Journal of Pharmacokinetics and Pharmacodynamics最新文献

Non-Small Cell Lung Cancer (NSCLC) remains one of the main causes of cancer death worldwide. In the urge of finding an effective approach to treat cancer, enormous therapeutic targets and treatment combinations are explored in clinical studies, which are not only costly, suffer from a shortage of participants, but also unable to explore all prospective therapeutic solutions. Within the evolving therapeutic landscape, the combined use of radiotherapy (RT) and checkpoint inhibitors (ICIs) emerged as a promising avenue. Exploiting the power of quantitative system pharmacology (QSP), we undertook a study to anticipate the therapeutic outcomes of these interventions, aiming to address the limitations of clinical trials. After enhancing a pre-existing QSP platform and accurately replicating clinical data outcomes, we conducted an in-depth study, examining different treatment protocols with nivolumab and RT, both as monotherapy and in combination, by assessing their efficacy through clinical endpoints, namely time to progression (TTP) and duration of response (DOR). As result, the synergy of combined protocols showcased enhanced TTP and extended DOR, suggesting dual advantages of extended response and slowed disease progression with certain combined regimens. Through the lens of QSP modeling, our findings highlight the potential to fine-tune combination therapies for NSCLC, thereby providing pivotal insights for tailoring patient-centric therapeutic interventions.

Conventional pharmacokinetic (PK) models contain several useful inductive biases guiding model convergence to more realistic predictions of drug concentrations. Implementing similar biases in standard neural networks can be challenging, but might be fundamental for model robustness and predictive performance. In this study, we build on the deep compartment model (DCM) architecture by introducing constraints that guide the model to explore more physiologically realistic solutions. Using a simulation study, we show that constraints improve robustness in sparse data settings. Additionally, predicted concentration-time curves took on more realistic shapes compared to unconstrained models. Next, we propose the use of multi-branch networks, where each covariate can be connected to specific PK parameters, to reduce the propensity of models to learn spurious effects. Another benefit of this architecture is that covariate effects are isolated, enabling model interpretability through the visualization of learned functions. We show that all models were sensitive to learning false effects when trained in the presence of unimportant covariates, indicating the importance of selecting an appropriate set of covariates to link to the PK parameters. Finally, we compared the predictive performance of the constrained models to previous relevant population PK models on a real-world data set of 69 haemophilia A patients. Here, constrained models obtained higher accuracy compared to the standard DCM, with the multi-branch network outperforming previous PK models. We conclude that physiological-based constraints can improve model robustness. We describe an interpretable architecture which aids model trust, which will be key for the adoption of machine learning-based models in clinical practice.

The new adjuvant chemotherapy of docetaxel, epirubicin, and cyclophosphamide has been recommended for treating breast cancer. It is necessary to investigate the potential drug-drug Interactions (DDIs) since they have a narrow therapeutic window in which slight differences in exposure might result in significant differences in treatment efficacy and tolerability. To guide clinical rational drug use, this study aimed to evaluate the DDI potentials of docetaxel, cyclophosphamide, and epirubicin in cancer patients using physiologically based pharmacokinetic (PBPK) models. The GastroPlus™ was used to develop the PBPK models, which were refined and validated with observed data. The established PBPK models accurately described the pharmacokinetics (PKs) of three drugs in cancer patients, and the predicted-to-observed ratios of all the PK parameters met the acceptance criterion. The PBPK model predicted no significant changes in plasma concentrations of these drugs during co-administration, which was consistent with the observed clinical phenomenon. Besides, the verified PBPK models were then used to predict the effect of other Cytochrome P450 3A4 (CYP3A4) inhibitors/inducers on these drug exposures. In the DDI simulation, strong CYP3A4 modulators changed the exposure of three drugs by 0.71-1.61 fold. Therefore, patients receiving these drugs in combination with strong CYP3A4 inhibitors should be monitored regularly to prevent adverse reactions. Furthermore, co-administration of docetaxel, cyclophosphamide, or epirubicin with strong CYP3A4 inducers should be avoided. In conclusion, the PBPK models can be used to further investigate the DDI potential of each drug and to develop dosage recommendations for concurrent usage by additional perpetrators or victims.

Pharmacokinetic modeling of monoclonal antibodies (mAbs) with non-linear binding is based on equations of the target-mediated drug disposition (Mager and Jusko, J Pharmacokinet Pharmacodyn 28:507-532, 2001). These equations demonstrated their utility in countless examples and drug development programs. The model assumes that the mAb drug and the target have only one binding site each while, in reality, most antibodies have two binding sites. Thus, the currently used model does not correspond to the biological process that it aims to describe. The correct mechanistic model should take into account both binding sites. We investigated, using simulations, whether this discrepancy is important and when it is advisable to use a model with correct stoichiometric 2-to-1 ratio. We show that for soluble targets when elimination rate of the drug-target complex is comparable with the elimination rate of the drug or lower, and when measurements of both total drug and total target concentrations are available, the model with 1-to-1 (monovalent) binding cannot describe data simulated from the model with 2-to-1 (bivalent) binding. In these cases, models with correct stoichiometric assumptions may be necessary for an adequate description of the observed data. Also, a model with allosteric binding that encompasses both 2-to-1 and 1-to-1 binding models as particular cases was proposed and applied. It was shown to be identifiable given the detailed concentration data of total drug and total target.

Authors' Response to Letter to Editor from Hinpetch Daungsupawong and Viroj Wiwanitkit.

The development of optimized dosing regimens plays a crucial role in oncology drug development. This study focused on the population pharmacokinetic modelling and simulation of docetaxel, comparing the pharmacokinetic exposure of oral docetaxel plus encequidar (oDox + E) with the standard of care intravenous (IV) docetaxel regimen. The aim was to evaluate the feasibility of oDox + E as a potential alternative to IV docetaxel. The article demonstrates an approach which aligns with the FDA's Project Optimus which aims to improve oncology drug development through model informed drug development (MIDD). The key question answered by this study was whether a feasible regimen of oDox + E existed. The purpose of this question was to provide an early GO / NO-GO decision point to guide drug development and improve development efficiency.

Methods:  A stepwise approach was employed to develop a population pharmacokinetic model for total and unbound docetaxel plasma concentrations after IV docetaxel and oDox + E administration. Simulations were performed from the final model to assess the probability of target attainment (PTA) for different oDox + E dose regimens (including multiple dose regimens) in relation to IV docetaxel using AUC over effective concentration (AUCOEC) metric across a range of effective concentrations (EC). A Go / No-Go framework was defined-the first part of the framework assessed whether a feasible oDox + E regimen existed (i.e., a PTA ≥ 80%), and the second part defined the conditions to proceed with a Go decision.

Results:  The overall population pharmacokinetic model consisted of a 3-compartment model with linear elimination, constant bioavailability, constant binding mechanics, and a combined error model. Simulations revealed that single dose oDox + E regimens did not achieve a PTA greater than 80%. However, two- and three-dose regimens at 600 mg achieved PTAs exceeding 80% for certain EC levels.

Conclusion:  The study demonstrates the benefits of MIDD using oDox + E as a motivating example. A population pharmacokinetic model was developed for the total and unbound concentration in plasma of docetaxel after administration of IV docetaxel and oDox + E. The model was used to simulate oDox + E dose regimens which were compared to the current standard of care IV docetaxel regimen. A GO / NO-GO framework was applied to determine whether oDox + E should progress to the next phase of drug development and whether any conditions should apply. A two or three-dose regimen of oDox + E at 600 mg was able to achieve non-inferior pharmacokinetic exposure to current standard of care IV docetaxel in simulations. A Conditional GO decision was made based on this result and further quantification of the "effective concentration" would improve the ability to optimise the dose regimen.

This is a correspondence on "Evaluation of ChatGPT and Gemini large language models for pharmacometrics with NONMEM". Additional concern on using ChatGPT and Gemini is provided.

Dose personalization improves patient outcomes for many drugs with a narrow therapeutic index and high inter-individuality variability, including busulfan. Non-compartmental analysis (NCA) and model-based methods like maximum a posteriori Bayesian (MAP) approaches are two methods routinely used for dose optimization. These approaches vary in how they estimate patient-specific pharmacokinetic parameters to inform a dose and the impact of these differences is not well-understood. Using busulfan as an example application and area under the concentration-time curve (AUC) as a target exposure metric, these estimation methods were compared using retrospective patient data (N = 246) and simulated precision dosing treatment courses. NCA was performed with or without peak extension, and MAP Bayesian estimation was performed using either the one-compartment Shukla model or the two-compartment McCune model. All methods showed good agreement on real-world data (correlation coefficients of 0.945-0.998) as assessed by Bland-Altman plots, although agreement between NCA and MAP methods was higher during the first dosing interval (0.982-0.994) compared to subsequent dosing intervals (0.918-0.938). In dose adjustment simulations, both NCA and MAP estimated high target attainment (> 98%) although true simulated target attainment was lower for NCA (63-66%) versus MAP (91-93%). The largest differences in AUC estimation were due to different assumptions for the shape of the concentration curve during the infusion phase, followed by how the methods considered time-dependent clearance and concentration-time points collected in earlier intervals. In conclusion, although AUC estimates between the two methods showed good correlation, in a simulated study, MAP lead to higher target attainment. When changing from one method to another, or changing infusion duration and other factors, optimum estimated exposure targets may require adjusting to maintain a consistent exposure.

Aprocitentan is a novel, potent, dual endothelin receptor antagonist that recently demonstrated efficacy in the treatment of difficult-to-treat (resistant) hypertension. The aim of this study was to develop a population pharmacokinetic (PK) model describing aprocitentan plasma concentration over time, to investigate relationships between subject-specific factors (covariates) and model parameters, and to quantify the influence of the identified covariates on the exposure to aprocitentan via model-based simulations, enabling judgment about the clinical relevance of the covariates.PK data from 902 subjects in ten Phase 1, one Phase 2, and one Phase 3 study were pooled to develop a joint population PK model. The concentration-time course of aprocitentan was described by a two-compartment model with absorption lag time, first-order absorption and elimination, and reduced relative bioavailability following very high doses of 300 and 600 mg.The population PK model described the observed data well. Volume and clearance parameters were associated with body weight. Renal function as reflected by estimated glomerular filtration rate (eGFR), hepatic impairment, and sex were identified as relevant covariates on clearance.The subject-specific characteristics of body weight, eGFR, hepatic impairment, and sex were shown to influence exposure parameters area under the concentration-time curve and maximum concentration in steady state to a limited extent, i.e., not more than 25% different from a reference subject, and therefore do not warrant dose adjustments.

The use of β-lactam (BL) and β-lactamase inhibitor (BLI) combinations, such as piperacillin-tazobactam (PIP-TAZ) is an effective strategy to combat infections by extended-spectrum β-lactamase-producing bacteria. However, in Gram-negative bacteria, resistance (both mutational and adaptive) to BL-BLI combination can still develop through multiple mechanisms. These mechanisms may include increased β-lactamase activity, reduced drug influx, and increased drug efflux. Understanding the relative contribution of these mechanisms during resistance development helps identify the most impactful mechanism to target in designing a treatment to counter BL-BLI resistance. This study used semi-mechanistic mathematical modeling in combination with antibiotic sensitivity assays to assess the potential impact of different resistance mechanisms during the development of PIP-TAZ resistance in a Klebsiella pneumoniae isolate expressing CTX-M-15 and SHV-1 β-lactamases. The mathematical models were used to evaluate the potential impact of several cellular changes as a sole mediator of PIP-TAZ resistance. Our semi-mechanistic model identified 2 out of the 13 inspected mechanisms as key resistance mechanisms that may independently support the observed magnitude of PIP-TAZ resistance, namely porin loss and efflux pump up-regulation. Simulation using the resulting models also suggested the possible adjustment of PIP-TAZ dose outside its commonly used 8:1 dosing ratio. The current study demonstrated how theory-based mechanistic models informed by experimental data can be used to support hypothesis generation regarding potential resistance mechanisms, which may guide subsequent experimental studies.