ESTRO 2023 - Abstract Book

S1560

Digital Posters

ESTRO 2023

Figure 1 : boxplots of the mean UH errors for various structures of the brain and pelvis cohorts The Dmean mean error was smaller than 2 % for PTVs and OARs but Dmean deviations could reach up to 15% for femoral heads (Figure 2). The mean D98% difference was -1.4% for the brain PTVs and -3.5% for the pelvis PTVs. The mean gamma passing rate was 90.8% for the brain and 80.2 % for the pelvis.

Figure 2 : boxplots of Dmean deviations collected for various various structures of the brain and pelvis cohorts

Conclusion This study demonstrated the feasibility of using AI-tools to generate clinically acceptable synthetic CT for MR-only radiotherapy workflows in the pelvic and brain regions. Deviations between sCT and CT were comparable to those reported in the literature for both image and dosimetric evaluations. Future work will include the comparison of this new sCT generator to other AI and non-AI solutions, as well as investigate the ability of using sCTs for patient positioning.

PO-1829 A computational framework for assessment of dose to lymphocytes in external radiotherapy treatments

I. Lopez-Martinez 1 , I. Espinoza 1 , I. Muñoz-Hernandez 1 , P. Muñoz-Schuffenegger 2 , B. Sánchez-Nieto 1

1 Pontificia Universidad Católica de Chile, Facultad de Física, Santiago, Chile; 2 Pontificia Universidad Católica de Chile, Radiation Oncology Unit, Department of Hematology-Oncology, Santiago, Chile Purpose or Objective Several studies have shown increased patient survival when combining radiotherapy (RT) and immunotherapy (IO) for several cancers. However, due to the increasing evidence of the negative patient survival correlation with radiation-induced lymphopenia (RIL), the effect of dose on the immune system has become a growing issue in RT planning. As circulating lymphocytes (CL) are in constant motion during RT treatments, they are often not considered in the planning system and are not counted as an organ at risk; furthermore, to achieve an immune-sparing RT is fundamental to assess the dose delivered to CL. This work presents a computational framework that allows for a patient-specific fast calculation of lymphocyte dose. Materials and Methods Based on the ICRP89 publication, a blood network model was generated in MATLAB (version R2021a). It represents the circulatory system as an array of blood flow from one organ to another. For each organ, a dose-volume histogram (DVH) is required. The software IS2aR[1] was used to generate a patient-specific whole-body virtual phantom(VP) based on a rigid registration of the planning CT and the ICRP110 reference phantom. DVHs of organs inside the 5% isodose are directly extracted from the TPS, while the remainder are calculated using Periphocal 3D[2] on the generated VP. The blood was simulated using "blood particles"(BP), which represent a small volume of blood, and they follow a "first in, first out" iterative movement in each organ. During each iteration, a BP will receive a dose equal to a random bin in the DVH (whichever organ it is) multiplied by a fraction of treatment time, thus obtaining the DVH of lymphocytes. Finally, using the model by Sung et al. [3] (modified to make the supply factor in concordance with our patient cohort), the lymphocytes survival fraction is obtained Results The model was preliminarily tested on data from 4 patients with hepatocarcinoma treated with RT. Lymphocyte counts were taken pre-treatment, between the first or second months after the beginning of RT (early measurement) and during the third month after RT (late measurement). In all but one patient, the maximum difference between counts and model predictions was 6.8% for early measurements. In late measurements, the average difference was 17.3%. (Fig 1.)