Pharmacokinetics And Individualized Administration Of Valproic Acid In Children With Epilepsy

Epilepsy is a common and diverse set of chronic neurological disorders of children. The mainstay of treatment of epilepsy is anticonvulsant medications. However, medication adherence is an important issue in the control of seizures. Delayed or missed doses occur often in the treatment of epileptic patients. The measurement of adherence of the regimen and provide practical recommendations for the patients and clinicians in the events of poor compliance, which were important in the management of epilepsy. Valporic acid (VPA) is the first-line antiepileptic drug used in the treatment of epilepsy. It is also used in combination with other antiepileptic drugs (AEDs) in some patients with epilepsy or epilepsy syndrome. The aim of this study were:①to establish the population pharmacokinetics model of VPA in Chinese epileptic children and investigate the impacts of physiology and combination of other AEDs on the clearance of VPA.②to develop the Bayesian decision method to measure the adherence of VPA based on the population pharmacokinetics model.③to study the effect of delayed and missed doses (poor compliance) on the pharmacokinetics of VPA and design corresponding rescue regimens.Serum samples at steady trough state (n=1108) were collected prospectively or retrospectively from904compliant epileptic patients during their routine clinical care from three medical center. Nonlinear mixed effect modeling software (NONMEM) was used to build the population pharmacokinetics model. The one compartment model with first order absorption and elimination was selected as the pharmacokinetic model. The absorption constant rate (Ka) of VPA was fixed to a literature value of1.9/h. The first-order conditional estimation method with η-ε interaction (FOCE-I) was used throughout the model-building procedure. Various statistical models describing the random residual variability were considered. The final population pharmacokinetics model for VPA was as follows: Age greater than3years:CL/F=0.256·(TBW/20.0)0236·1.50CBZ·exp(η1) Age less than or equal to3years:CL/F=0.218·(Age/2.0)0293·1.50CBZ·exp(η1)where CL/F represents apparent clearance, TBW represents total body weight, and CBZ represents carbamazepine. The between-subject variability (BSV) of CL/F (ωCL) were23%The additive residual variability (σ1) for three medical center were18.1,13.7and18.8mg/L, respectively.The sensitivity analysis results show that Ka has no significant impact on the estimation of pharmacokinetics parameters of VPA. The95%confident interval derived from bootstrap and log-likelihood profiling method were close to that calculated by NONMEM. The mean predict error of cross-validation were close to that from the original data. The numerical predictive check and visual predictive check indicated the good predictive ability of the model which can be used to generate simulated data for adherence study in the second and third part of this paper.The scenarios were based on various dosing events (N=1,2,3) and the corresponding number of scenarios were2,4,8. The concentration-time data of each scenarios were simulated based on the previously developed population pharmacokinetics model and in which the trough concentration data were collected to calculate the conditional probability in each concentration interval (1mg/L). The equiprobable prior distributions are used in the current work. For a given observation, the posterior probability were calculated using the Bayesian function based on the conditional probability and prior distributions, and decisions were made according the posterior probability. Moreover, the impact of30minutes window of last dose,1hour window of the closer dose of last dose, inter-subject variability, residual variability on the conditional probability and posterior probability were investigated. The decision of eight scenarios of N=3and four scenarios of N=2can be made for one time daily and two times daily regimen, respectively. Thirty minutes window of last dose,1hour window of the closer dose of last dose has no impact on the decision making. The low level of inter-subject variability and residual variability can improve the ability of making decision. The decision of various adherence scenarios can quickly got according to the simple computer program.Monte Carlo simulation was performed to generate VPA time-concentration profiles in various scenarios based on the population pharmacokinetic study in epilepsy patients. The scenarios were based on patients who were given multiple doses of VPA that ranged from100to500mg every12hours or from500to750mg one time daily. The therapeutic range of VPA for each scenario was estimated to assess the effect of delayed or missed doses and to design corresponding rescue regimens. Moreover, the impact of body weight, absorption rate and co-therapy with other antiepileptic drugs (carbamazepine) on the dosage recommendation were investigated in the event of poor compliance. The risk for a sub-therapeutic range of VPA was increased in a daily dose-dependent manner in both one and two times daily regimens when delayed or missed doses occurred. The effect of poor compliance is less significant on lower daily doses compared with the effect on higher daily doses. The dose recommendations, in the event of poor compliance, were time related and dose dependent. Patient body weight, absorption rate and co-therapy with carbamazepine had no significant impact on the dose recommendation. Moreover, taking a dose30minutes before or30minutes delayed from the scheduled time had no significant impact on the concentration profile of VPA. Patients should take the doses delay as soon as they remember, and partial missed doses may need to be taken near or at the next scheduled time. The pharmacokinetic simulation is a useful tool to analyze the impact of poor compliance and to provide rational information to patients in the case of delayed or missed doses.