertram, E., Cereghino, J.J., 1991. Felbamate for partial seizures: results of a controlled clinical trial. Neurology 41, 1785 – 1789. Nussinovitch, M., Soen, G., Voloviz, B., Varsano, I., Luttwak, P.Z., 1995. Urinary retention related to phenytoin therapy. Clin. Pediatr. (Phil.) 34, 382 – 383. Pellok, J.M., 1996. Utilization of new antiepileptic drugs in children. Epilepsia 37, S66 – S73. Rho, J.M., Donevain, S.D., Rogawski, M.A., 1994. Mechanism of action of the anti-convulsant felbamate: opposing effects on N-methyl-D-aspartate and GABA A receptors. Ann. Neurol. 35, 224 – 234. Sachdeo, R., Kramer, L.D., Rosemberg, A., Sachdeo, S., 1992. Felbamate monotherapy: controlled trial in patients with partial onset seizures. Ann. Neurol. 32, 386 – 392. Siegel, H., Kelley, K., Stertz, B., Reeves-Tyer, P., Flamini, R., Malow, B., Gaillard, W.D., Ko, D., Theodore, W.H., 1999. The efficacy of Felbamate as add-on therapy to valproic acid in the Lennox – Gastaut syndrome. Epilepsy Res. 34, 91 – 97. Steiner, I., Birmanns, B., 1993. Carbamazepine-induced urinary retention in long-standing diabetes mellitus. Neurology 43, 1855 – 1856. The Felbamate Study Group in Lennox – Gastaut Syndrome, 1993. Efficacy of Felbamate in childhood epileptic encephalopathy (Lennox – Gastaut syndrome). New Engl. J. Med. 328, 29 – 33.