Copenhagen, Denmark
Onsite/Online

ESTRO 2022

Session Item

Sunday
May 08
14:15 - 15:15
Poster Station 1
13: Brachytherapy
Angeles Rovirosa, Spain
Poster Discussion
Brachytherapy
Source path verification - Advanced Gynecological Brachytherapy Applicator (Venezia)
Hema Vaithianathan, Australia
PD-0562

Abstract

Source path verification - Advanced Gynecological Brachytherapy Applicator (Venezia)
Authors:

Hema Vaithianathan1, Tom Kupfer2

1Olivia Newton-John Cancer Research & Wellness Centre , Radiation Oncology, Heidelberg, Victoria, Australia; 2Olivia Newton-John Cancer Research & Wellness Centre, Radiation Oncology, Heidelberg, Victoria, Australia

Show Affiliations
Purpose or Objective

The department purchased first set of Elekta’s Advanced Gynecological Applicator (AGA, “Venezia”) in 2017. Since then, three AGA sets are added in 3 years. This work’s aim was to verify the accuracy of Oncentra, brachytherapy TPS’s modelled source (ver 4.6.0) dwell position (DP) against the actual source position especially in lunar ovoids curved source path, prior to clinical use.

Material and Methods

AGA consists of two lunar shaped ovoids (LO) along with intrauterine (IU) tandem. Three sizes of LO (22, 26 and 30 mm, inner lumen 3 mm diameter) and several IU tube lengths can be configured to accommodate various patient anatomies, while vaginal and perineal interstitial needles (Proguide 6F Round 294mm) allow conformation of the dose to the individual tumor shapes. DP verification in the IU tandems was done using autoradiographs (RTQA film). While in a lunar applicator, with source dwelling at various positions in a curved path, the centre of the source may not be exactly at the centre of the lumen.  However, with bulky LO, the film is too far from the source DP resulting in unacceptable uncertainties in source localization on autoradiographs. Instead, the position of the source within the LO was radiographed with a mobile x-ray unit and a computed radiography (CR) plate. For each LO, four DP were imaged separately to preserve image fidelity whilst avoiding saturation of the CR-detector. A composite image was then created in MatLab using grey-scale based image registration. Finally, the composite image was overlaid over an image created in Oncentra, which showed a detailed outline of the LO model inclusive of the modelled source DP (Nucletron Applicator Library Manager ver 1.0.1.1). The 2D vector distance between the imaged source and the corresponding model DP was measured, along with the offset. Further commissioning tests of the applicator included physical integrity checks, overall match of the applicator’s CT image to the applicator model and determining the distance of the most distal DP to the tip of the needle or to the tip of the applicator (offset).

Results

The physical source does not travel at the centre of the LO lumen, rather, it touches the wall of the lumen in different places as it advances into the ovoid (figure). This is partially reflected in the modelled DP, which agreed to within 1 mm with the centre of the source. The measured offsets (approximately 0.6mm) for IU tube, LO and the 6F needle were within 1 mm of vendor specifications. The CT images of the applicators matched the model well.


Conclusion

Oncentra’s AGA model agrees well with measurements in LO, IU and needles. CR is a useful method to determine the contorted source path within challenging applicator types when autoradiographs cannot accurately be used.  Subsequent LO commissioning is done using CT only which is efficient and less time consuming. Annual quality assurance of the applicators are also done using CT images