ESTRO 2024 - Abstract Book

S5518

RTT - Patient care, preparation, immobilisation and IGRT verification protocols

ESTRO 2024

Comparing different institutional setups for prostate cancer in proton therapy

Sin Ting Chiu, York Sze Wong, Yuen Yan Chan, Man Chun Lee

Hong Kong Sanatorium & Hospital, Department of Radiotherapy, Hong Kong, Hong Kong

Purpose/Objective:

Patient immobilization has always been a key concern for precise radiotherapy daily treatment. We strive to minimize all uncertainties to achieve narrow target margins with escalated target dose. In proton therapy, not only the stability and reproducibility should be considered, but also its impact on the dose and range uncertainty are factors that need to be weighed up. In proton therapy, opposing lateral beam arrangement is often used to reduce the bladder and rectal doses. However, the proton beams directly pass through the dense bones of the upper femurs and femoral heads before reaching the treatment target, i.e., the prostate. Thus, the displacement of femoral heads may increase the dose and range uncertainty in proton therapy treatment. In this retrospective study, we aim to determine the most appropriate setup with the least setup offset for prostate proton therapy treatments.

Material/Methods:

Forty-six patients who underwent prostate radiotherapy at Hong Kong Sanatorium and Hospital in 2022 were retrospectively recruited in this study. These patients were divided into three streams based on their treatment techniques: Stream 1–in-house foam leg separator for patients who had VMAT technique; Stream 2- Vaclok® with in house foam leg separator for patients who had Tomotherapy; and Stream 3 - indexed knee and foot support adopted in Unity for patients who had MR-LINAC treatment. Five pre-treatment verification images, including CBCT, MVCT, kVCT, and T2 weighted MRI for each patient, were rigidly registered with planning CT in MIM using the box-based assisted alignment for both femurs. The group mean, systematic error, random error of translational and rotational directions were used to evaluate the overall alignment quality between various immobilization devices.

Results:

A total of 230 daily pre-treatment images from 46 patients (14 from stream 1, 12 from stream 2 and 20 from stream 3) were retrospectively reviewed and analyzed. Table 1 illustrated the average deviations of translational directions (Right-Left (R-L), Superior-Inferior (S-I), Anterior-Posterior (A-P)), total vector error (TVE) and rotational directions (pitch, roll and yaw). Systematic error (Σ) and random error(σ) were calculated for all directions and summarized in Table 1.