SU-GG-T-496: The Sensitivity of BED and TCP Parameters to Dose Heterogeneity in Brachytherapy Treatments

Abstract

Purpose: To investigate the sensitivity of radiobiological parameters to dose heterogeneity in brachytherapy. Materials and Methods: Biological effective dose (BED) is calculated using a homogeneous and a heterogeneous model. The homogeneous model is based on the minimal peripheral dose to the target and does not take into account the heterogeneity of the dose distribution in brachytherapy. The heterogeneous model (recommended by the AAPM TG137) is capable of taking the dose heterogeneity into account. Different Monte Carlo calculation protocols were used for five breast permanent seed implant patients. The tumor control probability (TCP) was calculated for both the homogeneous and the heterogeneous models. Results: The homogeneous model is overestimating the BED in all calculation cases. This overestimation is higher for plans with higher D90 values. For example, by about 10% for a plan with D90=40 Gy, this overestimation increases to 30% when the D90=90 Gy. Depending on the D90, the homogeneous model overestimates the TCP vis-à-vis the heterogeneous model. Although the difference between the two models is large for intermediate D90 (between 40 Gy and 80 Gy), however, they agree well for higher D90 doses above 80 Gy. The 50% control dose is higher by 13 Gy while the γ50 parameter is lower by 60% in the heterogeneous compared to the homogeneous model. Conclusion: The BED is always overestimated by the homogeneous model for all brachytherapy plans. The difference between the homogeneous and the heterogeneous TCP models is large at intermediate D90 (40Gy to 80Gy) but they agree at higher D90. Because of the elevated dose heterogeneities, the homogeneous model is not suitable for brachytherapy. A more accurate estimation of the BED and the TCP will improve outcome predictions for brachytherapy treatments.