Author: Zakaria Aboulbanine, Greeshma A. Agasthya, Paul Inman, Anuj J. Kapadia, Anthony Hong Cheol Lim, Jayasai Ram Rajagopal, Chris C. Wang 👨🔬
Affiliation: Oak Ridge National Laboratory, Georgia Institute of Technology 🌍
Purpose: Radiobiological simulation via TOPAS-nBio requires significant computational resources and time to provide meaningful results. This study aims to decrease simulation time and computational resources for radiobiological track structure simulations by utilizing pre-calculated Standard DNA Damage (SDD) data. The method is tested on a uniform distribution of cells in spherical and cylindrical volumes exposed to various radioactive sources distributed in extracellular space.
Methods: A pre-calculated SDD library of electrons, protons, and alphas was generated using TOPAS-nBio with simulations exceeding 100 Gy dose per discrete energy level. The energy spectrum of individual radiation sources to cells was obtained using Geant-4. The spectrum was translated into track-weighted probability functions (TPFs) based on incident particles to each cell nucleus. TPFs select random single-track SDD data from the pre-calculated SDD library up to the desired dose or activity in the cell nucleus to create a superimposed SDD data file for each cell, and this step was repeated to a number of statistics fell below a precise level. Timestamps were added to the superimposed SDD files to incorporate dose-rate effects. The resulting files were then analyzed using MEDRAS to compute double-strand break (DSB) mis-repairs and chromosomal aberrations, which enable an assessment cell survival.
Results: The proposed approach significantly decreased the simulation time for radiation-induced DNA damage, reducing it from 10 minutes for electrons, 1.5 hours for protons, and 5 hours for alpha particles to 23.5±3.7 sec/Gy/cell. Extensive validation involving 10,000 simulations across 26 different radiation scenarios demonstrated a strong correlation with Monte Carlo results. This method efficiently simulated DNA damage in 5,000 cells and accurately predicted cell survival outcomes for Cf-252, Ac-225, Lu-177, and Y-90 sources.
Conclusion: We have demonstrated that the proposed method can estimate cell survival curves for complex radiation scenarios with 2-3 orders of magnitude faster calculations than direct track structure calculations.