Monte Carlo techniques for patient specific verification of complex radiation therapy treatments including TBI, VMAT and SBRT lung

The main objective of this thesis is to develop Monte Carlo (MC) techniques for verification of complex radiation therapy treatments with emphasis on total body irradiation (TBI) and Volumetric modulated arc therapy (VMAT). This work was motivated by an initial study including ten non-small cell lung cancer (NSCLC) patients which evaluated the dosimetric consequences of plans optimized using the treatment planning system (TPS) by recalculating them with MC. It was shown that poor modelling of electronic disequilibrium by the TPS lead to underdosage of the planning target volume (PTV) and that quality assurance (QA) procedures should be based on a MC approach. With the emergence of volumetric modulated arc therapy (VMAT), which is a complex type of treatment delivery, new developments in MC simulations are required. A patient specific MC based QA system for VMAT treatments was developed and implemented clinically. This system is able to assess machine delivery performance and dose calculation accuracy of the TPS. During a substantial portion of the treatment the radiation beam is attenuated by the treatment couch. The impact of the attenuation on QA results is found to be patient specific and is non negligible. A process to create a couch model for MC simulations and its implementation is presented. The accuracy of this system is demonstrated against experimental measurements. For TBI the most important contributors to mortality is interstitial pneumonitis (IP). Adequate lung shielding and accurate estimation of lung doses is critical to reduce incidence of IP. A MC based TBI verification system including all the treatment delivery characteristics as well as patient specific lung compensators is presented. For the purposes of treatment plan quality improvement, a study including five anonymized image data sets from previously treated patients is performed. It is shown that mean doses to lungs are systematically larger for the prone position treatment compared to the supine position due to anatomical deformation. Improvement in dose distribution is investigated using a new fast inverse dose optimization algorithm combined with a new treatment delivery technique. This thesis concludes that MC based verification for complex radiotherapy treatments is clinically feasible and outperforms current methods.Science, Faculty ofPhysics and Astronomy, Department ofGraduat