Eslicarbazepine Acetate: A Double-blind, Add-on, Placebo-controlled Exploratory Trial in Adult Patients with Partial-onset Seizures

Authors


Address correspondence and reprint requests to Patrício Soares-da-Silva at Department of Research and Development, BIAL, À Avenida da Siderugia Nacional, 4745-457 S. Mamedo do Coronado, Portugal. E-mail: psoares.silva@bial.com

Abstract

Summary: Objective: To explore the efficacy and safety of eslicarbazepine acetate (BIA 2-093), a new antiepileptic drug, as adjunctive therapy in adult patients with partial epilepsy.

Methods: A multicenter, double-blind, randomized, placebo-controlled study was conducted in 143 refractory patients aged 18–65 years with ≥4 partial-onset seizures/month. The study consisted of a 12-week treatment period followed by a 1-week tapering off. Patients were randomly assigned to one of three groups: treatment with eslicarbazepine acetate once daily (QD, n = 50), twice daily (BID, n = 46), or placebo (PL, n = 47). The daily dose was titrated from 400 mg to 800 mg and to 1,200 mg at 4-week intervals. The proportion of responders (patients with a ≥50% seizure reduction) was the primary end point.

Results: The percentage of responders versus baseline showed a statistically significant difference between QD and PL groups (54% vs. 28%; 90% CI =–∞, –14; p = 0.008). The difference between the BID (41%) and PL did not reach statistical significance (90% CI =–∞, –1; p = 0.12). A significantly higher proportion of responders in weeks 5–8 was found in the QD group than in the BID group (58% vs. 33%, respectively, p = 0.022). At the end of the 12-week treatment, the number of seizure-free patients in the QD and BID groups was 24%, which was significantly different from the PL group. The incidence of adverse events was similar between the treatment groups and no drug-related serious adverse events occurred.

Conclusion: Eslicarbazepine acetate was efficacious and well tolerated as an adjunctive therapy of refractory epileptic patients.

Eslicarbazepine acetate (ESL) or BIA 2-093 (S-(–)-10-acetoxy-10,11-dihydro-5H-dibenz/b,f/azepine-5-carboxamide) is a new drug whose mechanism of action is by blocking the voltage-gated sodium channel (VGSC) (Benes et al., 1999). ESL shares with carbamazepine (CBZ) and oxcarbazepine (OXC) the basic chemical structure of a dibenzazepine nucleus with the 5-carboxamide substituent, but is structurally different at the 10,11-position (Benes et al., 1999). This molecular variation results in the following differences in ESL metabolism: (a) unlike CBZ, ESL is not metabolized to CBZ-10,11-epoxide and, therefore, is not susceptible to enzyme induction or autoinduction; and (b) whereas OXC is a prodrug to both enantiomers of the OXC monohydroxy derivative (MHD), ESL is a prodrug of just S-MHD, also designated as S-licarbazepine or eslicarbazepine (Benes et al., 1999; Hainzl et al., 2001). ESL is an effective anticonvulsant in laboratory animals, exerting protecting effects against maximal electroshock seizure (MES) and a variety of chemically induced seizure models. Although data from animal studies do not necessarily extrapolate to man, in animal models ESL was found to be particularly active against MES-induced seizures with anticonvulsant potency similar to that of CBZ, but more potent than OXC (Benes et al., 1999). In comparison to CBZ and OXC, ESL was also found to cause less neurological impairment, resulting in a higher protective index (PI = TD50/ED50) (Benes et al., 1999).

Mechanistically, ESL does not interfere with receptors for benzodiazepines, gamma-aminobutyric acid (GABA), and glutamate. It behaves as a potent blocker of VGSC by competitively interacting with site 2 of the inactivated state of the channel (Benes et al., 1999; Ambrosio et al., 2001, 2002). Its affinity for this state of the channel was found to be similar to that of CBZ, while the affinity for the resting state of the channel was found to be about three-fold lower than that of CBZ (Bonifacio et al., 2001). This profile suggests an enhanced inhibitory selectivity of ESL for rapidly firing neurons over those displaying normal activity (Bonifacio et al., 2001).

Following oral administration to humans, ESL is rapidly and extensively hydrolyzed in the liver to S-MHD or the S(+)-enantiomer of licarbazepine [S-licarbazepine, (S)-(+)-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide] (Silveira et al., 2004; Almeida et al., 2005a). S-MHD is the primary active metabolite of ESL but after its formation a minor part of it (∼5%) undergoes chiral inversion, probably via metabolic oxidation to OXC and subsequent reduction to R-MHD (Hainzl et al., 2001). Following oral administration to humans ESL plasma concentrations have been below the limit of quantification (LOQ) of the assay (10 ng/ml), because it is rapidly and extensively converted to S-MHD (Almeida et al., 2003, 2004, 2005a; Silveira et al., 2004). When a nonchiral method is used, the assay does not distinguish between eslicarbazepine (S-MHD) and R-MHD, and thus, MHD plasma levels and exposure (area under the plasma drug concentration-time curve, AUC) have been reported as BIA 2-005, which is a mixture of S-MHD and R-MHD, with the former being the most prevalent enantiomer (95%) in plasma (Almeida et al., 2003, 2004; Fontes-Ribeiro et al., 2005). The bioavailability of ESL measured in terms of AUC of the active entity (metabolites) BIA 2-005 is 16% greater than that observed after oral dosing of an equivalent molar dose of OXC (Bialer et al., 2004).

In phase I studies ESL was administered to healthy subjects in single oral doses up to 2,400 mg (Almeida et al., 2003; Vaz-da-Silva et al., 2005) and in multiple doses ranging from 200 mg twice daily to 2,400 mg once daily (Almeida et al., 2004; Vaz-da-Silva et al., 2005). These studies showed that BIA 2-005 peak plasma concentration (Cmax) was obtained at 1–4 h after dosing (tmax) and the extent of systemic exposure (AUC) of BIA 2-005 was approximately dose proportional. Following multiple dosing BIA 2-005 steady-state plasma levels were attained after 4–5 days, consistent with an effective half-life of 20–24 h. The mean renal clearance of BIA 2-005 was 20–30 ml/min and the total amount of BIA 2-005 excreted in the urine was approximately 20% and 40% within 12 h and 24 h postdose, respectively. A study in healthy subjects showed that the presence of food did not change the pharmacokinetic (PK) profile of ESL or BIA 2-005 (Maia et al., 2005). No age-related or gender-related differences were found in the ESL PK of both young and elderly subjects (Almeida et al., 2005a).

The objectives of the current study were to assess efficacy (primary) and safety (secondary) of ESL as an adjunctive therapy in patients with treatment-refractory simple or complex partial seizures with or without secondary generalization. Preliminary results of this trial were previously reported in an abstract format (Almeida et al., 2005b).

METHODS

Study design

This was a multicenter, randomized, placebo-controlled, parallel-group, therapeutic exploratory (phase II) study comparing monthly ascending doses of ESL to placebo. The trial was conducted in 19 centers in five European countries: Croatia, Czech Republic, Germany, Lithuania, and Poland. The study was approved by local ethics committees and national regulatory authorities, and was conducted according to the principles of the Declaration of Helsinki and the Good Clinical Practice (GCP) recommendations. Written informed consent was obtained from each patient prior to enrolment in the study.

The study consisted of a retrospective 2-month baseline followed by a prospective 12-week treatment phase followed by a 1-week tapering-off phase. During the 12-week treatment phase patients were assigned to three treatment groups receiving ESL once daily (QD, once daily group), twice daily (BID, twice daily group), or placebo (PL, placebo group), respectively. Daily doses of ESL were increased at 4-week periods (400 mg, 800 mg, and 1,200 mg). The study design is presented in Fig 1. Investigational product was administered without regard for meals because a study in healthy subjects showed that presence of food has no effect on ESL PK (Maia et al., 2005). Concomitant antiepileptic drugs (AEDs) were continued unchanged. Tablets strengths of 200 mg, 400 mg, and 600 mg of ESL and respective placebos, manufactured by BIAL (S. Mamede do Coronado, Portugal) in accordance with Good Manufacturing Practice, were used.

Figure 1.

Study design diagram.

Patient population

Eligible patients were male and female aged 18–65 years with at least four partial-onset seizures per month in spite of treatment with one or two of the following AEDs: phenytoin, valproic acid, primidone, phenobarbital, lamotrigine, gabapentin, topiramate, and clonazepam in stable doses during at least 2 months prior to randomization.

Patients were not allowed to participate in case of any of the following exclusion criteria: vagus nerve stimulation; primarily generalized seizures; known progressive neurological disturbance; history of status epilepticus within the past 3 months; seizure of nonepileptic origin; restricted legal competence and incapability to follow trial instructions; major psychiatric disorders; concurrent drug therapy with monoamine oxidase inhibitors or calcium channel blockers; use of OXC or CBZ during the last 6 months before the randomization visit; known hypersensitivity to OXC or CBZ; abuse of alcohol, drugs, or medications; history of relevant medical disorder; second- or third-degree atrioventricular blockade not corrected with a pacemaker; abnormalities of sodium, hepatic function and white blood cell counts; pregnancy, nursing or inadequate contraception in women of childbearing potential; or participation in other clinical trials within the last 2 months.

Assessment procedures

Medical history and physical examination were performed at screening and updated at each visit. Blood pressure and heart rate measurements, neurological examination, and ECG recordings were obtained at each visit. Clinical laboratory tests (hematology, coagulation, plasma biochemistry, and urinalysis) were performed at screening and each of the following visits. Clinical laboratory tests were performed by a central laboratory. A pregnancy test was performed at screening and follow-up visits.

Patient and/or caregiver recorded the number and type of seizures in patient diaries throughout the study. Adverse events were obtained by means of questioning, patient's physical and neurological examination, reviewing of laboratory test results, and ECG parameters. Abnormalities at physical and neurological examination were commented by the investigator and clinically relevant changes, compared to baseline, were documented as adverse events. Changes of blood pressure, pulse rate, and ECG parameters, as compared to baseline, which were judged as being of clinical relevance by the investigator, were documented as adverse events. Patients were specifically questioned about the following side effects—ataxia, dizziness, fatigue, nausea, and somnolence.

Blood sampling

Blood samples (7 ml) for drug plasma assay were taken at predose before the daily dose (for “trough” concentrations assay). Blood samples were drawn into lithium heparin tubes and centrifuged at approximately 1,500 g for 10 min at 4°C. The resulting plasma was separated into two equal aliquots of 1 ml and stored at –20°C until required for analysis.

Assay of ESL and BIA 2-005 concentrations in plasma

Plasma concentrations of ESL and BIA 2-005 were determined using solid-phase extraction followed by high-performance liquid chromatography with mass spectrometric detection (LC-MS), as previously reported (Fontes-Ribeiro et al., 2005).

Statistical considerations

The intention-to-treat (ITT) population included all randomized patients with at least one administration of study medication. The per protocol (PP) population included all patients who had completed the study and who had no major protocol violation. The proportion of responders (patients with a 50% or greater reduction in seizure frequency) in the 12-week treatment period compared to the baseline period was previously defined as the primary efficacy end point. Testing for superiority of the test drug versus placebo with regard to the proportion of responders was done for the ITT population using the one-sided t-test (test of superiority). If appropriate, the normal approximation to the binomial distribution for the difference in proportions was used. Alternatively, exact methods were used. Additionally, the appropriate 90% confidence intervals (90% CI) were calculated. An exploratory analysis of the primary end point was also performed for the PP population. Data were interpreted assuming a significance level of α= 0.05.

The estimation of the sample size (the number of patients per group) was based on the following assumptions: significance level α= 0.1 (one sided), a power of 80%, and an equivalence region of 20% with regard to the proportion of responders in the treatment groups. For a proportion of responders of 15% in the PL group, a sample size of 39 patients in the PL group and 78 patients in the ESL groups was determined. Assuming an overall dropout rate of approximately 20%, it was estimated to enroll about 135 patients in order to obtain 117 evaluable patients.

Secondary efficacy variables included the reduction in total seizure frequency at each 4-week periods compared to baseline and the proportion of seizure-free patients. Patients who prematurely discontinued the participation in the study were not replaced. Their data in efficacy analyses were treated using “last observation carried forward” (LOCF) method.

Safety variables included the reported adverse events, relevant changes in clinical laboratory parameters, vital signs, and ECG parameters compared to baseline, and the proportion of patients experiencing ataxia, dizziness, fatigue, nausea, and somnolence.

The statistical package SAS (version 8.2, SAS Institute, Inc., Cary, NC, U.S.A.) was used for the statistical analysis.

RESULTS

Population

A total of 144 patients were enrolled in the study and 113 patients completed the 12-week treatment period. Disposition of patients is displayed in Fig. 2 and demographics are presented in Table 1. All patients were white. There were relatively more female patients in the BID group but results showed no gender-related statistical difference. Treatment groups were homogeneous with regard to age, height, and weight. Complex partial and partial evolving to secondary generalized seizures were more prevalent. At baseline, the mean (range) number of seizures per month was 14.1 (4–106), 13.6 (4–128), and 11.8 (4–120) in the QD, BID, and PL groups, respectively. Mean (SD) duration of epilepsy was 16.7 (11.7), 19.5 (12.6), and 20.0 (13.6) years in the QD, BID, and PL groups, respectively. Patients taking two concomitant AEDs were 70%, 61%, and 68% in the QD, BID, and PL groups, respectively; the remaining patients were taking just one concomitant AED. Valproic acid, the most concomitant used AED, was taken by 68%, 61%, and 66% of patients in the QD, BID, and PL groups, respectively. Topiramate was used by 36%, 35%, and 21%; lamotrigine by 30%, 28%, and 32%; and clonazepam by 14%, 17%, and 15% of patients, in the QD, BID, and PL groups, respectively. Relatively more patients were taking phenytoin in the PL group: 17% versus 2% and 4% in the QD and BID groups, respectively.

Figure 2.

Disposition of patients and patient populations.

Table 1. Patient baseline characteristics: ITT population (n = 143)
 Eslicarbazepine acetate QDEslicarbazepine acetate BIDPlacebo
  1. ITT, intent-to-treat population; QD, once daily; BID, twice daily; BMI, body mass index.

Number of patients504647
Gender (%)
 Male44.034.842.6
 Female56.065.257.4
Age (years)
 Mean ± SD39.3 ± 11.439.8 ± 11.940.4 ± 10.8
 Range19–5919–6119–58
Height (cm)
 Mean ± SD168 ± 9  169 ± 11 168 ± 8  
 Range152–187154–192152–188
Weight (kg)
 Mean ± SD70 ± 1372 ± 1773 ± 14
 Range45–11945–11651–115
BMI (kg/m2)
 Mean ± SD25 ± 4 25 ± 5 26 ± 4 
 Range17–3917–3820–39
Seizure type, n (%)
 Simple partial17 (34.0)17 (37.0)13 (27.7)
 Complex partial36 (72.0)33 (71.7)38 (80.9)
 Partial evolving to secondary generalized40 (80.0)37 (80.4)34 (72.3)

Efficacy results

Primary variable

The proportion of responders was compared between active treatment and placebo groups. Two types of statistical evaluation of the differences between the treatment groups were performed: one-sided 90% CI for the superiority of active treatment over placebo was calculated and testing of the statistical significance of the superiority was performed by continuity-adjusted χ2 test. In cases of differences of borderline statistical significance (as indicated by contradictory implications of 90% CI and continuity-adjusted χ2 test result), the result of continuity-adjusted χ2 test was accepted as more accurate.

The primary analysis of the main variable (percentage of responders at the end of the 12-week treatment period vs. baseline in the ITT population) showed a statistically significant difference between QD and PL groups (54% vs. 28%; 90% CI =–∞, –14; p = 0.008) (Fig. 3); the difference between the BID and PL group did not reach statistical significance (41% vs. 28%; 90% CI =–∞, –1; p = 0.12). An exploratory analysis of the primary variable in PP population showed similar results: a statistically significant difference between QD and PL groups (61% vs. 38% of responders; 90% CI =–∞, –8; p = 0.04), without significant differences between the BID and PL groups (53% vs. 38%; 90% CI =–∞; 1; p = 0.17). A significantly higher proportion of responders in the period of weeks 5–8 (800 mg/day) was found in the QD group when compared with the BID group (58.0% vs. 33%, respectively, p = 0.022, χ2 test with adjustment for continuity).

Figure 3.

Proportion of responders by treatment group (ITT population, 143 patients).

Secondary analyses of the main variable (percentage of responders) were performed for the treatment periods of weeks 1–4 (400 mg/day or placebo) and weeks 5–8 (800 mg/day or placebo) in the ITT population. For weeks 1–4, no significant differences in the proportion of responders (42%, 35%, and 34% in QD, BID, and PL groups, respectively) were found. For weeks 5–8, the proportion of responders reached 58% in the QD group versus 38% in the PL group (90% CI =–∞, –7; p = 0.04), while there was no statistical difference between the BID (33%) and PL groups (90% CI =–∞, 18; p = 0.36).

Secondary variables

Reduction in seizure frequency during each 4-week period compared to the baseline was calculated. As shown in Fig. 4, the greatest relative decrease in the seizure number was obtained in patients in the QD group between weeks 5–8 (800 mg/day) and between weeks 9–12 (1,200 mg/day). At all 4-week periods, the QD group showed a significantly higher reduction in the seizure number than the reduction observed in corresponding periods of the PL group (p = 0.037, p = 0.018, and 0.002, respectively; Wilcoxon rank sum test). During the 12-week treatment phase, the number of seizure−free patients increased in both ESL treatment groups (QD and BID groups) (Fig. 5). Fisher's exact test analysis showed that on the last 4 weeks of the treatment period there were statistically significantly more seizure−free patients in both QD and BID groups as compared to the PL.

Figure 4.

Evolution of the percentage of seizure reduction in relation to baseline, in the three 4-week treatment periods (ITT population, 143 patients).

Figure 5.

The percentage of seizure-free patients, in the three 4-week treatment periods (ITT population, 143 patients).

Pharmacokinetic results

Plasma levels of BIA 2-005, the active moiety found in plasma following administration of ESL, were assayed at trough (i.e., predose), at the end of every 4-week treatment intervals. Table 2 presents the mean BIA 2-005 plasma levels together with their 95% CI. Trough BIA 2-005 levels were proportional to dose administered and, as expected, were higher at the end of the dosing interval (12 h) of the BID group than at the end of the dosing interval (24 h) of the QD group. However, differences between groups were not statistically significant. ESL plasma levels were found to be below the LOQ (0.05 mg/L).

Table 2. BIA 2-005 (a mixture of ∼95% S-MHD and ∼5% R-MHD %) mean trough plasma concentration (Cmin) and its 95% confidence interval (95% CI) obtained following repetitive dosing of ESL 400, 800, and 1,200 mg once daily (QD) or twice daily (BID)
Daily doseMean Dosage regimenconcentration (mg/L)95% CI (mg/L)
400 mg/day 400 mg QD (n = 47)3.92.9–5.0
 200 mg BID (n = 41)4.94.1–5.7
800 mg/day 800 mg QD (n = 45)10.7  8.4–13.0
 400 mg BID (n = 37)13.5 10.0–16.9
1,200 mg/day1,200 mg QD (n = 43)  14.6 11.6–17.6
 600 mg BID (n = 38)15.5 12.3–18.7

Tolerability

Out of the 143 randomized patients who received at least one dose of ESL, 59 patients (41.3%) reported a total of 167 adverse events. The number of patients reporting adverse events was similar in the three treatment groups: 19 (38.0%) in QD, 19 (41.3%) in BID, and 21 (44.7%) in PL groups, respectively. A similar number of adverse events was observed in the QD and the PL groups, and a higher number in the BID group: 51, 50, and 66, respectively. Most of the adverse events—114 (68.3% of all adverse events)—were of mild intensity. Forty-seven adverse events (28.1%) were assessed as being of moderate intensity and only 6 (3.6%) were reported to be severe: one in the QD group, one in the BID group, and four in the PL group. The most often reported adverse event was headache (16 cases) followed by dizziness (10 cases) and nausea (9 cases). Table 3 displays those adverse events that were classified by the investigator as possibly/probably/definitely related with treatment.

Table 3. Summary of treatment-emergent adverse events; number (%) of patients
 Eslicarbazepine acetate QDEslicarbazepine acetate BIDPlacebo
  1. Some subjects reported more than one adverse event.

  2. QD, once daily; BID, twice daily.

Number of patients504647
Patients with any adverse event19 (38.0%)19 (41.3%)21 (44.7%)
Patients discontinued due to adverse events3 (6.0%)4 (8.7%)4 (8.5%)
Patients discontinued due to seizure frequency increase0 (0.0%)2 (4.3%)1 (2.1%)
Adverse events considered possibly related to treatment occurring in more than one patient
 Headache1 (2.0%)4 (8.7%)1 (2.1%)
 Dizziness1 (2.0%)3 (6.5%)0 (0.0%)
 Nausea2 (4.0%)3 (6.5%)2 (4.3%)
 Somnolence3 (6.0%)1 (2.2%)0 (0.0%)
 Vomiting0 (0.0%)3 (6.5%)0 (0.0%)
 Hair loss1 (2.0%)2 (4.3%)0 (0.0%)
 Dry mouth0 (0.0%)2 (4.3%)0 (0.0%)
 Concentration impaired0 (0.0%)2 (4.3%)0 (0.0%)
 Insomnia0 (0.0%)0 (0.0%)2 (4.3%)
 CK increase1 (2.0%)0 (0.0%)1 (2.1%)
 Drowsiness1 (2.0%)0 (0.0%)1 (2.1%)

No death occurred during the study. Three serious adverse events were reported during the study: one case of viral gastroenteritis, one case of bacterial gastroenteritis and one case of ischemic stroke in the right middle cerebral artery territory. These three serious adverse events were considered not related to drug treatment by the investigator. Fourteen patients discontinued the participation in the study prematurely because of the occurrence of an adverse event (AE) (3 patients in the QD group, 4 patients in the BID group, and 4 patients in the PL group) or exacerbation of seizures (0 patient in the QD group, 2 patients in the BID group, and 1 patient in the PL group).

Only three cases of abnormal laboratory results were assessed as clinically relevant and could have causal relationship to the administration of ESL: mild anemia, mild blood electrolyte imbalance, and moderately elevated creatine kinase level (all in the QD group). At admission in the study and at the end of the 12-week treatment phase, mean (SD) serum sodium levels were, respectively, 141.6 (2.66) and 140.2 (4.45) mmol/L in the QD group, 141.7 (3.24) and 139.6 (5.09) in the BID group, and 141.0 (3.28) and 141.1 (2.37) in the PL group. There were no clinically relevant overall trends in the vital signs or ECG during the study.

DISCUSSION

ESL is a new VGSC blocker with the potential to become a new central nervous system (CNS) drug for the treatment for epilepsy, bipolar disorder, and neuropathic pain. Studies in healthy subjects showed that ESL was safe and well tolerated. In humans, ESL was shown to be a prodrug to eslicarbazepine (S-licarbazepine or S-MHD), the active entity following oral administration of ESL (Almeida et al., 2005a).

The present phase II clinical trial is the first study of ESL in patients with partial-onset seizures refractory to treatment with 1 or 2 standard AEDs. Daily doses from 400 mg to 1,200 mg were tested against placebo, administered daily in one or two equally divided doses. The study dosage regimens (QD and BID) were chosen based on ESL PK data obtained in phase I studies. The present study showed that ESL was an effective add-on therapy in refractory epileptic patients. The results suggest that a once-daily regimen is as (or possibly more) efficacious as the same daily dose given twice a day. The once-daily dosing allows higher peak plasma concentration (Cmax) of the ESL active entity (MHD or BIA-2-005) than the twice-daily dosing of the same daily dose, which suggests that efficacy following once-daily ESL treatment correlates better with the Cmax (i.e., rate of exposure) than with exposure (AUC). However, taking into account the limited sample size of the study, this observation needs confirmation.

In randomized placebo-controlled studies of AEDs for epilepsy, the reported responder rate in the placebo arms usually ranges between 9.3% and 16.6% (Burneo et al., 2002). A relatively higher placebo-responder rate was observed in our study. Several causes may be responsible for this outcome. The relatively small sample size may be a contributing factor, because a slight variation in the number of patients with response to placebo represents a marked variation in the placebo-responder rate. Niklson et al. (2006) reported that the placebo response is higher in patients on only one AED, with later age at onset and shorter duration of epilepsy than in patients with more than one baseline AED, earlier age at onset, and longer duration of epilepsy. However, in the present study no differences between treatment groups were found concerning epilepsy time of diagnosis and number of concomitant AEDs. The lone remarkable difference between groups was the percentage of patients taking phenytoin (17% in the PL group vs. 2% and 4% in the QD and BID groups, respectively). It is expected that ongoing confirmatory studies in larger samples of patients will allow a better characterization of the ESL place in therapy.

The safety profile of the ESL was favorable. No clear dose relationship could be found in the profile of most adverse events. Hyponatremia is a relatively frequent laboratory abnormality with OXC and CBZ. With OXC, the reported overall incidence of serum sodium <135 mmol/L is 24.5% and <125 mmol/L is 3.0%, with an incidence of serum sodium levels <125 mmol/L of 3.8% in patients aged 18–64 years, and of 7.3% in patients over 65 years (Schmidt and Sachdeo, 2000). Hyponatremia associated with CBZ has also been well described and the incidence ranges from 1.8% to 40% depending on the patient population studied (Kuz and Manssourian, 2005). In the present study, just one patient (1.0%) among those treated with ESL showed a decrease in the sodium levels to <125 mmol/L. At randomization, this patient presented a 133 mmol/L sodium level, which is below the lower limit of the normal range (135 mmol/L) and following 12 weeks of treatment sodium decreased to 123 mmol/L, without clinical signs. In the study population, mean sodium levels remained unchanged during treatment with ESL.

In conclusion, ESL showed to be efficacious and well tolerated as an adjunctive therapy of partial-onset seizures refractory to treatment with one or two AEDs.

Appendix

Participant investigators and their respective sites

Croatia: Tomislav Babic (KBC Rebro, Kispaticeva 12, 10000 Zagreb), Danilo Hodoba (Psychiatric Hospital, Vrapce, 10090 Zagreb) and Ksenija Willheim (KBC Rijeka, Camberieva 17, 51000 Rijek); Czech Republic: Ivan Rektor (Neurologická klinika FN u sv. Anny, Pekařská 53, 656 91 Brno), Ladislav Pazdera (Neurologické oddělení Polikliniky, Jiráskova 1389, 516 01 Rychnov nad Kněžnou), Svojmil Petránek (Neurologické oddělení FN Bulovka, Budínova 2, 180 00 Praha 8), Karel Urbánek (Neurologická klinika FN Olomouc, I.P.Pavlova 6, 775 20 Olomouc) and Gerhard Waberzinek (Neurologická klinika FN Hradec Králové, Nezvalova 265, 500 02 Hradec Králové); Germany: Christian E. Elger (Department of Epileptology, University of Bonn, Sigmund Freud Str. 25, 53105 Bonn); Lithuania: Dalius Jatuzis (Clinic of Neurology Vilnius University Hospital “Santariskiu Klinikos,” Santariskiu 2, 2021, Vilnius), Aloyzas-Algimantas Skaringa (Department of Neurology Vilnius Mykolo Marcinkeviciaus Hospital, Kauno 7/2, 2006 Vilnius) and Danguole-Marija Vysniauskiene (Department of Neurology Panevezys Hospital, Smelynes 25, 5300 Panevezys); Poland: Piotr Czapiński (Centrum Leczenia Padaczki i Migreny, Ul. Kielecka 25, 31-523 Kraków), Anna Członkowska (Instytut Psychiatrii i Neurologii w Warszawie II Klinika Neurologii, Ul Sobieskiego 9, 02-957 Warszawa), Zofia Kazibutowska-Zaranska (Samodzielny Publiczny Szpital Kliniczny nr 7 Górnośla̧skie Centrum Medyczne ślAM, Klinika Neurologii, Ul. Ziołowa 45/47, 40-635 Katowice), Andrzej Klimek (Specjalistyczny Szpital im. M. Kopernika, Ul. Pabianicka 62, 93-513 Łódź), Krystyna Pierzchała (śla̧ska Akademia Medyczna II Katedra i Klinika Neurologii, ul 3 maja 13/15, 41-800 Zabrze), Zbigniew Stelmasiak (Samodzielny Publiczny Szpital Kliniczny nr 4 Katedra i Klinika Neurologii Akademii Medycznej, Ul. Jaczewskiego 8, 20-954 Lublin), and Jacek Gawłowicz (Wojewódzki Szpital Specjalistyczny Poradnia Przeciwpadaczkowa, Al. Kraśnicka 100, 20-718 Lublin).

Acknowledgments

Acknowledgments:  We acknowledge the staff of SCOPE International for their committed involvement in the study management. This study was sponsored by BIAL (Portela & C SA).

Ancillary