ESTRO 2024 - Abstract Book

S5557

RTT - Patient care, preparation, immobilisation and IGRT verification protocols

ESTRO 2024

1757

Digital Poster

Is the diaphragm a useful surrogate structure when matching a middle/lower third oesophagus CBCT?

Nathan Taylor, Karen Pilling, Alex Bradshaw, Rachel Brooks-Pearson

Newcastle upon Tyne Hospitals NHS Foundation Trust, Northern Centre for Cancer Care, Newacastle Upon Tyne, United Kingdom

Purpose/Objective:

On treatment verification is used to ensure radiotherapy treatment is delivered as planned. The radiographer assesses the image to check the patient position, the target position and organ at risk position. Soft tissue anatomy of the middle and lower third oesophagus can move due to tumour response, inflammation or eating (Gao et al., 2019). Geometric uncertainties and motion challenges in oesophageal radiotherapy treatment can be corrected through adaptive planning and motion management techniques (Voncken et al., 2020). However, adaptive planning is not yet widely available in UK NHS radiotherapy centres. Current practice at the investigating UK NHS radiotherapy centre is a daily 3D-CBCT bony match, followed by a visual check of target coverage, which is challenging due to poor CBCT image quality. Visser et al (2022) found in their study of 15 patients that when using weekly repeated 4DCT, the diaphragm was a suitable surrogate for target position. Surrogates for the target position are used in online IGRT to reduce inter-observer variability, as target matching can be subjective and therefore affect treatment accuracy and dose distributions. This study aimed to determine if the diaphragm could be a surrogate structure for online image matching, using on-treatment CBCTs in patients with middle/lower third oesophageal tumours.

Material/Methods:

An offline image match was conducted on 428 CBCTs from 20 patients by a single independent observer. This was 12 months of retrospective radical middle and lower third oesophageal radiotherapy patient data. Three matches were undertaken for each CBCT; bone match, target match and a diaphragm match. Due to the anatomical shape of the diaphragm, this was only matched in the longitudinal direction as evidence suggested that the diaphragm had the largest impact on longitudinal shifts, especially in left-oriented oesophageal tumours (Visser et al., 2022). Image registrations were performed offline using ARIA® offline review function, involving translations only. Bland-Altman limits of agreement were used to quantify the variation between each match.

Results:

The limits of agreement (LOA) were analysed for longitudinal, vertical, and lateral shifts shown in the table below. The limits of agreement were smallest in the “bone vs target” match which suggests that the current practice of matching to the bone is the most accurate surrogate for the target position. Wide variation in the limits of agreement,