Internal dosimetry for TARE therapies by means of GAMOS Monte Carlo simulations

Abstract

Three-dimensional internal dosimetry is increasingly used in planning Trans-Arterial Radio-Embolization (TARE) of HepatoCellular Carcinoma (HCC). Among the existing calculation approaches, Monte Carlo (MC) simulation is the gold standard. Aim of this work was to carry out a retrospective study of clinical cases of TARE to compare the performances of different computation approaches. We developed a procedure exploiting GAMOS (GEANT4-based Architecture for Medicine-Oriented Simulations) MC. Three dimensional absorbed dose maps, dose profiles and Dose Volume Histograms (DVHs) were produced for liver through MC simulations and convolution method implemented in STRATOS software. We compared the average absorbed doses with results of Medical International Radiation Dose (MIRD) approach. For most patients, a reasonable agreement was found, with relative differences in mean doses within (-20.2%,+15.6%) for MIRD vs. MC and (-12.1%, +7.6%) for STRATOS vs. MC. Discrepancies can mainly be related to the gamma-rays contribution, more precisely taken into account in MC. For one patient we evaluated through MC simulation a lung dose of about 2 Gy coming from pulmonary shunt (96%) and from irradiation from liver (4%), with values up to 4.5 Gy near liver-lung interface. 3D dosimetry for TARE treatments can be satisfactorily carried out with convolution methods as long as VOIs of regular shape are considered. MC simulations are more appropriate for VOIs where the contribution from gamma-rays has to be carefully taken into account. The absorbed dose distribution in presence of relevant tissue inhomogeneities can be assessed accurately by means of MC simulations only.

Keywords: GAMOS; GEANT4; Internal dosimetry; Monte Carlo; TARE.