ESTRO 2021 Abstract Book

S333

ESTRO 2021

previous protocol of 40 Gy in 15 daily fractions for 3 consecutive weeks. In the current study, we compare treatment uncertainties encountered in clinical routine in two cohorts of patients treated with ultrahypofractionated and standard hypofractionated RT schedule using an automated system for electronic portal imaging device (EPID)-based 2D in vivo dosimetry (IVD). Materials and Methods Data of the first 165 patients treated with ultrahypofractionated RT after BCS (i.e., 5fx-group) were retrospectively retrieved from the database (SunCHECK server). The control group consists of 165 consecutive patients treated with hypofractionated RT in 3 weeks (i.e., 15fx-group) after BCS one year before. EPID-based integrated transit dose images were acquired in the 5fx-group for each treatment fraction and in the 15fx-group on the first 3 days of treatment and weekly thereafter or more if a measured fraction did not pass the tolerance level (i.e., failed fraction; FF). Results were analyzed using a local gamma analysis with a threshold of 20%, a dose difference of 7%, a distance to agreement of 6 mm and a passing level of 90%. Moreover, the influence of the image-guided RT (IGRT) protocol was investigated: all patients in the 5fx-group had daily online IGRT with correction before treatment delivery, which was only performed in part of the 15fx-group according to our “no-action-level” (NAL) protocol. Results A total of 825 measured fractions (MFs) of IVD was acquired in the 5fx-group compared to a total of 944 MFs of IVD in the 15fx-group, of which 621 have been treated with online IGRT with correction before treatment delivery (i.e., IGRT 15fx-group). In the 15fx-group 13.7% of fractions did not pass the tolerance level compared to 3.9% in the 5fx-group. For the IGRT 15fx-group this was 9.5%. Reported causes for these FFs were patient positioning (5.9%), technical issues (5.8%) and breast swelling (1.9%) in the 15fx-group compared to 1.6%, 2.1% and 0.2% respectively in the 5fx-group and 2.6%, 5.6% and 1.3% respectively in the IGRT 15fx-group. There were no statistically significant differences in patient positioning (95% CI=[- 11.09,4.83], P=0.43) and breast swelling (95% CI=[-0.02,0.08], P=0.28) between the 5fx-group and IGRT 15fx-group, suggesting that the daily IGRT protocol might be the main reason that results are superior in the 5fx-group. Conclusion Two dimensional EPID-based IVD demonstrated that ultrahypofractionated RT for early-stage breast cancer results in less failed fractions as compared to hypofractionated RT given in 3 weeks by decreasing patient positioning, breast swelling and technical issues-related errors.

PH-0433 DIBH SABR for abdominal tumors while monitoring with fiducial markers and triggered images K. Kisivan 1 , C. Glavak 1 , Z. Laszlo 1 , M. Petone Csima 2 , E. Somogyine Ezer 3 , K. Mahr 4 , A. Gulyban 5 , Z. Kaposztas 6 , F. Lakosi 1 1 Moritz Kaposi General Hospital, Dr. Jozsef Baka Center, Department of Radiation Oncology, Kaposvar, Hungary; 2 Szent Istvan University, Kaposvar Campus, Faculty of Pedagogy, Kaposvar, Hungary; 3 Moritz Kaposi General Hospital, Department of Oncology, Kaposvar, Hungary; 4 Zala County Hospital Szent Rafael, Department of Oncology, Zalaegerszeg, Hungary; 5 Institut Jules Bordet, Medical Physics Department, Bruxelles, Belgium; 6 Moritz Kaposi General Hospital, Department of Surgery, Kaposvar, Hungary Purpose or Objective To test the feasibility of linac-based stereotactic ablative radiotherapy (SABR) using deep inspiration breath- hold (DIBH) with target monitoring using triggered imaging (TI) and auto-beam hold (ABH). Materials and Methods Six patients were treated with VMAT-based SABR (5x6.6/8 Gy, 3x15 Gy) for pancreatic cancer (4) and liver tumors (2). The patients were positioned before treatment using DIBH-CBCT (pre-CBCT). During treatment triggered images were acquired every three seconds and peritumoral surgical clips (5) or gold markers (1) were used for intrafractional tumor visualization. In case of ≥3mm (deviation limit: 3 mm, PTV margin 5 mm) displacements, treatment delivery was interrupted, and the geometric deviations were corrected using additional imaging (2D/3D match or CBCT) or a gating threshold adjustment. A post-RT DIBH CBCT (post-CBCT) was also acquired for visualization of 3D residual errors and dose adaptation. Beam interruptions and treatment time were recorded. The correlation between treatment time and number of corrections was estimated. CBCT image quality for OARs was evaluated on a scale of 1 to 4 (1: inadequate, 2: doubtful, 3: acceptable, 4: excellent). The Krippendorff's alpha coefficient (α ≥ 0.8) was used to compare agreement between opinions of radiation therapists (RTT) about image quality for OARs and to determine their