ESTRO 37 Abstract book

S1071

ESTRO 37

Conclusion This study confirms that retrospective automatic sorting of 4D-MR images acquired during free breathing is possible by using a ROI placed over the diaphragm on navigator slices as internal surrogates. The method is not sensitive to the ROI position as inspiration phases were reproducibly detected and respiratory cycle time intervals were automatically defined. Integration into treatment planning system was also possible in order to picture the dynamic behavior of the liver (and so the lesion). EP-1967 VMAT technique on left sided breast focusing on the dose to the heart A. Stravato 1 , V. Palumbo 1 , A. Fogliata 1 , G. Reggiori 1 , P. Mancosu 1 , L. Paganini 1 , F. De Rose 1 , D. Franceschini 1 , S. Tomatis 1 , M. Scorsetti 1 1 Istituto Clinico Humanitas, Department of radiotherapy and radiosurgery, Rozzano Milan, Italy Purpose or Objective This study focuses on the planning of patients with left sided breast cancer. Different volumetric modulated arc (VMAT RapidArc, RA) designs were considered in order to evaluate the dosimetric trade-offs for a clinically acceptable plan, trying to find the optimal balance between heart sparing and target coverage and conformity. Material and Methods Ten patients with left-sided breast cancer treated with VMAT technique were included in the hypofractionated clinical protocol. The daily dose prescription was 3.2 Gy to the boost region in simultaneous integrated boost (SIB) technique and 2.7 Gy to the whole breast for 15 fractions. The patients were simulated in a supine position with the arm abducted (90° or greater) on the disease side. Treatment plans were generated for a 6MV Varian UNIQUE linac, equipped with standard Millennium MLC with 120 leaves. Three different RA designs were optimized for each patient (Fig.1): "2F" : a basic two partial arcs were arranged covering the entire partial arc trajectory with total rotation of about 2x 220° (Fig.1A) "Flip" : each arc from the 2F plan was splitted in two "sub" arcs optimizing for each one the rotation of the collimator for heart dose reduction (Fig.1B) "Flip_Avoid" : the "Flip" plan was re-considered with an anterior avoidance sector of about 50-60 degree to preserve the heart on a couple of fields (Fig.1C) Plan optimization objectives were stringent on dose homogeneity, mean dose to the heart <4Gy, ipsilateral lung <8Gy, contralateral lung <3Gy, contralateral breast <3Gy. An evaluation in terms of DHV comparison was performed with an homemade software based on MatLab. Mean doses, high doses to organs at risk were considered. Homogeneity and conformity indexes were calculated. Results All plans achieved at least 95% of PTV receiving 95% of dose, with V 105% less than 5%. The main differences regarded the low dose distribution. Mean heart doses were significantly lower for the "Flip_Avoid" plans ( p <0.005) when compared to the "2F" plans. No significant differences were seen for the controlateral lung while an improvement was achieved for the V 5Gy and V 20Gy for the ipsilateral lung. The role of the avoidance and the collimator projection is crucial for the plan optimization and dose heart reduction. Conclusion The main differences among plans concerned the doses to the organs at risk, still maintaining an high dose coverage and conformity. "Flip" and "Flip_avoid" plans presented a reduction of the mean doses for all critical structures:

respiratory cycle is registered. Therefore the dynamic behavior and the organ deformation are not captured. This study aims at presenting a novel retrospective gating approach for dynamic MR imaging during free breathing. Material and Methods 4D-MR liver images are acquired on a 1.5T MR scan using an 18-channel body flex coil. 4D-MRI acquisition is performed on healthy volunteers who gave their informed consent. This innovative sequence is based on a modified bSSFP sequence and consists of an interleaved acquisition of 2D image slices and navigators. The navigator is set and acquired at a fixed position in sagittal orientation, while axial image position is changing in order to cover the entire volume of interest. 4D-MR images are acquired over 40 slices with a 2.5mm thickness and a 0.44ms slice- time-resolution. The sequence is repeated 20 times to cover the entire respiratory cycle. Audio-coaching in the MR scan was provided and volunteers were asked to breathe at a constant period of 6 sec. Total acquisition time is less than 12 min. For 4D-MR images reconstruction, the period of the respiratory cycle is divided in 6 bins: 0%, 16%, 33%, 50%, 66% and 83%. This strategy is copied from 4D-CT, in order to prepare consistent registration for accurate lesion contouring in the application cases of liver SBRT. Motion position is given by the mean intensity in a rectangular region of interest (ROI) placed at the level of the diaphragm on the sagittal navigator (Fig. 1).

Results Mean intensity time course within the ROI is able to identify automatically the inspiration position corresponding to the phase 0%. Acquisition time interval between two consecutive inspirations defines each experimental respiratory cycle for the 20 repeated MR acquisitions. 2D MR image slices are automatically binned according to the 6 respiratory phases and stacked in 3D volumes (Fig. 2). As the ROI is defined by the user, 5 ROI voluntarily positioned differently, but still at the level of the diaphragm, were tested. For all ROI, mean respiratory cycles is 5.68 sec [1.76-7.92] and 94.48% of the inspiration positions are found each time for the same navigator slices. The volunteer for whom these results are given recognizes that her respiration was not always synchronized with the audio-coaching.