Journal of Pharmacokinetics and Pharmacodynamics最新文献

Concentration-QTc (C-QTc) analysis is a model-based method widely used to assess the impact of drugs on QT interval duration. C-QTc modelling was enabled to be used after the publication of the International Council for Harmonisation (ICH) E14 Questions and Answers guidance document in 2015, followed by the Scientific White Paper on C-QTc modelling (Garnett et al. J Pharmacokinet Pharmacodyn 45(3):383-397 2018), which included technical details and recommendations on how to perform and report the modelling. This hands-on tutorial aims to provide a practical implementation of the recommended C-QTc modelling methodology, including R code to perform the complete analysis, from data formatting to model predictions. The target audience is scientists who will perform C-QTc analyses. The tutorial uses real data from a previously published QT study by (Johannesen et al.Clin Pharmacol Ther 96(5):549-558 2014), focusing on two active treatments (dofetilide and verapamil) and placebo to illustrate positive and negative QT signals. The methodology implemented in this tutorial follows the recommendations outlined in the White paper. This tutorial includes practical steps for preparing an analysis-ready dataset, conducting exploratory data analysis, fitting the linear mixed effects (LME) model, assessing model performance and estimating the upper limit of the two-sided 90% confidence interval (CI) of baseline and placebo-corrected QTc (ΔΔQTc). Reproducibility of this workflow is ensured through the use of pkgr to manage R packages. The R codes provided as part of this tutorial were successfully used for several projects within the AstraZeneca portfolio and accepted by health authorities as part of QTc submissions.

Erythropoiesis is a complex process that results in the production of erythrocytes from hematopoietic stem cells in the bone marrow. This work aimed to develop a population pharmacokinetic-pharmacodynamic (PKPD) model describing erythropoiesis and hemoglobin synthesis following bitopertin, an inhibitor of glycine transporter 1 (GlyT1), administration. Data from a Phase 1 clinical trial in 67 healthy subjects administered bitopertin (10, 30, or 60 mg) or placebo for 120 days were analyzed. Hematological assessments included erythrocyte and reticulocyte counts, immature reticulocyte fraction, hemoglobin concentration, and mean corpuscular hemoglobin. The proposed semi-mechanistic model, which leverages data and physiological knowledge, was found to adequately simultaneously describe the dose- and time-dependent changes in the biomarkers. The framework was used to illustrate the potential outcome of hypothetical drug-target interactions at distinct stages of erythropoiesis and hemoglobin synthesis, exemplifying its usefulness in a clinical setting.

Autosomal dominant tubulointerstitial kidney disease (ADTKD), caused by mutations in UMOD and MUC1 genes, leads to tubular damage and fibrosis, ultimately resulting in kidney failure (KF). This study investigated clinical and genetic factors influencing the rate and severity of ADTKD progression by developing quantitative models. An estimated glomerular filtration rate (eGFR) of 10 mL/min/1.73 m² was used to define KF, corresponding to dialysis initiation. Natural history data from the Wake Forest University School of Medicine study were used to develop the models for UMOD (n = 371) and MUC1 (n = 233) disease types (age ≥ 18 years). Longitudinal change in eGFR and time-to-KF were quantified using nonlinear mixed-effects and parametric time-to-event modeling approaches, respectively, in Monolix (version 2024R1). Sigmoid I_max functions with steepness parameters varying before and after inflection points best captured eGFR decline. Patients with UMOD and MUC1 disease variants exhibited a similar initial shallow steepness ( $\approx$ 1), but after inflection, each declined rapidly. MUC1 patients progressed faster than UMOD during the post-inflection phase (γ₂ = 10.23 vs. 6.34). eGFR at first clinic visit (eGFR_FCV) and age at first clinic visit (AFCV) significantly affected between-subject variability in eGFR decline. A Weibull hazard function best described the time to KF. In UMOD, males reached Te (the age at which approximately 36.8% of individuals remain free from KF) 4 years earlier than females on average (β_Te_Male = -0.07), indicating faster progression in males. Older AFCV was associated with slower progression to KF (β_Te_AFCV = 0.59 for UMOD and 0.81 for MUC1). These models may help enable quantitative data-driven subgroup analysis in the future, optimizing inclusion/exclusion criteria for ADTKD clinical trials.

Drug discovery balances many factors as it identifies compounds for clinical testing, including compound efficacy, safety, pharmacokinetic (PK) properties, commercial feasibility, competitive positioning, and organizational pressures to move quickly with limited knowledge. When considering target engagement within clinically acceptable dosing constraints, design elements often balance potency requirements against the required extent of target engagement, which subsequently inform the PK design criteria (e.g. absorption and half-life considerations). Hence, an early understanding of the magnitude and duration of target engagement can focus design teams by providing well defined design criteria. To this end, an in vitro target engagement assay has been developed to bin targets and compounds by the type of target engagement profile required for efficacy (cellular anti-proliferation). This in turn directionally informs on the required concentration profile most aligned with the efficacy readout, bucketing results into three primary categories that drive efficacy: high transient concentrations, average concentrations, and threshold concentrations. This manuscript will outline the methodology developed for this early target coverage assessment and provide examples with selected compounds spanning molecularly targeted and cytotoxic oncology small molecules.

Posaconazole is an effective broad-spectrum triazole antifungal used as prophylaxis or to treat invasive Aspergillus and Candida infections in adults and pediatric patients. Posaconazole is a known strong inhibitor of cytochrome P4503A4 (CYP3A4) and substrate of P-glycoprotein (P-gp), which may lead to drug-drug interactions (DDIs) when co-administered with CYP3A4-sensitive substrates and warrants modified dosing of sensitive drugs when administered concomitantly with posaconazole. Given the long elimination half-life of posaconazole (26-35 h), there is the potential for DDIs caused by posaconazole after discontinuing the antifungal. Our clinical studies revealed that the half-life of posaconazole is significantly prolonged in subjects with a body mass index (BMI) ≥ 35 kg/m², which may put this population at an increased risk of DDIs after stopping posaconazole. This manuscript describes the development, verification, and validation of a whole-body, physiologically-based pharmacokinetic (PBPK) model which describes the concomitant use and washout DDIs of posaconazole delayed-release tablet (DRT) with victim drugs ranolazine and lurasidone in healthy volunteers of normal weight and with obesity. The key findings of this model are 1) the half-life of posaconazole is significantly prolonged in patients with BMI ≥ 35 kg/m² and 2) the mechanism of inhibition of CYP3A4 by posaconazole appears to be irreversible in vivo. This model may be used moving forward to assess the potential for washout DDIs with CYP3A4-sensitive substrates during concomitant use with, and after discontinuing posaconazole in subjects with normal weight and obesity.

This work focuses on design of experiments for Pharmacokinetic (PK) and Pharmacodynamic (PD) studies. Non-Linear Mixed Effects Models (NLMEM) modelling allows the identification and quantification of covariates that explain inter-individual variability (IIV). The Fisher Information Matrix (FIM), computed by linearization, has already been used to predict uncertainty on covariate parameters and power of test to detect statistical significance. A covariate effect is deemed statistically significant if it is different from 0 according to a Wald comparison test and clinically relevant if the ratio of change it causes in the parameter is relevant according to a test inspired by the two one-sided tests (TOST) as in bioequivalence studies. FIM calculation was extended by computing its expectation on the joint distribution of the covariates, discrete and continuous. Three methods were proposed: using a provided sample of covariate vectors, simulating covariate vectors, based on provided independent distributions or on estimated copulas. Thereafter, CI of ratios, power of tests and number of subjects needed to achieve desired confidence were derived. Methods were implemented in a working version of the R package PFIM6.1. A simulation study was conducted under various scenarios, including different sample sizes, sampling points, and IIV. Overall, uncertainty on covariate effects and power of tests were accurately predicted. The method was applied to a population PK model of the drug cabozantinib including 27 covariate relationships. Despite numerous relationships, limited representation of certain covariates, FIM correctly predicted uncertainty, and is therefore suitable for rapidly computing number of subjects needed to achieve given powers.

The questions and answers (Q&A) document for ICH E14/S7B provides the following advancements for QTc assessment: concentration-QTc modeling (C-QTc) as the primary analysis, accepting alternative approaches (Q&A 5.1 and 6.1) to thorough QT (TQT) studies, and incorporating an integrated nonclinical risk assessment as supporting evidence. Based on QT study reports reviewed by the FDA between 2016 and 2024, changes to the E14 guideline have resulted in a 34% decrease in the proportion of TQT studies, while the use of C-QTc analysis as the primary analysis has significantly increased. Studies using C-QTc instead of by-time analysis as the primary analysis reduced median sample sizes by 67%, 42%, and 35% for parallel, nested crossover, and crossover studies, respectively. The white paper C-QTc model was used for 60% of drugs that prolonged the QTc interval. From 2020 to 2024, reviews incorporating an integrated nonclinical risk assessment have also increased. The advancements in QTc assessments have streamlined QTc assessment and made clinical trials less resource-intensive. As the advancements continue to evolve the drug safety evaluation is likely to become even more adaptive and enable more precise and targeted QTc assessment.

Quantitative Systems Pharmacology (QSP) has emerged as a cornerstone of modern drug development, providing a robust framework to integrate data from preclinical and clinical studies, enhance decision-making, and optimize therapeutic strategies. By modeling biological systems and drug interactions, QSP enables predictions of outcomes, optimization of dosing regimens, and personalized medicine applications. Recent advancements in artificial intelligence (AI) and machine learning (ML) hold the potential to significantly transform QSP by enabling enhanced data extraction, fostering the development of hybrid mechanistic ML models, and supporting the introduction of surrogate models and digital twins. This manuscript explores the transformative role of AI and ML in reshaping QSP modeling workflows. AI/ML tools now enable automated literature mining, the generation of dynamic models from data, and the creation of hybrid frameworks that blend mechanistic insights with data-driven approaches. Large Language Models (LLMs) further revolutionize the field by transitioning AI/ML from merely a tool to becoming an active partner in QSP modeling. By facilitating interdisciplinary collaboration, lowering barriers to entry, and democratizing QSP workflows, LLMs empower researchers without deep coding expertise to engage in complex modeling tasks. Additionally, the integration of Artificial General Intelligence (AGI) holds the potential to autonomously propose, refine, and validate models, further accelerating innovation across multiscale biological processes. Key challenges remain in integrating AI/ML into QSP workflows, particularly in ensuring rigorous validation pipelines, addressing ethical considerations, and establishing robust regulatory frameworks to address the reliability and reproducibility of AI-assisted models. Moreover, the complexity of multiscale biological integration, effective data management, and fostering interdisciplinary collaboration present ongoing hurdles. Despite these challenges, the potential of AI/ML to enhance hybrid model development, improve model interpretability, and democratize QSP modeling offers an exciting opportunity to revolutionize drug development and therapeutic innovation. This work highlights a pathway toward a transformative era for QSP, leveraging advancements in AI and ML to address these challenges and drive innovation in the field.

Oligonucleotide therapeutics hold promise for targeted gene silencing, yet achieving optimal tissue-specific delivery remains challenging. This study introduces a mechanistic whole-body physiologically based pharmacokinetic (PBPK) model to predict tissue uptake dynamics of both conjugated (targeted) and unconjugated oligonucleotides across species. The model incorporates two uptake pathways: a non-saturable nonspecific pathway for all oligonucleotides and receptor-mediated endocytosis (RME) specific to conjugated molecules. Parameters for nonspecific uptake were derived from plasma and tissue concentration data of unconjugated antisense oligonucleotides (ASOs) in rats, while RME parameters for N-acetylgalactosamine (GalNAc)-conjugated oligonucleotides targeting the asialoglycoprotein receptor (ASGPR) were obtained from literature. Model validation against experimental data for conjugated and unconjugated ASOs and small interfering RNAs (siRNAs) in rats and mice demonstrated good predictive performance, with median predicted-to-observed AUC ratios of 0.84 (Interquartile range [IQR] 0.434-1.22) in rats and 0.629 (IQR 0.3-1.6) in mice. Local sensitivity analyses identified key parameters and processes influencing organ uptake, including the unbound plasma fraction and receptor-mediated uptake efficiency. Simulations highlighted the potential of sustained-release formulations to improve targeting specificity by mitigating receptor saturation. This is the first whole-body PBPK model to describe oligonucleotide pharmacokinetics across species and modalities. The model provides critical mechanistic insights to optimize tissue-specific delivery, guide formulation strategies, and enhance therapeutic outcomes for targeted oligonucleotide therapeutics.

Advancements in large language models (LLMs) have suggested their potential utility for diverse pharmacometrics tasks. This study investigated the performance of LLM for generating structure diagrams, publication-ready tables, analysis reports, and conducting simulations using output files from pharmacometrics models. Forty-four NONMEM output files were obtained from the GitHub software repository. The performance of Claude 3.5 Sonnet (Claude) and ChatGPT 4o was compared with two other candidate LLMs: Gemini 1.5 Pro and Llama 3.2. Prompt engineering was conducted for Claude for pharmacometrics tasks such as generating model structure diagrams, parameter tables, and analysis reports. Simulations were conducted using ChatGPT. Claude Artifacts was used to visualize model structure diagrams, parameter tables, and analysis reports. A web-based R Shiny application was implemented to provide an accessible interface for automating pharmacometric model structure diagrams, parameter tables, and analysis reports tasks. Claude was selected for investigation following performance comparisons with ChatGPT 4o, Gemini 1.5 Pro, and Llama on model structure diagram and parameter table generation tasks. Claude successfully generated the model structure diagrams for 40 (90.9%) of the 44 NONMEM output files with the initial prompts, and the remaining were resolved with an additional prompt. Claude consistently generated accurate parameter summary tables and succinct model analysis reports. Modest variability in model structure diagrams generated for replicate prompts was identified. ChatGPT demonstrated simulation capabilities but revealed limitations with complex PK/PD models. LLMs have the potential to enhance key pharmacometrics modeling tasks. However, expert review of the results generated is essential.