ESTRO 2024 - Abstract Book

S4300

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

ESTRO 2024

Ten left-sided breast and lymph node cases previously treated with Ethos (and Eclipse optimization) were selected. An Ethos optimization template was tuned using standard 3 partial arc volumetric modulated arc therapy. Target volume prescriptions were 52.2 Gy to the breast and the internal mammary chain (IMC) and 49.3 Gy to the clavicular nodes (Nodes) in 29 fractions. CTV-PTV margins are set at 7mm for all locations except 5mm below the skin. Re planning was performed automatically for each patient using the template without editing. The generated plans were compared with the original clinical plans using PTV and CTV coverage, homogeneity indices (HI) and dose to organs at risk (OAR) via Dose-Volume Histogram metrics. Paired samples t-tests were performed with a significant p-value of < 0.05. In addition, two breast radiation oncologists performed a blinded assessment of the clinical acceptability of each of the two plans (original and automated) for each patient. The Ethos optimisation template was not shown to the physicians prior to scoring in order to avoid scoring bias.

Results:

The figure below shows the dosimetric results for the automated and the original plans. The Ethos optimization template produced plans with significantly higher breast and node CTV coverage than the original plans (p<0.05). However, there was no significant difference in PTV coverage between the two plans despite lower values for the automated plans. The homogeneity index for the 52.2 Gy isodose was 0.16 ± 0.05 for the original plans and 0.19 ± 0.06 for the automated plans. Regarding dose to organs at risk, 100% of the automated plans were within dose tolerance for the heart, ipsilateral lung, and contralateral breast, compared to 70% of the original plans. For the contralateral lung, the Dmean<5Gy constraint was met by 70% and 50% of the automated and original plans, respectively. Physicians 1 and 2 clinically validated 70% and 80% of the automated plans, respectively. The main reason for non-validation was insufficient PTV coverage of the breast and IMC.