Clinical Pharmacokinetics最新文献

Background and objective: Fosfomycin trometamol may be a valuable option for antimicrobial prophylaxis before urological surgery for benign prostatic hyperplasia. The objective is to develop a population pharmacokinetic model of a novel oral fosfomycin prophylactic scheme in the plasma and prostate of patients undergoing endoscopic surgery for benign prostatic hyperplasia.

Methods: One- and two-compartment plasma pharmacokinetic models were fitted to fosfomycin plasma data, and different plasma-prostate linked models were tested by means of non-linear mixed effects modelling. Monte Carlo simulation was used to obtain 1000-subject concentration-time profiles of fosfomycin in the prostate. Probabilities of target attainment ≥ 90% of an area under the plasma concentration-time curve/minimum inhibitory concentration (MIC) > 83.3 and of a 70%t > MIC in the prostate were considered as optimal. Cumulative fractions of response against both wild-type and extended-spectrum beta-lactamase-producing Escherichia coli were calculated.

Results: A total of 104 patients, each providing a pair of plasma and prostate concomitant samples were included in the study. A one-compartment pharmacokinetic model was used to describe plasma fosfomycin concentration. Fosfomycin plasma-prostate relationships were adequately described by a direct response model with a power function. Simulations showed that fosfomycin disposition in the prostate was closely related to that in plasma. Optimal probabilities of target attainments were ensured against Enterobacterales having an MIC up to 0.5-1 mg/L in the 12 h vulnerable period after completing the prophylactic scheme.

Conclusion: A prophylactic regimen of two doses of oral fosfomycin trometamol 3 g 12 h apart before undergoing prostatic surgery may grant effective concentrations in the prostatic tissue of patients for a 12 h vulnerable period.

Model-informed precision dosing relies on population pharmacokinetic models to personalize drug therapy and improve clinical outcomes. While hundreds of models are published, only a small fraction are externally evaluated and even fewer are applied in practice. This gap reflects a lack of clear standardized guidance on how to identify, assess, and implement published models in new clinical settings. This guidance provides a structured step-by-step framework for the external evaluation and selection of published population pharmacokinetic models to support their use in model-informed precision dosing. It outlines key considerations for defining the model's intended use, assessing the characteristics of the available dataset, screening for suitable candidate models, and applying prediction- and simulation-based diagnostics. By addressing methodological and practical challenges, this framework supports more reproduceable use of published models in real-world settings to help bridge the gap between model development and clinical application.

Background: Baricitinib is approved for the treatment of adults with moderate-to-severe atopic dermatitis (AD) who are candidates for systemic therapy and has received regulatory authorization in Europe for moderate-to-severe AD in patients 2 to <18 years.

Objective: This study aims to optimize dosing for baricitinib in pediatric patients with atopic dermatitis using pharmacokinetic/pharmacodynamic modeling leveraging adult data.

Methods: The phase III, randomized, double-blind, placebo-controlled study, BREEZE-AD-PEDS (NCT03952559, registration date: 2019-05-16), enrolled patients (aged 2 to <18 years) with moderate-to-severe AD. During a pharmacokinetic (PK) lead-in period, baricitinib concentration data from age-based dose cohorts (4 mg once daily [QD]: 10 to <18 years; 2 mg QD: 2 to <10 years) were compared with actual and simulated concentration values from adult patients receiving baricitinib 4 mg QD. A population PK model incorporating allometric scaling was developed to determine weight-based dosing in pediatric patients that matches adult exposures. The exposure-response (E-R) relationships were analyzed for the primary endpoint: a validated Investigator Global Assessment^® (vIGA-AD) score of 0 or 1 (clear to almost clear skin) with ≥2-point improvement from baseline at week 16. Baricitinib pharmacokinetics were characterized from 393 pediatric patients using a 2-compartment model with allometric scaling on clearance and volume of distribution.

Results: The age-based and subsequent weight-based dosing (2 mg for patients 10 to <30 kg and 4 mg for patients ≥30 kg) was comparable to the 4-mg adult exposure. A clear E-R relationship was observed for the primary endpoint when sorted for age or weight groups.

Conclusion: The population PK model developed using baricitinib concentrations from adult patients, with allometric scaling for weight on clearance and volume, adequately predicted exposures in the pediatric population. The PK modeling, with E-R analysis, informed an appropriate weight-based dosing regimen.

Background and objective: Tenofovir (TFV)-based regimens are backbones of both HIV treatment and pre-exposure prophylaxis during pregnancy. Multiple studies have shown up to one-third decreases in dried blood spot tenofovir-diphosphate concentrations during pregnancy among participants taking tenofovir disoproxil fumarate (TDF). Currently, there are no mechanism-based models describing the pharmacokinetics of tenofovir diphosphate (the active anabolite) in peripheral blood mononuclear cells (PBMCs) of pregnant individuals receiving TDF or tenofovir alafenamide (TAF), and the mechanisms behind observed differences between dried blood spots and PBMCs remain unclear.

Methods: To address this gap, we developed a semi-mechanistic model to simultaneously describe the pharmacokinetics of all clinically relevant TDF and TAF-derived moieties and conducted clinical trial simulations to compare TDF and TAF pharmacokinetics during pregnancy and postpartum.

Results: The pharmacokinetics of plasma TAF and TFV were best described by one-compartment and two-compartment models, respectively, with first-order absorption. A transit compartment was included to reflect the slower elimination rate of plasma TFV after receiving TAF. Cellular matrix PBMC and dried blood spots were included using a biophase model. For TDF, plasma TFV apparent clearance increased by 24.9% and 13.1% during the second and third trimesters of pregnancy, respectively, compared with non-pregnant populations. In the postpartum period, plasma TFV apparent clearance in pregnant women was 9.3% lower than in non-pregnant women. The bioavailability for TAF decreased by 17.3% and 5.1% during the second and third trimesters, respectively, and increased by 18% during the postpartum period relative to non-pregnant women. In pregnant women, simulations showed that TAF maintains approximately five times higher tenofovir diphosphate concentrations in PBMCs compared with TDF during the second and third trimesters, despite a decrease in PBMC tenofovir diphosphate concentrations for both drugs. This finding is consistent with the higher PBMC loading effect of TAF observed in non-pregnant populations.

Conclusions: Our semi-mechanistic model provides a framework for understanding pregnancy-associated pharmacokinetic changes and supports future research to refine dosing strategies for HIV treatment and prevention in pregnancy.

Background and objective: There is limited information on the pharmacokinetics and optimal dosing of fluconazole in patients undergoing mechanical circulatory support (MCS) device implantation.

Aim: The aim of this study was to describe fluconazole pharmacokinetics and identify dosing regimens that achieve pharmacokinetic/pharmacodynamic targets in this patient cohort.

Methods: In this prospective, single-centre study, adults undergoing MCS device implantation received intravenous fluconazole 200 mg or 400 mg, continued once daily for 5 days. Fluconazole concentrations were measured at four peri-operative time points, on return to the intensive care unit, and on days 3 and 5 following implantation. The area under the concentration-time curve from time zero to 24 h (AUC_0-24) was estimated, with target exposures defined as AUC_0-24/minimum inhibitory concentration (MIC) ≥ 50 for prophylaxis and AUC_0-24/MIC ≥ 100 for therapy. Population pharmacokinetic modelling was performed using non-linear mixed-effects methods and for Monte Carlo dosing simulations.

Results: Sixty-five fluconazole concentrations from 10 male patients (median age 51.5 years; IQR 50.0-57.0) were included. A two-compartment model including an additional renal replacement therapy (RRT)-dependent clearance pathway best described the data. Parameter estimates from the final model included a central volume of distribution of 5.75 L, non-RRT clearance of 0.45 L/h, and RRT clearance of 2.22 L/h. Simulations showed that doses of 800 mg or 12 mg/kg would be required for the highest probability of target attainment for Candida fluconazole MICs up to 4 mg/L (AUC_0-24/MIC ≥ 50) and 2 mg/L (AUC_0-24/MIC ≥ 100).

Conclusion: In patients undergoing MCS device implantation, intravenous fluconazole 200 mg administered pre-operatively and continued daily for 5 days is insufficient for Candida spp. with MICs > 1 mg/L (prophylaxis, AUC_0-24/MIC ≥ 50) and > 0.5 mg/L (treatment, AUC_0-24/MIC ≥ 100).

Gonadotropin-releasing hormone (GnRH) antagonists inhibit estrogen synthesis and secretion, making them promising treatment options for estrogen-dependent diseases, such as endometriosis. This study developed a population pharmacokinetic/pharmacodynamic (PK/PD) model for merigolix, a novel oral GnRH antagonist, to determine its optimal dosing in the treatment of endometriosis. Population PK/PD modeling was performed using NONMEM 7.4, incorporating data from phase I clinical studies involving single and multiple ascending dose (SAD and MAD) trials in healthy premenopausal volunteers. The PK profile was characterized using a two-compartment model incorporating first-order absorption and elimination processes. The temporal delay between merigolix concentration and subsequent estradiol (E2) suppression was described using an indirect response turnover model. The models were evaluated via visual predictive checks, goodness-of-fit plots, and bootstrap analysis. The PK model described merigolix concentrations across various doses (estimated clearance: 549 L/h, central volume of distribution: 1690 L). The PD model demonstrated dose-dependent E2 suppression (estimated maximum inhibitory effect [I_max]: 1, half-maximal inhibitory concentration [IC₅₀]: 0.209 ng/mL). Simulations suggested that, assuming a baseline E2 concentration of 100 pg/mL, daily doses of 120 and 160 mg achieved the clinically meaningful target E2 range of 20-40 pg/mL (partial suppression), while higher doses of 240 and 320 mg resulted in target E2 levels below 20 pg/mL (full suppression), effectively controlling symptoms and minimizing the risk of bone mineral density loss. This PK/PD model provides a quantitative framework for optimizing merigolix dosing and supports the selection of dosing regimens for future clinical trials, potentially offering a novel therapeutic option for endometriosis treatment.

Objectives: To develop and validate a physiologically based pharmacokinetic (PBPK) population model for the Chinese obese population.

Methods: A Chinese adult population database was established in PK-Sim through recalibration of the East Asian population database, using newly collected anatomical and physiological data from Chinese adults. All three drugs (dexmedetomidine, omeprazole, and propofol) possessing Chinese obese population PK data were selected, with their PBPK models then developed and validated using matched clinical data. These models with the fixed drug-specific parameters were applied to the Chinese adult database to simulate drug concentrations, with results compared with the built-in East Asian database. Then, physiological parameters were adjusted using real-world and literature data from Chinese patients with obesity to establish a Chinese obese adult database. Similarly, drug concentrations in this population were simulated and compared with the simulation results based on the published White obese population database.

Results: The predicted C_max and AUC_last values were within 0.5-2 fold of the observed values, demonstrating all drug models were validated. The Chinese adult database showed superior accuracy to the East Asian database (90% versus 75% of AUC_last and 80% versus 70% of C_max within 0.8-1.25 fold). Similarly, the Chinese obese database outperformed the White obese database (83% versus 33% for AUC_last and 33% versus 0% for C_max within 0.8-1.25 fold).

Conclusions: The validated drug models, combined with the Chinese adult and obese adult databases, reliably predicted drug concentrations in Chinese adults and adults with obesity, outperforming the East Asian population database and White obese population database.