Covariate modeling in pharmacometrics: General points for consideration.

Modeling the relationships between covariates and pharmacometric model parameters is a central feature of pharmacometric analyses. The information obtained from covariate modeling may be used for dose selection, dose individualization, or the planning of clinical studies in different population subgroups. The pharmacometric literature has amassed a diverse, complex, and evolving collection of methodologies and interpretive guidance related to covariate modeling. With the number and complexity of technologies increasing, a need for an overview of the state of the art has emerged. In this article the International Society of Pharmacometrics (ISoP) Standards and Best Practices Committee presents perspectives on best practices for planning, executing, reporting, and interpreting covariate analyses to guide pharmacometrics decision making in academic, industry, and regulatory settings.

Application of a patient-centered reverse translational systems-based approach to understand mechanisms of an adverse drug reaction of immune checkpoint inhibitors.

Immunotherapy became a key pillar of cancer therapeutics with the approvals of ipilimumab, nivolumab, and pembrolizumab, which inhibit either cytotoxic T-lymphocyte antigen-4 (CTLA-4) or programmed death-1 (PD-1) that are negative regulators of T-cell activation. However, boosting T-cell activation is often accompanied by autoimmunity, leading to adverse drug reactions (ADRs), including high grade 3-4 colitis and its severe complications whose prevalence may reach 14% for combination checkpoint inhibitors. In this research, we investigated how mechanistic differences between anti-CTLA-4 (ipilimumab) and anti-PD-1 (nivolumab and pembrolizumab) affect colitis, a general class toxicity. The data analytical platform Molecular Health Effect was utilized to map population ADR data from the US Food and Drug Administration (FDA) Adverse Event Reporting System to chemical and biological databases for hypothesis generation regarding the underlying molecular mechanisms causing colitis. Disproportionality analysis was used to assess the statistical relevance between adverse events of interest and molecular causation. We verified that the anti-CTLA-4 drug is associated with an approximately three-fold higher proportional reporting ratio associated with colitis than those of the anti-PD-1 drugs. The signal of the molecular mechanisms, including signaling pathways of inflammatory cytokines, was statistically insignificant to test the hypothesis that the severer rate of colitis associated with ipilimumab would be due to a greater magnitude of T-cell activation as a result of earlier response of the anti-CTLA-4 drug in the immune response. This patient-centered systems-based approach provides an exploratory process to better understand drug pair adverse events at pathway and target levels through reverse translation from postmarket surveillance safety reports.

A case study of a patient-centered reverse translational systems-based approach to understand adverse event profiles in drug development.

Adverse drug reactions (ADRs) of targeted therapy drugs (TTDs) are frequently unexpected and long-term toxicities detract from exceptional efficacy of new TTDs. In this proof-of-concept study, we explored how molecular causation involved in trastuzumab-induced cardiotoxicity changes when trastuzumab was given in combination with doxorubicin, tamoxifen, paroxetine, or lapatinib. The data analytical platform Molecular Health Effect was utilized to map population ADR data from the US Food and Drug Administration (FDA) Adverse Event Reporting System to chemical and biological databases (such as UniProt and Reactome), for hypothesis generation regarding the underlying molecular mechanisms causing cardiotoxicity. Disproportionality analysis was used to assess the statistical relevance between adverse events of interest and molecular causation. Literature search was performed to compare the established hypotheses to published experimental findings. We found that the combination therapy of trastuzumab and doxorubicin may affect mitochondrial dysfunction in cardiomyocytes through different molecular pathways such as BCL-X and PGC-1α proteins, leading to a synergistic effect of cardiotoxicity. We found, on the other hand, that trastuzumab-induced cardiotoxicity would be diminished by concomitant use of tamoxifen, paroxetine, and/or lapatinib. Tamoxifen and paroxetine may cause less cardiotoxicity through an increase in antioxidant activities, such as glutathione conjugation. Lapatinib may decrease the apoptotic effects in cardiomyocytes by altering the effects of trastuzumab on BCL-X proteins. This patient-centered systems-based approach provides, based on the trastuzumab-induced ADR cardiotoxicity, an example of how to apply reverse translation to investigate ADRs at the molecular pathway and target level to understand the causality and prevalence during drug development of novel therapeutics.

Semi-Mechanistic Pharmacokinetic Model to Guide the Dose Selection of Nimotuzumab in Patients with Autosomal Dominant Polycystic Kidney Disease.

Autosomal dominant polycystic kidney disease (ADPKD) is a genetic disease characterized by an overexpression of epidermal growth factor receptor (EGFR). Nimotuzumab is a recombinant humanized monoclonal antibody against human EGFR. The aim of this study was to develop a population pharmacokinetic model for nimotuzumab and to identify demographic and clinical predictive factors of the pharmacokinetic variability. The population pharmacokinetics (PopPK) of nimotuzumab was characterized using a nonlinear mixed-effect modeling approach with NONMEM®. A total of 422 log-transformed concentration-versus-time datapoints from 20 patients enrolled in a single-center phase I clinical trial were used. Quasi steady state approximation of the full TMDD (target-mediated drug disposition) model with constant target concentration best described the concentration-time profiles. A turnover mediator was included which stimulates the non-specific clearance of mAb in the central compartment in order to explain the reduced levels at higher doses. Covariates had no influence on the PK (pharmacokinetics) parameters. The model was able to detect that the maximum effective dose in ADPKD subjects is 100 mg. The developed PopPK model may be used to guide the dose selection for nimotuzumab during routine clinical practice in patients with polycystic kidney disease. The model will further support the ongoing investigations of the PK/PD relationships of nimotuzumab to improve its therapeutic use in other disease areas.

Quantitative Systems Pharmacology Model-Based Predictions of Clinical Endpoints to Optimize Warfarin and Rivaroxaban Anti-Thrombosis Therapy.

BACKGROUND: Tight monitoring of efficacy and safety of anticoagulants such as warfarin is imperative to optimize the benefit-risk ratio of anticoagulants in patients. The standard tests used are measurements of prothrombin time (PT), usually expressed as international normalized ratio (INR), and activated partial thromboplastin time (aPTT). OBJECTIVE: To leverage a previously developed quantitative systems pharmacology (QSP) model of the human coagulation network to predict INR and aPTT for warfarin and rivaroxaban, respectively. METHODS: A modeling and simulation approach was used to predict INR and aPTT measurements of patients receiving steady-state anticoagulation therapy. A previously developed QSP model was leveraged for the present analysis. The effect of genetic polymorphisms known to influence dose response of warfarin (CYP2C9, VKORC1) were implemented into the model by modifying warfarin clearance (CYP2C9 *1: 0.2 L/h; *2: 0.14 L/h, *3: 0.04 L/h) and the concentration of available vitamin K (VKORC1 GA: -22% from normal vitamin K concentration; AA: -44% from normal vitamin K concentration). Virtual patient populations were used to assess the ability of the model to accurately predict routine INR and aPTT measurements from patients under long-term anticoagulant therapy. RESULTS: The introduced model accurately described the observed INR of patients receiving long-term warfarin treatment. The model was able to demonstrate the influence of genetic polymorphisms of CYP2C9 and VKORC1 on the INR measurements. Additionally, the model was successfully used to predict aPTT measurements for patients receiving long-term rivaroxaban therapy. CONCLUSION: The QSP model accurately predicted INR and aPTT measurements observed during routine therapeutic drug monitoring. This is an exemplar of how a QSP model can be adapted and used as a model-based precision dosing tool during clinical practice and drug development to predict efficacy and safety of anticoagulants to ultimately help optimize anti-thrombotic therapy.

A clinical population pharmacokinetic/pharmacodynamic model for BIIB059, a monoclonal antibody for the treatment of systemic and cutaneous lupus erythematosus.

A population pharmacokinetic/pharmacodynamic (popPK/PD) model for BIIB059 (anti-blood dendritic cell antigen 2 [anti-BDCA2]), a humanized immunoglobulin G1 monoclonal antibody currently under development for the treatment of SLE and CLE, is presented. BIIB059 binds BDCA2, a plasmacytoid dendritic cell (pDC)-specific receptor that inhibits the production of IFN-I and other inflammatory mediators when ligated. Phase 1 PK and PD data of healthy adult volunteers (HV, n = 87) and SLE subjects (n = 22) were utilized for the development of the popPK/PD model. The data included single and multiple dosing of intravenous and subcutaneous BIIB059. BDCA2 internalization (PD marker) was measured for all subjects by monitoring reduction of BDCA2 on pDC cell surface and used for development of the popPD model. A two-compartment popPK model with linear plus non-linear elimination was found to best describe BIIB059 PK. BDCA2 levels were best captured using an indirect response model with stimulation of the elimination of BDCA2. Clearance in SLE subjects was 25% higher compared to HV (6.87 vs 5.52 mL/h). Bodyweight was identified as only other covariate on clearance and central volume. The estimates of EC50 and Emax were 0.35 µg/mL and 8.92, respectively. No difference in EC50 and Emax was observed between SLE and HV. The popPK/PD model described the data accurately, as evaluated by pcVPCs and bootstrap. The presented popPK/PD model for BIIB059 provides valuable insight into the dynamics and dose-response relationship of BIIB059 for the treatment of SLE and CLE and was used to guide dose selection for the Phase 2 clinical study (NCT02847598).

Semi-mechanistic modelling platform to assess cardiac contractility and haemodynamics in preclinical cardiovascular safety profiling of new molecular entities.

BACKGROUND AND PURPOSE: Cardiovascular safety is one of the most frequent causes of safety-related attrition both preclinically and clinically. Preclinical cardiovascular safety is routinely assessed using dog telemetry monitoring key cardiovascular functions. The present research was to develop a semi-mechanistic modelling platform to simultaneously assess changes in contractility (dPdtmax ), heart rate (HR) and mean arterial pressure (MAP) in preclinical studies. EXPERIMENTAL APPROACH: Data from dPdtmax , HR, preload (left ventricular end-diastolic pressure [LVEDP]) and MAP were available from dog telemetry studies after dosing with atenolol (n = 27), salbutamol (n = 5), L-NG -nitroarginine methyl ester (L-NAME; n = 4), milrinone (n = 4), verapamil (n = 12), dofetilide (n = 8), flecainide (n = 4) and AZ001 (n = 14). Literature model for rat CV function was used for the structural population pharmacodynamic model development. LVEDP was evaluated as covariate to account for the effect of preload on dPdtmax . KEY RESULTS: The model was able to describe drug-induced changes in dPdtmax , HR and MAP for all drugs included in the developed framework adequately, by incorporating appropriate drug effects on dPdtmax , HR and/or total peripheral resistance. Consistent with the Starling's law, incorporation of LVEDP as a covariate on dPdtmax to correct for the preload effect was found to be statistically significant. CONCLUSIONS AND IMPLICATIONS: The contractility and haemodynamics semi-mechanistic modelling platform accounts for diurnal variation, drug-induced changes and inter-animal variation. It can be used to hypothesize and evaluate pharmacological effects and provide a holistic cardiovascular safety profile for new drugs.

Performance of the SAEM and FOCEI Algorithms in the Open-Source, Nonlinear Mixed Effect Modeling Tool nlmixr.

The free and open-source package nlmixr implements pharmacometric nonlinear mixed effects model parameter estimation in R. It provides a uniform language to define pharmacometric models using ordinary differential equations. Performances of the stochastic approximation expectation-maximization (SAEM) and first order-conditional estimation with interaction (FOCEI) algorithms in nlmixr were compared with those found in the industry standards, Monolix and NONMEM, using the following two scenarios: a simple model fit to 500 sparsely sampled data sets and a range of more complex compartmental models with linear and nonlinear clearance fit to data sets with rich sampling. Estimation results obtained from nlmixr for FOCEI and SAEM matched the corresponding output from NONMEM/FOCEI and Monolix/SAEM closely both in terms of parameter estimates and associated standard errors. These results indicate that nlmixr may provide a viable alternative to existing tools for pharmacometric parameter estimation.

Nonlinear Mixed-Effects Model Development and Simulation Using nlmixr and Related R Open-Source Packages.

nlmixr is a free and open-source R package for fitting nonlinear pharmacokinetic (PK), pharmacodynamic (PD), joint PK-PD, and quantitative systems pharmacology mixed-effects models. Currently, nlmixr is capable of fitting both traditional compartmental PK models as well as more complex models implemented using ordinary differential equations. We believe that, over time, it will become a capable, credible alternative to commercial software tools, such as NONMEM, Monolix, and Phoenix NLME.

Physiologically Based Pharmacokinetic Modeling to Evaluate Formulation Factors Influencing Bioequivalence of Metoprolol Extended-Release Products.

The University of Florida Center for Pharmacometrics and Systems Pharmacology and the Food and Drug Administration Office of Generic Drugs have collaborated on a research project to develop a mechanism- and risk-based strategy that systematically investigates postmarketing reports of therapeutic inequivalence following the switch between brand and generic drug products. In this study we developed and qualified a physiologically based pharmacokinetic model to systematically investigate the influence of drug- and formulation-related properties on the oral absorption and bioequivalence of modified-release products using metoprolol as an example. Our findings show that the properties of the release-controlling polymer are the critical attributes for in vitro dissolution, in vivo absorption, and systemic exposure (ie, pharmacokinetics) and, thus, the bioequivalence of metoprolol extended-release products rather than the properties of the drug itself. Differences in dissolution rate can result in significant differences in maximum plasma concentration but not in area under the concentration-time curve.

Evaluating the Clinical Impact of Formulation Variability: A Metoprolol Extended-Release Case Study.

The objective of this research was to evaluate the impact of changes in the formulation of metoprolol extended-release (ER) tablets on dissolution, pharmacokinetic, and exercise-induced heart rate (EIHR) using a combined physiologically based absorption pharmacokinetic, and population pharmacokinetic/pharmacodynamic modeling and simulation approach. Using a previously developed physiologically based absorption pharmacokinetic model in DDDPlus and GastroPlus, we simulated the changes in drug release and exposure as the result of quantitative changes in the release-controlling excipient, hydroxylpropylmethylcellulose, for 50 and 200 mg. The similarity of dissolution profiles was assessed using the f2 test, and bioequivalence was tested on the simulated pharmacokinetic profiles. We used the simulated concentration-time profiles following formulation changes as pharmacokinetic input into a population pharmacokinetic/pharmacodynamic model newly developed in NONMEM to determine if changes in pharmacokinetics lead to clinically significant changes in pharmacodynamics. Pharmacodynamic assessment was based on the percentage reduction in the EIHR from baseline. Therapeutic effect was considered similar when the model-predicted EIHR was within 50% to 85% of the average maximum EIHR of healthy 30-year-old subjects. A 40% or more increase in the release rate constant resulted in dissimilarity in dissolution profiles and bioINequivalence in pharmacokinetics for both 50 and 200 mg. Formulation-related differences in drug release of metoprolol ER tablets can lead to differences in pharmacokinetics. However, the evaluated pharmacokinetic differences do not lead to clinically meaningful differences in EIHR, suggesting that EIHR may not be sensitive enough to detect changes in pharmacokinetics of metoprolol ER products.

Is Bioequivalence Established Based on the Reference-Scaled Average Bioequivalence Approach Relevant to Chronic Administration of Phenytoin? Perspectives Based on Population Pharmacokinetic Modeling and Simulations.

Phenytoin demonstrates time-dependent and nonlinear pharmacokinetics (PK) within the therapeutic range of 10 to 20 µg/mL. There are discussions on the relevance of bioequivalence (BE) demonstrated in a single-dose BE study in healthy subjects to exposure under chronic use conditions in patients, particularly given that phenytoin has a narrow therapeutic index. The objective of this study was to quantitatively evaluate the appropriateness of single-dose PK BE through simulations for the phenytoin extended-capsule products. A previously published population PK model was updated to account for the interoccasion variability using the dense PK data of the reference listed drug (Dilantin) from 5 single-dose, fully replicated BE studies (n = 124). BE studies with alternative designs were simulated using the developed PK model and subsequently analyzed accordingly: Scenario 1, multiple-dose, 2-period, crossover BE studies with an average BE approach; Scenario 2, single-dose, 4-period, fully replicated BE studies with a reference-scaled average BE approach as recommended in the product-specific guidance. In both scenarios, hypothetical phenytoin capsules with different formulation-related PK parameters, such as relative bioavailability and absorption rates, were included in the simulations. The results showed that the both scenarios provided the same results with respect to BE conclusions.

A pre-clinical quantitative model predicts the pharmacokinetics/pharmacodynamics of an anti-BDCA2 monoclonal antibody in humans.

BIIB059 is a novel humanized monoclonal antibody (mAb) that is currently under development for the treatment of Systemic Lupus Erythematosus and Cutaneous Lupus Erythematosus. BIIB059 is targeted against the blood dendritic cell antigen 2 (BDCA2), a receptor exclusively expressed on the surface of plasmacytoid dendritic cells (pDCs). Herein, we utilized pre-clinical pharmacokinetic (PK) and pharmacodynamic (PD) data to develop a non-human primate (NHP) model and to address whether the NHP model can be successfully scaled to predict the human PK/PD. In particular, PK data from 17 cynomolgus monkeys were utilized for PK model development, wherein BIIB059 was administered intravenously (1 and 10 mg/kg single-dosing and 5 mg/kg multiple-dosing) or subcutaneously (0.2 and 7.5 mg/kg single-dosing). Additionally, PD data (BDCA2 receptor density on pDCs) from 6 cynomolgus monkeys were used for the development of the PD model. The developed NHP two-compartment PK model, linked with an indirect response PD model, was subsequently scaled to humans by combining traditional allometric PK scaling with sensitivity-analysis-driven scaling of the PD. The scaled PK/PD model was then used to simulate the human PK/PD for different dose levels. When clinical data from the BIIB059 Phase I study became available, they were used to evaluate the predictability of the scaled PK/PD model and the model simulations were in agreement with the clinical data. Therefore, the presented approach is suggested to be employed in scaling pre-clinical mAb models to support the selection of safe first-in-human doses and, more broadly, the prediction of PK/PD in the clinic.

A Semi-Mechanistic Population Pharmacokinetic Model of Nusinersen: An Antisense Oligonucleotide for the Treatment of Spinal Muscular Atrophy.

A pharmacokinetic (PK) model was developed for nusinersen, an antisense oligonucleotide (ASO) that is the first approved treatment for spinal muscular atrophy (SMA). The model was built with data from 92 nonhuman primates (NHPs) following intrathecal doses (0.3-7 mg) and characterized the PK in cerebrospinal fluid (CSF), plasma, total spinal cord, brain, and pons. The estimated volumes were 13.6, 937, 4.5, 53.8, and 2.11 mL, respectively. Global sensitivity analysis demonstrated that the CSF-to-plasma drug distribution rate (0.09 hour-1 ) is a major determinant of the maximum nusinersen concentration in central nervous system (CNS) tissues. Physiological age-based and body weight-based allometric scaling was implemented with exponent values of -0.08 and 1 for the rate constants and the volume of distribution, respectively. Simulations of the scaled model were in agreement with clinical observations from 52 pediatric phase I PK profiles. The developed model can be used to guide the design of clinical trials with ASOs.

Perspective on the State of Pharmacometrics and Systems Pharmacology Integration.

Reliance on modeling and simulation in drug discovery and development has dramatically increased over the past decade. Two disciplines at the forefront of this activity, pharmacometrics and systems pharmacology (SP), emerged independently from different fields; consequently, a perception exists that only few examples integrate these approaches. Herein, we review the state of pharmacometrics and SP integration and describe benefits of combining these approaches in a model-informed drug discovery and development framework.

Model-Based Approach to Predict Adherence to Protocol During Antiobesity Trials.

Development of antiobesity drugs is continuously challenged by high dropout rates during clinical trials. The objective was to develop a population pharmacodynamic model that describes the temporal changes in body weight, considering disease progression, lifestyle intervention, and drug effects. Markov modeling (MM) was applied for quantification and characterization of responder and nonresponder as key drivers of dropout rates, to ultimately support the clinical trial simulations and the outcome in terms of trial adherence. Subjects (n = 4591) from 6 Contrave® trials were included in this analysis. An indirect-response model developed by van Wart et al was used as a starting point. Inclusion of drug effect was dose driven using a population dose- and time-dependent pharmacodynamic (DTPD) model. Additionally, a population-pharmacokinetic parameter- and data (PPPD)-driven model was developed using the final DTPD model structure and final parameter estimates from a previously developed population pharmacokinetic model based on available Contrave® pharmacokinetic concentrations. Last, MM was developed to predict transition rate probabilities among responder, nonresponder, and dropout states driven by the pharmacodynamic effect resulting from the DTPD or PPPD model. Covariates included in the models and parameters were diabetes mellitus and race. The linked DTPD-MM and PPPD-MM was able to predict transition rates among responder, nonresponder, and dropout states well. The analysis concluded that body-weight change is an important factor influencing dropout rates, and the MM depicted that overall a DTPD model-driven approach provides a reasonable prediction of clinical trial outcome probabilities similar to a pharmacokinetic-driven approach.

Development and validation of a LC-MS/MS assay for quantification of cisplatin in rat plasma and urine.

Till date, no analytical method published to detect Cisplatin has been validated according to the U.S. Food and Drug Administration (FDA) guidance using liquid chromatography mass spectrometry (LC-MS/MS). We report, a validated LC-MS/MS method for quantitative determination of cisplatin in rat plasma and urine according to FDA guidlines. Cisplatin is a platinum containing compound used for the treatment of different types of cancers. Quantitative determination of cisplatin has been carried out using atomic absorption spectroscopy, high pressure liquid chromatography with phosphorescence, ultra-violet detection, or with inductively coupled plasma mass spectrometry. Few LC-MS/MS methods have been reported for the analysis of cisplatin either for direct quantification or indirect by derivatizing with organic compounds but none of the reported methods have validated the method. The developed and validated assay presented here is a highly sensitive LC-MS/MS method developed and validated for the quantitative determination of cisplatin following derivatization with diethyldithiocarbamate (DDTC) in order to detect platinum (Pt) of cisplatin, suitable for pharmacokinetic studies in rats and to further use it to study human toxicology. Chromatographic separation was achieved using a Poroshell 120 EC-C18 column (3×50mm, 2.7µm) with a binary gradient mobile phase. Quantification was performed on a triple quadruple with electrospray ionization and detection was performed using multiple reaction monitoring. The method has a limit of detection of 1ng/mL, and the quantifiable range was 3-3000ng/mL in rat plasma and urine. The method was accurate and precise with an accuracy and precision for intra-day and inter-day of ±20% for lower limit of quantitation and of ±15% for low, mid and high quality control samples. This method was successfully applied to study the pharmacokinetic profile of cisplatin in rat plasma and urine given a range of doses from 0.5 to 3.5mg/kg.

Systems pharmacology to predict drug safety in drug development.

Ensuring that drugs are safe and effective is a very high priority for drug development and the US Food and Drug Administration review process. This is especially true today because of faster approval times and smaller clinical trials, especially in oncology and rare diseases. In light of these trends, systems pharmacology is seen as an essential strategy to understand and predict adverse drug events during drug development by analyzing interactions between drugs and multiple targets rather than the traditional "one-drug-one-target" approach. This commentary offers an overview of the current trends and challenges of using systems pharmacology to reduce the risks of unintended adverse events.