Commentary on the MID3 Good Practices Paper

Bioequivalence risk assessment of oral formulations containing racemic ibuprofen through a chiral physiologically based pharmacokinetic model of ibuprofen enantiomers

The characterization of the time course of ibuprofen enantiomers can be useful in the selection of the most sensitive analyte in bioequivalence studies. Physiologically based pharmacokinetic (PBPK) modelling and simulation represents the most efficient methodology to virtually assess bioequivalence outcomes. In this work, we aim to develop and verify a PBPK model for ibuprofen enantiomers administered as a racemic mixture with different immediate release dosage forms to anticipate bioequivalence outcomes based on different particle size distributions. A PBPK model incorporating stereoselectivity and non-linearity in plasma protein binding and metabolism as well as R-to-S unidirectional inversion has been developed in Simcyp®. A dataset composed of 11 Phase I clinical trials with 54 scenarios (27 per enantiomer) and 14,452 observations (7129 for R-ibuprofen and 7323 for S-ibuprofen) was used. Prediction errors for AUC0-t and Cmax for both enantiomers fell within the 0.8-1.25 range in 50/54 (93 %) and 42/54 (78 %) of scenarios, respectively. Outstanding model performance, with 10/10 (100 %) of Cmax and 9/10 (90 %) of AUC0-t within the 0.9-1.1 range, was demonstrated for oral suspensions, which strongly supported its use for bioequivalence risk assessment. The deterministic bioequivalence risk assessment has revealed R-ibuprofen as the most sensitive analyte to detect differences in particle size distribution for oral suspensions containing 400 mg of racemic ibuprofen, suggesting that achiral bioanalytical methods would increase type II error and declare non-bioequivalence for formulations that are bioequivalent for the eutomer.

An industry perspective on current QSP trends in drug development

2023 marks the 10th anniversary of Natpara's submission to the US FDA, which led to the first recorded regulatory interaction where a decision was supported by Quantitative and Systems Pharmacology (QSP) simulations. It had taken about 5 years for the timid QSP discipline to emerge as an effective Model-Informed Drug Development (MIDD) tool with visible impact in the pharmaceutical industry. Since then, the presence of QSP in the regulatory environment has continued to increase, to the point that the Agency reported 60 QSP submissions in 2020 alone, representing ~ 4% of their annual IND submissions [1]. What sort of industry mindset has enabled QSP to reach this level of success? How does QSP fit within the MIDD paradigm? Does QSP mean the same to Discovery and to Clinical Development projects? How do 'platforms' compare to 'fit-for-purpose' QSP models in an industrial setting? Can QSP and empirical Pharmacokinetic-Pharmacodynamic (PKPD) modelling be complementary? What level of validation is required to inform drug development decisions? This article reflects on all these questions, in particular addressing those audiences with limited line-of-sight into the drug industry decision-making machinery.

Phys-DAT: A physiologically-based pharmacokinetic model for unraveling the dissolution, transit and absorption processes using PhysPK®

BACKGROUND AND OBJECTIVE

In silico methods have become the key for efficiently testing and qualifying drug properties. Due to the complexity of the LADME processes and drug characteristics associated to oral drug absorption, there is a growing demand in the development of Physiologically-based Pharmacokinetic (PBPK) software with greater flexibility. Thus, the aims of this work are (i) to develop a mechanistic-based modeling framework of dissolution, transit and absorption (Phys-DAT) processes in the PhysPK platform and (ii) to assess the predictive power of the acausal MOOM methodology embedded in Phys-DAT versus reference ODE-based PBPK software.

METHODS

A PBPK model was developed including unreleased, undissolved and dissolved thermodynamic states of the drug. The gastrointestinal tract (GI) was represented by nine compartments and first-order transit kinetics was assumed for the drug fractions. Dissolution processes were described using solubility-independent or solubility-dependent mechanisms and pH effects. Linear transit and linear absorption mechanisms including gradual decrease absorption rate were considered to represent the passive diffusion process. Internal validation of the Phys-DAT model was performed through simulation-based analysis, considering different theoretical scenarios. External validation was carried out using in silico and in vivo data of GI segments and plasma concentrations. Both BCS I and II class drugs were included.

RESULTS

The model predicts plasma-concentration profiles of each compartment for undissolved, dissolved, and absorbed fractions using PhysPK® v.2.4.1. Internal and external validations demonstrate that the model aligned with the theoretical assumptions and accurately predicted Cmax, Tmax, and AUC 0-t for both BCS I and II drugs. Average Fold Error (AFE), Absolute Average Fold Error (AAFE), and Percent Prediction Error (PPE) calculations indicate good predictive performance, with predicted/observed ratios falling within the acceptable range.

CONCLUSIONS

Phys-DAT represents a mechanistic model for predicting oral absorption, including the dissolution, pH effect, transit, and absorption processes. PhysPK has shown to be a tool with strong prediction accuracy, similar to the obtained by ODE-based PBPK reference software, and the results obtained with the Phys-DAT model for oral administered drugs showed predictive reliability in healthy volunteers, setting the basis to determine the interchangeability of the acausal MOOM methodology with other modeling approaches.

An AI Approach to Generating MIDD Assets Across the Drug Development Continuum

Model-informed drug development involves developing and applying exposure-based, biological, and statistical models derived from preclinical and clinical data sources to inform drug development and decision-making. Discrete models are generated from individual experiments resulting in a single model expression that is utilized to inform a single stage-gate decision. Other model types provide a more holistic view of disease biology and potentially disease progression depending on the appropriateness of the underlying data sources for that purpose. Despite this awareness, most data integration and model development approaches are still reliant on internal (within company) data stores and traditional structural model types. An AI/ML-based MIDD approach relies on more diverse data and is informed by past successes and failures including data outside a host company (external data sources) that may enhance predictive value and enhance data generated by the sponsor to reflect more informed and timely experimentation. The AI/ML methodology also provides a complementary approach to more traditional modeling efforts that support MIDD and thus yields greater fidelity in decision-making. Early pilot studies support this assessment but will require broader adoption and regulatory support for more evidence and refinement of this paradigm. An AI/ML-based approach to MIDD has the potential to transform regulatory science and the current drug development paradigm, optimize information value, and increase candidate and eventually product confidence with respect to safety and efficacy. We highlight early experiences with this approach using the AI compute platforms as representative examples of how MIDD can be facilitated with an AI/ML approach.

In Vitro to In Vivo Extrapolation Linked to Physiologically Based Pharmacokinetic Models for Assessing the Brain Drug Disposition

Drug development for the central nervous system (CNS) is a complex endeavour with low success rates, as the structural complexity of the brain and specifically the blood-brain barrier (BBB) poses tremendous challenges. Several in vitro brain systems have been evaluated, but the ultimate use of these data in terms of translation to human brain concentration profiles remains to be fully developed. Thus, linking up in vitro-to-in vivo extrapolation (IVIVE) strategies to physiologically based pharmacokinetic (PBPK) models of brain is a useful effort that allows better prediction of drug concentrations in CNS components. Such models may overcome some known aspects of inter-species differences in CNS drug disposition. Required physiological (i.e. systems) parameters in the model are derived from quantitative values in each organ. However, due to the inability to directly measure brain concentrations in humans, compound-specific (drug) parameters are often obtained from in silico or in vitro studies. Such data are translated through IVIVE which could be also applied to preclinical in vivo observations. In such exercises, the limitations of the assays and inter-species differences should be adequately understood in order to verify these predictions with the observed concentration data. This report summarizes the state of IVIVE-PBPK-linked models and discusses shortcomings and areas of further research for better prediction of CNS drug disposition.

Quantitative Proteomic Map of Enzymes and Transporters in the Human Kidney: Stepping Closer to Mechanistic Kidney Models to Define Local Kinetics

The applications of translational modeling of local drug concentrations in various organs had a sharp increase over the last decade. These are part of the model-informed drug development initiative, adopted by the pharmaceutical industry and promoted by drug regulatory agencies. With respect to the kidney, the models serve as a bridge for understanding animal vs. human observations related to renal drug disposition and any consequential adverse effects. However, quantitative data on key drug-metabolizing enzymes and transporters relevant for predicting renal drug disposition are limited. Using targeted and global quantitative proteomics, we determined the abundance of multiple enzymes and transporters in 20 human kidney cortex samples. Nine enzymes and 22 transporters were quantified (8 for the first time in the kidneys). In addition, > 4,000 proteins were identified and used to form an open database. CYP2B6, CYP3A5, and CYP4F2 showed comparable, but generally low expression, whereas UGT1A9 and UGT2B7 levels were the highest. Significant correlation between abundance and activity (measured by mycophenolic acid clearance) was observed for UGT1A9 (Rs = 0.65, P = 0.004) and UGT2B7 (Rs = 0.70, P = 0.023). Expression of P-gp ≈ MATE-1 and OATP4C1 transporters were high. Strong intercorrelations were observed between several transporters (P-gp/MRP4, MRP2/OAT3, and OAT3/OAT4); no correlation in expression was apparent for functionally related transporters (OCT2/MATEs). This study extends our knowledge of pharmacologically relevant proteins in the kidney cortex, with implications on more prudent use of mechanistic kidney models under the general framework of quantitative systems pharmacology and toxicology.

Scientific and regulatory evaluation of mechanistic in silico drug and disease models in drug development: Building model credibility

The value of in silico methods in drug development and evaluation has been demonstrated repeatedly and convincingly. While their benefits are now unanimously recognized, international standards for their evaluation, accepted by all stakeholders involved, are still to be established. In this white paper, we propose a risk‐informed evaluation framework for mechanistic model credibility evaluation. To properly frame the proposed verification and validation activities, concepts such as context of use, regulatory impact and risk‐based analysis are discussed. To ensure common understanding between all stakeholders, an overview is provided of relevant in silico terminology used throughout this paper. To illustrate the feasibility of the proposed approach, we have applied it to three real case examples in the context of drug development, using a credibility matrix currently being tested as a quick‐start tool by regulators. Altogether, this white paper provides a practical approach to model evaluation, applicable in both scientific and regulatory evaluation contexts.

Opening a debate on open-source modeling tools: Pouring fuel on fire versus extinguishing the flare of a healthy debate

As model‐informed drug development becomes an integral part of modern approaches to the discovery of new therapeutic entities and showing their safety and effectiveness, modalities of incorporating the paradigm into widespread practice require a revisit. Traditionally, modeling and simulation (M&S) have been performed by specialized teams who create bespoke models for each case and have reservations about letting modeling be done by the greater mass of scientists engaged in various stages of drug development. An analogy can be drawn between M&S and automobiles: typical drivers of ordinary cars use them for daily tasks, such as going from point A to B whereas specialized Formula 1 drivers using bespoke individually made cars to test the latest technologies. The reliability and robustness of ordinary cars for the first group requires elements related to quality and endurance that are very different from those applicable to any Formula 1 car supported by a large team of engineers. In this commentary, we frame and analyze the problems concerning the structure and setup of various M&S tools, and their pros and cons. We demonstrate that many misconceptions have precluded having an open discussion on what each modality of M&S tools strives to achieve, and we provide data and evidence that support the move of M&S to main stream use by many, as opposed to specialized usage by few. Parallels are drawn in many other areas involving laboratory instrumentation, statistical analyses, and so on.

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

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

Evaluation of Assumptions Underpinning Pharmacometric Models

Assumptions inherent to pharmacometric model development and use are not routinely acknowledged, described, or evaluated. The aim of this work is to present a framework for systematic evaluation of assumptions. To aid identification of assumptions, we categorise assumptions into two types: implicit and explicit assumptions. Implicit assumptions are inherent in a method or model and underpin its derivation and use. Explicit assumptions arise from heuristic principles and are typically defined by the user to enable the application of a method or model. A flowchart was developed for systematic evaluation of assumptions. For each assumption, the impact of assumption violation ('significant', 'insignificant', 'unknown') and the probability of assumption violation ('likely', 'unlikely', 'unknown') will be evaluated based on prior knowledge or the result of an additional bespoke study to arrive at a decision ('go', 'no-go') for both model building and model use. A table of assumptions with standardised headings has been proposed to facilitate the documentation of assumptions and evaluation of results. The utility of the proposed framework was illustrated using four assumptions underpinning a top-down model describing the warfarin-coagulation proteins' relationship. The next step of this work is to apply the framework to a series of other settings to fully assess its practicality and its value in identifying and making inferences from assumptions.

Pharmacokinetics of Intravitreal Anti-VEGF Drugs in Age-Related Macular Degeneration

Intravitreal administration of anti-vascular endothelial growth factor (VEGF) antibodies has become the standard treatment for Age-Related Macular Degeneration; however, the knowledge of their pharmacokinetics is limited. A comprehensive review of the preclinical and clinical pharmacokinetic data that were obtained in different studies with intravitreal bevacizumab, ranibizumab, and aflibercept has been conducted. Moreover, the factors that can influence the vitreous pharmacokinetics of these drugs, as well as the methods that were used in the studies for analytical determination, have been exposed. These anti-VEGF drugs present different charge and molecular weights, which play an important role in vitreous distribution and elimination. The pharmacokinetic parameters that were collected differ depending on the species that were involved in the studies and on physiological and pathological conditions, such as vitrectomy and lensectomy. Knowledge of the intravitreal pharmacokinetics of the anti-VEGF drugs that were used in clinical practice is of vital importance.

Current status and future perspective on preclinical pharmacokinetic and pharmacodynamic (PK/PD) analysis: Survey in Japan pharmaceutical manufacturers association (JPMA)

Preclinical pharmacokinetic/pharmacodynamic (PK/PD) analysis is an efficient tool for the translational research and proof of mechanism/concept in animals. The questionnaire survey on the practice of preclinical PK/PD analysis was conducted in the member companies of the Japan Pharmaceutical Manufacturers Association (JPMA). According to the survey, 60% of companies conducted preclinical PK/PD analysis and its impact for drug development was different between each of the companies. The frequently analyzed therapeutic areas of preclinical PK/PD analysis were neurology, inflammation and metabolic disease, and those are different from the therapeutic area (infectious disease and oncology) in which PK/PD analysis was considered as effective by the present survey. Many companies which have used preclinical PK/PD analysis for the translation to human PK/PD and for the prediction of dose/regimen had good communication with other research & development (R&D) departments (e.g. pharmacology/clinical pharmacology). The increase in resources for preclinical PK/PD analysis including education was highly demanded. As a future perspective, the closer collaboration between pharmacokinetics scientists, pharmacologists, toxicologists and clinical pharmacologists and the increase in resources including upskilling and the comprehension of preclinical PK/PD analysis by the project team are considered to lead to efficient contributions to improve the success ratio of drug discovery and development.