PDF, Englisch Download (36MB) | Nutzungsbedingungen

Abstract

A thorough understanding of drug-target relationships is essential in preclinical and clinical translation studies. However, there is a gap of knowledge in the quantitative understanding of dose-response relationships at the target site. To fill that gap, a particularly promising approach is quantitative systems pharmacology (QSP) where, mechanistic and hence comprehensive models of dose-effect relationships are used to guide the design of clinical and translational studies. In this thesis, I present for the first time a QSP approach for a therapeutic protein, interferon alpha (IFN-a), by coupling physiologically based pharmacokinetic (PBPK) models at the whole-body level with intracellular models of signal transduction in the liver. Whole-body distribution models of an injected dose of IFN-a calibrated to quantitative measurements of the plasma concentration are established for humans and mice. They are then coupled to mechanistic intracellular models of the triggered JAK/STAT signalling cascade that describes the dynamic response in the expression of the antiviral mRNAc of IRF9 for humans and antiviral protein Mx2 for mice on the cellular scale. By doing so, I am able to establish the quantitative dose-effect relationship of the injected IFN-a dose to the responding interferon stimulated genes (ISGs) triggered at the target site, the liver. The established multi-scale physiologically based pharmacokinetic/pharmacodynamic (PBPK/PD) model of human predict a reduced response of IRF9 mRNAc to IFN-a under physiological in vivo conditions as compared to in vitro. The QSP model also elicits the large impact of the IFN-receptors on the clearance of IFN-a in the liver, thus, not only providing mechanistic insights into the pharmacodynamic (PD) response but also elucidating the influence of receptor variability on the response. Although IFN-a is specifically used in humans, in preclinical studies, it is also tested in mice for understanding the medical impact of IFN-a for other diseases. Therefore, I elaborate an analogous QSP model for the IFN-a response in mice to illustrate possibilities of model-based cross species translation. Like the human model, a whole body PBPK/PD mouse model was also established to follow the response of antiviral protein Mx2. The model clarified the differences between the pharmacokinetics of human and murine IFN-a injection in mice and will support quantitative crossspecies extrapolation in the future. Finally, as heterogeneity in ISGs reflects inter-cell variability in response to IFN-a, I study the impact of sources of this heterogeneity by implementing the mechanistic stochastic model of the JAK/STAT signalling pathway. The model was developed on the basis of time-resolved flow cytometry data of two ISGs, MxA and IFIT1, in Huh7.5 cells. The model analysed intrinsic variability in the concentration of the molecules of the pathway and generated a graded response of MxA and IFIT1 instead of an all-or-none response. Ultimately, the model concludes that the stochasticity in the initiation of the signalling pathway, i.e., at the receptor level, can be buffered by the system and a more robust response of ISGs, MxA and IFIT1 is induced.