ESTRO 2020 Abstract book

S1108 ESTRO 2020

kV CBCT are equipped in TomoTherapy and linear accelerator respectively. Recently, the latest generation of TomoTherapy named Radixact ® X9 has been introduced to Hong Kong. However, the comparison between the novel MVCT and kV CBCT has not been conducted. The purpose of this study was to compare the verification accuracy, feasibility to be used for adaptive radiation therapy and imaging dose between MVCT and kV CBCT. Material and Methods Planning computed tomography (pCT) images of a respiratory motion phantom (QUASAR™) were acquired with and without motion. Known deviations from the pCT reference position were introduced and drawn onto the phantom in three translational and three rotational directions. The assigned deviated positions were automatically verified with MVCT, 3D kV CBCT and Symmetry TM 4D kV CBCT images. The discrepancies between the detected and actual positional shifts (Absolute Mean Detection Errors (δD )) were analyzed statistically. CATPHAN ® 700 was used to evaluate image quality while anthropomorphic phantom was utilized to compare the calculated dose agreement to pCT based on a nasopharyngeal treatment plan between two verification images. MOSFET dosimeters were placed in the anthropomorphic phantom to measure the dose to selected depths by the two verification systems.

Results Due to minor anatomical inconsistency between pMR and pCT mostly due to bowel and breathing motion, the mean dose of bowel D max among all the 10 subjects resulted in a significant difference of 1.1 Gy (Z=-2.07, p=0.038). The mean PTV target dose difference between the plans was 0.2 Gy (Z=-0.65, p=0.515) in D 95 coverage, 3.7% (Z=-0.39, p=0.694) in V 105 and 0.2 Gy (Z=-0.35, p=0.730) in D max , showing no statistically significant change between the two plans. The paired samples t-test between CT plan and CT actual showed that there was significant average PTV dose difference of 1.2 Gy in D 95 (t 40 =8.84, p<0.001), 4.6% in V 105 (t 40 =4.83, p<0.001) and 0.4 Gy (t 40 =3.28, p=0.002) in D max . In comparison with MR-MR matching, MR actual resulted in a target delivered dose that was better in agreement with its planned dose compared with kVCT-MVCT matching. The mean PTV dose difference of D 95 , V 105 and D max were 0.6 Gy (t 40 =5.55, p<0.001), 3.6% (t 40 =3.40, p=0.002) and 0.4 Gy (t 40 =3.540, p=0.001) respectively. In terms of OAR sparing, CT actual had a reduced average dose of 0.3 Gy (t 9 =2.31, p=0.046) from CT plan in bowel D mean . Whereas in MR actual delivered dose, rectum D max and D mean showed a slightly reduced dose difference of 0.7 Gy (t 8 =-3.29, p=0.011) and 0.6 Gy (t 8 =-3.32, p=0.011) respectively.

Conclusion In this study, MR-based planning and MR-based verification have been proven to be feasible where CT was not needed for electron density information and image verification procedure for pelvis HT. The planned dose effect has shown to be insignificant. In terms of daily positional verification, we can facilitate more precise delivery with better target coverage if MR images could be used. PO‐1890 Comparative assessment of the verification performance, ART and imaging dose between MVCT and kVCBCT Y.S. Tang 1 , A.M. Chow 2 1 Hong Kong Baptist Hospital, Charles Cheng Oncology Centre, Hong Kong, Hong Kong SAR China ; 2 The University of Hong Kong, Department of Diagnostic Radiology, Hong Kong, Hong Kong SAR China Purpose or Objective Megavoltage computed tomography (MVCT) and kilovoltage cone-beam computed tomography (kV CBCT) are two commonly used volumetric verification modalities across radiation therapy centres in Hong Kong. MVCT and

Results Regardless of motion, verification by 3D or 4D kV CBCT showed significantly smaller mean δDs than MVCT in all directions except yaw rotation. The image quality of both verification modalities was not optimal enough for adaptive radiation therapy (ART) use. Various dosimetric parameters and dose-volume histograms (DVHs) showed that the doses calculated with MVCT and kV CBCT were overestimated and underestimated respectively as compared to that with pCT. Imaging dose was found to depend on the scanning protocol or pitch-reconstruction interval combination selected. For MVCT, dose increased with depth since MV radiation is more penetrating than kV.