Challenges and opportunities with modelling and simulation in drug discovery and drug development

Deep-PK: deep learning for small molecule pharmacokinetic and toxicity prediction

Evaluating pharmacokinetic properties of small molecules is considered a key feature in most drug development and high-throughput screening processes. Generally, pharmacokinetics, which represent the fate of drugs in the human body, are described from four perspectives: absorption, distribution, metabolism and excretion-all of which are closely related to a fifth perspective, toxicity (ADMET). Since obtaining ADMET data from in vitro, in vivo or pre-clinical stages is time consuming and expensive, many efforts have been made to predict ADMET properties via computational approaches. However, the majority of available methods are limited in their ability to provide pharmacokinetics and toxicity for diverse targets, ensure good overall accuracy, and offer ease of use, interpretability and extensibility for further optimizations. Here, we introduce Deep-PK, a deep learning-based pharmacokinetic and toxicity prediction, analysis and optimization platform. We applied graph neural networks and graph-based signatures as a graph-level feature to yield the best predictive performance across 73 endpoints, including 64 ADMET and 9 general properties. With these powerful models, Deep-PK supports molecular optimization and interpretation, aiding users in optimizing and understanding pharmacokinetics and toxicity for given input molecules. The Deep-PK is freely available at https://biosig.lab.uq.edu.au/deeppk/.

MolToxPred: small molecule toxicity prediction using machine learning approach

Different types of chemicals and products may exhibit various health risks when administered into the human body. For toxicity reasons, the number of new drugs entering the market through the conventional drug development process has been reduced over the years. However, with the advent of big data and artificial intelligence, machine learning techniques have emerged as a potential solution for predicting toxicity and ensuring efficient drug development and chemical safety. An ML model for toxicity prediction can reduce experimental costs and time while addressing ethical concerns by drastically reducing the need for animals and clinical trials. Herein, MolToxPred, an ML-based tool, has been developed using a stacked model approach to predict the potential toxicity of small molecules and metabolites. The stacked model consists of random forest, multi-layer perceptron, and LightGBM as base classifiers and Logistic Regression as the meta classifier. For training and validation purposes, a comprehensive set of toxic and non-toxic molecules is curated. Different structural and physicochemical-based features in the form of molecular descriptors and fingerprints were employed. MolToxPred utilizes a comprehensive feature selection process and optimizes its hyperparameters through Bayesian optimization with stratified 5-fold cross-validation. In the evaluation phase, MolToxPred achieved an AUROC of 87.76% on the test set and 88.84% on an external validation set. The McNemar test was used as the post-hoc test to determine if the stacked models' performance was significantly different compared to the base learners. The developed stacked model outperformed its base classifiers and an existing tool in the literature, reaffirming its better performance. The hypothesis is that the incorporation of a diverse set of data, the subsequent feature selection, and a stacked ensemble approach give MolToxPred the edge over other methods. In addition to this, an attempt has been made to identify structural alerts responsible for endpoints of the Tox21 data to determine the association of a molecule with a plausible downstream pathway of action. MolToxPred may be helpful for drug discovery and regulatory pipelines in pharmaceutical and other industries for in silico toxicity prediction of small molecule candidates.

Prediction of pharmacokinetics of an anaplastic lymphoma kinase inhibitor in rat and monkey: application of physiologically based pharmacokinetic model as an alternative tool to minimise animal studies

The pharmacokinetic (PK) and toxicokinetic profile of a drug from its preclinical evaluation helps the researcher determine whether the drug should be tested in humans based on its safety and toxicity.Preclinical studies require time and resources and are prone to error. Moreover, according to the United States Food and Drug Administration Modernisation Act 2, animal testing is no longer mandatory for new drug development, and an animal-free alternative, such as cell-based assay and computer models, can be used.Different physiologically based PK models were developed for an anaplastic lymphoma kinase inhibitor in rats and monkeys after intravenous and oral administration using its physicochemical properties and in vitro characterisation data.The developed model was validated against the in vivo data available in the literature, and the validation results were found within the acceptable limit. A parameter sensitivity analysis was performed to identify the properties of the compound influencing the PK profile.This work demonstrates the application of the physiologically based PK model to predict the PKs of a drug, which will eventually assist in reducing the number of animal studies and save time and cost of drug discovery and development.

toxCSM: comprehensive prediction of small molecule toxicity profiles

Drug discovery is a lengthy, costly and high-risk endeavour that is further convoluted by high attrition rates in later development stages. Toxicity has been one of the main causes of failure during clinical trials, increasing drug development time and costs. To facilitate early identification and optimisation of toxicity profiles, several computational tools emerged aiming at improving success rates by timely pre-screening drug candidates. Despite these efforts, there is an increasing demand for platforms capable of assessing both environmental as well as human-based toxicity properties at large scale. Here, we present toxCSM, a comprehensive computational platform for the study and optimisation of toxicity profiles of small molecules. toxCSM leverages on the well-established concepts of graph-based signatures, molecular descriptors and similarity scores to develop 36 models for predicting a range of toxicity properties, which can assist in developing safer drugs and agrochemicals. toxCSM achieved an Area Under the Receiver Operating Characteristic (ROC) Curve (AUC) of up to 0.99 and Pearson's correlation coefficients of up to 0.94 on 10-fold cross-validation, with comparable performance on blind test sets, outperforming all alternative methods. toxCSM is freely available as a user-friendly web server and API at http://biosig.lab.uq.edu.au/toxcsm.

Microfluidic Organ-on-a-Chip System for Disease Modeling and Drug Development

An organ-on-a-chip is a device that combines micro-manufacturing and tissue engineering to replicate the critical physiological environment and functions of the human organs. Therefore, it can be used to predict drug responses and environmental effects on organs. Microfluidic technology can control micro-scale reagents with high precision. Hence, microfluidics have been widely applied in organ-on-chip systems to mimic specific organ or multiple organs in vivo. These models integrated with various sensors show great potential in simulating the human environment. In this review, we mainly introduce the typical structures and recent research achievements of several organ-on-a-chip platforms. We also discuss innovations in models applied to the fields of pharmacokinetics/pharmacodynamics, nano-medicine, continuous dynamic monitoring in disease modeling, and their further applications in other fields.

Assessment of the predictive capability of modelling and simulation to determine bioequivalence of inhaled drugs: A systematic review

OBJECTIVES

There are a multitude of different modelling techniques that have been used for inhaled drugs. The main objective of this review was to conduct an exhaustive survey of published mathematical models in the area of asthma and chronic obstructive pulmonary disease (COPD) for inhalation drugs. Additionally, this review will attempt to assess the applicability of these models to assess bioequivalence (BE) of orally inhaled products (OIPs).

EVIDENCE ACQUISITION

PubMed, Science Direct, Web of Science, and Scopus databases were searched from 1996 to 2020, to find studies that described mathematical models used for inhaled drugs in asthma/COPD.

RESULTS

50 articles were finally included in this systematic review. This research identified 22 articles on in silico aerosol deposition models, 20 articles related to population pharmacokinetics and 8 articles on physiologically based pharmacokinetic modelling (PBPK) modelling for inhaled drugs in asthma/COPD. Among all the aerosol deposition models, computational fluid dynamics (CFD) simulations are more likely to predict regional aerosol deposition pattern in human respiratory tracts. Across the population PK articles, body weight, gender, age and smoking status were the most common covariates that were found to be significant. Further, limited published PBPK models reported approximately 29 parameters relevant for absorption and distribution of inhaled drugs. The strengths and weaknesses of each modelling technique has also been reviewed.

CONCLUSION

Overall, while there are different modelling techniques that have been used for inhaled drugs in asthma and COPD, there is very limited application of these models for assessment of bioequivalence of OIPs. This review also provides a ready reference of various parameters that have been considered in various models which will aid in evaluation if one model or hybrid in silico models need to be considered when assessing bioequivalence of OIPs.

Quantitative analysis of an impact of P-glycoprotein on edoxaban's disposition using a human physiologically based pharmacokinetic (PBPK) model

The purpose of this study was to evaluate the impact of P-glycoprotein (P-gp) efflux on edoxaban absorption in gastrointestinal tracts quantitatively by a physiologically based pharmacokinetic (PBPK) model constructed with clinical and non-clinical observations (using GastroPlus™ software). An absorption process was described by the advanced compartmental absorption and transit model with the P-gp function. A human PBPK model was constructed by integrating the clinical and non-clinical observations. The constructed model was demonstrated to reproduce the data observed in the mass-balance study. Thus, elimination pathways can be quantitatively incorporated into the model. A constructed model successfully described the difference in slopes of plasma concentration (Cp)-time curve at around 8 - 24 hr post-dose between intravenous infusion and oral administration. Furthermore, the model without P-gp efflux activity can reproduce the Cp-time profile in the absence of P-gp activity observed from the clinical DDI study results. Since the difference of slopes between intravenous infusion and oral administration also disappeared by the absence of P-gp efflux activity, P-gp must be a key molecule to govern edoxaban's PK behavior. The constructed PBPK model will help us to understand the significant contribution of P-gp in edoxaban's disposition in gastrointestinal tracts quantitatively.

Optimization of hydroxychloroquine dosing scheme based on COVID-19 patients' characteristics: a review of the literature and simulations

During the recent COVID-19 outbreak hydroxychloroquine (HCQ) has been proposed as a safe and effective therapeutic option. However, a wide variety of dosing schemes has been applied in the clinical practice and tested in clinical studies. An extended literature survey was performed investigating the pharmacokinetics, the efficacy and safety of HCQ in COVID-19 treatment. Population pharmacokinetic models were retrieved from the literature and after evaluation and assessment one was selected in order to perform simulations. The most commonly applied dosing schemes were explored for patients with different weights and different levels of HCQ clearance impairment. Model-based simulations of HCQ concentrations revealed that high initial doses followed by low and sparse doses may offer significant benefits to patients by decreasing the viral load without reaching levels considered to produce adverse effects. For instance, the dosing scheme proposed for a 70 kg adult with moderate COVID-19 symptoms would be 600 mg upon diagnosis, 400 mg after 12 h, 300 mg after 24 h, 200 mg after 36 h, followed by 200 mg BID for 4 d, followed by 200 mg OD for 5 d. Based on the results from simulations performed and the currently published knowledge regarding HCQ in COVID-19 treatment, this study provides evidence that a high loading dose followed by sparse doses could offer significant benefits to the patients.

Computational Approaches in Preclinical Studies on Drug Discovery and Development

Because undesirable pharmacokinetics and toxicity are significant reasons for the failure of drug development in the costly late stage, it has been widely recognized that drug ADMET properties should be considered as early as possible to reduce failure rates in the clinical phase of drug discovery. Concurrently, drug recalls have become increasingly common in recent years, prompting pharmaceutical companies to increase attention toward the safety evaluation of preclinical drugs. In vitro and in vivo drug evaluation techniques are currently more mature in preclinical applications, but these technologies are costly. In recent years, with the rapid development of computer science, in silico technology has been widely used to evaluate the relevant properties of drugs in the preclinical stage and has produced many software programs and in silico models, further promoting the study of ADMET in vitro. In this review, we first introduce the two ADMET prediction categories (molecular modeling and data modeling). Then, we perform a systematic classification and description of the databases and software commonly used for ADMET prediction. We focus on some widely studied ADMT properties as well as PBPK simulation, and we list some applications that are related to the prediction categories and web tools. Finally, we discuss challenges and limitations in the preclinical area and propose some suggestions and prospects for the future.

Physiologically Based Pharmacokinetic Modelling for First-In-Human Predictions: An Updated Model Building Strategy Illustrated with Challenging Industry Case Studies

Physiologically based pharmacokinetic modelling is well established in the pharmaceutical industry and is accepted by regulatory agencies for the prediction of drug-drug interactions. However, physiologically based pharmacokinetic modelling is valuable to address a much wider range of pharmaceutical applications, and new regulatory impact is expected as its full power is leveraged. As one example, physiologically based pharmacokinetic modelling is already routinely used during drug discovery for in-vitro to in-vivo translation and pharmacokinetic modelling in preclinical species, and this leads to the application of verified models for first-in-human pharmacokinetic predictions. A consistent cross-industry strategy in this application area would increase confidence in the approach and facilitate further learning. With this in mind, this article aims to enhance a previously published first-in-human physiologically based pharmacokinetic model-building strategy. Based on the experience of scientists from multiple companies participating in the GastroPlus™ User Group Steering Committee, new Absorption, Distribution, Metabolism and Excretion knowledge is integrated and decision trees proposed for each essential component of a first-in-human prediction. We have reviewed many relevant scientific publications to identify new findings and highlight gaps that need to be addressed. Finally, four industry case studies for more challenging compounds illustrate and highlight key components of the strategy.

Model-Informed Drug Discovery and Development Strategy for the Rapid Development of Anti-Tuberculosis Drug Combinations

The increasing emergence of drug-resistant tuberculosis requires new effective and safe drug regimens. However, drug discovery and development are challenging, lengthy and costly. The framework of model-informed drug discovery and development (MID3) is proposed to be applied throughout the preclinical to clinical phases to provide an informative prediction of drug exposure and efficacy in humans in order to select novel anti-tuberculosis drug combinations. The MID3 includes pharmacokinetic-pharmacodynamic and quantitative systems pharmacology models, machine learning and artificial intelligence, which integrates all the available knowledge related to disease and the compounds. A translational in vitro-in vivo link throughout modeling and simulation is crucial to optimize the selection of regimens with the highest probability of receiving approval from regulatory authorities. In vitro-in vivo correlation (IVIVC) and physiologically-based pharmacokinetic modeling provide powerful tools to predict pharmacokinetic drug-drug interactions based on preclinical information. Mechanistic or semi-mechanistic pharmacokinetic-pharmacodynamic models have been successfully applied to predict the clinical exposure-response profile for anti-tuberculosis drugs using preclinical data. Potential pharmacodynamic drug-drug interactions can be predicted from in vitro data through IVIVC and pharmacokinetic-pharmacodynamic modeling accounting for translational factors. It is essential for academic and industrial drug developers to collaborate across disciplines to realize the huge potential of MID3.

Physiologically Based Pharmacokinetic and Pharmacodynamic Analysis Enabled by Microfluidically Linked Organs-on-Chips

Physiologically based pharmacokinetic (PBPK) modeling and simulation approaches are beginning to be integrated into drug development and approval processes because they enable key pharmacokinetic (PK) parameters to be predicted from in vitro data. However, these approaches are hampered by many limitations, including an inability to incorporate organ-specific differentials in drug clearance, distribution, and absorption that result from differences in cell uptake, transport, and metabolism. Moreover, such approaches are generally unable to provide insight into pharmacodynamic (PD) parameters. Recent development of microfluidic Organ-on-a-Chip (Organ Chip) cell culture devices that recapitulate tissue-tissue interfaces, vascular perfusion, and organ-level functionality offer the ability to overcome these limitations when multiple Organ Chips are linked via their endothelium-lined vascular channels. Here, we discuss successes and challenges in the use of existing culture models and vascularized Organ Chips for PBPK and PD modeling of human drug responses, as well as in vitro to in vivo extrapolation (IVIVE) of these results, and how these approaches might advance drug development and regulatory review processes in the future.

Quantitative Analysis of Tozadenant Using Liquid Chromatography-Mass Spectrometric Method in Rat Plasma and Its Human Pharmacokinetics Prediction Using Physiologically Based Pharmacokinetic Modeling

Tozadenant is one of the selective adenosine A2a receptor antagonists with a potential to be a new Parkinson's disease (PD) therapeutic drug. In this study, a liquid chromatography-mass spectrometry based bioanalytical method was qualified and applied for the quantitative analysis of tozadenant in rat plasma. A good calibration curve was observed in the range from 1.01 to 2200 ng/mL for tozadenant using a quadratic regression. In vitro and preclinical in vivo pharmacokinetic (PK) properties of tozadenant were studied through the developed bioanalytical methods, and human PK profiles were predicted using physiologically based pharmacokinetic (PBPK) modeling based on these values. The PBPK model was initially optimized using in vitro and in vivo PK data obtained by intravenous administration at a dose of 1 mg/kg in rats. Other in vivo PK data in rats were used to validate the PBPK model. The human PK of tozadenant after oral administration at a dose of 240 mg was simulated by using an optimized and validated PBPK model. The predicted human PK parameters and profiles were similar to the observed clinical data. As a result, optimized PBPK model could reasonably predict the PK in human.

Application of Physiologically Based Pharmacokinetic Modeling in Understanding Bosutinib Drug-Drug Interactions: Importance of Intestinal P-Glycoprotein

Bosutinib is an orally available Src/Abl tyrosine kinase inhibitor indicated for the treatment of patients with Ph+ chronic myelogenous leukemia at a clinically recommended dose of 500 mg once daily. Clinical results indicated that increases in bosutinib oral exposures were supraproportional at the lower doses (50-200 mg) and approximately dose-proportional at the higher doses (200-600 mg). Bosutinib is a substrate of CYP3A4 and P-glycoprotein and exhibits pH-dependent solubility with moderate intestinal permeability. These findings led us to investigate the factors influencing the underlying pharmacokinetic mechanisms of bosutinib with physiologically based pharmacokinetic (PBPK) models. Our primary objectives were to: 1) refine the previously developed bosutinib PBPK model on the basis of the latest oral bioavailability data and 2) verify the refined PBPK model with P-glycoprotein kinetics on the basis of the bosutinib drug-drug interaction (DDI) results with ketoconazole and rifampin. Additionally, the verified PBPK model was applied to predict bosutinib DDIs with dual CYP3A/P-glycoprotein inhibitors. The results indicated that 1) the refined PBPK model adequately described the observed plasma concentration-time profiles of bosutinib and 2) the verified PBPK model reasonably predicted the effects of ketoconazole and rifampin on bosutinib exposures by accounting for intestinal P-glycoprotein inhibition/induction. These results suggested that bosutinib DDI mechanism could involve not only CYP3A4-mediated metabolism but also P-glycoprotein-mediated efflux on absorption. In summary, P-glycoprotein kinetics could constitute an element in the PBPK models critical to understanding the pharmacokinetic mechanism of dual CYP3A/P-glycoprotein substrates, such as bosutinib, that exhibit nonlinear pharmacokinetics owing largely to a saturation of intestinal P-glycoprotein-mediated efflux.

Computational modeling of human oral bioavailability: what will be next?

INTRODUCTION

The oral route is the most convenient way of administrating drugs. Therefore, accurate determination of oral bioavailability is paramount during drug discovery and development. Quantitative structure-property relationship (QSPR), rule-of-thumb (RoT) and physiologically based-pharmacokinetic (PBPK) approaches are promising alternatives to the early oral bioavailability prediction. Areas covered: The authors give insight into the factors affecting bioavailability, the fundamental theoretical framework and the practical aspects of computational methods for predicting this property. They also give their perspectives on future computational models for estimating oral bioavailability. Expert opinion: Oral bioavailability is a multi-factorial pharmacokinetic property with its accurate prediction challenging. For RoT and QSPR modeling, the reliability of datasets, the significance of molecular descriptor families and the diversity of chemometric tools used are important factors that define model predictability and interpretability. Likewise, for PBPK modeling the integrity of the pharmacokinetic data, the number of input parameters, the complexity of statistical analysis and the software packages used are relevant factors in bioavailability prediction. Although these approaches have been utilized independently, the tendency to use hybrid QSPR-PBPK approaches together with the exploration of ensemble and deep-learning systems for QSPR modeling of oral bioavailability has opened new avenues for development promising tools for oral bioavailability prediction.

Forecasting oral absorption across biopharmaceutics classification system classes with physiologically based pharmacokinetic models

OBJECTIVES

The aim of this study was (1) to determine how closely physiologically based pharmacokinetic (PBPK) models can predict oral bioavailability using a priori knowledge of drug-specific properties and (2) to examine the influence of the biopharmaceutics classification system class on the simulation success.

METHODS

Simcyp Simulator, GastroPlus^™ and GI-Sim were used. Compounds with published Biowaiver monographs (bisoprolol (BCS I), nifedipine (BCS II), cimetidine (BCS III), furosemide (BCS IV)) were selected to ensure availability of accurate and reproducible data for all required parameters. Simulation success was evaluated with the average fold error (AFE) and absolute average fold error (AAFE). Parameter sensitivity analysis (PSA) to selected parameters was performed.

KEY FINDINGS

Plasma concentration-time profiles after intravenous administration were forecast within an AAFE < 3. The addition of absorption processes resulted in more variability in the prediction of the plasma profiles, irrespective of biopharmaceutics classification system (BCS) class. The reliability of literature permeability data was identified as a key issue in the accuracy of predicting oral drug absorption.

CONCLUSION

For the four drugs studied, it appears that the forecasting accuracy of the PBPK models is related to the BCS class (BCS I > BCS II, BCS III > BCS IV). These results will need to be verified with additional drugs.

Quantifying and Communicating Uncertainty in Preclinical Human Dose-Prediction

Human dose-prediction is fundamental for ranking lead-optimization compounds in drug discovery and to inform design of early clinical trials. This tutorial describes how uncertainty in such predictions can be quantified and efficiently communicated to facilitate decision-making. Using three drug-discovery case studies, we show how several uncertain pieces of input information can be integrated into one single uncomplicated plot with key predictions, including their uncertainties, for many compounds or for many scenarios, or both.

Encapsulated cells for long-term secretion of soluble VEGF receptor 1: Material optimization and simulation of ocular drug response

Anti-angiogenic therapies with vascular endothelial growth factor (VEGF) inhibiting factors are effective treatment options for neovascular diseases of the retina, but these proteins can only be delivered as intravitreal (IVT) injections. To sustain a therapeutic drug level in the retina, VEGF inhibitors have to be delivered frequently, every 4-8weeks, causing inconvenience for the patients and expenses for the healthcare system. The aim of this study was to investigate cell encapsulation as a delivery system for prolonged anti-angiogenic treatment of retinal neovascularization. Genetically engineered ARPE-19 cells secreting soluble vascular endothelial growth factor receptor 1 (sVEGFR1) were encapsulated in a hydrogel of cross-linked collagen and interpenetrating hyaluronic acid (HA). The system was optimized in terms of matrix composition and cell density, and long-term cell viability and protein secretion measurements were performed. sVEGFR1 ARPE-19 cells in the optimized hydrogel remained viable and secreted sVEGFR1 at a constant rate for at least 50days. Based on pharmacokinetic/pharmacodynamic (PK/PD) modeling, delivery of sVEGFR1 from this cell encapsulation system is expected to lead only to modest VEGF inhibition, but improvements of the protein structure and/or secretion rate should result in strong and prolonged therapeutic effect. In conclusion, the hydrogel matrix herein supported the survival and protein secretion from the encapsulated cells. The PK/PD simulation is a convenient approach to predict the efficiency of the cell encapsulation system before in vivo experiments.

Ubiquity: a framework for physiological/mechanism-based pharmacokinetic/pharmacodynamic model development and deployment

Practitioners of pharmacokinetic/pharmacodynamic modeling routinely employ various software packages that enable them to fit differential equation based mechanistic or empirical models to biological/pharmacological data. The availability and choice of different analytical tools, while enabling, can also pose a significant challenge in terms of both, implementation and transferability. A package has been developed that addresses these issues by creating a simple text-based format, which provides methods to reduce coding complexity and enables the modeler to describe the components of the model based on the underlying physiochemical processes. A Perl script builds the system for multiple formats (ADAPT, MATLAB, Berkeley Madonna, etc.), enabling analysis across several software packages and reducing the chance for transcription error. Workflows can then be built around this package, which can increase efficiency and model availability. As a proof of concept, tools are included that allow models constructed in this format to be run with MATLAB both at the scripting level and through a generic graphical application that can be compiled and run as a stand-alone application.

Computational approaches to analyse and predict small molecule transport and distribution at cellular and subcellular levels

Quantitative structure-activity relationship (QSAR) studies and mechanistic mathematical modeling approaches have been independently employed for analysing and predicting the transport and distribution of small molecule chemical agents in living organisms. Both of these computational approaches have been useful for interpreting experiments measuring the transport properties of small molecule chemical agents, in vitro and in vivo. Nevertheless, mechanistic cell-based pharmacokinetic models have been especially useful to guide the design of experiments probing the molecular pathways underlying small molecule transport phenomena. Unlike QSAR models, mechanistic models can be integrated from microscopic to macroscopic levels, to analyse the spatiotemporal dynamics of small molecule chemical agents from intracellular organelles to whole organs, well beyond the experiments and training data sets upon which the models are based. Based on differential equations, mechanistic models can also be integrated with other differential equations-based systems biology models of biochemical networks or signaling pathways. Although the origin and evolution of mathematical modeling approaches aimed at predicting drug transport and distribution has occurred independently from systems biology, we propose that the incorporation of mechanistic cell-based computational models of drug transport and distribution into a systems biology modeling framework is a logical next step for the advancement of systems pharmacology research.

Recent advances in 2D and 3D in vitro systems using primary hepatocytes, alternative hepatocyte sources and non-parenchymal liver cells and their use in investigating mechanisms of hepatotoxicity, cell signaling and ADME

This review encompasses the most important advances in liver functions and hepatotoxicity and analyzes which mechanisms can be studied in vitro. In a complex architecture of nested, zonated lobules, the liver consists of approximately 80 % hepatocytes and 20 % non-parenchymal cells, the latter being involved in a secondary phase that may dramatically aggravate the initial damage. Hepatotoxicity, as well as hepatic metabolism, is controlled by a set of nuclear receptors (including PXR, CAR, HNF-4α, FXR, LXR, SHP, VDR and PPAR) and signaling pathways. When isolating liver cells, some pathways are activated, e.g., the RAS/MEK/ERK pathway, whereas others are silenced (e.g. HNF-4α), resulting in up- and downregulation of hundreds of genes. An understanding of these changes is crucial for a correct interpretation of in vitro data. The possibilities and limitations of the most useful liver in vitro systems are summarized, including three-dimensional culture techniques, co-cultures with non-parenchymal cells, hepatospheres, precision cut liver slices and the isolated perfused liver. Also discussed is how closely hepatoma, stem cell and iPS cell-derived hepatocyte-like-cells resemble real hepatocytes. Finally, a summary is given of the state of the art of liver in vitro and mathematical modeling systems that are currently used in the pharmaceutical industry with an emphasis on drug metabolism, prediction of clearance, drug interaction, transporter studies and hepatotoxicity. One key message is that despite our enthusiasm for in vitro systems, we must never lose sight of the in vivo situation. Although hepatocytes have been isolated for decades, the hunt for relevant alternative systems has only just begun.

Simulation of the pharmacokinetics of bisoprolol in healthy adults and patients with impaired renal function using whole-body physiologically based pharmacokinetic modeling

AIM

To develop and evaluate a whole-body physiologically based pharmacokinetic (WB-PBPK) model of bisoprolol and to simulate its exposure and disposition in healthy adults and patients with renal function impairment.

METHODS

Bisoprolol dispositions in 14 tissue compartments were described by perfusion-limited compartments. Based the tissue composition equations and drug-specific properties such as log P, permeability, and plasma protein binding published in literatures, the absorption and whole-body distribution of bisoprolol was predicted using the 'Advanced Compartmental Absorption Transit' (ACAT) model and the whole-body disposition model, respectively. Renal and hepatic clearances were simulated using empirical scaling methods followed by incorporation into the WB-PBPK model. Model refinements were conducted after a comparison of the simulated concentration-time profiles and pharmacokinetic parameters with the observed data in healthy adults following intravenous and oral administration. Finally, the WB-PBPK model coupled with a Monte Carlo simulation was employed to predict the mean and variability of bisoprolol pharmacokinetics in virtual healthy subjects and patients.

RESULTS

The simulated and observed data after both intravenous and oral dosing showed good agreement for all of the dose levels in the reported normal adult population groups. The predicted pharmacokinetic parameters (AUC, C(max), and T(max)) were reasonably consistent (<1.3-fold error) with the observed values after single oral administration of doses ranging from of 5 to 20 mg using the refined WB-PBPK model. The simulated plasma profiles after multiple oral administration of bisoprolol in healthy adults and patient with renal impairment matched well with the observed profiles.

CONCLUSION

The WB-PBPK model successfully predicts the intravenous and oral pharmacokinetics of bisoprolol across multiple dose levels in diverse normal adult human populations and patients with renal insufficiency.

BioDMET: a physiologically based pharmacokinetic simulation tool for assessing proposed solutions to complex biological problems

We developed a detailed, whole-body physiologically based pharmacokinetic (PBPK) modeling tool for calculating the distribution of pharmaceutical agents in the various tissues and organs of a human or animal as a function of time. Ordinary differential equations (ODEs) represent the circulation of body fluids through organs and tissues at the macroscopic level, and the biological transport mechanisms and biotransformations within cells and their organelles at the molecular scale. Each major organ in the body is modeled as composed of one or more tissues. Tissues are made up of cells and fluid spaces. The model accounts for the circulation of arterial and venous blood as well as lymph. Since its development was fueled by the need to accurately predict the pharmacokinetic properties of imaging agents, BioDMET is more complex than most PBPK models. The anatomical details of the model are important for the imaging simulation endpoints. Model complexity has also been crucial for quickly adapting the tool to different problems without the need to generate a new model for every problem. When simpler models are preferred, the non-critical compartments can be dynamically collapsed to reduce unnecessary complexity. BioDMET has been used for imaging feasibility calculations in oncology, neurology, cardiology, and diabetes. For this purpose, the time concentration data generated by the model is inputted into a physics-based image simulator to establish imageability criteria. These are then used to define agent and physiology property ranges required for successful imaging. BioDMET has lately been adapted to aid the development of antimicrobial therapeutics. Given a range of built-in features and its inherent flexibility to customization, the model can be used to study a variety of pharmacokinetic and pharmacodynamic problems such as the effects of inter-individual differences and disease-states on drug pharmacokinetics and pharmacodynamics, dosing optimization, and inter-species scaling. While developing a tool to aid imaging agent and drug development, we aimed at accelerating the acceptance and broad use of PBPK modeling by providing a free mechanistic PBPK software that is user friendly, easy to adapt to a wide range of problems even by non-programmers, provided with ready-to-use parameterized models and benchmarking data collected from the peer-reviewed literature.

Application of IVIVE and PBPK modeling in prospective prediction of clinical pharmacokinetics: strategy and approach during the drug discovery phase with four case studies

Prospective simulations of human pharmacokinetic (PK) parameters and plasma concentration-time curves using in vitro in vivo extrapolation (IVIVE) and physiologically based pharmacokinetic (PBPK) models are becoming a vital part of the drug discovery and development process. This paper presents a strategy to deliver prospective simulations in support of clinical candidate nomination. A three stage approach of input parameter evaluation, model selection and multiple scenario simulation is utilized to predict the key components influencing human PK; absorption, distribution and clearance. The Simcyp® simulator is used to illustrate the approach and four compounds are presented as case studies. In general, the prospective predictions captured the observed clinical data well. Predicted C(max) was within 2-fold of observed data for all compounds and AUC was within 2-fold for all compounds following a single dose and three out of four compounds following multiple doses. Similarly, t(max) was within 2-fold of observed data for all compounds. However, C(last) was less accurately captured with two of the four compounds predicting C(last) within 2-fold of observed data following a single dose. The trend in results was towards overestimation of AUC and C(last) , this was particularly apparent for compound 2 for which clearance was likely underestimated via IVIVE. The prospective approach to simulating human PK using IVIVE and PBPK modeling outlined here attempts to utilize all available in silico, in vitro and in vivo preclinical data in order to determine the most appropriate assumptions to use in prospective predictions of absorption, distribution and clearance to aid clinical candidate nomination.

Simulation of human intravenous and oral pharmacokinetics of 21 diverse compounds using physiologically based pharmacokinetic modelling

BACKGROUND

The importance of predicting human pharmacokinetics during compound selection has been recognized in the pharmaceutical industry. To this end there are many different approaches that are applied.

METHODS

In this study we compared the accuracy of physiologically based pharmacokinetic (PBPK) methodologies implemented in GastroPlus™ with the one-compartment approach routinely used at Pfizer for human pharmacokinetic plasma concentration-time profile prediction. Twenty-one Pfizer compounds were selected based on the availability of relevant preclinical and clinical data. Intravenous and oral human simulations were performed for each compound. To understand any mispredictions, simulations were also performed using the observed clearance (CL) value as input into the model.

RESULTS

The simulation results using PBPK were shown to be superior to those obtained via traditional one-compartment analyses. In many cases, this difference was statistically significant. Specifically, the results showed that the PBPK approach was able to accurately predict passive distribution and absorption processes. Some issues and limitations remain with respect to the prediction of CL and active transport processes and these need to be improved to further increase the utility of PBPK modelling. A particular advantage of the PBPK approach is its ability to accurately predict the multiphasic shape of the pharmacokinetic profiles for many of the compounds tested.

CONCLUSION

The results from this evaluation demonstrate the utility of PBPK methodology for the prediction of human pharmacokinetics. This methodology can be applied at different stages to enhance the understanding of the compounds in a particular chemical series, guide experiments, aid candidate selection and inform clinical trial design.

Physiologically-based pharmacokinetics in drug development and regulatory science

The application of physiologically-based pharmacokinetic (PBPK) modeling is coming of age in drug development and regulation, reflecting significant advances over the past 10 years in the predictability of key pharmacokinetic (PK) parameters from human in vitro data and in the availability of dedicated software platforms and associated databases. Specific advances and contemporary challenges with respect to predicting the processes of drug clearance, distribution, and absorption are reviewed, together with the ability to anticipate the quantitative extent of PK-based drug-drug interactions and the impact of age, genetics, disease, and formulation. The value of this capability in selecting and designing appropriate clinical studies, its implications for resource-sparing techniques, and a more holistic view of the application of PK across the preclinical/clinical divide are considered. Finally, some attention is given to the positioning of PBPK within the drug development and approval paradigm and its future application in truly personalized medicine.

The pharmacokinetic/pharmacodynamic pipeline: translating anticancer drug pharmacology to the clinic

Progress in an understanding of the genetic basis of cancer coupled to molecular pharmacology of potential new anticancer drugs calls for new approaches that are able to address key issues in the drug development process, including pharmacokinetic (PK) and pharmacodynamic (PD) relationships. The incorporation of predictive preclinical PK/PD models into rationally designed early-stage clinical trials offers a promising way to relieve a significant bottleneck in the drug discovery pipeline. The aim of the current review is to discuss some considerations for how quantitative PK and PD analyses for anticancer drugs may be conducted and integrated into a global translational effort, and the importance of examining drug disposition and dynamics in target tissues to support the development of preclinical PK/PD models that can be subsequently extrapolated to predict pharmacologic characteristics in patients. In this article, we describe three different physiologically based (PB) PK modeling approaches, i.e., the whole-body PBPK model, the hybrid PBPK model, and the two-pore model for macromolecules, as well as their applications. General conclusions are that greater effort should be made to generate more clinical data that could validate scaled preclinical PB-PK/PD tumor-based models and, thus, stimulate a framework for preclinical to clinical translation. Finally, given the innovative techniques to measure tissue drug concentrations and associated biomarkers of drug responses, development of predictive PK/PD models will become a standard approach for drug discovery and development.

Physiologically-based pharmacokinetic modeling for absorption, transport, metabolism and excretion

The seminal paper on the liver physiologically-based pharmacokinetic (PBPK) model by Rowland et al. (J Pharmacokinet Biopharm 1:123-136, 1973) that described the influence of blood flow, intrinsic clearance, and binding on hepatic clearance had inspired further development of PBPK modeling of the liver, kidney and intestine as well as whole body. Shortly thereafter, a series of papers from Pang and Rowland compared the well-stirred and parallel-tube liver models and sparked further development on clearance concepts in the liver, including those described by the dispersion model. From 2005 onwards, several seminal papers by Rodgers and Rowland, in their recognition of the binding of molecules to tissue acidic and neutral phospholipids, improved the methodology in providing estimates of the tissue-to-plasma coefficient and rendering easy calculation of these hard-to-get constants. The improvement has strongly consolidated the basic premise on PBPK modeling and simulations and these basics have allowed scientists to focus on other important variables: membrane barriers, and transporter and enzyme and their heterogeneities that further impact drug disposition. In particular, the PBPK models have delved into sequential metabolism and futile cycling to illustrate how transporters and enzymes could affect the metabolism of drugs and metabolites. PBPK models that are especially pertinent to metabolite kinetics are being utilized in drug studies and risk assessment. These types of PBPK modeling reveal differences in kinetics between the formed vs. preformed metabolite, showing special considerations for membrane barriers, and the influence of competing pathways and competing organs.

Current and future directions of pharmacokinetic and pharmacokinetic-pharmacodynamic modelling and simulation: population approach group in Europe 19th annual meeting

The Population Approach Group in Europe meeting is the annual meeting of a group of scientists interested in quantitative pharmacology approaches. This meeting provides insights on new modelling approaches and applications thereof in different therapeutic areas and in different phases of drug development, in which modelling and simulation is set 'at work'. These meeting highlights mainly focus on the contributions pertinent to this journal, such as physiology-based pharmacokinetic modelling, modelling of pharmacokinetic-pharmacodynamic relationships (including binary/categorical end points) in non-clinical studies and also include presentations of junior scientists in the annual Lewis Sheiner Student Session.

A pH-dilution method for estimation of biorelevant drug solubility along the gastrointestinal tract: application to physiologically based pharmacokinetic modeling

Physiologically based pharmacokinetic (PBPK) modeling tools have become an integral part of the modern drug discovery-development process. However, accurate PK prediction of enabling formulations of poorly soluble compounds by applying PBPK modeling has been very limited. This is because current PBPK models rely only on thermodynamic drug solubility inputs (e.g., pH-solubility profile) and give little consideration to the dynamic changes in apparent drug solubility (e.g., supersaturation) that occur during gastrointestinal (GI) transit of an enabling formulation of a water insoluble drug. Moreover, biorepresentative and predictive in vitro tools to measure formulation dependent solubility changes during GI transit remain underdeveloped. In this work, we have developed an in vitro dual pH-dilution method based on rat physiology to estimate the apparent drug concentration in solution along the GI tract during release from solubility enabling formulations. This simple dual pH-dilution method was evaluated using various solubility enabling formulations (i.e., cosolvent solution, amorphous solid dispersions) made using a model early development drug candidate with poor aqueous solubility. The in vitro drug concentration-time profiles from the enabling formulations were used as solubility inputs for PBPK modeling using GastroPlus software. This resulted in excellent predictions of the in vivo oral plasma concentration-time profiles, as compared to using the traditional inputs of thermodynamic pH-solubility profiles. In summary, this work describes a novel in vitro method for facile estimation of formulation dependent GI drug concentration-time profiles and demonstrates the utility of PBPK modeling for oral PK prediction of enabling formulations of poorly soluble drugs.

Predictions of metabolic drug-drug interactions using physiologically based modelling: Two cytochrome P450 3A4 substrates coadministered with ketoconazole or verapamil

Nowadays, evaluation of potential risk of metabolic drug-drug interactions (mDDIs) is of high importance within the pharmaceutical industry, in order to improve safety and reduce the attrition rate of new drugs. Accurate and early prediction of mDDIs has become essential for drug research and development, and in vitro experiments designed to evaluate potential mDDIs are systematically included in the drug development plan prior to clinical assessment. The aim of this study was to illustrate the value and limitations of the classical and new approaches available to predict risks of DDIs in the research and development processes. The interaction of cytochrome P450 (CYP) 3A4 inhibitors (ketoconazole and verapamil) with midazolam was predicted using the inhibitor concentration/inhibition constant ([I]/K(i)) approach, the static approach with added variability (Simcyp(R)), and whole-body physiologically based pharmacokinetic (WB-PBPK) modelling (acslXtreme(R)). Then an in-house reference drug was used to challenge the different approaches based on the midazolam experience. Predicted values (pharmacokinetic parameters, the area under the plasma concentration-time curve [AUC] ratio and plasma concentrations) were compared with observed values obtained after intravenous and oral administration in order to assess the accuracy of the prediction methods. With the [I]/K(i) approach, the interaction risk was always overpredicted for the midazolam substrate, regardless of its route of administration and the coadministered inhibitor. However, the predictions were always satisfactory (within 2-fold) for the reference drug. For the Simcyp(R) calculations, two of the three interaction results for midazolam were overpredicted, both when midazolam was given orally, whereas the prediction obtained when midazolam was administered intravenously was satisfactory. For the reference drug, all predictions could be considered satisfactory. For the WB-PBPK approach, all predictions were satisfactory, regardless of the substrate, route of administration, dose and coadministered inhibitor. DDI risk predictions are performed throughout the research and development processes and are now fully integrated into decision-making processes. The regulatory approach is useful to provide alerts, even at a very early stage of drug development. The 'steady state' approach in Simcyp(R) improves the prediction by using physiological knowledge and mechanistic assumptions. The DDI predictions are very useful, as they provide a range of AUC ratios that include individuals at the extremes of the population, in addition to the 'average tendency'. Finally, the WB-PBPK approach improves the predictions by simulating the concentration-time profiles and calculating the related pharmacokinetic parameters, taking into account the time of administration of each drug - but it requires a good understanding of the absorption, distribution, metabolism and excretion properties of the compound.

Prediction of pharmacokinetic profile of valsartan in humans based on in vitro uptake-transport data

The aim of this study was to evaluate a physiologically based pharmacokinetic (PBPK) model for predicting PK profiles in humans based on a model refined in rats and humans in vitro uptake-transport data using valsartan as a probe substrate. Valsartan is eliminated unchanged, mostly through biliary excretion, both in humans and rats. It was, therefore, chosen as model compound to predict in vivo elimination based on in vitro hepatic uptake-transport data using a fully mechanistic PBPK model. Plated rat and human hepatocytes, and cell lines overexpressing human OATP1B1 and OATP1B3 were used for in vitro uptake experiments. A mechanistic two-compartment model was used to derive the active and passive transport parameters, namely uptake Michaelis-Menten parameters (V(max) and K(m,u)) together with passive diffusion (P(dif)). These transport parameters were then used as input in a whole body physiologically based pharmacokinetic (PBPK) model. The uptake rate of valsartan was higher for rat hepatocytes (K(m,u)=28.4+/-3.7 microM, V(max)=1320+/-180 pmol/mg/min, and P(dif) =1.21+/-0.42 microl/mg/min) compared to human hepatocytes (K(m,u)=44.4+/-14.6 microM, V(max)=304+/-85 pmol/mg/min, and P(dif)=0.724+/-0.271 microl/mg/min). OATP1B1 and -1B3 parameters were correlated to human hepatocyte data, using experimentally established relative activity factors (RAF). Resulting PBPK simulations were compared for plasma- (humans and rats) and bile- (rats) concentration-time profiles following iv bolus administration of valsartan. Plasma clearances (CL(P)) for rats and humans were predicted within twofold relative to predictions based on respective in vitro data. The simulations were extended to simulate the impact of either OATP1B1 or -1B3 inhibition on plasma profile. The limited data set indicates that the mechanistic model allowed for accurate evaluation of in vitro transport data; and the resulting hepatic uptake transport kinetic parameters enabled the prediction of in vivo PK profiles and plasma clearances, using PBPK modelling. Moreover, the interspecies difference in elimination rate observed in vivo was correctly reflected in the transport parameters determined in vitro.