Nonlinear mixed effect (NLME) models are the gold standard for the analysis of patient response following drug exposure. However, these types of models are complex and time‐consuming to develop. There is great interest in the adoption of machine‐learning methods, but most implementations cannot be reliably extrapolated to treatment strategies outside of the training data. In order to solve this problem, we propose the deep compartment model (DCM), a combination of neural networks and ordinary differential equations. Using simulated datasets of different sizes, we show that our model remains accurate when training on small data sets. Furthermore, using a real‐world data set of patients with hemophilia A receiving factor VIII concentrate while undergoing surgery, we show that our model more accurately predicts a priori drug concentrations compared to a previous NLME model. In addition, we show that our model correctly describes the changing drug concentration over time. By adopting pharmacokinetic principles, the DCM allows for simulation of different treatment strategies and enables therapeutic drug monitoring.
Aims The aim of this study was to investigate the population pharmacokinetics (PK) of clonidine in intensive care unit (ICU) patients in order to develop a dosing regimen for sedation. Methods We included 24 adult mechanically ventilated, sedated patients from a mixed medical and surgical ICU. Intravenous clonidine was added to standard sedation in doses of 600, 1200 or 1800 μg/d. Within each treatment group, 4 patients received a loading dose of half the daily dose administered in 4 hours. Patients gave an average of 12 samples per individual. In total, 286 samples were available for analysis. Model development was conducted with NONMEM and various covariates were tested. After modelling, doses to achieve a target steady‐state plasma concentration of >1.5 μg/L were explored using stochastic Monte Carlo simulations for 1000 virtual patients. Results A 2‐compartment model was the best fit for the concentration‐time data. Clearance (CL) increased linearly with 0.213%/h; using allometric scaling, body weight was a significant covariate on the central volume of distribution (V1). Population PK parameters were: CL 17.1 (L/h), V1 124 (L/70 kg), intercompartmental CL 83.7 (L/h), and peripheral volume of distribution 178 (L), with 33.3% CV interindividual variability on CL and 66.8% CV interindividual variability on V1. Simulations revealed that a maintenance dose of 1200 μg/d provides target sedation concentrations of >1.5 μg/L in 95% of the patients. Conclusion A population PK model for clonidine was developed in an adult ICU. A dosing regimen of 1200 μg/d provided a target sedation concentration of >1.5 μg/L.
In population pharmacokinetic (PK) models, interindividual variability is explained by implementation of covariates in the model. The widely used forward stepwise selection method is sensitive to bias, which may lead to an incorrect inclusion of covariates. Alternatives, such as the full fixed effects model, reduce this bias but are dependent on the chosen implementation of each covariate. As the correct functional forms are unknown, this may still lead to an inaccurate selection of covariates. Machine learning (ML) techniques can potentially be used to learn the optimal functional forms for implementing covariates directly from data. A recent study suggested that using ML resulted in an improved selection of influential covariates. However, how do we select the appropriate functional form for including these covariates? In this work, we use SHapley Additive exPlanations (SHAP) to infer the relationship between covariates and PK parameters from ML models. As a case-study, we use data from 119 patients with hemophilia A receiving clotting factor VIII concentrate peri-operatively. We fit both a random forest and a XGBoost model to predict empirical Bayes estimated clearance and central volume from a base nonlinear mixed effects model. Next, we show that SHAP reveals covariate relationships which match previous findings. In addition, we can reveal subtle effects arising from combinations of covariates difficult to obtain using other methods of covariate analysis. We conclude that the proposed method can be used to extend ML-based covariate selection, and holds potential as a complete full model alternative to classical covariate analyses. Study Highlights WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC?Covariate selection in pharmacokinetic (PK) modeling is a complex process and is sensitive to bias. Machine-learning (ML) techniques might help to simplify and potentially improve this process, but are difficult to interpret as is.
Ketorolac is a nonsteroidal anti-inflammatory racemic drug with analgesic effects only attributed to its S-enantiomer. The aim of this study is to quantify enantiomer-specific maturational pharmacokinetics (PK) of ketorolac and investigate if the contribution of both enantiomers to the total ketorolac concentration remains equal between infants and adults or if a change in target racemic concentration should be considered when applied to infants. Methods: Data were pooled from 5 different studies in adults, children and infants, with 1020 plasma concentrations following single intravenous ketorolac administration. An allometry-based enantiomer-specific population PK model was developed with NONMEM 7.3. Simulations were performed in typical adults and infants to investigate differences in Sand R-ketorolac exposure. Results: Sand R-ketorolac PK were best described with a 3-and a 2-compartment model, respectively. The allometry-based PK parameters accounted for changes between populations. No maturation function of ketorolac clearance could be identified. All model parameters were estimated with adequate precision (relative standard error <50%). Single dose simulations showed that a previously established analgesic concentration at half maximal effect in adults of 0.37 mg/L, had a mean S-ketorolac concentration of 0.057 mg/L, but a mean S-ketorolac concentration of 0.046 mg/L in infants. To match the effective adult S-ketorolac-concentration (0.057 mg/L) in typical infants, the EC 50-racemic should be increased to 0.41 mg/L. Conclusion: Enantiomer-specific changes in ketorolac PK yield different concentrations and Sand R-ketorolac ratios between infants and adults at identical racemic concentrations. These PK findings should be considered when studies on maturational pharmacodynamics are considered.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.