Current practices and challenges in inverse planning for gynecologic brachytherapy: insights from a multi-institutional study.

Abstract

This study investigated the feasibility and reproducibility of inverse planning (IP) in gynecologic high-dose-rate brachytherapy (BT) across ten institutions (five IP and five manual). Two pre-defined cases (high-risk clinical target volume, CTVHR: 62.8 and 27.4 cm3) included the organs at risk (OARs), a tandem, ovoids, and 8 and 3 interstitial needles, respectively. Each institution generated BT plans according to its policy, assuming 30 Gy/15 fractions of whole-pelvis irradiation and 20 Gy/10 fractions of central shielding. Plans were evaluated using dose-volume histogram (DVH) parameters. Dwell-time distributions were compared with conventional Manchester-based pear-shaped plans to identify deviations (e.g. needle dominance and high-dose characteristics) that may not be captured by DVH alone. Institutional interviews assessed clinical acceptability and optimization strategies. All plans were considered clinically acceptable, with no significant differences between IP and non-IP groups in CTVHR  D90% and OARs ${{D}}_{{2}\ {\mathrm{cm}}^{{3}}}$. IP institutions used institution-specific strategies to preserve uterine dose conformity, including fixing dwell times for selected applicators and applying optimization constraints based on Manchester-derived contours, delineated ovoids or delineated uterine structures. Overall, IP plans were comparable to non-IP plans for CTVHR  D90%, CTVHR  V100% (cm3) and OARs ${{D}}_{{2}\ {\mathrm{cm}}^{{3}}}$. In contrast, differences in dose distribution around the uterine fundus and the wide range of the body V100% (cm3) suggested varying institutional strategies for fundus dose. These findings highlight the need for planning QA that incorporates dwell-time review in addition to DVH parameters, and provide a foundation for optimizing and standardizing IP in gynecologic BT to improve planning reproducibility and patient outcomes.