SEARCH

SEARCH BY CITATION

Keywords:

  • Adjunctive therapy;
  • Partial seizures;
  • Randomized controlled trial;
  • Refractory epilepsy;
  • Zonisamide

Abstract

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgments
  7. REFERENCES

Summary: Purpose: To evaluate the safety and efficacy of zonisamide (ZNS) as adjunctive treatment in patients with refractory localization-related epilepsy.

Methods: This was a double-blind, placebo-controlled study of adjunctive ZNS in 351 patients with refractory partial seizures receiving a stable regimen of one to three antiepileptic drugs (AEDs). Patients were randomized to placebo or ZNS, 100 mg, 300 mg, or 500 mg/day (2:1:1:2) after a 12-week baseline. Dose titration was undertaken over a 6-week titration phase, which was followed by an 18-week fixed-dose assessment phase. Primary efficacy parameters were the differences between ZNS, 500 mg/day, and placebo in the change from baseline in frequency of complex partial (CP) seizures during the fixed-dose assessment phase and in the proportion of CP responders (≥50% decrease from baseline in seizure frequency). Safety and tolerability also were assessed.

Results: Compared with placebo, the highest dose of ZNS (500 mg/day) resulted in a significantly greater decrease in CP seizure frequency from baseline (51.2% vs. 16.3%; p < 0.0001) and a significantly higher proportion of CP responders (52.3% vs. 21.3%; p < 0.001). Both ZNS, 500 mg/day, and 300 mg/day were statistically superior to placebo in reducing the frequency of “all seizures” and simple partial (SP) + CP seizures. For all seizures, a significant dose–response relation was observed (p < 0.0001).The most common adverse events were somnolence, headache, dizziness, and nausea during the titration phase and headache and pharyngitis during the fixed-dose assessment phase.

Conclusions: ZNS provides dose-dependent, effective, and generally well-tolerated adjunctive therapy in patients with partial seizures.

An antiepileptic drug (AED) selected as adjunctive therapy should possess a range of pharmacologic actions and demonstrable efficacy against diverse types of seizures coupled with a good tolerability and safety profile.

Zonisamide (ZNS) has multiple pharmacodynamic actions that may contribute to the antiepileptic effects observed in animal models (1). These include blockade of voltage-dependent T-type calcium channels (2) and voltage-sensitive sodium channels (3), blockade of potassium-evoked glutamate response (4), reduction of glutamate-mediated synaptic excitation (5), and increased γ-aminobutyric acid (GABA) release (6). These observations suggest that ZNS may have a broad spectrum of action. Indeed, benefit has been demonstrated in the treatment of various seizure types including partial seizures, myoclonic seizures, and other difficult-to-treat epilepsies (7,8). ZNS is licensed in the United States for the adjunctive treatment of partial seizures in adults, and in Japan, as monotherapy and as adjunctive therapy for children and adults with both generalized and partial seizures. Worldwide experience with ZNS approximates to 2 million patient-years of exposure.

In placebo-controlled clinical studies in patients with refractory partial seizures, ZNS has demonstrated efficacy when used as adjunctive therapy (9–11). However, the steady-state treatment periods in these studies were ≤8 weeks, and the dosing regimens differed among the trials. Two studies evaluated median doses of ∼400 mg/day that could be adjusted based on clinical response (9,10), whereas the third study used a complex design to evaluate doses of 100 mg, 200 mg, and 400 mg/day (11). Accordingly, the safety and efficacy of ZNS across a range of doses remains to be confirmed in randomized studies involving longer duration of treatment.

The primary objective of this 24-week, randomized, double-blind study was to evaluate the efficacy and safety of ZNS, 500 mg/day, compared with placebo when used in combination with existing AEDs in patients with refractory partial seizures; secondary objectives included evaluation of the safety and efficacy of ZNS, 100 mg/day and 300 mg/day, and assessment of a dose–response relation with ZNS.

METHODS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgments
  7. REFERENCES

Patients

Eligible patients were at least 12 years old with partial seizures with or without secondary generalization unsatisfactorily controlled despite a stable regimen of one to three AEDs. Seizures were classified according to International League Against Epilepsy (ILAE) criteria (12) into simple partial (SP) seizures, complex partial (CP) seizures, and partial seizures with secondary generalization (SGS). Generalized seizures and unclassified seizures also were recorded and counted in the “all seizures” category. Documented evidence of a surface electroencephalogram was required to support the clinical diagnosis of localization-related epilepsy. Preexisting results from appropriate brain magnetic resonance imaging or computerized tomography scans were needed to exclude patients with progressive neurologic disease. Patients were otherwise in general good health.

Patients with a history of nonepileptic seizures, alcoholism, drug abuse, or significant drug sensitivity were excluded, as were those with progressive neurologic disease. Concurrent treatment with monoamine oxidase inhibitors, antidepressants, or antipsychotics was not permitted from 14 days before screening (or 5 half-lives of the parent drug and/or active metabolite, whichever was longer). Pregnant and lactating women also were excluded, and female patients of childbearing potential were required to use an adequate form of contraception.

The study protocol and related patient information were approved by independent ethics committees before study commencement. Patients provided written informed consent in accordance with the International Conference on Harmonisation Good Clinical Practice Guidelines. If the patient was younger than 18 years, or 16 years in some countries, written informed consent was obtained from the patient's legal guardian.

Study design

This was a multicenter, randomized, double-blind, placebo-controlled, parallel-group study to assess the efficacy and safety of ZNS as adjunctive therapy in patients with refractory partial seizures. The study was conducted at 54 centers, across 18 European countries (49 centers) and South Africa (five centers). After screening, patients underwent a 12-week prospective baseline period during which they recorded all seizures on diary cards. Patients were included if they experienced ≥12 partial seizures, with no more than a 3-week seizure-free interval, during this baseline period.

In total, 351 patients were randomized to receive one of four treatments (placebo; ZNS, 100 mg/day; ZNS, 300 mg/day; or ZNS, 500 mg/day) in the ratio of 2:1:1:2. Patients were randomized sequentially in blocks of six. Treatments were blinded by using a double-dummy technique throughout the study (i.e., all patients received the same number of capsules in any particular phase).

Randomized patients entered a 6-week dose-titration phase (weeks 13–18). The titration schedule is shown in Fig. 1. Study medication was given in divided doses in the morning (a.m.) and evening (p.m.). All those receiving active treatment were given 25 mg (a.m.) + 25 mg (p.m.) ZNS on days 1–3, increasing to 50 mg + 50 mg on days 4–7. This concluded dose escalation for patients in the ZNS, 100 mg/day, group. The dose was increased at weekly intervals for patients in the 300- or 500-mg/day groups, reaching 100 mg + 200 mg daily in week 16. This concluded dose escalation for patients in the ZNS, 300 mg/day, group. Dose increments were continued for patients in the 500-mg/day group, reaching 200 mg + 300 mg daily in week 18.

image

Figure 1. Study design. ZNS, zonisamide; AEDs, antiepileptic drugs.

Download figure to PowerPoint

Patients were required to continue on these doses for a further 18-week fixed-dose assessment phase (weeks 19–36), during which they attended clinic visits at 4-week intervals for efficacy and safety evaluations. After this phase, patients were provided with appropriate blinded medication and either entered an open-label extension study or entered a 4-week, controlled-withdrawal down-titration period. Patients in the extension study were maintained on or titrated to ZNS, 500 mg/day.

Study measures

Patients were instructed to record all seizures by date and type in a diary throughout the study. Seizure-frequency data from the diaries were normalized to 28-day periods.

The two primary efficacy parameters were (a) the median percentage change in frequency of CP seizures without secondary generalization, from baseline to the fixed-dose assessment phase, normalized to a 28-day period; and (b) the proportion of responders (defined as patients with a ≥50% decrease from baseline in CP seizure frequency) during the fixed-dose assessment phase, normalized to a 28-day period. Median percentage changes from baseline in the frequency of all partial seizures without secondary generalization (defined as SP+CP seizures) and the frequency of all seizures (defined as SP+CP seizures, secondarily generalized seizures, generalized and unclassified seizures) as well as the corresponding responder rates were evaluated as secondary variables. The number of seizure-free days per 28-day period also was assessed.

Safety was assessed by evaluation of treatment-emergent adverse events (AEs), clinical laboratory tests, electrocardiogram (ECG) results, physical and neurologic examinations, vital signs, and weight. AE data were collected throughout the study. Laboratory tests were performed on blood and urine samples taken at screening, week 12 (end of prospective baseline phase), and at 4-weekly intervals from week 16 to week 36. ECGs were carried out at screening (or week 4 at latest) and week 40 (discontinuation visit). Neurologic and physical examinations, including vital signs and weight, were carried out at screening and at weeks 20, 36, and 40.

Samples for determination of ZNS serum concentrations were obtained periodically throughout the study. These data were analyzed by using a population pharmacokinetic approach, and the results are not reported here.

Statistical analysis

All patients who received at least one dose of study drug were included in the safety analysis. Primary and secondary efficacy variables were assessed for three populations: the intent-to-treat (ITT) population (all patients who received at least one dose of ZNS, and who had postbaseline partial seizure frequency data); the primary efficacy-analysis population (all patients in the ITT population with partial seizure frequency data collected during the fixed-dose assessment phase); the efficacy-evaluable population (all patients in the ITT population who averaged four seizures/month during the 12-week baseline phase and took ≥10 weeks of study medication, including ≥2 weeks from the fixed-dose assessment phase). As stated in the protocol, the primary efficacy analysis was conducted by using the primary efficacy-analysis population; sensitivity analyses were performed by using the ITT and efficacy-evaluable populations. Subjects must have had data available at the time point under evaluation to be included in the efficacy analysis; no imputation methods were used to include missing data. Subjects were excluded from analyses of change from baseline in CP or SP+CP seizure frequency if they did not have the appropriate seizure type, but they were included in the denominator for the analyses of responders.

The sample-size calculation was based on the expected proportion of responders in the primary efficacy-analysis population (rather than median percentage change in seizure frequency from baseline) because this parameter was less sensitive to treatment differences and, therefore, required more subjects to achieve the specified 80% power. A two-group Pearson χ2 test with a 0.05 two-sided significance level had ≥80% power to detect a difference between response proportions of 0.17 in the placebo group and 0.34 in the ZNS, 500 mg/day, group when the sample size in each group was 106. Allowing for the 2:1:1:2 randomization scheme, the total sample size required was 318 subjects. Assuming an 18% dropout rate after screening, the required number of subjects to be recruited was 388.

Statistical analyses were performed by using SAS. All hypothesis tests were two-tailed and performed at the 5% significance level, apart from interactions, which were tested at the 10% level of significance. Where appropriate, 95% confidence intervals (CIs) were determined. A hierarchical relation was defined such that the second primary efficacy parameter, the proportion of responders, was tested only if the first parameter, percentage change in seizure frequency from baseline, was statistically significant at the 5% level.

The primary efficacy analyses were determined by using the primary efficacy-analysis population. The first primary efficacy parameter (median percentage change from baseline in CP seizure frequency) was analyzed by using an analysis of variance (ANOVA) model on rank-transformed data. The model included the main effects of treatment and center group (centers pooled by country). A term for the interaction between treatment and center group was included if found to be statistically significant at the 10% level. A step-down procedure was used to compare each active treatment group with placebo in turn, starting with 500 mg/day versus placebo, to protect against multiplicity effects. If any of the comparisons did not achieve a p value of ≤0.05, then the next step-down comparison was not made. The difference between the medians with 95% CI and associated p value was calculated by the Hodges-Lehman test. The second primary efficacy parameter (proportion of responders for the CP seizure type) was assessed by using the Cochran–Mantel–Haenszel (CMH) test, with center group as the stratification variable. The primary efficacy analyses were repeated for analyses of all seizures and SP+CP seizures. In the analyses by seizure group, only patients with appropriate seizures during the baseline period could be included in the analysis. A Mantel–Haenszel χ2 test (13) for linear trend (using scores of 0, 1, 3, and 5) in the proportion of responders for all seizures was performed by using the ITT population. Seizure-free days were summarized by using descriptive statistics for the ITT population.

RESULTS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgments
  7. REFERENCES

Patients

Of the 444 patients who were screened, 351 were randomized in the 2:1:1:2 schedule, and 349 received at least one dose of study drug and were evaluable for safety (Table 1). The ITT population comprised 347 patients (119 placebo; 55 ZNS, 100 mg; 55 ZNS, 300 mg; 118 ZNS, 500 mg), as no postbaseline seizure-frequency data were available for four randomized patients. Of these, 312 patients (112 placebo; 54 ZNS, 100 mg; 45 ZNS, 300 mg; 101 ZNS, 500 mg) had partial seizure-frequency data collected during the fixed-dose assessment phase and so were included in the primary efficacy-analysis population, and 294 patients (109 placebo; 54 ZNS, 100 mg; 43 ZNS, 300 mg; 88 ZNS, 500 mg) met the criteria for inclusion in the efficacy-evaluable population.

Table 1. Patient dispositionThumbnail image of

Demographic characteristics and disease histories are shown in Table 2. In general, the groups were well balanced demographically. However, patients in the ZNS, 300 mg/day, group were younger than those in the other groups. These patients also had a shorter median time since epilepsy diagnosis (overall and for each seizure group), a substantially lower historic CP frequency, and a higher historic SP seizure frequency, compared with the other groups. Across all treatment groups, 75% of patients were receiving two or more concomitant AEDs at baseline. The use of three concomitant AEDs was more frequent in the ZNS, 100 mg/day and 500 mg/day, groups compared with the other groups. Across all treatment groups, the most common concomitant AEDs at baseline across treatment groups were carbamazepine (CBZ), valproic acid/sodium valproate (VPA), and lamotrigine (LTG) (Table 3). Most patients in each treatment group were taking an enzyme-inducing AED at baseline.

Table 2. Patient demographics and baseline characteristics (safety population)
  Placebo (n = 120)ZNS 100 mg/day (n = 56)ZNS 300 mg/day (n = 55)ZNS 500 mg/day (n = 118)
  1. AEDs, antiepileptic drugs; CP, complex partial; SG, secondarily generalized; SP, simple partial; SD, standard deviation; ZNS, zonisamide.

  2. aAlthough seven patients were reported as taking more than three AEDs, these were not considered protocol violators, as the fourth medication was used only as rescue therapy.

Gender: no (%)
 Male68 (56.7)30 (53.6)27 (49.1)53 (44.9)
 Female52 (43.3)26 (46.4)28 (50.9)65 (55.1)
Age at screening (yr)
 Mean36.536.132.936.1
 Range12–6412–6512–7312–77
Time since epilepsy onset (mo)
 Median254.0279.5188.0227.0
 Range22–5865–6727–67011–776
Seizure start date: median (range) (yr)
 CP seizure20.0 (0–48)16.0 (0–56)14.0 (3–55)16.0 (1–64)
 SP+CP seizures19.5 (0–48)16.0 (0–56)13.5 (0–55)16.0 (1–64)
 All seizures21.0 (1–48)22.5 (0–56)14.0 (0–55)17.0 (1–64)
Historic SP frequency/28 days
 Mean (SD)10.9 (44.5)9.2 (24.7)12.5 (32.0)9.5 (33.7)
 Median (range)  0.0 (0–459)  0.0 (0–139)  3.0 (0–212)  0.0 (0–318)
Historic CP frequency/28 days
 Mean (SD)12.5 (22.0)11.8 (2.1)  9.6 (13.0)12.7 (20.3)
 Median (range)  5.7 (0–153)  6.3 (0–121)  2.9 (0–56)  6.2 (0–119)
Historic SG frequency/28 days
 Mean (SD)2.7 (9.5)2.7 (5.5)2.4 (6.5)2.9 (8.6)
 Median (range) 0.0 (0–81) 0.0 (0–27) 0.0 (0–45) 0.0 (0–64)
Concomitant AEDs: N (%)
 00000
 122 (18.3)17 (30.4)15 (27.3)32 (27.1)
 268 (56.7)21 (37.5)27 (49.1)50 (42.4)
 326 (21.7)17 (30.4)13 (23.6)34 (28.8)
>3a4 (3.3)1 (1.8)02 (1.7)
Table 3. Concomitant antiepileptic drugs (%) taken by ≥10% of subjects in any treatment group at baseline (safety population)
  Placebo (n = 120)ZNS 100 mg/day (n = 56)ZNS 300 mg/day (n = 55)ZNS 500 mg/day (n = 118)
Carbamazepine80 (66.7)38 (67.9)30 (54.5)78 (66.1)
Clobazam14 (11.7) 6 (10.7) 3 (5.5) 12 (10.2)
Gabapentin10 (8.3)  4 (7.1)  7 (12.7)12 (10.2)
Lamotrigine34 (28.3)19 (33.9)19 (34.5)28 (23.7)
Phenobarbital 5 (4.2)  7 (12.5)3 (5.5)14 (11.9)
Phenytoin24 (20.0) 6 (10.7) 7 (12.7)21 (17.8)
Topiramate10 (8.3)  6 (10.7) 5 (9.1) 13 (11.0)
Valproate34 (28.3)15 (26.8)16 (29.1)31 (26.3)

Primary efficacy-analysis population

During the fixed-dose assessment phase, treatment with ZNS, 500 mg/day, produced a significantly greater median reduction in CP seizure frequency from baseline than did placebo [51.2% (n = 86 subjects included in the analysis) vs. 16.3% (n = 89)] (Fig. 2) for the primary efficacy-analysis population; the difference between ZNS, 500 mg/day, and placebo groups in the median percentage reduction from baseline in CP seizure frequency was 31.2% (95% CI, 15.7–44.6; p < 0.0001). The proportion of responders for CP seizures was significantly higher (p < 0.001) in the ZNS, 500 mg/day, group compared with placebo (52.3% vs. 21.3%; Fig. 3); the odds ratio (95% CI) for the ZNS, 500 mg/day, group relative to placebo was 4.07 (1.94–8.56). Similar statistically significant treatment differences in seizure-frequency reduction from baseline and in response rates also were found for both the ITT and the efficacy-evaluable populations (p < 0.002 in all cases) and are detailed later. Although relatively few patients were aged 12–17 years, similar proportions of responders existed for adolescents as for adult patients.

image

Figure 2. Median percentage reduction in seizure frequency from baseline with adjunctive zonisamide (primary efficacy-analysis population). SP, simple partial; CP, complex partial.

Download figure to PowerPoint

image

Figure 3. Proportions of responders to adjunctive zonisamide (primary efficacy-analysis population). SP, simple partial; CP, complex partial.

Download figure to PowerPoint

For all seizures and for SP+CP seizures, median baseline frequencies were broadly similar across treatment groups. The median percentage reduction in all seizure frequency from baseline was significantly greater than that with placebo [18.1% (n = 112)] for both ZNS, 500 mg/day [51.3% (n = 101); p < 0.0001] and 300 mg/day [41.8% (n = 45); p = 0.0005]. The median reduction in SP+CP seizure frequency from baseline also was significantly greater for both ZNS, 500 mg/day [50.6% (n = 99); p < 0.0001], and 300 mg/day [46.4% (n = 42); p = 0.0007] than for placebo [19.4% (n = 109)]. The median reduction in the frequency of all seizures and SP+CP seizures with ZNS, 100 mg/day, was not statistically different from placebo.

For all seizures, the proportion of responders was higher in each ZNS group (500 mg, 52.5%; 300 mg, 42.2%; 100 mg, 29.6%) than for placebo (17.9%). The treatment difference was significant (p < 0.001) for the ZNS, 500 mg/day, group compared with placebo; the odds ratio (95% CI) for the ZNS, 500 mg/day group relative to placebo was 4.63 (2.28–9.39). For SP+CP seizures, the proportion of responders also was higher in each ZNS group (500 mg, 50.5%; 300 mg, 42.9%; 100 mg, 28.8%) than in the placebo group (20.2%). Again, the treatment difference was significant (p < 0.001) for the ZNS, 500 mg/day, group compared with placebo; the odds ratio (95% CI) for the ZNS, 500 mg/day, group relative to placebo was 4.25 (2.01–8.95).

Patients treated with ZNS, 300 mg/day, showed a greater median reduction in CP seizure frequency from baseline than those who had received placebo [40.1% (n = 25) vs. 16.3% (n = 89)] in the primary efficacy population, but this difference was not statistically significant. Treatment with ZNS, 300 mg/day, also produced a higher CP seizure-responder rate than placebo (36.0% vs. 21.3%). The CP responder rate with ZNS, 100 mg/day, was similar to that of placebo.

ITT population

During the fixed-dose assessment phase, treatment with ZNS, 500 mg/day, produced a significantly greater median reduction in CP seizure frequency from baseline than did placebo [44.1% (n = 102) vs. 16.9% (n = 94)] for the ITT population; the difference between ZNS, 500 mg/day, and placebo groups in the median percentage reduction from baseline in CP seizure frequency was 22.3% (95% CI, 7.65–36.30; p = 0.0017). The proportion of responders for CP seizures was significantly higher (p < 0.001) in the ZNS, 500 mg/day, group compared with placebo (45.1% vs. 22.3%); the odds ratio (95% CI) for the ZNS, 500 mg/day, group relative to placebo was 3.04 (1.48–6.24).

For all seizures and for SP+CP seizures, mean baseline frequencies were similar across treatment groups. The median percentage reduction in all seizure frequency from baseline was significantly greater than placebo [17.4% (n = 119)] for both ZNS, 500 mg/day [46.1% (n = 118); p < 0.0001] and 300 mg/day [38.5% (n = 55); p = 0.0034]. The median reduction in SP+CP seizure frequency from baseline also was significantly greater for both ZNS, 500 mg/day [45.1% (n = 115); p = 0.0001] and 300 mg/day [41.4% (n = 52); p = 0.006] than for placebo [19.0% (n = 114)]. The median reduction in the frequency of all seizures and SP+CP seizures with ZNS, 100 mg/day, was not statistically different from placebo.

For all seizures, the proportion of responders was higher in each ZNS group (500 mg, 46.6%; 300 mg, 34.5%; 100 mg, 30.9%) than for placebo (17.6%). The treatment difference was significant (p < 0.001) for the ZNS, 500 mg/day, group compared with placebo; the odds ratio (95% CI) for the ZNS, 500 mg/day, group relative to placebo was 3.78 (1.91–7.49). For SP+CP seizures, the proportion of responders was also higher in each ZNS group (500 mg, 44.3%; 300 mg, 34.6%; 100 mg, 30.2%) than in the placebo group (20.2%). Again, the treatment difference was significant (p < 0.001) for the ZNS, 500 mg/day, group compared with placebo; the odds ratio (95% CI) for the ZNS, 500 mg/day, group relative to placebo was 3.21 (1.57–6.58).

Patients treated with ZNS, 300 mg/day, showed a greater median reduction in CP seizure frequency from baseline than did those who had received placebo [32.4% (n = 33) vs. 16.9% (n = 94)] in the ITT population, but this difference was not statistically significant. Treatment with ZNS, 300 mg/day, also produced a higher CP seizure responder rate than placebo (30.3% vs. 22.3%). The CP responder rate with ZNS, 100 mg/day, was similar to that of placebo.

The proportion of responders to ZNS showed an apparent dose–response relation for all seizures, SP+CP seizures, and CP seizures. The proportion of responders for all seizures was significantly associated with ZNS dose (p < 0.0001), as shown by linear trend analysis. This analysis showed that, compared with placebo, subjects in each ZNS group were more likely to show a response to treatment (observed odds ratio compared with placebo: 500 mg, 4.07; 300 mg, 2.46; 100 mg, 2.09).

The median number of seizure-free days per 28-day period increased from baseline by ∼3 days with ZNS, 300 mg/day, and ZNS, 500 mg/day, during the fixed-dose assessment period to ∼21 and 23 days, respectively. The increase in seizure-free days with ZNS, 100 mg/day, was similar to that with placebo (1.2 days). Six (5%) patients treated with ZNS, 500 mg/day, and two (2%) treated with placebo achieved freedom from all seizures throughout the titration and fixed-dose assessment phases.

Conversely, an increase in the frequency of all seizures by >25% was observed in only seven, four, and five patients (6%, 7%, and 9%) treated with ZNS, 500, 300, or 100 mg/day, respectively, compared with 19 (16%) patients treated with placebo.

Efficacy-evaluable population (primary end points)

During the fixed-dose assessment phase, treatment with ZNS, 500 mg/day, produced a significantly greater median reduction in CP seizure frequency from baseline than placebo [49.8% (n = 74) vs. 16.3% (n = 86)] for the efficacy-evaluable population; the difference between ZNS, 500 mg/day, and placebo groups in the median percentage reduction from baseline in CP seizure frequency was 29.0% (95% CI, 13.59–42.20; p = 0.0004). The proportion of responders for CP seizures was significantly higher (p < 0.001) in the ZNS, 500 mg/day, group compared with placebo (50.0% vs. 20.9%); the odds ratio (95% CI) for the ZNS, 500 mg/day, group relative to placebo was 3.82 (1.78–8.23).

Safety and tolerability

The majority of patients reported at least one AE: 82 (68.3%) patients in the placebo group and similar proportions of those receiving ZNS, 100 mg/day or 300 mg/day [38 patients (67.9%) and 39 patients (70.9%), respectively; Table 4). However, AEs were more common with the ZNS, 500 mg/day, dose [96 patients (81.4%)]. The majority of AEs in each treatment group were of mild or moderate severity.

Table 4. Summary (%) of adverse events (safety population)
  Placebo (n = 120)ZNS 100 mg/day (n = 56)ZNS 300 mg/day (n = 55)ZNS 500 mg/day (n = 118)
  1. AE, adverse event.

  2. aMost patients who completed the study entered a follow-up study and did not undergo down-titration.

Any AE82 (68.3)38 (67.9)39 (70.9)96 (81.4)
Titration60 (50.0)32 (57.1)34 (61.8)73 (61.9)
Fixed-dose assessment58 (48.3)29 (51.8)23 (41.8)59 (50.0)
Down-titrationa3 (2.5)1 (1.8)5 (9.1)4 (3.4)
AE leading to withdrawal12 (10.0)1 (1.8)10 (18.2)32 (27.1)
Titration7 (5.8) 0 7 (12.7)23 (19.5)
Fixed-dose assessment4 (3.3)1 (1.8)2 (3.6)10 (8.5) 
Down-titrationa2 (1.7) 01 (1.8) 0
Serious AE10 (8.3)  05 (9.1)9 (7.6)
Death2 (1.7) 0 0 0

Serious AEs had an incidence in the ZNS treatment groups (0 to 9.1%) that was either lower or similar to that in the placebo group (8.3%). Three deaths occurred during the study. One patient died before being randomized to receive study medication. Two patients, both in the placebo group, died of causes considered by the investigator to be unrelated to treatment (suffocation during an epileptic seizure and unknown cause of death). The incidence of withdrawals due to AEs was higher for the 300-mg/day and 500-mg/day groups (18.2% and 27.1%, respectively) than for the 100-mg/day and placebo groups (1.8% and 10.0%, respectively). The higher incidence of AEs leading to withdrawal in the 500-mg/day group compared with other groups was mainly because of more events of dizziness, difficulty concentrating, nausea, and somnolence during the titration phase.

Common AEs (defined as an incidence of ≥10% of patients in any treatment group; Table 5) were somnolence, dizziness, headache, and nausea during the titration phase and headache and pharyngitis during the fixed-dose assessment phase. Headache was more frequently reported in the placebo group than in the ZNS groups during the fixed-dose assessment phase. With the exception of somnolence during the titration phase, no AE had a >10% higher incidence in a ZNS treatment group than in placebo. For treatment-related AEs, only dizziness, with an incidence of ≤11%, was a common event during the titration phase. No common treatment-related AEs were found during the fixed-dose assessment phase. Safety data during down titration (Table 4) are limited, as most patients continued treatment in a follow-on extension study; no serious treatment-related AEs were reported during down-titration.

Table 5. Adverse events (%) reported by ≥10% of patients during the titration phase and during the fixed-dose assessment phase (safety population)
  Placebo (n = 120)ZNS 100 mg/day (n = 56)ZNS 300 mg/day (n = 55)ZNS 500 mg/day (n = 118)
  1. AE, adverse event; ZNS, zonisamide.

Titration phase
 Any AE60 (50.0)32 (57.1)34 (61.8)73 (61.9)
 Dizziness3 (2.5) 1 (1.8)  4 (7.3)14 (11.9)
 Headache8 (6.7) 4 (7.1)  7 (12.7) 8 (6.8) 
 Nausea7 (5.8) 2 (3.6)  6 (10.9) 9 (7.6) 
 Somnolence3 (2.5) 3 (5.4)  2 (3.6) 17 (14.4)
Fixed-dose assessment phase
 Any AE58 (48.3)29 (51.8)23 (41.8)59 (50.0)
 Headache15 (12.5) 6 (10.7) 6 (10.9) 5 (4.2) 
 Pharyngitis 3 (2.5)  6 (10.7) 1 (1.8)  3 (2.5) 

Median clinical chemistry or hematologic laboratory results were similar over time in each of the treatment groups. Slight changes in mean weight occurred during the study. At week 36, mean decreases in weight of 1.7 kg were observed with ZNS, 500 mg/day, and 1.0 kg with ZNS, 100 mg/day; small increases were seen in the ZNS, 300 mg/day (mean, 0.1 kg) and placebo (mean, 0.2 kg) treatment groups.

No clinically significant differences were found between the three doses of ZNS and placebo with regard to vital signs, physical or neurologic examination, and ECG results.

DISCUSSION

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgments
  7. REFERENCES

Results from this randomized, double-blind, placebo-controlled study demonstrate that ZNS, 500 mg/day, was significantly superior to placebo in reducing the frequency of CP seizures from baseline during the fixed-dose assessment phase and in increasing the proportion of responders. It should be noted that although the primary efficacy parameters concerned CP seizures, patients were enrolled on the basis of the frequency of their SP+CP seizures at baseline. Consequently, assessments of CP seizures could include only patients with CP seizures during the baseline period.

ZNS, 500 mg/day, also was significantly superior to placebo in reducing the frequency of all seizures and SP+CP seizures. ZNS, 300 mg/day, was significantly superior to placebo in reducing the frequency of all seizures and SP+CP seizures, and numerically, but not statistically, superior to placebo in reducing the frequency of CP seizures. This latter finding may be a result of the low CP seizure frequency at baseline in the ZNS, 300 mg/day group, compared with the other treatment groups and the smaller population size in this dose group compared with the 500-mg/day and placebo groups because of the unequal randomization scheme. The efficacies of ZNS, 100 mg/day, and placebo were generally similar for each seizure group.

Although it is difficult to compare studies directly because of differences in populations and protocols, ZNS produced higher responder rates in the present study with an 18-week assessment period than was observed in previous ZNS studies that had assessment periods of ≤8 weeks. The short-term ZNS studies showed responder rates for all seizures of 29.9% and 28.2% in two studies with variable ZNS doses of around 400 mg/day (9,10) and 42% in a third study that evaluated a fixed dose of ZNS, 400 mg/day (11). In the present study, odds ratios indicate that patients treated with ZNS, 500 mg/day, were ∼4 times as likely to have a response to treatment compared with those given placebo: this is higher than in a short-term ZNS study (14). Notably, a statistically significant dose–response relation was found for the all-seizures category in the present study (ITT analysis), indicating that the proportion of subjects with at least a 50% reduction of all seizures increased as the dose of ZNS increased.

Seizure freedom remains the goal for patients with epilepsy. Achievement of total freedom from seizures in relatively few patients treated with ZNS is not unexpected in the highly refractory patient population included in this study. Similar rates have been reported with topiramate (15) and levetiracetam (16). Assessments of seizure freedom in observational outcome studies that closely adhere to routine clinical practice will demonstrate the long-term utility of the newer AEDs (17).

It also is important to note that only 6–9% of patients had an increase in seizure frequency of >25% when ZNS was added as adjunctive therapy, compared with 16% of patients who received placebo. Older AED agents have been reported to have a potential to exacerbate certain types of seizures. This is considered to be less likely for ZNS because of its broad range of pharmacologic effects, and numerous reports indicate that ZNS is effective in other types of epilepsy including generalized seizures (18,19), infantile spasms (20,21), progressive myoclonus epilepsy and myoclonic syndromes (22–25), and absence seizures (26).

As with other AEDs, a balance between efficacy, safety, and tolerability must be achieved. ZNS was generally well tolerated in this study. As may be expected with any potent therapeutic agent, the numbers of patients reporting AEs and AEs leading to withdrawal increased at higher doses. During the titration phase, the dose was generally increased at 1-week intervals, and a slightly longer interval may increase tolerability for some patients. However, a low incidence of serious AEs was seen during the study in all treatment groups. Although serious rash or renal stones in patients receiving ZNS have been reported previously (16), these were not reported during this study. The most frequently reported AEs associated with ZNS were generally mild to moderate in severity, comprising somnolence, dizziness, headache, and nausea during the up-titration and headache and pharyngitis in the fixed-dose assessment phase. This relatively benign safety profile of ZNS is supported by extensive postmarketing data both from Japan (where ZNS has been in use since 1989 and is approved for use as monotherapy and adjunctive therapy in adults and children) and the United States (where ZNS was approved for adjunctive use in adults in 2000) and includes data from ∼1 million patients and 2 million patient-years of exposure.

Flexibility of dosing also is an important factor for any adjunctive AED therapy. ZNS has a long half-life in plasma or serum (∼60 h in healthy volunteers), although this is reduced by approximately one half in patients taking concurrent liver enzyme–inducing medication (27,28). The long half-life of ZNS minimizes any fluctuation in plasma or serum concentrations in the event of a missed dose. ZNS does not induce or inhibit cytochrome P450 3A4 and may, therefore, be added to any AED regimen without the need for dosage modification (28). Studies in patients with epilepsy confirm that ZNS has minimal effects on the pharmacokinetics of CBZ (29), PHT (30), VPA (31), or LTG (32).

In conclusion, the dose-dependent efficacy of ZNS as adjunctive therapy for partial seizures has been clearly demonstrated in this study. Despite having highly refractory epilepsy, a high proportion of patients responded to ZNS. In addition, effective doses were generally well tolerated. Because of the simplicity of combining ZNS with an existing AED schedule and its multiple modes of action, ZNS represents an appropriate and rational option for first-choice adjunctive therapy in patients with partial epilepsy who have not responded to monotherapy.

Acknowledgments

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgments
  7. REFERENCES

Acknowledgment:  Funding for the study and manuscript preparation were provided by Elan Pharmaceuticals. We are grateful to the support of our co-investigators on the study: Tomislav Babic (Croatia), Attila Balogh (Hungary), Elinor Ben-Menachem (Sweden), Bernard Boothman [United Kingdom (U.K.)], Valmantas Budrys (Lithuania), James Butler (Republic of South Africa), Pamela Crawford (U.K.), Piotr Czapinski (Poland), Norman Delanty (Ireland), Vida Demarin (Croatia), Richard Grunewald (U.K.), Stephen Howell (U.K.), Andriy Dubenko (Ukraine), Yevgen Dubenko (Ukraine), Jacek Gawowicz (Poland), Gilles Geraud (France), Pierre Marie Gonnaud (France), Peter Halasz (Hungary), Yngve Hallstrom (Sweden), Lyubomir Haralanov (Bulgaria), Edouard Hirsch (France), Jouko Isojarvi (Finland), Simon Kesler (Republic of South Africa), Aleksandrs Kovaldins (Latvia), Edward Bernard Lee Pan (Republic of South Africa), Martine Lemesle (France), Thomas Mayer (Germany), Dimitar Minchev (Bulgaria), Ludmyla Muzychuk (Ukraine), Ragnar Palm (Sweden), Markku Partinen (Finland), Claude Remy (France), Josemir Sander (U.K.), Valentin Sander (Estonia), Bettina Schmitz (Germany), Vincenzo Sgro (Italy), Mohammad Sharief (U.K.), David Smith (U.K.), Johannes Albertus Smuts (Republic of South Africa), Zbigniew Stelmasiak (Poland), Pilla Taba (Estonia), Carlo Tassinari (Italy), Dimitar Tchakarov (Bulgaria), Nerija Vaiciene (Lithuania), Wim Van Paesschen (Belgium), F. Verster (Republic of South Africa), Egils Vitols (Latvia), Elaine Wilson (U.K.), and Stephen Wroe (U.K.).

REFERENCES

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgments
  7. REFERENCES
  • 1
    Macdonald RL. Mechanisms of action. In: LevyRH, et al., eds. Antiepileptic drugs. 5th ed. Philadelphia : Lippincott Williams & Wilkins, 2002: 86772.
  • 2
    Suzuki S, Kawakami K, Nishimura S, et al. Zonisamide blocks T-type calcium channel in cultured neurons of rat cerebral cortex. Epilepsy Res 1992;12: 217.DOI: 10.1016/0920-1211(92)90087-A
  • 3
    Schauf CL. Zonisamide enhances slow sodium inactivation in Myxicola. Brain Res 1987;413: 1858.DOI: 10.1016/0006-8993(87)90168-5
  • 4
    Okada M, Kawata Y, Mizuno K, et al. Interaction between Ca2+, K+, carbamazepine and zonisamide on hippocampal extracellular glutamate monitored with a microdialysis electrode. Br J Pharmacol 1998;124: 127785.
  • 5
    Zhu W, Rogawski MA. Zonisamide depresses excitatory synaptic transmission by a presynaptic action [Abstract]. Epilepsia 1999;40(suppl 7):245.
  • 6
    Kawai M, Hiramatsu M, Endo A, et al. Effect of zonisamide on the release of aspartic acid and gamma-aminobutyric acid from the hippocampal slices of E1 mice. Neurosciences 1994;20: 1159.
  • 7
    Jain KK. An assessment of zonisamide as an anti-epileptic drug. Expert Opin Pharmacother 2000;1: 124560.
  • 8
    Seino M, Fujitani B. Clinical efficacy and use in epilepsy. In: LevyRH et al., eds. Antiepileptic drugs. 5th ed. Philadelphia : Lippincott Williams & Wilkins, 2002: 885.
  • 9
    Schmidt D, Jacob R, Loiseau P, et al. Zonisamide for add-on treatment of refractory partial epilepsy: a European double-blind trial. Epilepsy Res 1993;15: 6773.DOI: 10.1016/0920-1211(93)90011-U
  • 10
    Sackellares JC, Ramsay RE, Wilder BJ, et al. Randomized, controlled clinical trial of zonisamide as adjunctive treatment for refractory partial seizures. Epilepsia 2004;45: 610.DOI: 10.1111/j.0013-9580.2004.11403.x
  • 11
    Faught E, Ayala R, Montouris GG, et al. Randomized controlled trial of zonisamide for the treatment of refractory partial-onset seizures. Neurology 2001;57: 17749.
  • 12
    Commission on the Classification and Terminology of the International League Against Epilepsy. Proposal for revised clinical and electroencephalographic classification of epileptic seizures. Epilepsia 1981;22: 489501.
  • 13
    Hollander M, Wolfe DA. Nonparametric statistical methods. 1st ed. New York : Wiley, 1973.
  • 14
    Marson AG, Kadir ZA, Chadwick DW. New antiepileptic drugs: a systematic review of their efficacy and tolerability. BMJ 1996;313: 116974.
  • 15
    Peeters K, Adriaenssen I, Wapenaar R, et al. A pooled analysis of adjunctive topiramate in refractory partial epilepsy. Acta Neurol Scand 2003;108: 915.DOI: 10.1034/j.1600-0404.2003.00113.x
  • 16
    Cereghino JJ, Biton V, Abou-Khalil B, et al. Levetiracetam for partial seizures: results of a double-blind, randomized clinical trial. Neurology 2000;55: 23642.
  • 17
    Mohanraj R, Brodie MJ. Measuring the efficacy of antiepileptic drugs. Seizure 2003;12: 41343.DOI: 10.1016/S1059-1311(03)00047-5
  • 18
    Oguni H, Hayashi K, Fukuyama Y, et al. Phase III clinical study of the new antiepileptic drug, AD-810 (zonisamide), in patients with childhood epilepsy. Jpn J Pediatr 1988;42: 43950.
  • 19
    Valeriano J, Lane C. Efficacy and tolerability of zonisamide in generalized seizures [Abstract]. Epilepsia 2001;42(suppl 7):188.
  • 20
    Suzuki Y, Nagai T, Ono J, et al. Zonisamide monotherapy in newly diagnosed infantile spasms. Epilepsia 1997;38: 10358.
  • 21
    Kishi T, Nejihashi Y, Kajiyama M, et al. Successful zonisamide treatment for infants with hypsarrhythmia. Pediatr Neurol 2000;23: 2747.DOI: 10.1016/S0887-8994(00)00183-1
  • 22
    Kyllerman M, Ben-Menachem E. Zonisamide for progressive myoclonus epilepsy: long-term observations in seven patients. Epilepsy Res 1998;29: 10914.DOI: 10.1016/S0920-1211(97)00069-7
  • 23
    Vossler DG and the Zonisamide PME Study Group. Multicenter, open-label, safety and efficacy study of zonisamide in patients with progressive myoclonic epilepsy [Abstract]. Neurology 2002;58(suppl 3):A2967.
  • 24
    Ganguly G, Selco S, Ko DY. Zonisamide is effective in the treatment of myoclonus status epilepticus and myoclonus in Jakob-Creutzfeldt disease [Abstract]. Epilepsia 2002;43(suppl 7):218.
  • 25
    Bourgeois BFD. Zonisamide (ZNS) in the treatment of myoclonic astatic epilepsy (MAE, Doose syndrome) [Abstract]. Epilepsia 2002;43(suppl 7):56.DOI: 10.1046/j.1528-1157.43.s.9.13.x
  • 26
    Hershkowitz L. Zonisamide therapy in patients with absence seizures [Abstract]. Epilepsia 2002;43:(suppl 7):193.DOI: 10.1046/j.1528-1157.2002.20901.x
  • 27
    Kochak GM, Page JG, Buchanan RA, et al. Steady-state pharmacokinetics of zonisamide, an antiepilepsy agent for treatment of refractory complex partial seizures. J Clin Pharmacol 1998;38: 16671.
  • 28
    Mather GG, Shah J. Drug interactions. In: LevyRH et al., eds. Antiepileptic drugs. 5th ed. Philadelphia : Lippincott Williams & Wilkins, 2002: 880.
  • 29
    Rosenfeld WE, Bergen D, Garnett W, et al. Steady-state drug interaction study of zonisamide and carbamazepine in patients with epilepsy [Abstract]. Neurology 2001;56(suppl 3):A336.
  • 30
    Levy RH, Ragueneau-Majlessi I, Garnett WR, et al. Lack of a clinically significant effect of zonisamide on phenytoin steady-state pharmacokinetics in patients with epilepsy. J Clin Pharmacol 2004;44: 12304.DOI: 10.1177/0091270004268045
  • 31
    Smith D, Brodie M, Dunkley D, et al. Steady-state drug interaction study of zonisamide and sodium valproate in epileptic patients [Abstract]. Neurology 2001;56(suppl 3):A338.
  • 32
    Brodie M, Wilson E, Smith D, et al. Steady-state drug interaction study of zonisamide and lamotrigine in epileptic patients [Abstract]. Neurology 2001;56(suppl 3):A337.