Abstract Book

ESTRO 37

S569

Investigacion en Radiofisica e Instrumentacion Nuclear en Medicina IRIMED, Valencia, Spain

Purpose or Objective EMBRACE-II recommended reporting vaginal dose points in high-dose-rate brachytherapy (HDRBT) and external beam radiotherapy (EBRT) in cervical cancer treatments. This study aims to evaluate the robustness of these dose points according to interobserver analysis. Material and Methods The vaginal dose points of five cervical cancer patients treated with HDRBT and EBRT were evaluated by 5- blinded medical physicists from the same center. The planning images were MR and CT images for HDRBT and EBRT, respectively. A consensus was previously established by all observers to determine the anatomical references criteria. The vaginal point in Posterior-Inferior Border of Symphysis (PIBS), PIBS+2cm and PIBS–2cm in the caniocaudal direction dose points were obtained for HDRBT and EBRT (Figure 1). Also, the vaginal top dose points were evaluated for HDRBT treatments (Figure 1). The mean, range and standard deviation for these dose points were evaluated for each of the 5 observers. The interobserver variation was assessed by coefficient of variation (COV). The effect of interobserver variation in the total dose (HDRBT+EBRT) recorded was analyzed by estimating the biologically equivalent dose (EQD 2α/β =3), based on two nominal 7 Gy cervical HDRBT plans (two fractions per plan) and 45 Gy cervical EBRT plan. Figure 1. Vaginal dose points: (a) PIBS, PIBS+2cm and PIBS-2cm in MR sagittal image for HDRBT treatment, (b) Vaginal top points in MR axial image for HDR treatment, (c) PIBS, PIBS+2cm and PIBS-2cm in CT sagittal image for EBRT treatment.

Results Interobserver COV for PIBS, PIBS+2cm and PIBS-2cm were 2.2%, 40.2% and 6.5%, respectively. Dose variability in PIBS+2cm could be due to the proximity to the active positions of the source in some HDRBT treatments cases. COV was very similar between the vaginal top dose points, which was around 15% in all of them. The highest interobserver dose point variation yielded a vaginal dose difference up to 24.3 Gy EQD2, without taking into account the PIBS+2cm where a very extreme difference resulted due to its proximity to the active length tandem. The mean of estimated intraobserver variation for the reported dose points was around 5%. The mean, minimum, maximum and dose difference received by the vaginal dose points for each parameter are shown in the Table 1.