Quantification of uncertainty within systems pharmacology to optimise personalised therapy

The proposed research will demonstrate how the quantification of uncertainty within theoretical pharmacology models can be used to optimise personalised therapy. The current one-dose-fits-all viewpoint, dosing for the 'average' human and empirical derivations of dosing regimens are clearly not sufficient to effectively treat a diverse population. Personalised medicine, with the determination of treatment plans influenced by genetic factors and historical cases of toxicity, represents a step towards improving clinical outcomes. However, when an individual is deemed not suitable for treatment with the standard medicine/dose based on pharmacogenetics, an alternative must be found which is likely to either be costlier, less effective or potentially toxic. This scenario can be improved by having a greater understanding of both the mechanisms of toxicity in subpopulations and a proper quantification of risk and efficacy. Mathematical and statistical tools can help to improve the understanding of the mechanisms of toxicity, optimise treatment (e.g. alternative dosing regimens, multidrug administration, assess alternative medicine strategies) and communicate the risk of suggested therapy based on uncertainty in the model simulations. To develop and test this proposed quantitative framework during the fellowship, two exemplar drugs will be studied based on adverse reactions in anticoagulation therapy. Warfarin will be used to develop and validate the framework, while the newer direct anticoagulants (e.g. dabigatran, rivaroxaban), whose metabolism and toxicity pathways are less well characterised, will be used for more predictive research. Statistical techniques will be applied to define and quantify uncertainties within model output and define risk when translating to optimised treatment strategies incorporating different genetic and non-genetic factors associated with variability in drug-induced adverse reactions. The integration of applied mathematics and statistics comprises a potentially exciting new field bringing two modelling approaches together to better quantify uncertainty and inform industrial drug development and therapy.

The research programme I propose for the fellowship focuses on the important scientific health theme of stratified medicine. The ability to accurately determine how to optimise therapy or even develop a drug is highly dependent on the accuracy of the predictive information being supplied. Proper quantification of the uncertainty associated with this information is therefore paramount when being potentially used for important diagnostic and therapeutic purposes. The ability to detect toxicity, including idiosyncratic responses, within individuals of populations, rather than the prediction for an idealised 'average' person, would be a significant step change in current approaches and could therefore potentially offer significant impacts to public health. The mechanistic nature of the modelling methodology in this proposed programme therefore makes the research ideally suited for applications within the priority area of stratified medicine. Specifically, distinct drug responses within a population can be simulated with the identification of a perturbation in an underlying biological process. The mechanistic model allows for direct manipulation of this part of the drug-response pathway. Furthermore, information regarding the prevalence of a given subgroup through pharmacogenetic data analysis would allow the prediction of the likelihood of that response. The integration of these mechanistic mathematical approaches with statistics and uncertainty quantification techniques significantly improves upon current mathematical models by employing rigorous methods to understand, communicate and minimise unwanted variation (modelling-related uncertainty) whilst preserving and exploiting wanted variation (inherent population variability).

The key aim of this research programme is to develop my skills in statistical methods so that I can improve the predictivity and impact of mechanistic-based models used within pharmacology and toxicology to optimise personalised therapy. The integration of applied mathematics techniques, used to develop mechanistic models of potential drug dosing and toxicity, with statistical tools and analysis, used to quantify uncertainty, comprises a potentially exciting new field bringing two different modelling approaches together to better inform industrial drug development, testing, regulation and therapy. A key novel application area is the incorporation of uncertainty information within alternative treatment strategies coupled to PBPK models. The additional complexity of multiscale heterogeneity within these systems will present a challenge to the current methods of uncertainty quantification and could therefore open up new opportunities for method development, which would form the basis of new research. Additionally, I anticipate that the novelty of this proposed work, coupled with its importance within pharma, will lead to a number of high profile publications as well as industrial impact.