Volumetric and dosimetric comparison of computed tomography- and trans-abdominal ultrasound-based volumes for image-guided cervix cancer brachytherapy.

Abstract

Purpose: In cervical cancer, brachytherapy is a key component of multi-disciplinary treatment. Wide scale studies, such as EMBRACE, demonstrate benefits of volume-oriented and image-guided brachytherapy. However, MRI and CT are too expensive for health systems with scarce resources. The aim of the study was to assess whether ultrasound- and CT-based volumes of high-risk clinical target volume (HR-CTV) in organs at risk (OARs) as well as dose values in a given plan are comparable.

Material and methods: Eighteen applications were evaluated. After CT scans, axial ultrasound images were acquired in 5 mm steps, serving as a secondary set. Post-prescription, HR-CTV, bladder, and rectum were contoured on ultrasound images. Image sets were compared using CT-based volumes as primary reference. Dice coefficient and Jaccard index were calculated. CT-based dosimetric plan was then compared with ultrasound-based volumes to determine D₉₀ HR-CTV, D_2cc bladder, and D_2cc rectum, and these were compared with CT-based values.

Results: The mean dose differences between CT and ultrasound volumes for HR-CTV and D_2cc rectum were less than 5%, with the bladder slightly above 5%. Wilcoxon test showed no significant difference between the mean doses for CT and ultrasound. Dice coefficients indicated good to very good correlation for these volumes.

Conclusions: Ultrasound-based volume acquisition appears comparable with CT volumetric acquisition in both dosimetric and volumetric terms. Further studies are needed to validate this technique, potentially offering a more affordable and feasible volume-based brachytherapy option for low-income health systems.