Reduction of respiratory pancreas motion using an MRI and proton therapy compatible abdominal corset

Objectives:

An MRI- and particle therapy compatible patient-individualized abdominal corset was developed and its efficacy to reduce respiratory-induced pancreas motion was evaluated by orthogonal 2D-cine and 4D-MRI.

Patients & Methods:

Nine patients (6 female; average age 72.9±9.6 years) with tumors of the pancreas (7), gallbladder (1), or liver (1) provided written informed consent to be scanned on a 3T MRI scanner (Philips Healthcare) by means of orthogonal 2D-cine and 4D-MRI, without and with a patient-individualized abdominal corset, respectively. A polyethylene (PE) abdominal corset (ORD Dresden GmbH) was customized based on an optical 3D-surface scan (Artec Eva®). For MR imaging the patients were positioned supine on a flat tabletop using an anterior coil holder to avoid compression of the chest wall. For 2D-cine MRI, coronal and sagittal slices were selected to image the pancreatic head. The 4D-MRI dataset was reconstructed by retrospectively resorting a multi-slice 2D acquisition into 10 respiratory phases. Pancreas motion was determined as center-of-mass displacement in three orthogonal directions (inferior-superior (IS), anterior-posterior (AP), left-right (LR)).

Results and Conclusion:

MRI revealed predominant motion in IS direction, which was reduced by 42% (p<0.05) in both 2D-cine (5.0±3.7 mm without corset; 3.8±1.1 mm with corset) and 4D-MRI (8.6±3.7 mm without corset; 5.0±2.4 mm with corset) through the corset which furthermore reduced the motion variability by 40% (p<0.01). The corset had no effect on motion in AP and LR direction. This confirms earlier results obtained with a custom-made polyurethane corset [1]. However, the homogeneous material properties of PE allow the corset to be integrated in PT without increasing range uncertainties. Pancreas motion in IS direction can be significantly reduced in patients scheduled for PT by an individualized abdominal corset.

[1] Heerkens et al, Phys Imaging Radiat Oncol 2017; 2:7–10.