External evaluation of a published population pharmacokinetic model of valproic acid in Thai manic patients

Fixed parameters in the population pharmacokinetic modeling of valproic acid might not be suitable: external validation in Chinese adults with epilepsy or after neurosurgery

PURPOSE

This study aims to assess the predictive performance of published valproic acid (VPA) population pharmacokinetic (PPK) models using an external data set in Chinese adults with epilepsy or after neurosurgery.

METHODS

A total of 384 concentrations from 290 Chinese adults with epilepsy or after neurosurgery were used for external validation. Data on published VPA PPK models were extracted from the literature. Prediction-based diagnostics (such as F20 and F30), simulation-based diagnostics, and Bayesian forecasting were used to evaluate the predictability of models.

RESULTS

The results of prediction-based diagnostics of all models were unsatisfactory. Models B, F, and H showed the best prediction performance in simulation-based diagnostics and Bayesian forecasting, demonstrating superior precision and accuracy. Bayesian forecasting demonstrated significant improvements in the model predictability.

CONCLUSION

The published PPK models showed extensive variation in predictive performance for extrapolation among Chinese adults with epilepsy or after neurosurgery patients. Fixed parameters of Vd and Ka in the PPK modeling of VPA might be the reason for the unsatisfied predictive performance. Bayesian forecasting significantly improved model predictability and may help to individualize VPA dosing.

Valproic acid monitoring: Serum prediction using a machine learning framework from multicenter real-world data

BACKGROUND

Our study employs machine learning to predict serum valproic acid (VPA) concentrations, aiming to contribute to the development of non-invasive assays for therapeutic drug monitoring.

METHODS

Medical records from 2002 to 2019 were obtained from the Taiwan Chang Gung Research Database. Using various machine learning algorithms, we developed predictive models to classify serum VPA concentrations into two categories (1-50 μg/ml or 51-100 μg/ml) and predicted the exact concentration value. The models were trained on 5142 samples and tested on 644 independent samples. Accuracy was the main metric used to evaluate model performance, with a tolerance of 20 μg/ml for continuous variables. Furthermore, we identified important features and developed simplified models with fewer features.

RESULTS

The models achieved an average accuracy of 0.80-0.86 for binary outcomes and 0.72-0.88 for continuous outcome. Ten top features associated with higher serum VPA levels included higher VPA last and daily doses, bipolar disorder or schizophrenia spectrum disorder diagnoses, elevated levels of serum albumin, calcium, and creatinine, low platelet count, low percentage of segmented white blood cells, and low red cell distribution width-coefficient of variation. The simplified models had an average accuracy of 0.82-0.86 for binary outcome and 0.70-0.86 for continuous outcome.

LIMITATIONS

The study's predictive model lacked external test data from outside the hospital for validation.

CONCLUSIONS

Machine learning models have the potential to integrate real-world data and predict VPA concentrations, providing a promising tool for reducing the need for frequent monitoring of serum levels in clinical practice.

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.

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.