CPT: Pharmacometrics & Systems Pharmacology最新文献

Fludarabine (Flu), administered as a prodrug Flu monophosphate, is a lymphodepleting agent used prior to hematopoietic stem cell transplantation (HSCT) which exhibits substantial pharmacokinetics (PK) variability, contributing to suboptimal outcomes. This study developed and validated a physiologically based pharmacokinetic (PBPK) model using literature-based data and a middle-out approach for Flu and its two main metabolites in adults and children, and evaluated its performance to predict individual Flu exposures in 28 pediatric HSCT patients using a virtual twin (VT) approach using PK-Sim software. Different informed models with individual demographic and biological characteristics were assessed by comparing predicted and observed plasma exposures (AUC_0➔24h) via fold error metrics and regression analyses. The PBPK model accurately reproduced observed Flu and metabolites concentrations in both adults and children. In the VT cohort, informing the model with plasma protein scaling factor and nuclear GFR improved drug exposure predictions in total (mean fold error (MFE) = 0.91) and unbound (MFE = 0.88) compartments with minimal bias in both Deming and Bland–Altman analyses. This model provided superior agreement between observed and predicted exposures, achieving improved agreement across statistical and regression approaches compared to the model with estimated GFR. PBPK-based VT modeling enables accurate, individualized prediction of Flu PK in pediatric HSCT patients. These results support the implementation of model-informed precision dosing to achieve personalized pediatric dosing of Flu in HSCT.

Long-acting injectables (LAI) are of increasing interest as they facilitate improved medication adherence and exposure, with target plasma concentration levels maintained over weeks/months. Biopredictive in vitro dissolution tests can aid formulation development of LAIs and guide quality control dissolution testing by facilitating accelerated test development. However, it is not easy to develop such tests when mechanisms underlying in vivo dissolution are not fully understood. The question of interest (QOI) and context of use (COU) of this study involve quantifying the impact of in vivo parameters which are critical bioavailability attributes (CBAs), using physiologically based pharmacokinetic (PBPK) models generated for LAI methylprednisolone acetate. Simulated dissolution profiles from the PBPK models can provide a design space for biopredictive in vitro dissolution testing methods. The five CBAs explored in this study were particle size, solubility, diffusion layer thickness, diffusion coefficient, and depot volume. Although the best performing models displayed good predictive ability, they used different (literature/prediction derived) attribute values. Simulated in vivo dissolution profiles generated suggested much slower dissolution rates, with 80–100% dissolved after 1200 h, than in vitro dissolution tests from FDA ‘Dissolution Methods Database,’ where almost 90% was dissolved in 90 min. To conclude, in vitro dissolution conditions resulting in larger effective particle sizes and diffusion layer thickness, suggesting low fluid velocities, need to be explored to generate biopredictive dissolution profiles. The current approach illustrates how using models with plausible CBA value ranges can be used to simulate a target dissolution profile design space, assisting in vitro LAI dissolution test development.

IPSOS (NCT03191786) is a Phase III trial comparing atezolizumab (atezo) monotherapy to single-agent chemotherapy (gemcitabine or vinorelbine) in patients with treatment-naïve locally advanced or metastatic NSCLC unsuitable for platinum-doublet chemotherapy. The study demonstrated significant overall survival (OS) improvement in the atezo arm compared to single-agent chemotherapy, with a stratified hazard ratio (HR) of 0.78 (95% CI: 0.63–0.97; p = 0.028). Since the IPSOS control arm only allowed gemcitabine or vinorelbine, a model-based meta-analysis (MBMA) was conducted, extracting OS data from published literature in similar patients, adjusting for population differences across trials, to estimate the HR between IPSOS arms versus historical trials which utilized single-agent chemotherapies. The aim was to demonstrate the non-inferiority of the IPSOS control arm versus historical controls. The literature search included patients who were chemotherapy-naïve, had advanced or metastatic NSCLC, were platinum-ineligible, ≥ 70 years or had ECOG ≥ 2, and were treated with single-agent paclitaxel, docetaxel, gemcitabine, pemetrexed, or vinorelbine. Summary-level OS data were extracted by digitizing Kaplan-Meier curves, resulting in a database of 26 trials with 41 arms and 3637 participants. A nonparametric approach modeling the conditional probability of OS data was implemented. After adjusting for ECOG PS (the only significant covariate), the model-predicted HR for the IPSOS control arm relative to historical trials was 0.543 (95% CI: 0.435–0.677), and the HR for the IPSOS atezo monotherapy arm was 0.418 (95% CI: 0.335–0.522). Overall, the MBMA results support the benefit of atezo seen in the IPSOS trial.

N.-A. T. Vu, Y. M. Song, S. K. Kim, et al., “Beyond the Michaelis–Menten: Evaluation of a tQSSA-Based IVIVE Approach for Predicting In Vivo Intrinsic Clearance From Hepatocyte Assays,” CPT: Pharmacometrics & Systems Pharmacology 15, no. 2 (2026): e70169, https://doi.org/10.1002/psp4.70169.

In the article cited above, the authors inadvertently omitted the Acknowledgments section, which is shown below.

The authors regret this error.

Global efforts to eliminate onchocerciasis are hampered by the lack of a macrofilaricidal drug capable of killing adult parasites. Oxfendazole, a veterinary anthelminthic, exhibits macrofilaricidal activity and holds promise to shorten treatment durations. Phase 1 studies in healthy Caucasian adults demonstrated favorable pharmacokinetics and safety using a veterinary oral liquid formulation. More recently, a Phase 1 bioavailability trial (NCT04920292) evaluated a field-applicable tablet formulation in healthy African adults. This study presents a secondary analysis to (i) characterize the population pharmacokinetics of oxfendazole and its major metabolites in healthy African adults receiving the tablet formulation and (ii) propose a dosing regimen for Phase 2 evaluation in patients with onchocerciasis. Thirty healthy African adults were enrolled, and plasma concentration–time profiles of oxfendazole, fenbendazole, and oxfendazole sulfone were obtained from 24 participants who received oxfendazole (8 per dose group: 100 mg single dose, 400 mg single dose, 400 mg once daily for 5 days). All cohorts were pooled and analyzed using nonlinear mixed effects modeling. Oxfendazole absorption was best described by first-order kinetics with first-pass metabolism. Dose-limited bioavailability was evident. Disposition was best described by one-compartment models with linear elimination. Simulations suggested that 400 mg once daily (or 50 mg twice daily) for 5 days is required to achieve putative exposure targets (> 200 ng/mL for 5 days), with low risk of safety concerns. The population pharmacokinetic model adequately described oxfendazole pharmacokinetics in healthy African adults and supports dosing selection for future clinical trials.

Trial Registration: ClinicalTrials.gov Identifier: NCT04920292

Model-informed drug development (MIDD) has emerged as a cornerstone paradigm in global pharmaceutical innovation. Historically underutilized in China, MIDD methodologies gained momentum following the National Medical Products Administration's (NMPA) 2020 release of the Model-Informed Drug Development Technical Guideline, which was subsequently augmented by supplementary technical guidelines to systematically promote and institutionalize MIDD adoption. This study conducts a longitudinal analysis of MIDD implementation in China-approved innovative drugs from 2018 to 2024, spanning pre- and post-guideline eras.

Checkpoint inhibitors that target PD-1 or PD-L1 have had a profound effect in a variety of cancers, both as a single therapy and in combinations. Meta-analyses suggest that monoclonal antibodies (mAbs) targeting PD-1 may yield better survival outcomes compared to anti-PD-L1 mAbs, however these conclusions are limited by a lack of direct clinical comparisons between the two classes. There is a shared hypothesis for the mechanism of action of these drugs: inhibition of the PD-1:PD-L1 signaling pathway through binding to either target. Using a Quantitative Systems Pharmacology (QSP) model-based analysis, we test whether differential inhibition of PD-1:PD-L1 complex formation (a surrogate for inhibition of the signaling pathway) is sufficient to explain the efficacy difference between anti-PD-1 and anti-PD-L1 mAbs observed in clinical meta-analyses. The model predicts that high levels of PD-1:PD-L1 complex inhibition are achieved by all the considered mAbs at their clinical dosing regimens, but it does not indicate that anti-PD-1 mAbs yield higher inhibition over anti-PD-L1s, in contrast to the meta-analyses. Significant model parameter variability and a bootstrap sampling analysis mirroring the comparison from Duan et al. (2020) do not change this conclusion. This suggests that anti-PD-1 and anti-PD-L1 mAbs are not differentiable based on PD-1:PD-L1 complex inhibition alone, and that the hypothesized shared mechanism of action of the two classes of drugs is incomplete.

Covariate selection in population pharmacokinetics modeling is essential for understanding interindividual variability in drug response and optimizing dosing. Traditional stepwise covariate modeling is often time-consuming, compared to the new machine learning alternatives. This study investigates the use of neural networks with stochastic gates for automated covariate selection, aiming to efficiently identify relevant covariates while penalizing excessive covariate inclusion. On various synthetic datasets, the approach demonstrated robustness in detecting important covariates, overcoming challenges such as high correlations, low covariate frequencies, high interindividual variability, and complex covariate dependencies. In real clinical data from a monalizumab study, the method successfully identified covariates that matched those found by experts. However, for tixagevimab/cilgavimab, it identified a superset of covariates, indicating a potential need for further pruning. This machine learning-based method enhances the covariate preselection process in population pharmacokinetics model development, offering significant time savings and improving efficiency even under challenging scenarios.

Hereditary angioedema (HAE) is a rare disorder linked to kallikrein-kinin system dysregulation, which leads to uncontrolled activation of plasma prekallikrein. Donidalorsen is an antisense oligonucleotide designed to selectively degrade prekallikrein messenger RNA and thereby reduce prekallikrein production. We aimed to develop population pharmacokinetic and pharmacokinetic/pharmacodynamic models of donidalorsen and evaluate the impact of potential intrinsic/extrinsic covariates on exposure and prekallikrein response. Plasma donidalorsen and prekallikrein data were obtained from phase 1 to 3 studies in healthy volunteers (NCT03263507, 721744-CS9) and adult and adolescent patients with HAE (NCT04030598, NCT05139810). The evaluated doses were 20, 40, 60, and 80 mg every 4 weeks (Q4W) and 80 mg every 8 weeks (Q8W), administered subcutaneously over 13–21 weeks. Donidalorsen pharmacokinetics were well described by a linear 2-compartment model with first-order absorption. The population terminal elimination half-life was 31.4 days. Prekallikrein was well described by an indirect response model with inhibition of prekallikrein production by donidalorsen. Covariate analysis identified body weight as the main factor affecting pharmacokinetic exposure; however, this effect was not considered clinically significant. The developed population pharmacokinetic/pharmacodynamic model well characterized the donidalorsen exposure–prekallikrein response relationship. Modeling analyses support that no dose adjustment is needed with respect to intrinsic/extrinsic factors in adults and adolescents with HAE. The nearly identical simulated pharmacokinetic or prekallikrein time courses for Q4W versus monthly dosing and for Q8W versus every-2-month dosing regimens support switching to more convenient regimens for patients.

Oncolytic viruses, specifically Sindbis virus (SINV), combined with cytokines show promising results in slowing glioma progression, but a quantitative understanding of their effects remains limited. In this study, we use an ordinary differential equation (ODE) model to examine the effect of adding cytokines to oncolytic SINV therapy. We fit the mathematical model to data extracted from published tumor growth curves to estimate key model parameters. We find that there are statistically significant differences between the infection rates of SINV and cytokine-bearing SINV, as well as differences in the cytokine's ability to reduce viral production. Model simulations show that the addition of cytokines causes an almost immediate reduction in the tumor size caused by the increased viral infection rate. The simultaneous reduction in viral production caused by the cytokines results in oscillations in virus, cytokines, and tumor volume. By providing parameter estimates for key biological processes, our model can help optimize treatment strategies and guide future research in oncolytic virotherapy.