Population Pharmacokinetics of Valproic Acid in Adult Chinese Epileptic Patients and its Application in an Individualized Dosage Regimen

Joint use of population pharmacokinetics and machine learning for prediction of valproic acid plasma concentration in elderly epileptic patients

BACKGROUND

Valproic acid (VPA) is a commonly used broad-spectrum antiepileptic drug. For elderly epileptic patients, VPA plasma concentrations have a considerable variation. We aim to establish a prediction model via a combination of machine learning and population pharmacokinetics (PPK) for VPA plasma concentration.

METHODS

A retrospective study was performed incorporating 43 variables, including PPK parameters. Recursive Feature Elimination with Cross-Validation was used for feature selection. Multiple algorithms were employed for ensemble model, and the model was interpreted by Shapley Additive exPlanations.

RESULTS

The inclusion of PPK parameters significantly enhances the performance of individual algorithm model. The composition of categorical boosting, light gradient boosting machine, and random forest (7:2:1) with the highest R2 (0.74) was determined as the ensemble model. The model included 11 variables after feature selection, of which the predictive performance was comparable to the model that incorporated all variables.

CONCLUSIONS

Our model was specifically tailored for elderly epileptic patients, providing an efficient and cost-effective approach to predict VPA plasma concentration. The model combined classical PPK with machine learning, and underwent optimization through feature selection and algorithm integration. Our model can serve as a fundamental tool for clinicians in determining VPA plasma concentration and individualized dosing regimens accordingly.

Population pharmacokinetics of valproic acid in children with epilepsy: Implications for dose tailoring when switching from oral syrup to sustained-release tablets

Significant pharmacokinetic (PK) differences exist between different forms of valproic acid (VPA), such as syrup and sustained-release (SR) tablets. This study aimed to develop a population pharmacokinetic (PopPK) model for VPA in children with epilepsy and offer dose adjustment recommendation for switching dosage forms as needed. The study collected 1411 VPA steady-state trough concentrations (Ctrough) from 617 children with epilepsy. Using NONMEM software, a PopPK model was developed, employing a stepwise approach to identify possible variables such as demographic information and concomitant medications. The final model underwent internal and external evaluation via graphical and statistical methods. Moreover, Monte Carlo simulations were used to generate a dose tailoring strategy for typical patients weighting 20-50 kg. As a result, the PK characteristics of VPA were described using a one-compartment model with first-order absorption. The absorption rate constant (ka) was set at 2.64 and 0.46 h-1 for syrup and SR tablets. Body weight and sex were identified as significant factors affecting VPA's pharmacokinetics. The final PopPK model demonstrated acceptable prediction performance and stability during internal and external evaluation. For children taking syrup, a daily dose of 25 mg/kg resulted in the highest probability of achieving the desired target Ctrough, while a dose of 20 mg/kg/day was appropriate for those taking SR tablets. In conclusion, we established a PopPK model for VPA in children with epilepsy to tailor VPA dosage when switching between syrup and SR tablets, aiming to improve plasma VPA concentrations fluctuations.

Population Pharmacokinetics and Exposure-Safety of Lipophilic Conjugates Prodrug DP-VPA in Healthy Chinese Subjects for Dose Regime Exploring

Phospholipid-valproic acid (DP-VPA)is a prodrug for treating epilepsy. The present study explored the pharmacokinetics (PK) and exposure safety of DP-VPA to provide a basis for future studies exploring the safe dosage and therapeutic strategies for epilepsy. The study included a randomized placebo-controlled dose-escalation tolerance evaluation trial and a randomized triple crossover food-effect trial in healthy Chinese volunteers. A population pharmacokinetic (PopPK) model was established to analyze the PK of DP-VPA and active metabolite VPA. The exposure safety was assessed with the adverse drug reaction (ADR) in CNS. The PopPK of DP-VPA and metabolite VPA fitted a two-compartment model coupling one-compartment with Michaelis-Menten metabolite kinetics and first-order elimination. The absorption processes after single oral administration of DP-VPA tablet demonstrated nonlinear characteristics, including 0-order kinetic phase and time-dependent phase fitting Weibull distribution. The final model indicated that the DP-VPA PK was significantly affected by dosage and food. The exposure-safety relationship demonstrated a generalized linear regression; mild/moderate ADRs occurred in some subjects with 600 mg and all subjects with 1500 mg of DP-VPA, and no severe ADRs were reported up to 2400 mg. In conclusion, the study established a PopPK model describing the processing of DP-VPA and VPA in healthy Chinese subjects. DP-VPA showed good tolerance after a single dose of 600-2400 mg with nonlinear PK and was affected by dosage and food. Based on the association between neurological ADRs and higher exposure to DP-VPA by exposure-safety analysis, 900-1200 mg was recommended for subsequent study of safety and clinical effectiveness.

Handling Delayed or Missed Dose of Antiseizure Medications: A Model-Informed Individual Remedial Dosing

BACKGROUND AND OBJECTIVES

Delayed or missed antiseizure medications (ASMs) doses are common during long-term or lifelong antiepilepsy treatment. This study aims to explore optimal individualized remedial dosing regimens for delayed or missed doses of 11 commonly used ASMs.

METHODS

To explore remedial dosing regimens, Monte Carlo simulation was used based on previously identified and published population pharmacokinetic models. Six remedial strategies for delayed or missed doses were investigated. The deviation time outside the individual therapeutic range was used to evaluate each remedial regimen. The influences of patients' demographics, concomitant medication, and scheduled dosing intervals on remedial regimens were assessed. RxODE and Shiny in R were used to perform Monte Carlo simulation and recommend individual remedial regimens.

RESULTS

The recommended remedial regimens were highly correlated with delayed time, scheduled dosing interval, and half-life of the ASM. Moreover, the optimal remedial regimens for pediatric and adult patients were different. The renal function, along with concomitant medication that affects the clearance of the ASM, may also influence the remedial regimens. A web-based dashboard was developed to provide individualized remedial regimens for the delayed or missed dose, and a user-defined module with all parameters that could be defined flexibly by the user was also built.

DISCUSSION

Monte Carlo simulation based on population pharmacokinetic models may provide a rational approach to propose remedial regimens for delayed or missed doses of ASMs in pediatric and adult patients with epilepsy.

Machine learning advances the integration of covariates in population pharmacokinetic models: Valproic acid as an example

Background and Aim: Many studies associated with the combination of machine learning (ML) and pharmacometrics have appeared in recent years. ML can be used as an initial step for fast screening of covariates in population pharmacokinetic (popPK) models. The present study aimed to integrate covariates derived from different popPK models using ML.

Methods: Two published popPK models of valproic acid (VPA) in Chinese epileptic patients were used, where the population parameters were influenced by some covariates. Based on the covariates and a one-compartment model that describes the pharmacokinetics of VPA, a dataset was constructed using Monte Carlo simulation, to develop an XGBoost model to estimate the steady-state concentrations (Css) of VPA. We utilized SHapley Additive exPlanation (SHAP) values to interpret the prediction model, and calculated estimates of VPA exposure in four assumed scenarios involving different combinations of CYP2C19 genotypes and co-administered antiepileptic drugs. To develop an easy-to-use model in the clinic, we built a simplified model by using CYP2C19 genotypes and some noninvasive clinical parameters, and omitting several features that were infrequently measured or whose clinically available values were inaccurate, and verified it on our independent external dataset.

Results: After data preprocessing, the finally generated combined dataset was divided into a derivation cohort and a validation cohort (8:2). The XGBoost model was developed in the derivation cohort and yielded excellent performance in the validation cohort with a mean absolute error of 2.4 mg/L, root-mean-squared error of 3.3 mg/L, mean relative error of 0%, and percentages within ±20% of actual values of 98.85%. The SHAP analysis revealed that daily dose, time, CYP2C19*2 and/or *3 variants, albumin, body weight, single dose, and CYP2C19*1*1 genotype were the top seven confounding factors influencing the Css of VPA. Under the simulated dosage regimen of 500 mg/bid, the VPA exposure in patients who had CYP2C19*2 and/or *3 variants and no carbamazepine, phenytoin, or phenobarbital treatment, was approximately 1.74-fold compared to those with CYP2C19*1/*1 genotype and co-administered carbamazepine + phenytoin + phenobarbital. The feasibility of the simplified model was fully illustrated by its performance in our external dataset.

Conclusion: This study highlighted the bridging role of ML in big data and pharmacometrics, by integrating covariates derived from different popPK models.

Comparative pharmacokinetics of valproic acid among Pakistani and South Korean patients: A population pharmacokinetic study

Purpose

The pharmacokinetics of valproic acid have been evaluated in a variety of populations however, the comparison in two different populations was yet to be reported. This study is aimed to compare the pharmacokinetics of valproic acid in Pakistani and South Korean patients.

Method

The therapeutic drug monitoring (TDM) data of valproic acid from 92 Pakistani patients with 218 samples was combined with the data of 99 South Korean patients with 335 samples in order to form a pooled dataset of 191 patients with 553 samples. Population pharmacokinetic model was developed on NONMEM® software by using first order conditional estimation method for estimation of pharmacokinetic parameters. The influence of different covariates including ethnicity was evaluated the stepwise covariate modelling. The final model was evaluated for predictive performance and robustness by using goodness of fit plots and bootstrap analysis respectively.

Results

The data was better described by one compartment model with first order elimination. The value for clearance (CL) of valproic in pooled data was 0.931 L/h with 43.4% interindividual variability (IIV) while volume of distribution (Vd) was 16.6 L with 22.3% IIV. In covariate analysis, ethnicity and body weight were significant covariates for CL while body weight was also significant for Vd.

Conclusion

A significant difference in CL of valproic acid among Pakistani and South Korean patients was observed. The model can be used for the dose tailoring of valproic acid based on ethnicity and body weight of Pakistani and South Korean patients.

Published population pharmacokinetic models of valproic acid in adult patients: a systematic review and external validation in a Chinese sample of inpatients with bipolar disorder

BACKGROUND

This study reviewed all published valproic acid (VPA) population pharmacokinetic (PPK) models in adult patients and assessed them using external validation methods to determine predictive performance.

METHODS

Thirteen published PPK models (labeled with letters A to M) not restricted to children were identified in PubMed, Embase, and Web of Science databases. They were evaluated in a sample totaling 411 serum concentrations from 146 adult inpatients diagnosed with bipolar disorder in a Chinese hospital. Serum concentrations of VPA were analyzed by validated ultra-performance liquid chromatography-tandem mass spectrometry. Performance was assessed by 4 tests (prediction-based diagnostics, visual predictive checks, normalized prediction distribution error, and Bayesian forecasting).

RESULTS

Models K and L, developed in large samples of Chinese and Thai patients, showed good performance in our Chinese dataset. Models H and J demonstrated good performance in Tests 2 and 3 of the 4 tests, respectively. Another 7 models exhibited intermediate performance. The models with the worst performance, F and M, could not be improved by Bayesian forecasting.

CONCLUSION

In our validation study the most important factors contributing to good performance were absence of children, Asian ethnicity, one-compartment models and inclusion of body weight and VPA dose in previously published models.

Population Pharmacokinetics of Valproic Acid in Pediatric and Adult Caucasian Patients

(1) Background: The aim of this study was to explore the valproic acid (VPA) pharmacokinetic characteristics in a large population of pediatric and adult Caucasian patients and to establish a robust population pharmacokinetic (PopPK) model. (2) Methods: A total of 2527 serum VPA samples collected from 1204 patients included in a therapeutic drug monitoring program were retrospectively analyzed. Patients were randomly assigned to either a model development group or an external evaluation group. PopPK analysis was performed on 1751 samples from 776 patients with NONMEM using a nonlinear mixed-effect modelling approach. The influence of demographic, anthropometric, treatment and comedication variables on the apparent clearance (CL/F) of VPA was studied. The bootstrap method was used to evaluate the final model internally. External evaluation was carried out using 776 VPA serum samples from 368 patients. (3) Results: A one-compartment model with first-order absorption and elimination successfully described the data. The final model included total body weight, age and comedication with phenytoin, phenobarbital and carbamazepine with a significant impact on VPA elimination. Internal and external evaluations demonstrated the good predictability of the model. (4) Conclusions: A PopPK model of VPA in Caucasian patients was successfully established, which will be helpful for model-informed precision dosing approaches in clinical patient care.

Population pharmacokinetics of valproic acid in adult Chinese patients with bipolar disorder

PURPOSE

To develop and validate a population pharmacokinetic (PPK) model of valproic acid (VPA) in adult Chinese patients with bipolar disorder, and provide guidance for individualized therapy in this population.

METHODS

A total of 1104 serum concentrations from 272 patients were collected in this study. The data analysis was performed using a nonlinear mixed-effects modeling approach. Covariates included demographic parameters, biological characteristics, and concomitant medications. Bootstrap validation (1000 runs), normalized prediction distribution error (NPDE), and external validation of 50 patients were employed to evaluate the final model.

RESULTS

A one-compartment model with first-order absorption and elimination was developed for VPA extended-release tablets. VPA clearance was significantly influenced by three variables: sex (12% higher in male patients), daily dose (increasing with the 0.13 exponent), and body weight (increasing with the 0.56 exponent). Typical values for the absorption rate constant (K_a), apparent clearance (CL/F), and apparent distribution volume (V/F) for a female patient weighing 70 kg administered VPA 1000 mg/day were 0.18 h^-1, 0.46 L/h, and 12.84 L, respectively. The results of model evaluation indicated a good stable and precise performance of the final model.

CONCLUSIONS

A qualified PPK model of VPA was developed in Chinese patients with bipolar disorder. This model could be used as a suitable tool for the personalization of VPA dosing for bipolar patients.

Remedial dosing recommendations for delayed or missed doses of valproic acid in patients with epilepsy based on Monte Carlo simulations

OBJECTIVE

Delayed or missed doses are unavoidable in the pharmacotherapy of epilepsy and significantly compromise the efficacy of antiepileptic drug treatment. An inappropriate remedial regimen can cause seizure relapse or serious adverse events. This study investigated the effect of delayed or missed doses on the pharmacokinetics (PK) of valproic acid (VPA) in patients with epilepsy and established remedial dosing recommendations for nonadherent patients.

METHODS

Monte Carlo simulations are based on all previous population pharmacokinetic models for pediatric, adult and elderly patients with epilepsy. The following four remedial strategies were investigated for each delayed dose: A) A partial dose or a regular dose is taken immediately; a regular dose is taken at the next scheduled time. B) The delayed dose was administered immediately, followed by a partial dose at the next scheduled time. C) The delayed dose and a partial dose are taken; the next scheduled time is skipped, and the regular regimen is resumed. D) Double doses are taken when missed one dose or two doses, and the regular regimen at the subsequent scheduled time is resumed.

RESULTS

The recommended remedial dose was related to the delay duration and daily dose. Remedial dosing strategies A and B were almost equivalent, whereas Strategy C was recommended when the delayed dose was close to the next scheduled dose. Strategy D was only suggested for delayed two doses.

CONCLUSION

Simulations provide quantitative insight into the remedial regimens for nonadherent patients, and clinicians should select the optimal regimen for each patient based on the individual's status.

Predictive ability of published population pharmacokinetic models of valproic acid in Thai manic patients

WHAT IS KNOWN AND OBJECTIVE

Population pharmacokinetic (PopPK) models of valproic acid (VPA) have been developed to aid individualized drug dosing, but most of these have been based on the treatment of epileptic patients and recent evidence shows that VPA clearance (CL_VPA ) in manic patients differs from that in epileptic patients. In the light of this, the predictive ability of selected VPA PopPK models based on epileptic patients was assessed to determine whether they could be used with patients with mania.

METHODS

VPA PopPK models that were based on the treatment of epileptic patients and developed using a non-linear mixed-effect approach with a one-compartment structure were selected and used to predict the VPA concentrations of a validation data set. The mean absolute prediction error (MAPE) and root mean square error (RMSE) were used to assess the accuracy and precision of the model.

RESULTS

The validation data set consisted of 235 Thai manic patients with a mean age of 39.6 years and a mean weight of 62.8 kg. Five models were selected to predict VPA concentrations in patients suffering from mania, and these were labelled A, C, E, F and G. The results showed that all models sufficiently predicted VPA concentrations in patients with mania, and of the models studied, G provided the most accurate and precise predictions, with MAPE and RMSE of 23% and 29.75, respectively.

WHAT IS NEW AND CONCLUSION

VPA PopPK models developed using patients with epilepsy can also be used for individualized dosing of patients with mania, but before implementation, the accuracy of these models' predictions should be assessed in the target population.

Estimation of apparent clearance of valproic acid in adult Saudi patients

Background Valproic acid is one of several antiepileptic medications requiring therapeutic drug monitoring due to its complex and wide pharmacokinetic interindividual variability. Objective The objective of this study was to determine the population pharmacokinetics of valproic acid in adult Saudi patients and to identify factors that explain its pharmacokinetic variability. Setting Tertiary referral teaching hospital, Riyadh, Saudi Arabia. Method A retrospective chart review was performed at King Saud University Medical City of patients who received oral valproic acid. The population pharmacokinetic models were developed using Monolix 4.4. After development of the base model, we investigated several covariates including age, sex, weight, total daily dose, and cotherapy with carbamazepine and phenytoin. Main outcome measures the pharmacokinetic parameters of valproic acid and the variables that contributing towards its inter-individual variability. Results The analysis included a total of 54 valproic acid plasma concentrations from 54 patients (42.5% male). The data were sufficiently described by a one-compartment model with linear absorption and elimination processes. Average parameter estimates for valproic acid apparent clearance (CL/F) and apparent volume of distribution (V/F) were 0.14 L/h and 37.7 L (fixed), respectively. The inter-individual variability (coefficients of variation) in CL/F was 12%. The most significant covariates for valproic acid CL/F were age, body weight, total daily dose, and cotherapy with carbamazepine and phenytoin. Conclusion This model showed significant inter-individual variability between subjects. Our findings showed that patient age, body weight, total daily dose, and cotherapy with carbamazepine and phenytoin are the most significant covariates of valproic acid clearance. Collectively, healthcare providers should take these factors in consideration for optimal valproic acid dosage regimen.

Population pharmacokinetics of valproic acid in epileptic children: Effects of clinical and genetic factors

Valproic acid (VPA) is a first-line anti-epileptic drug that is used in the treatment of generalized and partial seizures. Gene variants had been proved to influence the pharmacokinetics (PK) of VPA and contribute to its inter-individual variability (IIV). The aim of this study was to systematically investigate the effects of candidate gene variants (CYPs, UGTs, ABC transporters, and nuclear receptors) on VPA PK in Chinese children with epilepsy. A total of 1065 VPA serum trough concentrations at steady state were collected from 264 epileptic pediatric patients aged 3 months to 16 years. The population pharmacokinetic (PPK) model was developed using a nonlinear mixed effects modelling (NONMEM) approach. For the final PPK model, the oral clearance (CL/F) of VPA was estimated to be 0.259 L/h with IIV of 13.3%. The estimates generated by NONMEM indicated that the VPA CL/F was significantly influenced by patient body weight (increased by an exponent of 0.662), co-administration with carbamazepine (increased CL/F by 22%), and daily dose of VPA (increased by an exponent of 0.22). CL/F in patients with the LEPR rs1137101 variant (668 AG and GG genotypes) was much lower than in patients with the AA genotype (17.8% and 22.6% lower, respectively). However, none of the CYPs or UGTs gene variants was found to influence the PK of VPA in this study. Evaluation by bootstrap and normalized prediction distribution error (NPDE) showed that the final model was stable. The predictive performance was evaluated by goodness-of-fit (GOF) plots and visual predictive checks (VPC), and the results indicated satisfactory precision. Our model suggests a correlation between VPA CL/F and LEPR rs1137101 variants, which might be beneficial in the context of individual dose optimization.

A population pharmacokinetic model taking into account protein binding for the sustained-release granule formulation of valproic acid in children with epilepsy

PURPOSE

The objective of this work was to develop a population pharmacokinetic model for a prolonged-release granule formulation of valproic acid (VPA) in children with epilepsy and to determine the doses providing a VPA trough concentration (C_trough) within the target range (50-100 mg/L).

METHODS

Ninety-eight children (1-17.6 years, 325 plasma samples) were included in the study. The model was built with NONMEM 7.3. The probability to obtain C_trough between 50 and 100 mg/L was determined by the Monte Carlo simulations for doses of 20, 30, 40, and 60 mg/kg/day and body weights between 10 and 70 kg.

RESULTS

A one compartment model, with first-order absorption and flip-flop parameterization and linear elimination, but taking protein binding into account, was used to describe the data. Typical values for unbound VPA clearance and distribution volume were 6.24 L/h/70 kg and 130 L/h/70 kg respectively. Both parameters were related to body weight via allometric models. The highest probability to obtain a C_trough within the target range for 10-kg children was obtained with a 40 mg/kg daily dose, whereas daily doses of 30 and 20 mg/kg were found appropriate for 20 to 30- and ≥ 40-kg children respectively. However, for these same doses, the exposure to unbound VPA could differ by 40%.

CONCLUSIONS

If the present study supports the current dose recommendations of 20-30 mg/kg/day, except for children under 20 kg, who may need higher doses, it also highlights the need for further research on the pharmacokinetics/pharmacodynamic profile of unbound VPA.

A systematic review of population pharmacokinetics of valproic acid

AIMS

Population pharmacokinetics is an essential tool that helps guide individualized dosing regimens. The aims of this systematic review are to provide knowledge concerning population pharmacokinetics of valproic acid (VPA) and to identify factors influencing VPA pharmacokinetic variability.

METHODS

PubMed and Embase databases were systematically searched from inception to June, 2017. Relevant articles from reference lists were also included. All population pharmacokinetic studies of VPA conducted in humans and that employed a nonlinear mixed effect modelling approach were included in this review.

RESULTS

Twenty-six studies were included in this review. Most studies characterized VPA pharmacokinetics as a one-compartment model. Three studies reported a two-compartment model. Body weight, dose and age were significant predictors for VPA volume of distribution (V_d ). The estimated V_d for one-compartment models ranged from 8.4 to 23.3 l. For two-compartment models, peripheral volumes of distribution ranged from 4.08 to 42.1 l. Frequently reported significant predictors for VPA clearance (CL_VPA ) included body weight, VPA dose, concomitant medications, gender and age. The estimated CL_VPA ranged from 0.206 to 1.154 l h^-1 and the inter-individual variability ranged from 13.40 to 35.90%. Two studies reported population pharmacokinetics/pharmacodynamics of VPA in patients with epilepsy. Seventeen studies evaluated the performance of their final models.

CONCLUSIONS

Significant predictors influencing VPA pharmacokinetics as well as model methodologies are highlighted in this review. For clinical application, CL_VPA could be predicted using body weight, VPA dose, concomitant medications, gender or age. For future research, there is a knowledge gap regarding population pharmacokinetics/pharmacodynamics of VPA in a population other than epileptic patients.

Population pharmacokinetics analysis of intravenous busulfan in Chinese patients undergoing hematopoietic stem cell transplantation

There are several reports describing population pharmacokinetic (popPK) models of busulfan (BU). However, limited information is available in Chinese hematopoietic stem cell transplantation (HSCT) patients. The present study aimed to establish a popPK model of intravenous BU in Chinese HSCT patients for individualized drug therapy. The popPK model of BU was developed from a total of 284 concentration-time points from 53 patients. The effects of demographic and biochemical covariates were investigated by nonlinear mixed effect model (NONMEM) software. Plots, visual predictive check (VPC), bootstrap and normalized prediction distribution error (NPDE) were performed to determine the stability and the reliability of the final model. A one-compartment model with first-order elimination process was confirmed as the final structural model for BU. For a typical patient whose body surface area (BSA) is 1.7 m² , the population typical values of CL and Vd were 11.86 L/h, and 48.2 L, respectively. The result suggested BSA showed significant influence on CL and Vd (P<.001). Plots revealed the final model was performing a goodness fit. The steady rate verified by bootstrap was 100%, relative deviation was less than 4.00%, estimated value of final model was in the 95% confidence interval (CI). The VPC results showed the observed values were almost all positioned within the 5th and 95th CIs. The mean and variance of the NPDE were 0.0363 (Wilcoxon signed-rank test, 0.298) and 0.877 (Fisher variance test, 0.134; SW test of normality, 0.108), respectively. The global adjusted P value was 0.305, which indicated that the prediction of the BU popPK model was adequate. A physician-friendly Microsoft Excel-base tool was implemented using the final popPK model for designing individualized dosing regimens.

Determination of the Optimal Concentration of Valproic Acid in Patients with Epilepsy: A Population Pharmacokinetic-Pharmacodynamic Analysis

Valproic acid (VPA) is one of the most widely prescribed antiepileptic drugs for the treatment of epileptic seizures. Although it is well known that the doses of VPA and its plasma concentrations are highly correlated, the plasma concentrations do not correlate well with the therapeutic effects of the VPA. In this study, we developed a population-based pharmacokinetic (PK)-pharmacodynamic (PD) model to determine the optimal concentration of VPA according to the clinical characteristics of each patient. This retrospective study included 77 VPA-treated Japanese patients with epilepsy. A nonlinear mixed-effects model best represented the relationship between the trough concentrations of VPA at steady-state and an over 50% reduction in seizure frequency. The model was fitted using a logistic regression model, in which the logit function of the probability was a linear function of the predicted trough concentration of VPA. The model showed that the age, seizure locus, the sodium channel neuronal type I alpha subunit rs3812718 polymorphism and co-administration of carbamazepine, clonazepam, phenytoin or topiramate were associated with an over 50% reduction in the seizure frequency. We plotted the receiver operating characteristic (ROC) curve for the logit(Pr) value of the model and the presence or absence of a more than 50% reduction in seizure frequency, and the areas under the curves with the 95% confidence interval from the ROC curve were 0.823 with 0.793-0.853. A logit(Pr) value of 0.1 was considered the optimal cut-off point (sensitivity = 71.8% and specificity = 80.4%), and we calculated the optimal trough concentration of VPA for each patient. Such parameters may be useful to determine the recommended therapeutic concentration of VPA for each patient, and the procedure may contribute to the further development of personalized pharmacological therapy for epilepsy.