In radiotherapy treatment, the outcome is conditioned by the intrinsic radiosensitivity of the tumor cells; one of the major players in determining the radiosensitivity is the local oxygen partial pressure, which is strictly dependent on the vasculature, usually underdeveloped in tumour tissues. Tumour hypoxia plays a crucial role in this context, as tissues with poor oxygenation present a resistance to radiation up to three times higher with respect to oxygenated tissues; indeed, the oxygen has the property to make permanent the indirect damages produced by the free radicals released by the interaction of radiation with cells molecules. The aim of this research project is to develop in silico tumour models, which will include in a first step the tumour intrinsic radiosensitivity, mainly characterized by the oxygen diffusion from the blood vessels; in this framework several vascular geometries will be assessed and validated against empirical data through a finite difference method; critical in this context will be the size of the domain, as having a resolution of 5-10 micrometers will require the computation of tens of millions of voxels, in order to model tumours of several centimeters in size. Further steps in this project will include other parameters, such as the density of clonogenic cells or the rate of their metabolic activity. Such models will allow fast comparisons between different treatment schedules varying in total dose, fractionation schemes, eventual local boosts, with the key function of keeping into consideration the local microenvironment and radiation resistance.