SU-E-T-493: 4D Considerations for Active-Scanning Proton Beams

Abstract

Purpose: To optimize delivery parameters for mobile targets treated with active-scanning proton beams. Methods: An algorithm was developed to sort delivery of scanned proton plans into temporal bins. The binning reflects various dose rates and respiratory periods. A single-field proton plan was generated to treat a target within a phantom. The delivery of the plan was sorted into temporal bins, and the bins were displaced spatially to simulate residual motion within a respiratory gated delivery. The binned delivery was repeated multiple times with random starting points. Weighted combinations of the varied start points were formed to simulate repainting as a means of neutralizing intra-gate motion. This process was extended to a plan generated for a single phase of a 4DCT scan of a lung tumor. Intra-gate motion was simulated by binning the delivery and calculating portions on neighboring 4DCT phases. The dose matrices were deformably remapped back onto the planned 4DCT phase and the matrices were summed to yield the delivered dose. As with the phantom plan, various ranges of motion and repainting were simulated.Results: Simulated motion of a phantom tumor volume yields shallower DVH fall off than a static displacement of the same magnitude, although both hot and cold spots are observed. Simulated repaintings of the volume mitigated coverage inhomogeneities. Because of lower density tissue surrounding the lung tumor, simulated intra-gate motion results in a reduction of dose to the tumor. Conclusions: By varying the range of simulated motion and number of repaintings we will optimize duty cycle and delivery parameters for scanning proton treatment of a mobile tumor. J Kruse is supported in part by a grant from Varian Medical Systems, Inc.