ESTRO 2021 Abstract Book

S16

ESTRO 2021

Purpose or Objective Changes during treatment in anatomy compared to the planning CT can result in reduced coverage of the target volume. This can be due to changes in motion or tissue deformations. With increasing use of Volumetric modulated Arc Therapy (VMAT) in breast cancer with nodal irradiation it is imperative to ensure that plans produced are robust to these changes. At our centre we have adopted a wide tangent (WT) technique which delivers open fields (2cm flash from skin) for ~70% of the prescribed dose and tangential Intensity Modulated Radiotherapy (IMRT) segments for the remainder without flash. As we transition to a VMAT solution for these patients, we sought to compare the robustness of our WT technique with 3 other VMAT flash approaches. Materials and Methods 10 (5 right, 5 left) WT deep inspiration breath-hold (DIBH) patients who had 40Gy/15 fractions to the chest wall and nodes including the internal mammary nodes (IMN) were retrospectively re-planned with 6MV dual partial arc VMAT. The replans consisted of a VMAT with no flash margin (VMAT NF) with a CTV-PTV margin of zero anteriorly and laterally, VMAT with manually applied flash (VMAT MF) on all allowable control points by 2cm from skin and a simulated organ motion robust optimisation plan (VMAT SOM RO) based on a CTVp motion of 1.5cm anteriorly and laterally. The evaluation of robustness was done by shifting the isocenter posteriorly by 5mm, recalculating on 5 deformed CT scans with deformed CTVp contours (SOM1-SOM5) as well as recalculating the plans on the daily CBCTs in the treatment position from the WT treatments. The CBCT daily doses were deformed back to the planning CT and summed. The SOM deformed CTs consisted of 5 scenarios with motion of the CTVp by 1.5cm laterally (SOM1), 1.5cm anteriorly (SOM2), 0.75cm laterally (SOM3), 0.75 anteriorly (SOM4) and 0.79 laterally with 1.28cm anteriorly (SOM5). SOM1, 2 and 5 were used during the VMAT SOM RO plans only. Coverage of the CTVp and a CTVp Superficial (CTVp within 1cm of skin) was assessed and dose to 0.5cc CTVp. Results 840 evaluation dose calculations were performed. VMAT NF clearly gave poor coverage of CTVp in the simulations and had its smallest change of -3.2% to CTVp V 38Gy on CBCT. VMAT SOM RO plans overall were the most robust and its difference to WT and VMAT MF became more apparent as larger anatomical changes were present. CTVp D0.5cc decreased on CBCT and increased highest for the VMAT MF plans (case max increase of 3.3Gy on SOM1) in simulations. The VMAT SOM RO plans were relatively stable with a max case increase of 1.2Gy on SOM1.