Simulation of oxygen and hypoxia PET-tracer distributions in 3D vascular architectures of tumours

Abstract

The outcome of radiotherapy is strongly influenced by hypoxia (low level of oxygenin the tumour), which leads to increased radioresistance and poor outcome. Positron Emission Tomography (PET) can be used to identify hypoxic regions within tumours. Hypoxia-specific tracers have been developed for this purpose. Deriving the tumour oxygen distribution from a hypoxia PET-tracer distribution is still not straightforward and computer simulations may help to better understand the limitations of this technique. In this thesis, hypoxia PET-tracer distributions were simulated in three dimensions by using mathematical models to calculate the spatiotemporal distributions of both oxygen and tracer in the tumour. The impact of necrosis in the tracer uptake was also considered. The partial differential equations in the model were solved, using the Finite Element Method for different vascular architectures. The correct functioning of the solving method was checked for simple architectures by changing parameter values and analysing the obtained distributions. Time-Activity Curves were calculated from the simulated concentrations and the impact of the vascular fraction was determined, for vascular architectures of different complexity. This study improves the understanding of the signal in hypoxia-PET studies. Future work should take into account even more complex vascular architectures, as well as acute hypoxia.