Abstract
Lung cancer radiation therapy (RT) is associated with complex geometrical uncertainties, such as respiratory motion, differential baseline shifts between primary tumor and involved lymph nodes, and anatomical changes due to treatment response. Generous safety margins required to account for these uncertainties limit the potential of dose escalation to improve treatment outcome. Four dimensional inverse planning incorporating pretreatment patient-specific respiratory motion information into the treatment plan already improves treatment plan quality. More importantly, repetitive imaging during treatment quantifies patient-specific intrafraction, interfraction, and progressive geometrical variations. These patient-specific parameters subsequently can drive adaptive plan modification correcting for systematic errors while incorporating random errors. Adaptive RT therefore has the potential to considerably improve the accuracy of RT, reducing the exposure of organs at risk, facilitating safe dose escalation, and improving local control as well as overall survival.