SU-E-T-391: Is 1D Anisotropy Function Recommended by TG-43U1 Protocol a Good Candidate for Clinical Brachytherapy Dosimetry?

Abstract

Purpose: The effect of the anisotropy on dose distributions for real permanent prostate implants with 125I seeds has been evaluated and compared with Monte Carlo (MC) calculations. Methods: soSeed (Bebig) was used. Dose distribution of the prostate implants was calculated using MCNP5 code and compared to TG43 calculation. A DVH comparison between 2D and 1D approximations was performed for prostate, urethra and rectum for 20 patients. Results: A DVHs comparison between the use of 1D & 2D anisotropy functions are presented for 20 prostates implants. For prostate, Dmean, D90 and D98 are underestimated by 1D approximation by about 3%, 1.5% and 1% respectively. It leads also to inconsistent results for Dmin and D100. 1D approximation overestimates the dose for about of 15% in planes situated at the two ends of implant containing no implanted seeds while it underestimates dose about 5% in the other planes. This yields a compensating effect in the resulting DVH's. These over dosage and under dosage must be attributed to an averaging effect between 2D and 1D anisotropy function. It can be noticed that the effect of 1D approximation on D98 and Dmin of prostate varies from patient to another. This is due to the many factors, for example, prostate volume, number of seeds and seeds positions relative to prostate…etc. D90 is always underestimated by 1D approximation where for 70% of patients D90 exceeds 185 Gy. For the urethra, 1D approximation underestimates the Dmax in which 60% of patients, their maximum dose for the urethra exceeds the limits. Finally, 1D approximation underestimates V100 about 50% les than 2D approximation Conclusions: The use of the 2D anisotropy function is preferred to the 1D anisotropy function as this approximation yields more consistent results for the 20 patients studied. The observed overestimation and the variability among the patients can influence dose—outcome.