ESTRO 2024 - Abstract Book

S4262

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

ESTRO 2024

Adèle Gabillaud 1 , Louis Rigal 2 , Antoine Simon 2 , Raphaël Martins 2 , Loïg Duvergé 3 , Nolwenn Delaby 1 , Karim Benali 4,2 , Renaud de Crevoisier 2 , Julien Bellec 1 1 CLCC Centre Eugène Marquis, Medical Physics Department, Rennes, France. 2 Univ Rennes, CHU Rennes, CLCC Eugene Marquis, Inserm, LTSI – UMR 1099, LTSI, Rennes, France. 3 CLCC Centre Eugène Marquis, Radiation Oncology Department, Rennes, France. 4 Saint-Etienne University Hospital, F-42270, France, Cardiology, Saint-Priest-En-Jarez, France

Purpose/Objective:

This study evaluates the potential dosimetric benefits of the mid-position (MidP) concept compared with Internal Target Volume approach for treatment planning of Stereotactic Arrhythmia Radioablation (STAR).

Material/Methods:

Ten patients who underwent STAR for refractory ventricular tachycardia in our institution were included in this retrospective planning study. All patients were treated in a VersaHD (Elekta, Crawley, UK) using a VMAT technique with in-treatment coneBeam CT image guidance. For each patient, a cardiac-gated four-dimensional computed tomography scan (4D-CT card ) and a respiratory-gated scan (4D-CT respi ) were acquired and both reconstructed in 10 phases. The CTV was delineated on the end-of-diastole phase of the 4D-CT card and was propagated firstly to the other cardiac phases by deformable registration and secondly to all respiratory phases by rigid image registration [1]. The mean time-weighted position of the CTV was deduced from the DICOM coordinates of the CTV in the different image frames. MidP-based PTV margins were calculated based on van Herk’s margin recipe considering cardiac and respiratory motions as patient-specific random errors. For comparison purpose, an ITV-based PTV was generated by performing the union of the CTV indexed on each phase of the 4D-CT card and 4D-CT respi and using a 3 mm safety margin. For each patient, a VMAT treatment plan using 2 arcs was simulated on RayStation v12A TPS (RaySearch) for both MidP-based and ITV-based PTVs. Treatment plans were optimized such as the prescription isodose of 25 Gy (or 20 Gy for patients for which the left coronaries arteries were nearby the CTV) encompassed the PTV while respecting Organs at risk (OAR) dose constraints of the RAVENTA trial [2]. All treatments plans were then normalized such that 95% of the PTV received 100% of the prescription dose. For each patient and treatment plans, the dose delivered to the CTV was evaluated by 4D dose accumulation by incorporating firstly the different positions of the CTV during cardiac and respiratory cycles extracted from the 4D-CT dataset and secondly the intra-fraction positioning errors retrospectively evaluated based on in-treatment CBCT images.

Results:

Compared with ITV strategy, the MidP strategy resulted in a mean [min-max] relative PTV volume reduction of 29% [22%, 41%] (p<0.001, Wilcokson signed rank test). The mean [min-max] D95% CTV coverage was 102% [99%-106%] and 105% [100%-108%] of the prescription dose for MidP and ITV-based plans, respectively. The doses delivered to surrounding OAR for the ITV-based and MidP-based plans are reported in Table1. The median heart dose was significantly lower with MidP-based plans with a mean difference of -0.2 Gy (p<0.001). The near-maximum dose (D1%) delivered to left coronary arteries, aorta and stomach were systematically lower with the MidP-based plans.