SU‐E‐J‐156: Preclinical Inverstigation of Dynamic Tumor Tracking Using Vero SBRT Linear Accelerator: Motion Phantom Dosimetry Study

Abstract

Purpose:Following the ‘end‐to‐end testing’ paradigm of Dynamic Target Tracking option in our Image‐Guided dedicated SBRT VeroTM linac, we verify the capability of the system to deliver planned dose to moving targets in the heterogeneous thorax phantom (CIRSTM). The system includes gimbaled C‐band linac head, robotic 6 degree of freedom couch and a tumor tracking method based on predictive modeling of target position using fluoroscopically tracked implanted markers and optically tracked infrared reflecting external markers.Methods:4DCT scan of the motion phantom with the VisicoilTM implanted marker in the close vicinity of the target was acquired, the ‘exhale’=most prevalent phase was used for planning (iPlan by BrainLabTM). Typical 3D conformal SBRT treatment plans aimed to deliver 6‐8Gy/fx to two types of targets: a)solid water‐equivalent target 3cm in diameter; b)single VisicoilTM marker inserted within lung equivalent material. The planning GTV/CTV‐to‐PTV margins were 2mm, the block margins were 3 mm. The dose calculated by MonteCarlo algorithm with 1% variance using option Dose‐to‐water was compared to the ion chamber (CC01 by IBA Dosimetry) measurements in case (a) and GafchromicTM EBT3 film measurements in case (b). During delivery, the target 6 motion patterns available as a standard on CIRSTM motion phantom were investigated: in case (a), the target was moving along the designated sine or cosine4 3D trajectory; in case (b), the inserted marker was moving sinusoidally in 1D.Results:The ion chamber measurements have shown the agreement with the planned dose within 1% under all the studied motion conditions. The film measurements show 98.1% agreement with the planar calculated dose (gamma criteria: 3%/3mm).Conclusion:We successfully verified the capability of the SBRT VeroTM linac to perform real‐time tumor tracking and accurate dose delivery to the target, based on predictive modeling of the correlation between implanted marker motion and external surrogate of breathing motion.