Monte Carlo calculations of critical organ doses in radioembolization therapy of primary liver tumors via 90Y microspheres

Abstract

Background: Dosimetry calculations in radioembolization therapy are known to include some uncertainties due to working assumptions. First, the microspheres used in the procedure are homogeneously distributed within the tumor volume. Second, Medical Internal Radiation Dosimetry (MIRD) method of dose calculation involves a mono-compartmental model only. To minimize the impact of these uncertainties, this study proposes Monte Carlo (MC) simulations as an alternative to MIRD method to verify the absorbed doses in the volumes of interest (tumor and its surroundings). Material and Methods: Lung, liver, and tumor volumes of 30 radioembolization patients were defined in a mathematical whole-body phantom and MC simulations were performed using Monte Carlo N-Particle code. Absorbed doses were calculated for these tissues both in addition to stomach wall, pancreas, spleen, and kidneys which are close to the tumor volume being treated with microspheres of radioembolization therapy containing the beta-emitting 90Y radioisotope. Results: The doses absorbed by tumor, lung, and liver volumes of each patient were calculated by both MIRD methodology and MC simulations. The differences between the two methods were evaluated for normal lung tissue and tumor tissues in the liver where maximum differences were observed for tumor tissues (16.18%) and lungs (11.69%). Furthermore, it was observed through MC simulations; the organs that are close to the liver being treated were also exposed to the radiation for which absorbed doses could not be calculated by MIRD method. Conclusion: MC simulations may offer significant advantages for dose verification in radioembolization therapy.