ESTRO 37 Abstract book

S1103

ESTRO 37

intraabdominal target. A prospective study was carried out to compare the plan quality of volumetric modulated arc therapy (VMAT) with a linear accelerator (LINAC) to that of respiration-gated intensity modulated radiation therapy (IMRT) with magnetic-resonance (MR) image- This study was designed to test the null hypothesis that dose-volume parameter of LINAC and that of tri- 60 Co system plans are the same. The study was designed to test given hypothesis with alpha 0.01 and beta 0.1 using paired T-test. Target accrual was adjusted to incorporate the 20% dropout rate. LINAC plans were generated from 4D CT simulation images with abdominal compression and forced shallow breathing, whereas tri- 60 Co system plans were generated from breath hold fused CT simulation and MR simulation images. Competing plans from each set-up were generated and dose-volumetric parameters from both plans were obtained for analysis. Results A total of 102 patients were enrolled, and 27 patients were excluded due to the innate imbalance in a treatment plan. There were 32 patients with liver tumor (Group A), 21 patients with pancreas or extrahepatic bile duct cancer (Group B), and 22 patients with metastatic abdominal lymph node or adrenal gland tumor (Group C). There were 20 with LINAC plans and 12 with tri- 60 Co plans in Group A, 13 with LINAC plans and 8 with tri- 60 Co plans in Group B, and 14 with LINAC plans and 8 with tri- 60 Co plans in Group C. The average planning target volumes were significantly higher with LINAC vs. tri- 60 Co plans in all groups ( p < 0.001, < 0.001, and 0.001, respectively). In addition, mean doses to liver were increased with tri- 60 Co vs. LINAC ( p = 0.004, < 0.001, and < 0.001, respectively). In Group A, D 1cc of stomach and duodenum were higher in tri- 60 Co vs. LINAC, respectively ( p = 0.004 and 0.045, respectively). In Group B and C, however, D 1cc of stomach and duodenum were not significantly different between two plans. But, mean doses and D 1cc of kidney were significantly higher in tri- 60 Co plan (mean dose, p = 0.049 and 0.003; D 1cc , p = 0.003 and 0.01, respectively). Tri- 60 Co system was selected for patients whose GTV abutted hollow organ in overall patients. In Group B and C, patients with the tumor in contact with hollow organ and small respiratory motion (diaphragm movement < 7mm) showed a trend towards likely to be treated with tri- 60 Co system plan. Conclusion The quality of VMAT plans was superior to that of tri- 60 Co plans for the majority of patients undergoing radiotherapy for an abdominal target. However, respiratory gated image-guided radiotherapy with tri- 60 Co system offered benefits to patients with organs at risk abutted to target. Furthermore, in minority of patients, only MR guided respiration-gated radiotherapy met planning parameters and was thus clinically applicable. EP-2017 Assessing the dosimetric impact of intra- fractional prostate motion during SBRT. S. Novikov 1 , J. Melnik 1 , N. Ilin 1 , M. Girshovitch 1 , S. Kanaev 1 1 Prof. N.N. Petrov Research Institute of Oncology, Radiation oncology and nuclear medicine, St. Petersburg, Russian Federation Purpose or Objective To determine average and maximum values of prostate displacement during every fraction of stererotactic radiotherapy (SBRT) for prostate cancer. To analyze impact of this displacement on final dose distribution and to recommend PTV margins. Material and Methods Prostate intra-fractional motion was determined in 31 patients with prostate cancer that received 156 fractions of SBRT on Novalis TX system. Three reflective spheres guided tri- 60 Co system. Material and Methods

were fixed on the skin and 3 fiducial markers were implanted in the prostate of every patient. SBRT was performed in supine position without immobilization devices under Exac-Trac infrared real time control of patient position on the table. Pre-treatment target (prostate) positioning was performed by Cone beam CT (CBCT) and subsequent verification CBCT was done after the end of each fraction with recording the information about markers shifts. On the last stage on “Eclipse” 3D planning system we performed modelling of prostate displacement during every fraction and determined final “real” dose distribution in prostate and rectum after SBRT with various PTV margins. Results Results: According to postfraction CBCT average marker shifts in cranio-caudal direction was 0,8+/-1,2mm, in lateral - 0.9+/-1,0mm, in anterior-posterior - 1.3+/- 1,3mm. In 5% - 8,3 % cases value of cranio-caudal and lateral shifts were above 2 mm with maximal shift -5мм. Modelling of dose distribution in prostate with considering average intrafractional shifts indicate acceptable dose distribution in clinical target volume (prostate) in patients with PTV with 1mm posterior and 3 mm – anterior, lateral and cranio-caudal margins: V100prostate – 98%, D90prostate – 101%. At the same time reduction of “standard” (3mm- posterior, 5mm – in other directions) margins to above mentioned (1mm- posterior, 3 mm – other directions) can reduce rectum volume that covered by 80% isodose. Conclusion due to determined small intrafractional displacement of prostate during SBRT sessions recommended margins for PTV can be as follows: posterior – 1mm, anterior – 3mm, lateral – 3mm and cranio-caudal – 3mm. EP-2018 Components of prostate displacement during hypofractionated radiotherapy treatment M. Roch 1 , P. Castro 1 , A. Zapatero 2 , D. Hernández 1 , L. Perez 1 1 Hospital Universitario La Princesa, Radiophysics / Radiation Oncology, Madrid, Spain 2 Hospital Universitario La Princesa, Radiation Oncology, Madrid, Spain Purpose or Objective The first objective of this study was to determine the prostate intrafraction motion using pre- and post- treatment cone-beam CT (CBCT) images with three implanted gold seed fiducial markers. The second objective was to decompose this movement in two components: the patient displacement and the prostate gland motion from pelvic bony anatomy. Material and Methods The study cohort consisted of 15 prostate carcinoma patients that were treated with a hypofractionated radiotherapy protocol: 67.5 Gy in 25 fractions, 2.7 Gy per fraction, with a simultaneous boost irradiation of 69 Gy. Planning CT images were acquired in supine position and with a knee wedge and foot block for immobilization. Three fiducial gold markers were implanted into the gland under ultrasound guidance at least two weeks before the CT planning acquisition. The patients were reminded to follow an empty rectum and fill bladder preparation protocol before the TC acquisition and each fractional treatment. At each fraction a pre-treatment CBCT (CBCT pre-fiducial ) was acquired and a rigid 3D fiducial markers-registration to the planning CT was performed to an accurate positioning of the patient (Eq. 1). On alternate days, a post- treatment CBCT (CBCT post-fiducial ) was acquired with the patient still in the treatment position (Eq 2). Two more rigid 3D pelvic bony anatomy-registrations were performed off-line: one using the pre-treatment CBCT (CBCT pre-bones ) (Eq. 3) and another one using post - treatment CBCT (CBCT post-bones ) (Eq. 4).