Abstract

PurposeTo study the feasibility and the effectiveness of a novel implementation of robust optimization on 2 sets of computed tomography (CT) data simultaneously for skin flashing in intensity modulated radiation therapy for breast cancer. Method and MaterialsFive patients who received treatment to the breast and regional lymphatics were selected for this study. For each patient, 3 plans were generated using 3 different skin-flashing methods, including (1) a manual flash plan with optimization on the nominal planning target volume (PTV) not extending beyond the skin that required manually postplanning the opening of the multi-leaf collimator and jaw to obtain flash; (2) an expanded PTV plan with optimization on an expanded PTV that included the target in the air beyond the skin; and (3) a robust-optimized (RO) plan using robust optimization that simultaneously optimizes on the nominal CT data set and a simulated geometry error CT data set. The feasibility and the effectiveness of the robust optimization approach was investigated by comparing it with the 2 other methods. The robustness of the plan against target position variations was studied by simulating 0-, 5-, 10-, and 15-mm geometry errors. ResultsThe RO plans were the only ones able to meet acceptable criteria for all patients in both the nominal and simulated geometry error scenarios. The expanded PTV plans developed major deviation on the maximum dose to the PTV for 1 patient. For the manual flash plans, every patient developed major deviation either on 95% of the dose to the PTV or the maximum dose to the PTV in the simulated geometry error scenarios. The RO plan demonstrated the best robustness against the target position variation among the 3 methods of skin flashing. The doses to the lung and heart were comparable for all 3 planning techniques. ConclusionUsing robust optimization for skin flash in breast intensity modulated radiation therapy planning is feasible. Further investigation is warranted to confirm the clinical effectiveness of this novel approach.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call