SU-F-T-331: The Effect of Optimization Grid Size On Helical Tomotherapy for Nasopharyngeal Carcinama




To investigate the effect of plan optimization grid size on helical tomotherapy for nasopharyngeal carcinoma (NPC)


Ten patients of NPC were enrolled. For each patient, three plans with different dose optimization grid size (fine, normal, coarse) were created by Helical Tomotherapy Planning Station (version: 5.0, GPU Acceleration and VoLo).The time consumed by 500 optimization iterations, the monitor units and the dose volume histogram factors were compared. Paired-Wilcoxon test was used in statistical analysis.


The iterations time, monitor units and plan quality have significant difference from various optimization grid size: the mean iterations time of fine optimization grid size(138.89min) is 3.02 times, 6.29times to normal (45.99min) and coarse (22.11min) respectively; The monitor units of fine(6811.90MU) is 1.25% (6896.90MU, p=0.285) and 6.28% (7239.90MU, p=0.005)lower than normal and coarse; the mean homogeneity index (HI) of fine is 9.32% and 68.25% better than normal (p=0.075) and coarse (p=0.005); the conformity index (CI) of fine is 5.86% and 16.90% better than normal(p=0.005) and coarse (p=0.005). Compared to normal and coarse, fine optimization grid size reduced D2%,D98% of PTV by 0.47Gy, 3.43Gy and 0.11Gy, 0.23Gy respectively. Larger grid size caused a mean dose increase to the structure of parotids. There are no significant difference between normal and fine for the maximum dose of spinal cord, brainstem and optic nerves, but both fine and normal have obvious advantages than coarse. Three optimization grid size have no significant difference on the maximum dose of lens and eyes.


In NPC helical tomotherapy planning, larger optimization grid size required higher MUs to achieve the same target coverage. Fine optimization grid size is recommended while normal will improve calculation efficiency. Considering the disadvantage of coarse optimization grid size, it is not recommended to be used.