ESTRO 2024 - Abstract Book

S4023

Physics - Inter-fraction motion management and offline adaptive radiotherapy

ESTRO 2024

in an unnecessary workload for the department and dose to the patient. In this prospective study we investigated whether dose evaluations on corrected CBCTs (cCBCT) could be used to identify patients that need a plan adaptation.

Material/Methods:

20 photon breast cancer patients (6 locoregionally, 14 locally), treated between June 2023 and October 2023, were included in this prospective study. The photon treatment plans consisted of a 70% conformal (tangential) and a 30% VMAT technique. All patients received daily online CBCT imaging prior to radiation. In case of significant (≥ 7 mm) anatomical deformations visible on the CBCT within the breast area the patients received an rCT for dosimetric evaluation. The rCT was acquired within one day of the CBCT which showed the anatomical deformations. In this prospective study, a dosimetric analysis was performed on the rCT (clinical practice) and the cCBCT. The cCBCT was created with a dedicated CBCT conversion algorithm in the treatment planning system (TPS) [1], [2]. Deformable registration was performed on the planningCT (pCT), cCBCT and the rCT to warp the whole breast CTV and heart from the pCT to the cCBCT and rCT. All warped structures were checked and if necessary, adapted by the medical doctor (MD). Thereafter, the original treatment plan was recalculated on the cCBCT and rCT. The MD evaluated both dose distributions and decided whether a plan adaptation was necessary. The decisions based on the cCBCT were compared with those on the rCT (gold standard) and translated into true positives (TP), true negatives (TN), false negatives (FN) and false positives (FP). To quantitatively compare the dose between the cCBCT and rCT, dose-volume statistics were obtained for the CTV for multiple dose-volume parameters.

Results:

In 16 out of 20 patients the outcome (requiring plan adaptation or not) was equal for both rCT and cCBCT. However, in 4 out of 20 patients, the results of the evaluation were inconsistent. For 3 patients no adaptation was required based on the rCT but was indicated on the cCBCT (FP). In all 3 patients, the reason was a different positioning between the CBCT and rCT acquisition. One patient received a plan adaptation due to a systematically new positioning on the rCT compared to the CBCT (FN). In Figure 1, the relative to the prescribed dose difference between the cCBCT and rCT is plotted. For the D98, the dose differences were between 0.57% and -0.96%, with an average of -0.20%, and this is considered clinically acceptable. The MDs regarded cCBCT quality sufficient for assessing and modifying the CTV for local breast cancer patients. However, evaluating the nodal and boost areas was more challenging due to the inferior image quality compared to rCTs.

Table 1: Outcomes of dose evaluation on repeat CT (rCT) and corrected CBCT (cCBCT). TN = True negative, TP = True positive, FP = False positive, FN = False negative.

Total

Local

Locoregional

TN

14

11

3

TP

2

2

0

FP

3

1

3

FN

1

0

1