Room: Karl Dean Ballroom A2
Purpose: This work aims to reduce the dose to organs at risks without compromising the target coverage by combining contour-scanning and energy-layer specific collimation for proton pencil beam scanning (PBS).
Methods: Four techniques are compared for a clinical PBS field in a skull-base tumour patient geometry: Spot positioning on a grid (grid-scanning), spot positioning following the contour of the target (contour-scanning), adding energy-layer specific collimation on the contour scanning plan without re-optimizing the pencil beam weights (collimation) and re-optimizing the contour scanning plan to take the collimation into account (optimized collimation). The optimization is performed by including an analytical model for the collimated pencil beams, which is based on the geometrical collimation setup and the properties of the un-collimated pencil beams, in the treatment planning system. Finally, all dose distributions are recalculated using the TOPAS Monte Carlo toolkit. A multi-leaf collimator is included in the Monte Carlo for the collimation and optimized collimation fields.
Results: Monte Carlo recalculated fields agree well with the analytical dose distributions. Compared to the conventional grid-scanning approach contour-scanning lowers the mean dose to the brainstem by 9.7% without compromising the dose to the PTV (agreement of mean dose within 0.3%). While collimation alone lowers the mean dose to the brainstem by 17.4%, cold spots in the target volume degrade the plan quality (2.3% lower mean dose to PTV). Re-optimizing the field to take collimation into account leads to a recovery of PTV coverage, while lowering the dose to the brainstem by 15.4% compared to grid-scanning alone.
Conclusion: Combining contour-scanning and energy-layer specific collimation substantially lowers the dose to the brainstem when compared to grid-scanning. To prevent cold spots in the target volume from degrading the treatment quality, it is critical to take the collimation into account during the optimization process.
Funding Support, Disclosures, and Conflict of Interest: This work was supported by a research grant from Varian Medical Systems Particle Therapy GmbH, Germany.