Program Information
Clinical Implications When Using TG-186 (ACE) Heterogeneity Software
A Likhacheva , E Grade , A Sadeghi , T Sokolowski*, Arizona Cancer Specialists, Mesa, AZ
Presentations
SU-F-T-22 (Sunday, July 31, 2016) 3:00 PM - 6:00 PM Room: Exhibit Hall
Purpose: The purpose of this study is to compare dosimetric calculations using traditional TG-43 formalism and Oncentra Brachy Advanced Collapsed cone Engine (ACE) TG-186 calculation algorithm in clinical setting.
Methods: We analyzed dosimetry of four patients treated with accelerated partial breast irradiation using a multi-channel intracavitary device (SAVI). All patients were treated to 34 Gy in 10 fractions using a high-dose-rate (192) Ir source. The plans were designed and treated using the TG-43 model. ACE was used to assess the effect heterogeneity correction on various dosimetric parameters. Mass density was estimated using Hounsfield units.
Results: Compared to TG-43 formalism, ACE estimated lower doses to targets and organs at risk. The mean difference was 19.8% (range 15.3-24.1%) for PTV_eval V200, 12.0% (range 9.7-17.7%) for PTV_eval V150, 4.3% (range 3.3-6.5%) for PTV_eval D95, 3.3% (range 1.4-5.4%) for PTV_eval D90, 5.4% (range 2.9-9.9%) for maximum rib dose, and 5.7% (2.4-7.4%) for maximum skin dose. There was no correlation between the magnitude of the difference and the PTV_eval volume, air volume, or tissue-applicator conformance.
Conclusion: Based on our preliminary study, the TG-43 algorithm appears to overestimate the dose to targets and organs at risk when compared to the ACE TG-186 software. We hypothesize that air adjacent to the SAVI struts contributes to lack of scatter thereby contributing a significant difference in dose calculation when using ACE. We believe that ACE calculation provides a more realistic isodose distribution than TG-43. We plan to further investigate the impact of heterogeneity correction on brachytherapy planning for a wide variety of clinical scenarios, include skin, cervix/uterus, prostate, and lung
Contact Email: