Clinical Validation of Proton Pencil Beam Scanning on an Anthropomorphic Lung Phantom

Abstract

To validate the use of proton pencil beam scanning for the treatment of moving lung tumors. Pencil beam scanning (PBS) proton plans are generated for the standard thorax phantom designed by Imaging and Radiation Oncology Core (IROC). Two planning techniques, namely robust optimization against range and setup uncertainties, as well as layer and volumetric repainting, are used for these PBS plans. Patient specific QA measurements are performed using both a water tank and a custom heterogeneous phantom. Tumor motion in this anthropomorphic thorax phantom is 2 cm in magnitude. A custom movable insert is used with the moving IROC thorax phantom to find the optimal number of volumetric repainting. Both pencil beam convolution (PBC) and Monte Carlo (MC) algorithms are used for dose calculation and their accuracy are compared with actual measurements. Single ionization chamber, two-dimensional ionization chamber array, thermoluminescent dosimeter (TLD) and films are utilized to check the point dose and dose distribution accuracy. The optimal number of volumetric repainting is found to be 4 times in our system. The mean dose overestimation on moving target by PBC and MC algorithms are found to be 5.1% and 2.0%, respectively. The mean gamma passing rate for PBC and MC algorithms are 91% and 96% respectively. MC dose algorithm calculation is found to have a good agreement with measurements. Moving lung tumors can be treated using proton pencil beam scanning with robust optimization, volumetric repainting, and Monte Carlo dose calculation.