PO97

Abstract

Purpose To compare dosimetric values for interstitial HDR brachytherapy cases using both manual and inverse planning techniques and refine optimization results for clinical use. Materials and Methods Ten plans for prior interstitial brachytherapy were selected for analysis representing a variety of treatments: Elekta's Venezia applicator with needles (4), Best Medical's Syed/Neblett gynecological template (3), and Best Medical's prostate template (3). Each plan, previously manually optimized (MO), was optimized in Oncentra (Elekta) using both IPSA and HIPO inverse planning algorithms. For the first plan of each type, optimization parameters were iteratively adjusted from comparison to the MO treated plan. The parameters were then saved as a template to apply to future plans of the same type. Dosimetric quantities were recorded for each optimization type for comparison. For the optimized cases, the metrics collected were clinically relevant values representing target coverage and OAR constraints. Results For target coverage (HRCTV D90%), IPSA produced lower coverage on average for Venezia (-15.5%) and Syed (-0.2%) cases when compared to the MO plan and higher for prostate (4.3%). HIPO resulted in higher coverage for Venezia (1.3%) and prostate (1.5%) and lower for Syed (-0.7%). OAR doses were assessed normalized to HRCTV D90% equal to prescription dose. IPSA had lower OAR metrics on average for Syed (-8.3%) and prostate (-3.2%) and higher for Venezia (0.1%). HIPO gave lower OAR metrics for Venezia (-1.9%) and Syed (-4.2%) and higher for prostate (2.2%). Conclusions Overall, HIPO was more consistent in comparable or improved results to the clinically treated MO plan. Treatment planning time for clinical interstitial cases has reduced, and we have adopted a hybrid optimization approach starting with HIPO inverse optimization and then performing manual changes as needed. Future work includes refining optimization parameters to be globally applicable for each treatment type and warrant less manual optimization. To compare dosimetric values for interstitial HDR brachytherapy cases using both manual and inverse planning techniques and refine optimization results for clinical use. Ten plans for prior interstitial brachytherapy were selected for analysis representing a variety of treatments: Elekta's Venezia applicator with needles (4), Best Medical's Syed/Neblett gynecological template (3), and Best Medical's prostate template (3). Each plan, previously manually optimized (MO), was optimized in Oncentra (Elekta) using both IPSA and HIPO inverse planning algorithms. For the first plan of each type, optimization parameters were iteratively adjusted from comparison to the MO treated plan. The parameters were then saved as a template to apply to future plans of the same type. Dosimetric quantities were recorded for each optimization type for comparison. For the optimized cases, the metrics collected were clinically relevant values representing target coverage and OAR constraints. For target coverage (HRCTV D90%), IPSA produced lower coverage on average for Venezia (-15.5%) and Syed (-0.2%) cases when compared to the MO plan and higher for prostate (4.3%). HIPO resulted in higher coverage for Venezia (1.3%) and prostate (1.5%) and lower for Syed (-0.7%). OAR doses were assessed normalized to HRCTV D90% equal to prescription dose. IPSA had lower OAR metrics on average for Syed (-8.3%) and prostate (-3.2%) and higher for Venezia (0.1%). HIPO gave lower OAR metrics for Venezia (-1.9%) and Syed (-4.2%) and higher for prostate (2.2%). Overall, HIPO was more consistent in comparable or improved results to the clinically treated MO plan. Treatment planning time for clinical interstitial cases has reduced, and we have adopted a hybrid optimization approach starting with HIPO inverse optimization and then performing manual changes as needed. Future work includes refining optimization parameters to be globally applicable for each treatment type and warrant less manual optimization.