Antiepileptic Drugs, 5th Edition



Drug Interactions

Gary G. Mather PhD, DABT*

Jaymin Shah PhD**

* Program Director, Preclinical ADME/TOX, Myriad Pharmaceuticals Inc., Salt Lake City, Utah

** Director, Clinical Pharmacology, Elan Pharmaceuticals, South San Francisco, California

Zonisamide [3-(sulfamoylmethyl)-1,2-benzisoxazole] is a new antiepileptic drug (AED) with a novel mechanism of action and a broad spectrum of activity. Although zonisamide was approved in the United States only recently, it has been prescribed in Japan for over 10 years. Experience with zonisamide suggests that clinical responses can be achieved at a targeted steady-state plasma concentration close to 20 µg/mL, with individual patients maintained successfully throughout a range of 10 to 40 µg/mL (1). This range is approximately equivalent to 45 to 200 µmol/L zonisamide.

Zonisamide has proven to be useful in the treatment of several types of seizures (2,3) in combination with other AEDs and thus has the potential to interact with AEDs. Zonisamide also is used in the treatment of patients with manic disorders (4) and thus can interact with a variety of other drugs.


Zonisamide metabolism has been described in detail in Chapter 92, and metabolic features relevant to drug interactions are reviewed here. The largest portion of an oral dose of zonisamide is excreted in the urine as unchanged zonisamide, 2-sulfamoylacetylphenol (SMAP), and SMAP conjugates (5). The open-ring metabolite SMAP is formed by enzyme-dependent reductive cleavage of the 1,2-benzisoxazole ring catalyzed largely by cytochrome P450 (CYP) isoenzyme CYP3A4 (6). The in vitro Km describing this reaction has been estimated at approximately 312 µmol/L (7). It recently has been reported that the expressed enzymes CYP2C19 and CYP3A5 also are capable of catalyzing zonisamide reduction (8). However, the intrinsic clearance of CYP3A4 is much higher than those of CYP2C19 and CYP3A5. Therefore, from the point of view of enzyme quantity and relative intrinsic clearances, it appears that CYP3A4 is the primary enzyme responsible for zonisamide metabolism in vivo. The enzyme aldehyde oxidase in the cytosol of several mammalian species also has been shown to catalyze formation of SMAP in vitro (9). Similar studies with enzymes of human origin have not been reported.

Based on the in vitro data, it can be anticipated that inhibitors of CYP3A4-mediated formation of SMAP may decrease zonisamide clearance and increase its plasma concentrations. The degree of interaction depends not only on the potency of the inhibitor but also on the extent to which this metabolic pathway contributes to the overall clearance of zonisamide. Conversely, inducers of CYP3A4 may increase zonisamide clearance and reduce half-life and plasma concentrations. It also can be inferred that the percentage of zonisamide dose cleared by reductive metabolism to SMAP will be increased in patients also treated with drugs known to induce CYP3A4, such as phenytoin, carbamazepine, and phenobarbital. Therefore, the pharmacokinetic effects of CYP3A4 inhibitors are expected to be greatest in patients treated with polytherapy.



The average zonisamide half-life was 26.8 hours in eight patients also treated with phenytoin (n = 7) or phenytoin and carbamazepine (n = 1) (10), a time considerably shorter than the 60 hours reported in untreated volunteers (11). A half-life of 27.1 hours, lower zonisamide maximum plasma concentration (Cmax), and decreased area under the concentration-time curve (AUC) were accompanied by increased clearance in a similar study in patients treated with phenytoin (12). These data are consistent with the knowledge that phenytoin increases the clearance of other substrates of CYP3A4 such as oral contraceptives and carbamazepine


(13). Most recently, zonisamide was added in a stepwise fashion to therapy of patients maintained at steady state with phenytoin. After 2 weeks of treatment with 200 mg twice daily, the zonisamide half-life was 27.3 hours (14). Protein binding characteristics of zonisamide in human serum are not altered by phenytoin treatment, and therefore no changes in zonisamide clearance are anticipated due to altered levels of unbound zonisamide (15).


In a recent study, zonisamide was added in a stepwise manner to therapy of patients taking only carbamazepine. After maintaining treatment with zonisamide at 200 mg twice daily for at least 2 weeks, its half-life was 39 hours (16). This is consistent with earlier studies, where the average zonisamide half-life was 36.4 hours in patients also treated with carbamazepine and the oral clearance was 20.6 mL/hr/kg (12). The extent of the reduction of zonisamide half-life by carbamazepine appears to be less than that observed with phenytoin. In a population pharmacokinetics study, the clearance of zonisamide was found to be increased by 13% in 37 patients also treated with carbamazepine (17).


McJilton et al. (18) reported two patients treated with zonisamide whose serum levels became elevated when lamotrigine was introduced. Both patients had been treated with zonisamide for approximately 3 years and had a history of stable zonisamide serum concentrations. After introduction of lamotrigine with incremental doses of 25 mg/wk, signs and symptoms of toxicity appeared as lamotrigine doses reached 400 mg/day. Trough zonisamide levels were increased from approximately 27 or 33 µg/mL to 61.8 or 64.6 µg/mL, respectively, for the two patients. Toxicities subsided as lamotrigine was withdrawn and recurred when the patients were rechallenged. The mechanism for this interaction is unclear but would be consistent with reduction of zonisamide clearance. Conversely, Brodie et al. (19) has reported a zonisamide half-life of 51 hours in healthy epileptic patients also treated with lamotrigine, a value shorter than estimated after extended treatment with 400 mg/day (63 to 69 hours) in a group of 24 healthy volunteers (20).


In a study of volunteers treated with phenobarbital, the Cmax and time to maximal plasma concentration (Tmax) of a single dose of zonisamide were unaffected, but clearance was increased and half-life was decreased to 38 hours (21). These data are similar to previous reports demonstrating that clearance of zonisamide is significantly increased and half-life reduced in subjects also treated with phenobarbital (22).

Valproic Acid

The free fraction of zonisamide was slightly but significantly increased in vitro when measured in the presence of valproic acid. Although the increase was small (approximately 3%), a similar increase in free fraction was noted in patients also treated with valproic acid (23). In a separate study, steady-state pharmacokinetics were determined after administration of zonisamide to patients also treated with valproate. The observed zonisamide half-life was 52 hours, similar to previously reported data (24).


The high concentration of zonisamide in erythrocytes suggests that a drug-drug interaction with other sulfonamides may be possible. Although binding to serum proteins is not altered by other sulfonamides (15), zonisamide bound to erythrocyte membranes can be readily displaced by other sulfonamides in vitro (25), with a potency similar to their dissociation constants for carbonic anhydrase. In a study performed in rats, plasma and tissue levels of zonisamide did not change significantly even though erythrocyte levels of zonisamide decreased in the presence of other sulfonamides (25). This is most likely because the tissue and plasma compartments of zonisamide are large compared with the erythrocyte compartment at therapeutic dose levels.


The formation of SMAP was inhibited by approximately 40% by high concentrations of cimetidine (1 mmol/L) in human liver microsomes in vitro (6), suggesting the possibility of interaction. However, no significant differences were reported in a recent study (26). The zonisamide AUC, mean elimination half-life, and Cmax were all similar in groups administered a single 300-mg dose of zonisamide either with or without coadministration of cimetidine 300 mg four times daily. Although cimetidine is known to elevate phenytoin plasma concentrations, the interaction with phenytoin is most likely mediated through inhibition of CYP2C19, an enzyme inhibited by cimetidine at plasma concentrations.

Other Drugs

The in vitro formation of SMAP was inhibited by ketoconazole, cyclosporine, dihydroergotamine, itraconazole, miconazole, triazolam, and fluconazole, in order of potency, with inhibition constants (Ki) ranging from 0.18 to 61.4


µmol/L (8). Using these data and the maximal unbound concentration of each inhibitor, these investigators predicted that only ketoconazole, cyclosporine, and miconazole would decrease zonisamide clearance by more than 10%. Because the degree of inhibition estimated was based on the assumption that zonisamide is eliminated entirely by metabolism, these predictions estimate maximum levels of inhibition. However, “tight binders” (inhibitors) of CYP3A4 may exhibit more significant inhibition than can be predicted in vitro by their unbound concentrations (27). Although clinical interactions have not been reported to date, it seems prudent to be aware of this possibility whenever potent inhibitors of CYP3A4 are added to zonisamide treatment regimens.



Reports of the interaction between zonisamide and carbamazepine have been conflicting. Early studies suggested that average serum concentrations of carbamazepine after addition of zonisamide (300 to 600 mg/day for 2, 4, 8 and 12 weeks) were not significantly different from baseline values in a group of 10 patients (28). Mean carbamazepine concentrations during zonisamide treatment were unchanged in one, increased in three, and decreased in six patients. In other preliminary studies, initiation of zonisamide therapy was accompanied by higher carbamazepine plasma concentrations. Sackellares et al. (29) reported a significant elevation in carbamazepine plasma concentration as zonisamide was added. Carbamazepine levels increased from a mean of 6.8 to 11.7 µg/mL in seven patients treated concomitantly with two, three, or four other AEDs. However, one additional AED (phenytoin for four patients, phenobarbital for two patients, and primidone for one patient) was eliminated over the same period. Thus, the increases in carbamazepine levels may have resulted from withdrawal of an inducing drug rather than from introduction of zonisamide. Minami et al. (30) reported that the average level-to-dose ratio of carbamazepine in 16 pediatric patients receiving zonisamide was lower than that in patients treated with carbamazepine alone. In the same study, the ratio of carbamazepine epoxide to carbamazepine nearly doubled as zonisamide dosages increased from zero to 8.1 ± 3.8 mg/kg/day and zonisamide serum levels increased to 27.7 ± 7.8 µg/mL. Conversely, the epoxide-to-carbamazepine ratio was decreased significantly and the carbamazepine level-to-dose ratio was slightly increased in 15 patients when zonisamide was added to their treatment regimen (31). The reasons for these discrepancies are unclear but may result from differences in the time between drug administration and sample collection. Carbamazepine is metabolized by enzymes found in the gastrointestinal mucosa. It is conceivable that the local concentration of zonisamide in the gastrointestinal tract may affect the rate of carbamazepine metabolism during the absorption phase.


In a study of 18 patients whose seizures were controlled with lamotrigine, zonisamide was added after two baseline measurements of plasma lamotrigine (7 days apart). Zonisamide was titrated from initial dosages of 100 mg/day to 200 mg twice daily within 3 weeks. After 2 weeks of stable treatment, the mean lamotrigine Cmax, Tmax, and AUC were similar to baseline measurements, with no statistically significant differences detected (19).


The effect of zonisamide on plasma concentrations of phenytoin is unclear. Zonisamide initially was reported to increase the serum concentrations of phenytoin. Indeed, phenytoin levels were higher after addition of zonisamide in two of three patients also treated with phenytoin and at least one additional AED (29); however, phenobarbital was withdrawn from one patient and carbamazepine from the other over the same period. Thus, the change in phenytoin may have resulted from withdrawal of an inducing drug rather than introduction of zonisamide. Consistent with this hypothesis, a study conducted in 10 patients revealed that the average serum concentration of phenytoin was not significantly different from baseline values after addition of zonisamide (300 to 600 mg/day for 2, 4, 8, and 12 weeks). Mean phenytoin concentrations increased in three patients, decreased in six patients, and remained unchanged in the remaining patient (28). A population pharmacokinetics study revealed a significant effect of zonisamide on the disposition of phenytoin. Slight but significant increases in the serum phenytoin concentrations were observed in patients receiving zonisamide. The apparent clearance of phenytoin at a given dose was decreased by 14% by zonisamide. Moreover, the apparent Km for phenytoin was increased by 16% in the presence of zonisamide (32). Phenytoin protein binding in human serum was not changed significantly in the presence of zonisamide (21.3 µg eq/mL) in vitro (15); therefore, no changes in phenytoin clearance due to changes in unbound phenytoin should be anticipated.

Valproic Acid

In one study, zonisamide was added to the therapy of 16 patients with epilepsy controlled with valproic acid alone. Zonisamide doses were titrated to 200 mg twice daily within 3 weeks. The mean Cmax, Tmax, and AUC of valproic acid at 35 days were similar to baseline measurements without zonisamide (24). Similar results were reported in previous studies (21).



Oral Contraceptives

Numerous studies have shown that induction of metabolism and specifically induction of CYP3A4 increases the metabolic clearance of contraceptive compounds. This leads to reductions in their plasma concentrations and ultimately decreased efficacy. Enzyme induction by zonisamide is unclear. Zonisamide administration has been accompanied by either a slight increase or a slight decrease in the level of carbamazepine, a drug also cleared predominantly by CYP3A4 (33). There also is limited evidence of autoinduction with chronic zonisamide treatment. Although zonisamide clearance decreased with increasing dose, the Vmax (maximum velocity) increased in six patients from a mean of 0.36783 L/kg/hr to 0.59588 L/kg/hr with chronic dosing (34). No change was noted in two additional patients. In a more recent study, single-dose pharmacokinetic parameters of levonorgestrel and ethinylestradiol were measured in a randomized, two-way crossover study in 12 healthy volunteers. Volunteers were given a single dose of Eugynon 50 (Schering AG, Montville, NJ) (50 µg ethinylestradiol and 250 µg levonorgestrel) alone or after approximately 4 weeks of zonisamide treatment. Zonisamide was initiated at 100 mg/day and escalated to 200 mg twice daily over 15 days. The AUC and Cmax for both ethinylestradiol and levonorgestrel were similar regardless of zonisamide treatment (unpublished data). These data suggest that zonisamide does not increase the clearance of these compounds. Therefore, it appears that zonisamide can be used without decreasing the efficacy of these oral contraceptives.


  1. Oommen KJ, Mathews S. Zonisamide: a new antiepileptic drug. Clin Neuropharmacol1999;22:192-200.
  2. Leppik IE. Zonisamide. Epilepsia1999;40:S23-S29.
  3. Perucca E. The clinical pharmacokinetics of the new antiepileptic drugs. Epilepsia1999;40[Suppl 9]:S7-S13.
  4. Kanba S, Yagi G, Kamijima K, et al. The first open study of zonisamide, a novel anticonvulsant, shows efficacy in mania. Prog Neuropsychopharmacol Biol Psychiatry1994;18:707-715.
  5. Ito T, Yamaguchi T, Miyazaki H, et al. Pharmacokinetic studies of AD-810, a new antiepileptic compound. Arzneimittelforschung1982;32:1581-1586.
  6. Nakasa H, Komiya M, Ohmori S, et al. Characterization of human liver microsomal cytochrome P450 involved in the reductive metabolism of zonisamide. Mol Pharmacol1993;44: 216-221.
  7. Mather GG, Carlson S, Trager WF, et al. Prediction of zonisamide interactions based on metabolic isozymes. Epilepsia1997;38[Suppl 8]:108.
  8. Nakasa H, Nakamura H, Ono S, et al. Prediction of drug-drug interactions of zonisamide metabolism in humans from in vitro data. Eur J Clin Pharmacol1998;54:177-183.
  9. Sugihara K, Kitamura S, Tatsumi K. Involvement of mammalian liver cytosols and aldehyde oxidase in reductive metabolism of zonisamide. Drug Metab Dispos1996;24:199-202.
  10. Browne TR, Szabo GK, Kres J. Elimination half-life of zonisamide (CI-912) after chronic administration. J Clin Pharmacol1986;26:555.
  11. Seino M, Miyazaki H, Ito T. Zonisamide. Epilepsy Res1991; [Suppl 3].
  12. Ojemann LM, Shastri RA, Wilensky AJ, et al. Comparative pharmacokinetics of zonisamide (CI-912) in epileptic patients on carbamazepine or phenytoin monotherapy. Ther Drug Monit1986; 8:293-296.
  13. Mather GG, Levy RH. Anticonvulsants. In: Levy RH, Thummel KE, Trager WF, et al., eds. Metabolic drug interactions.Philadelphia: Lippincott Williams & Wilkins, 2000:217-243.
  14. Garnett WR, Towne AR, Rosenfeld WE, et al. Steady-state pharmacokinetic interaction study of zonisamide (Zonegran) and phenytoin in subjects with epilepsy. American Academy of Neurology, 2000.
  15. Matsumoto K, Miyazaki H, Fujii T, et al. Absorption, distribution and excretion of 3-(sulfamoyl[14C]methyl)-1,2-benzisoxazole (AD-810) in rats, dogs and monkeys and of AD-810 in men. Arzneimittelforschung1983;33:961-968.
  16. Rosenfeld WE, Bergen D, Garnett W, et al. Steady state drug interaction study of zonisamide and carbamazepine in epileptic patients. American Academy of Neurology, 2000.
  17. Hashimoto Y, Odani A, Tanigawara Y, et al. Population analysis of the dose-dependent pharmacokinetics of zonisamide in epileptic patients. Biol Pharm Bull1994;17:323-326.
  18. McJilton J, DeToledo J, DeCerce J, et al. Cotherapy of lamotrigine/zonisamide results in significant elevation of zonisamide levels. Epilepsia1996;37[Suppl 5]:173.
  19. Brodie M, Wilson E, Smith D, et al. Steady state drug interaction study of zonisamide and lamotrigine in epileptic patients. American Academy of Neurology, 2000.
  20. Kochak GM, Page JG, Buchanan RA, et al. Steady-state pharmacokinetics of zonisamide, an antiepileptic agent for treatment of refractory complex partial seizures. J Clin Pharmacol1998; 38:166-171.
  21. Buchanan RA, Page JG, French JA, et al. Zonisamide drug interactions. Epilepsia1997;38[Suppl 9]:107.
  22. Schentag JJ, Bengo FM, Wilton JH, et al., Influence of Phenobarbital, cimetidine, and renal disease on zonisamide kinetics. Pharm Res1987;4(S):S79.
  23. Kimura M, Tanaka N, Kimura Y, et al. Factors influencing serum concentration of zonisamide in epileptic patients. Chem Pharm Bull (Tokyo)1992;40:193-195.
  24. Smith D, Brodie M, Dunkley D, et al. Steady state drug interaction study of zonisamide and sodium valproate in epileptic patients. American Academy of Neurology, 2000.
  25. Matsumoto K, Miyazaki H, Fujii T, et al. Binding of sulfonamides to erythrocyte proteins and possible drug-drug interaction. Chem Pharm Bull (Tokyo)1989;37:2807-2810.
  26. Groves L, Wallace J, Shellenberger K. Effect of cimetidine on zonisamide pharmacokinetics in healthy volunteers. Epilepsia1996;39[Suppl 6]:191.
  27. Thummel KE, Wilkinson GR. In vitro and in vivo drug interactions involving human CYP3A. Annu Rev Pharmacol Toxicol1998;38:389-430
  28. Browne TR, Szabo GK, Kres J, et al. Drug interactions of zonisamide (CI-912) with phenytoin and carbamazepine. J Clin Pharmacol1986; 26:555.
  29. Sackellares JC, Donofrio PD, Wagner JG, et al. Pilot study of zonisamide (1,2-benzisoxazole-3-methanesulfonamide) in patients with refractory partial seizures. Epilepsia1985; 26: 206-211.
  30. Minami T, Ieiri I, Ohtsubo K, et al. Influence of additional therapy with zonisamide (Excegran) on protein binding and metabolism of carbamazepine. Epilepsia1994; 35:1023-1025.



  1. Shinoda M, Akita M, Hasegawa M, et al. The necessity of adjusting the dosage of zonisamide when coadministered with other anti-epileptic drugs. Biol Pharm Bull1996;19:1090-1092.
  2. Odani A, Hashimoto Y, Takayanagi K, et al. Population pharmacokinetics of phenytoin in Japanese patients with epilepsy: analysis with a dose-dependent clearance model.Biol Pharm Bull1996;19:444-448.
  3. Kerr BM, Thummel KE, Warden CJ, et al. Human liver carbamazepine metabolism: role of CYP3A4 and CYP2C8 in 10,11-epoxide formation. Biochem Pharmacol1994;47:1969-1979.
  4. McJilton JS, Ramsay RE, Vasquez D, et al. Pharmacokinetics of zonisamide: results from a randomized double-blind multicenter placebo controlled study. Epilepsia2000;41 [Suppl 7]:224.