Noncancer Risk Assessment: Principles and Practice in Environmental and Occupational Settings

Abstract

The approach to assessing the risks of noncancer toxicity has differed historically from that used to assess the potential risks of carcinogenicity. Assessment of risks of carcinogenicity has historically assumed that a small number of molecular events can evoke mutagenic changes in a single cell, ultimately leading to self-replicating damage and carcinogenicity. Generally, this is considered a nonthreshold effect because presumably all levels of exposure may pose a small, but finite, probability of generating a response. In contrast, it is most often assumed that noncarcinogenic changes have a threshold, a dose level below which a response is unlikely, because homeostatic, compensating, and adaptive mechanisms in the cell protect against toxic effects. Modern understanding of mode of action (MOA), loosely defined as how a chemical causes the observed effect, has led to refinements in this dichotomy. Rather than considering cancer versus noncancer effects, the focus is on whether or not the chemical causes its effects by a mutagenic MOA, specifically DNA interaction. Nonthreshold approaches are generally used for effects resulting from interaction with DNA, while effects resulting from a nonmutagenic MOA (including both cancer and noncancer endpoints) are generally evaluated using threshold approaches. A recent NRC publication [1], however, recommended linear extrapolation under certain conditions for noncancer endpoints that do not involved interaction with DNA. The issues raised by that publication are addressed later in this chapter. Recognizing both the historical approach and the importance of evaluation of MOA, this chapter will continue to use the term “noncancer risk,” but the methods described here should be understood to apply to both noncancer endpoints and cancer endpoints for which MOA information indicates that a threshold applies. This chapter describes the general framework for noncancer risk assessment and some salient principles for evaluating the quality of data and formulating judgments about the nature and magnitude of the hazard. Highlights of noncancer risk assessment methods used by a variety of agencies and organizations, and examples of how occupational risk assessment is moving toward a more systematic use of risk assessment principles are presented. This chapter also has several specific aims. The first is to provide scientifically supportable quantitative risk assessment procedures to meet the risk assessment goals listed in the following paragraph. A second aim is to provide a scientific rationale that may be used to determine whether new quantitative risk assessment procedures not specifically examined in this chapter are scientifically supportable. The final aim of this chapter is to provide a basis for developing new or improved quantitative risk assessment procedures. The quantitative risk assessment procedures described in this chapter have been developed to meet a variety of risk assessment goals. Although the protection of the public and occupational health are common themes that run through these separate risk assessment goals, the goals are sufficiently different to warrant separate and distinct procedures. Examples of such goals are as follows: to rank chemicals as to possible hazard; to prioritize chemicals for further evaluation, in combination with exposure information; to screen chemicals (e.g., new chemicals or ones under development), for purposes such as identifying which ones are appropriate for further development; to determine and/or estimate a level of daily exposure that is likely to be without an appreciable risk of deleterious effects during a lifetime; to determine and/or estimate the likely human response to exposure to various levels of a particular chemical or mixture. Moreover, differing amounts of toxicity data are needed for various quantitative procedures. Thus, both the problem being addressed and the amount of data available affect the choice of procedure. Keywords: benchmark dose; categorical regression; dose–response; dosimetry; hazard characterization; mode of action; physiologically based pharmacokinetic modeling; probabilistic RfD; problem formulation; risk characterization; uncertainty factor; weight of evidence