CPT: Pharmacometrics & Systems Pharmacology最新文献

The use of synthetic data in pharmacology research has gained significant attention due to its potential to address privacy concerns and promote open science. In this study, we implemented and compared three synthetic data generation methods, CT-GAN, TVAE, and a simplified implementation of Avatar, for a previously published pharmacogenetic dataset of 253 patients with one measurement per patient (non-longitudinal). The aim of this study was to evaluate the performance of these methods in terms of data utility and privacy trade off. Our results showed that CT-GAN and Avatar used with k = 10 (number of patients used to create the local model of generation) had the best overall performance in terms of data utility and privacy preservation. However, the TVAE method showed a relatively lower level of performance in these aspects. In terms of Hazard ratio estimation, Avatar with k = 10 produced HR estimates closest to the original data, whereas CT-GAN slightly underestimated the HR and TVAE showed the most significant deviation from the original HR. We also investigated the effect of applying the algorithms multiple times to improve results stability in terms of HR estimation. Our findings suggested that this approach could be beneficial, especially in the case of small datasets, to achieve more reliable and robust results. In conclusion, our study provides valuable insights into the performance of CT-GAN, TVAE, and Avatar methods for synthetic data generation in pharmacogenetic research. The application to other type of data and analyses (data driven) used in pharmacology should be further investigated.

Quantitative systems toxicology (QST) models are increasingly being applied for predicting and understanding toxicity liabilities in pharmaceutical research and development. A European Federation of Pharmaceutical Industries and Associations (EFPIA)-wide survey was completed by 15 companies. The results provide insights into the current use of QST models across the industry. 73% of responding companies with more than 10,000 employees utilize QST models. The most applied QST models are for liver, cardiac electrophysiology, and bone marrow/hematology. Responders indicated particular interest in QST models for the central nervous system (CNS), kidney, lung, and skin. QST models are used to support decisions in both preclinical and clinical stages of pharmaceutical development. The survey suggests high demand for QST models and resource limitations were indicated as a common obstacle to broader use and impact. Increased investment in QST resources and training may accelerate application and impact. Case studies of QST model use in decision-making within EFPIA companies are also discussed. This article aims to (i) share industry experience and learnings from applying QST models to inform decision-making in drug discovery and development programs, and (ii) share approaches taken during QST model development and validation and compare these with recommendations for modeling best practices and frameworks proposed in the literature. Discussion of QST-specific applications in relation to these modeling frameworks is relevant in the context of the recently proposed International Council for Harmonization (ICH) M15 guideline on general principles for Model-Informed Drug Development (MIDD).

Empirical maturation models (e.g., Johnson and Rhodin models) for glomerular filtration rate (GFR) are commonly used as scaling factors for predicting pediatric renal clearance, but their predictive performance for drugs featured with tubular reabsorption is poorly understood. This study investigated the adequacy of GFR-based scaling models for predicting pediatric renal clearance in drugs with passive tubular reabsorption by comparing with a mechanistic kidney model (Mech-KiM) that encompasses the physiological processes of glomerular filtration, tubular secretion, and reabsorption. The analysis utilized hypothetical drugs with varying fractions of tubular reabsorption (F_reabs), alongside the model drug metronidazole, which has a F_reabs of 96%. Our simulations showed that when F_reabs is ≤70%, the discrepancies between the GFR-based scaling methods and the Mech-KiM model in predicting pediatric renal clearance were generally within a twofold range throughout childhood. However, for drugs with substantial tubular reabsorption (e.g., F_reabs > 70%), discrepancies greater than twofold were observed between the GFR-based scaling methods and the Mech-KiM model in predicting renal clearance for young children. In neonates, the differences ranged from 5- to 10-fold when the adult F_reabs was 95%. Pediatric physiologically based pharmacokinetic (PBPK) modeling of metronidazole revealed that using a GFR-based scaling method (Johnson model) significantly overestimated drug concentrations in children under 2 months, whereas utilizing the Mech-KiM model for renal clearance predictions yielded estimates closely aligned with observed concentrations. Our study demonstrates that using GFR-based scaling models to predict pediatric renal clearance might be inadequate for drugs with extensive passive tubular reabsorption (e.g., F_reabs > 70%).

Opioid use disorders (OUD) are a major issue in the U.S. Current treatments for pregnant women, like methadone and buprenorphine require daily dosing and have adverse effects. Monthly injectable naltrexone (XR-NTX) mitigates these adverse effects but is not recommended during pregnancy due to limited pharmacokinetic and safety data. This study developed a physiologically based pharmacokinetic (PBPK) model to describe XR-NTX pharmacokinetics during pregnancy, and to predict dosing recommendations. Model predictions were successfully validated with observed data. Maternal plasma XR-NTX profiles were simulated for 400 non-pregnant virtual females at the approved dose of 380 mg, then randomized to continue with either 380, 285, 190, or 95 mg during pregnancy. The non-pregnant virtual females had a mean predicted C_max, AUC_0-7days, and AUC_0-28days of 23.3 ng/mL, 142 ng·d/mL, and 148 ng·d/mL, respectively. Maternal XR-NTX exposure (AUC_0-28days) were predicted to increase by 1.37, 1.43, and 1.72 times during the first, second, and third trimester of pregnancy. However, the fetal-to-maternal exposure (AUC_0-28days) was lower in the first (15%), second (7%), and third (9%) trimesters. A dose of 285 mg of XR-NTX in pregnancy during the first/second trimester and dose of 190 mg in the third trimester were predicted to provide maternal exposures that were comparable to non-pregnant levels at the standard dose. This study provides crucial insights into XR-NTX pharmacokinetics and proposes a dosing strategy during pregnancy, potentially aiding further clinical investigations and decision making regarding XR-NTX use during pregnancy.

Immunogenicity is the propensity of a therapeutic protein to generate an immune response to itself. While reporting of antidrug antibodies (ADAs) is increasing, model-based analysis of such data is seldom performed. Model-based characterization of factors affecting the emergence and dissipation of ADAs may inform drug development and/or improve understanding in clinical practice. This analysis aimed to predict ADA dynamics, including the potential influence of individual covariates, following subcutaneous satralizumab administration. Satralizumab is a humanized IgG2 monoclonal recycling IL-6 receptor antagonist antibody approved for treating neuromyelitis optica spectrum disorder (NMOSD). Longitudinal pharmacokinetic (PK) and ADA data from 154 NMOSD patients in two pivotal Phase 3 studies (NCT02028884, NCT02073279) and PK data from one Phase 1 study (SA-001JP) in 72 healthy volunteers were available for this analysis. An existing population PK model was adapted to derive steady-state concentration without ADA for each patient. A mixed hidden Markov model (mHMM) was developed whereby three different states were identified: one absorbing Markov state for non-ADA developer, and two dynamic and inter-connected Markov states—transient ADA negative and positive. Satralizumab exposure and body mass index impacted transition probabilities and, therefore, the likelihood of developing ADAs. In conclusion, the mHMM model was able to describe the time course of ADA development and identify patterns of ADA development in NMOSD patients following treatment with satralizumab, which may allow for the formulation of strategies to reduce the emergence or limit the impact of ADA in the clinical setting.

Quantitative model-based clinical trial simulation tools play a critical role in informing study designs through simulation before actual execution. These tools help drug developers explore various trial scenarios in silico to select a clinical trial design to detect therapeutic effects more efficiently, therefore reducing time, expense, and participants' burden. To increase the usability of the tools, user-friendly and interactive platforms should be developed to navigate various simulation scenarios. However, developing such tools challenges researchers, requiring expertise in modeling and interface development. This tutorial aims to address this gap by guiding developers in creating tailored R Shiny apps, using an example of a model-based clinical trial simulation tool that we developed for Duchenne muscular dystrophy (DMD). In this tutorial, the structural framework, essential controllers, and visualization techniques for analysis are described, along with key code examples such as criteria selection and power calculation. A virtual population was created using a machine learning algorithm to enlarge the available sample size to simulate clinical trial scenarios in the presented tool. In addition, external validation of the simulated outputs was conducted using a placebo arm of a recently published DMD trial. This tutorial will be particularly useful for developing clinical trial simulation tools based on DMD progression models for other end points and biomarkers. The presented strategies can also be applied to other diseases.

Type I interferons contribute to the pathogenesis of several autoimmune disorders, including systemic lupus erythematosus (SLE), systemic sclerosis, cutaneous lupus erythematosus, and myositis. Anifrolumab is a monoclonal antibody that binds to subunit 1 of the type I interferon receptor (IFNAR1). Results of phase IIb and phase III trials led to the approval of intravenous anifrolumab 300 mg every 4 weeks (Q4W) alongside standard therapy in patients with moderate-to-severe SLE. Here, we built a population physiology-based pharmacokinetic (PBPK) model of anifrolumab by utilizing the physiochemical properties of anifrolumab, binding kinetics to the Fc gamma neonatal receptor, and target-mediated drug disposition properties. A novel relative transcriptomics approach was employed to determine IFNAR1 expression in tissues (blood, skin, gastrointestinal tract, lungs, and muscle) using mRNA abundances from bioinformatic databases. The IFNAR1 expression and PBPK model were validated by testing their ability to predict clinical pharmacokinetics over a large dose range from different clinical scenarios after subcutaneous and intravenous anifrolumab dosing. The validated PBPK model predicted high unbound local concentrations of anifrolumab in blood, skin, gastrointestinal tract, lungs, and muscle, which exceeded its IFNAR1 dissociation equilibrium constant values. The model also predicted high IFNAR1 occupancy with subcutaneous and intravenous anifrolumab dosing. The model predicted more sustained IFNAR1 occupancy ≥90% with subcutaneous anifrolumab 120 mg once-weekly dosing vs. intravenous 300 mg Q4W dosing. The results informed the dosing of phase III studies of anifrolumab in new indications and present a novel approach to PBPK modeling coupled with relative transcriptomics in simulating pharmacokinetics of therapeutic monoclonal antibodies.

Pharmacokinetic studies are important for understanding drug disposition in the human body. However, pregnant and lactating women are often excluded from primary pharmacokinetic studies and as such there is often limited dosing information regarding drug use in pregnant and/or lactating women. The objectives of this interim analysis were to define the transfer of rifampicin to a breastfed infant and to determine the area under the concentration–time curve of rifampicin in maternal plasma, breastmilk and infant plasma. Performing this interim analysis enabled us to substantiate whether prior assumptions we made on several study design issues including patient sample size and pharmacokinetic sampling times held and whether we needed to amend our protocol or not. We enrolled lactating mothers on treatment for tuberculosis with their breastfeeding infants (below 12 months of age), performed intensive pharmacokinetic sampling (0–24 h post-dose) on plasma samples from both the mother, infant(s) and breastmilk samples from the mother on two separate occasions (once during the initiation phase and another during the continuation phase of tuberculosis treatment). The initial study design, including sampling times, was informed by a stochastic simulation and estimation exercise, with very limited prior breastmilk data. An interim analysis after recruiting 6 mother–infant pairs ascertained that our initial assumptions were ideal for achieving our study objectives and no amendments to the sampling times were necessary. Initial data from 6 mother–infant pairs show that rifampicin penetrates breastmilk with an approximate milk-to-plasma ratio of 0.169 and 0.189 on two separate visits. However, it was undetectable in most infants.

Gastrointestinal first-pass metabolism plays an important role in bioavailability and in drug–drug interactions. Physiologically-based pharmacokinetic (PBPK) modeling is a powerful tool to integrate these processes mechanistically. However, a correct bottom-up prediction of GI first-pass metabolism is challenging and depends on various model parameters like the level of enzyme expression and the basolateral intestinal mucosa permeability (P_mucosa). This work aimed to investigate if cytochrome P450 (CYP) 3A4 expression could help predict the first-pass effect using PBPK modeling or whether additional factors like P_mucosa do play additional roles using PBPK modeling. To this end, a systematic review of the absolute CYP3A expression in the human gastrointestinal tract and liver was conducted. The resulting CYP3A4 expression profile and two previously published profiles were applied to PBPK models of seven CYP3A4 substrates (alfentanil, alprazolam, felodipine, midazolam, sildenafil, triazolam, and verapamil) built-in PK-Sim®. For each compound, it was assessed whether first-pass metabolism could be adequately predicted based on the integrated CYP3A4 expression profile alone or whether an optimization of P_mucosa was required. Evaluation criteria were the precision of the predicted interstudy bioavailabilities and area under the concentration–time curves. It was found that none of the expression profiles provided upfront an adequate description of the extent of GI metabolism and that optimization of P_mucosa as a compound-specific parameter improved the prediction of most models. Our findings indicate that a pure bottom-up prediction of gastrointestinal first-pass metabolism is currently not possible and that compound-specific features like P_mucosa must be considered as well.