Predicting blood-to-plasma concentration ratios of drugs from chemical structures and volumes of distribution in humans

Abstract

Despite their importance in determining the dosing regimen of drugs in the clinic, only a few studies have investigated methods for predicting blood-to-plasma concentration ratios (Rb). This study established an Rb prediction model incorporating typical human pharmacokinetics (PK) parameters. Experimental Rb values were compiled for 289 compounds, offering reliable predictions by expanding the applicability domain. Notably, it is the largest list of Rb values reported so far. Subsequently, human PK parameters calculated from plasma drug concentrations, including the volume of distribution (Vd), clearance, mean residence time, and plasma protein binding rate, as well as 2702 kinds of molecular descriptors, were used to construct quantitative structure–PK relationship models for Rb. Among the evaluated PK parameters, logVd correlated best with Rb (correlation coefficient of 0.47). Thus, in addition to molecular descriptors selected by XGBoost, logVd was employed to construct the prediction models. Among the analyzed algorithms, artificial neural networks gave the best results. Following optimization using six molecular descriptors and logVd, the model exhibited a correlation coefficient of 0.64 and a root-mean-square error of 0.205, which were superior to those previously reported for other Rb prediction methods. Since Vd values and chemical structures are known for most medications, the Rb prediction model described herein is expected to be valuable in clinical settings. Graphical abstract

The blood-to-plasma ratio and predicted GABAA-binding affinity of designer benzodiazepines

“…Calculations from the same source report a bloodto-plasma ratio of 0.51 ± 0.10 for diazepam which is also similar to the value derived in this work. Chlorpromazine had an experimental blood-to-plasma ratio of 1.43 ± 0.32 in this work compared to a literature range of 1.17-1.56 [23,28,29]. Quinine had an experimental blood-to-plasma ratio of 1.66 ± 0.52 in this work compared to a range of 1.46-2.00 [18,23].…”

“…Chlorpromazine had an experimental blood-to-plasma ratio of 1.43 ± 0.32 in this work compared to a literature range of 1.17-1.56 [23,28,29]. Quinine had an experimental blood-to-plasma ratio of 1.66 ± 0.52 in this work compared to a range of 1.46-2.00 [18,23].…”

“…Stock solutions were stored for no longer than 1 week. Appropriate volumes of working solutions were added to blood or plasma to yield a final compound concentration of 5 µM (final solvent concentrations were 0.1%) [23]. The treated blood or plasma samples were incubated for 1 h at 37 °C.…”

“…There is a large variation in reported blood-to-plasma ratios for many compounds in the literature. When blood-to-plasma ratios are not known, they are often assumed to be equal to 1.0 or equal to the blood-to-plasma ratios for other animals [23,35].…”

Novel QSAR Approach for a Regression Model of Clearance That Combines DeepSnap-Deep Learning and Conventional Machine Learning

“…Construction of QSAR prediction models is based mainly on molecular descriptors or fingerprints as features of compounds and conventional machine learning (ML) algorithms such as partial least squares regression, neural networks, random forest, support vector machines, and XGBoost. 4 − 9 In this study, we defined conventional ML as a method that uses molecular descriptors and fingerprints as features and applies algorithms other than deep learning (DL) algorithms. Among these conventional ML algorithms, XGBoost has been used successfully in cheminformatics, and for some parameters, it produced results that were as good as those obtained using DL algorithms.…”