Abstract
Prescribing medicines requires careful selection of doses in order to derive specific desirable effects while avoiding or mitigating the undesirable effects. This is typically achieved through a standard (unified) dose given to all patients. However, with many drugs, a one-size-fits-all dose would result in many patients not achieving treatment target, either as treatment failure due to under dosing or toxicity due to an overdose. Such a problem is more prominent with drugs that have a narrow therapeutic window and/or significant variability in the pharmacokinetic characteristics and/or pharmacodynamic behaviour. Dose individualisation is necessary for optimal therapeutic benefits to be derived from such drugs. Essential to dose individualisation tools is the identification and quantification for the source(s) and magnitude of variability in pharmacokinetic characteristics and pharmacodynamic behaviour. Based on that, an algorithm can be developed to account for such variability in dose calculation.
The work conducted in this thesis involves the application of various pharmacometric techniques to create tools and techniques for dose individualisation using anticoagulation in children as a motivating example. First, the thesis starts with introducing the main topics and concepts discussed throughout and reviewing the pharmacology of unfractionated heparin, a drug used as an example in this thesis.
Afterwards, bottom-up approaches for dose-individualisation were investigated. This included the development of a quantitative systems pharmacology model of coagulation in children which is then used to explore the influence of maturation of the haemostatic system on the effect and response of unfractionated heparin. It also included investigation of methods and techniques for simplification of quantitative systems pharmacology models and the proposal of a novel machine learning-based model simplification technique.
Finally, the thesis focused on developing a top-down population pharmacokinetic model of enoxaparin in children. The model was used to evaluate the current enoxaparin dosing guidelines in different paediatric subpopulation and propose more appropriate dosing recommendations.