ESTRO 2021 Abstract Book

S1611

ESTRO 2021

for PTV coverage based on the first set of patient plans. Materials and Methods

We analyse the plans of the first 13 SABR prostate patients treated on LINAC. The plans were prepared by experienced senior planners and independently checked by different senior planner. We create additional volumes of ´ Rectum_overlap , Bladder_overlap , Bowel_overlap ´ in ECLIPSE for every plan, ´_overlap´ meaning the overlap with PTV. We define an independent set of variables and one dependent variable so that we can make a regression analysis and obtain a regression model. Independent set of variables are represented by ´Rectum_overlap, Bladder_overlap, Bowel_overlap´ and PTV coverage is the dependent variable. We use the Data Analysis tool in Excel to create a multilinear regression model. Its statistical significance is described by R square value, standard error and p-value. Results We created the regression models from data of 10, 11, 12, 13 patients. These 4 models manifest their statistical significance by: R 2 of 0.88, 0.73, 0.65, 0.64, respectively; standard error of 1.1, 1.5, 20.3, 19.6, respectively; p-values of 0.004, 0.02, 0.03 and 0.02, respectively. Summarising the results, we are able to predict PTV coverage with 88% significance and standard deviation of 1.1 %, using multilinear model from the data of 10 patients. Conclusion We created a multilinear regression model that helps us to predict PTV coverage with the significance of 88% and standard deviation of 1.1 %. As soon as the planner finds out the absolute volumes of rectum-, bladder- and boweloverlaps with PTV, he will be able to get the estimated PTV coverage that helps him to recognise the proposed plan quality. PO-1893 The dosimetric effect of varying seroma during proton and photon therapy in breast cancer patients E.R. Skarsø 1 , L.H. Refsgaard 2 , L.B. Stick 1 , B.V. Offersen 3,4,1 , S.S. Korreman 1,3,5 1 Aarhus University Hospital, Danish Center for Particle Therapy, Aarhus, Denmark; 2 Aarhus University Hospital, Department of Experimental Clinical Oncology, Aarhus , Denmark; 3 Aarhus University Hospital, Department of Oncology, Aarhus, Denmark; 4 Aarhus University Hospital, Department of Experimental Clinical Oncology, Aarhus, Denmark; 5 Aarhus University, Department of Clinical Medicine, Aarhus, Denmark Purpose or Objective Novel research has shown that pencil beam scanning proton therapy can reduce the exposure of heart and lung compared to photon therapy in selected breast cancer (BC) patients. However, a shortcoming of proton therapy is its sensitivity to anatomical changes, and mastectomized BC patients are known to form seromas in the breast region. This study has investigated and compared the dosimetric effects of varying seroma during proton and photon therapy. Materials and Methods Nine women receiving post-mastectomy chest wall and nodal photon radiotherapy at our institution were included in this retrospective study. Photon treatment plans (TPs) and comparative proton TPs were created in Eclipse TPS (v13.6 Varian Medical Systems) according to the Danish Breast Cancer Group (DBCG) guidelines. All photon TPs were composed of two opposing tangential fields, two opposing periclavicular fields (AP-PA) and supplementary fields-in-fields. The proton plans were robustly optimized (±3.5% range and ±0.5cm setup error) using two enface beams (single-field optimization), with angles ranging between 5° - 15° and 34° - 45° and a 5cm range shifter. Seromas were simulated by adding structures to the CT scans of water equivalent material, in eight different locations. For each location, the seromas were raised above the skin by 3, 5, 10, 15 and 20mm and added to the predefined structures: CTVp_chestwall (CTVp) and body outline, see fig 2. The eight locations were created based on the CTVp, see fig. 1. The TPs were recalculated with constant MU. Target

coverage was compared for all plans, and the compliance according to DBCG guidelines defined as

= 47.5Gy.

Results

The photon TPs could encompass seromas with 20mm elevation with no noticeable influence on the

CTVp. For

proton TPs, even with a 3mm elevation, the

was influenced greatly, with a median

ranging from 37.9Gy

to 44.6Gy depending on location, and thus not complying with the DBCG guideline = 47.5Gy. Seromas located laterally influenced the CTVp, with a dose loss between 8.9% - 59.4% compared to no seroma, and thus the dose loss

correlated with the seroma location, see fig 2.

CTVp for proton and photon TPs can be seen in fig. 1