Margriet Kwint

Chapter 1 14 rotation. The images made by the CBCT are registered to the images of the RT-planning CT based on anatomical structures (e.g. vertebrae, carina or the primary tumor). This registration can be used for tumor alignment, to observe anatomical changes and for dosimetric purposes (41). The accuracy of radiotherapy is affected by a diversity of geometrical uncertainties (e.g. set-up errors, baseline shifts and respiratory motion). The goal of IGRT is to increase this accuracy during a radiotherapy fraction (intra- fraction) and between different fractions (inter-fraction). The repetitive CBCT’s made us also aware of intra thoracic anatomical changes during treatment in lung cancer. In the Netherlands, CBCT’s are typically analyzed by radiation therapy technician (RTT); the radiation oncologist is informed in case a change is observed. With the increased use of daily CBCT imaging, there was a clinical need for a clear and practical decision support system to guide the RTTs in prioritizing the anatomical changes. Since 2012, daily CBCT’s with online position verification and correction are made for lung cancer patients treated with radical intent in the Netherlands Cancer Institute. Schaake et al. (42) demonstrated that the PTV margins can be reduced when this daily online CBCT position verification is used. Subsequently, a PTV margin reduction with expected decrease in toxicity was clinically introduced in 2015 in our institute. These daily CBCT’s also resulted in an increase of imaging data of tumor volume changes during treatment. The predictive value of tumor volume changes has been studied previously (43-45). These studies hint towards a predictive potential for OS, when there was tumor volume change during treatment, but the observed associations were inconsistent and the performed studies included few patients. Furthermore, the performed analyses could have been too simplistic. For example, observed GTV-changes during treatment were dichotomized below or above the median (43-45). The assumption that 2 groups (above/below median) represent the treatment response of all NSCLC patients might be a misconception (46). In order to identify various subgroups of patients with distinct treatment responses, more advanced statistical techniques, such as latent class mixed modelling, could be useful (47-50). Hence, more research is needed to analyze the predictive potential of tumor volume change during radiotherapy for treatment outcome. Acute esophagus toxicity The addition of chemotherapy concurrent with radiotherapy provokes a radiosensitizing effect leading to an improved local tumor control and OS, compared