Drug-resistant epilepsy and topiramate: Plasma concentration and frequency of epileptic seizures.

Summary: Topiramate (TPM) is a second‐generation antiepileptic drug (AED), acting on drug‐resistant epilepsy. The aim of the study was to evaluate the influence of the dose, use of other AEDs on TPM plasma concentration (C_p), and frequency of epileptic seizures. A cross‐sectional analytical study was developed with patients aged 18‐60 years, for diagnosis of drug‐resistant epilepsy, using TPM in monotherapy or associated with other AEDs. The following variables were analyzed: age, frequency of epileptic seizures, pharmacotherapeutic regimen with its respective doses, adherence to medication treatment, and adverse events score. Thirty‐seven patients were included, 83.8% of the patients presented C_p below the therapeutic range. Multiple linear regression estimated that the increase of 1.0 mg/kg/d promoted an increase of 0.68 μg/mL in TPMC_p, while the use of inducers predicted a reduction of 2.97 μg/mL (P < .001). Multiple Poisson regression predicts that an increase of 1.0 μg/mL in TPMC_p decreased the patient's chance of presenting seizures, and patients using AED inducers were about ten times more likely to present seizures than those who do not use (P < .001). In addition, for patients using AED inducers with C_p below the therapeutic range, the mean number of seizures per month was greater than those with C_p within the therapeutic range. The prescribed dose and the use of AED inducers influence C_p of TPM, likewise the low C_p of first‐line AEDs and of the adjuvant in the treatment, TPM, as well as low TPM dose seem to affect the control of epileptic seizures.

anticonvulsants; drug monitoring; drug‐resistant epilepsy

The management of epilepsy is complex, especially in drug‐resistant epilepsy where combination therapy is often used.[1] [2]

Topiramate (TPM) is a second‐generation antiepileptic drug (AED) that is effective as monotherapy and adjuvant therapy, acting on drug‐resistant epilepsy.[3] TPM has linear pharmacokinetics, apparent volume of distribution of 0.6‐0.8 L/kg and low plasma protein binding (about 15.0%). The elimination half‐life ranges from 19 to 30 hours, and apparent plasma clearance (Cl/F) ranges from 15 to 50 mL/h/kg. Approximately 60.0% of TPM is excreted unchanged in urine, and the remainder is metabolized, but no active metabolite is known. The therapeutic range of TPM is between 5 and 20 μg/mL and therapeutic monitoring is an important tool for the rational prescription of TPM in clinical practice, since the association of TPM with AED inducers of cytochrome P450 (CYP450) metabolism promotes increased TPM metabolism. Furthermore, valproic acid (VPA) inhibitors of CYP450 also promote an increase in the metabolism of TPM.[4] [5] [6] [7] [8]

The aim of the study was to evaluate the influence of dose, age, use of other AEDs, and adherence to drug treatment correlating with TPM plasma concentration (Cp) and frequency of epileptic seizures in adult patients with drug‐resistant epilepsy.

Thirty‐nine patients were included in the study. However, two were excluded because of changes in liver function tests, thus 37 patients remained in the study (Figure ). Demographics, diagnosis of epilepsy, epileptic seizure type, frequency of seizures, and adverse events (AE) score are given in Table .

Demographic and clinical characteristics of the 37 patients with drug‐resistant epilepsy using topiramate

The most prescribed pharmacotherapeutic regimen was the association of TPM, carbamazepine (CBZ) and clobazam (CLB), with a significantly higher number of patients in this regimen (P = .022). Among the patients in combination therapy, 75.7% used three or more AEDs (Table ). Regarding the potential pharmacokinetic interactions, it was observed that 32 patients (86.5%) used a pharmacotherapeutic regimen in which TPM was associated with metabolism inducer drugs of TPM (MID‐TPM).

Pharmacotherapeutic characteristics of the 37 patients with drug‐resistant epilepsy using topiramate

1 AEDs, antiepileptic drugs, CBZ, carbamazepine; CLB, clobazam; LTG, lamotrigine; PB, Phenobarbital; PHT, phenytoin; TPM, topiramate; VPA, valproic acid.

2 Clonazepam, gabapentin and oxcarbazepine.

The mean TPM Cp found was 3.21 μg/mL (standard deviation (SD) 2.76; Table ). When correlating TPM Cp and the dose prescribed for the patients, a linear relationship between the two variables can be observed (r = .75, P < .001; Figure ). The means for MID‐TPM Cp are reported in Table . Regarding MID‐TPM Cp the mean CBZ, PB, LTG and VPA Cp were within the therapeutic range, however 10.7% (n = 03), 20.0% (n = 01), 50.0% (n = 04) and 66.6% (n = 2) of the patients, respectively, presented Cp below the lower limit of the recommended range.

Topiramate plasma concentration profile of the 37 patients with drug‐resistant epilepsy

• 3 NA, Not applicable; Min, minimum; Max, maximum.
• 4 Lower detectable concentration value detectable by the method.
• 5 Refers to absolute number.

Metabolism inducer drugs of topiramate plasma concentration profile of the 32 patients with drug‐resistant epilepsy

• 6 Min, minimum; Max, maximum.
• 7 Patients used a pharmacotherapeutic regimen in which TPM was associated with metabolism inducer drugs of TPM.

In this study, 62.2% of the patients were considered adherent to the drug treatment, and the mean Cp of the adherent patients was 3.38 μg/mL (SD 3.13) and non‐adherent patients 2.93 μg/mL (SD 2.12) [mean difference = −0.45, P = .605 (CI 95% −2.20, 1.30)], and the mean TPM dose was 5.27 and 4.58 mg/kg/d respectively [mean difference = 0.69, P = .417 (95% CI −1.01 and 2.39)].

When analyzing the patients in groups that use MID‐TPM and those who do not (non‐MID‐TPM), the mean TPM Cp was 2.60 μg/mL (SD 1.89) and 7.07 μg/mL (SD 4.42) respectively, with the mean difference between the groups being 4.46 μg/mL (P < .001 (95% CI 2.19, 6.73); Figure ). Likewise, there was evidence of a difference between the mean TPM dose used between the groups of patients who used inducers and those who did not, being 4.52 (SD 2.23) and 6.92 (SD 3.13) mg/kg/d respectively (mean difference = 2.39, P = .041 (95% CI 0.99, 4.69); Figure ).

Also, calculating the ratio between Cp and TPM dose of patients using inducers and those who did not, the mean values were 0.57 and 0.99 respectively, with the mean difference between the groups of 0.42 (P = .004 (95% CI 0.15, 0.69)).

No evidence of difference was found regarding age (P = .503), gender (P = .348), adherence to drug treatment (P = 1.00), number of AEDs used (P = .078), and AE score (P = .210) in patients who used and did not use MID‐TPM.

Therefore, the multiple linear regression model adjusted for the patients’ age was used to verify the possible influence of the dose and the use of inducers on TPM Cp. There was evidence that Cp variability was related to the dose and the use of inducers (R2 0.69, F 23.87, P < .001), so it was estimated that the increase of 1.0 mg/kg/d promoted an increase of 0.68 μg/mL in TPM Cp, while the use of inducers predicted a reduction of 2.97 μg/mL.

Among the 31 patients presenting Cp <5 μg/mL, 29 (93.6%) used AEDs inducing the metabolism of TPM, with 24 and 5 patients using one and two inducers, respectively. When comparing Cp of the patients of these two groups, there was no evidence that the mean TPM Cp was different between them, with a mean difference of −0.64 μg/mL (P = .314 (95% CI −1.93, 0.64)).

The other two patients who had Cp < 5 μg/mL and did not use metabolism inducing AEDs, used a lower dose than patients who had Cp within the therapeutic interval.

Furthermore, the 31 patients with Cp <5 μg/mL were separated into groups according to the number of AEDs used in the pharmacotherapeutic regimen. The comparison of mean TPM Cp of these groups by the ANOVA test did not find evidence of any difference between them (F = 0.92, P = .467).

The calculation of Cl/F showed a mean of 88.1 mL/h per kg (SD 65.1). Patients using MID‐TPM had a mean Cl/F of 94.5 mL/h per kg (SD 67.6) and patients who did not use inducers of 47.2 mL/h per kg (SD 17.6); there was no evidence of difference in mean Cl/F between the groups (mean difference = −6.4, P = .135).

Regarding epileptic seizures, when analyzing the patients who presented TPM Cp <5.0 and ≥5.0 μg/mL, the mean frequency was 16.5 (SD 0.7) and 6.5 (SD 1.0) seizures/mo respectively, and evidence was found that the mean number of seizures was higher among patients with TPM Cp <5.0 μg/mL (mean difference = 10.02, P < .001 (95% CI 7.52, 12.51)). Likewise, there was evidence of a difference between the frequency of seizures between patients using MID‐TPM and those who did not (mean difference = 16.06, P < .001 (95% CI 14.38, 17.74)) and the mean frequency of seizures in patients using inducers was 17.06 (SD 0.73) and those who did not was 1.00 (SD 0.44) seizures/mo.

The multiple Poisson regression model (Χ2 153.16, P < .001) predicts that TPM Cp and the use of MID‐TPM explain 94% of the variability of the frequency of seizures, whereby an increase of 1.0 μg/mL in Cp decreased the patient's chance of presenting seizures by 14% (OR = 0.86 (95% CI 0.83, 0.92)), and patients using MID‐TPM were about ten times more likely to present seizures than those who do not (OR = 10.5 (95% CI 4.3, 25.5)).

When analyzing the patients that use AED inducers in groups with Cp below and within the therapeutic range, the mean frequencies of seizures were 23.7 (SD 1.62) and 12.1 (SD 0.66) seizures/mo respectively, and evidence was found that the mean number of seizures was higher among patients with Cp below the therapeutic range (mean difference = 11.60, P < .001 (95% CI 8.17, 15.02)).

The knowledge of the Cp of a drug and the identification of the factors that can influence it have the objective to help in the success of the treatment of the patient with drug‐resistant epilepsy.

The prevalence of patients in combination therapy found in this study is in agreement with other studies conducted with patients with drug‐resistant epilepsy,[9] [10] [11] [12] [13] since one of the main strategies for the control of seizures in drug‐resistant epilepsy is the association of AEDs.

First‐generation AEDs are effective with a high level of scientific evidence in the control of partial seizures in adults, which justifies the prevalence of prescription of these drugs. In addition, despite the possibility of pharmacokinetic interactions, in the case of patients with drug‐resistant epilepsy, the association of AEDs is considered safer than sequential monotherapy, especially in cases of severe and frequent epileptic seizures. Therefore first‐generation AEDs are not replaced in the pharmacotherapeutic regimen, which may lead to the use of more AEDs by these patients.[2] [14] [15]

When assessing and analyzing the ratio between Cp and TPM dose of the groups of patients using and not using MID‐TPM, the ratio of the first group being lower than the second group was observed. However, when analyzing the variables of this ratio it is observed that the group using the inducer obtained the lowest value because the values of Cp and dose are smaller than the group that does not use the inducer, when in fact patients using MID‐TPM should use a higher dose of this drug because of the influence of the inducers on the pharmacokinetics of TPM to maintain Cp within the therapeutic range.

The lower dose for the group that uses MID‐TPM could be justified by the occurrence of AE, but there was no evidence of a difference in the AE profile between the two groups, which does not justify the use of lower TPM doses for patients using AED inducers.

The results show that probably the dose and pharmacokinetic interactions with MID‐TPM, regardless of the number of associated inducers, were the two main factors for the high prevalence of patients with TPM Cp below the reference interval. These data corroborate with other studies that have also identified the influence of dose and the use of MID‐TPM on TPM Cp.[7] [13] [16] [17] [18]

The analysis of this study that the frequency of seizures was higher in the group of patients with Cp <5 μg/mL and in the group that uses MID‐TPM, should be evaluated with caution. The non‐use of inducers decreasing the patient's likelihood of presenting a seizure should be carefully analyzed considering the pharmacokinetics of the drugs. The efficacy of AED inducers on the management of epileptic seizures should be considered. In addition, for patients using AED inducers with Cp below the therapeutic range, the mean number of seizures/mo was greater than those with Cp within the therapeutic range. Therefore, the lower dose of TPM among patients using these AED inducers is probably the main cause of this result, since the AED association aims to increase the pharmacological response, always considering the risk of AE, pharmacokinetic, and pharmacodynamic interactions.[14] Due to this being a cross‐sectional study, it was not possible to associate the dose and the frequency of seizures in the regression model since the increase in dose may be a consequence of the increase in the number of seizures, and it is not possible to predict temporality of the event.

One of the limitations of this study was that the number of patients included did not allow individual evaluation of the influence of each MID‐TPM on the Cp of this drug. In addition, the number of patients with drug‐resistant epilepsy using TPM in monotherapy did not allow more specific conclusions regarding the drug, due to the influence of the other AEDs on the control of seizures and AE.

Currently, therapeutic monitoring of TPM is not performed in the General Hospital of the Ribeirão Preto Medical School, University of São Paulo (HCFMRP‐USP), this being the first study that presented the profile of TPM Cp.

In this study it was possible to establish the prescribed dose and the use of MID‐TPM influence the Cp of the drug. In addition, epileptic seizure control is probably affected by the low Cp of first‐line AEDs and the adjuvant in treatment, TPM, as well as the low TPM dose used by patients, and when due to the use of MID‐TPM the dose should be adjusted. In view of this, TPM therapeutic monitoring and other AEDs are suggested as an auxiliary tool in clinical practice for the optimization of pharmacotherapy and clinical response in patients with drug‐resistant epilepsy.

A cross‐sectional analytical study was developed between October 2013 and November 2014, in the epilepsy outpatient clinic at HCFMRP‐USP, Brazil. The local ethics committee approved the study (030/2014).

Patients’ records that have TPM prescribed were evaluated. Inclusion criteria were: (i) diagnosis of drug‐resistant epilepsy; (ii) age 18‐60 years; (iii) being on TPM in monotherapy or associated with other AEDs, for at least 10 days in unchanged dosages and intake intervals; (iv) not presenting a diagnosis of moderate to severe cognitive impairment; (v) not being pregnant; and (vi) not having liver, kidney, gastrointestinal, infectious or progressive diseases. The established exclusion criteria were clinically relevant changes in liver and kidney laboratory tests performed on the day of data collection.

Patients who agreed and signed the informed consent form were included in the study, and venous blood samples were collected into heparin‐containing Vacutainer tubes prior to the administration of the next dose of TPM (in the trough period). The following variables were analyzed: age, gender, body weight (kg), frequency of epileptic seizures (in the last 30 days), pharmacotherapeutic regimen with its respective doses (mg/kg/d), adherence to medication treatment, and AE score. Subsequently, the Cl/F of each patient was calculated. The diagnosis of epilepsy (classification, aetiology and type of epileptic seizure) was also checked in the patients’ records.

Adherence to medication treatment was evaluated by the Modified Morisky Scale,[19] and patients who obtained a percentage of correctness ≥80% were considered as adherent.

The evaluation of AE was performed through the Liverpool Adverse Events Profile,[20] [21] the test allows a score between 19 and 76 points to be achieved, where the higher the score, the higher the prevalence and the intensity of the AE.

A plasma aliquot (1 mL) was added with internal standard (IS) (primidone (PRM); Sigma‐Aldrich (St Louis, MO, USA), 100 μg/mL, 50 μL) and sodium borate buffer solution (pH 9.0, 0.3 mol/L, 1.5 mL). The extraction procedure occurred with the addition of extraction solvent (ethyl acetate, 4.0 mL) followed by mechanical agitation and centrifugation. The extractor phase (3.0 mL) was collected, evaporated under a gentle stream of room temperature compressed air. The dried residue was solubilized in 100 μL of methanol, and TPM Cp was measured by a gas chromatography mass spectrometry system (GCMS) A Shimadzu GCMS system (GCMS‐QP2010) coupled with a Shimadzu autosampler (AOC‐5000; Kyoto, Japan). The GC BP1 column had 0.25 μm film thickness, 3 m length and 0.25 mm internal diameter (SGE Analytical Science, Sydney, Australia). The initial oven temperature was 230°C for 3 minutes and heated to 290°C at 10°C/min. The final temperature of 290°C was maintained for 1 minute. The sample volume injected was 2 μL on the column in the split injection mode (1:40) with injector temperature of 250°C. The carrier gas was high purity helium with a head pressure of 1.6 kgf/cm2. The ion source and interface temperature were set to 250°C. The electron multiplier voltage was held at 70 eV. Ions monitored were m/z 146 for PRM (IS) and m/z 324 for TPM. The retention time was 3.9 min for PRM and 3.9 min for TPM.

This method was validated following the guideline of the European Medicines Agency.[22] Calibration curves were prepared by internal standardization considering the peak‐area ratios of TPM/IS, and the linear range was 0.1‐40.0 μg/mL of TPM in blank plasma (r > .99). The lower limit of quantification for this method was 0.1 μg/mL.

The qualitative variables were described in absolute and relative frequencies and the quantitative variables were described in mean and SD or median and interquartile range (IQR). The linear association between two quantitative variables was determined by the Pearson correlation coefficient (r).

The unpaired Student t test was used to compare mean TPM Cp, TPM dose, TPM Cp /dose ratio, age, AE score, and Cl/F between a group of patients in combination therapy with MID‐TPM (CBZ, PB, PHT, VPA) (MID‐TPM group) and a group of patients on monotherapy and combination therapy with drugs that do not induce the metabolism of TPM (non‐MID‐TPM group). Likewise, the mean TPM Cp and the mean TPM dose were compared by dividing the patients into the following groups: adherent and non‐adherent to the drug treatment and using one and two MID‐TPM.

Fisher's exact test was used to identify differences in gender and adherence to drug therapy between the MID‐TPM and non‐MID‐TPM groups.

The frequency of patients’ epileptic seizures was compared between groups using the Poisson model, using the following three analyzes: patients with TPM Cp ≥ 5 μg/mL and TPM Cp < 5 μg/mL, based on the lowest value of the therapeutic range; patients in the MID‐TPM and the non‐MID‐TPM groups; and patients using AED inducers with Cp both below and within the therapeutic range.

The ANOVA test was performed to compare TPM Cp among the groups formed by the number of AEDs used by the patients.

The linear regression model was adjusted to analyze the association of TPM Cp with dose, use of MID‐TPM, and age. Meanwhile, the Poisson regression model verified the association of TPM Cp and the use of MID‐TPM with the frequency of epileptic seizures.

The level of significance (α) was set at 5% and statistical analyzes were performed with the assistance of the Statistical Package for Social Sciences (SPSS Inc., version 17.1.0).

PHOTO (COLOR): Flow diagram of patient recruitment. HCFMRP‐USP: General Hospital of the Ribeirão Preto Medical School, University of São Paulo; TPM: topiramate

PHOTO (COLOR): Association between plasma concentration and topiramate dose (mg/kg/d) of the 37 patients with drug‐resistant epilepsy included in the study

PHOTO (COLOR): Topiramate plasma concentration values in patients not using metabolism inducer drugs of topiramate (non‐MID‐TPM) and in patients using metabolism inducer drugs of topiramate (MID‐TPM). Horizontal lines indicate the mean topiramate plasma concentration of the groups

PHOTO (COLOR): Topiramate dose values in groups that do not use metabolism inducer drugs of topiramate (non‐MID‐TPM) and those who do use (MID‐TPM). Horizontal lines indicate the mean topiramate dose of the groups