Virtual patient populations for quantitative pharmacology with its application in antibiotic treatment optimization

Abstract

This thesis advances approaches for generating realistic virtual patient populations and demonstrates their application in quantitative pharmacological modelling to characterize variability in pharmacokinetics (PK) and pharmacodynamics (PD) and its implications for antibiotic treatment.

The first part focuses on the development and application of copula-based approaches for generating virtual patient populations that capture the complex dependencies and variability observed in real-world patient data. These approaches provide a flexible framework for representing multidimensional patient characteristics and generating realistic virtual populations for quantitative pharmacological applications.

The second part focuses on characterizing variability in PK and PD, and evaluating its impact on antibiotic treatment response. Through quantitative mechanism-based modelling and simulation, these studies provide insights into the sources and consequences of variability in antibiotic exposure and response, supporting a more individualized understanding of antibiotic treatment.

Together, the studies demonstrate the value of realistic virtual populations in quantitative pharmacological modelling as an approach for understanding variability and supporting antibiotic treatment optimization. The developed approaches provide a framework that can be extended to other patient populations, therapeutic indications, and applications in drug development and patient care.