ESTRO 2025 - Abstract Book

S260

Brachytherapy - Gynaecology

ESTRO 2025

The reference isodose was 100% of prescription dose.For each plan, we built the isodose100 structure (iso100), corresponding to the prescription dose line from which we removed the applicator volume, to consider the actual dose to tissues surrounding the applicator ( Figure1 ).Since the scope of this study is related to the gradient impact, we force the analysis to avoid possible volumes of underdosage. To quantify the dose increment inside iso100, we considered as valuable metric the 75th percentile of the voxel-based gradient calculated from dose distribution (vGI). The influence of vGI on the tumor control probability (TCP), calculated as described by [ doi:10.7150/ijms.5.41 ], was investigated. We considered the iso100 volume, the distance of OARS (mHD) defined as the minimum Hausdorff distance between iso100 and involved organs at risk, as well as the planner as variables that could impact on vGI.

Results: We analyzed 37 dose distributions. The TCP was calculated for every treatment fraction with median[min;max] values of 88.90% [80.95;95.56]. Over all plans, large variability of vGI, Iso100 volume and mHD were found with values of (median[min; max]) 1.40 [1.11;1.83], 45.28cm 3 [21.30;74.62] and 3.89mm[2.43;11.29], respectively. We found correlation between vGI and TCP with a Pearson correlation coefficient (PCC) of 0.83, (p<0.01) ( Figure2a ). The vGI resulted correlated with both iso100 volume and mHD ( Figure 2b/2c ) with PCC=0.64 (p<0.01) and 0.67 (p<0.01) with iso100 volume and mHD, respectively. No significant correlation was found between vGI variability and planners with PCC=0.087 p=0.62.