Sources of interindividual variability in IVIVE of clearance: an investigation into the prediction of benzodiazepine clearance using a mechanistic population-based pharmacokinetic model

Physiologically-Based Pharmacokinetic Modeling and Dosing Optimization of Cefotaxime in Preterm and Term Neonates

BACKGROUND

Cefotaxime is commonly used in treating bacterial infections in neonates. To characterize the pharmacokinetic process in neonates and evaluate different recommended dosing schedules of cefotaxime, a physiologically-based pharmacokinetic (PBPK) model of cefotaxime was established in adults and scaled to neonates.

METHODS

A whole-body PBPK model was built in PK-SIM® software. Three elimination pathways are composed of enzymatic metabolism in the liver, passive filtration through glomerulus, and active tubular secretion mediated by renal transporters. The ontogeny information was applied to account for age-related changes in cefotaxime pharmacokinetics. The established models were verified with realistic clinical data in adults and pediatric populations. Simulations in neonates were conducted and 100 % of the dosing interval where the unbound concentration in plasma was above the minimum inhibitory concentration (fT>MIC) was selected as the target index for dosing regimen evaluation.

RESULTS

The developed PBPK models successfully described the pharmacokinetic process of cefotaxime in adults and were scaled to the pediatric population. Good verification results were achieved in both adults' and neonates' PBPK models, indicating a good predictive performance. The optimal dosage regimen of cefotaxime was proposed according to the postnatal age (PNA) and gestational age (GA) of neonates. For preterm neonates (GA < 36 weeks), dosages of 25 mg/kg every 8 h in PNA 0-6 days and 25 mg/kg every 6 h in PNA 7-28 days were suggested. For term neonates (GA ≥ 36 weeks), dosages of 33 mg/kg every 8 h in PNA 0-6 days and 33 mg/kg every 6 h in PNA 7-28 days were recommended.

CONCLUSIONS

Our study may provide useful experience in practicing PBPK model-informed precision dosing in the pediatric population.

Exploring the impact of ethnicity on drug pharmacokinetics using PBPK models: A case study with lansoprazole in Japanese subjects

AIMS

The aim of this study is to demonstrate the use of PBPK modelling to explore the impact of ethnic differences on drug PK.

METHODS

A PBPK model developed for lansoprazole was used to predict the clinical PK of lansoprazole in Japanese subjects by incorporating the physiological parameters of a Japanese population into the model. Further verification of the developed Japanese population with clinical studies involving eight other CYP substrates-omeprazole, ticlopidine, alprazolam, midazolam, nifedipine, cinacalcet, paroxetine and dextromethorphan-was also carried out.

RESULTS

The PK of lansoprazole in both Caucasian and Japanese subjects was well predicted by the model as the observed data were within the 5th and 95th percentiles across all the clinical studies. In age- and sex-matched simulations in both the Caucasian and Japanese populations, the predicted PK (mean ± SD) of a single oral dose of 30-mg lansoprazole was higher in the Japanese population in all cases, with more than twofold higher AUC of 5.98 ± 6.43 mg/L.h (95% CI: 4.72, 7.24) vs. 2.46 ± 2.45 mg/L.h (95% CI: 1.98, 2.94) in one scenario. In addition, in two out of the nine clinical DDIs of lansoprazole and the additional CYP substrates simulated using the Japanese population, the predicted DDI in Japanese was more than 1.25-fold that in Caucasians, indicating an increased DDI liability.

CONCLUSIONS

By accounting for various physiological parameters that characterize a population in a PBPK model, the impact of the different identified interethnic differences on the drug's PK can be explored, which can inform the adoption of drugs from one region to another.

Evaluation of the Success of High-Throughput Physiologically Based Pharmacokinetic (HT-PBPK) Modeling Predictions to Inform Early Drug Discovery

Minimizing in vitro and in vivo testing in early drug discovery with the use of physiologically based pharmacokinetic (PBPK) modeling and machine learning (ML) approaches has the potential to reduce discovery cycle times and animal experimentation. However, the prediction success of such an approach has not been shown for a larger and diverse set of compounds representative of a lead optimization pipeline. In this study, the prediction success of the oral (PO) and intravenous (IV) pharmacokinetics (PK) parameters in rats was assessed using a “bottom-up” approach, combining in vitro and ML inputs with a PBPK model. More than 240 compounds for which all of the necessary inputs and PK data were available were used for this assessment. Different clearance scaling approaches were assessed, using hepatocyte intrinsic clearance and protein binding as inputs. In addition, a novel high-throughput PBPK (HT-PBPK) approach was evaluated to assess the scalability of PBPK predictions for a larger number of compounds in drug discovery. The results showed that bottom-up PBPK modeling was able to predict the rat IV and PO PK parameters for the majority of compounds within a 2- to 3-fold error range, using both direct scaling and dilution methods for clearance predictions. The use of only ML-predicted inputs from the structure did not perform well when using in vitro inputs, likely due to clearance miss predictions. The HT-PBPK approach produced comparable results to the full PBPK modeling approach but reduced the simulation time from hours to seconds. In conclusion, a bottom-up PBPK and HT-PBPK approach can successfully predict the PK parameters and guide early discovery by informing compound prioritization, provided that good in vitro assays are in place for key parameters such as clearance.

Pharmacokinetics and safety of rilpivirine in healthy Japanese subjects and exploration of ethnic sensitivity of rilpivirine pharmacokinetics with physiologically based pharmacokinetic model approach

Rilpivirine is a non-nucleoside reverse transcriptase inhibitor, used for the treatment of human immunodeficiency virus type-1 infection. An open label study was conducted to investigate the pharmacokinetics (PK) and safety of a single oral dose of rilpivirine 25 mg in Japanese healthy adult subjects. No adverse events were reported. The mean C_max (144.3 ng/mL) and AUC_inf (4542 ng h/mL) in Japanese subjects were approximately 30 % higher than those reported from a similar study in Caucasian healthy subjects, whereas the median t_max and mean t_1/2 values were comparable between studies. A simple physiologically based PK model was developed to characterize the rilpivirine PK profile. The model adequately described rilpivirine PK profiles, and well-predicted drug-drug interactions. With exploration using the model, body size and CYP3A4 abundance were identified as factors which explained the observed inter-ethnic difference in rilpivirine exposure. The inter-ethnic difference in rilpivirine exposure was however considered not clinically relevant, since inter-individual variabilities of those intrinsic factors are larger than inter-ethnic ones; and the observed AUC_inf in Japanese subjects was within the range of AUC_tau associated with efficacy and safety in Phase 3 studies. This study results support the use of rilpivirine without dose modification specific to Japanese patients.

Precisely adjusting the hepatic clearance of highly extracted drugs using the modified well-stirred model

Hepatic clearance has been widely studied for over 50 yr. Many models have been developed using either theoretical or empirical tests to predict drug metabolism. The well-stirred, parallel-tube, and dispersion metabolic models have been extensively discussed. However, to our knowledge, these models cannot fully describe all relevant scenarios in hepatic clearance. We addressed this issue using the isolated perfused rat liver technique with minor modifications. Diazepam was selected to illustrate different levels of drug plasma-protein binding by changing the added concentration of human serum albumin. The free fractions of diazepam at different albumin concentrations were assayed by rapid equilibrium dialysis. The experimental data provide new insights concerning an accepted formula used to describe hepatic clearance. Regarding drug concentrations passing through the liver, the driving force concentration (C_H,ss) in terms of C_in (influx in the liver) or C_out (efflux from the liver) needs to be carefully considered when determining drug hepatic and intrinsic clearances. The newly established model, termed the modified well-stirred model, which was derived from the original formula, successfully estimated hepatic drug metabolism. Using the modified well-stirred model, a theoretical driving force concentration of diazepam passing through the liver was evaluated. The model was further used to assess the predictability of in vitro to in vivo extrapolation. This study was not intended to refute the existing models, but rather to augment them using experimental data. The results stress the importance of proper calculation of dose when the drug clearance deviates from the prediction of the well-stirred model.

A latent variable approach to account for correlated inputs in global sensitivity analysis

In drug development decision-making is often supported through model-based methods, such as physiologically-based pharmacokinetics (PBPK). Global sensitivity analysis (GSA) is gaining use for quality assessment of model-informed inference. However, the inclusion and interpretation of correlated factors in GSA has proven an issue. Here we developed and evaluated a latent variable approach for dealing with correlated factors in GSA. An approach was developed that describes the correlation between two model inputs through the causal relationship of three independent factors: the latent variable and the unique variances of the two correlated parameters. The latent variable approach was applied to a set of algebraic models and a case from PBPK. Then, this method was compared to Sobol’s GSA assuming no correlations, Sobol’s GSA with groups and the Kucherenko approach. For the latent variable approach, GSA was performed with Sobol’s method. By using the latent variable approach, it is possible to devise a unique and easy interpretation of the sensitivity indices while maintaining the correlation between the factors. Compared methods either consider the parameters independent, group the dependent variables into one unique factor or present difficulties in the interpretation of the sensitivity indices. In situations where GSA is called upon to support model-informed decision-making, the latent variable approach offers a practical method, in terms of ease of implementation and interpretability, for applying GSA to models with correlated inputs that does not violate the independence assumption. Prerequisites and limitations of the approach are discussed.

Physiologically-based pharmacokinetic modeling to predict drug interactions of lemborexant with CYP3A inhibitors

Lemborexant, a recently approved dual orexin receptor antagonist for treatment of adults with insomnia, is eliminated primarily by cytochrome P450 (CYP)3A metabolism. The recommended dose of lemborexant is 5 mg once per night, with a maximum recommended dose of 10 mg once daily. A physiologically-based pharmacokinetic (PBPK) model for lemborexant was developed and applied to integrate data obtained from in vivo drug-drug interaction (DDI) assessments, and to further explore lemborexant interaction with CYP3A inhibitors and inducers. The model predictions were in good agreement with observed pharmacokinetic data and with DDI results from clinical studies with CYP3A inhibitors, itraconazole and fluconazole. The model further predicted that DDI effects of weak CYP3A inhibitors (fluoxetine and ranitidine) are weak, and effects of moderate inhibitors (erythromycin and verapamil) are moderate. Based on the PBPK simulations and clinical efficacy and safety data, the maximum daily recommended lemborexant dose when administered with weak CYP3A inhibitors is 5 mg; co-administration of moderate and strong inhibitors should be avoided except in countries where 2.5 mg has been approved.

Implementation of a Physiologically Based Pharmacokinetic Modeling Approach to Guide Optimal Dosing Regimens for Imatinib and Potential Drug Interactions in Paediatrics

Long-term use of imatinib is effective and well-tolerated in children with chronic myeloid leukaemia (CML) yet defining an optimal dosing regimen for imatinib in younger patients is a challenge. The potential interactions between imatinib and coadministered drugs in this "special" population also remains largely unexplored. This study implements a physiologically based pharmacokinetic (PBPK) modeling approach to investigate optimal dosing regimens and potential drug interactions with imatinib in the paediatric population. A PBPK model for imatinib was developed in the Simcyp Simulator (version 17) utilizing in silico, in vitro drug metabolism, and in vivo pharmacokinetic data and verified using an independent set of published clinical pharmacokinetic data. The model was then extrapolated to children and adolescents (aged 2-18 years) by incorporating developmental changes in organ size and maturation of drug-metabolising enzymes and plasma protein responsible for imatinib disposition. The PBPK model described imatinib pharmacokinetics in adult and paediatric populations and predicted drug interaction with carbamazepine, a cytochrome P450 (CYP)3A4 and 2C8 inducer, with a good accuracy (evaluated by visual inspections of the simulation results and predicted pharmacokinetic parameters that were within 1.25-fold of the clinically observed values). The PBPK simulation suggests that the optimal dosing regimen range for imatinib is 230-340 mg/m2/d in paediatrics, which is supported by the recommended initial dose for treatment of childhood CML. The simulations also highlighted that children and adults being treated with imatinib have similar vulnerability to CYP modulations. A PBPK model for imatinib was successfully developed with an excellent performance in predicting imatinib pharmacokinetics across age groups. This PBPK model is beneficial to guide optimal dosing regimens for imatinib and predict drug interactions with CYP modulators in the paediatric population.

Requirements to Establishing Confidence in Physiologically Based Pharmacokinetic (PBPK) Models and Overcoming Some of the Challenges to Meeting Them

When scientifically well-founded, the mechanistic basis of physiologically based pharmacokinetic (PBPK) models can help reduce the uncertainty and increase confidence in extrapolations outside the studied scenarios or studied populations. However, it is not always possible to establish mechanistically credible PBPK models. Requirements to establishing confidence in PBPK models, and challenges to meeting these requirements, are presented in this article. Parameter non-identifiability is the most challenging among the barriers to establishing confidence in PBPK models. Using case examples of small molecule drugs, this article examines the use of hypothesis testing to overcome parameter non-identifiability issues, with the objective of enhancing confidence in the mechanistic basis of PBPK models and thereby improving the quality of predictions that are meant for internal decisions and regulatory submissions. When the mechanistic basis of a PBPK model cannot be established, we propose the use of simpler models or evidence-based approaches.

Quantitative Prediction of CYP3A4- and CYP3A5-Mediated Drug Interactions

We verified a physiologically-based pharmacokinetic (PBPK) model to predict cytochrome P450 3A4/5-mediated drug-drug interactions (DDIs). A midazolam (MDZ)-ketoconazole (KTZ) interaction study in 24 subjects selected by CYP3A5 genotype, and liquid chromatography and mass spectroscopy quantification of CYP3A4/5 abundance from independently acquired and genotyped human liver (n = 136) and small intestinal (N = 12) samples, were conducted. The observed CYP3A5 genetic effect on MDZ systemic and oral clearance was successfully replicated by a mechanistic framework incorporating the proteomics-informed CYP3A abundance and optimized small intestinal CYP3A4 abundance based on MDZ intestinal availability (F_G ) of 0.44. Furthermore, combined with a modified KTZ PBPK model, this framework recapitulated the observed geometric mean ratio of MDZ area under the curve (AUCR) following 200 or 400 mg KTZ, which was, respectively, 2.7-3.4 and 3.9-4.7-fold in intravenous administration and 11.4-13.4 and 17.0-19.7-fold in oral administration, with AUCR numerically lower (P > 0.05) in CYP3A5 expressers than nonexpressers. In conclusion, the developed mechanistic framework supports dynamic prediction of CYP3A-mediated DDIs in study planning by bridging DDIs between CYP3A5 expressers and nonexpressers.

A Comprehensive Framework for Physiologically-Based Pharmacokinetic Modeling in Matlab

Physiologically-based pharmacokinetic (PBPK) models are useful tools to predict clinical scenarios for special populations for whom there are high hurdles to conduct clinical trials such as children or the elderly. However, the coding of a PBPK model in a mathematical programming language can be challenging. This tutorial illustrates how to build a whole-body PBPK model in Matlab to answer specific pharmacological questions involving drug disposition and magnitudes of drug-drug interactions in different patient populations.

Use of generally recognized as safe or dietary compounds to inhibit buprenorphine metabolism: potential to improve buprenorphine oral bioavailability

The present study evaluated the potential of five generally recognized as safe (GRAS) or dietary compounds (α-mangostin, chrysin, ginger extract, pterostilbene and silybin) to inhibit oxidative (CYP) and conjugative (UGT) metabolism using pooled human intestinal and liver microsomes. Buprenorphine was chosen as the model substrate as it is extensively metabolized by CYPs to norbuprenorphine and by UGTs to buprenorphine glucuronide. Chrysin, ginger extract, α-mangostin, pterostilbene and silybin were tested for their inhibition of the formation of norbuprenorphine or buprenorphine glucuronide in both intestinal and liver microsomes. Pterostilbene was the most potent inhibitor of norbuprenorphine formation in both intestinal and liver microsomes, with IC50 values of 1.3 and 0.8 μM, respectively, while α-mangostin and silybin most potently inhibited buprenorphine glucuronide formation. The equipotent combination of pterostilbene and ginger extract additively inhibited both pathways in intestinal microsomes. Since pterostilbene and ginger extract showed potent CYP and/or UGT inhibition of buprenorphine metabolism, their equipotent combination was tested to assess the presence of synergistic inhibition. However, because the combination showed additive inhibition, it was not used while performing IVIVE analysis. Based on quantitative in vitro-in vivo extrapolation, pterostilbene (21 mg oral dose) appeared to be most effective in improving the mean predicted Foral and AUC∞ PO of buprenorphine from 3 ± 2% and 340 ± 330 ng*min/ml to 75 ± 8% and 36,000 ± 25,000 ng*min/ml, respectively. At a 10-fold lower dose of pterostilbene, the predicted buprenorphine Foral approximated sublingual bioavailability (~35%) and showed a 2-4 fold reduction in the variability around the predicted AUC∞ PO of buprenorphine. These results demonstrate the feasibility of using various GRAS/dietary compounds to inhibit substantially the metabolism by CYP and UGT enzymes to achieve higher and less variable oral bioavailability. This inhibitor strategy may be useful for drugs suffering from low and variable oral bioavailability due to extensive presystemic oxidative and/or conjugative metabolism.

Prediction of inter-individual variability on the pharmacokinetics of CYP2C8 substrates in human

Inter-individual variability in pharmacokinetics can lead to unexpected side effects and treatment failure, and is therefore an important factor in drug development. CYP2C8 is a major drug-metabolizing enzyme known to be involved in the metabolism of over 100 drugs. In this study, we predicted the inter-individual variability in AUC/Dose of CYP2C8 substrates in healthy volunteers using the Monte Carlo simulation. Inter-individual variability in the hepatic intrinsic clearance of CYP2C8 substrates (CL_int,h,2C8) was estimated from the inter-individual variability in pharmacokinetics of pioglitazone, which is a major CYP2C8 substrate. The coefficient of variation (CV) of CL_int,h,2C8 was estimated to be 40%. Using this value, the CVs of AUC/Dose of other major CYP2C8 substrates, rosiglitazone and amodiaquine, were predicted to validate the estimated CV of CL_int,h,2C8. As a result, the reported CVs of both substrates were within the 2.5-97.5 percentile range of the predicted CVs. Furthermore, the CVs of AUC/Dose of the CYP2C8 substrates loperamide and chloroquine, which are affected by renal clearance, were also successfully predicted. Combining this value with previously reported CVs of other CYPs, we were able to successfully predict the inter-individual variability in pharmacokinetics of various drugs in clinical.

Identifying populations sensitive to environmental chemicals by simulating toxicokinetic variability

The thousands of chemicals present in the environment (USGAO, 2013) must be triaged to identify priority chemicals for human health risk research. Most chemicals have little of the toxicokinetic (TK) data that are necessary for relating exposures to tissue concentrations that are believed to be toxic. Ongoing efforts have collected limited, in vitro TK data for a few hundred chemicals. These data have been combined with biomonitoring data to estimate an approximate margin between potential hazard and exposure. The most "at risk" 95th percentile of adults have been identified from simulated populations that are generated either using standard "average" adult human parameters or very specific cohorts such as Northern Europeans. To better reflect the modern U.S. population, we developed a population simulation using physiologies based on distributions of demographic and anthropometric quantities from the most recent U.S. Centers for Disease Control and Prevention National Health and Nutrition Examination Survey (NHANES) data. This allowed incorporation of inter-individual variability, including variability across relevant demographic subgroups. Variability was analyzed with a Monte Carlo approach that accounted for the correlation structure in physiological parameters. To identify portions of the U.S. population that are more at risk for specific chemicals, physiologic variability was incorporated within an open-source high-throughput (HT) TK modeling framework. We prioritized 50 chemicals based on estimates of both potential hazard and exposure. Potential hazard was estimated from in vitro HT screening assays (i.e., the Tox21 and ToxCast programs). Bioactive in vitro concentrations were extrapolated to doses that produce equivalent concentrations in body tissues using a reverse dosimetry approach in which generic TK models are parameterized with: 1) chemical-specific parameters derived from in vitro measurements and predicted from chemical structure; and 2) with physiological parameters for a virtual population. For risk-based prioritization of chemicals, predicted bioactive equivalent doses were compared to demographic-specific inferences of exposure rates that were based on NHANES urinary analyte biomonitoring data. The inclusion of NHANES-derived inter-individual variability decreased predicted bioactive equivalent doses by 12% on average for the total population when compared to previous methods. However, for some combinations of chemical and demographic groups the margin was reduced by as much as three quarters. This TK modeling framework allows targeted risk prioritization of chemicals for demographic groups of interest, including potentially sensitive life stages and subpopulations.

Incorporation of the Time-Varying Postprandial Increase in Splanchnic Blood Flow into a PBPK Model to Predict the Effect of Food on the Pharmacokinetics of Orally Administered High-Extraction Drugs

Following a meal, a transient increase in splanchnic blood flow occurs that can result in increased exposure to orally administered high-extraction drugs. Typically, physiologically based pharmacokinetic (PBPK) models have incorporated this increase in blood flow as a time-invariant fed/fasted ratio, but this approach is unable to explain the extent of increased drug exposure. A model for the time-varying increase in splanchnic blood flow following a moderate- to high-calorie meal (TV-Q _Splanch) was developed to describe the observed data for healthy individuals. This was integrated within a PBPK model and used to predict the contribution of increased splanchnic blood flow to the observed food effect for two orally administered high-extraction drugs, propranolol and ibrutinib. The model predicted geometric mean fed/fasted AUC and C _max ratios of 1.24 and 1.29 for propranolol, which were within the range of published values (within 1.0-1.8-fold of values from eight clinical studies). For ibrutinib, the predicted geometric mean fed/fasted AUC and C _max ratios were 2.0 and 1.84, respectively, which was within 1.1-fold of the reported fed/fasted AUC ratio but underestimated the reported C _max ratio by up to 1.9-fold. For both drugs, the interindividual variability in fed/fasted AUC and C _max ratios was underpredicted. This suggests that the postprandial change in splanchnic blood flow is a major mechanism of the food effect for propranolol and ibrutinib but is insufficient to fully explain the observations. The proposed model is anticipated to improve the prediction of food effect for high-extraction drugs, but should be considered with other mechanisms.

The Constraints, Construction, and Verification of a Strain-Specific Physiologically Based Pharmacokinetic Rat Model

The use of in vitro-in vivo extrapolation (IVIVE) techniques, mechanistically incorporated within physiologically based pharmacokinetic (PBPK) models, can harness in vitro drug data and enhance understanding of in vivo pharmacokinetics. This study's objective was to develop a user-friendly rat (250 g, male Sprague-Dawley) IVIVE-linked PBPK model. A 13-compartment PBPK model including mechanistic absorption models was developed, with required system data (anatomical, physiological, and relevant IVIVE scaling factors) collated from literature and analyzed. Overall, 178 system parameter values for the model are provided. This study also highlights gaps in available system data required for strain-specific rat PBPK model development. The model's functionality and performance were assessed using previous literature-sourced in vitro properties for diazepam, metoprolol, and midazolam. The results of simulations were compared against observed pharmacokinetic rat data. Predicted and observed concentration profiles in 10 tissues for diazepam after a single intravenous (i.v.) dose making use of either observed i.v. clearance (CL_iv) or in vitro hepatocyte intrinsic clearance (CL_int) for simulations generally led to good predictions in various tissue compartments. Overall, all i.v. plasma concentration profiles were successfully predicted. However, there were challenges in predicting oral plasma concentration profiles for metoprolol and midazolam, and the potential reasons and according solutions are discussed.

A physiologically based pharmacokinetic drug-disease model to predict carvedilol exposure in adult and paediatric heart failure patients by incorporating pathophysiological changes in hepatic and renal blood flows

Background and Objective

Chronic diseases are associated with pathophysiological changes that could have profound impacts on drug pharmacokinetic behaviour, with a potential need to modify the administered drug therapy. It is important to acknowledge that most patients with chronic illnesses do not have a single, predominant condition but suffer from multiple comorbidities. The rapid advancement in physiologically based pharmacokinetic (PBPK) modelling, as well as the increasing quantitative knowledge of disease-related pathophysiological changes, facilitate building of drug–disease models. However, there are only a few published examples of PBPK models incorporating the pathophysiological changes that occur with chronic diseases. The objective of this study was to develop PBPK models that incorporate the haemodynamic changes in hepatic and renal blood flows occurring in chronic heart failure (CHF) and to evaluate these changes in adults and children, using carvedilol as a model drug.

Methods

After a comprehensive literature search to select the model input parameters, two PBPK models were developed. Model 1 was based on human liver and intestinal microsome clearances, and model 2 was based on clearance by specific cytochrome P450 enzymes. After evaluation of both models in healthy adults, the reduced hepatic and renal blood flows were incorporated into the developed models to predict carvedilol exposure in the adult CHF population. The adult carvedilol models were scaled down to children by using Simcyp^® (Simcyp Ltd, Sheffield, UK). In order to show the impact of reduced organ blood flows on carvedilol disposition, the predictions in the CHF population were made with and without reductions in organ blood flows.

Results

The predictions made by both models in healthy adults were comparable and within the 2-fold error range. In adults with CHF, the mean observed/predicted ratio [ratio_(Obs/Pred)] for oral clearance (CL/F) without reductions in organ blood flows was outside the 2-fold error range, i.e. 0.34 (95 % confidence interval [CI] 0.31–0.37), with use of both models. The mean CL/F ratio_(Obs/Pred) values after incorporation of reduced organ blood flows were 1.0 (95 % CI 0.92–1.08) and 0.95 (95 % CI 0.88–1.03) with use of models 1 and 2, respectively. The mean ratio_(Obs/Pred) values for the pharmacokinetic parameters were not improved after incorporation of reduced blood flows in paediatric patients, except in those above 17 years of age, who were categorized according to the New York Heart Association classification of CHF, where the CL/F ratio_(Obs/Pred) values in two patients were closer to unity.

Conclusion

There was a strong connection between a decrease in hepatic clearance of carvedilol and an increase in the severity of CHF, especially in adults and in paediatric patients above 17 years of age. The incorporated reductions in hepatic and renal blood flows occurring in moderate and severe CHF patients resulted in improved predictions of carvedilol exposure. The developed models can be extended to predict exposures of drugs with high hepatic extraction in the CHF population.

Electronic supplementary material

The online version of this article (doi:10.1007/s40262-015-0253-7) contains supplementary material, which is available to authorized users.

Prediction of Drug-Drug Interactions Arising from CYP3A induction Using a Physiologically Based Dynamic Model

Using physiologically based pharmacokinetic modeling, we predicted the magnitude of drug-drug interactions (DDIs) for studies with rifampicin and seven CYP3A4 probe substrates administered i.v. (10 studies) or orally (19 studies). The results showed a tendency to underpredict the DDI magnitude when the victim drug was administered orally. Possible sources of inaccuracy were investigated systematically to determine the most appropriate model refinement. When the maximal fold induction (Ind_max) for rifampicin was increased (from 8 to 16) in both the liver and the gut, or when the Ind_max was increased in the gut but not in liver, there was a decrease in bias and increased precision compared with the base model (Ind_max = 8) [geometric mean fold error (GMFE) 2.12 vs. 1.48 and 1.77, respectively]. Induction parameters (mRNA and activity), determined for rifampicin, carbamazepine, phenytoin, and phenobarbital in hepatocytes from four donors, were then used to evaluate use of the refined rifampicin model for calibration. Calibration of mRNA and activity data for other inducers using the refined rifampicin model led to more accurate DDI predictions compared with the initial model (activity GMFE 1.49 vs. 1.68; mRNA GMFE 1.35 vs. 1.46), suggesting that robust in vivo reference values can be used to overcome interdonor and laboratory-to-laboratory variability. Use of uncalibrated data also performed well (GMFE 1.39 and 1.44 for activity and mRNA). As a result of experimental variability (i.e., in donors and protocols), it is prudent to fully characterize in vitro induction with prototypical inducers to give an understanding of how that particular system extrapolates to the in vivo situation when using an uncalibrated approach.

Mechanistic understanding of the nonlinear pharmacokinetics and intersubject variability of simeprevir: A PBPK-guided drug development approach

Simeprevir, a hepatitis C virus (HCV) NS3/4A protease inhibitor, displays nonlinear pharmacokinetics (PK) at therapeutic doses. Using physiologically based PK modeling, various drug-drug interactions were simulated with simeprevir as victim drug to identify whether saturation of the predominant metabolic enzyme (CYP3A4) or the active hepatic transporters (organic anion-transporting polypeptide (OATP)1B1/3) could account for the nonlinear PK. Interactions with ritonavir, a strong CYP3A4 inhibitor that does not affect OATP (at 100 mg dose), erythromycin, a moderate CYP3A4 inhibitor, and efavirenz, a moderate CYP3A inducer that does not affect OATP, demonstrated the involvement of CYP3A4. Interaction studies with low-dose cyclosporine confirmed the role of OATP. The interplay between hepatic uptake and CYP3A4 metabolism was verified by simulations with rifampicin, a potent CYP3A4 inducer and OATP1B1/3 inhibitor, and maintenance doses of cyclosporine. Saturation of gut and liver metabolism by CYP3A4, and saturation of hepatic uptake by OATP1B1/3, seem to account for the observed nonlinear PK of simeprevir.

Physiologically Based Pharmacokinetic (PBPK) Modeling and Simulation Approaches: A Systematic Review of Published Models, Applications, and Model Verification

Modeling and simulation of drug disposition has emerged as an important tool in drug development, clinical study design and regulatory review, and the number of physiologically based pharmacokinetic (PBPK) modeling related publications and regulatory submissions have risen dramatically in recent years. However, the extent of use of PBPK modeling by researchers, and the public availability of models has not been systematically evaluated. This review evaluates PBPK-related publications to 1) identify the common applications of PBPK modeling; 2) determine ways in which models are developed; 3) establish how model quality is assessed; and 4) provide a list of publically available PBPK models for sensitive P450 and transporter substrates as well as selective inhibitors and inducers. PubMed searches were conducted using the terms "PBPK" and "physiologically based pharmacokinetic model" to collect published models. Only papers on PBPK modeling of pharmaceutical agents in humans published in English between 2008 and May 2015 were reviewed. A total of 366 PBPK-related articles met the search criteria, with the number of articles published per year rising steadily. Published models were most commonly used for drug-drug interaction predictions (28%), followed by interindividual variability and general clinical pharmacokinetic predictions (23%), formulation or absorption modeling (12%), and predicting age-related changes in pharmacokinetics and disposition (10%). In total, 106 models of sensitive substrates, inhibitors, and inducers were identified. An in-depth analysis of the model development and verification revealed a lack of consistency in model development and quality assessment practices, demonstrating a need for development of best-practice guidelines.

Role of DNA Methylation on the Expression of the Anthracycline Metabolizing Enzyme AKR7A2 in Human Heart

The intracardiac synthesis of anthracycline alcohol metabolites by aldo-keto reductases (AKRs) contributes to the pathogenesis of anthracycline-related cardiotoxicity. AKR7A2 is the most abundant anthracycline reductase in hearts from donors with and without Down syndrome (DS), and its expression varies between individuals (≈tenfold). We investigated whether DNA methylation impacts AKR7A2 expression in hearts from donors with (n = 11) and without DS (n = 30). Linear models were used to test for associations between methylation status and cardiac AKR7A2 expression. In hearts from donors without DS, DNA methylation status at CpG site -865 correlated with AKR7A2 mRNA (Pearson's regression coefficient, r = -0.4051, P = 0.0264) and AKR7A2 protein expression (r = -0.5818, P = 0.0071). In heart tissue from donors with DS, DNA methylation status at CpG site -232 correlated with AKR7A2 protein expression (r = 0.8659, P = 0.0025). Multiple linear regression modeling revealed that methylation at several CpG sites is associated with the synthesis of cardiotoxic daunorubicinol. AKR7A2 methylation status in lymphoblastoid cell lines from donors with and without DS was examined to explore potential parallelisms between cardiac tissue and lymphoid cells. These results suggest that DNA methylation impacts AKR7A2 expression and the synthesis of cardiotoxic daunorubicinol.

Prediction of in vivo clearance and associated variability of CYP2C19 substrates by genotypes in populations utilizing a pharmacogenetics-based mechanistic model

It is important to examine the cytochrome P450 2C19 (CYP2C19) genetic contribution to drug disposition and responses of CYP2C19 substrates during drug development. Design of such clinical trials requires projection of genotype-dependent in vivo clearance and associated variabilities of the investigational drug, which is not generally available during early stages of drug development, but is essential for CYP2C19 substrates with multiple clearance pathways. This study evaluated the utility of pharmacogenetics-based mechanistic modeling in predicting such parameters. Hepatic CYP2C19 activity and variability within genotypes were derived from in vitro S-mephenytoin metabolic activity in genotyped human liver microsomes (N = 128). These data were then used in mechanistic models to predict genotype-dependent disposition of CYP2C19 substrates (i.e., S-mephenytoin, citalopram, pantoprazole, and voriconazole) by incorporating in vivo clearance or pharmacokinetics of wild-type subjects and parameters of other clearance pathways. Relative to the wild-type, the CYP2C19 abundance (coefficient of variation percentage) in CYP2C19*17/*17, *1/*17, *1/*1, *17/null, *1/null, and null/null microsomes was estimated as 1.85 (117%), 1.79 (155%), 1.00 (138%), 0.83 (80%), 0.38 (130%), and 0 (0%), respectively. The subsequent modeling and simulations predicted, within 2-fold of the observed, the means and variabilities of urinary S/R-mephenytoin ratio (36 of 37 genetic groups), the oral clearance of citalopram (9 of 9 genetic groups) and pantoprazole (6 of 6 genetic groups), and voriconazole oral clearance (4 of 4 genetic groups). Thus, relative CYP2C19 genotype-dependent hepatic activity and variability were quantified in vitro and used in a mechanistic model to predict pharmacokinetic variability, thus allowing the design of pharmacogenetics and drug-drug interaction trials for CYP2C19 substrates.

Physiologically based pharmacokinetic modelling to predict single- and multiple-dose human pharmacokinetics of bitopertin

BACKGROUND

Bitopertin (RG1678) is a glycine reuptake inhibitor currently in phase 3 trials for treatment of schizophrenia. This paper describes the use of physiologically based pharmacokinetic (PBPK) modelling and preclinical data to gain insights into and predict bitopertin clinical pharmacokinetics.

METHODS

Simulations of pharmacokinetics were initiated early in the drug discovery stage by integrating physicochemical properties and in vitro measurements into a PBPK rat model. Comparison of pharmacokinetics predicted by PBPK modelling with those measured after intravenous and oral dosing in rats and monkeys showed a good match and thus increased confidence that a similar approach could be applied for human prediction. After comparison of predicted plasma concentrations with those measured after single oral doses in the first clinical study, the human model was refined and then applied to simulate multiple-dose pharmacokinetics.

RESULTS

Clinical plasma concentrations measured were in good agreement with PBPK predictions. Predicted area under the plasma concentration-time curve (AUC) was within twofold of the observed mean values for all dose levels. Maximum plasma concentration (C max) at higher doses was well predicted but approximately twofold below observed values at the lower doses. A slightly less than dose-proportional increase in both AUC and C max was observed, and model simulations indicated that when the dose exceeded 50 mg, solubility limited the fraction of dose absorbed. Refinement of the absorption model with additional solubility and permeability measurements further improved the match of simulations to observed single-dose data. Simulated multiple-dose pharmacokinetics with the refined model were in good agreement with observed data.

CONCLUSIONS

Clinical pharmacokinetics of bitopertin can be well simulated with a mechanistic PBPK model. This model supports further clinical development and provides a valuable repository for pharmacokinetic knowledge gained about the molecule.

Predicting drug-drug interactions: application of physiologically based pharmacokinetic models under a systems biology approach

The development of in vitro-in vivo extrapolation (IVIVE), a 'bottom-up' approach, to predict pharmacokinetic parameters and drug-drug interactions (DDIs) has accelerated mainly due to an increase in the understanding of the multiple mechanisms involved in these interactions and the availability of appropriate in vitro systems that act as surrogates for delineating various elements of the interactions relevant to absorption, distribution, metabolism and elimination. Recent advances in the knowledge of the population variables required for IVIVE (demographic, anatomical, genetic and physiological parameters) have also contributed to the appreciation of the sources of variability and wider use of this approach for different scenarios within the pharmaceutical industry. Initially, the authors present an overview of the integration of IVIVE into 'static' and 'dynamic' models for the quantitative prediction of DDIs. The main purpose of this review is to discuss the application of IVIVE in conjunction with physiologically based pharmacokinetic modeling under a systems biology approach to characterize the potential DDIs in individual patients, including those who cannot be investigated in formal clinical trials for ethical reasons. In addition, we address the issues related to the prediction of complex DDIs involving the inhibition of cytochrome P- and transporter-mediated activities through multiple drugs.

Inhibitory effect of ketoconazole on the pharmacokinetics of a multireceptor tyrosine kinase inhibitor BMS-690514 in healthy participants: assessing the mechanism of the interaction with physiologically-based pharmacokinetic simulations

BMS-690514, a selective inhibitor of the ErbB and vascular endothelial growth factor receptors, has shown antitumor activity in early clinical development. The compound is metabolized by multiple enzymes, with CYP3A4 responsible for the largest fraction (34%) of metabolism. It is also a substrate of P-glycoprotein (P-gp) in vitro. To assess the effect of ketoconazole on BMS-690514 pharmacokinetics, 17 healthy volunteers received 200 mg BMS-690514 alone followed by 100 mg BMS-690514 with ketoconazole (400 mg once daily for 4 days). The AUC(∞) of 100 mg BMS-690514 concomitantly administered with ketoconazole was similar to that of 200 mg BMS-690514 alone. The dose-normalized C(max) and AUC(∞) of BMS-690514 from the 100-mg BMS-690514/400-mg ketoconazole treatment increased by 55% and 127%, respectively, relative to those from 200 mg BMS-690514 alone. Prediction of the drug-drug interaction (DDI) using a population-based simulator (Simcyp) indicated that, in addition to CYP3A4 inhibition, the inhibition of P-gp by ketoconazole in the intestine, liver, and kidneys must be invoked to fully account for the DDI observed. This finding suggests that the inhibition of P-gp by ketoconazole, along with its effect on CYP3A4, needs to be considered when designing a DDI study of ketoconazole with a victim drug that is a dual substrate.

Use of physiologically based pharmacokinetic modeling for assessment of drug-drug interactions

Interactions between co-administered medicines can reduce efficacy or lead to adverse effects. Understanding and managing such interactions is essential in bringing safe and effective medicines to the market. Ideally, interaction potential should be recognized early and minimized in compounds that reach late stages of drug development. Physiologically based pharmacokinetic models combine knowledge of physiological factors with compound-specific properties to simulate how a drug behaves in the human body. These software tools are increasingly used during drug discovery and development and, when integrating relevant in vitro data, can simulate drug interaction potential. This article provides some background and presents illustrative examples. Physiologically based models are an integral tool in the discovery and development of drugs, and can significantly aid our understanding and prediction of drug interactions.

Application of a systems approach to the bottom-up assessment of pharmacokinetics in obese patients: expected variations in clearance

BACKGROUND AND OBJECTIVES

The maintenance dose of a drug is dependent on drug clearance, and thus any biochemical and physiological changes in obesity that affect parameters such as cardiac output, renal function, expression of drug-metabolizing enzymes and protein binding may result in altered clearance compared with that observed in normal-weight subjects (corrected or uncorrected for body weight). Because of the increasing worldwide incidence of obesity, there is a need for more information regarding the optimal dosing of drug therapy to be made available to prescribers. This is usually provided via clinical studies in obese people; however, such studies are not available for all drugs that might be used in obese subjects. Incorporation of the relevant physiological and biochemical changes into predictive bottom-up pharmacokinetic models in order to optimize dosage regimens may offer a logical way forward for the cases where no clinical data exist. The aims of the current report are to apply such a 'systems approach' to identify the likelihood of observing variations in the clearance of drugs in obesity and morbid obesity for a set of compounds for which clinical data, as well as the necessary in vitro information, are available, and to provide a framework for assessing other drugs in the future.

METHODS

The population-specific changes in demographic, physiological and biochemical parameters that are known to be relevant to obese and morbidly obese subjects were collated and incorporated into two separate population libraries. These libraries, together with mechanistic in vitro-in vivo extrapolations (IVIVE) within the Simcyp Population-based Simulator™, were used to predict the clearance of oral alprazolam, oral caffeine, oral chlorzoxazone, oral ciclosporin, intravenous and oral midazolam, intravenous phenytoin, oral theophylline and oral triazolam. The design of the simulated studies was matched as closely as possible with that of the clinical studies. Outcome was measured by the predicted ratio of the clearance of the drug in obese and lean subjects ± its 90% confidence interval, compared with observed values. The overall statistical measures of the performance of the model to detect differences in compound clearance between obese and lean populations were investigated by measuring sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). A power calculation was carried out to investigate the impact of the sample size on the overall outcome of clinical studies.

RESULTS

The model was successful in predicting clearance in obese subjects, with the degree to which simulations could mimic the outcome of in vivo studies being greater than 60% for six of the eight drugs. A clear difference in the clearance of chlorzoxazone was correctly picked up via simulation. The overall statistical measures of the performance of the Simcyp Simulator were 100% sensitivity, 66% specificity, 60% PPV and 100% NPV. Studies designed on the basis of the ratio of the absolute values required substantial numbers of participants in order to detect a significant difference, except for phenytoin and chlorzoxazone, where the ratios of the weight-normalized clearances generally showed statistically significant differences with a smaller number of subjects.

CONCLUSION

Extension of a mechanistic predictive pharmacokinetic model to accommodate physiological and biochemical changes associated with obesity and morbid obesity allowed prediction of changes in drug clearance on the basis of in vitro data, with reasonable accuracy across a range of compounds that are metabolized by different enzymes. Prediction of the effects of obesity on drug clearance, normalized by various body size scalars, is of potential value in the design of clinical studies during drug development and in the introduction of dosage adjustments that are likely to be needed in clinical practice.

Are there differences in the catalytic activity per unit enzyme of recombinantly expressed and human liver microsomal cytochrome P450 2C9? A systematic investigation into inter-system extrapolation factors

The 'relative activity factor' (RAF) compares the activity per unit of microsomal protein in recombinantly expressed cytochrome P450 enzymes (rhCYP) and human liver without separating the potential sources of variation (i.e. abundance of enzyme per mg of protein or variation of activity per unit enzyme). The dimensionless 'inter-system extrapolation factor' (ISEF) dissects differences in activity from those in CYP abundance. Detailed protocols for the determination of this scalar, which is used in population in vitro-in vivo extrapolation (IVIVE), are currently lacking. The present study determined an ISEF for CYP2C9 and, for the first time, systematically evaluated the effects of probe substrate, cytochrome b5 and methods for assessing the intrinsic clearance (CL(int) ). Values of ISEF for S-warfarin, tolbutamide and diclofenac were 0.75 ± 0.18, 0.57 ± 0.07 and 0.37 ± 0.07, respectively, using CL(int) values derived from the kinetic values V(max) and K(m) of metabolite formation in rhCYP2C9 + reductase + b5 BD Supersomes™. The ISEF values obtained using rhCYP2C9 + reductase BD Supersomes™ were more variable, with values of 7.16 ± 1.25, 0.89 ± 0.52 and 0.50 ± 0.05 for S-warfarin, tolbutamide and diclofenac, respectively. Although the ISEF values obtained from rhCYP2C9 + reductase + b5 for the three probe substrates were statistically different (p < 0.001), the use of the mean value of 0.54 resulted in predicted oral clearance values for all three substrates within 1.4 fold of the observed literature values. For consistency in the relative activity across substrates, use of a b5 expressing recombinant system, with the intrinsic clearance calculated from full kinetic data is recommended for generation of the CYP2C9 ISEF. Furthermore, as ISEFs have been found to be sensitive to differences in accessory proteins, rhCYP system specific ISEFs are recommended.