Radiobiological calculation for the treatment of patients in the field of FLASH radiotherapy

My thesis work concerns the study of FLASH radiotherapy. Recent in vivo results have shown prominent tissue sparing effect of radiotherapy with ultra-high dose rates (FLASH) compared to conventional dose rates (CONV). To nowadays several hypotheses are being investigated to correctly explain the FLASH effect. It has been suggested, among the most relevant theories, that this apparently beneficial effect is due to differences in oxygenation in tumor and healthy tissues. The differential response between FLASH-RT and CONV-RT may be due to radiochemical oxygen depletion at very high doses and the resulting radioresistance conferred on the irradiated tissue. It is largely accepted that hypoxic tissues are more radioresistant than well-oxygenated tissues. This is because in the presence of molecular oxygen there is a fixation of radiation-induced indirect DNA damage. To this end, several models on the kinetics of oxygen during irradiation have been realized to develop a time-dependent model on the behaviour of oxygen. This model aims to analyse, in terms of dose and dose-rate, the oxygen enhancement because of the therapy. At this point, it becomes apparent how important it is to mechanistically model the effect of oxygen in the water radiolysis process to fully understand how FLASH radiation therapy works. However, due to the computational costs of many-body interaction, oxygen is often ignored in simulations with common microscopic Monte Carlo tools. To make the MC code even more versatile and apply it to Oxygen Enhancement Ratio (OER) studies, it is necessary to include more types of molecules other than free radicals generated by the initial radiation in the chemical phase simulation. However, due to the computational complexity of the "many-body" problem and the long-time duration of the chemical phase, a step-by-step simulation of these relevant processes on conventional CPU computational platforms can be time-consuming. Under the constraint of computational resources, studies typically suffer from a narrow simulation region or short time duration, limiting their broad applications. To overcome these obstacles, Graphical Processing Unit (GPU)-based parallel computing can be a cost-effective option. An example is an open-source, GPU-based microscopic MC simulation toolkit, gMicroMC. This toolkit represents the tool used during my thesis work to perform MC simulations. So, the final purpose is to evaluate the response of FLASH therapy to the variation of significant parameters at physical and biological.