Modelling and simulation of [18F]fluoromisonidazole dynamics based on histology-derived microvessel maps

摘要: Hypoxia can be assessed non-invasively by positron emission tomography (PET) using radiotracers such as [(18)F]fluoromisonidazole (Fmiso) accumulating in poorly oxygenated cells. Typical features of dynamic Fmiso PET data are high signal variability the first hour after tracer administration and slow formation a consistent contrast. The purpose this study is to investigate whether these characteristics explained current conception underlying microscopic processes identify fundamental effects. This achieved modelling simulating tissue oxygenation dynamics on scale. In simulations, vessel structures histology-derived maps act sources sinks for oxygen well molecules. Molecular distributions extravascular space determined reaction-diffusion equations, which solved numerically two-dimensional finite element method. Simulated time activity curves (TACs), though not directly comparable TACs, reproduce major clinical curves, indicating that model parameter values adequate. Evidence dependence early vascular fraction found. Further, possible effects leading late contrast potential implications quantification discussed.