ESTRO 2020 Abstract book

S1125 ESTRO 2020

the SC has been performed on the CT2. We collected maximum doses (Dmax) at BS and SC on both scanners. The decision to do a second plan treatment was made in case of exceeding the dose constraints in CT2. Results For the BS, we found an increase in Dmax in 75% of patients and exceeded constraints in only one case. The average was 0.75 ± 2.1Gy (p = 0.06). For SC, the increase in Dmax was statistically significant (0.85 ± 0.38 Gy,p = 0.05. This increase was observed in 75% of cases with an overload in one case. For the first case, after replanification, the maximum BS dose decreased from 62Gy to 55.93 Gy. For the second case, the maximum dose in the SC was reduced from 46 Gy to 42.3 Gy. These changes were made without altering the coverage at target volumes. Conclusion An increase in neurological structures’ doses was found at 38 Gy for patients undergoing IMRT for NPC. The generation of a new treatment plan at mid of course radiotherapy reduced the organs at risk doses. This dosimetric gain should be confirmed and validated by prospective trials. PO-1923 Intrafractional uncertainty of the heart position in DIBH and free breathing radiotherapy S. Fortin Jørgensen 1 , W. Ottoson 2 , C.F. Behrens 2 , G.F. Persson 3 , S.N. Bekke 2 1 Københavns Professionshøjskole, Radiography, Copenhagen, Denmark ; 2 Herlev Hospital- University of Copenhagen, Radiotherapy Research Unit, Herlev, Denmark ; 3 Herlev Hospital- University of Copenhagen, Department of Oncology, Herlev, Denmark Purpose or Objective Deep-Inspiration Breath-Hold (DIBH) radiotherapy is standard treatment for left sided breast cancer patients in order to minimize dose to the heart and lung. However, knowledge of the position uncertainty of the heart during radiotherapy in DIBH and free breathing (FB) is lacking. Current study estimates this uncertainty using data from patients treated with lung stereotactic radiotherapy (SBRT). Material and Methods This retrospective study includes 21 consecutive DIBH plans (16 patients, 5 with dual targets) and 12 randomly chosen FB plans delivered between April 2018 to March 2019. Treatments were delivered in 3 or 5 fractions with volumetric modulated arc therapy (VMAT) using 2-4 half arcs and patients were immobilized using an upper body vacuum bag, Vacfix. Daily cone beam CTs (CBCTs) were acquired before and in the middle of the treatment for target position verification. In order to estimate the intrafractional uncertainties, each CBCT was automatically registered to the planning CT using the intensity range [-1000; 250] with match volume of interest around the heart. Manual adjustments were made when necessary. The intrafractional uncertainties were assessed by calculating the match differences between the CBCT registrations on the heart, and statistics were applied (t- test and F-test). Results The motion of the heart ranged from -10.1 mm to 7.7 mm in DIBH, and from -3.5 mm to 5.1 mm in FB (Fig. 1). The variation of the heart motion was significantly increased in all directions for the patients treated in DIBH compared to the patients treated in FB ( p < 0.05). However, there was only a significantly different mean value of the motion between DIBH and FB in the longitudinal direction (DIBH: - 1.1 mm; FB: 0.2 mm, p = 0,01).

treatment observed. Ensuring patient compliance of the full bladder drinking protocol and daily ultrasound imaging before each fraction of radiotherapy is suggested to minimise changes in bladder volume and therefore reduce uterus and cervix movement. PO-1921 Comparison of dynamic tumor volumes in 4D- CT with two types of respiratory motion sensor S.H. Tsai 1 , Y. Huang 1 , Y. Chen 1 , P. Juang 1 , C. Kang 1 , C. Huang 1 1 Kaohsiung Chang Gung Memorial Hospital, Proton Therapy Center- Department of Radiation Oncology, Kaohsiung, Taiwan Purpose or Objective Our proton therapy system is equipped with Anzai Respiratory Gating System (AZ-733VI, Anzai Medical Co., Ltd., Tokyo, Japan), in which there are two types of respiratory motion sensor, a patient-non-contact type (Laser Sensor, LS) and a patient-contact type (Load Cell Sensor, LCS). To evaluate the consistency between these two types of sensor, we compared dynamic tumor volume deviations in four-dimensional computed tomography (4DCT) (Discovery CT590 RT, GE Healthcare, Chicago, IL, USA). Material and Methods We utilized Xsight Lung Tracking Phantom Kit (CIRS Model 18023-A, Computerized Imaging Reference Systems, Inc., Norfolk, VA, USA) to simulate respiratory motion. The sphere phantom of tumor is 2.5 cm in diameter. Five seconds of one respiratory cycle and 30 mm of respiratory motion were set. With two types of respiratory motion sensor, we acquired full amplitude of respiratory motion waveform data and images with 1.25 mm slice thickness from 4DCT using both 5 seconds and 6 seconds for each respiratory cycle scanning. The calculated dynamic tumor volumes of most useful phases for clinical gating, T40-T60 phases, from each type of sensor were compared. Results The calculated tumor volume in static CT scanning (7.26 cm3) was slightly smaller than its real volume (8.18 cm3). With 5 seconds for each respiratory cycle scanning in 4DCT, the tumor volumes of T40-T60 phases from LS were 7.38, 7.22 and 6.84 cm3, and those from LCS were 7.35, 7.19, 6.80 cm3. With 6 seconds scanning, the tumor volumes of T40-T60 phases from LS were 7.15, 7.26 and 7.39 cm3, and those from LCS were 5.84, 6.66, 6.80 cm3. There was no significant difference of the dynamic tumor volumes between two types of sensor (p=0.700 for 5 seconds scanning; p=0.100 for 6 seconds scanning). Conclusion Dynamic tumor volumes of T40-T60 phases have no significant difference between two types of respiratory motion sensor. The LCS is consistent with LS for clinical application. PO-1922 Dosimetric replanification benefit in exceeded neurological structures doses during IMRT for NPC M. Wafa 1 , H. Daoud 1 , F. Najla 1 , S. Tarak 1 , S. Wicem 1 , F. Leila 1 , D. Jamel 1 1 Hbib Bourguiba Hospital, radiotherapy, sfax, Tunisia Purpose or Objective To quantify the dose change observed at mid-treatment in neurological structures: brainstem (BS) and spinal cord (SC)and to evaluate the replanification dosimetric impact during intensity-modulated radiotherapy (IMRT) for nasopharyngeal carcinomas (NPC). Material and Methods Our study collected 20 patients undergoing IMRT treatment for NPC. For each patient, a second dosimetric (CT2) scan was performed at 38 Gy and was fused with the initial planning scanner (CT1). Manual contouring of the BS and CBCT position was