Abstracts

Rationale: Poor medication adherence is a major contributor to increased morbidity and mortality, however it is often under-addressed in clinical practice. For patients with epilepsy, this may result in worse seizure control. Due to ethical concerns, it is impossible to study the impact of medication nonadherence in clinical trials; however, our lab has developed a novel medication-in-food automated delivery system to evaluate the effect of anti-seizure medication (ASM) nonadherence on seizure control in preclinical animal models. Perampanel (PER) and levetiracetam (LEV) are clinically approved ASMs, however, their efficacy has not yet been studied in this context. Therefore, we investigated their antiseizure efficacy in the post-SE rat following chronic oral ASM dosing in a fully adherent and variably nonadherent paradigm.

Methods: Low-dose systemic kainic acid (KA) was used to induce SE in adult male Sprague-Dawley rats approximately 3 weeks prior to implanting cortical EEG electrodes. Following surgical recovery and video EEG confirmation of spontaneous recurring seizures, animals were enrolled on placebo (PCBO) chow for 4 weeks (15 g/kg, p.o., q.i.d.) to establish their baseline seizure burden. Animals were then randomly assigned to the maximally tolerated dose of PER (2.5 mg/kg q.i.d, p.o.) or LEV (75 mg/kg, q.i.d., p.o.) either 100% or randomly 50% of the time during a 4-week treatment phase. Total seizure burden was monitored for each animal and the temporal relationship between a missed dose and seizure occurrence was evaluated over the course of the study. Weekly plasma samples were collected to monitor systemic levels of PER or LEV throughout the intervention phase of the study.

Results: Complete adherence (e.g., 100%) to PER or LEV results in better seizure control than variable nonadherence. Animals receiving 100% PER had a median seizure reduction of 27% compared to baseline, whereas 100% LEV animals had a median reduction of 49%. Conversely, animals receiving 50% PER or 50% LEV had a median increase of 31% and 85% in seizure frequency, respectively. Systemic concentrations PER and LEV were significantly elevated in fully adherent animals compared to nonadherent animals. In all animals, total plasma concentrations were significantly correlated with the degree of seizure control. No significant relationship was identified between a single missed dose of PER and the risk of seizure occurance, however, early data suggests a greater risk of a seizure if the first, second, or third dose of LEV is missed immediately preceding that event.

Conclusions: Preliminary analyses suggest that the relationship between a missed dose and a seizure cannot be explained by the pharmacokinetics (PK) of PER; however, there may be a PK dependence between missed LEV and seizure occurrence. While full adherence to PER or LEV provides better seizure control than variable nonadherence, PER may provide a certain forgiveness factor if a dose is missed within a 24-h period.  

Funding: Funding was provided by a grant from Eisai Inc.