P240] Pre study for MRI linac: Interfractional anatomical and dosimetric changes based on CBCT images for rectal cancer

Abstract

Purpose The aim is to study anatomical changes and to predict the dosimetric consequences to evaluate the usefulness of magnetic resonance imaging (MRI) linear accelerator (Linac) for rectum cancer treatment. Methods 51 rectal cancer patients treated with preoperative radiotherapy were included in this retrospective study. Each patient had a plan computed tomography (pCT) and daily cone beam CT (CBCT). We evaluated the rectum interfraction changes by delineated the rectum, on daily CBCTs and on the pCT. Deformable image registration (DIR) was applied to the structures and pCT to match the weekly CBCT. The reference treatment plan calculated on the pCT was subsequently transferred to the weekly deformed CT, and the dose distribution was recalculated. Results For 10 patients, the quality of the pCT and CBCTs was not sufficient to delineate the rectum. For 32 patients, the rectal volume decreased during the course of treatment, for 4 patients the rectal volume was stable, and for 4 patients the rectal volume increased. The mean rectal volume for all patients decreased significantly between the pCT compared to the last CBCT (p = 0.0009). The sum (Boolean operation) of the rectal volume delineated on the CBCTs for a given patient was larger than the plan clinical target volume (CTV) but smaller than the planned treatment volume (PTV) for all patients. For 11 patients, CBCTs covered all the PTV volume. For these patients, the DIR was applied. Except for one patient, the CTV-Tumor was well covered at all treatment fractions. For 54% of the patients, the CTV-Elective minimum dose was less than 95% of the prescribed dose as is the clinical constraint. The organ at risk dose (OAR) was approximately constant during the treatment, except for the bladder. Conclusions For 20% of the patients, the quality of the pCT and CBCT is not sufficient to delineate the rectum. The tumor is not visible on any pCT or CBCT. We assumed that MRI Linac can be used to delineate the rectum and/or the tumor more precisely, making it possible to reduce the PTV margin. Except bladder and CTV-Elective, all the structures received similar dose at all treatment fractions.