Exhibit Hall | Forum 3
Purpose: To investigate a method of using a narrow virtual bolus on both the medial and the lateral sides of the breast, forcing the optimization algorithm to create skin flash.
Methods: Five left breast patients with coverage of regional nodes including IM, SCV, and axillary nodes were used. Two narrow virtual boluses of 1.5cm thickness and 1.5cm width were created on the medial and lateral edge of left breast. A narrow virtual PTV of 0.5cm wide and 2.0cm thick was created at the center of the bolus: 0.5cm inside the body and 1.5cm inside bolus. Gantry angle from 300° to 179° was used in the VMAT plan. Virtual bolus was used in optimization. Besides breast and nodes prescribed to the prescription dose of 50Gy in 25 fractions, additional constraints were added to the virtual PTV in optimization to achieve skin flash. PTVbreastEval cropped 0.3cm from skin was the true PTV. The optimized plan was normalized to the mean dose of PTVbreastEval. VMAT plans using this technique were compared to conventional VMAT plans using 1.5cm thickness uniform bolus covering the whole breast target and a virtual PTV extending beyond skin by 1.5cm for optimization.
Results: Both methods achieved excellent skin flash. However, in the conventional VMAT plan using uniform thickness bolus, the coverage of PTVbreastEval degraded significantly after virtual bolus removal: V90% reduced from 96.8% to 94.8%; and V95% from 94.8% to 88.4% on average. However, the two narrow virtual bolus method was more robust to keep the coverage of PTVbreastEval: V90 was reduced from 96.7% to 96.4% and V95% from 93.7% to 92.4% on average, after virtual bolus removal.
Conclusion: Using the two narrow virtual bolus on the medial and lateral edge of breast for VMAT breast plan is a better way in dosimetry to create skin flash.
Not Applicable / None Entered.
Not Applicable / None Entered.