Journal of Pharmacokinetics and Pharmacodynamics最新文献

PK-QTc analyses are an integral part of drug development programs. These analyses are often based on phase I study data, and the question may be asked whether the design of these phase I studies has an impact on the precision of the corresponding PK-QT analysis. More precisely, we are interested whether one can increase the power of such analyses when using interleaved ascending dose designs rather than parallel group ascending dose designs. Based on a simulation study, previous authors have concluded that this is the case. Their conclusions, however, are based on assumptions regarding the magnitude of the random effect variances, and on a very specific set-up of their simulation study. In this paper we provide a study re-analysis of historical QTc data. The resulting estimates of these random effect variances are much smaller than those used by the previous authors. We also propose a simulation set-up that adequately mimics the data generation process and the correlation between the primary endpoint change from baseline and the covariate baseline. We present a simulation study using the revised simulation set-up and random effect variances as observed in our study re-analysis. We did not find major differences in power between the different designs when the number of observations is the same. We also provide a justification based on causal analysis why we think our simulation set-up is more adequate for situations when change from baseline is the primary endpoint, specifically when baseline is used as a covariate.

Prolongation of the QT interval in the ECG is a critical finding that signifies an extended duration from the onset of ventricular depolarization to the end of ventricular repolarization. It can predispose patients to life-threatening arrhythmias, such as Torsades de Pointes (TdP). Long QT syndromes (LQTS) are defined by mutations in ion channel genes, particularly those encoding cardiac potassium and sodium channels and are characterized by a significant risk for sudden cardiac death if untreated. However, besides these clearly defined entities various medications have been implicated in causing QT interval prolongation. There is increasing evidence for a genetically determined risk for drug-induced QT prolongation. In addition, due to numerous clinical factors influencing the QT interval, QT prolongation increases the risk of TdP particularly in multi-morbid patients necessitating vigilant monitoring in at-risk populations. This review gives an overview of mechanisms and conditions which induce QT prolongation, the clinical assessment of QT interval duration, thereby highlighting quantitative variations in measurement techniques and heart-rate correction, as well as in demographic interpretation of normal values. The risk of cardiac arrhythmia is discussed, in both patients with congenital LQTS and acquired QT prolongation, along with influencing pharmacokinetic/pharmacodynamic, non-pharmacologic and genetic risk factors for TdP. Finally, clinical implications for individual patient management, including risk-adapted drug-prescription and use of ECG monitoring to mitigate the risks associated with QT prolongation, are summarized. Understanding the interplay between pharmacokinetics, pharmacodynamics, genetic predisposition and co-morbidities is essential for optimizing treatment in the context of prolonged QT intervals, preventing adverse cardiovascular events, and improving cardiac safety. Comprehensive drug labelling regarding exposure-QT relationships and available pharmacovigilance data are important sources of information enhancing patient safety.

Nonlinear mixed-effects models rely on adequate initial parameter estimates for efficient parameter optimization. Poor initial estimates can result in failed model convergence or termination with incorrect parameter estimates. Non-compartmental analysis (NCA) and other manual methods have typically been used to derive initial estimates for pharmacokinetic (PK) parameters. However, NCA struggles with sparse data and recent advances in automated modeling increasingly emphasize the need for initial estimates that require no user input. This study aimed to develop an integrated pipeline for the computation of initial estimates applicable to various data types and model structures. The designed pipeline incorporated a custom-designed algorithm that leveraged data-driven methods to generate initial estimates for both structural and statistical parameters in population pharmacokinetic (PopPK) base models. The pipeline's performance was evaluated across twenty-one simulated datasets and thirteen real-life datasets. The results suggested that this pipeline performed well in all test cases. Initial estimates recommended by the pipeline resulted in final parameter estimates closely aligned with pre-set true values in simulated datasets or with literature references in the case of real-life data. This study provides an efficient and reliable tool for delivering PK initial estimates for population pharmacokinetic modeling in both rich and sparse data scenarios. An open-source R package has been created.

Despite tremendous clinical responses and patient benefit in hematological malignancies, chimeric Antigen Receptor (CAR) T cells have demonstrated limited success in solid tumors. Herein, we have scaled and augmented our previously described murine PBPK-PD model (Singh et al. mAbs, 2020) to characterize cellular kinetics (CK) and anti-tumor activity in patients with solid tumor malignancies. The model was able to integrate (1) differential kinetics of effector- and memory-phenotypes in peripheral blood (PB), solid tumors and other pertinent tissues (n = 8), (2) host-immune system dynamics with or without prior lymphodepletion chemotherapy (LDC) and its impact of CAR-T cell kinetics and (3) antigenic heterogeneity in patients. Model was developed based on digitized individual level CK, categorical antitumor activity and percentage tumor antigen expression dataset from following phase-1 dose-escalation studies: (A) anti-mesothelin CAR-T in multiple cancer indications (n = 15, cohorts w/ and w/o LDC), (B) gavocabtagene autoleucel (n = 7, w/ and w/o LDC) in multiple indications, (C) anti-glypican 3 CAR-T in advanced hepatocellular carcinoma (n = 13, dose-range 0.7-5.18 billion) and (D) anti-PSMA/TGFβ CAR-T in prostate cancer (n = 10, w/ and w/o LDC). The developed clinical PBPK-PD model was able to simultaneously characterize the CK and categorical anti-tumor longitudinal dataset(s) for each case study while accounting for antigen-expressing tumor burden in each patient. Moreover, model accounted for host-T cell population dynamics post LDC, which competed with CAR-T cell towards overall expansion and persistence post-treatment. Using model simulation, CAR-T cell expansion was found to be dependent on initial tumor burden and antigen positive tumor fraction. The developed PBPK-PD model could be leveraged as an effective tool in future to provide mechanistic understanding on CK-PD behavior of cell therapies targeting solid tumors.

Solving models comprised of nonlinear differential equations (DEs) in NONMEM using ADVAN6 or ADVAN13 typically requires substantially longer run times than models comprised of linear DEs, which in some cases allow for analytical solutions. Often the need to use nonlinear DE solvers results from pharmacokinetic (PK) variations over the dosing interval introducing the nonlinearity via a nonlinear transfer function, as is the case for indirect-response models and enzyme induction models. As long run times hinder model development, it is desirable to derive suitable approximations to speed up model solutions. The zero-order hold, a concept used in the field of advanced process control to optimize control decisions, provides an attractive approximation for these situations that often results in a sequential system of simpler DEs that in some cases can be solved analytically. Two examples, an indirect-response model and an enzyme induction model, demonstrate that the zero-order hold approximation provides a substantial reduction in computational time (up to ~ 140-fold) without unduly biasing the parameter estimates. These examples demonstrate that the zero-order hold approximation offers an attractive method for efficiently solving models where time-varying PK leads to a nonlinear system of DEs.

When a tumor is treated with an antibody-drug conjugate (ADC) complex biochemistry occurs in a domain-the tumor-whose size and structure are changing. Some parts of the tumor may be growing because tumor cells proliferate. Other parts may be stagnant, or nearly so, because the cells there have been damaged by the cytotoxin. Still others may be shrinking because the cells there have been killed by the cytotoxin and are being cleared. Chemical concentrations within the tumor, which influence kinetics and transport, change as the tumor grows or shrinks. Cell surface antigen, to which ADCs are designed to bind, is lost when cells are cleared and is freshly introduced when cells proliferate. For these reasons, and because shrinking the tumor by killing its cells is the purpose of ADC treatment, it is important in a quantitative systems pharmacology (QSP) approach to the problem to model the evolution of tumor size and structure over the course of ADC treatment. In this paper we present a partial differential equation (PDE) model of ADC transport and kinetics in a growing and shrinking Krogh cylinder tumor. We present results of several studies we performed with the model, including an antigen concentration study that shows tumor growth inhibition to be non-monotone as a function of antigen concentration, and a study of the effects of co-administration of mAb and ADC that shows that the greater the delay between mAb and ADC administration the less the effect of co-administration, and which suggests the mechanism for this effect.

Tisotumab vedotin (TV), a tissue factor-directed antibody-drug conjugate (ADC), is approved in the US at 2.0 mg/kg every 3 weeks (Q3W) for adult patients with recurrent or metastatic cervical cancer following disease progression on or after chemotherapy. Previous logistic regression analysis showed a positive association between TV exposure and ocular adverse events (OAEs), which were identified as prespecified AEs of interest in TV clinical studies. To further optimize TV dose from a safety perspective, we developed a discrete-time Markov model (DTMM) to characterize exposure-response (E-R) relationships of exposures of both ADC and the microtubule-disrupting agent monomethyl auristatin E to the incidence, severity, and longitudinal time course of grade ≥ 2 OAEs in patients with advanced solid tumors. A total of 757 patients who received TV as monotherapy or combination (with carboplatin, bevacizumab, or pembrolizumab) across seven clinical studies were included in this analysis. Of multiple covariates modeled, implementation of an eye care plan was the only covariate to significantly reduce risk of grade ≥ 2 OAEs. The DTMM suggested an association between ADC exposure and risk of grade ≥ 2 OAEs. Based on the totality of data from clinical outcomes, pharmacokinetics, and E-R analyses, as well as DTMM modeling results, TV 1.7 mg/kg every 2 weeks may provide higher efficacy with slightly increased risk of OAEs compared with 2.0 mg/kg Q3W, although these OAEs are manageable with an appropriate eye care plan. ClinicalTrials.gov ID (first submission): NCT03485209 (2018-03-08), NCT03657043 (2018-08-22), NCT03438396 (2018-02-08), NCT03786081 (2018-12-13), NCT03913741 (2019-03-29), NCT02001623 (2013-11-14), and NCT02552121 (2015-09-14).

Covariate clinical relevance (CCR) is commonly assessed in population pharmacokinetics using forest plots visualizing parameter changes across covariate values. In our previous work (Philipp et al. 2024), CCR was evaluated using a [0.80-1.20] reference area and a 90% confidence interval for both relevance and significance assessment. However, more conventional thresholds include a broader reference area of [0.80-1.25] and the use of a 5% type I error to assess statistical significance. This commentary extends our previous analysis by evaluating CCR decisions under these more conventional thresholds, in order to assess whether the full model, the stepwise covariate modeling (SCM) and its enhanced version SCM+ remain robust. A comparison with the previous results is also provided. The revised CCR evaluation gave satisfactory results across all three approaches. For covariates with a simulated effect, the full model and SCM/SCM+ provided consistent conclusions with those of the true model. For covariates without a simulated effect, the full model mainly found them non-relevant (NR) non-significant or insufficient information (II) non-significant, while SCM/SCM+ mainly did not select them. These results align with our previous findings. Conclusions for covariates with a simulated effect were almost unchanged. For covariates without a simulated effect, the more conventional threshold allowed the full model to conclude more frequently to their NR instead of II, likely due to the broader reference area and stricter type I error control. Overall, the consistency of our results across different thresholds demonstrates their robustness and supports their generalizability.

Automatic differentiation (AD), a key method for accurately and efficiently computing derivatives in modern machine learning, is now implemented in Phoenix® NLME™ 8.6 for the first time and applied to the first-order conditional estimation extended least squares (FOCE ELS), Laplacian, and adaptive Gaussian quadrature (AGQ) algorithms. We name the AD implementation as 'automatic-differentiation-assisted parametric optimization' (ADPO), which can be enabled by checking the 'Fast Optimization' option. We present in detail how ADPO is implemented in the frequently used FOCE ELS algorithm, and analyze its performance from the benchmarks based on four PK/PD models. We show both ADPO and traditional FOCE ELS which uses gradients obtained from finite difference (FD) are reasonably accurate and robust, while the main advantage of ADPO being that it considerably reduces computation time no matter what ODE solvers are used: in general ADPO reduces the total run time by around 20% to 50% compared to traditional FOCE ELS. In a case for the realistic voriconazole model using 'auto-detect' ODE solver, 95% reduction in the total run time is observed.