Pharmacometric Models to Improve the Treatment and Development of Drugs against Tuberculosis

  • Datum:
  • Plats: B21, BMC, Husargatan 3, Uppsala
  • Doktorand: Svensson, Robin J.
  • Om avhandlingen
  • Arrangör: Institutionen för farmaceutisk biovetenskap
  • Kontaktperson: Svensson, Robin J.
  • Disputation

The aim of this thesis was to develop pharmacometric models to optimise the development and use of current and future drugs for treating tuberculosis.

With 10 million new infections yearly, tuberculosis has a major impact on the human well-being of the world. Most patients have infections susceptible to a first-line treatment with a treatment success rate of 80%, a number that can potentially be improved by optimising the first-line treatment. Besides susceptible disease, each year half a million patients are infected by tuberculosis with resistance to first-line treatment where only 50% of patients get cured. Thus, new drugs against resistant tuberculosis are desperately needed but given the inefficiency of developing new anti-tuberculosis drugs, enough new drugs will not reach patients in time. The aim of this thesis was to develop pharmacometric models to optimise the development and use of current and future drugs for treating tuberculosis.

A population pharmacokinetic model for rifampicin, the most prominent first-line drug, was developed and later used for developing exposure-response models followed by clinical trial simulations. The developed exposure-response models were based on liquid culture data and were expanded to describe the relationship between liquid culture results and a new biomarker, the molecular bacterial load assay which is a quicker alternative to liquid culture and is also contamination-free.

The in vitro-derived semi-mechanistic Multistate Tuberculosis Pharmacometric (MTP) model was applied to clinical rifampicin and clofazimine colony forming unit datasets. This novel application of the MTP model allowed detection of statistically significant exposure-response relationships between rifampicin and clofazimine for the specific killing of non-multiplying, persister bacteria. Furthermore, the MTP model was compared to conventional statistical analyses for detecting drug effects in Phase IIa. If designing and analysing Phase IIa using the MTP model, the required sample size for detecting drug effects can be lowered. An improved design and analysis of pre-clinical treatment outcome assessments was developed which increased the information gain compared to a conventional design yet kept the animal use at a minimum. Lastly, a therapeutic drug monitoring approach was suggested based on updated targets for rifampicin, a framework easily expandable to second-line drugs.

In conclusion this thesis presents the development of pharmacometric models which will streamline both the development and use of drugs against tuberculosis.