Model-Based Interspecies Scaling of Glucose Homeostasis

Pharmacokinetic Considerations for Optimizing Inhaled Spray-Dried Pyrazinoic Acid Formulations

Tuberculosis (TB), caused by Mycobacterium tuberculosis (Mtb), remains a leading cause of death with 1.6 million deaths worldwide reported in 2021. Oral pyrazinamide (PZA) is an integral part of anti-TB regimens, but its prolonged use has the potential to drive the development of PZA-resistant Mtb. PZA is converted to the active moiety pyrazinoic acid (POA) by the Mtb pyrazinamidase encoded by pncA, and mutations in pncA are associated with the majority of PZA resistance. Conventional oral and parenteral therapies may result in subtherapeutic exposure in the lung; hence, direct pulmonary administration of POA may provide an approach to rescue PZA efficacy for treating pncA-mutant PZA-resistant Mtb. The objectives of the current study were to (i) develop novel dry powder POA formulations, (ii) assess their feasibility for pulmonary delivery using physicochemical characterization, (iii) evaluate their pharmacokinetics (PK) in the guinea pig model, and (iv) develop a mechanism-based pharmacokinetic model (MBM) using in vivo PK data to select a formulation providing adequate exposure in epithelial lining fluid (ELF) and lung tissue. We developed three POA formulations for pulmonary delivery and characterized their PK in plasma, ELF, and lung tissue following passive inhalation in guinea pigs. Additionally, the PK of POA following oral, intravenous, and intratracheal administration was characterized in guinea pigs. The MBM was used to simultaneously model PK data following administration of POA and its formulations via the different routes. The MBM described POA PK well in plasma, ELF, and lung tissue. Physicochemical analyses and MBM predictions suggested that POA maltodextrin was the best among the three formulations and an excellent candidate for further development as it has: (i) the highest ELF-to-plasma exposure ratio (203) and lung tissue-to-plasma exposure ratio (30.4) compared with POA maltodextrin and leucine (75.7/16.2) and POA leucine salt (64.2/19.3) and (ii) the highest concentration in ELF (CmaxELF: 171 nM) within 15.5 min, correlating with a fast transfer into ELF after pulmonary administration (KPM: 22.6 1/h). The data from the guinea pig allowed scaling, using the MBM to a human dose of POA maltodextrin powder demonstrating the potential feasibility of an inhaled product.

Pharmacokinetic considerations for optimizing inhaled spray-dried pyrazinoic acid formulations

Tuberculosis (TB), caused by Mycobacterium tuberculosis ( Mtb ), remains a leading cause of death with 1.6 million deaths worldwide reported in 2021. Oral pyrazinamide (PZA) is an integral part of anti-TB regimens, but its prolonged use has the potential to drive development of PZA resistant Mtb . PZA is converted to the active moiety pyrazinoic acid (POA) by the Mtb pyrazinamidase encoded by pncA , and mutations in pncA are associated with the majority of PZA resistance. Conventional oral and parenteral therapies may result in subtherapeutic exposure in the lung, hence direct pulmonary administration of POA may provide an approach to rescue PZA efficacy for treating pncA- mutant PZA-resistant Mtb . The objectives of the current study were to i) develop novel dry powder POA formulations ii) assess their feasibility for pulmonary delivery using physicochemical characterization, iii) evaluate their pharmacokinetics (PK) in the guinea pig model and iv) develop a mechanism based pharmacokinetic model (MBM) using in vivo PK data to select a formulation providing adequate exposure in epithelial lining fluid (ELF) and lung tissue. We developed three POA formulations for pulmonary delivery and characterized their PK in plasma, ELF, and lung tissue following passive inhalation in guinea pigs. Additionally, the PK of POA following oral, intravenous and intratracheal administration was characterized in guinea pigs. The MBM was used to simultaneously model PK data following administration of POA and its formulations via the different routes. The MBM described POA PK well in plasma, ELF and lung tissue. Physicochemical analyses and MBM predictions suggested that POA maltodextrin was the best among the three formulations and an excellent candidate for further development as it has: (i) the highest ELF-to-plasma exposure ratio (203) and lung tissue-to-plasma exposure ratio (30.4) compared with POA maltodextrin and leucine (75.7/16.2) and POA leucine salt (64.2/19.3); (ii) the highest concentration in ELF ( Cmac ELF : 171 nM) within 15.5 minutes, correlating with a fast transfer into ELF after pulmonary administration ( k PM : 22.6 1/h). The data from the guinea pig allowed scaling, using the MBM to a human dose of POA maltodextrin powder demonstrating the potential feasibility of an inhaled product.

Table of Contents TOC/Abstract Graphic

Interspecies and in vitro-in vivo scaling for quantitative modeling of whole-body drug pharmacokinetics in patients: Application to the anticancer drug oxaliplatin

Quantitative systems pharmacology holds the promises of integrating results from laboratory animals or in vitro human systems into the design of human pharmacokinetic/pharmacodynamic (PK/PD) models allowing for precision and personalized medicine. However, reliable and general in vitro-to-in vivo extrapolation and interspecies scaling methods are still lacking. Here, we developed a translational strategy for the anticancer drug oxaliplatin. Using ex vivo PK data in the whole blood of the mouse, rat, and human, a model representing the amount of platinum (Pt) in the plasma and in the red blood cells was designed and could faithfully fit each dataset independently. A "purely physiologically-based (PB)" scaling approach solely based on preclinical data failed to reproduce human observations, which were then included in the calibration. Investigating approaches in which one parameter was set as species-specific, whereas the others were computed by PB scaling laws, we concluded that allowing the Pt binding rate to plasma proteins to be species-specific permitted to closely fit all data, and guaranteed parameter identifiability. Such a strategy presenting the drawback of including all clinical datasets, we further identified a minimal subset of human data ensuring accurate model calibration. Next, a "whole body" model of oxaliplatin human PK was inferred from the ex vivo study. Its three remaining parameters were estimated, using one third of the available patient data. Remarkably, the model achieved a good fit to the training dataset and successfully reproduced the unseen observations. Such validation endorsed the legitimacy of our scaling methodology calling for its testing with other drugs.

Integrated experimental-computational analysis of a HepaRG liver-islet microphysiological system for human-centric diabetes research

Microphysiological systems (MPS) are powerful tools for emulating human physiology and replicating disease progression in vitro. MPS could be better predictors of human outcome than current animal models, but mechanistic interpretation and in vivo extrapolation of the experimental results remain significant challenges. Here, we address these challenges using an integrated experimental-computational approach. This approach allows for in silico representation and predictions of glucose metabolism in a previously reported MPS with two organ compartments (liver and pancreas) connected in a closed loop with circulating medium. We developed a computational model describing glucose metabolism over 15 days of culture in the MPS. The model was calibrated on an experiment-specific basis using data from seven experiments, where HepaRG single-liver or liver-islet cultures were exposed to both normal and hyperglycemic conditions resembling high blood glucose levels in diabetes. The calibrated models reproduced the fast (i.e. hourly) variations in glucose and insulin observed in the MPS experiments, as well as the long-term (i.e. over weeks) decline in both glucose tolerance and insulin secretion. We also investigated the behaviour of the system under hypoglycemia by simulating this condition in silico, and the model could correctly predict the glucose and insulin responses measured in new MPS experiments. Last, we used the computational model to translate the experimental results to humans, showing good agreement with published data of the glucose response to a meal in healthy subjects. The integrated experimental-computational framework opens new avenues for future investigations toward disease mechanisms and the development of new therapies for metabolic disorders.

Microphysiological systems (MPS) are powerful tools to unravel biological knowledge underlying disease. MPS provide a physiologically relevant, human-based in vitro setting, which can potentially yield better translatability to humans than current animal models and traditional cell cultures. However, mechanistic interpretation and extrapolation of the experimental results to human outcome remain significant challenges. In this study, we confront these challenges using an integrated experimental-computational approach. We present a computational model describing glucose metabolism in a previously reported MPS integrating liver and pancreas. This MPS supports a homeostatic feedback loop between HepaRG/HHSteC spheroids and pancreatic islets, and allows for detailed investigations of mechanisms underlying type 2 diabetes in humans. We show that the computational model captures the complex dynamics of glucose-insulin regulation observed in the system, and can provide mechanistic insight into disease progression features, such as insulin resistance and β-cell dynamics. Furthermore, the computational model can explain key differences in temporal dynamics between MPS and human responses, and thus provides a tool for translating experimental insights into human outcome. The integrated experimental-computational framework opens new avenues for future investigations toward disease mechanisms and the development of new therapies for metabolic disorders.

A quantitative systems pharmacology approach to predict the safe-equivalent dose of doxorubicin in patients with cardiovascular comorbidity

Patients with cardiovascular comorbidity are less tolerant to cardiotoxic drugs and should be treated with reduced doses to prevent cardiotoxicity. However, the safe‐equivalent dose of antitumor drugs in patients with cardiovascular disease/risk is difficult to predict because they are usually excluded from clinical trials as a result of ethical considerations. In this study, a translational quantitative system pharmacology‐pharmacokinetic‐pharmacodynamic (QSP‐PK‐PD) model was developed based on preclinical study to predict the safe‐equivalence dose of doxorubicin in patients with or without cardiovascular disease. Virtual clinical trials were conducted to validate the translational QSP‐PK‐PD model. The model replicated several experimental and clinical observations: the left ventricular ejection fraction (LVEF) was reduced and the left ventricular end‐diastolic volume (LVEDV) was elevated in systolic dysfunction rats, the LVEF was preserved and LVEDV reduced in diastolic dysfunction rats, and patients with preexisting cardiovascular disease were more vulnerable to doxorubicin‐induced cardiac dysfunction than cardiovascular healthy patients. A parameter sensitivity analysis showed that doxorubicin‐induced cardiovascular dysfunction was mainly determined by the sensitivity of cardiomyocytes to cardiotoxic drugs and the baseline value of LVEDV, reflected in LVEF change percentage from the baseline. Blood pressure was the least sensitive factor affecting doxorubicin‐induced cardiotoxicity.

An Updated Organ-Based Multi-Level Model for Glucose Homeostasis: Organ Distributions, Timing, and Impact of Blood Flow

Glucose homeostasis is the tight control of glucose in the blood. This complex control is important, due to its malfunction in serious diseases like diabetes, and not yet sufficiently understood. Due to the involvement of numerous organs and sub-systems, each with their own intra-cellular control, we have developed a multi-level mathematical model, for glucose homeostasis, which integrates a variety of data. Over the last 10 years, this model has been used to insert new insights from the intra-cellular level into the larger whole-body perspective. However, the original cell-organ-body translation has during these years never been updated, despite several critical shortcomings, which also have not been resolved by other modeling efforts. For this reason, we here present an updated multi-level model. This model provides a more accurate sub-division of how much glucose is being taken up by the different organs. Unlike the original model, we now also account for the different dynamics seen in the different organs. The new model also incorporates the central impact of blood flow on insulin-stimulated glucose uptake. Each new improvement is clear upon visual inspection, and they are also supported by statistical tests. The final multi-level model describes >300 data points in >40 time-series and dose-response curves, resulting from a large variety of perturbations, describing both intra-cellular processes, organ fluxes, and whole-body meal responses. We hope that this model will serve as an improved basis for future data integration, useful for research and drug developments within diabetes.

Hybrid modelling for stroke care: Review and suggestions of new approaches for risk assessment and simulation of scenarios

Stroke is an example of a complex and multi-factorial disease involving multiple organs, timescales, and disease mechanisms. To deal with this complexity, and to realize Precision Medicine of stroke, mathematical models are needed. Such approaches include: 1) machine learning, 2) bioinformatic network models, and 3) mechanistic models. Since these three approaches have complementary strengths and weaknesses, a hybrid modelling approach combining them would be the most beneficial. However, no concrete approach ready to be implemented for a specific disease has been presented to date. In this paper, we both review the strengths and weaknesses of the three approaches, and propose a roadmap for hybrid modelling in the case of stroke care. We focus on two main tasks needed for the clinical setting: a) For stroke risk calculation, we propose a new two-step approach, where non-linear mixed effects models and bioinformatic network models yield biomarkers which are used as input to a machine learning model and b) For simulation of care scenarios, we propose a new four-step approach, which revolves around iterations between simulations of the mechanistic models and imputations of non-modelled or non-measured variables. We illustrate and discuss the different approaches in the context of Precision Medicine for stroke.

A mechanism-based pharmacokinetic model of remdesivir leveraging interspecies scaling to simulate COVID-19 treatment in humans

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) outbreak initiated the global coronavirus disease 2019 (COVID-19) pandemic resulting in 42.9 million confirmed infections and > 1.1 million deaths worldwide as of October 26, 2020. Remdesivir is a broad-spectrum nucleotide prodrug shown to be effective against enzootic coronaviruses. The pharmacokinetics (PKs) of remdesivir in plasma have recently been described. However, the distribution of its active metabolite nucleoside triphosphate (NTP) to the site of pulmonary infection is unknown in humans. Our objective was to use existing in vivo mouse PK data for remdesivir and its metabolites to develop a mechanism-based model to allometrically scale and simulate the human PK of remdesivir in plasma and NTP in lung homogenate. Remdesivir and GS-441524 concentrations in plasma and total phosphorylated nucleoside concentrations in lung homogenate from Ces1c-/- mice administered 25 or 50 mg/kg of remdesivir subcutaneously were simultaneously fit to estimate PK parameters. The mouse PK model was allometrically scaled to predict human PK parameters to simulate the clinically recommended 200 mg loading dose followed by 100 mg daily maintenance doses administered as 30-minute intravenous infusions. Simulations of unbound remdesivir concentrations in human plasma were below 2.48 μM, the 90% maximal inhibitory concentration for SARS-CoV-2 inhibition in vitro. Simulations of NTP in the lungs were below high efficacy in vitro thresholds. We have identified a need for alternative dosing strategies to achieve more efficacious concentrations of NTP in human lungs, perhaps by reformulating remdesivir for direct pulmonary delivery.

Mathematical Modeling for the Physiological and Clinical Investigation of Glucose Homeostasis and Diabetes

Mathematical modeling in the field of glucose metabolism has a longstanding tradition. The use of models is motivated by several reasons. Models have been used for calculating parameters of physiological interest from experimental data indirectly, to provide an unambiguous quantitative representation of pathophysiological mechanisms, to determine indices of clinical usefulness from simple experimental tests. With the growing societal impact of type 2 diabetes, which involves the disturbance of the glucose homeostasis system, development and use of models in this area have increased. Following the approaches of physiological and clinical investigation, the focus of the models has spanned from representations of whole body processes to those of cells, i.e., from in vivo to in vitro research. Model-based approaches for linking in vivo to in vitro research have been proposed, as well as multiscale models merging the two areas. The success and impact of models has been variable. Two kinds of models have received remarkable interest: those widely used in clinical applications, e.g., for the assessment of insulin sensitivity and β-cell function and some models representing specific aspects of the glucose homeostasis system, which have become iconic for their efficacy in describing clearly and compactly key physiological processes, such as insulin secretion from the pancreatic β cells. Models are inevitably simplified and approximate representations of a physiological system. Key to their success is an appropriate balance between adherence to reality, comprehensibility, interpretative value and practical usefulness. This has been achieved with a variety of approaches. Although many models concerning the glucose homeostasis system have been proposed, research in this area still needs to address numerous issues and tackle new opportunities. The mathematical representation of the glucose homeostasis processes is only partial, also because some mechanisms are still only partially understood. For in vitro research, mathematical models still need to develop their potential. This review illustrates the problems, approaches and contribution of mathematical modeling to the physiological and clinical investigation of glucose homeostasis and diabetes, focusing on the most relevant and stimulating models.

Connecting Rodent and Human Pharmacokinetic Models for the Design and Translation of Glucose-Responsive Insulin

Despite considerable progress, development of glucose-responsive insulins (GRIs) still largely depends on empirical knowledge and tedious experimentation-especially on rodents. To assist the rational design and clinical translation of the therapeutic, we present a Pharmacokinetic Algorithm Mapping GRI Efficacies in Rodents and Humans (PAMERAH) built upon our previous human model. PAMERAH constitutes a framework for predicting the therapeutic efficacy of a GRI candidate from its user-specified mechanism of action, kinetics, and dosage, which we show is accurate when checked against data from experiments and literature. Results from simulated glucose clamps also agree quantitatively with recent GRI publications. We demonstrate that the model can be used to explore the vast number of permutations constituting the GRI parameter space and thereby identify the optimal design ranges that yield desired performance. A design guide aside, PAMERAH more importantly can facilitate GRI's clinical translation by connecting each candidate's efficacies in rats, mice, and humans. The resultant mapping helps to find GRIs that appear promising in rodents but underperform in humans (i.e., false positives). Conversely, it also allows for the discovery of optimal human GRI dynamics not captured by experiments on a rodent population (false negatives). We condense such information onto a "translatability grid" as a straightforward, visual guide for GRI development.

A Model-Informed Drug Discovery and Development Strategy for the Novel Glucose-Responsive Insulin MK-2640 Enabled Rapid Decision Making

A model‐informed drug discovery and development strategy played a key role in the novel glucose‐responsive insulin MK‐2640’s early clinical development strategy and supported a novel clinical trial paradigm to assess glucose responsiveness. The development and application of in silico modeling approaches by leveraging substantial published clinical insulin pharmacokinetic–pharmacodynamic (PKPD) data and emerging preclinical and clinical data enabled rapid quantitative decision making. Learnings can be applied to define PKPD properties of novel insulins that could become therapeutically meaningful for diabetic patients.

Applications of Quantitative Systems Pharmacology in Model-Informed Drug Discovery: Perspective on Impact and Opportunities

Quantitative systems pharmacology (QSP) approaches have been increasingly applied in the pharmaceutical since the landmark white paper published in 2011 by a National Institutes of Health working group brought attention to the discipline. In this perspective, we discuss QSP in the context of other modeling approaches and highlight the impact of QSP across various stages of drug development and therapeutic areas. We discuss challenges to the field as well as future opportunities.

Glucose, fructosamine, and insulin measurements in saliva of dogs: variations after an experimental glucose administration

The aim of this study was to evaluate if glucose, fructosamine, and insulin levels can be measured in saliva of dogs and assess the changes in these compounds after an experimental glucose administration. Automated spectrophotometric assays for glucose and fructosamine and an ELISA assay for insulin measurements were validated in saliva of dogs, by evaluating precision, accuracy, and limits of detection. In addition, an intravenous glucose bolus was administrated to 10 beagles and fasting serum and saliva samples were obtained immediately before and 5, 10, 20, 30, and 45 min after glucose infusion. The results of the between-run imprecision gave mean CVs of 6.16, 9.40, and 3.10% for glucose, fructosamine, and insulin, respectively. Linearity under dilution showed coefficient of correlation of 0.999, 0.994, and 0.990 for glucose, fructosamine, and insulin, respectively. The LDs were 0.04 mg/dL, 4.08 μmol/L, and 0.02 μg/mL for glucose, fructosamine, and insulin, respectively. The glucose administration caused an increase in serum and salivary levels of glucose with a peak in salivary levels at 30 min and of insulin with a peak in salivary levels at 45 min, while fructosamine did not change. No correlations between serum and salivary concentrations were found for any compound. It is concluded that glucose, fructosamine, and insulin can be measured in saliva of dogs, and an experimental administration of glucose in this species can lead to increases in glucose and insulin in saliva.