K. Mizuno,1 E. V. Capparelli,2 T. Fukuda,1 M. Dong,1 A. A. Vinks,1 T. A. Glauser3; 1Division of Clinical Pharmacology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, 2Department of Pediatrics, University of California San Diego, La Jolla, CA, 3Division of Pediatric Neurology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH

BACKGROUND: Ethosuximide (ETX) has recently been identified as the preferred first choice in childhood absence epilepsy (CAE) (Glauser et al. NEJM, 362:790-9, 2010). Despite superior efficacy and fewer adverse attentional effects, up to 50% of ETX treated patients in this randomized trial failed to the treatment. ETX PK-PD relationship needs to be clarified to investigate ETX target concentrations for further individualized dosing. The purpose of this study was to explore the ETX exposure-response relationship using data collected as part of the randomized trial in children with newly diagnosed CAE.
METHODS: A total of 1,320 observations collected at 4 week intervals was available from 211 CAE patients (3.5-12.8 years). Dose titrated over 16 weeks until patients were seizure free. Individual area under the concentration-curves (AUCs) were determined by Bayesian estimation using population PK modeling (NONMEM 7.2). Seizure status data were available in 101 patients at week 16. The response probability was predicted by logistic regression analysis using predose concentrations and AUC (glm function, R 3.0.3).
RESULTS: High inter-patient variability was observed in ETX clearance (41.5% CV). Responders showed a wide range of AUCs (420 - 2437 mg∙h/L). 14 out of 17 non-responders at the maximum dose had AUCs that were significantly lower than in responders at the same dose (p<0.05). This is indicative of PK and PD variability impacting the response probability. The ETX PK-PD relationship could be described by the sigmoidal logistic regression relationship with a 50% of maximum effective AUC of 1230 mg∙h/L and an EC50 of 79 mg/L based on predose concentrations.
CONCLUSION: This is the first study to successfully identify the ETX PK-PD relationship using logistic regression modeling in CAE patients.