Implementation of a brachytherapy Ir-source in an in-house system and comparison of simulation results with EGSnrc, VMC++ and PIN

Abstract

Today's brachytherapy planning systems perform computation of energy deposition in patients by assuming homogeneous water medium and multiplicative transmission factors for shielding. Patient heterogeneities, shape and size are not fully taken into account. Aim of this study is the implementation of the microSelectron 192Ir high dose rate brachytherapy source in the Swiss Monte Carlo Plan, an in-house developed MC environment, where a defined geometry can be simulated by using one out of three different transport algorithms: EGSnrc, VMC++ or PIN. Additionally, the impact of different phantom shapes and clinical relevant inhomogeneities on dose distributions are studied. Radial dose functions, dose rate constants and anisotropy functions according to the AAPM TG-43 formalism have been determined. The implemented source has been validated by comparing dose distributions in a 30 × 30 × 30 cm3water phantom derived from MC simulations using the three transport algorithms with literature data. Dose rate constants, radial dose functions and anisotropy functions agree within 3% with literature data. Placing the source toward the surface of a water phantom can result in local underdosage of up to 17% when compared with the dose distribution around the source at the centre of a 30 × 30 × 30 cm3water phantom. Taking into account the presence of an air and cortical bone inclusion positioned at 1 cm from the source can lead to dose deviations in the region behind these inhomogeneities of up to +7% and -4%, respectively, if compared with the dose in a 30 × 30 × 30 cm3water phantom. The same geometry can be used to compare different transport codes within one Monte Carlo system. Apart from the inverse square law, the impact of the size and the geometry of the phantom as well as heterogeneities on dose distributions have to be considered.