SU-E-J-57: MRI-Linac (MRL) Guided Treatment for Esophageal Cancer.

ABSTRACT

For radiotherapy, oesophageal cancer is located in a difficult area. Spatial control of the dose distribution is difficult to achieve with current CT-based radiation techniques, as on CT, soft-tissue contrast is too low. Furthermore, the oesophagus moves and organs at risk (e.g. lung, heart, liver, spinal cord) are in close proximity. An 1.5 T MRI-accelerator (MRL) has sufficient soft-tissue tumour visualization possibilities to allow for precise real-time, online, position verification and for dose escalation without organ at riskoverdose. Our research consists of the preparatory work for the first clinical study on the MRL for patients with oesophageal cancer. To improve image quality and reduce the motion artefacts, the benefit of cardiac triggering and breath holds is evaluated on fifteen oesophageal patients. Results show the superb image quality of these MRI sequences. The use of this high quality MRI gives the possibility for non-invasive real-time visualization andtracking of the tumour. We quantify oesophageal tumour motion on cineMRI. The tumour is tracked on sequential mixed T1/T2w images (acquisition time 60s, temporal resolution 0.5s, slice thickness 7mm) of a single coronal and sagittal slice using a Minimum Output Sum of Squared Error (MOSSE) adaptive correlation filter. Tumour registration within the individual images can typically be done at a millisecond time scale. Motion of oesophageal tumours can well be tracked and is highly variable between patients. The greatest mobility is seen in cranio-caudal direction, with amaximum peak-to-peak amplitude of tumour movement of 24.5mm followed by the dorso-ventral and the medio-lateral direction. Movement seems greatest in tumours located in the lower part of the oesophagus. This study shows both the superb image quality for GTV localisation and the possibility for on-line and real time tumour tracking. The study opens thepossibility for tracked radiation delivery with a 1.5T MRI accelerator. Partial funding has been obtained by Elekta and Philips.