BEST IN PHYSICS (THERAPY): Dose Painting: A Fast Treatment Planning System for Dual-Robot Kilovoltage Radiotherapy
Abstract
Purpose
To develop a fast TPS for the dual-robot Kilovoltage Optimized Adaptive Radiotherapy (KOALA) system utilizing a 225-kVp x-ray beam and iris collimation.
Methods
A TPS was developed to optimize dose delivery with an x-ray tube mounted on a 6-degree-of-freedom robotic arm by means of dose painting enabled by isocenter shifts. First, a uniformly distributed set of treatment beam angles with a source-to-isocenter distance (SID) of 40-cm was selected, determined by the location of the planning target volume (PTV) and constrained by the range of motion of the robotic arm. At each treatment angle, a subset of overlapping circular beams with different isocenters optimally covering the 2D projection of the PTV was identified and doses for each treatment beam analytically calculated using the linear Boltzmann transport equation (LBTE). A treatment plan was optimized by sequentially selecting treatment angles that minimized a cost function.
Results
KOALA treatment plans were successfully generated for lung, osteosarcoma and glioma patients in 5-15 minutes. Unlike KOALA single-isocenter plans, dose-painted plans successfully conformed PTV dose to non-spherical tumors. For the lung patient with a 66.2 cm3 PTV volume (D95%=49Gy), the maximum PTV dose was 68, 72, and 64 Gy in the single-isocenter, dose-painted and VMAT plans, respectively. The mean dose to the spinal cord, liver and body was by 35.5, 194, 286% higher in the KOALA dose-painted plan than in the 6MV VMAT plan. However, the mean dose to right lung, heart and esophagus was by 19.7, 74.7, 15.7% lower in the KOALA dose-painted plan than in the VMAT plan.
Conclusion
These results demonstrate the successful performance of the KOALA TPS. The process is sufficiently fast for adaptive radiotherapy applications. However, increasing the effective energy of the treatment beam should be explored to further reduce integral dose and dose to bone.