ESTRO 2024 - Abstract Book

S4408

Physics - Intra-fraction motion management and real-time adaptive radiotherapy

ESTRO 2024

3262

Digital Poster

Monitoring of PET-Avid OAR Moving in the Treatment Area During Biology-guided Radiotherapy Delivery

Murat Surucu 1 , Bin Han 1 , Harshali Bal 2 , Linxi Shi 2 , Shiyu Xu 2 , Yevgen Voronenko 2 , Girish Bal 2 , Jeffrey Schmall 2 , Nataliya Kovalchuk 1

1 Stanford University, Radiation Oncology, Stanford, USA. 2 Reflexion, Inc., Hayward, USA

Purpose/Objective:

Biology guided radiation therapy (BgRT) is a novel technology that uses limited time sampled PET reconstructed images to deliver external beam radiation to the targets. A biology tracking zone (BTZ) contour, defined during planning, generally demarcates the range of the expected target motion without containing any organ-at-risk (OAR). However, unexpected sudden patient movement or organ motion during treatment may result in the PET-avid OAR moving into the BTZ, resulting in accidental dose delivery to the OAR. In this work, the goal was to design an approach to monitor the PET counts within the BTZ, to detect gross movement of a PET-avid OAR into the BTZ as a secondary mitigation.

Material/Methods:

The proposed method is based on statistically assessing the PET image metric trend that serves as a surrogate marker for OAR movement with respect to BTZ. To evaluate the algorithm, a clinically realistic large anthropomorphic phantom (LAP) is designed, which is a large water filled phantom was designed with two articulating arms. Each arm was attached to a 26 mm sphere filled with FDG and the spheres were placed in the same transaxial plane. The posterior sphere was designated as the target with a 26x26x26 mm3 GTV and a BTZ of 51x51x45.6 mm3 while the anterior sphere was designated as the PET-avid OAR. The phantom was filled with FDG with a background concentration of 4.84 kBq/ml, while the target and OAR were filled at a ratio of 8:1 and 10:1 respectively. The OAR was moved posteriorly by 3.9 cm such that 50% of the OAR was inside the BTZ. The PET signal statistics emanating from within the BTZ as well as from a 6 mm shell around the BTZ were continuously measured to detect any changes, such as intrusion of the OAR into the BTZ. In BgRT, we use 4 passes where the patient couch is moved back and forth by the length of the BTZ in the superior-inferior direction during treatment with 2.1mm step motions, while spending up to 10 seconds at any beam station to deliver BgRT. To detect OAR motion during delivery, multiple metrics were used. For example, the system checks whether the total PET counts within the BTZ and the PET counts in a 6mm BTZ expansion vary significantly from the prescan PET. Further, it also verifies if the standard deviation of the counts within the BTZ, but outside the target, varies more than a certain threshold. For the evaluation of the PET metrics, two approaches were used to generate the accumulated treatment PET. In the first approach, the pre-scan PET data is progressively replaced by the currently measured beam-station PET data for analyzing the PET shape change index. Once the first pass is completed, the beam-station counts from the second pass replaces the PET data from the previous pass. The second approach uses the data from the last seven beam stations for the correlation and intensity indices analysis and compares it against the pre-scan PET data.