You are here:
HH: Toxicity Assessment
- Non-Cancer Effects
- Cancer Effects
- Toxicity Values
- Chemical-Specific Resources and Guidance
The basic objective of a toxicity assessment is to identify what adverse health effects a chemical causes and how the appearance of these adverse effects depends on exposure level (dose). The toxic effects of a chemical frequently depend on the route of exposure (oral, inhalation, dermal) and the duration of exposure (subchronic, chronic, or lifetime). Thus, a full description of the toxic effects of a chemical includes a listing of what adverse health effects the chemical may cause and how the occurrence of these effects depends upon dose, route, and duration of exposure.
The toxicity assessment process is usually divided into two parts: the first characterizes and quantifies the non-cancer effects of the chemical, while the second addresses the cancer effects of the chemical. This two-part approach is employed because there are typically major differences in the time-course of action and the shape of the dose-response curve for cancer and non-cancer effects.
Essentially all chemicals can cause non-cancer adverse health effects if given at a high enough dose. However, when the dose is sufficiently low, typically no adverse effect is observed. Thus, in characterizing the non-cancer effects of a chemical, the key parameter is the threshold dose at which an adverse effect first becomes evident. Doses below the threshold are considered to be safe, while doses above the threshold are likely to cause an effect.
The threshold dose is typically estimated from toxicological data (derived from studies of humans and/or animals) by determining the highest dose that does not produce an observable adverse effect and the lowest dose which does produce an effect. These are referred to as the "no-observed-adverse-effect-level" (NOAEL) and the "lowest-observed-adverse-effect-level" (LOAEL), respectively. The threshold is presumed to lie in the interval between the NOAEL and the LOAEL.
However, in order to be conservative (protective), non-cancer risk evaluations are not based directly on the threshold exposure level, but on a value referred to as the Reference Dose (RfD). The RfD is an estimate (with uncertainty spanning perhaps an order of magnitude) of a daily exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime.
The RfD is derived from the NOAEL (or the LOAEL if a reliable NOAEL is not available) by dividing by an "uncertainty factor." If the data are from studies in humans and if the observations are considered to be very reliable, the uncertainty factor may be as small as 1.0. However, the uncertainty factor is normally at least 10 and can be much higher if data are limited. The purpose of dividing the NOAEL or the LOAEL by an uncertainty factor is to ensure that the RfD is not higher than the true threshold level for adverse effects. Thus, there is always a "margin of safety" built into an RfD, and doses equal to or less than the RfD are nearly certain to be without any risk of adverse effect. Doses higher than the RfD may carry some risk, but because of the margin of safety, a dose above the RfD does not mean that an effect will necessarily occur.
For cancer effects, the toxicity assessment process has two components. The first is a qualitative evaluation of the weight of evidence (WOE) that the chemical does or does not cause cancer in humans. For chemicals that are believed to be capable of causing cancer in humans, the second part of the toxicity assessment is to describe the carcinogenic potency of the chemical. This is done by quantifying how the number of cancers observed in exposed animals or humans increases as the dose increases. Typically, it is assumed that the dose-response curve for cancer has no threshold (i.e., there is no dose other than zero that does not increase the risk of cancer), arising from the origin and increasing linearly until high doses are reached. Thus, the most convenient descriptor of cancer potency is the slope of the dose-response curve at low doses (where the slope is still linear). This is referred to as the Slope Factor (SF), which has dimensions of risk of cancer per unit dose.
Estimating the cancer Slope Factor is often complicated by the fact that observable increases in cancer incidence usually occur only at relatively high doses, frequently in the part of the dose-response curve that is no longer linear. Thus, it is necessary to use mathematical models to extrapolate from the observed high dose data to the desired (but unmeasurable) slope at low dose. In order to account for the uncertainty in this extrapolation process, EPA typically chooses to employ the upper 95th confidence limit of the slope as the Slope Factor. That is, there is a 95 percent probability that the true cancer potency is lower than the value chosen for the Slope Factor. This approach ensures that there is a margin of safety in cancer as well as non-cancer risk estimates.
You will need Adobe Reader to view some of the files on this page. See EPA’s About PDF page to learn more.
Toxicity values for use in Superfund human health risk assessments are available from several sources. The recommended hierarchy for these alternative sources is described in Human Health Toxicity Values in Superfund Risk Assessments (PDF) (Memorandum, OSWER 9285.7-53, December 2003) (4 pp, 228 K) and is summarized below:
Tier 1 - IRIS
The preferred source of toxicity data is EPA's Integrated Risk Information System (IRIS) database. Values in this database have been derived by expert toxicologists at EPA and most values have undergone thorough review and validation both within and outside EPA. If a toxicity value is available in IRIS, that value should be used in preference to any other value.
Tier 2 - PPRTVs
If toxicity values for a contaminant of potential concern are not available in IRIS, the next source to consult is EPA's Provisional Peer Reviewed Toxicity Values (PPRTVs). This source includes toxicity values that have been developed by the Office of Research and Development/National Center for Environmental Assessment/Superfund Health Risk Technical Support Center (STSC). This database is not available to the general public, but is accessible to EPA risk assessors via EPA's intranet at http://hhpprtv.ornl.gov/. Contact the Region 8 Superfund toxicologists for assistance with these values.
Tier 3 - Other Toxicity Values
Tier 3 includes additional EPA and non-EPA sources of toxicity information, including:
- The California EPA (Cal/EPA)'s Toxicity Criteria Database Exit contains toxicity values that are peer-reviewed and address both cancer and non-cancer effects.
- The Agency for Toxic Substances and Disease Registry (ATSDR)'s Minimal Risk Levels (MRLs) for Hazardous Substances Exit are peer-reviewed estimates of the daily human exposure to a hazardous substance that is likely to be without appreciable risk of adverse non-cancer health effects over a specified duration of exposure.
- Toxicity values in EPA's Health Effects Assessment Summary Tables (HEAST) (EPA-540-R-97-036, July 1997) are also Tier 3 values. The HEAST values on chemical contaminants are not currently available on an EPA internet site. They may be obtained by contacting a Superfund risk assessor.
In considering these different Tier 3 sources, priority should be given to those that are the most current, the basis for which is transparent and publicly available, and which have been peer-reviewed. Consultation with the STSC or headquarters program office is recommended regarding the use of the Tier 3 values for Superfund response decisions when the contaminant appears to be a risk driver for the site.
Toxicity Values for Radionuclides
Neither IRIS nor the PPRTV database contains radionuclide slope factors. Because EPA's Office of Radiation and Indoor Air (ORIA) obtains peer review on the radionuclide slope factors contained in the HEAST Radionuclides Table (formerly Table 4), routine consultation with STSC is generally not necessary on these values even when they may be a risk driver on a Superfund site. These radionuclide slope factors have been adopted by EPA in its Preliminary Remediation Goals for Radionuclides calculator and the Soil Screening Guidance for Radionuclides documents.
Details on the Methods Used to Derive Toxicity Factors
As noted above, development of human health toxicity factors is not a normal component of a Superfund risk assessment. Therefore, it is not necessary for Superfund risk assessors to be familiar with the detailed methods and approaches that are used. However, a basic understanding of these methods may be valuable in discussing the uncertainties associated with toxicity values. Links to documents that describe the methods for deriving toxicity factors include:
Guidelines for Carcinogen Risk Assessment (EPA/630/P-03/001F, March 2005) and Supplemental Guidance for Assessing Susceptibility from Early-Life Exposure to Carcinogens (EPA/630/R-03/003F, March 2005)
Methods for Derivation of Inhalation Reference Concentrations and Application of Inhalation Dosimetry (EPA/600/8-90/066F, October 1994)
The Benchmark Dose Approach
- The Use of the Benchmark Dose Approach in Health Risk Assessment (EPA/630/R-94/007, February 1995)
- Benchmark Dose Technical Guidance
- Benchmark Dose Software (BMDS)
Dioxin is the "shorthand" name for 2,3,7,8-tetrachlorodibenzodioxin (TCDD). This is the most potent of a series of related polychlorinated dibenzodioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs). This compound and its related congeners are often of special concern to EPA because dioxin has a very high potency for causing cancer and other adverse effects in exposed individuals. The following link provides a helpful introduction to dioxin:
Facts About Dioxin (Region 8, November 1999)
TEF/TEQ Approach for Dealing with Dioxin Mixtures
In general, the toxicity of the different PCDD and PCDF congeners depends on the number and arrangement of the chlorine atoms on the dibenzodioxin or dibenzofuran ring structures. The relative toxicity of a congener compared to TCDD is expressed in terms of a "toxicity equivalency factor" (TEF). The following links to the 2010 update of EPA's recommended TEF approach for dioxin:
Given measured concentration values for each congener (actually, groups of congeners) and the TEF for that group, the "toxicologically equivalent concentration" (TEQ) for a mixture of PCDDs and PCDFs is calculated as follows:
TEQ = Σ (TEFi × Ci)
TEFi = Toxicity equivalency factor for congener group "i"
Ci = Concentration of congener group "i"
The TEQ is the concentration of TCDD that is predicted to be of equal toxicity to the sum of the toxicity of all the different PCDDs and PCDFs present at the site. Risks from the calculated TEQ are therefore based on the toxicity factors for TCDD.
Toxicity of TCDD
The toxicity of TCDD has been and remains a complex topic, with many research publications on the nature of the adverse effects caused by exposure, the mechanism of action, and the dose-response relationships. EPA has recently updated its assessment of dioxin, and the findings and conclusions are presented in the following:
EPA's Reanalysis of Key Issues Related to Dioxin Toxicity and Response to NAS Comments, Volume 1 (PDF) (EPA/600/R-10/038F, February 2012) (344 pp, 4 MB)
The World Health Organization also has recently updated its assessment of dioxin, and their findings are summarized in the following:
Executive Summary: Assessment of the health risk of dioxins:re-evaluation of the Tolerable Daily Intake (TDI) (PDF) (WHO Consultation, Geneva, Switzerland, May 1998) (28 pp, 179K) Exit
Background Levels of Dioxins
One of the most difficult issues that must be considered in any risk assessment for dioxins and furans is the fact the both PCDDs and PCDFs are formed under a wide variety of conditions, both from anthropogenic and natural sources. Thus, PCDDs and PCDFs are ubiquitous in the environment, and simply because they are observed at a site does not imply that a site-specific release has occurred. There are a number of documents that provide useful information on the level of PCDDs and/or PCDFs in soil and other media under "background" conditions, including the following:
USA Background Dioxin Soils Concentrations (Region 8, November 1999)
Denver Front Range Study of Dioxins and Surface Soil
Appendices are available upon request. Please contact Susan Griffin (email@example.com).
EPA has developed several guidance documents in order to help risk assessors and risk managers evaluate whether it is necessary to perform a detailed investigation of dioxins in site media. For soil, EPA recommends that 1 ppb (TEQs) be used as a screening level of concern for residential land uses. For commercial/industrial exposure scenarios, a soil level within the range of 5 ppb to 20 ppb (TEQs) is recommended unless extenuating site-specific circumstances warrant a different level. More details are presented in the following links:
Approach for Addressing Dioxin in Soil at CERCLA and RCRA Sites (PDF) (Memorandum from T. Fields, Jr., OSWER 9200.4-26, April 1998) (6 pp, 43 K)
For groundwater, the default remedial level is the Maximum Contaminant Level (MCL) for dioxin, which is published in the National Primary Drinking Water Regulations:
Polychlorinated biphenyls (PCBs) are another class of compounds consisting of many different congeners that differ with respect to the number and placement of chlorine atoms on the biphenyl ring structure. Although less potent than dioxins, PCBs are thought to act via a mechanism similar to dioxins, and PCBs cause many of the same types of adverse effects as dioxins. Useful data on the toxicity of PCBs are presented in the following:
PCBs: Structure-Function Relationships and Mechanism of Action (S. Safe et al., Environmental Health Perspectives 60:47-56, 1985)
Toxicity data for PCBs are based mainly on studies of commercial PCB mixtures (Aroclors), with toxicity tending to increase as function of the chlorine content of the mixture. However, when commercial PCB mixtures are released into the environment, differences between congeners with respect to properties such as volatility, solubility, chemical and biological degradability, and bioaccumulation tend to cause the composition of the PCBs in soil, water, and animal tissue (e.g., fish) to differ from the composition of the original Aroclor mixture. For this reason, it is inherently difficult to use toxicity data from Aroclor mixtures to predict the toxicity of environmental PCBs. To help address this problem, EPA has developed two alternative strategies for evaluating risks from environmental PCB mixtures.
The first approach is essentially the same as for dioxins (see above). That is, the World Health Organization has developed Toxicity Equivalency Factors (TEFs) for PCB congener groups. If data on the concentration levels of the various PCB congener classes are available, the total toxicity of the PCB mixture may be estimated from the slope factor for TCDD using the TEQ approach.
The second approach is based on studies of PCB (rather than dioxin) toxicity, taking into account the effect of environmental fate processes and the expected nature of PCB mixtures that are expected to occur in water and soil. The details of this method are presented in the following:
PCBs: Cancer Dose-Response Assessment and Application to Environmental Mixtures (EPA/600/P-96/001F, September 1996)
Because of the tendency of PCBs to bioaccumulate in the food chain, one of the most important routes by which humans may be exposed to PCBs is via fish ingestion. Guidance on the level of concern for PCBs in fish tissue is provided in the following:
Polychlorinated Biphenyls (PCBs) Update: Impact on Fish Advisories (PDF) (EPA-823-F-99-019, September 1999) (7 pp, 107 K)
The amount of chemical that is actually absorbed into the body (bioavailability) is an important consideration in the risk assessment process, especially for exposures to metals in soil. A description of the issues and a summary of valuable bioavailability data assembled by EPA Region 8 and others is presented on the Bioavailability page.