ESTRO 38 Abstract book

S1026 ESTRO 38

including one who demonstrated significant breast swelling, the robust plans were compared to the corresponding non-robust plans by comparing the nominal dose distributions on the planning CT and by calculating daily delivered doses using cone beam CT (CBCT) images (n = 67) taken prior to each treatment delivery. Results No significant differences were observed between the robust and non-robust plans for the nominal dose distributions on the planning CT with both techniques able to meet all mandatory clinical goals taken from the HeartSpare Plus trial. Figure 1 illustrates the skin flash that can be achieved using organ-motion based robust optimisation.

clinical and AP plans mean rectal dose was 40.2 Gy and 36.8 Gy, mean bladder dose was 44.6 Gy and 44.1 Gy, and mean anal canal dose was 29.9 Gy and 27.7 Gy.

When considering the daily delivered doses, the robust plans demonstrated significantly better whole breast clinical target volume (CTV) coverage (Table 1) and this is particularly evident in the patient who demonstrated significant breast swelling. The robust plans demonstrated a significant advantage in covering the superficial part of the breast CTV with prescription dose. Delivered doses to organs at risk and IMC and Axilla CTVs were similar for the robust and non-robust plans (Table 1).

Figure 1. Mean dose volume histograms for prostate bed target volumes and organs at risk for clinical plans and RapidPlan plans. Conclusion Automated prostate bed VMAT planning with RapidPlan results in clinically acceptable and possibly favorable treatment plans from a DVH perspective. Dose distributions should also be thoroughly analyzed before clinical implementation. EP-1889 Evaluation of organ-motion based robust optimisation for RT of the breast, axilla, and IMC I. Blasiak-Wal 1 , A. Dunlop 1 , R. Colgan 1 , A. Ranger 2 , A. Kirby 2 1 The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, Radiotherapy Physics, Sutton, United Kingdom ; 2 ICR and Royal Marsden NHS Foundation Trust, Clinical Oncology, Sutton, United Kingdom Purpose or Objective Simultaneous irradiation of the internal mammary chain (IMC) alongside treating the whole breast and axilla nodes using VMAT (in breath-hold) is increasing. A concern of VMAT delivery for breast RT is the lack of skin flash that is utilised in tangent-based delivery to account for setup uncertainties and increases in the patient contour. We aimed to develop and validate a methodology using organ- motion based robust optimisation, to generate VMAT plans for simultaneous irradiation of the whole breast, IMC, and axilla nodes. Material and Methods Non-robust plans were initially generated in order to determine a patient-specific optimal objective function. In order to account for changes in position, size, and shape of the breast target tissue, organ-motion simulation available within a commercial treatment planning system (TPS) (RayStation 7.0, RaySearch Laboratories, Stockholm) was used to simulate CTs exhibiting worst-case breast swelling scenarios. Robustly-optimised plans were generated using the simulated CTs and the objective function from the non-robust plan. For five patients,

Conclusion Organ-motion-based robust optimisation available within a commercial TPS is able to generate plans that are clinically acceptable in the nominal (planning CT) scenario and are more robust, when compared to the non-robustly optimised plans, to both typical changes in patient shape and cases of significant breast swelling. EP-1890 Patient specific conversion of CBCT images for proton therapy treatment planning N. Krah 1 1 CNRS / University Lyon, Creatis lab UMR 5220, Lyon, France