ESTRO 36 Abstract Book

S887 ESTRO 36 2017 _______________________________________________________________________________________________

posture at the linac is slightly in better agreement with the posture at the CT for absolute couch positioning. On the ML images, a residual deviation in lung wall position of 5 mm or more was present for 5.1% of the fractions for relative couch positioning and for 2.7% for absolute couch positioning. This also implies a slightly improved patient setup using absolute couch positioning.

and 0.81cm AP for the supraclavicular region and 0.84cm LR, 0.89cm CC and 0.98cm AP for axilla region. Conclusion For the introduction of a VMAT planning technique for breast and axilla irradiation specific PTV margins adapted for supraclavicular and axilla inter-fraction motion need to be introduced. 1 Topolnjak, et al, IJROBP, Volume 78, Issue 4, 15 November 2010, Pages 1235–1243 EP-1657 Clinical use of transit dosimetry to analyze inter-fraction motion errors F. Ebrahimi Tazehmahalleh 1 , C. Moustakis 1 , U. Haverkamp 1 , H.T. Eich 1 1 Universitaetsklinikum Muenster, Klinik fuer Strahlentherapie, Münster, Germany Purpose or Objective The aim of this work was to inquire the correlation between the target and organ at risks motions and volume discrepancy with the dosimetric variations at hospital levels. The high resolution, large active area, and effectiveness of the Electronic portal imaging devices offers it to be used for in vivo dosimetry during radiation therapy as an additional dose delivery check. The transit dosimetry has the potential of testifying dose delivery, the accuracy of MLC leaf positioning and the calculation of dose to a patient or phantom. Material and Methods In total 42 patients with stereotactic plans were evaluated. Delivery was carried out on a Varian TrueBeam linac equipped with an aS1000 EPID. Continuous portal imaging was performed at each treatment fraction during the delivery of treatment for all beams. To validate the method, we executed treatment plans on a commercial respiratory motion phantom containing plastic spheres as target. Phantom CT scans were made in different phases. First phase were done by applying sinusoidal breathing cycle in different motion amplitudes (-20, -10, 0, 10, 20 mm) in superior/inferior direction and second phase was done by pre-defined breathing simulation with a short pause after exhalation in oscillation mode. Three techniques: 3D-CRT, IMRT and VMAT-SBRT were generated and on board transit dose was collected by EPID during the treatment. The daily obtained portal image were compared with the reference image using the gamma evaluation method with criterion 2% dose difference and 2 mm distance to agreement (DTA) criteria with a threshold value of 5% of maximum value. Results The area gamma passing rate per arc in most of the plans was higher than the acceptable limit but in some arcs it had lower agreement, the lowest value was 3.7%. Besides irradiating phantom in planned respiratory motion, we re- irradiated the same plans due to displacement of the target by stopping the movement or changing the breathing speed. Gamma parameters such as maximum gamma, average gamma, and percentage of the field area with a gamma value>1.0 were analyzed. For all the VMAT arcs in phantom measurements, the gamma evaluations were within the tolerance limits (γmax = 3.5, γavg = 0.5 and γ% >1 = 2%) tough in some measurement 20 mm target displacement was applied. For IMRT fields, measurements were not in good agreement in different tumor motion. 3DCRT fields showed poorest gamma agreement in portal dosimetry analysis. Conclusion This research increases the need of a tool for monitoring inter-fraction errors by confirming the tumor position within the treatment field over the course of therapy. Using daily EPID images over the course of treatment could potentially provide accurate verification of dose delivery to heterogeneous anatomical regions in patients receiving 3D-CRT and IMRT radiation therapy treatments. However, further studies are required to assess 3D IN VIVO dose

2.

Conclusion The patient posture as well as patient setup for breast cancer patients is slightly improved using the predicted, or absolute, couch position. 1 W.J. de Kruijf, R.J. Martens, Reducing patient posture variability using the predicted couch position. Med. Dosim, 40:218-21; 2015. EP-1656 The inter-fraction variation of the supraclavicular- and the axilla-area in breast cancer patients S. Gerrets 1 , L. Kroon- van der 1 , M. Buijs 1 , P. Remeijer 1 1 Netherlands Cancer Institute Antoni van Leeuwenhoek Hospital, Radiotherapy, Amsterdam, The Netherlands Purpose or Objective Currently a volumetric modulated arc technique (VMAT) for whole breast irradiation, including the supraclavicular- and axilla-area is being implemented at our institute. In contrast to our currently used tangential fields, VMAT requires a CTV and PTV with corresponding margins. From our clinical experience we know that the setup of the shoulder can be very challenging..The purpose of this study was to quantify the inter-fractional variation of the supraclavicular- and the axilla-area in order to quantify So far 6 right sided and 6 left sided breast cancer patients, were randomly selected in this ongoing study. Patients were positioned on a Macromedics MBLXI breastboard with upper- and lower- arm trays. During the acquisition of the planning CT skin marks were drawn extended to the humerus to improve reproducibility of the arm positioning. Setup verification and correction was performed based on bony anatomy registration (ribs and sternum) using Cone beam CT and an offline shrinking action level (SAL) protocol. Retrospectively, the residual inter-fraction errors of the supraclavicular area and the axilla were measured by performing bony anatomy registrations using a rectangular region of interest representative for these areas/regions (see Figure 1 'Region of interest”), and determining their difference from the registration on ribs and sternum. From these residual errors, the random and systematic errors were computed and corrected for the use of a SAL protocol (N=3 and α=9mm). Using previously determined setup data from Topolnjak et al [1], Subsequently, the CTV to PTV margins were determined according to the standard margin recipe: 2.5∑+0.7σ. Results In total 88 Cone beam CT were analyzed; 5-10 scans per patient. Computed residual errors for the supraclavicular region and axilla region are shown in Table 1.The random and systematic residual errors for the axilla regions are larger than the supraclavicular region, as expected. Notable is the small residual error for supraclavicular in LR-direction. The total margins are 0.59cm LR, 0.76cm CC CTV to PTV margins. Material and Methods