Raphaël Saporta: Model-based translation of antibiotic-bacteria-neutrophil interactions

Abstract

The rapid emergence of antibiotic resistance warrants the development of new antibiotics and the optimisation of the use of existing ones. Novel translational methods could help maximise the information gained from preclinical data generated during drug development. Despite the pivotal role of the host’s immune system in infection control, interactions between antibiotics, bacteria and the immune system are not fully understood. Pharmacokinetic-pharmacodynamic (PKPD) modelling approaches, which allow to describe changes in bacterial load over time, were used in this thesis to evaluate model-based translation for antibiotics and characterise the impact of immune response on bacterial killing and antibiotic PKPD.

PKPD models were built for meropenem and afabicin to describe bacterial dynamics in vitro using static time-kill data, and in vivo using mouse thigh or lung infection data. Neutropenic and immunocompetent mouse infection models were used to investigate the contribution of immune response. Translation of antibiotic effects was explored by standard PK/PD index-based approaches and by using the developed PKPD models to perform predictions of bacterial dynamics in different systems. The impact of the design of mouse dose fractionation studies on antibiotic PKPD analyses was also assessed in a simulation-based analysis.

Through model-based analyses, the PKPD relationships for the effects of meropenem and afabicin on bacterial killing, with and without an intact immune system, were quantitatively characterised. An in vivo effect of meropenem against resistant Escherichia coli and Klebsiella pneumoniae was identified. By studying mice with varying degrees of immunosuppression, differences in meropenem PK and a reduced bacterial killing attributable to meropenem in presence of an immune system were described. A model-based translation framework was successfully developed for afabicin, integrating in vitro data and both neutropenic and immunocompetent mouse thigh infection data. The framework demonstrated the ability to translate the activity of afabicin against Staphylococcus aureus from in vitro to in vivo settings and across bacterial strains, and ultimately predicted bacterial killing in patients. In conclusion, the studies performed in this thesis contributed to advancements in the development and application of model-based translation approaches for antibiotics and enhanced the understanding of interactions between antibiotics, bacteria and the immune system.