ESTRO 2021 Abstract Book

S1669

ESTRO 2021

rotational shift was obviously much lower in 4DCT in comparison to 3DCT simulated patients. Specifically, the longitudinal shift was almost 2mm less in the 4DCT simulation, which meant the 4DCT may improve the longitudinal positioning accuracy in the daily IGRT setup. One patient example of 4DCT and 3DCT matched with daily CBCT was shown in Fig 1.

Conclusion Compared with regular 3DCT simulation, the 4DCT simulation may effectively reduce the longitudinal setup error in daily IGRT for lung cancer radiotherapy. There were almost 2 mm setup difference in the longitudinal direction when 4DCT simulation was applied in our preliminary 12-patient study. Our future study will provide data analysis in more patients and more various immobilization devices. PO-1962 Impact of different immobilization systems on setup accuracy for radiotherapy of BCT vs. PMRT P. Yang 1 , Z. Peng 1 , D. Li 1 , X. Li 1 , Z. Zhang 1 , Y. Wang 1 , Y. Liu 1 1 Peking University Shenzhen Hospital, Radiation Oncology, shenzhen, China Purpose or Objective The positioning accuracy may affect tumor control and overall survival for patients undergoing breast radiotherapy. Hence it is essential to determine immobilization devices to reduce setup errors for individual patient. In this study, we aim to investigate setup errors of different immobilization devices in patients undergoing breast conserving therapy (BCT) and postmesectomy radiotherapy (PMRT), which will guide our future clinical practice to achieve better positioning accuracy. Materials and Methods Forty-two patients with breast cancer undergoing radiotherapy at Peking University Shenzhen Hospital were retrospectively selected for this study. Among them, there were 24 patients undergoing BCT and 18 patients undergoing PMRT. Patients were randomly assigned into two groups: group 1 (27 patients) with vaclok and group 2 (15 patients) with bracket. All patients underwent CT simulation (SIEMENS Definition AS OPEN) with slice thickness of 3mm and image resolution of 512x512. During the first three fractions and once a week afterwards, kV on-board imaging （ OBI ） of each patient was acquired prior to treatment on the linac (Trilogy, Varian Medical Systems, Palo Alto, USA) and registered with digitally reconstructed radiography (DRR) generated from the planning CT. The setup errors in the longitudinal, lateral and vertical directions and rotation were compared between group #1 (vaclok) and #2 (bracket). Subsequently, patient setup errors of BCT and PMRT were also analyzed. Results The setup errors in lateral (LAT), longitudinal (LNG) , vertical (VRT) directions and rotation (RTN) of the two group was shown in Table1. There was no significant difference in setup errors of LAT, LNG and VRT directions between the two immobilization devices (p ＞ 0.05). However, there was a significant difference observed in rotational setup (p ＜ 0.05). The setup errors of patient undergoing BCT and PMRT were shown in Table 2. There was no significant difference in all directions (P ＞ 0.05) in BCT patients either with vaclok or bracket. On the other hand, for pateints undergoing PMRT, the setup errors of using Vac-Lok or bracket in LAT, and VRT directions showed no significant difference. Howeve r, there was significant difference in LNG direction (P ＜ 0.05).