Population variability in biological adaptive responses to DNA damage and the shapes of carcinogen dose–response curves

Carcinogen dose–response curves for both ionizing radiation and chemicals are typically assumed to be linear at environmentally relevant doses. This assumption is used to ensure protection of the public health in the absence of relevant dose–response data. A theoretical justification for the assumption has been provided by the argument that low dose linearity is expected when an exogenous agent adds to an ongoing endogenous process. Here, we use computational modeling to evaluate (1) how two biological adaptive processes, induction of DNA repair and cell cycle checkpoint control, may affect the shapes of dose–response curves for DNA-damaging carcinogens and (2) how the resulting dose–response behaviors may vary within a population. Each model incorporating an adaptive process was capable of generating not only monotonic dose–responses but also nonmonotonic (J-shaped) and threshold responses. Monte Carlo analysis suggested that all these dose–response behaviors could coexist within a population, as the spectrum of qualitative differences arose from quantitative changes in parameter values. While this analysis is largely theoretical, it suggests that (a) accurate prediction of the qualitative form of the dose–response requires a quantitative understanding of the mechanism, (b) significant uncertainty is associated with human health risk prediction in the absence of such quantitative understanding and (c) a stronger experimental and regulatory focus on biological mechanisms and interindividual variability would allow flexibility in regulatory treatment of environmental carcinogens without compromising human health.