Author: Xun Jia, Youfang Lai 👨🔬
Affiliation: Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University, Johns Hopkins University 🌍
Purpose: Ultrahigh dose rate FLASH (>40 Gy/s) radiotherapy (RT) has attracted significant attention. The mechanism remains unclear, hindering clinical translation. This study investigated the behavior of radical modulation by dose rate, oxygen and scavenger concentrations and modeled the experimental measuring supercoiled plasmid DNA damages. We developed a novel computational scheme combining microscopic Monte Carlo simulation (MMCS) and ordinary differential equations (ODEs) to allow modeling of radiation chemistry for the real experiment time duration.
Methods: We considered a cubic region of interest with 2 µm in size. 20 keV electrons were injected randomly in space and temporally following an exponential distribution with mean time between electrons 190 µs for FLASH dose rate 100 Gy/s and 190 ms for conventional dose rate. After each electron injection, MMCS captured initial radical heterogeneity within 2 µs, while ODEs modeled radical reactions over the remaining period between electrons. Periodic boundary condition was used. We constructed a model to compute intact supercoiled plasmid ratio Sp as a function of the integral of hydroxyl and hydrated electron concentrations with time and dose: Sp=exp(-α∫nOH⋅dt-β∫nehdt-γD). Model parameters were determined using data under conventional dose rate, and the model predictions at FLASH dose rate were compared to experimental results.
Results: The fitted parameters were α=34.7 μm3s-1, β=0.66 μm3s-1, γ=0.0018 Gy-1. Root mean square errors between model predictions and experimental results were 0.03 (conventional) and 0.04 (FLASH). The model predicts less supercoiled plasmid loss at FLASH dose rate with oxygen, which is consistent with experiments. Oxygen transformed hydrated electrons and hydrogen atoms of short lifetime into superoxide and hydroperoxyl of longer lifetime, enabling the inter-track recombination between radicals that can be modulated by dose rate.
Conclusion: This simulation framework effectively captured behavior of radiochemistry modulations caused by dose rate, oxygen and scavenger concentrations and predict supercoiled plasmid damages in FLASH condition.