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. 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 ( ) 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 genotypes and co-administered antiepileptic drugs. To develop an easy-to-use model in the clinic, we built a simplified model by using 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. 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, and/or variants, albumin, body weight, single dose, and genotype were the top seven confounding factors influencing the of VPA. Under the simulated dosage regimen of 500 mg/bid, the VPA exposure in patients who had and/or variants and no carbamazepine, phenytoin, or phenobarbital treatment, was approximately 1.74-fold compared to those with genotype and co-administered carbamazepine + phenytoin + phenobarbital. The feasibility of the simplified model was fully illustrated by its performance in our external dataset. This study highlighted the bridging role of ML in big data and pharmacometrics, by integrating covariates derived from different popPK models.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9621318 | PMC |
http://dx.doi.org/10.3389/fphar.2022.994665 | DOI Listing |
Br J Clin Pharmacol
December 2024
Department of Pharmacology, Toxicology and Pharmacovigilance, CHU de Limoges, Limoges, France.
Aims: Mycophenolic acid (MPA), the active component of enteric-coated mycophenolate sodium (EC-MPS), exhibits highly variable pharmacokinetics. Only a few population pharmacokinetic (popPK) models and Bayesian estimators (MAP-BE) exist for estimating MPA AUC and all in renal transplantation. This study aimed to develop a popPK model and MAP-BE for MPA AUC estimation using a limited sampling strategy (LSS) in solid organ transplant (SOT), haematopoietic stem cell (HSC) recipients and patients with autoimmune diseases (AID) on EC-MPS.
View Article and Find Full Text PDFClin Pharmacokinet
December 2024
Pharmacy Service, Division of Medicines, Hospital Clinic of Barcelona, Universitat de Barcelona, Barcelona, Spain.
In recent years, many population pharmacokinetic (popPK) models have been developed for echinocandins to better understand the pharmacokinetics (PK) of these antifungals. This comprehensive review aimed to summarize popPK models of echinocandins (micafungin, caspofungin, anidulafungin, and rezafungin), by focusing on dosage optimization to maximize the probability of attaining the PK/PD target proposed in special populations. A search in PubMed, Embase, Web of Science, and Scopus, supplemented by the bibliography of relevant articles, was conducted from inception to March 2024, including both observational and prospective trials.
View Article and Find Full Text PDFAAPS J
December 2024
Center of Clinical Pharmacology, The Third Xiangya Hospital, Central South University, Changsha, China.
NH600001 is a new general anaesthetic drug with a structure similar to etomidate. The objective of this study was to investigate the relationship between concentrations of NH600001 and sedation efficacy based on data from phase I-II studies and factors influencing the pharmacokinetics and pharmacodynamics of NH600001. The dataset consisted of 2 phase I studies in healthy subjects and 1 phase II study in patients undergoing gastroscopy.
View Article and Find Full Text PDFIntroduction: High-dose MTX is used to treat pediatric acute lymphoblastic leukemia (ALL). The drug has a low therapeutic index and a highly interindividual variability in systemic exposure. These characteristics necessitate dose adjustments and therapeutic drug monitoring protocols, while population pharmacokinetic (POP/PK) models may enable more precise drug dosing.
View Article and Find Full Text PDFAdv Ther
December 2024
Acadia Pharmaceuticals Inc., 12830 El Camino Real, Suite 400, San Diego, CA, 92130, USA.
Introduction: Weight-banded trofinetide dosing improved physician- and caregiver-rated efficacy measures and had acceptable tolerability in patients aged 2‒4 years (DAFFODIL study) and 5‒20 years (LAVENDER study) with Rett syndrome (RTT). Selection of weight-banded dosing regimens for these studies was based on population pharmacokinetic (popPK) modeling and exposure simulations. This study applied an updated popPK model to confirm steady-state trofinetide exposures achieved in DAFFODIL patients were within target range.
View Article and Find Full Text PDFEnter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!