ESTRO 36 Abstract Book

S894 ESTRO 36 _______________________________________________________________________________________________

Results In initial plans, the mean calculated dose in parotid glands ranged from 24.1 to 26.2 Gy. There is significant variation in parotid contouring. The degree of variation varied from patient to patient, with maximum differences up to 23% in mean dose to parotid glands. Maximum differences in mean dose to parotid glands due to the uncorrected setup- shifts was up to 12%. Conclusion Intraobserver variability in contouring of OARs and daily variations in patient setup are significant contributors to uncertainty in radiotherapy treatment planning, and consequently in delivered dose. Our analysis indicates that the not-precise contouring can lead to larger difference between delivered and calculated dose. EP-1647 Validation of a set up procedure for IMRT/VMAT breast treatment using in vivo dosimetry with EPID S. Kang 1 , J. Li 1 , P. WANG 1 , X. Liao 1 , M. Xiao 1 , B. Tang 1 , X. Xin 1 , L.C. Orlandini 1 1 Sichuan Cancer Hospital, Radiation Oncology, Chengdu, China Purpose or Objective In vivo dosimetry (IVD) is an important tool able to verify the accuracy of the treatment delivered and its reproducibility. The change of a consolidated existing procedure can be performed if an indisputable evidence of improvement may be proved and the feasibility in a clinical workflow is guarantee. In this work IVD performed with electronic portal imaging device (EPID) was used to evaluate VMAT and IMRT breast treatment performed with a new set up and immobilization procedure. Material and Methods IVD with EPID was performed over 32 patients that underwent an IMRT or VMAT breast plus supraclavicular treatment. Half of the patients followed the standard set up procedure (SP) of the department, consisting of a thermoplastic mask covering the district to be treated, patient marks over the mask, bolus applied over the mask; the others followed a new procedure (NP) and were immobilized with a breast board and a knee support, patient marks over the skin and bolus applied over the skin. The accuracy of the treatment was evaluated with a commercial software (SOFDISO, Best Medical Italy) that provided two indexes: the ratio R between the reconstructed (D iso ) and planned (D tps ) isocenter dose (R=D iso /D tps ) which can represent the accuracy of the dose delivered, and a Pγ% obtained performing a gamma analysis between the first EPID image and the next ones acquired immobilized. Three consecutives tests were scheduled during the first week of treatment and successively two IVD test per week. The MLC log files of the treatments delivered where analysed with a commercial software and compared with the planned treatment in order to discriminate the deviation coming from the patient (anatomy and set up), from the deviation coming from the linac.

Results

Only the IVD test coming from a delivery with the machines log file in tolerance were considered. The results of 545 IVD tests obtained over 32 patients were reported in Table 1. Every treatment performed with IMRT and VMAT resulted in 100% of the patients with R and Pγ indexes in tolerance as for SP as for NP. The percentage of Pγ index in tolerance as for VMAT as for IMRT increased with NP. A 10% of, off tolerance tests persisted. The IVD tests off tolerance were reported in the acceptable threshold before the next fraction. Discussion: the new immobilization procedure enabled a direct localization of the patient skin and of the bolus positioned over it. The use of the mould mask, positioned over the patient’s lead to a non-direct evaluation of the patient rotation and accommodation inside it. The beam can lack of reproducibility if considering its path: air gap between the bolus, the mask and the patient skin not considered in the treatment planning. This aspect is moreover important for IMRT treatment where for some beam entry this situation can be more evident. Conclusion IVD in is a powerful tool that can be helpful in the validation of new set up and immobilization procedures. EP-1648 Thermoplastic mask dependency with interfractional uncertainties for head and neck VMAT treatments E.M. Ambroa Rey 1 , R. Gómez Pardos 1 , D. Navarro Giménez 1 , A. Ramirez Muñoz 1 , M. Colomer Truyols 1 1 Consorci Sanitari de Terrassa, Medical Physics Unit- Radiation Oncology Department, Terrassa, Spain Purpose or Objective Volumetric-modulated arc therapy (VMAT) techniques have the ability to deliver a highly conformal dose distribution to the target and high dose gradient at the interface between the tumor and the normal tissues, decreasing the irradiated volume and sparing OARs. However inaccurate alignment of the radiation beam with the patient can lead to critical organs to receive an unwanted high dose or the tumor to receive a reduced dose producing a loss in tumor control. Radiation therapy for head-and-neck (H&N) cancer requires a reliable immobilization for an accurate treatment. The purpose of this study is to establish the interfractional setup error for VMAT H&N patients, using a kilovoltage cone beam CT (CBCT) and a robotic treatment couch (HexaPOD) for accurate patient positioning in six degrees of freedom and analyze the differences between two types of thermoplastic masks (Head mask (H) and Head and Shoulder (HS) mask).