Workshop report: identifying key issues underpinning the selection of linear or non-linear dose-response extrapolation for human health risk assessment of systemic toxicants.

The report of an Expert Panel convened by the National Research Council (NRC) of the National Academy of Sciences (NAS), entitled Science and Decisions: Advancing Risk Assessment (National Research Council, 2009a), includes a recommendation to use, as a default approach, low-dose linear extrapolation for systemic toxicity. This recommendation represents a significant departure from long-standing risk assessment practices for non-cancer toxicity, where the most appropriate No Observed Adverse Effect Level (NOAEL) or Benchmark Dose (BMD) of the critical effect in the key study is selected, and then a "safe exposure" level is derived by applying uncertainty factors to account for dataset completeness, potential greater sensitivity of humans when compared with experimental animals, and for potential variability of sensitivity in humans. A workshop was held to "frame" issues raised by the NAS report that needed further study. Workshop objectives included the following: (1) identify the issues raised by the 2009 NRC report and discuss the extent to which existing science may (or may not) align with the NAS analyses and recommendations, and (2) identify/develop possible actions to assist in advancing deeper and broader considerations of some of the critical issues presented by the 2009 NAS Panel. Experts invited to this "Framing" Workshop encompassed a full spectrum of toxicology and risk assessment disciplines; in particular, expertise in molecular interactions and dose-response of biological systems were well represented. The recommendations developed at this Framing Workshop provide specific ideas on possible further steps to facilitate deeper and broader consideration of the issues underpinning dose-response extrapolation in the risk assessment of systemic toxicants.