Studying pancreas motion to improve pancreas radiotherapy planning techniques

Abstract

Pancreatic radiotherapy treatment represents a huge challenge, in terms of delineating the treatment margins and choosing the best method for immobilization. This problem mainly results from the pancreas motion. Various studies have examined pancreatic tumor motion, without examining the movement patterns associated with the healthy pancreas. Similarly, existing studies have focused on the comparison of various immobilization methods, with little focus on studying the effects of patient positioning. Therefore, this study was performed to increase our understanding of normal pancreas motion patterns and to identify new methods for the delineation of treatment margins. In addition, this study was designed to increase our understand of how different positions and surfaces may affect treatment outcomes. To fulfil these aims, two healthy participants were scanned using different magnetic resonance imaging (MRI) methods. The first participant was scanned normally, on two separate days, whereas the second participant was scanned on the same day, in three different circumstances: on a normal MRI table (Normal), on an MRI table without pads (Pads), and on a Mediboard positioning table (Mediboard). For each surface, the participant was scanned in four positions: supine, prone, left oblique, and right oblique. To obtain motion details, ImageJ [25] software was used to contour the pancreas and obtain measurements. The motion was quantified based on centroid measurements, and the pancreatic margins were delineated automatically, using ImageJ, based on the area measurements. The results showed that mean pancreatic movement values for the first participant, on day 1, were 3.60 mm, in the anterior-posterior direction, and 0.92 mm, in the lateral-medial direction. On the second day of measurements, the mean values for the first participant were 1.95 mm, in the anterior-posterior direction, and 0.20 mm, in the lateral-medial, direction. The calculated clinical target volume (CTV) and planning target volume (PTV) margins for both days 1 and 2 were 10 mm and 5mm, respectively, in all directions. In the second participant, the mean pancreas motion values, in all directions, for the supine, prone, left oblique, and right oblique positions were 5.43 mm, 5.3 mm, 6.66 mm, and 7.12 mm, respectively, for the Normal condition; 10.28 mm, 21.84 mm, 9.12 mm, and 5.4 mm for the Pads condition; and 4.66 mm, 4.39 mm, 7 mm, and 7.3 mm for the Mediboard condition. These results indicated that the movement patterns of the healthy pancreas can differ from day to day, even in the same patient. Moreover, delineating custom treatment margins can spare the critical organs from exposed to excessive radiation and provide solutions for various treatment problems. Furthermore, positioning the patient differently may provide an advantage in terms of reducing both the pancreas size and motion. Although the present study reports important findings that may benefit researchers in the field of pancreatic radiotherapy, additional studies remain necessary to determine the optimal method for the treatment of pancreatic tumours.