Topiramate (TPM) is a highly oxygenated sulfamate-substituted monosaccharide that is structurally distinct from other anticonvulsant medications. Available in the United States as Topamax (Ortho-McNeil Pharmaceutical), it is a broad-spectrum agent that has been extensively studied in double-blind, randomized, controlled trials in adults and children. Initially approved for use as adjunctive therapy in partial-onset seizures, primary generalized tonic-clonic seizures, and multiple seizure types associated with Lennox-Gastaut syndrome, it is currently under evaluation for approval as first-line therapy in the United States.

TPM (2,3:4,5-di-O-isopropylidene-β-D-fructopyranose sulfamate; Fig. 61.1) is a white crystalline powder, which is freely soluble in acetone, chloroform, dimethylsulfoxide, and ethanol. TPM is supplied as 25-, 50-, 100-, and 200-mg tablets and as 15- and 25-mg sprinkle capsules that can be opened and sprinkled onto soft food for children and for patients who may have difficulty swallowing tablets.

TPM has a unique combination of activities at various receptor sites and ion channels, which may account for its broad-spectrum profile in epilepsy and other neurologic disorders. It blocks the kainate/AMPA (α-amino-3-hydroxy-5-methy-lisoxazole-4-propionic acid) subtype of the glutamate receptor (1, 2, 3, 4), with no direct effect on NMDA (N-methyl-D-aspartate) receptor activity; blocks voltage-activated sodium channels to limit sustained repetitive firing (5, 6, 7, 8, 9); enhances γ-aminobutyric acid (GABA)-mediated chloride flux at GABA_A receptors (10,11); reduces the amplitude of high-voltage-activated calcium currents (12,13); and activates potassium conductance (14,15). It has been hypothesized that effects of TPM on voltage-activated sodium channels, high-voltage-activated calcium channels, GABA_A receptors, and AMPA/kainate receptors reflect a common modulator involving protein phosphorylation (16). TPM is also a weak inhibitor of carbonic anhydrase isoenzymes (CA II and CA IV), which may modulate pH-dependent activation of volt-age-and receptor-gated ion channels (17); its inhibitory effect is less than acetazolamide.

The anticonvulsant properties of TPM have been demonstrated in several animal models of epilepsy; these have been discussed elsewhere (16,18, 19, 20, 21). Experimental studies have shown that TPM reduces seizure-induced hippocampal neuronal injury (22) and prevents spontaneous seizures following status epilepticus (23). In an experimental model of neonatal hypoxia/ischemia, TPM suppressed acute seizures and reduced subsequent susceptibility to neuronal injury and seizures induced by a second insult (kainate) (24).

Renal elimination, low protein binding, and a long half-life make TPM relatively easy to manage from a pharmacokinetic perspective.

Steady-state plasma concentrations of TPM are generally linear, with dose-proportional increases in plasma concentration (26). Mean plasma concentrations achieved during maintenance in randomized, controlled trials of TPM monotherapy were: 50 mg per day, 1.6 and 1.9 μg/mL; 97 mg per day, 3.8 μg/mL; 189 mg per day, 6.4 μg/mL; 313 mg per day, 11.7 μg/mL; 367 mg per day, 12.4 μg/mL (30). Studies of TPM as monotherapy have provided the opportunity to examine the relationship between TPM blood levels and clinical response. In a study comparing 50 and 500 mg per day TPM as monotherapy, plasma concentrations >9.91 μg/mL were associated with better seizure control compared with plasma concentrations of 1.77 to 9.91 μg/mL and ≤1.76 μg/mL (31). However, because of the intraindividual variations in blood levels associated with seizure control and side effects, a traditional “therapeutic range” cannot be identified. As expected, plasma concentrations are higher when TPM is administered as monotherapy (6.4 to 12.4 μg/mL with ˜200 to 400 mg per day) versus its use as add-on to enzyme-inducing AEDs (1.4 to 5.3 μg/mL with ˜200 to 400 mg per day). Despite the substantially higher plasma concentration with monotherapy, the incidence of central nervous system (CNS)-related adverse events, particularly cognitive effects, was substantially lower with TPM monotherapy than with adjunctive therapy. This finding underscores the contribution of pharmacodynamic interactions to the occurrence of adverse events during TPM polytherapy and the limited benefit of therapeutic drug monitoring in TPM-treated patients.

The relationship between TPM dose and plasma level was examined in children in whom TPM was titrated to clinical response or side effects (32). Among 21 children aged 6 to 12 years, TPM plasma levels were predictably related to dose (1:1 ratio). With monotherapy, a mean dose of 9.7 mg/kg per day (range, 5.5 to 16.5 mg/kg per day) resulted in a mean plasma level of 9.8 μg/mL (range, 3.4 to 16.6 μg/mL). For 20 younger children (younger than 6 years of age), however, higher monotherapy doses were needed (mean, 22.5 mg/kg per day; range, 11 to 35 mg/kg per day) to achieve seizure control; mean plasma level was 14.8 μg/mL (range, 6.1 to 23.7 μg/mL). When TPM was administered with an enzyme-inducing drug, the TPM dosage in younger children (mean, 14.2 mg/kg per day) was double that in older children (7.0 mg/kg per day) (32).

Predominantly renal elimination and low protein binding minimize the potential for drug interactions.
Pharmacokinetic interactions between TPM and other AEDs are limited primarily to the effects of enzyme-inducing drugs on TPM. TPM plasma levels are approximately 50% lower when TPM is given with an enzyme-inducing AED (33, 34, 35) compared to TPM use alone or in combination with nonenzyme-inducing drugs (33, 34, 35, 36). The addition of TPM does not significantly affect plasma concentrations of carbamazepine (33), valproate (36), phenobarbital/primidone (34), or lamotrigine (37). However, phenytoin plasma levels may be increased as much as 25% in some patients, particularly those in whom phenytoin metabolism may be at or near saturation (35). Studies of TPM in models designed to predict drug interactions related to the cytochrome P450 (CYP450) enzyme system have shown inhibition of only the CYP2C19 isozyme, which may account for the potential interaction with phenytoin (38). Although pharmacokinetic interactions between TPM and other AEDs are limited, the lower incidence of adverse effects with TPM monotherapy (31,39,40) suggests that pharmacodynamic interactions may affect tolerability when TPM is added to existing therapy.

Interaction studies evaluating the effect of TPM on combination oral contraceptives showed that TPM has no effect on the progestin (norethindrone, 1.0 mg) component (41,42). At doses of ≤200 mg per day, TPM has no significant effect on estrogen (ethinyl estradiol, 35 μg) concentrations (41,42). At higher doses (400 and 800 mg per day), TPM was associated with 21% and 30% reductions, respectively, in ethinyl estradiol concentrations, suggesting a modest induction of estrogen clearance (42). The level of induction is substantially less than that associated with potent enzyme-inducing agents such as carbamazepine (42% reduction in estrogen concentration) (41). The dose-related effect of TPM on estrogen clearance is consistent with the concentration-dependent induction of CYP450 CYP3A4 activity measured in vitro (43). TPM induced CYP3A4 enzymes only at concentrations >50 μM, a concentration that is unlikely to be achieved with dosages up to 400 mg per day; enzyme induction was still less than that associated with known inducers (phenobarbital and rifampicin) used in this study.

A slight decrease in digoxin clearance has been observed with the addition of TPM (44), but generally does not require dosage adjustments. Changes in metformin pharmacokinetics suggest that diabetic control should be monitored when TPM is added or withdrawn (45).