BLUE RIBBON POSTER THERAPY: Linear Energy Transfer Distributions In Dose-Optimized Pencil Beam Scanning Proton Therapy Delivered Using a Dynamic Collimation System
Abstract
Purpose
Dose-optimized proton pencil beam scanning (PBS) treatments that are collimated using the Dynamic Collimation System (DCS) have greater dosimetric sparing of healthy tissue compared to uncollimated treatment plans, but differences in linear energy transfer (LET) have yet to be studied. A Monte Carlo (MC) framework was developed to compare LET distributions between DCS-collimated and conventional uncollimated PBS proton treatment plans.
Methods
Three intracranial, three-field intensity modulated proton therapy (IMPT) treatment plans were generated to provide maximum tumor dose conformity for uncollimated and DCS-collimated treatments using the FDA-cleared Astroid (.decimal, Sanford, FL) treatment planning system. Treatment delivery was simulated within a custom DCS MC module developed using Geant4. Dose and dose-averaged LET (DLET) distributions were produced for each treatment plan with sufficient histories to achieve statistical dose uncertainties within 2% for voxels containing over 5% of the maximum dose within a 3 mm dose grid resolution.
Results
The gamma pass rate of all dose distribution comparisons between MC and the commissioned analytical dose calculator was greater than 94% for 3%/3 mm gamma criteria. Proximity of specific OARs to high-dose gradients impacted the distribution of DLET delivered to these structures, with 6 OARs showing higher DLET with the DCS than without (max +0.349 keV/µm, median +0.181 keV/µm), 2 OARs showing lower DLET with the DCS than without (min −0.873 keV/µm, median −0.452 keV/µm), and 4 OARs showing no significant change. For the 10 mm ring surrounding the PTV, DCS collimation reduced the mean dose by an average of 20.3% (+/- 5%) while increasing the mean DLET by an average of 9.1% (+/-5%).
Conclusion
DCS dose distributions optimized using a maximum conformity-based approach provided consistent dose reductions to OARs but did not consistently reduce LET. In many structures, LET increased when dose decreased, motivating LET-aware evaluation and multi-objective planning for DCS treatments.