ESTRO 2025 - Abstract Book

S3243

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

ESTRO 2025

Purpose/Objective: Cervix and pelvic nodes cancer patients can be treated with cone beam computed tomography (CBCT)-guided online adaptive radiotherapy (oART) using a Varian Ethos linac. Although treatments are adapted to account for inter-fractional changes, planning target volume (PTV) margins are still required due to intra-fractional motion. This work derived uterine PTV margins from intra-fractional displacement and applied them to clinical treatments, in a pilot study, to assess their routine viability. Material/Methods: Workflow for the oART treatment sessions is shown in figure 1, with the uterus (CTV1) contoured separately each fraction (Shelley 2023). Intra-fractional displacement was assessed by measuring the uterine displacement between the first CBCT scan and the second verification CBCT, acquired just before treatment. Isotropic population margins were derived from the 90 th percentile of uterine intra-fractional displacement, for 19 patients treated with 459 fractions, with 2mm additional delineation uncertainty.

Derived margins were then used for a small cohort of pilot clinical treatments. However, a non-adaptive plan, with larger margins, was available for all fractions if uterine displacement was too large. Target coverage was assessed for the pilot treatments by measuring the largest intra-fractional uterine displacement in the six cardinal directions for each fraction, before comparison with the size of the added margin. Fractions where uterine displacement was greater than the margin size were dosimetrically assessed. The bladder and CTV1 were delineated on the verification CBCT, and registered to the dose cube exported from the Ethos treatment session to assess dose coverage. Results: Population uterine margins of 8mm were derived from intra-fractional displacements (see figure 2). Pilot treatments had CTV1 encompassed by the PTV in 95% of fractions treated. CTV1 displacement for patients treated in the pilot study varied from 4.2-9.1mm. For all fractions with CTV1 displacement beyond the PTV, CTV1 D99%>95% and Dmean>99% of the prescribed dose, which falls within the requirements of the local protocol. CTV1 was displaced in the superior direction in most cases, due to intra-fractional bladder filling; this moved it closer to nodal target volumes.