Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Jan;42(1):79-91.
doi: 10.1007/s11095-025-03817-3. Epub 2025 Jan 22.

Prediction Trough Concentrations of Valproic Acid Among Chinese Adult Patients with Epilepsy Using Machine Learning Techniques

Nannan Yao  1 , Qiongyue Zhao  1 , Ying Cao  1 , Dongshi Gu  2 , Ning Zhang  3
Affiliations

Prediction Trough Concentrations of Valproic Acid Among Chinese Adult Patients with Epilepsy Using Machine Learning Techniques

Nannan Yao et al. Pharm Res. 2025 Jan.

Abstract

Objective: This study aimed to establish an optimal model based on machine learning (ML) to predict Valproic acid (VPA) trough concentrations in Chinese adult epilepsy patients.

Methods: A single-center retrospective study was carried out at the Jinshan Hospital affiliated with Fudan University from January 2022 to December 2023. A total of 102 VPA trough concentrations were split into a derivation cohort and a validation cohort at a ratio of 8:2. Thirteen ML algorithms were developed using 27 variables in the derivation cohort and were filtered by the lowest mean absolute error (MAE) value. In addition, feature selection was applied to optimize the model.

Results: Ultimately, the extra tree algorithm was chosen to establish the personalized VPA trough concentration prediction model due to its best performance (MAE = 13.08). The SHapley Additive exPlanations (SHAP) plots were used to visualize and rank the importance of features, providing insights into how each feature influences the model's predictions. After feature selection, we found that the model with the top 9 variables [including daily dose, last dose, uric acid (UA), platelet (PLT), combination, gender, weight, albumin (ALB), aspartate aminotransferase (AST)] outperformed the model with 27 variables, with MAE of 6.82, RMSE of 9.62, R2 value of 0.720, relative accuracy (±20%) of 61.90%, and absolute accuracy (±20 mg/L) of 90.48%.

Conclusion: In conclusion, the trough concentration prediction model for VPA in Chinese adult epileptic patients based on the extra tree algorithm demonstrated strong predictive ability which is valuable for clinicians in medication guidance.

Keywords: chinese adult epileptic patients; machine learning; prediction; trough concentration; valproic acid.

PubMed Disclaimer

Conflict of interest statement

Declarations. Ethics Statement: The study involving humans was approved by the Hospital Ethics Committee of the Jinshan Hospital affiliated with Fudan University (Approval No. JIEC 2024-S67). The study was performed in accordance with the Declaration of Helsinki. Written informed consent from patients was waived by the Hospital Ethics Committee. Conflict of Interest: The authors declare no competing interests in this work.

References

    1. Vossler DG, Weingarten M, Gidal BE. Summary of antiepileptic drugs available in the United States of America: working toward a world without epilepsy. Epilepsy Curr. 2018;18(4 Suppl 1):1–26. - PubMed - PMC
    1. Chen S, Wu H, Klebe D, et al. Valproic acid: a new candidate of therapeutic application for the acute central nervous system injuries. Neurochem Res. 2014;39(9):1621–33. - PubMed
    1. Shen X, Chen X, Lu J, et al. Pharmacogenetics-based population pharmacokinetic analysis and dose optimization of valproic acid in Chinese southern children with epilepsy: effect of ABCB1 gene polymorphism. Front Pharmacol. 2022;13:1037239. - PubMed - PMC
    1. Pascali JP, Giorgetti A, Barone R, et al. Valproic acid determination by liquid chromatography coupled to mass spectrometry (LC-MS/MS) in whole blood for forensic purposes. Drug Test Anal. 2023;15(1):128–33. - PubMed
    1. Mei S, Feng W, Zhu L, et al. Effect of CYP2C19, UGT1A8, and UGT2B7 on valproic acid clearance in children with epilepsy: a population pharmacokinetic model. Eur J Clin Pharmacol. 2018;74(8):1029–36. - PubMed

MeSH terms

Supplementary concepts