Advanced Computational Approaches for Characterizing Stochastic Cellular Responses to Low Dose, Low Dose Rate Exposures

Abstract

OAK - B135 This project final report summarizes modeling research conducted in the U.S. Department of Energy (DOE), Low Dose Radiation Research Program at the Lovelace Respiratory Research Institute from October 1998 through June 2003. The modeling research described involves critically evaluating the validity of the linear nonthreshold (LNT) risk model as it relates to stochastic effects induced in cells by low doses of ionizing radiation and genotoxic chemicals. The LNT model plays a central role in low-dose risk assessment for humans. With the LNT model, any radiation (or genotoxic chemical) exposure is assumed to increase one¡¯s risk of cancer. Based on the LNT model, others have predicted tens of thousands of cancer deaths related to environmental exposure to radioactive material from nuclear accidents (e.g., Chernobyl) and fallout from nuclear weapons testing. Our research has focused on developing biologically based models that explain the shape of dose-response curves for low-dose radiation and genotoxic chemical-induced stochastic effects in cells. Understanding the shape of the dose-response curve for radiation and genotoxic chemical-induced stochastic effects in cells helps to better understand the shape of the dose-response curve for cancer induction in humans. We have used a modeling approach that facilitated model revisions over time, allowing for timely incorporation of new knowledge gained related to the biological basis for low-dose-induced stochastic effects in cells. Both deleterious (e.g., genomic instability, mutations, and neoplastic transformation) and protective (e.g., DNA repair and apoptosis) effects have been included in our modeling. Our most advanced model, NEOTRANS2, involves differing levels of genomic instability. Persistent genomic instability is presumed to be associated with nonspecific, nonlethal mutations and to increase both the risk for neoplastic transformation and for cancer occurrence. Our research results, based on applications of NEOTRANS2, indicate that nonlinear threshold-type, dose-response relationships for excess stochastic effects (problematic nonlethal mutations, neoplastic transformation) should be expected after exposure to low linear energy transfer (LET) gamma rays or gamma rays in combination with high-LET alpha radiation. Similar thresholds are expected for low-dose-rate low-LET beta irradiation. We attribute the thresholds to low-dose, low-LET radiation induced protection against spontaneous mutations and neoplastic transformations. The protection is presumed mainly to involve selective elimination of problematic cells via apoptosis. Low-dose, low-LET radiation is presumed to trigger wide-area cell signaling, which in turn leads to problematic bystander cells (e.g., mutants, neoplastically transformed cells) selectively undergoing apoptosis. Thus, this protective bystander effect leads to selective elimination of problematic cells (a tissue cleansing process in vivo). However, this protective bystander effects is a different process from low-dose stimulation of the immune system. Low-dose, low-LET radiation stimulation of the immune system may explain why thresholds for inducing excess cancer appear much larger (possibly more than 100-fold larger) than thresholds for inducing excess mutations and neoplastic transformations, when the dose rate is low. For ionizing radiation, the current risk assessment paradigm is such that the relative risk (RR) is always ¡Ý 1, no matter how small the dose. Our research results indicate that for low-dose or low-dose-rate, low-LET irradiation, RR < 1 may be more the rule than the exception. Directly tied to the current RR paradigm are the billion-dollar cleanup costs for radionuclide-contaminated DOE sites. Our research results suggest that continued use of the current RR paradigm for which RR ¡Ý 1 could cause more harm than benefit to society (e.g., by spreading unwarranted fear about phantom excess risks associated with low-dose low-LET radiation). Such phantom risks also may arise from risk assessments conducted for combined exposure to low- and high-LET radiations when based on the LNT or other models that exclude RR < 1. Our results for high-LET radiation are consistent with the LNT hypothesis but only where there is no additional low-LET contribution (e.g., gamma rays) to the total dose. For high-LET neutron sources, gamma rays arise (especially in vivo) for large mammals such as humans from neutron interactions with tissue. The gamma rays might provide some protection from low-dose-related stochastic effects via inducing the protective bystander apoptosis effect that is considered to contribute to tissue cleansing via removal of problematic cells.