ESTRO 2023 - Abstract Book

S1667

Digital Posters

ESTRO 2023

Conclusion IMRT and VMAT significantly reduce the mean dose to carotids and this difference is robust for the inter-fraction differences. The uncertainty plan is a poor estimator of patient inter-fraction differences, it doesn’t account for non-rigid movements. Robustness is better for 3D (CTV) and IMRT (PTV). It improves when adding the shoulders as an avoidance volume. 2D transit IVD: false positives due to the position of the shoulders at the beam exit. Good estimator of inter-fraction positioning differences , useful to flag patients that need daily imaging. 3D dose recalculation (logfiles and CBCT): good estimator even if sometimes differs from Eclipse recalculation due to differences in the body contour and dose calculation algorithms.

PO-1916 Parameters influencing plan adaptation in particle therapy are non-dosimetrical?

A. Vai 1 , E.L. Chilug 2 , S. Molinelli 1 , A. Barcellini 3 , G. Magro 1 , A. Bazani 1 , A. Mirandola 1 , E. Rossi 1 , S. Russo 1 , L. Trombetta 1 , B. Vischioni 3 , E. Orlandi 3 , M. Ciocca 1 1 CNAO Foundation, Medical Physics Department, Pavia, Italy; 2 Horia Hulubei National Institute for R&D in Physics and Nuclear Engineering (IFIN-HH), Physics and Nuclear Engineering (IFIN-HH), Magurele-Ilfov, Romania; 3 CNAO Foundation, Radiotherapy Department, Pavia, Italy Purpose or Objective Finding parameters influencing the need of plan adaptation might be beneficial in particle therapy (PT) to optimize the balance between plan quality along the treatment course and clinical workload. Materials and Methods A single-center, retrospective statistical analysis was carried out on PT treatments (protons and carbon ions) delivered within 1 year at our facility (n=557). Correlations (Fisher’s exact test and Pearson’s chi test) were investigated between more than 20 parameters divided in the following groups: patient data (sex, age, weight), tumor histology and location, treatment schedule timeline (including re-evaluation imaging), particle type, nominal plan parameters (e.g. beam geometry, robustness), dose evaluators for target and OARs. In particular, focus was given to patients which received at least one re-evaluative CT scan (RvCT) during the treatment course (n=340) and might require treatment adaptation. For these patients, relevant dosimetrical parameters for target and OARs of the nominal plan recalculated on each RvCT were compared to nominal values. Analysis was primarily focused on H&N group given the expected larger rate of replanning compared to other tumor sites. Results Of the 485 pts which received a CT-based treatment plan, n=340 (70.1%) received at least 1 re-evaluative CT during treatment course. N=87(17.9%) were re-optimized once and n= 20 (5.9%) 2 times or more. With reference to tumor location (brain/abdomen and gyneco/pelvis and column /H&N) we registered the following rate of replanning, respectively: 2.3% / 27.6% / 13.8% /56.3%. Among the H&N, malignant mucosal melanoma (27.0%) and squamous cell carcinoma (21.6%) were the most replanned. Dosimetrical analysis showed plans were mostly re-optimized due to: a) dose deviations from nominal values for OARs not included in the robust optimization; b)unpredictable interfractional variations; c) target sub-optimal coverage due to anatomical changes in tissues within and surrounding the target. Tumor location and type of ion were significantly related to the rate of replanning (p<0.05).