ESTRO 2024 - Abstract Book

S4059

Physics - Inter-fraction motion management and offline adaptive radiotherapy

ESTRO 2024

1863

Proffered Paper

HyperSight CBCT directly suitable for offline adaptive radiotherapy for prostate, but not for lung

Nienke D. Sijtsema, Joan J. Penninkhof, Agustinus J.A.J. van de Schoot, Britt Kunnen, Judith H. Sluijter, Maarten L.P. Dirkx, Steven F. Petit

Erasmus MC Cancer Institute, Department of Radiotherapy, Rotterdam, Netherlands

Purpose/Objective:

The recently introduced HyperSight CBCT (Varian) promises HU accuracy similar to conventional CT scanners, and an enlarged field-of-view to allow sufficiently accurate CBCT-based dose calculation needed for radiotherapy treatment planning. Additionally, the improved image quality should enable accurate delineation of the tumor and organs at risk. Consequently, additional CT imaging could be omitted for offline plan adaptation, resulting in less patient burden and shortening of the preparation time of adaptive plans. In this work, we evaluated the accuracy of dose calculation on HyperSight CBCT in prostate and lung cancer patients.

Material/Methods:

First, a commercially available electron density (ED) phantom (CIRS) was used to assess HyperSight CBCT-to-ED calibration for the available reconstruction algorithms (i.e. Acuros and metal artifact reduction (MAR)). Additionally, a dynamic thorax phantom (CIRS) with different tumor motion amplitudes and frequencies was used to simulate breathing motion, to assess the effect of breathing motion on HyperSight CBCT images. Second, a retrospective clinical study was performed to evaluate the dosimetric accuracy of HyperSight CBCT in patients. Twelve consecutive prostate cancer patients and nine consecutive lung cancer patients with written informed consent (IRB protocol MEC-2022-0815) were included. All HyperSight CBCTs were acquired in free-breathing. Due to the short CBCT scan time (6 s), only lung cancer patients with a tumor motion less than 10 mm and a breathing frequency of 8 breaths/min or larger were included to avoid a possible mismatch in average tumor position. For each patient, the HyperSight CBCT obtained at the first treatment fraction was rigidly matched (translations and rotations) to the planning CT, and subsequently transformed and resampled, using in-house software. Next, clinical treatment plans were recalculated on the transformed HyperSight CBCT in Eclipse using the Acuros 17.0.1 dose algorithm. Differences in dose distribution between the clinical plan on the planning CT and the recalculated plan on the HyperSight CBCT were quantified using 3D global gamma comparison, with either 1%/1mm or 2%/2mm as reference criteria, and a threshold of 10% of the prescribed dose. PTV D mean , V 95% and V 107% , were assessed for all plans. For prostate cancer patients, rectum V 100% (i.e. the relative volume receiving the prescribed dose), V 28Gy and V 38.4Gy (for a dose prescription of 7x6.1 Gy), or V 50Gy and V 56Gy (for a dose prescription of 20x3.0/3.1 Gy), anus D mean , bladder V 100% , and D max in the left and right femoral head were assessed. For lung cancer patients, lungs-primary GTV (Lungs-GTVp) D mean , V 20Gy and V 5Gy , D max in SpinalCord+3 mm planning risk volume (PRV), esophagus V 45Gy and heart D mean were assessed.