Margriet Kwint

Safety and efficacy of reduced dose and margins 29 2 the CT compared to the baseline CT. The gross tumour volume (GTV) of the primary tumour as well as the GTV of the lymph nodes were both expanded to a planning target volume (PTV). Subsequently, the PTV-margins were individualized according to the peak-to-peak respiratory amplitude movement of the tumour and lymph nodes. The margins in all directions from GTV to PTV consisted of 12 mm plus ¼ of the peak-to-peak amplitude in orthogonal directions as measured in the 4D-CT [18]. An isotropic PTV margin of 12 mm was used for the lymph nodes. The PTV-margins were adapted from June 2015 when a bony anatomy based correction strategy was replaced with a carina based correction strategy on the CBCT [15]. Reduction of the PTV-margins was applicable to all directions with a maximum in the cranio-caudal direction of 3.8mm for the lymph nodes and 2.1mm for the primary tumour. From then on, the PTV-margins varied between 9-11 mm. The following organs at risk (OAR) were delineated according to our institutional protocol: heart, spinal cord, lungs and oesophagus. For this study, the heart (sub)structures were delineated automatically using an automatic segmentation method [10]. The planning constraints were: oesophagus D max ≤66 Gy and V 50Gy ≤50% (EQD2 10 ), MLD ≤20 Gy (EQD2 3 ), spinal cord ≤52 Gy (EQD2 2 ), total heart ≤40 Gy and ⅔ of the heart ≤50 Gy and ⅓ of the heart ≤66 Gy (EQD2 3 ). IMRT-plans were calculated using 10 MV photons. Dose distributions were calculated using collapsed cone inhomogeneity corrections (Pinnacle versions 9.2-9.10, Philips, Best, The Netherlands). The dose inhomogeneity within the PTV was between the 90% and 115%. Treatment verification was done using daily cone beam CT-scans (CBCT) according to an on-line setup correction protocol and correction was performed immediately before the start of every fraction. Replanning was done in case of significant changes of the anatomy with an anticipated clinically relevant influence on the dose distribution [19]. Toxicity and follow-up (FU) Toxicity was scored using the Common Toxicity Criteria for Adverse Events version 4.0. Acute toxicity was scored from start of RT until 3 months after the last fraction. Late toxicity was calculated from 3 months after the last fraction. A CT-thorax was performed 6-8 weeks after treatment to evaluate the treatment response. Subsequently, FU was performed every 3-6 months with chest X-ray or CT-thorax. After 2 years, FU was performed every 6 months.