PO91

Abstract

Purpose To commission and retrospectively compare dosimetry of Elekta's Advanced Collapsed Cone Engine (ACE) calculation algorithm with the standard TG 43-based dose formalism on an IPSA-optimized and CT-based HDR clinical breast plan. Materials and Methods At our institution, we commissioned ACE for HDR brachytherapy using the “Breast Test Case” data from iroc.mdandderson.org. This used a generic Ir-192 source and was published by the AAPM Working Group on Commissioning of Model-Based Dose Calculation Algorithms in Brachytherapy (UN35). A previous breast cancer case treated with Hologic's Contura multi-catheter HDR applicator was selected from our clinical database to evaluate the ACE dose engine. The Oncentra treatment planning system (v. 4.6) was used in this study. The HDR plan was first calculated with the TG-43 dose formalism and optimized using inverse planning by a simulated annealing algorithm (IPSA). Using the same relative dwell weights provided by IPSA from the TG-43 plan, dose was then recalculated using the ACE algorithm. The ACE calculation used Oncentra's High Accuracy level with a 1.0-mm-cubed dose grid. The planning target volume (PTV) and organs at risk (including regions with significant inhomogeneity) were contoured. Dose Volume Histograms (DVH's) of the PTV_eval and organs at risk were calculated with both TG-43 and ACE dose formalisms for each plan. Absolute and percent differences were also calculated for each metric. Results As shown in Table 1, the ACE calculated dose relative to the TG-43 calculated dose is consistently lower for all dosimetric parameters. The dose calculation differed by up to 9% or 112 cGy in the selected parameters. See comparison table for more details. The calculation time was 16 minutes using the standard GPU provided by Elekta hardware. Conclusion The IROC “Breast Test Case” ACE commissioning was a straightforward procedure and easy installation. Total commissioning time was less than an hour. The ACE calculated dose relative to the TG-43 calculated dose in the selected patient case was consistently lower for all dosimetric parameters. This algorithm offers a more accurate reconstruction of the dose distribution (by comparison to Monte Carlo, per UN35). The highest percent differences in dose calculation were found in the PTV_eval and skin region for this case. This study provides a framework for future studies of ACE dosimetric impact for breast cancer brachytherapy, and more breast cancer patient plans will be analyzed with summary statistics reported. This will be a particularly useful framework when correlating lower skin dose estimate with cosmetic outcomes. To commission and retrospectively compare dosimetry of Elekta's Advanced Collapsed Cone Engine (ACE) calculation algorithm with the standard TG 43-based dose formalism on an IPSA-optimized and CT-based HDR clinical breast plan. At our institution, we commissioned ACE for HDR brachytherapy using the “Breast Test Case” data from iroc.mdandderson.org. This used a generic Ir-192 source and was published by the AAPM Working Group on Commissioning of Model-Based Dose Calculation Algorithms in Brachytherapy (UN35). A previous breast cancer case treated with Hologic's Contura multi-catheter HDR applicator was selected from our clinical database to evaluate the ACE dose engine. The Oncentra treatment planning system (v. 4.6) was used in this study. The HDR plan was first calculated with the TG-43 dose formalism and optimized using inverse planning by a simulated annealing algorithm (IPSA). Using the same relative dwell weights provided by IPSA from the TG-43 plan, dose was then recalculated using the ACE algorithm. The ACE calculation used Oncentra's High Accuracy level with a 1.0-mm-cubed dose grid. The planning target volume (PTV) and organs at risk (including regions with significant inhomogeneity) were contoured. Dose Volume Histograms (DVH's) of the PTV_eval and organs at risk were calculated with both TG-43 and ACE dose formalisms for each plan. Absolute and percent differences were also calculated for each metric. As shown in Table 1, the ACE calculated dose relative to the TG-43 calculated dose is consistently lower for all dosimetric parameters. The dose calculation differed by up to 9% or 112 cGy in the selected parameters. See comparison table for more details. The calculation time was 16 minutes using the standard GPU provided by Elekta hardware. The IROC “Breast Test Case” ACE commissioning was a straightforward procedure and easy installation. Total commissioning time was less than an hour. The ACE calculated dose relative to the TG-43 calculated dose in the selected patient case was consistently lower for all dosimetric parameters. This algorithm offers a more accurate reconstruction of the dose distribution (by comparison to Monte Carlo, per UN35). The highest percent differences in dose calculation were found in the PTV_eval and skin region for this case. This study provides a framework for future studies of ACE dosimetric impact for breast cancer brachytherapy, and more breast cancer patient plans will be analyzed with summary statistics reported. This will be a particularly useful framework when correlating lower skin dose estimate with cosmetic outcomes.