ESTRO 38 Abstract book

S504 ESTRO 38

was performed before each fraction. Plans were recalculated on the maximum inspiration and expiration phases of the 4D-CT series, and on CBCTs. Statistical significance was examined using a Wilcoxon signed rank- test for related samples and set at p≤0.05. The computations were carried out with the STATA 13.0. Results A total of 392 plans were calculated and analyzed. All plans fully accomplished the objectives. Example of dose distributions for the two approaches is reported in figure 1. A general higher variability between planners was found for INH (the highest differences for ITV-D50%). Excluding obvious results, HOM plans resulted in higher body V50% sparing (P=0.16), lower total MU (P<0.001), lower gradient index (GI=body V50%/V95%) with GI HOM =4.1±0.3 and GI INH =3.8±0.1 (P<0.001), lower PTV-D98% (95.1±0.3% vs 98% - P<0.001). In the 4D-CT series, CTV-D98% was relatively reduced of 2.1±0.3% and 9.3±3.5% for HOM and INH (P<0.001), however absolute CTV-D98% values were higher for INH. In the CBCT series, a similar behavior for both approaches was observed with patient and day-by- day variations.

processed using the RS script. Spatial and volumetric comparison of the resulting PTV was performed against the Reggui-generated one. Statistical significance was assessed using a paired t-test at a 0.05 significance level. Results Patient population consisted in 15 peripheral and 5 central lesions with average 3D motion vector of 10.2 (SD 6.6) mm. The spatial position of the gravity centre showed no statistical differences (<0.1mm) between the Reggui and RS-based PTV. While the script-generated target volume was systematically larger by an average value of 0.42 (0.85) cc (p<0.05), the GTV to PTV margin was found significantly larger only in the right-left and anterior- posterior direction by an average of 0.1 (0.1) mm and by 0.2 (0.5) mm in inferior-superior (Figure 1).

Conclusion MidP-like solution was successfully integrated into the treatment planning system using built-in scripting capabilities. Hence, workflow was simplified while user inputs and manual interventions were minimized. The development of a tumour deformation handling solution is currently ongoing. PO-0936 To be or not to be homogeneous in SBRT plans? a systematic multi-planning study P. Mancosu 1 , G. Reggiori 1 , A. Gaudino 1 , F. Lobefalo 1 , L. Paganini 1 , V. Palumbo 1 , A. Stravato 1 , S. Tomatis 1 , M. Scorsetti 1,2 1 Humanitas Research Hospital, Radiotherapy and Oncology, Rozzano Milan, Italy ; 2 Humanitas University, Biomedical Sciences, Rozzano Milano, Italy Purpose or Objective In stereotactic body radiation therapy (SBRT) delivered with modulated plans, two opposing approaches could be followed. ICRU 83 recommends to normalize modulated plans at mean dose to the target (homogeneous approach - HOM). AAPM WG 101 recommends to normalize SBRT plans at a specific isodose (Inhomogeneous approach - INH). The new ICRU91 for stereotactic treatments does not recommend an explicit normalization but it advises to report doses at near maximum (D2%), minimum (D98%) and mean dose to the target. In this study, a systematic study on the dosimetric effect of the two approaches was performed. Material and Methods Four patients were randomly selected by the internal database of candidates to lung SBRT. Internal Target Volume (ITV) was generated as the union of clinical target volumes (CTV) contoured on 4D-CT series. Planning target volume (PTV) was defined as ITV+5mm in the 3 directions. Volumetric modulated arc therapy (VMAT) plans were optimized by seven experts who performed two different plans on each patient: HOM (normalization at mean dose to PTV; objectives: CTV-V98%>98%, PTV-V95%>95%, PTV- V105%<5%, hotspots only inside ITV) and INH (normalization at PTV-V98%=98%; objectives: PTV-D2%< 150%, PTV-D2%>140%). Plans were all prescribed to 48 Gy in 4 fractions. Daily image guided radiotherapy by CBCT

Figure 1: Axial views for (Left) HOM and (Right) INH approaches for two planners. Colorwash ranging between 30-70Gy Conclusion The HOM approach resulted in more reproducible dose distribution between different planners and lower mean dose to the body. The INH approach, however, produced significantly higher mean doses to the target and steeper dose gradients. When defining a clinical protocol, physicians should be aware of advantages and drawbacks of the two approaches. PO-0937 Can butterfly VMAT in DIBH reduce dose of LAD in left breast cancer radiotherapy? M. Maffei 1 , S. Bou Selman 1 , S. Hofer 2 , M. Haller 2 , P. Ferrari 2 1 Azienda sanitaria dell'Alto Adige, Servizio di radioterapia oncologica, Bolzano, Italy ; 2 Azienda sanitaria dell'Alto Adige, Servizio di fisica sanitaria, Bolzano, Italy Purpose or Objective to assess the feasibility of a multi partial arc in deep inspiration breath hold to minimize dose of left anterior descending coronary artery (LAD) and compare dose between conformal 3D and VMAT both in free breathing and deep inspiration breath hold conditions. Material and Methods 21 patients underwent left breast cancer radiotherapy were considered, prescribed dose was 56 Gy in 28 fractions. A total of four treatment combinations were compared (84 plans): 3D-conformal free-breathing (FB), 3D-conformal deep inspiration breath hold (DIBH), 2 partial arcs (from arc angle 290° to 160° clockwise and counterclockwise) free-breathing VMAT (VMAT) and 4 partial arcs using butterfly technique (20 sec per arc, from arc angle 290° to 350°, 120° to 160°, both clockwise and counterclockwise), in order to warranty the treatment