Is the science underlying recent alternatives for harmonizing quantitative risk approaches robust enough to replace toxicity thresholds in noncancer risk assessment?

SUMMARY Threshold and linear approaches to quantitative risk assessment are two fundamentally different paradigms that have coexisted for decades, primarily for noncancer and cancer risks, respectively. There have been several efforts in the risk assessment community over the years to “harmonize” cancer and noncancer risk assessment into a single unified approach. Two notable recent proposals for this are those by (1) the US National Academy of Sciences (NAS) and (2) the World Health Organization’s International Programme on Chemical Safety (WHO/IPCS). Both have recommended alternatives to traditional approaches for noncancer risk assessment. Herein, we review the salient aspects and ambitions of these two approaches to harmonizing cancer and noncancer dose–response analysis—the NAS “bottom-up” and the WHO/IPCS “top-down” proposals—focusing on how they apply to noncancer risk evaluation. We articulate issues and caveats and explore additional information needs. We also address how these proposals deal with the long-held concept in toxicology of exposure thresholds for noncancer effects, below which any substantial impact of exposures is not expected. We conclude there is solid evidence for the existence of thresholds and that the approach of estimating variation in susceptibility in these methods ought not be applied to low risk levels and very low exposures, since any such application would be expected to give answers that are unreliable and contingent mainly on the assumptions made rather than on the data about the agent’s actual dose–response properties. Unless and until thresholds are disproven or the reliability of these alternative methods improves, threshold-based approaches to noncancer risk assessment remain the most scientifically defensible.