Abstract
Mathematical modeling is an integral part of the drug development process. Models are developed to describe tumor dynamics or drug concentration to answer questions such as: What concentration is required to reach the desirable treatment outcome? What drug dose and frequency should a drug prescription specify to achieve this concentration? Models are also used for simulation, reducing the need to perform additional animal trials.
In this thesis, we consider how to model dynamical systems that incorporate biologically relevant phenomena such as drug elimination, tumor growth, and the effect of combination therapies consisting of anticancer drugs and radiation treatment. Survival analysis and time-to-event modeling are also discussed as well as how to combine these types of probabilistic models with the dynamical system models in a so-called joint model. All discussed models contain parameters that must be estimated using experimental data and how this estimation is done is considered along with how to deal with variability on different levels, e.g., between individuals and between species.
Furthermore, appended are five papers/manuscripts where this is applied to real-world problems. (I) concerns modeling of radiation therapy in combination with radiosensitizers, (II) presents a translational approach for predicting clinical results using a preclinical model, and (III) focuses on predicting progression-free survival using joint modeling. The last two are in manuscript form and (IV) presents a parametric model for sample size calculations and (V) considers how predictions of progression-free survival are distributed under different models.
Date: Wednesday, 20 March, 10:15 am
Location: Pascal, Hörsalsvägen 1