Abstract
Keywords
INTRODUCTION
Parkinson’s disease (PD) is a common neurodegenerative disorder characterized by motor and non-motor symptoms [1]. PD is positively associated with age, with a prevalence of approximately 0.3% in industrialized countries [2–4]. Currently, patients with PD require levodopa therapy for motor symptoms; however, long-term levodopa therapy is associated with an increased risk of motor complications such as motor fluctuations and dyskinesia [5–8]. Therefore, several drugs such as dopamine receptor agonists, monoamine oxidase-B (MAO-B) inhibitors, and catechol-O-methyltransferase (COMT) inhibitors have also been approved for the treatment of later PD in combination with levodopa [9–11]. A recent meta-analysis [12] showed that dopamine agonists, MAO-B inhibitors, and COMT inhibitors were superior to placebo in improving the Unified Parkinson’s Disease Rating Scale (UPDRS) Part III scores (weighted mean difference [WMD], –4.77, –2.9, and –2.02 for dopamine agonists, MAO-B inhibitors, and COMT inhibitors, respectively) and reducing the wearing-off time (WMD, –1.57, –0.93, and –0.83 for dopamine agonists, MAO-B inhibitors, and COMT inhibitors, respectively) in later PD. Furthermore, their effect sizes within each outcome were similar. In fact, the National Institute for Health and Care Excellence (NICE) guideline stated that it was evident from reviewing the evidence base that there was no single drug of choice in the pharmacotherapy of later PD [9]. A recent meta-analysis [12] showed that there were differences in risk of adverse events among these drugs for later PD. Specifically, they identified dyskinesia, dizziness, hallucinations, hypotension, nausea, and somnolence associated with dopamine receptor agonists; vomiting associated with MAO-B inhibitors; and dyskinesia, constipation, dizziness, dry mouth, insomnia, nausea, somnolence, and vomiting associated with COMT inhibitors [12].
Murata and colleagues accidentally found that when they gave zonisamide to a PD patient with convulsive attacks incidentally, zonisamide improved the attacks as well as the parkinsonian symptoms [13]. Zonisamide (1,2-benzisoxazole-3-methanesulfonamide) is an antiepileptic drug [14] that was shown to exert multiple mechanisms of action in animal studies on rats including sodium and T-type calcium channel blockage [15], inhibition of glutamate release [16], enhancement of gamma aminobutyric acid-mediated neuronal inhibition [17], inhibition of MAO-B activity [18], activation of dopamine synthesis and release (this mechanism is dose-dependent at low doses in animal models, whereas higher doses reduce intracellular dopamine) [18], and potential neuroprotective properties [19]. Based on its inhibition of MAO-B activity and activation of dopamine synthesis and release, zonisamide was suggested to have beneficial effects on motor symptoms of patients with PD [20].
Several recent randomized placebo-controlled trials (RCTs) were conducted to evaluate combination therapy with zonisamide and antiparkinson drugs in patients with PD [21–24] (Table 1). However, the results were discrepant between the trials. For example, while two studies [21, 23] reported that zonisamide (25 mg or 50 mg) was superior to placebo in improving UPDRS Part III scores [25], another study [22] did not find a significant difference between the treatments. One reason that can explain this discrepancy is the small sample sizes. In fact, the statistical power of UPDRS Part III scores in the study by Murata et al. [22] with an alpha error level of 5% was only 29.4% because of the small sample size. Meta-analysis provides a weighted summary result, with more weight given to studies with larger sample sizes. In addition, quality of the study design can be evaluated with consideration of individual studies by performing a systematic review (Supplementary Figure 1). By combining results from multiple studies, a meta-analysis can increase the statistical power of individual studies; however, whether there is significant heterogeneity among the studies included in a meta-analysis needs to be assessed as well [26]. Therefore, we conducted a systematic review and meta-analysis to examine whether combination therapy with zonisamide and antiparkinson drugs was beneficial for the treatment of PD.
METHODS
This meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [27]. We performed a systematic literature search according to the following PICO (Patient, Intervention, Comparator, and Outcomes) criteria: Patients of interest (P): patients with primary/idiopathic PD who received antiparkinson drugs including amantadine, COMT inhibitors, dopamine agonists, levodopa, and/or MAO-B inhibitors (all were eligible); Intervention of interest (I): zonisamide; Comparator of interest (C): placebo; Outcomes of interest (O): UPDRS score, wearing-off time, discontinuation rate (all causes, adverse events, inefficacy, and death), and individual adverse events.
All doses of antiparkinson drugs were eligible for inclusion in this study.
Inclusion criteria, search strategy, data extraction, and outcome measures
We included only double-blinded RCTs where zonisamide was given to patients with PD. To identify relevant studies, we searched major healthcare databases (the Cochrane Central Register of Controlled Trials, MEDLINE, Embase, PsycINFO, and PubMed) and clinical trial registries (ClinicalTrials.gov, ISRCTN, and the WHO portal). There were no language restrictions, and we considered all published studies from the inception of the databases to May 3, 2016. We used the following keywords: (“Parkinson Disease” [Mesh]) OR “Parkinson disease” OR “Parkinson’s disease” AND “zonisamide.” Additional eligible studies were sought from the reference lists of primary articles and relevant reviews.
We selected the studies in two stages. In the first stage, two authors (S.M. and T.K.) independently screened all titles and abstracts for eligibility. In the second stage, we assessed full texts of the retrieved papers. Two authors (S.M. and T.K.) independently selected the papers against prespecified inclusion criteria. Any discrepancies were resolved after discussion among the three authors (S.M., T.K., and N.I.). The first two authors (S.M. and T.K.) independently extracted, reviewed, and entered the data into Review Manager (Version 5.3 for Windows, Cochrane Collaboration, http://ims.cochrane.org/revman). We contacted the authors of included studies to request pertinent endpoint data if the relevant data needed for the meta-analysis were not available in the articles. All extracted data from the studies included in this meta-analysis are shown in forest plots (Supplementary Figures 2 and 3).
Data synthesis and statistical analysis
We synthesized data when data were available from at least two studies for a particular outcome. The primary outcome measures for efficacy were change in the UPDRS Part III score (motor function) and the wearing-off time. For analysis of the wearing-off time, we combined the data on wearing-off time derived from the scores for UPDRS Part IV item 39, which stated “what proportion of the waking day is the patient “off” on average?,” in patient diaries. The primary outcome measure for safety was discontinuation rates due to all causes. Secondary outcome measures were scores for UPDRS Part I (mentation, behavior, and mood), UPDRS Part II (activities of daily living [ADL]), UPDRS Part IV (complications of therapy), and UPDRS total and discontinuation rates due to adverse events, inefficacy, and death. In addition, we pooled the incidence rates of individual adverse events.
We based our analyses on intent-to-treat (ITT) or modified ITT data (i.e., at least one dose or one follow-up assessment).
The meta-analysis was performed using Review Manager. To combine studies, the random effects model by DerSimonian and Laird was used [28]. The random effects model is more conservative than the fixed effects model and produces a wider confidence interval (CI). For continuous data, we used WMD and 95% CI when data that were measured using the same scale were combined. To combine data measured using different scales, we used standardized mean difference (SMD), combining effect size (Hedges’ g), and 95% CIs. For dichotomous data, the risk ratios (RRs) were estimated with 95% CIs. When there were several dose groups, we used the combined data from all-dose groups for dichotomous data. We explored study heterogeneity using I2 statistics, with values of 50% or higher considered as evidence of considerable heterogeneity [29]. In cases with I2≥50% for primary outcome measures, we planned to conduct sensitivity analyses to determine the reasons for heterogeneity. We examined the following confounding factors: UPDRS Part III scores at baseline (higher [≥21] or lower [<21] scores for each subgroup based on the mean baseline UPDRS Part III score of 21 among the four RCTs) and differences in concomitant drugs among the studies (rate of COMT inhibitors use: studies with 39% COMT inhibitor use versus studies with 0–1% COMT inhibitor use). Additionally, as the approved zonisamide doses for PD in Japan were 25 and 50 mg/day, we performed a subgroup meta-analysis of the primary outcomes divided by the approved zonisamide dose (25 mg/day and 50 mg/day). We also assessed the methodological quality of the included articles based on the Cochrane risk of bias criteria (Cochrane Collaboration; http://www.cochrane.org/). Funnel plots were visually inspected to assess the possibility of publication bias in primary outcomes when included studies were≥10 [30].
RESULTS
Study characteristics
The primary search yielded a total of 368 studies, of which 116 were duplicates (Fig. 1). We excluded 235 studies after reviewing the titles and abstracts because they were case reports (n = 2), non-clinical trials (n = 29), non-PD studies (n = 25), non-zonisamide studies (n = 32), ongoing trials (n = 1), review articles (n = 140), or single-arm studies (n = 6); further, 14 studies were excluded after full-text review because they were case reports (n = 2), data based on same samples (n = 3), non-zonisamide study (n = 1), review article (n = 1), or single-arm studies (n = 7). From a review article [31], we found an unpublished RCT from Japan [24]. AD810N-202-1 study was one of the phase II trials of zonisamide for patients with PD (trial duration: January, 2002 to May, 2003, please contact Sumitomo Dainippon Pharma Co., Ltd for further information on the study); this study was also included in the current meta-analysis. Finally, four RCTs testing zonisamide for PD treatment were included in the meta-analysis [21–24]. There were no differences in the results of the present literature search between the twoauthors.
Characteristics of the included trials are shown in Table 1. The four RCTs included a total of 1,068 patients with PD (zonisamide, n = 750; placebo, n = 318) who already received antiparkinson drugs (levodopa, 100%; dopamine receptor agonists, 90%; MAO-B inhibitors, 47%; amantadine, 32%; COMT inhibitors, 15%). Eight-five percent of the patients included in the current meta-analysis were classified as later PD based on the NICE guideline. Because three of the four RCTs were conducted by the same group of researchers (Murata et al.) and as two were recently published in 2015, we asked the researchers and the company (Sumitomo Dainippon Pharma Company Ltd.) whether the same patients were included in multiple studies analyzed in the present meta-analysis and were able to confirm that no patients were included in more than one study. Three trials [21–23] lasted for 12 weeks, whereas one trial [24] lasted for 10 weeks. Study sample sizes ranged from 136 to 389 patients. Mean age of the pooled study population was 64 years. All studies were sponsored by the pharmaceutical industry (Sumitomo Dainippon Pharma Company Ltd.) and were conducted in Japan. We evaluated the methodological quality of all studies using the Cochrane risk of bias criteria (Supplementary Figure 1). All studies were of a high-quality design (double-blind, parallel, RCT, ITT, or modified ITT population). One study did not set a primaryoutcome [24].
Results of the meta-analysis with all doses of zonisamide
Results of the meta-analysis for primary outcomes with all doses of zonisamide: UPDRS Part III scores
Zonisamide combination therapy significantly decreased the UPDRS Part III scores compared to placebo (WMD, –2.56; 95% CI, –4.20 to –0.92; p = 0.002; I2 = 62%; N = 4, n = 1,008; Table 2).
Sensitivity analyses of primary outcomes with all doses of zonisamide: UPDRS Part III scores
There was significant heterogeneity in UPDRS Part III scores among the studies (I2 = 62%). Therefore, we performed two sensitivity analyses for identifying confounding factors affecting the UPDRS Part III scores. When divided according to the UPDRS Part III scores at baseline, the significant heterogeneity disappeared in the high-score subgroup; moreover, there were significant subgroup differences between the high-score and the low-score subgroups (high-score subgroup [≥21]: WMD, –3.44; 95% CI, –4.74 to –2.14; p < 0.00001; N = 3; I2 = 0%; low-score subgroup [<21]: WMD, –0.9; 95% CI, –2.15 to 0.35; p = 0.16; N = 1; I2 = not applicable; test for subgroup differences, I2 = 87%; p = 0.006). Analysis for differences in the rate of concomitant drugs showed that all studies had similar frequencies of use for levodopa (100%), dopamine receptor agonists (88% to 92%), MAO-B inhibitors (40% to 51%), and amantadine (25% to 39%); however, COMT inhibitor use varied (0% to 39%) (Table 1). When studies were divided into subgroups with high-frequency (39%) and low-frequency (0% to 1%) use of concomitant COMT inhibitors, the significant heterogeneity disappeared in the subgroup with the low-frequency use of concomitant COMT inhibitors; moreover, there were significant subgroup differences between the low-frequency and high-frequency concomitant COMT inhibitor use (low-frequency concomitant COMT inhibitor use subgroup: WMD, –3.44; 95% CI, –4.74 to –2.14; p < 0.00001; N = 3; I2 = 0%; high-frequency concomitant COMT inhibitor use subgroup: WMD, –0.9; 95% CI, –2.15 to 0.35; p = 0.16; N = 1; I2 = not applicable; test for subgroup differences: I2 = 87%; p = 0.006).
Results of the meta-analysis for primary outcomes with all doses of zonisamide: Wearing-off time
Zonisamide combination therapy significantly decreased the wearing-off time compared to placebo (SMD, –0.24; 95% CI, –0.39 to –0.09; p = 0.001; I2 = 0%; N = 4, n = 852; Table 2). There was no significant heterogeneity in the wearing-off time among the studies (I2 = 0%).
Results of the meta-analysis for primary outcomes with all doses of zonisamide: discontinuation rates due to all causes
The discontinuation rates due to all causes were similar between the zonisamide and placebo groups (RR, 1.29; 95% CI, 0.90 to 1.84; p = 0.16; I2 = 0%; N = 4; n = 1,068; Table 2). There was no significant heterogeneity in the discontinuation rate among the studies (I2 = 0%).
Results of the meta-analysis for secondary outcomes with all doses of zonisamide
Zonisamide combination therapy significantly decreased both the UPDRS Part II (off-time) and UPDRS total scores compared to placebo (UPDRS Part II [off-time] scores: WMD, –0.79; 95% CI, –1.33 to –0.24; p = 0.005; I2 = 0%; N = 4; n = 849; UPDRS total scores: WMD, –2.51; 95% CI, –3.81 to –1.21; p = 0.0001; I2 = 12%; N = 4; n = 1,008; Table 2). In contrast, the UPDRS Part I, UPDRS Part II (on time), and UPDRS Part IV scores were similar between the zonisamide and placebo groups (Table 2).
The discontinuation rates due to inefficacy (no study provided a detailed definition of inefficacy) and death were similar between the zonisamide and placebo groups; however, a marginal trend was found in the discontinuation rates due to adverse events with zonisamide compared to those observed with placebo (RR, 1.88; 95% CI, 0.97 to 3.62; p = 0.06; I2 = 0%; N = 4; n = 1,068; Table 2).
No significant differences were found between the groups for the incidence of at least one adverse or serious adverse event (Table 2). There were no significant differences in the incidence of any adverse events between the groups (Supplementary Figure 2). However, a marginal trend was found for the association of zonisamide with higher incidence of somnolence compared to placebo (RR, 1.81; 95% CI, 0.99 to 3.29; p = 0.05; I2 = 0%; N = 4; n = 1,044). In contrast, marginal trends were found in the association of zonisamide with lower incidence rates of right bundle branch block, joint dislocation, abnormal behavior, athymia, and hives, compared to the placebo group (right bundle branch block: RR, 0.13; 95% CI, 0.01 to 1.27; p = 0.08; I2 = 0%; N = 4; n = 1,044; joint dislocation: RR, 0.07; 95% CI, 0.00 to 1.40; p = 0.08; I2 = not applicable; N = 4; n = 1,044; abnormal behavior: RR, 0.14; 95% CI, 0.01 to 1.31; p = 0.08; I2 = 0%; N = 4; n = 1,044; athymia: RR, 0.25; 95% CI, 0.06 to 1.10; p = 0.07; I2 = not applicable; N = 4; n = 1,044; hives: RR, 0.14; 95% CI, 0.01 to 1.32; p = 0.09; I2 = 0%; N = 4; n = 1,044).
Results of the meta-analysis with the approved zonisamide doses (25 mg/day and 50 mg/day)
Results of the meta-analysis for primary outcomes with the approved zonisamide doses (25 mg/day and 50 mg/day): UPDRS Part III scores
Zonisamide combination therapy significantly decreased the UPDRS Part III scores compared to placebo (WMD, –2.86; 95% CI, –4.85 to –0.88; p = 0.005; I2 = 71%; N = 4; n = 859; Table 3).
Sensitivity analyses of primary outcomes with the approved zonisamide doses (25 mg/day and 50 mg/day): UPDRS Part III scores
There was significant heterogeneity in UPDRS Part III scores among the studies (I2 = 71%). Therefore, we performed two sensitivity analyses to identify confounding factors affecting the UPDRS Part III scores. When divided according to the UPDRS Part III scores at baseline, the significant heterogeneity disappeared in the high-score subgroup; moreover, there were significant subgroup differences between the high-score and low-score subgroups (high-score subgroup [≥21]: WMD, –3.79; 95% CI, –5.17 to –2.41; p < 0.00001; N = 3; I2 = 0%; low-score subgroup [<21]: WMD, –0.9; 95% CI, –2.15 to 0.35; p = 0.16; N = 1; I2 = not applicable; test for subgroup differences, I2 = 89.3%; p = 0.002). Analysis for differences in the rate of concomitant drugs showed that all studies had similar frequencies of levodopa (100%), dopamine receptor agonists (88% to 93%), MAO-B inhibitors (35% to 50%), and amantadine (25% to 40%), except for COMT inhibitors (0% to 39%). When studies were divided into subgroups with high-frequency (39%) and low-frequency (0% to 1%) use of concomitant COMT inhibitors, the significant heterogeneity disappeared in the subgroup with low frequency use of concomitant COMT inhibitors. Moreover, there were significant subgroup differences between the low-frequency and high-frequency concomitant COMT inhibitor use subgroups (low-frequency concomitant COMT inhibitor use subgroup: WMD, –3.79; 95% CI, –5.17 to –2.41; p < 0.00001; N = 3; I2 = 0%; high-frequency concomitant COMT inhibitor use subgroup: WMD, –0.9; 95% CI, –2.15 to 0.35; p = 0.16; N = 1; I2 = not applicable; test for subgroup differences, I2 = 89.3%; p = 0.002).
Results of the meta-analysis for primary outcomes with the approved zonisamide doses (25 mg/day and 50 mg/day): wearing-off time
Zonisamide combination therapy significantly decreased the wearing-off time compared to placebo (SMD, –0.24; 95% CI, –0.39 to –0.09; p = 0.002; I2 = 0%; N = 4, n = 751; Table 3). There was no significant heterogeneity in the wearing-off time among the studies (I2 = 0%).
Results of the meta-analysis for primary outcomes with the approved zonisamide doses (25 mg/day and 50 mg/day): discontinuation rates due to all causes
The discontinuation rates due to all causes were similar between the zonisamide and placebo groups (RR, 1.14; 95% CI, 0.79 to 1.66; p = 0.48; I2 = 0%; N = 4; n = 911; Table 3). There was no significant heterogeneity in the discontinuation rate among the studies (I2 = 0%).
Results of the meta-analysis for secondary outcomes with the approved zonisamide doses (25 mg/day and 50 mg/day)
Zonisamide combination therapy significantly decreased the UPDRS Part II (off-time) and UPDRS total scores compared to placebo (UPDRS Part II [off-time] scores: WMD, –0.88; 95% CI, –1.54 to –0.22; p = 0.009; I2 = 18%; N = 4; n = 750; UPDRS total scores: WMD, –3.12; 95% CI, –5.02 to –1.23; p = 0.001; I2 = 47%; N = 4; n = 859; Table 3). In contrast, the UPDRS Part I, UPDRS Part II (on time), and UPDRS Part IV scores were similar between the zonisamide and placebo groups (Table 3).
The discontinuation rates due to adverse events, inefficacy (no study provided a detailed definition of inefficacy), and death were similar between the zonisamide and placebo groups (Table 3).
No significant differences in the incidence of at least one adverse or serious adverse event were found between the groups (Table 3). There were no significant differences in the incidence of any adverse events between the groups (Supplementary Figure 3). In the incidences of individual adverse events, the zonisamide group exhibited a lower incidence of decreased appetite than the placebo group (RR, 0.51; 95% CI, 0.29 to 0.92; p = 0.03; I2 = 0%; N = 4; n = 891; number needed to harm [NNH], non-significant; Supplementary Figure 3).
DISCUSSION
To our knowledge, this is the first comprehensive meta-analysis of RCTs on the efficacy and tolerability of zonisamide combination therapy for treatment of patients with PD who were treated with antiparkinson drugs. The main finding of our study was that zonisamide combination therapy significantly improved motor functions and reduced the wearing-off time compared to placebo in patients with PD who received antiparkinson drugs. In addition, zonisamide significantly improved ADL (as measured by UPDRS Part II off-time) compared to placebo. However, due to the significant heterogeneity in motor function scores between the treatment groups, we performed a sensitivity analysis to detect any confounding factors and found that the significant heterogeneity might be caused by the severity of motor functions at baseline and differences in the rate of concomitant COMT inhibitors. The subgroups of these two sensitivity analyses included the same studies (AD810N-202-1 [24], Murata 2007 [21], and Murata 2015B [23] studies versus Murata 2015A [22] study). Therefore, zonisamide might provide benefit in PD patients with severe motor symptoms and/or PD patients who are not concomitantly treated with COMT inhibitors. However, because the number of studies included in the current meta-analysis was small, further studies using larger samples are needed for conclusive results.
There were also no significant differences in discontinuation rates due to all causes or death among studies. Moreover, zonisamide did not worsen mentation, behavior, mood, or complications of therapy compared to placebo. However, although zonisamide was marginally associated with a higher incidence of discontinuation rates due to adverse events (RR, 1.88; 95% CI, 0.97 to 3.62, p = 0.06) and somnolence (RR, 1.81; 95% CI, 0.99 to 3.29; p = 0.05), these significant differences disappeared with subgroup meta-analysis using data for the approved zonisamide dose for PD in Japan (25 and 50 mg/day) (Supplementary Figure 3). Moreover, the group using approved zonisamide doses exhibited a lower incidence of decreased appetite than the placebo group (RR, 0.51; 95% CI, 0.29 to 0.92, p = 0.03; NNH, non-significant). The results were also similar between the all-dose group and the approved dose subgroup meta-analyses (Table 3, Supplementary Figure 3). These results indicated that the approved dose of zonisamide in combination therapy improved motor symptoms and ADL, decreased the wearing-off time, and was well tolerated in PD patients who received antiparkinson drugs.
Bermejo et al. indicated that zonisamide might be associated with increased risk of psychosis, restless legs syndrome, weight loss, and fatigue [20]. However, we did not detect any significant differences in the rates of these adverse events between the zonisamide and placebo groups.
A recent meta-analysis [12] showed that dopamine agonists, MAO-B inhibitors, and COMT inhibitors were superior to placebo in improving the UPDRS Part III scores (WMD, –4.77, –2.9, and –2.02, respectively). In addition, the same meta-analysis showed that dopamine agonists, MAO-B inhibitors, and COMT inhibitors were superior to placebo in reducing the off-time (WMD, –1.57, –0.93, and –0.83, respectively). The authors also found that the discontinuation rates due to adverse events were significantly increased with COMT inhibitors (odds ratio [OR], 1.46) but not with dopamine agonists (OR, 1.01) or MAO-B inhibitors (OR, 0.62) [12]. Moreover, they reported the risk of adverse events of these drugs as follows: dopamine receptor agonists: dyskinesia (OR, 2.66), dizziness (OR, 1.48), hallucinations (OR, 2.54), hypotension (OR, 1.34), nausea (OR, 1.64), somnolence (OR, 1.62); MAO-B inhibitors: vomiting (OR, 3.25); COMT inhibitors: dyskinesia (OR, 2.50), constipation (OR, 3.22), dizziness (OR, 1.95), dry mouth (OR, 3.07), insomnia (OR, 1.59), nausea (OR, 2.07), somnolence (OR, 2.03), vomiting (OR, 3.68) [12]. Although we did not compare the results of the present study with those of the previous meta-analysis [12], based on the comparable findings between the two studies, zonisamide might be beneficial for treatment of PD patients and was well tolerated, similar to other antiparkinson drugs.
Limitations of our study were as follows: First, the number of studies in this meta-analysis was small. A funnel plot, which is used only if ten or more studies are included in the meta-analysis, was therefore not utilized for investigating potential publication bias. Moreover, the individual studies included relatively small patient samples. In addition, because all studies were conducted in Japanese population, it was unclear whether zonisamide combination therapy may be beneficial for PD in other populations. Second, all included studies were industry-sponsored clinical trials and potential sponsorship bias for the sponsored products could not be controlled [32]. Third, since the durations of all included studies were short (approximately 10–12 weeks), we could not determine the long-term efficacy and safety of zonisamide for patients with PD based on the results of this study.
In conclusion, the results of the present meta-analysis suggested that zonisamide combination therapy was beneficial for patients with PD who received antiparkinson drugs including levodopa, in Japanese population. Furthermore, the approved dose of zonisamide (<50 mg/day) in combination therapy appeared to be well tolerated due to the lack of no significant differences in incidence of any individual adverse events between the approved dose of zonisamide and placebo treatment groups. Future studies should investigate the long-term efficacy and generate more safety data for zonisamide for treatment of PD using larger samples and in populations other than the Japanese population.
