BACKGROUND: There are several reports describing population pharmacokinetic (PPK) models of valproic acid (VPA). However, little was known in Chinese adult patients with epilepsy. The present study aimed to establish a PPK model for VPA in Chinese adult epileptic patients and to demonstrate its use for dose individualization. METHODS: Data were obtained from a prospective study of 199 adult epileptic patients at 5 hospitals. The trough concentrations at steady state were measured by fluorescence polarization immunoassay. Data were analyzed using the Nonlinear Mixed Effects Model software. The serum trough concentrations at steady state were also measured using samples (n = 20) collected prospectively from a different hospital from those providing the data for deriving the original model. These independent samples served as an evaluation group. RESULTS: The important determinants of apparent VPA clearance were daily dose, body weight, and combination with carbamazepine, phenytoin, or phenobarbital. The final model predicted the individualized doses accurately. A total of 85% of the trough concentrations in the evaluation group were accurately predicted by the final model, whereas the prediction errors of the other patients were all < ± 31%. CONCLUSIONS: A PPK model was developed to estimate the individual clearance for patients taking VPA and could be applied for individualizing doses in the target population.
BACKGROUND: There are several reports describing population pharmacokinetic (PPK) models of valproic acid (VPA). However, little was known in Chinese adult patients with epilepsy. The present study aimed to establish a PPK model for VPA in Chinese adult epilepticpatients and to demonstrate its use for dose individualization. METHODS: Data were obtained from a prospective study of 199 adult epilepticpatients at 5 hospitals. The trough concentrations at steady state were measured by fluorescence polarization immunoassay. Data were analyzed using the Nonlinear Mixed Effects Model software. The serum trough concentrations at steady state were also measured using samples (n = 20) collected prospectively from a different hospital from those providing the data for deriving the original model. These independent samples served as an evaluation group. RESULTS: The important determinants of apparent VPA clearance were daily dose, body weight, and combination with carbamazepine, phenytoin, or phenobarbital. The final model predicted the individualized doses accurately. A total of 85% of the trough concentrations in the evaluation group were accurately predicted by the final model, whereas the prediction errors of the other patients were all < ± 31%. CONCLUSIONS: A PPK model was developed to estimate the individual clearance for patients taking VPA and could be applied for individualizing doses in the target population.
Authors: Paulo Teixeira-da-Silva; Jonás Samuel Pérez-Blanco; Dolores Santos-Buelga; María José Otero; María José García Journal: Pharmaceutics Date: 2022-04-07 Impact factor: 6.525