Resolving an Open Science-policy question: Should the LNT still be an omnibus regulatory assumption?

Abstract

Regulatory analyses, modeling the carcinogenic effect of ionizing radiations (IR) (e.g., alpha and beta particles, x-, and gamma rays, neutrons) and chemicals continue to use the linear no-threshold (LNT) model from zero to some low dose. The LNT is an omnibus causal default in regulatory occupational and health risk analysis. Its use raises four issues that make this default an open question. The first is that the LNT applied to study a single agent excludes co-exposure to other known risk factors: physical, dietary, socio-economic, and other. Causation is inappropriately specified because cancer incidence is imputed to the single agent's doses, although most cancers are multifactorial diseases. The second, linear interpolation from high to zero dose and response, is incorrect because biological and epidemiological evidence identify different mechanisms and modes of action at those doses. Third, additivity of exposure effect to background effect is questionable and certainly variable. Fourth, the default overestimates the probabilities and consequences at low doses, supplanting rational decision-making in which alternative models may be more or less likely to be correct. Recent converging scientific evidence against the LNT hypothesis answers the open question. The LNT use in regulation conflates science with administrative ease and risk aversion by policymakers. It should be replaced by models that are based on biologically motivated mechanistic understandings within an evolutionary biology framework that integrates adaptive strategies/processes in their formulation.

Keywords: Chemicals; Dose-response; Epidemiology; LNT; Policy; Radiations.