Contributions to Clinical Dose Individualisation

Medicines are a crucial component in the treatment and prevention of many diseases. Tailoring dose size and frequency to achieve a therapeutic target that meets the needs of an individual patient is a fundamental objective of treatment. Pharmacokinetic-pharmacodynamic models allow for the dose-concentration-effect relationship, and associated uncertainty to be characterised. The use of a model informed approach can promote the use of rational rather than empirical dosing regimens and optimise dose for individual patients. The research performed in this thesis aimed to examine how the quality and safety of vancomycin and warfarin treatment may be improved by using model-informed approaches to individualise therapy. In this thesis, models supported by pharmacological theory were implemented in the clinic to inform vancomycin and warfarin treatment. For vancomycin, a before-after design was used to compare (model-informed) AUC-guided approach with (historical) trough guided treatment. Model-informed treatment was associated with improved target attainment (acceptable concentration observations: 59% v 23.4%, P < 0.001) and reduced nephrotoxicity (odds ratio: 0.356, P = 0.035) without compromise in cure. For warfarin, a model-informed approach was compared to standard of care using a randomised experimental design. Patients allocated to model-informed treatment spent a higher proportion of time within the acceptable range (days 0-28: 63% v 56%, P = 0.13) and experienced a lower hazard of minor bleeding event (hazard ratio: 2.4, P = 0.005). In both studies the model proposed doses were widely accepted and prescribed by clinical staff. The performance of the models implemented for clinical use were evaluated against other published models. For warfarin, external evaluation was performed to examine model performance and identify the best model to implement for the clinical trial. An iterative approach to external evaluation identified the initial advantage of including genotype as a covariate as well as the loss of usefulness following INR feedback. For vancomycin external evaluation was performed to benchmark the model used in the clinical study and found comparable performance for predicting and forecasting concentration.

Name thesis:degree_name

PhD

Level thesis:degree_level

Doctoral

Discipline thesis:degree_discipline

Clinical Pharmacology

Grantor dc:publisher

ResearchSpace@Auckland

Year dc:date.issued

2022

Author dc:creator

• Ma, Guangda

Advisors dc:contributor.advisor

• Holford, Nick
• Hannam, Jacqui
• Harrison, Jeff

Statement dc:rights

• Items in ResearchSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Licence dc:rights.uri

• https://researchspace.auckland.ac.nz/docs/uoa-docs/rights.htm

Handle dc:identifier.uri

https://hdl.handle.net/2292/64872

OAI identifier oai:identifier

oai:researchspace.auckland.ac.nz:2292/64872