Approaches to modeling chemical reaction pathways in radiobiology

Recent interest in understanding effects of high dose-rate ('FLASH') radiobiology has prompted a number of groups to model the chemical reactions that might be involved, either to estimate radiolytic oxygen consumption in tissues, or the yields and persistence of specific reactive intermediates or products. However, most models have been either not biomimetic and/or inadequately supported by kinetic data. This review summarizes issues which should be addressed in developing models for chemical reactions in radiobiology.A model should be based on mechanistic pathways that lead to well-defined chemical and biological endpoints: crucially, the pathways should be plausibly similar in both the model and cells or tissues, and reflect the Law of Mass Action. Complex calculations of radiolytic yields are unnecessary, as reasonable estimates based on experimental data are generally available. Different parts of the intracellular milieu (such as the cytoplasm, nucleus, or phospholipid membranes) should be addressed separately, or with two-compartment models where appropriate. Homogeneous kinetics can be used as a first step in modeling, but the heterogeneity - both of radiolytic damage distribution and of cellular reactants - will need to be addressed. Major problems arise in choosing appropriate rate constants and estimating intracellular concentrations of reactants in the different organelles. It helps to identify and focus on the key reactions, as complex models may mask deficiencies and/or uncertainties; but it is still important to include all reactions and reactants that can have a significant effect on the model, as well as build upon experience in modeling chemical pathways in biology.