Assessing dose variance from immobilization devices in VMAT head and neck treatment planning: A retrospective case...

Immobilization devices serve several purposes in radiation oncology including improved positional reproducibility and decreased patient movement. If immobilization device beam attenuation is unaccounted for in the treatment planning process, planning target volume (PTV) coverage can be compromised. A retrospective case study was performed to evaluate beam attenuation from head and neck (HN) immobilization devices in volumetric-modulated arc therapy using the Eclipse treatment planning system and the Anisotropic Analytical Algorithm. Three facilities retrospectively evaluated 15 patients with various HN carcinomas. A test plan was created from each patient's existing plan in Eclipse to include immobilization devices in the dose calculation. Plan comparisons were performed using dose-volume histograms, which assessed the maximum dose, percentage of the PTV that received 95% of the prescription dose (V₉₅), and dose covering 100% of the PTV (D₁₀₀). The test plan normalization values were then adjusted to match the original PTV coverage. The maximum dose difference calculated ranged from 0.3% to 4.1%. The absolute difference in V₉₅ coverage between the HN and the test plans ranged from 0.4% to 2.3%. The difference in the D₁₀₀ was more statistically significant, reporting a range of 1.2% to 71.4%. Finally, to achieve the same PTV coverage as observed in the HN plan, an adjustment to the normalization value was needed, which required a change ranging from 0.1 to 7.7. Attenuation from immobilization devices, when unaccounted for, can cause a misrepresentation of the dose delivered to the PTV. It is not a standardized practice among institutions to include immobilization devices within the body contour for dose calculations. However, with the increasing use of volumetric-modulated arc therapy for treatment of the HN region, precision is of the utmost importance, and inclusion of immobilization devices should be standardized.