Abstracts

Rationale: It is estimated that approximately 30% of patients with epilepsy do not have adequate seizure control with their currently prescribed anti-seizure drug (ASD). Despite the fact that over 20 ASDs have come to market in the last 20 years, the percentage of therapy-resistant patients has remained relatively constant. Preclinical differentiation of promising compounds in clinically-relevant rodent models may provide a means to increase seizure control in pharmacoresistant individuals, and offer early-stage efficacy information. This work validates a new preclinical platform for screening investigational compounds for epilepsy that more closely mimics chronic clinical pharmacological treatment of newly diagnosed patients with epilepsy. Methods: Here we describe the results obtained using a novel experimental platform that utilizes an automated computer-controlled medication dosing system to deliver drug-in-food to newly diagnosed epileptic rats. Rats were implanted with EEG electrodes and treated with low-dose kainate to induce SE, which results in the development of epilepsy. Enrollment into a drug treatment paradigm began after the first observed spontaneous convulsive seizure. Two proof-of-concept studies were conducted. Study 1 evaluated the effect of increasing doses of carbamazepine (CBZ) on seizure control. Following enrollment, CBZ was administered at three different doses: 37.5 mg/kg, 75mg/kg, and 150mg/kg delivered q.i.d. Rats were randomid to receive each dosing regimen for 3 weeks per dose using a Latin-square protocol (9 weeks total, followed by a 3 week placebo treatment period). In Study 2, efficacy of CBZ (75mg/kg, q.i.d.) was compared to the performance of 2 commonly prescribed ASDs: valproic acid (VPA; 500mg/kg q.i.d.), or levetiracetam (LEV; 50mg/kg q.i.d.). Animals were first administered CBZ (n=9) for 3 weeks, and then randomly crossed over to either VPA (n=5) or LEV (n=4) for 3 weeks. Following this, animals were crossed over to the alternate anticonvulsant, i.e. LEV to VPA, for 3 weeks. Animals were then crossed back to CBZ for the final 3 weeks. Results: In Study 1, there was a significant inverse effect of CBZ dose on seizure severity during the 9-week study. In Study 2, there was a significant worsening of seizure burden observed during the LEV and VPA treatments when compared to the first CBZ treatment. There was no significant difference in seizure burden between LEV and VPA treatments, or between the first and final CBZ treatment epoch. Conclusions: Herein, we demonstrate the preclinical applicability of this platform for the screening and differentiation of potential anticonvulsant compounds through dose-response and cross-over administration studies. This automated feeder paradigm has preclinical utility for evaluating novel investigational therapies in chronically epileptic animals. While more work is necessary to fully characterize this system, the present results demonstrate the preclinical viability of this model as a drug differentiation platform for potential anticonvulsant compounds. Supported by NINDS#HHSN27120110029C.