ESTRO 2024 - Abstract Book

S3731

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

ESTRO 2024

The anatomical changes experienced by many head and neck (H&N) cancer patients throughout the radiation treatment course may compromise the planned dose distribution. Evaluating which patients may benefit from offline adaptive radiotherapy and in which session to replan is a time consuming and generally subjective task. The purpose of this work is to quantify dosimetric differences between reference and delivered doses in a large cohort of patients from two different institutions, and to establish a foundation for automated and evidence based decision-making.

Material/Methods:

All patients included in this study underwent VMAT treatments with a TrueBeam (Varian) linac, but for one patient, who was treated with a 9-field IMRT plan. For each treatment session, the planning CT was deformed to the daily CBCT using Velocity (Varian), and the dose distribution was computed in Eclipse (Varian) treatment planning system using the calculated treatment plan. Three levels of CTV and up to five organs at risk (spinal canal, parotids – left and right -, oral cavity and mandible) have been included in the analysis. A database was created with clinical information (e.g. initial weight, tumor site) and data from the treatment plan (e.g. number of sessions, prescribed dose). This was completed with information from each treatment session, for instance, the volumetric changes and the relevant dosimetric indices, i.e., D95% for the CTVs, D2cc for the mandible and spinal canal, and mean dose for the parotids and oral cavity. The portal images of each treatment session were also processed, comparing them with the images of the reference session (generally the first one), and computing some gamma passing rates (local analysis with a 10% dose threshold and with different gamma criteria, varying from 1% to 3% in dose and from 1 to 3 mm in distance).

The cumulative variation between the delivered and reference dose indices was then calculated, and a Spearman, ρ, correlation was evaluated between different variables from the database.

Results:

Data from 127 patients and 4077 treatment sessions delivered in two independent hospitals have been included. From all these patients, 26 (20.5%) were replanned following the current local criteria. The main tumor places were: 35 (27.5%) oropharynx, 31 (24.4%) larynx, 24 (18.9%) oral cavity, 12 (9.4%) hypopharynx and 11 (8.7%) nasopharynx. Regarding the radiation treatment, for 80 (63.0%) patients it was primary, whereas for the other 47 (37.0%) it was adjuvant. Figure 1 shows the cumulative (after the complete treatment) dose index variation, differentiating between hospitals and between those patients who were replanned and those who were not. Most cumulative differences are within ±3 Gy (about 4.5% of the prescribed dose) after the complete treatment course, although there are some cases with higher differences.