Abstract
Paclitaxel is a major chemotherapeutic drug used to treat a variety of tumour types. Through targeting microtubules, paclitaxel induces abnormal or arrested cell mitosis, leading to tumour shrinkage. The cytotoxicity of paclitaxel limits its clinical use, it is effective only at treating certain tumour types and it is not possible to predict which patients will respond well to treatment. The newer anti-mitotic drugs that have been developed to overcome some of these problems have thus far been less effective than paclitaxel in the clinic. One property of paclitaxel that distinguishes it from many of these other anti-mitotic drugs is its ability to concentrate within cells, a property that could allow intra-tumour concentrations to remain higher for longer following drug dosing. In this paper we seek to develop a mathematical model that can explain observations of paclitaxel uptake in isolated monolayer cultures. We perform a series of experiments on HeLa cell monolayers in which intracellular paclitaxel concentrations are measured under different treatment protocols. We then derive a spatially homogeneous model of paclitaxel uptake and use Bayesian inference to identify model parameters. As a prediction from this model was found to be inconsistent with a further set of experimental results, we consider a generalisation of the model to account for spatio-temporal dynamics and resolve the disparity between theory and experiment. The subsequent inclusion of the spatio-temporal dynamics provides a theoretical framework for the model to be extended to explain drug retention within multilayered tissues. This is important because paclitaxel penetration and release is expected to depend on local 3D tissue architecture.