ESTRO 2024 - Abstract Book

S3652

Physics - Dose prediction, optimisation and applications of photon and electron planning

ESTRO 2024

Cancer Research Consortium (DKTK), Partner Site Tübingen, a partnership between DKFZ and University Hospital Tübingen, Germany

Purpose/Objective:

Despite the current possibilities of image-guided high-precision patient position verification and online adaptation, the standard radiotherapy (RT) treatment pathway still involves a dedicated planning CT (pCT). Previous research has shown that a simulation-free workflow is realizable in palliative care and that the demands on staff and time resources can be significantly reduced [1-4]. The purpose of this study was to assess the feasibility and dosimetric quality of integrating a simulation-free treatment planning workflow for online adaptive RT on a 1.5 T MR-Linac in prostate cancer (PC) using diagnostic CT (dCT) scans.

Material/Methods:

A total of ten PC patients, which were previously treated on a 1.5 T MR-Linac (Unity, Elekta AB, Sweden) with adaptive RT (42.7 Gy in 7 fractions) were retrospectively included in this study. Simulation-free reference plans were created on existing dCT data (e.g. abdominal CTs and whole-body CTs from positron emission tomography) acquired at most one year before the start of RT. Contouring of the organs at risk (OAR) was performed automatically using commercial software (ARTplan TheraPanacea, Paris, France), and the definition of the target volume was accomplished through a uniform margin concept. The electron density information of the anatomical structures were substituted with population-weighted median values using the bulk density approach, according to [5]. Online plan adaptation at the MR-Linac was simulated for the simulation-free reference plans based on the MR image dataset including the structure set used at the first clinical treatment fraction in the standard adaptive workflow. The evaluation was conducted through a comparison of reference plans and adapted plans derived from both the standard and simulation-free planning workflows using the institutional dosimetric criteria for plan acceptance. Statistical assessment of the dosimetric criteria was carried out using a Wilcoxon signed rank-test (significance level α = 0.05). In both the reference plans and the adapted plans derived from the simulation-free workflow, compliance with institutional dosimetric criteria was consistently achieved in all cases. Statistical analysis of the dosimetric criteria revealed no significant differences between standard and simulation-free workflow for all structures except the penile bulb (p = 0.014) considering the reference plans and the clinical target volume (CTV) (p = 0.010) in the adapted plans, respectively. The evaluation of the adapted plans showed a median (range) D98% of the CTV of 41.3 Gy (40.9 - 41.5 Gy) for the standard treatment plans and 41.4 Gy (41.2 – 41.5 Gy) for the simulation-free plans. Regarding the PTV, the median D98% resulted in 40.1 Gy (39.4– 40.3 Gy) for the standard treatment plans and 40.1 Gy (39.7– 40.4 Gy) for the simulation-free plans (cf. Figure 1). Also in terms of plan complexity, no significant differences were observed regarding total number of monitor units (reference plans p= 0.845, adapted plans p = 0.625) or number of segments (reference plans p= 0.910, adapted plans p = 0.703). Results: