ESTRO 2025 - Abstract Book

S258

Brachytherapy - Gynaecology

ESTRO 2025

Material/Methods: Ten cervical cancer patients who underwent definitive chemoradiation who underwent MR at baseline, prebrachytherapy were evaluated to study the incremental value of incorporating DWI/ADC over T2 sequence alone for target delineation. MR scans were acquired on a 1.5/3 Tesla scanner, with a b value of 1000 for DWI. The GTV initial and GTV brachytherapy were delineated using only the T2 sequence as the hyperintense signal and gray zones and on the co-registered T2/DWI/ADC images as the hyperintense signal on DWI along with the anatomical information from T2w, and with qualitative ADC for interpretation of DWI, when necessary. The upper limit of acceptable registration error in the T2 and DWI images in the region of interest was 2mm. The contours were finalised on the consensus of radiation oncologist and radiologist and both were blinded to the clinical findings. Additional, response assessment scans of 8 patients available were studied and any residual if present, was delineated. The absolute volumes, near maximum distances (NMD's) at the level of uterine artery, 1 cm below, and 2 cm below on either side, DICE coefficient, correlation coefficient of the delineated volumes on T2 w imaging were compared. Results: At diagnosis, the percentage difference between the volumes, GTV_T2 initial and GTV_ DWI/ADCinitial was a median of 5.17%, ranging from + 4.5% to -51.1% with a DICE coefficient of 0.73 (range-0.51-0.78). The variation in NMD’s at diagnosis on the two sides ranged from -0.2 to 1.7 mm and -0.7 mm to 0.4 mm. At brachytherapy, the percentage difference of volumes at brachytherapy, GTV_T2 brachytherapy and GTV_ DWI/ADC brachytherapy ranged from - 44% to +100% with a median DICE coefficient of 0.45 (range- from 0.25 to 0.74) and NMD’s difference ranged from - 0.04 to +2.1 and -0.3 to 0.46 mm.