Li, Xia: Application of population, physiologically based, and semi-physiological pharmacokinetic modeling to assess sources of pharmacokinetic variability in individual drugs. - Bonn, 2020. - Dissertation, Rheinische Friedrich-Wilhelms-Universität Bonn.
Online-Ausgabe in bonndoc:
author = {{Xia Li}},
title = {Application of population, physiologically based, and semi-physiological pharmacokinetic modeling to assess sources of pharmacokinetic variability in individual drugs},
school = {Rheinische Friedrich-Wilhelms-Universität Bonn},
year = 2020,
month = jun,

note = {Large inter- and intra-individual variability in the response of drugs causes difficulties for clinicians to choose appropriate dosing regimens to target their therapeutic range. While underexposure might cause sub-therapeutic effects, overexposure might lead to adverse effects. Therefore exploring the sources of variability and quantifying the effect of respective covariates are critical for dosing optimization. We applied population, physiologically based, and semi-physiological pharmacokinetic modeling to assess sources of pharmacokinetic variability in individual drugs.
In the first project, we aimed to identify the best predictor for ciprofloxacin dosing in ICU patients, considering both liver and kidney dysfunction. Therefore, based on the datasets from the clinical study of ciprofloxacin in ICU patients, we used a population pharmacokinetic approach to test the relationship between ciprofloxacin clearance and parameters related to kidney and liver function. Further, age, sex, and total bilirubin plasma concentrations were identified as significant covariates, which in total explained 60% of inter-individual variability. Then model-based simulations for efficacy and toxicity were performed to provide the optimal dosage for individual patients according to their age, sex, and bilirubin levels. A dose reduction to 400 mg seems reasonable for female patients with higher age and considerably increased bilirubin if the MIC values of the causative strains are ≤0.25 mg/L.
For the second project, our aim was to investigate the metabolism of voriconazole in detail to understand dose- and time-dependent alterations and genetic polymorphism of CYP2C19 in the PK of voriconazole. Therefore, we performed in vitro assay to confirm the time-dependent inhibition of voriconazole on CYP3A and integrated into the PBPK model of voriconazole, together with the reference CYP2C19 expression values in each genotype group. Both in vitro results and in vivo datasets from clinical trials supported the development of a whole-body PBPK model of voriconazole. The PBPK model evaluation demonstrated a good performance of the model, with 71% of predicted/observed aggregate AUC ratios and all aggregate Cmax ratios from 28 evaluation datasets being within a 0.5- to 2-fold range. The results of model-based simulation showed that 200 mg voriconazole twice daily oral dosing is sufficient for CYP2C19 IMs (intermediate metabolizers: *1/*2, *1/*3, *2/*17, and *2/*2/*17) to reach the tentative therapeutic range of >1-2 mg/L to <5-6 mg/L for Ctrough (trough concentrations for multiple dosings), while 400 mg might be more suitable for RMs (rapid metabolizers: *1/*17, *17/*17) and NMs (normal metabolizers, *1/*1). When the model was integrated with independently developed CYP3A4 substrate models (midazolam and alfentanil), the observed AUC change of substrates by voriconazole was inside the 90% confidence interval (CI) of the predicted AUC change, indicating that CYP3A4 inhibition was appropriately incorporated into the voriconazole model.
My third project was to quantitatively assess hepatic and intestinal CYP3A inhibition. We designed a novel clinical study to quantify the time-course of CYP3A inhibition in liver and intestine. The combination of intravenous and intestinal infusions of midazolam and voriconazole turned out to be suitable to investigate the systemic and pre-systemic CYP3A metabolism, enabling to identify the contribution of hepatic and intestinal metabolism. Semi-physiological pharmacokinetic modeling was applied to describe the substrate and inhibitor concentrations in liver and intestine, and to further access the inhibitory effects of voriconazole on the hepatic and intestinal CYP3A activity.
In my fourth project, cyclosporine A is susceptible to drug-drug interactions since it is metabolized mainly by CYP3A and it is a substrate of P-glycoprotein. Baicalin’s main metabolite baicalein can inhibit CYP3A and P-gP. This study was carried out to investigate the effect of baicalin on cyclosporine A pharmacokinetics and to explore the safety co-administration of cyclosporine A and baicalin in humans. Subjects received a single 200 mg oral cyclosporine A dose alone in the reference period and combination with 500 mg baicalein in the test period. Our non-compartmental analysis resulted that 90 % CIs of AUC and Cmax test-to-reference ratios were within the bioequivalence boundaries of 80-125%, indicating no relevant effect of baicalin co-administration on CsA PK was identified.
My fifth project was mainly carried out to evaluate the effect of acute alcohol consumption on the activity of intestinal and hepatic CYP3A4. Within this project, semi-physiological pharmacokinetic modeling was used to describe the pharmacokinetics of ethanol and to describe the effect of ethanol on intestinal and hepatic CYP3A activity. The assessment of the ethanol effect on midazolam metabolism at both sites resulted in that ethanol reduced intestinal midazolam extraction to be 0.77-fold (90% CI 0.69–0.86) but had no significant effect on midazolam hepatic clearance. Thus, a high ethanol exposure when occurring during treatment with drugs having extensive first-pass metabolism by CYP3A may occasionally cause a relevant increase in drug exposure.
All approaches have powerful descriptive and predictive abilities and can support decision-making during drug development. In the case of voriconazole presented here, even a combination of models was required to improve the understanding of the properties of the drug. Without the development of the PBPK model of voriconazole, which provided the appropriate mechanistic components, the development of a semi-physiological model with the additional options, has not been possible. The selection of the most suitable of the modeling approaches ultimately depends on the specific questions to be answered and the type of data available.},

url = {}

Die folgenden Nutzungsbestimmungen sind mit dieser Ressource verbunden: