NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Research Council (US) Committee on Applications of Toxicogenomic Technologies to Predictive Toxicology. Applications of Toxicogenomic Technologies to Predictive Toxicology and Risk Assessment. Washington (DC): National Academies Press (US); 2007.

Cover of Applications of Toxicogenomic Technologies to Predictive Toxicology and Risk Assessment

Applications of Toxicogenomic Technologies to Predictive Toxicology and Risk Assessment.

Show details

COverview of Risk Assessment

The objective of chemical risk assessment methodologies is to facilitate both scientific and data-informed decision making and also is increasingly expected to provide more accurate predictions of actual risk. The validity of a prediction of risk derived from a risk assessment depends largely on the quality and accuracy of data. Where data do not exist or are contradictory, regulatory agencies are forced to rely on default values, uncertainty factors, and modeling approaches to fill in the blanks. These defaults and extrapolations introduce uncertainty into the risk estimates. New methodologies and testing methods could fill key data gaps, clarify data inconsistencies, or otherwise reduce uncertainty. If applied appropriately, these approaches have the potential to improve the accuracy and scientific credibility of regulatory decision making.

An Overview of Current Risk Assessment Practice

Government agencies charged with protecting public and worker health are required to review, quantify, and ultimately regulate chemicals, physical agents, and pharmaceuticals in a manner that will protect and enhance the public health and the environment. One of these regulatory responsibilities is to assess the risk to human health from chemical exposures. This section provides a brief overview of the aspects of regulatory risk assessment practices most relevant to toxicogenomic technologies. Further details about the risk-assessment process can be found in several references (EPA 1986 , 2004 , 2005; PCCRARM 1997).

Human health risk assessment is the process of analyzing information to determine whether an environmental hazard might cause harm to exposed persons (EPA 2004). The risk-assessment process integrates many disciplines of toxicology. It has both qualitative and quantitative components and consists of four general steps: hazard identification, dose-response assessment, exposure assessment, and risk characterization (NRC 1983, 1994).

Step 1: Hazard identification entails identifying the contaminants that are suspected to pose health hazards, quantifying the concentrations at which they are present in the environment, describing the specific forms of toxicity (neurotoxicity, carcinogenicity, etc.) that the contaminants of concern can cause, and evaluating the conditions under which these forms of toxicity might be expressed in exposed humans.

Step 2: Dose-response assessment entails further evaluating the conditions under which the toxic properties of a chemical might be manifested in exposed people, with particular emphasis on the quantitative relation between the dose and the toxic response. The development of this relationship may involve the use of mathematical models. This step may include an assessment of variations in response—for example, differences in susceptibility between young and old people.

Step 3: Exposure assessment involves specifying the population that might be exposed to the agent of concern, identifying the routes through which exposure can occur, and estimating the magnitude, duration, and timing of the doses that people might receive as a result of their exposure.

Step 4: Risk characterization involves integrating information from the first three steps to develop a qualitative or quantitative estimate of the likelihood that any of the hazards associated with the agent of concern will be realized in exposed people. This is the step in which risk-assessment results are expressed. Risk characterization should also include a full discussion of the uncertainties associated with the estimates of risk (Adapted from NRC 1994).

Toxicogenomic information has a potential role in all aspects of the risk-assessment process. For example, in hazard identification, toxicogenomic data could inform the types of hazard a chemical presents (for example, whether it poses cancer or noncancer risks) and the modes and mechanisms of toxic action1 through which it acts. Information on the mode of action is also a component in deciding the appropriate approach to dose-response assessment (described further below). Toxicogenomic approaches could support exposure assessment by indicating cellular exposure to toxicants. Toxicogenomic data may also be used to better understand areas of uncertainty, including variability in the human population, extrapolation of data from one species to another, identification of susceptible subpopulations, and provision of quantitative data to improve risk assessments.

Quantification of Risk

Different analytical techniques are used in cancer and noncancer risk assessments to quantify risk. The end result of a noncancer risk assessment can be the determination of a quantitative human reference dose (RfD) for oral exposures or a reference concentration (RfC) for inhalation exposures (both are generically referred to as a “reference value”).2

As summarized by the U.S. Environmental Protection Agency (EPA) (EPA 2004): When developing a noncancer reference value (a RfD or RfC) for a chemical substance, EPA surveys the scientific literature and selects a critical study and a critical effect. The critical effect is defined as the adverse effect, or its known precursor, that occurs at the lowest dose identified in the most sensitive species as the dose rate of an agent increases. When a no-observed-adverse-effect level (NOAEL)3 can be identified in a critical study, it becomes the basis for the reference value derivation. If NOAEL cannot be identified, then a lowest-observed-adverse-effect level (LOAEL)4 is identified instead. Recently, benchmark doses (BMDs) (EPA 2000) from the modeling of dose-response data have been used instead of the traditional NOAEL/LOAEL approach; however, most RfDs are based on NOAELs (EPA 2004). The NOAEL, LOAEL, or BMD is divided by appropriate uncertainty factor to derive the final reference value. The uncertainty factors are generally 10-fold (but can be higher or lower if informed by data) and are intended to account for uncertainty in the available data from (1) variation in the human population, (2) extrapolating animal data to humans, (3) extrapolating from less-than-lifetime exposures to lifetime exposure, (4) extrapolating from a LOAEL rather than from a NOAEL, and (5) using incomplete databases.5 For carcinogenic compounds, data from human epidemiologic studies are preferred, but, in the absence of human epidemiologic data, animal data are used.6 Dose-response curves are constructed from these studies; however, the range of doses is frequently above the levels of environmental interest. To estimate the risks below the levels tested, the observed data are used to derive a point of departure7 followed by extrapolation to lower exposures (EPA 2005). Linear or nonlinear approaches can be used to extrapolate to low doses, and the choice of methods is critical because the derived risk estimates vary by technique. In general, linear approaches produce more conservative risk estimates than nonlinear approaches (NRC 2006).

The selection of the various models used to extrapolate to low doses is informed by a compound’s mode of action. The EPA cancer guidance (EPA 2005) states that “when available data are insufficient to establish the mode of action for a tumor site and when scientifically plausible based on the available data, linear extrapolation is used as a default approach.”8 Further, “A nonlinear approach should be selected when there are sufficient data to ascertain the mode of action and conclude that it is not linear at low doses and the agent does not demonstrate mutagenic or other activity consistent with linearity at low doses.” For cancer risk assessments, extrapolations from the point of departure can be used to calculate a cancer slope factor9 (for linear extrapolation) and a RfD or RfC among other outputs (for nonlinear extrapolation) (EPA 2005).

The values derived from the cancer and noncancer risk assessments are used to protect the public from unacceptable chemical exposures and can be the basis of regulatory decision making (for example, in establishing standards for water and air quality and requirements for environmental cleanup). These values have a range of implications to stakeholders and, because of the uncertainty inherent in the risk assessment process, their derivation can be quite controversial (e.g., NRC 1999 , 2001 , 2005 , 2006). As a result, tools such as toxicogenomics that can be used in the risk-assessment process to increase the certainty of risk estimates are of great importance for protecting public health.


  1. EPA (U.S. Environmental Protection Agency). Guidelines for Carcinogen Risk Assessment EPA/630/R-00/004. Washington, DC: Risk Assessment Forum, U.S. Environmental Protection Agency; 1986.
  2. EPA (U.S. Environmental Protection Agency). External Review Draft. Washington, DC: Risk Assessment Forum, U.S. Environmental Protection Agency; October 2000, 2000. [accessed April 13, 2007]. Benchmark Dose Technical Guidance Document. EPA/630/R-00/001. [online]. Available:http://www​​/pdfs/bmds/BMD-External​_10_13_2000.pdf.
  3. EPA (U.S. Environmental Protection Agency). External Review Draft. Washington, DC: Office of Research and Development, U.S. Environmental Protection Agency; August 2001, 2001. Trichloroethylene Health Risk Assessment: Synthesis and Characterization. EPA/600/P-01/002A.
  4. EPA (U.S. Environmental Protection Agency). Prepared for the U.S. Environmental Protection Agency by Members of the Risk Assessment Task Force. Washington, DC: Office of the Science Advisor, U.S. Environmental Protection Agency; 2004. [accessed Jan.13, 2006]. An Examination of EPA Risk Assessment Principles and Practices. Staff Paper. EPA/100/B-04/001. [online]. Available: http://www​​.pdf.
  5. EPA (U.S. Environmental Protection Agency). Guidelines for Carcinogen Risk Assessment. Washington, DC: Risk Assessment Forum, U.S. Environmental Protection Agency; March 2005, 2005. [accessed April 13, 2007]. EPA/630/P-03/001F. [online]. Available:http://www​​/iris/cancer032505.pdf.
  6. EPA (U.S. Environmental Protection Agency). Glossary of IRIS Terms. Integrated Risk Information System, U.S. Environmental Protection Agency; 2007. [accessed April 13, 2007]. [online]. Available:http://www​
  7. NRC (National Research Council). Risk Assessment in the Federal Government: Managing the Process. Washington, DC: National Academy Press; 1983. [PubMed: 25032414]
  8. NRC (National Research Council). Science and Judgment in Risk Assessment. Washington, DC: National Academy Press; 1994.
  9. NRC (National Research Council). Arsenic in Drinking Water. Washington, DC: National Academy Press; 1999.
  10. NRC (National Research Council). Arsenic in Drinking Water: 2001 Update. Washington, DC: National Academy Press; 2001.
  11. NRC (National Research Council). Health Implications of Perchlorate Ingestion. Washington, DC: National Academy Press; 2005.
  12. NRC (National Research Council). Assessing the Human Health Risks of Trichloroethylene: Key Scientific Issues. Washington, DC: The National Academies Press; 2006.
  13. PCCRARM (Presidential/Congressional Commission on Risk Assessment and Risk Management). Framework for Environmental Health Risk Management. Vol. 1. Washington, DC: The Commission; 1997.



The EPA (EPA 2005) provides the following definitions: “The term ‘mode of action’ is defined as a sequence of key events and processes, starting with interaction of an agent with a cell, proceeding through operational and anatomical changes, and resulting in cancer formation. A ‘key event’ is an empirically observable precursor step that is itself a necessary element of the mode of action or is a biologically based marker for such an element. Mode of action is contrasted with ‘mechanism of action,’ which implies a more detailed understanding and description of events, often at the molecular level, than is meant by mode of action. The toxicokinetic processes that lead to formation or distribution of the active agent to the target tissue are considered in estimating dose but are not part of the mode of action as the term is used here. There are many examples of possible modes of carcinogenic action, such as mutagenicity, mitogenesis, inhibition of cell death, cytotoxicity with reparative cell proliferation, and immune suppression.”


Reference concentration (RfC): An estimate (with uncertainty spanning perhaps an order of magnitude) of a continuous inhalation exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime. Reference dose (RfD): An estimate (with uncertainty spanning perhaps an order of magnitude) of a daily oral exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime. (Source: EPA 2007)


The highest exposure level at which there are no statistically or biologically significant increases in the frequency or severity of adverse effects between the exposed population and its appropriate control is called the no-observed-adverse-effect level (NOAEL) (EPA 2004).


“A LOAEL is the lowest exposure level at which there are biologically significant increases (with or without statistical significance) in frequency or severity of adverse effects between the exposed population and its appropriate control group. The NOAEL is generally presumed to lie between zero and the LOAEL, so an UF [uncertainty factor] (generally 10 but sometimes 3 or 1) is applied to the LOAEL to derive a nominal NOAEL” (EPA 2004).


In some cases, the largest divisor the EPA will use is 3,000 because of the uncertainty when so many uncertainty factors are applied (for example, see the risk assessment for trichloroethylene (EPA 2001).


As described by the EPA (2005): “When animal studies are the basis of the analysis, the estimation of a human-equivalent dose should utilize toxicokinetic data to inform cross-species dose scaling if appropriate and if adequate data are available. Otherwise, default procedures [described the EPA 2005] should be applied.”


As described by the EPA (2005): “A ‘point of departure’ (POD) marks the beginning of extrapolation to lower doses. The POD is an estimated dose (usually expressed in human-equivalent terms) near the lower end of the observed range, without significant extrapolation to lower doses.”


The 2005 Cancer Guidance also states: “Linear extrapolation should be used when there are MOA [mode of action] data to indicate that the dose-response curve is expected to have a linear component below the POD [point of departure]. Agents that are generally considered to be linear in this region include: agents that are DNA-reactive and have direct mutagenic activity, or agents for which human exposures or body burdens are high and near doses associated with key precursor events in the carcinogenic process, so that background exposures to this and other agents operating through a common mode of action are in the increasing, approximately linear, portion of the dose-response curve.”


An upper bound, approximating a 95% confidence limit, on the increased cancer risk from a lifetime exposure to an agent. This estimate is usually expressed in units of proportion (of a population) affected per mg/kg-day (Adapted from EPA 2007).

Copyright © 2007, National Academy of Sciences.
Bookshelf ID: NBK10201


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (2.1M)

Related information

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...