Optimal PTV Margin Expansions Along Six Anatomic Directions in Oropharyngeal IGRT
A Yock1,2*, L Zhang1, L Court1, A Garden1, L Dong3, (1) UT MD Anderson Cancer Center, Houston, TX (2) UT Graduate School of Biomedical Sciences, Houston, TX (3) Scripps Proton Therapy Center, San Diego, CATH-A-BRA-6 Thursday 8:00:00 AM - 9:55:00 AM Room: Ballroom A
To determine the optimal PTV expansions along six anatomic directions for a population of patients receiving CT-guided IMRT of the oropharynx.
For 20 patients, daily-acquired CT-on-rails images used for patient alignment to the C2 vertebral body were deformably registered to the patients' planning CT. The positions of volume elements originally within the CTV (target voxels) or within a 1cm shell surrounding the CTV (normal tissue voxels) on the patients' planning CT were determined on the daily CTs. The geometric target coverage and included normal tissue were the proportion of target voxels and normal tissue voxels, respectively, that were within a margin for a particular proportion of treatment fractions.
Geometric target coverage was evaluated for 15,625 margins created in MatLab representing the 56 combinations of 1, 2, 3, 4, and 5mm expansions along the posterior, anterior, lateral, medial, inferior, and superior directions.
The optimal margin minimized the normal tissue values at 10, 50, and 90% of treatment fractions while providing adequate geometric target coverage for at least 90% of patients. Four target coverage criteria were selected requiring at least 90 or 95% of target voxels be covered during at least 90 or 95% of treatment fractions.
Optimization according to each geometric target coverage criterion yielded a unique, anisotropic margin. Margins ranged from 1-5, 3-5,1-3, 4-5, 2-5, and 3-5mm in the posterior, anterior, lateral, medial, inferior, and superior directions, respectively. These optimal margins included 9.8 to 44.8% less normal tissue than isotropic margins while providing equivalent target coverage.
Due to characteristics of non-rigid setup errors and anatomical changes during the course of treatment, the optimal margin for oropharyngeal patients was anisotropic. Anisotropic margin expansions could reduce normal tissue exposure without compromising geometric target coverage.