ESTRO 2025 - Abstract Book

S3214

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

ESTRO 2025

868

Proffered Paper Dose-optimised MLC tracking for stereotactic arrhythmia radioablation: striving toward eliminating treatment margins Emily A Hewson 1 , Suzanne Lydiard 2,1 , Lars Mejnertsen 3,1 , Paul J Keall 1 1 Image X Institute, The University of Sydney, Sydney, Australia. 2 Kathleen Kilgour Centre, Tauranga Hospital, Tauranga, New Zealand. 3 Department of Engineering, SeeTreat Medical, Sydney, Australia Purpose/Objective: Stereotactic Arrhythmia Radioablation (STAR) is a promising, non-invasive treatment for cardiac arrythmias [1]. Accurate delivery is challenging due to complex cardiorespiratory motion and target proximity to critical structures. To enable safe treatment an effective motion adaptation strategy is crucial. Geometry-optimised MLC tracking for STAR has previously been investigated, [2] but some limitations in intrafraction adaptation capabilities were identified. The aim of this study was to investigate whether dose-optimised MLC tracking could improve STAR intrafraction adaptation accuracy and achieve the prescribed target dose without target volume expansions. Material/Methods: STAR treatment plans targeting the left and right pulmonary vein antra delivered on a linac equipped with a Millenium 120-leaf MLC were simulated using data output from treatment logs [2]. The impact on delivered dose using dose-optimised MLC tracking was investigated using three dual arc VMAT treatment plans with PTV margin expansions of 0 mm, 3 mm, and 5 mm, simulated using three human subject-measured motion traces. Each treatment delivery was simulated using the geometry-optimised MLC tracked apertures, and with apertures determined using dose-optimised MLC tracking (Figure 1). The dose-optimised MLC tracking method performed a fast, simplified dose accumulation in the cardiac frame of reference to calculate the ideal and delivered doses to anatomy undergoing intrafraction motion during treatment delivery [3]. The MLC leaf positions were optimised to minimise the difference between the ideal and delivered doses at each timepoint, constrained by the MLC leaf speed limit of 3.6 cm s -1 [4]. Doses accumulated using each treatment strategy were compared to the planned doses using dose metrics and a 3D gamma comparison.

Results: The mean differences in D100% compared to the plan for the PTV using dose-optimised MLC tracking were -2.5%, - 1.8%, and -1.9% for the 0 mm, 3 mm, and 5 mm margins respectively. Using geometry-optimised MLC tracking the