EPA/1 OO/R 10/005 | December 2010 | www.epa.gov/osa
United States
Environmental Protection
Agency
Recommended Toxicity Equivalence
Factors (TEFs) for Human Health Risk
Assessments of 2,3,7,8-
Tetrachlorodibenzo-p-dioxin and
Dioxin-Like Compounds
Office of the Science Advisor
Risk Assessment Forum
-------
EP A/100/R-10/005
December 2010
Recommended Toxicity Equivalence Factors
(TEFs) for Human Health Risk Assessments
of 2,357,8-Tetrachlorodibenzo-/;-dioxin and
Dioxin-Like Compounds
Risk Assessment Forum
U.S. Environmental Protection Agency
Washington, DC 20460
-------
NOTICE
This report has been subjected to the Agency's peer and administrative review and has
been approved for publication as an EPA document. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
ABSTRACT
This document describes the U.S. Environmental Protection Agency's (EPA's) updated
approach for evaluating the human health risks from exposures to environmental media
containing dioxin-like compounds (DLCs). 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and
DLCs are structurally and lexicologically related halogenated aromatic hydrocarbons. The EPA
recommends that the toxicity equivalence factor (TEF) methodology, a component mixture
method, be used to evaluate human health risks posed by these mixtures, using TCDD as the
index chemical. The EPA recommends the use of the consensus TEF values for TCDD and the
DLCs published in 2005 by the World Health Organization. EPA Program Offices and Regions
have historically used TEF values in their risk assessments; this document recommends the 2005
WHO consensus TEFs, but does not address specific risk assessment applications of TEFs. The
EPA recommends these TEFs be used for all effects mediated through aryl hydrocarbon receptor
binding by the DLCs including cancer and noncancer effects. Using information that
summarizes the range of relative toxi cities of the DLCs, the EPA recommends that, for major
risk assessments as determined by U.S. EPA Program Offices or Regions, the conduct of a
sensitivity analysis be considered to illustrate the impact the TEFs have on the toxicity
equivalence (TEQ) value. The EPA will update all of these recommendations in the future based
on the evaluation of new toxicity data for the DLCs, updates to available relative potency (ReP)
data, including statistical summaries of RePs for individual DLCs, and the results of new
consensus processes undertaken to update the TEF approach.
Preferred citation:
U.S. EPA (Environmental Protection Agency). (2010) Recommended Toxicity Equivalence Factors (TEFs) for
Human Health Risk Assessments of 2,3,7,8-Tetrachlorodibenzo-p-dioxin and Dioxin-Like Compounds. Risk
Assessment Forum, Washington, DC. EPA/600/R-10/005.
11
-------
TABLE OF CONTENTS
Page
LIST OF TABLES iv
LIST OF ABBREVIATIONS v
LIST OF ABBREVIATIONS OF DIOXINS AND DIOXIN-LIKE COMPOUNDS vi
KEY TERMS vii
PREFACE viii
AUTHORS, TECHNICAL PANEL, AND REVIEWERS ix
INTRODUCTION 1
THE TEF METHODOLOGY 2
BACKGROUND 4
UNCERTAINTIES IN THE TEF APPROACH 10
UNCERTAINTY IN TEF METHOD AS SUMPTIONS 10
UNCERTAINTY IN THE PROCESSES AND DATA USED TO DERIVE TEFs 11
RECOMMENDATIONS 12
SENSITIVITY ANALYSIS 15
SENSITIVITY ANALYSIS LIMITATIONS 17
CONCLUSIONS 23
REFERENCES 25
in
-------
LIST OF TABLES
No. Title Page
1. Background and history of TEFs for risk assessment of DLCs 5
2. Recommended toxicity equivalence factors (TEFs) for human health risk
assessment of poly chlorinated dibenzo-^-dioxins, dibenzofurans, and
dioxin-like poly chlorinated biphenyls 13
3. Percentiles of in vivo ReP values 18
4. Percentiles of combined in vivo and in vitro ReP values 20
5. Summary of risk characterization recommendations for TEF applications 24
IV
-------
LIST OF ABBREVIATIONS
AhR aryl hydrocarbon receptor
DLC dioxin-like compound
ECEH European Centre for Environmental Health
EDso effective dose that causes an effect in 50% of the test units
IPCS International Programme on Chemical Safety
NAS National Academy of Science
ReP relative potency or relative effect potency
RePiggy World Health Organization ReP database developed in 1997
TCDD 2,3,7,8-tetrachlorodibenzo-^-dioxin
TEF toxicity equivalence factor
TEQ toxicity equivalence
EPA U.S. Environmental Protection Agency
WHO World Health Organization
-------
LIST OF ABBREVIATIONS OF DIOXINS AND DIOXIN-LIKE COMPOUNDS
Poly chlorinated biphenyls:
TCB tetrachlorinated biphenyl
PeCB pentachlorinated biphenyl
HxCB hexachlorinated biphenyl
HpCB heptachlorinated biphenyl
OCB octachlorinated biphenyl
PCB polychlorinated biphenyl
Poly chlorinated dibenzo-p-dioxins:
TCDD tetrachlorinated dibenzo^p-dioxin
PeCDD pentachlorinated dibenzo-p-dioxin
HxCDD hexachlorinated dibenzo-p-dioxin
HpCDD heptachlorinated dibenzo-p-dioxin
OCDD octachlorinated dibenzo-^-dioxin
PCDD polychlorinated dibenzo-p-dioxin
Polychlorinated dibenzofurans:
TCDF tetrachlorinated dibenzofuran
PeCDF pentachlorinated dibenzofuran
HxCDF hexachlorinated dibenzofuran
HpCDF heptachlorinated dibenzofuran
OCDF octachlorinated dibenzofuran
PCDF polychlorinated dibenzofuran
VI
-------
KEY TERMS
Dioxin-like: A description used for compounds that have chemical structures, physico-chemical
properties, and toxic responses similar to 2,3,7,8-tetrachlorodibenzo-/>-dioxin (TCDD). Because
of their hydrophobic nature and resistance towards metabolism, these chemicals persist and
bioaccumulate in fatty tissues of animals and humans. Certain members of the dioxin, furan, and
polychlorinated biphenyl (PCB) family are termed "dioxin-like" in this document and are
assigned toxic equivalence factor (TEF) values.
Index Chemical: The chemical selected as the basis for standardization of toxicity of
components in a mixture. The index chemical must have a clearly defined dose-response
relationship. For dioxin like compounds (DLCs), TCDD is typically specified as the index
chemical. (In some studies used to develop RePs, PCBi26 has been used as the index chemical.)
Relative Potency (ReP): The ratio of the potency of a compound to the standard toxicant in that
specific study; a concept similar to toxic equivalence but based on a single study, species, or
matrix, etc., and not integrated with other RePs to obtain a general TEF.
Toxic Equivalence Factors (TEFs): TEFs are consensus estimates of compound-specific
toxicity/potency relative to the toxicity/potency of an index chemical. TEFs are the result of
expert scientific judgment using all of the available data and taking into account uncertainties in
the available data.
Toxic Equivalence (TEQ): TEQ is the product of the concentration of an individual DLC in an
environmental mixture and its corresponding TCDD TEF for that compound.
vn
-------
PREFACE
This document updates the U.S. Environmental Protection Agency's (EPA's) approach
for evaluating the human health risks from exposures to environmental media containing
2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and dioxin-like compounds (DLCs). It is intended
for guidance only. It provides guidance to EPA Regional and Program Offices. EPA Program
Offices and Regions have historically used TEF values in their risk assessments; this document
recommends the 2005 WHO consensus TEFs, but does not address specific risk assessment
applications of TEFs. It does not establish any substantive "rules" under the Administrative
Procedure Act or any other law and will have no binding effect on EPA or any regulated entity.
Rather, it represents a statement of current policy. The EPA's National Center for
Environmental Assessment developed the initial draft of this document, which was then
reviewed and completed by a Technical Panel under the auspices of EPA's Risk Assessment
Forum. EPA made the document available for public comment during a 30 day public comment
period in September 2009, and an expert peer-review panel discussed the document in a
teleconference open to the public on October 22, 2009. The public comments received by EPA
were provided to the peer-review panel members prior to the October 2009 teleconference for
their consideration in making comments and recommendations to EPA. The peer-review report,
and EPA response to comments, is available at http://www.epa.gov/raf/hhtefguidance/index.htm.
The Risk Assessment Forum was established to promote scientific consensus within EPA
on difficult and controversial risk assessment issues and to ensure that this consensus is
incorporated into appropriate risk assessment guidance. To accomplish this, the Risk
Assessment Forum assembles experts from throughout EPA in a formal process to study and
report on these issues from an Agency-wide perspective.
Vlll
-------
AUTHORS, TECHNICAL PANEL, AND REVIEWERS
This document was prepared by authors from EPA's Office of Research and
Development and was then reviewed and completed by a Technical Panel under the auspices of
EPA's Risk Assessment Forum.
AUTHORS
Belinda Hawkins, U.S. EPA, Office of Research and Development, National Center for
Environmental Assessment, Cincinnati, OH 45268
Janet Hess-Wilson, U.S. EPA, Office of Research and Development, National Center for
Environmental Assessment, Cincinnati, OH 45268
Glenn Rice (Document Co-lead), U.S. EPA, Office of Research and Development, National
Center for Environmental Assessment, Cincinnati, OH 45268
Jeff Swartout, U.S. EPA, Office of Research and Development, National Center for
Environmental Assessment, Cincinnati, OH 45268
Linda K. Teuschler (Document Co-lead), U.S. EPA, Office of Research and Development,
National Center for Environmental Assessment, Cincinnati, OH 45268
TECHNICAL PANEL
Randy Wentsel (Chair), U.S. EPA, Office of Research and Development, Office of the Assistant
Administrator, Washington, DC 20460
Marlene Berg, U.S. EPA, Office of Solid Waste and Emergency Response, Office of Superfund
Remediation Technology Innovation, Washington, DC 20460
David E. Cooper, U.S. EPA, Office of Solid Waste and Emergency Response, Office of
Superfund Remediation Technology Innovation, Washington, DC 20460
Michael DeVito, National Institute of Environmental Health Sciences, National Institutes of
Health, Research Triangle Park, NC 27709 (formerly a member of U.S. Environmental
Protection Agency, National Health and Environmental Effects Research Laboratory)
Tala Henry, U.S. EPA, Office of Prevention, Pesticides and Toxic Substances, Office of
Pollution Prevention and Toxics, Risk Assessment Division, Washington, DC 20460
Margaret McDonough, U.S. EPA, Region 1, Boston, MA 02114
Marian Olsen, U.S. EPA, Region 2, New York, NY 10007
Glenn Rice, U.S. EPA, Office of Research and Development, National Center for Environmental
Assessment, Cincinnati, OH 45268
Dan Stralka, U.S. EPA, Region 9, San Francisco, CA, 94105
Linda K. Teuschler, U.S. EPA, Office of Research and Development, National Center for
Environmental Assessment, Cincinnati, OH 45268
Dwain Winters, U.S. EPA, Office of Prevention, Pesticides and Toxic Substances, Retired
IX
-------
AUTHORS, TECHNICAL PANEL, AND REVIEWERS cont.
RISK ASSESSMENT FORUM STAFF
Seema Schappelle, U.S. EPA, Office of the Science Advisor, Washington, DC 20460
Kathleen Raffaele, U.S. EPA, Office of the Science Advisor, Washington, DC 20460
Elizabeth Lee Hofmann, U.S. EPA, Office of the Science Advisor, Washington, DC 20460
This document has been provided for review to EPA scientists and interagency reviewers
from other federal agencies and White House offices.
INTERNAL PEER REVIEWERS
Mike DeVito, National Institute of Environmental Health Sciences, National Institutes of Health,
Research Triangle Park, NC 27709 (formerly a member of U.S. Environmental Protection
Agency, National Health and Environmental Effects Research Laboratory)
Lynn Flowers, U.S. EPA, Office of Research and Development, National Center for
Environmental Assessment, Washington, DC 20460
Glenn Suter, U.S. EPA, Office of Research and Development, National Center for
Environmental Assessment, Cincinnati, OH 45268
EXTERNAL PEER REVIEWERS
Peter L. deFur (lead), Environmental Stewardship Concepts, President, Henrico, VA 23238
Moiz Mumtaz, Agency for Toxic Substances and Disease Registry (ATSDR), Atlanta, GA
30341
Thomas B. Starr, TBS Associates, Raleigh, NC 27615
Martin van den Berg, University of Utrecht, Utrecht, Netherlands
Nigel J. Walker, National Institute of Environmental Health Sciences, Environmental
Toxicology Program, Research Triangle Park, NC 27709
-------
INTRODUCTION
This document describes the U.S. Environmental Protection Agency's (EPA's) updated
approach for evaluating the human health risks from exposures to environmental media
containing 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and dioxin-like compounds (DLCs).
TCDD and DLCs, including polychlorinated dibenzo-/>-dioxins (PCDDs), polychlorinated
dibenzofurans (PCDFs), and polychlorinated biphenyls (PCBs), are structurally and
lexicologically related halogenated dicyclic aromatic hydrocarbons.1
EPA's chemical mixtures guidelines and guidance documents (U.S. EPA, 1986, 2000)
call for the use of whole mixture data or data on a sufficiently similar mixture as preferred risk
assessment methods. However, when data are not sufficient to apply these methods, the EPA
also recommends component-based approaches. In such situations, the EPA has recommended
use of the Toxicity Equivalence Factor (TEF) Methodology and the World Health Organization's
(WHO's) TEFs to evaluate the risks associated with exposure to mixtures of TCDD and DLCs
for human health (U.S. EPA, 1987, 1989, 2003) and ecological risk assessments (U.S. EPA,
2008). The WHO has used a process based on consensus judgment of scientific expert panels to
develop TEFs for mammals, birds, and fish and has re-evaluated them on a schedule of
approximately every 5 years (Ahlborg et al., 1994; van den Berg et al., 1998, 2006; also see
WHO's Web site for the dioxin TEFs, available at:
http://www.who.int/ipcs/assessment/tef_update/en/). After evaluating the empirical data on
TCDD and some DLCs, WHO reconfirmed that the combined effects of these compounds
generally are consistent with dose additivity, a key underlying assumption of the TEF
methodology (van den Berg et al., 2006). In this document, the EPA is updating its human
health approach by adopting the mammalian TEFs for DLCs recommended in the WHO's 2005
reevaluation of TEFs for human exposures to DLCs (van den Berg et al., 2006). EPA Program
Offices and Regions have historically used TEF values in their risk assessments; this document
recommends the 2005 WHO consensus TEFs, but does not address specific risk assessment
applications of TEFs.
^orfurther information on the chemical structures of these compounds, see U.S. EPA (2003, 2008).
-------
THE TEF METHODOLOGY
This section briefly describes the TEF methodology, which is based on the concept of
dose addition. Application of this methodology in human health risk assessment has been
described and reaffirmed for use by the Agency in EPA's Supplementary Guidance for
Conducting Health Risk Assessment of Chemical Mixtures (U.S. EPA, 2000). Under dose
addition, the toxicokinetics and the toxicodynamics of all components are assumed to be similar
and the dose-response curves of the components of a mixture are assumed to be similarly
shaped.2 Following these assumptions, the combined toxicity of the individual components can
be estimated using the sum of their doses, which are scaled for potency relative to that of another
component of the mixture for which adequate dose-response information is available (U.S. EPA,
2000).
In practice, the scaling factor for each DLC is typically based on a comparison of its toxic
potency to that of a designated index chemical. For DLCs, TCDD is typically specified as the
index chemical. However, the WHO 2005 (van den Berg et al., 2006) panel also used PCBi26 as
an index chemical for some DLCs in some studies used to develop relative potency estimates;
the panel invoked transitivity, that is, by quantifying both the toxicity of a DLC relative to
PCBne and PCBi26 to TCDD, the toxicity of the DLC relative to TCDD was estimated (RePs;
Haws et al., 2006).3 The index chemical is well-studied lexicologically and must have a
dose-response function to apply the methodology to an environmental mixture. The
2 The TEF methodology has traditionally required that the dose response curves of the DLCs be parallel. In recent
years, EPA's guidance documents on chemical mixtures risk assessment have moved away from the strict dose-
response requirement of parallelism because of the variability inherent in showing such a phenomenon when dose-
response data across mixture components are typically from different labs, different experimental designs or dose
levels, and various strains, species, and genders of experimental animals. Further, it can be difficult to evaluate the
shapes of dose response curves from experimental studies in the low dose region of interest in risk assessment. For
the EPA's relative potency factor method, which is based on dose-addition, only similarly shaped dose response
curves are required (satisfied, for example, by modeling the mixture components using the same dose-response
functional form, or grouping chemicals by common slope parameters or by a common maximum effect) and may be
limited to a range of exposure conditions, including dose level, frequency and route (U.S. EPA, 2000, 2002).
3 For some compounds in some toxicity studies, the WHO panel compared the toxicity of DLCs to that of PCB126
during their development of estimates of RePs (Haws et al., 2006). When developing RePs based on comparing
effects of DLCs to those of PCB126, the WHO panel invoked transitivity; that is, by quantifying both the toxicity of a
DLC relative to PCB126 and PCB126 to TCDD, one could estimate the toxicity of the DLC relative to TCDD. Given
the TEF for PCB126 was 0.1, WHO (2005) multiplied the PCB126-based ReP by 0.1. Based on Hawes et al. (2006), a
total 114 RePs were developed for the mono-ortho PCBs in the TEF database. PCB126 served as the index
chemical for 29 (25.4%) of these. Forthe nonortho-PCBs in the same database, if PCB126 is excluded from the
nonortho PCBs in the TEF database, then PCB 126 served as the index chemical for 18 of 91 (20%) of the RePs.
-------
toxicological data considered for these comparisons of toxic potency are from both in vitro and
in vivo studies as well as structure-activity relationships and are based on the following classes
of measure: biochemical changes, toxicity, and carcinogenicity. A comparative measure from an
individual toxicity assay is termed an estimate of relative potency (ReP).4 Based on the RePs
that may be estimated from multiple toxicological assays, each individual PCDD, PCDF, and
PCB is assigned a single scaling factor termed the TEF. By definition, the TEF for TCDD is 1.0;
when PCBi26 serves as an index chemical the value of its TEF is 0.1 (U.S. EPA, 1989, 2000,
2003, 2008; van den Berg et al., 1998, 2006).
To apply TEFs to an environmental mixture of DLCs, each individual compound's
exposure concentration is multiplied by its specific TEF, yielding the individual PCDD, PCDF,
or PCB dose that is equivalent to a dose of the index chemical. These index chemical equivalent
doses are then summed. To estimate risk associated with the mixture, the dose-response function
for the index chemical is evaluated at this sum, which is an estimate of the total index chemical
equivalent dose for the mixture components being considered.
Equation 1 is the formula for calculating exposure concentration for n DLCs in a mixture
in TCDD toxic equivalence (TEQ). Exposure to the ith individual PCDD, PCDF, or PCB
compound is expressed in terms of an equivalent exposure of TCDD by computing the product
of the concentration of the individual compound (d) and its assigned TEF,. TEQ is then
calculated by summing these products across the n DLC present in the mixture. For human
health risk assessment, the TEQ may be evaluated using TCDD dose-response data and used to
assess the risk posed by exposures to mixtures of TCDD and DLCs.
(Eq.
4The term "relative effect potency" (ReP) also is used at times. This term is distinguished from the 'relative potency
factors' (RPF) method, which is a general dose additive method described in U.S. EPA (2000). van den Berg et al.
(2006) evaluated RePs based on biochemical and toxicological endpoints (also see related discussion in Haws et al.,
2006).
-------
BACKGROUND
There is a long history of the development of TEFs and the TEF methodology, dating
back to the 1980s (see Table 1 for details). Early EPA documents recommended the use of the
TEF approach for specific PCDDs and PCDFs for environmental risk assessment (U.S. EPA,
1987, 1989). The PCBs that displayed dioxin-like activity were added to the available TEFs for
DLCs in 1994 (Ahlborg et al., 1994). Then, in 1997, consensus TEFs were assigned to the DLCs
during a meeting held by the WHO (van den Berg et al., 1998); in 2003, EPA recommended the
use of the 1997 WHO mammalian TEFs for human health risk assessment (U.S. EPA, 2003).
Besides the inherent assumption of dose additivity that underpins the TEF approach (i.e.,
the toxicokinetics and the toxicodynamics of all components are assumed to be similar and the
dose-response curves of the components of a mixture are assumed to be similarly shaped),
limitations in the available toxicity data for the DLCs resulted in a number of additional
assumptions that were associated with this approach as implemented. These assumptions
included:
• the Ah receptor mediates most if not all of the biologic and toxic effects of TCDD and
the DLCs;
• the applicability of extrapolations from short-term bioassays to long-term health effects;
• similarities between interspecies kinetics and potency;
• appropriateness of high-dose to low-dose extrapolations; and
• the constancy of TEF relationships for different exposure routes, health endpoints, and
dose levels
(U.S. EPA, 1989, 2000, 2003; see also Birnbaum and DeVito [1995] and Birnbaum [1999]).
Toxic effects of a DLC induced through mechanisms other than the Ah receptor are not
accounted for in this method. Similarly, the TEF methodology does not account for the
interactions of TCDD and DLCs with each other or with other chemicals to which individuals
are exposed. (U.S. EPA [2000] defines the term "interaction" to refer to effects resulting from a
mixture of chemicals that are greater than or less than those anticipated to occur as a
-------
Table 1. Background and history of TEFs for risk assessment of DLCs
Publication
OME, 1984
U.S. EPA, 1986
Eadonetal., 1986
U.S. EPA, 1987
NATO, 1988
U.S. EPA, 1989
Barnes et al.,
1991
Ahlborg et al.,
1994
van den Berg et
al., 1998
U.S. EPA, 2000
U.S. EPA, 2003
(NAS Review draft)
Haws et al., 2006
van den Berg et
al., 2006
NAS, 2006
U.S. EPA, 2008
U.S. EPA, 2010
(this document)
Description of historical context
First to conclude that PCDDs and PCDFs share a common mechanism of action
(activation of the AhR) and that a toxic equivalency approach should be used to
compare equivalent group concentrations to TCDD.
EPA Guidelines for chemical mixtures risk assessment endorse EPA use of dose
addition approaches for chemicals with the same mode of action.
First to describe a TEF-like approach.
Recommends EPA use a TEF approach, applying it to specific PCDDs and PCDFs
instead of to equivalent group concentrations.
Concludes TEF approach is the best available interim approach for PCDD/PCDF risk
assessment. Presents an international TEF scheme.
EPA adopts the international TEF scheme developed by NATO (1988) for use in
developing interim estimates of risk from exposure to PCDDs and PCDFs.
EPA holds workshop. Guiding criteria for TEF approaches are developed.
Concludes that PCBs displaying dioxin-like activity meet the criteria for inclusion in
the TEF scheme.
Develops first set of global consensus TEFs. Adds PCBs, including di-ortho
congeners.
Develops second set of global consensus TEFs. Uses database compiled by the
Karolinska Institute. Deletes di-ortho PCBs from the concept. Recognizes that TEFs
for fish and birds need to be differentiated from humans. Acknowledges that in vivo
results are more important than in vitro results.
Supplemental guidance for chemical mixtures risk assessment describes TEF and
Relative Potency Factor methods. Endorses these for use by EPA.
This draft document recommends van den Berg et al. (1998) TEFs for EPA human
health risk assessment. Provides details on historical development of TEFs.
Refines Karolinska Institute ReP database. Updates the literature. Deletes duplicate
entries. Presents study exclusion criteria and deletes RePs based on studies not
meeting the criteria. Presents statistical summaries of the RePs for each DLC.
Develops third set of global consensus TEFs. Uses Haws et al. (2006) database.
Incorporates new literature including NTP (2006) study results. Holds stakeholder
meeting at the beginning of the evaluation. Articulates shortcomings of the present
TEF system. Identifies other potential compounds for inclusion in the TEF scheme.
Supports the use of the TEF approach by EPA to assess DLCs.
Recommends van den Berg et al. (2006) TEFs for EPA ecological risk assessments.
Recommends van den Berg et al. (2006) TEFs for EPA human health risk
assessments. Recommends the conduct of a sensitivity analysis be considered for
major assessments as determined by U.S. EPA Regions or Program Offices.
AhR = aryl hydrocarbon receptor; NATO = North Atlantic Treaty Organization.
-------
consequence of a specified definition of additivity, typically dose-addition or response addition.)
To capture the uncertainty in these assumptions, all TEFs were provided as order-of-magnitude
estimates, and the EPA described their application as a "useful interim approach"
(U.S. EPA, 1989).
A set of guiding criteria were developed for TEF approaches (Barnes et al., 1991;
U.S. EPA, 1991, 2000). These criteria included the development of TEFs through scientific
consensus. The assignment of global consensus TEFs for the DLCs, including the dioxin-like
PCBs, has been reevaluated as new data have become available (e.g., Ahlborg et al., 1994) and
through consensus judgment of expert panels (e.g., WHO deliberations detailed in van den Berg
et al., 1998, 2006). The TEF values published in van den Berg et al. (1998) were recommended
for use by EPA in its National Academy of Science (NAS) review draft dioxin reassessment
(U.S. EPA, 2003). In its review, NAS supported the use of the TEF approach (NAS, 2006, p. 8),
stating that "Even with the inherent uncertainties, the committee concludes that the TEF
methodology provides a reasonable, scientifically justifiable, and widely accepted method to
estimate the relative potency of DLCs."
In 2005, a WHO expert panel updated TEF values for DLCs (van den Berg et al., 2006).
They reaffirmed the characteristics necessary for inclusion of a compound in the WHO's TEF
approach (van den Berg et al., 1998). These include:
• Structural similarity to polychlorinated dibenzo-/?-dioxins or polychlorinated
dibenzofurans;
• Capacity to bind to the aryl hydrocarbon receptor (AhR);
• Capacity to elicit AhR-mediated biochemical and toxic responses; and
• Persistence and accumulation in the food chain.
van den Berg et al. (2006) also reevaluated the support for assuming dose additivity and
observing parallel dose-response curves. Evaluations of a number of studies of DLCs, including
a mixture study from the National Toxicology Program that evaluated neoplastic and
non-neoplastic endpoints (Walker et al., 2005), led the panel to state that the observed toxicity is
consistent generally with these two assumptions underlying the TEF approach. In addition, the
-------
NAS supported the use of an additivity assumption in its report on EPA's NAS review draft
dioxin reassessment (U.S. EPA, 2003), concluding that "from an overall perspective, this
assumption appears valid, at least in the context of risk assessment. Additivity in biochemical
and toxic responses by the indicated DLCs has been supported by numerous controlled mixture
studies in vitro and in vivo and is scientifically justifiable" (NAS, 2006, p. 80).
The TEF values were revised further by evaluating new toxicological data in conjunction
with statistical summaries of available in vivo RePs formed using a mammalian ReP database
(Haws et al., 2006). The database was comprised of ReP values from all identified studies that
could yield an estimate of a ReP for a DLC; the RePs were not weighted according to study
characteristics (e.g., in vivo, in vitro, chronic, acute, etc.). Haws and collaborators extended the
original WHO ReP database, developed at the Karolinska Institute (RePi997 database) in which
some studies were represented more than once in the form of dissertations, conference
proceedings, and/or peer-reviewed publications.5 In the development of a refined ReP database,
Haws et al. (2006) applied a set of study exclusion criteria to the ReP 1997 database to identify
RePs that likely provided "the most representative measure of a biological response." If a study
met any of the exclusion criteria, the RePs derived from the study were not included in the
quantitative analyses of all RePs. Haws et al. (2006) modified the RePi99? database using the
following exclusion criteria:
Replicate RePs, when RePs from the same original study were presented in multiple
publications.
Multiple RePs from a single study that used different assays to measure the same
response. In this case an effort was made to identify the single most representative ReP
from a study.
Study included only a single dose level of test and/or reference compound.
Data omitted from the final peer-reviewed publication.
5The RePiggv database was used in the WHO-European Centre for Environmental Health (ECEH)TInternational
Programme on Chemical Safety (IPCS) TEF evaluation in 1997 and included not only published manuscripts, but
also manuscripts in press, conference proceedings, theses, dissertations, and unpublished studies through June of
1997 that compared compounds to TCDD or PCB 126. Since the RePi997 database was intended to be all inclusive,
some studies are represented more than once in the form of dissertations, conference proceedings, and/or peer-
reviewed publications.
-------
• Authors indicated in the original publication that the ReP is not valid due to experimental
problems.
• Data entry errors.
• ReP based on replicates in an in vitro study (average value calculated and retained).
• ReP based on non-AhR-mediated response.
• ReP based on nonmammalian species.
• Response for test or reference compound not statistically different from controls and not
biologically meaningful.
• Reference compound (e.g., TCDD) not included in study or in identical study from the
same laboratory.
• Multiple RePs derived from the same data using different calculation techniques.
• Multiple RePs reported for laboratory validation study (samples sent to two different labs
for analysis and RePs calculated for both).
• Multiple RePs calculated based on different test conditions.
• RePs based on data at end of study and at end of some extended recovery period.
• ReP based on mixtures study.
• ReP from an unpublished study that could not be obtained.
The most recent WHO TEFs were developed using a refined approach. The WHO expert
panel considered data from Haws et al. (2006) who present summary statistics of the RePs for
each DLC, calculated from the assembled in vivo and in vitro studies that were not eliminated by
the exclusion criteria. For each individual DLC, the WHO expert panel examined where the
existing TEF value from van den Berg et al. (1998) fell within that DLC's in vivo ReP statistical
summary developed in Haws et al. (2006). If it fell above the 75th percentile of the ReP
statistical range, then they reviewed the basis of the 1998 TEF value, evaluated whether new data
would impact the TEF and either confirmed the 1998 value or derived an updated TEF value. If
it fell below the 75th percentile, the panel examined the database to identify the RePs having the
most influence on the TEF value, evaluated the new data, and derived an updated TEF value (van
den Berg et al., 2006). Because the ReP statistical ranges were unweighted relative to study type
-------
and quality, the TEFs were determined using point estimates from toxicological studies, not by
using specific points within the ReP ranges. A stepwise scale was used to assign the TEFs using
half order of magnitude increments on a logarithmic scale (e.g., 0.03, 0.1, 0.3, etc.) instead of the
increments used in previous efforts (e.g., 0.01, 0.05, 0.1, etc.), with uncertainty assumed to be at
least ± half a log.6
6For example, the uncertainty for a TEF of 0.1 can be described as being within the interval of 0.03 and 0.3, and for
a TEF value of 0.3, within an interval of 0.1 and 1. These estimates are generated by multiplying (dividing) the TEF
value by half a log (i.e., 3.16).
-------
UNCERTAINTIES IN THE TEF APPROACH
As is true for any risk assessment approach, uncertainties exist relative to data quality and
evaluation, strength of biological rationale, and ability to determine whether the assumptions of
the method being applied have been met. Application of the TEF approach to the human health
risk assessment of DLCs carries with it some of these uncertainties which have been discussed in
detail elsewhere in the literature. (For example, see discussions in Haws et al. [2006], NAS
[2006], EPA [2000, 2003], and van den Berg et al. [1998, 2006].) The following uncertainties
associated with application of the TEF approach are briefly described for the reader:
UNCERTAINTY IN TEF METHOD ASSUMPTIONS
• Dose additivity under the TEF method assumes a common mode of toxic action mediated
through AhR binding and downstream biochemical and toxic responses. There is some
evidence suggesting that some toxicities associated with some DLCs may be mediated
through other ligands and processes (i.e., not mediated through the AhR). Effects
mediated by other mechanisms (AhR independent) are not accounted for by the TEF
method.
• Dose additivity under the TEF method assumes parallel dose-response curves. This is
supported by some empirical data, but, in practice, parallelism is difficult to show for all
DLCs and exposure scenarios, particularly in the low response region of most interest in
environmental risk assessment.
• Dose additivity under the TEF method assumes that toxicological interactions are not
occurring at environmental levels of the DLCs. Some data suggest that combined
exposures of some DLCs may have antagonistic, rather than additive, effects; these could
be species-specific. It may also be noted that joint toxic action of dioxins with non
dioxin-like compounds could result in additive or nonadditive responses.
• Under the TEF method, the TEF of a DLC is assumed to be equivalent for all exposure
scenarios, for all end points of concern, and all are full agonists. The ranges of RePs
shown in the Haws et al. (2006) database demonstrate the uncertainty in this assumption
as the ranges represent RePs from various study types and endpoints.
• Under the TEF method, it is assumed that RePs from animal studies are predictive of
RePs in humans. However, the human AhR demonstrates some differences when
compared to the AhR from experimental animal species.
10
-------
UNCERTAINTY IN THE PROCESSES AND DATA USED TO DERIVE TEFs
Expert scientific judgment, which depends on the knowledge and evaluations of the
expert scientists involved, was used to select the DLCs included in the WHO TEF
approach by evaluating experimental data against specific criteria (van den Berg et al.,
2006). It may be noted that not all of the DLCs identified in releases from anthropogenic
sources are included.
Expert judgment and a consensus process were used to derive the WHO 2005 TEFs (van
den Berg et al., 2006), including evaluation of information from the Haws et al. (2006)
database.
The kinds of information available for comparing the responses to individual DLCs to
those of the index compound are highly variable across chemicals, including many types
of and numbers of in vivo (including different test species) and in vitro studies. In
addition, a number of different methods are employed to calculate REP values
(Haws et al., 2006). (See additional discussions of this below under the section on
Sensitivity Analysis Limitations.)
The uncertainty in TEQ estimates and in the TEF methodology accounts for only some of the
overall uncertainty in a risk assessment of DLCs. TEQ uncertainty only pertains to the
confidence associated with the estimation of TCDD equivalents in a mixture. There is also
uncertainty associated with assessing exposures to environmental mixtures of TCDD and DLCs
and with quantitatively linking health effects to the TCDD and DLC exposures. In addition, the
value of a TEQ is highly dependent on the DLC exposure estimates used in the TEQ
calculations.
11
-------
RECOMMENDATIONS
When data on a whole mixture or a sufficiently similar mixture are not available for
DLCs, the EPA recommends use of the WHO consensus mammalian TEF values from van den
Berg et al. (2006) in the assessment of human health risks posed by exposure to mixtures of
TCDD and DLCs, using TCDD as the index chemical. These TEFs are presented in Table 2.
The TEF methodology is most applicable to situations where exposures are predominantly to
mixtures of dioxins, furans and PCBs, and the goal of the assessment is to analyze the health
risks posed by the mixture, not from exposure to individual compounds or single classes of
compounds. Thus, other approaches may be considered when exposures are to single
compounds or chemical classes.7
The EPA agrees with van den Berg et al. (2006) that the TEFs are most appropriate for
dioxin exposures via the oral exposure route. The bioavailability of DLCs encountered through
various sources of oral exposure needs to be evaluated in risk analyses. The TEFs may be
applied to other exposure routes (i.e., dermal or inhalation), as an interim estimate or as a
component of the sensitivity analysis, assuming exposures to DLCs via these routes can be
quantified. Uncertainties associated with such applications should be identified. EPA
recommends that, if considered in an assessment, the fractional contribution of oral, dermal, and
inhalation route exposures to the predicted TEQ be identified.
TCDD and DLCs are associated with several different human health effects. Nearly all
TCDD and DLC experimental data appear to be consistent with the hypothesis that binding to
the AhR is the first step in a series of biochemical, cellular, and tissue changes that ultimately
lead to toxic responses observed in both experimental animals and humans. The general basis
for the TEF scheme is the assumption that the AhR mediates most if not all of the dioxin-like
biological and toxic effects induced by compounds included in the WHO 2005 TEF approach
(Safe, 1990; Okey et al., 1994; Birnbaum, 1994; Hankinson, 1995). Binding to the receptor
7For example, if the exposure is dominated by the single class of PCBs, then an alternative approach for evaluating
human health risk might include use of the PCB cancer slope factors on Integrated Risk Information System
(U.S. EPA, 1997). Also, when PCB exposures do not involve significant amounts of PCDDs and PCDFs, EPA
(1996) provides another alternative methodology that might be useful for PCB mixture cancer dose-response
assessment. However, in these cases, risks associated with other chemical exposures, i.e., not PCBs, would still
need to be addressed.
12
-------
Table 2. Recommended toxicity equivalence factors (TEFs) for human
health risk assessment of polychlorinated dibenzo-p-dioxins, dibenzofurans,
and dioxin-like polychlorinated biphenyls
Compound
Polychlorinated dibenzo-p-dioxins (PCDDs)
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
Polychlorinated dibenzofurans (PCDFs)
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF
Polychlorinated biphenyls* (PCBs)
3,3',4,4'-TCB (77)
3,4,4',5-TCB (81)
3,3',4,4',5-PeCB (126)
3,3',4,4',5,5'-HxCB (169)
2,3,3',4,4'-PeCB (105)
2,3,4,4',5-PeCB(114)
2,3',4,4',5-PeCB(118)
2',3,4,4',5-PeCB (123)
TEF
1
1
0.1
0.1
0.1
0.01
0.0003
0.1
0.03
0.3
0.1
0.1
0.1
0.1
0.01
0.01
0.0003
0.0001
0.0003
0.1
0.03
0.00003
0.00003
0.00003
0.00003
13
-------
Table 2. Recommended toxicity equivalence factors (TEFs) for human
health risk assessment of pol\ chlorinated dibenzo-p-dioxins, dibenzofurans,
and dioxin-like pol\ chlorinated biphenyls (continued)
Compound
2,3,3',4,4', 5-HXCB(156)
2,3,3',4,4',5'-HxCB (157)
2,3',4,4',5,5'-HxCB (167)
2,3,3',4,4',5,5'-HpCB(189)
TEF
0.00003
0.00003
0.00003
0.00003
*Note: TEFs that were previously assigned to PCB 170 and PCB 180 (Ahlborg et al., 1994) were withdrawn during
the WHO-ECEH/IPCS TEF re-evaluation in 1997, and a TEF for PCB 81 was established, such that the number of
PCB compounds with TEFs assigned was reduced from 13 to 12 (van den Berg et al., 1998). The numbers in
parentheses following each PCB are the PCB congener numbers.
Source: van den Berg et al. (2006); WHO's Web site on dioxin TEFs, available at:
http://www.who.int/ipcs/assessment/tef_update/en/.
appears to be necessary—but not sufficient—to generate the wide variety of toxic effects caused
by dioxin-like halogenated aromatic hydrocarbons (Sewall and Lucier, 1995; DeVito and
Birnbaum, 1995). In this document EPA assumes that all cancer and noncancer effects of TCDD
and DLCs are AhR dependent. The EPA recommends these TEFs be used for all cancer and
noncancer effects that appear to be mediated through AhR binding by the DLCs. EPA
recognizes that this issue will require further evaluation as additional toxicity data become
available. Eventually, endpoint-specific TEFs or separate TEFs for systemic toxicity and
carcinogenicity endpoints may need to be developed.
van den Berg et al. (2006) also identified a number of candidate compounds that may
need to be included in future developments of TEFs for DLCs:
• PCB 37
• Polybrominated dibenzo-p-dioxins and polybrominated dibenzofurans (PBDFs)
• Mixed halogenated dibenzo-^-dioxins and mixed halogenated dibenzofurans
• Hexachlorobenzene
14
-------
• Polychlorinated naphthalenes and polybrominated naphthalenes
• Polybrominated biphenyls
EPA will consider an update of the recommendations in this document when TEFs for these
candidate compounds are developed. At a minimum, if occurrence or exposure data are
available for these candidate compounds, this information should be included as part of a
qualitative risk characterization.
For analytic transparency, the EPA recommends that the fraction of the TEQ attributable
to each PCDD, PCDF, or PCB compound be identified in the risk characterization (Table 2 lists
the DLCs considered to be members of PCDD, PCDF, or PCB groups.) Further, the
contributions of each chemical class, i.e., the PCDDs, PCDFs, and dioxin-like PCBs, should also
be identified. Alternatively, the analysis could examine 2,3,7,8-TCDD alone, all dioxin
congeners, and the dioxin-like compounds (PCBs and PCDFs) in three separate analyses. The
compounds and class(es) making the largest contributions to the TEQ should be specified as
appropriate to the assessment (see example in Text Box 1). In addition, the implications of the
fraction of the TEQ attributable to TCDD should be
discussed in the analyses because the dose-response
data for TCDD are used to evaluate risks, and the
confidence in the risk estimate increases with
increases in the fraction of the TEQ attributable to
TCDD. Finally, if multiple routes are considered in
an assessment, the fractional contribution of the
compounds and class(es) to each exposure route to
the predicted TEQ should be identified.
Text Box 1. Example Risk
Characterization
U.S. EPA (2003) notes that the majority of
the TEQ (based on van den Berg et al., 1998)
from dietary exposures is typically associated
with the concentrations of only five
compounds (i.e., TCDD, 1,2,3,7,8-PCDD,
2,3,4,7,8-PeCDF, 1,2,3,6,7,8-HxCDD, PCB
126) whose ReP variability appears to be
small relative to other compounds.* Thus, if
dietary exposures are important to the
assessment being conducted, the fraction of
the TEQ attributable to these five compounds
should be presented and discussed in the risk
characterization.
*Note that the TEF for 2,3,4,7,8-PeCDF
changed from 0.5 to 0.3 from van den Berg
et al., 1998 to 2006, respectively.
SENSITIVITY ANALYSIS
The EPA recommends that, for major risk
assessments, as determined by U.S. EPA Program
Offices or Regions, the conduct of a sensitivity analysis be considered to illustrate the impact the
TEFs have on the TEQ value, which is consistent with good risk assessment practices
(U.S. EPA, 2000). While ideally a full quantitative uncertainty analysis is desirable, currently
15
-------
available ReP data that could be used to characterize the distributions of the TEFs are not
suitable for use in simulation procedures (e.g., a Monte Carlo analysis) that are typically
undertaken. Characterization of the underlying statistical distributions of the ReP data would be
needed as input to a quantitative uncertainty analysis; the true probability distributions of the
TEFs are not known at this time. The limitations in both the underlying ReP data and in the
ability to statistically analyze them preclude a detailed evaluation of the various sources of
heterogeneity inherent in a quantitative analysis of uncertainty. However, insightful sensitivity
analyses can be conducted using estimated ranges of the TEFs.
A TEF sensitivity analysis has at least two purposes: (1) to identify plausible upper and
lower estimates of the TEQ to assess the potential range the TEQ may have, and (2) to identify
the influence of TEF values for specific compounds on the TEQ. One quantitative approach for
identifying upper and lower TEQ estimates is presented in Eq. 2 and 3 below for n compounds
with TCDD represented by compound / = 1 (see discussion of limitations of this approach
below).
(Eq.2)
(Eq.3)
where:
TEQu = upper estimate of TEQ range
TEQL = lower estimate of TEQ range
d = concentration of the rth individual compound
= upper estimate of the rth compound's TEF; for/= 1, TEFiu= 1
= lower estimate of the rth compound's TEF; for/= 1, TEF1L= 1.
For the TEQu and TEQi estimates that are generated using Eq. 2 and 3, the fraction of the TEQ
attributable to TCDD and to each DLC should be identified.
EPA is aware of two possible data choices for identifying compound specific TEFnj and
values. First, van den Berg et al. (2006) state that the TEFs are assumed to have
16
-------
uncertainty of at least ± half a log (i.e., 3.16); thus, multiplying and dividing the compound
specific TEFs by 3.16 could provide estimates of TEFnj and /', respectively.
Second, the EPA is aware that Haws et al. (2006) has summarized statistical descriptions
of the ReP values. Although limited to the available ReP data (i.e., not necessarily an unbiased
sample of equivalence factors), the ReP ranges developed by Haws et al. (2006) may provide
another source of data for TEFtu and TEFn values to use in Eq. 2 and 3. Tables 3 and 4 present
specific percentiles of the Haws et al. (2006) statistical summaries for the RePs derived from in
vivo data and combined in vitro and in vivo data, respectively. The values for TEFm and TEFn,
for example, could be based on the minimum and maximum data, the 10th and 90th percentiles, or
the interquartile ranges from either Tables 3 or 4. Over time, this set of ReP values is expected
to change with the availability of additional relevant studies.
To identify the influence of specific compounds on the TEQ, EPA recommends that the
list of compounds that are most influential to the TEQ, as defined in Eq. 1, be further explored.
For each of these, the sensitivity of the TEQ to changes in the TEF values for the individual
compounds may be conducted (i.e., varying the TEF value for one compound at a time). The
same statistical ranges described above can be used to identify alternative TEF values.
SENSITIVITY ANALYSIS LIMITATIONS
The suggested summations of TEF; times Q should not be interpreted as upper or lower
bounds on confidence limits for the TEQ. These calculations only provide crude estimates of the
range of the TEQ, and they are useful for comparing the impact that the TEF; have on the TEQ in
a sensitivity analysis. A summation using a specific percentile does not result in an estimate of
the same percentile of the TEQ, but would likely overestimate that percentile for upper bound
estimates and likely underestimate that percentile for lower bound percentiles. Thus, an
overestimation of the TEQ range will increase as higher (lower) TEF percentiles are used in the
summation.
Issues with the assignment of the WHO 2005 TEFs (van den Berg et al., 2006) and the
construction of the Haws et al. (2006) ReP database preclude the conduct of a quantitative
uncertainty analysis and the calculation of confidence limits. Both of these issues may be
important in interpreting the results of a sensitivity analysis. The WHO 2005 individual TEFs
are not central tendency estimates of the available values (van den Berg et al., 2006), but instead
17
-------
Table 3. Percentiles of in vivo ReP values
Congener
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
1,2,3,4,7,8-HxCDD
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDD
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDD
1,2,3,7,8,9-HxCDF
1,2,3,7,8-PeCDD
1,2,3,7,8-PeCDF
2,3,4,6,7,8-HxCDF
2,3,4,7,8-PeCDF
OCDD
OCDF
PCB105
PCB114
PCB118
PCB123
PCB126
n
12
0
0
15
6
0
11
1
0
36
20
3
82
1
6
16
2
15
2
86
min
0.001
N/A
N/A
0.008
0.01
N/A
0.003
0.03
N/A
0.04
0.003
0.02
0.007
0.0003
0.000004
0.0000005
0.0002
0.0000004
0.00003
0.0001
0.1
0.004
N/A
N/A
0.03
0.03
N/A
0.01
0.03
N/A
0.1
0.009
0.02
0.05
0.0003
0.00002
0.000002
0.0002
0.000002
0.00004
0.02
0.25
0.007
N/A
N/A
0.05
0.04
N/A
0.02
0.03
N/A
0.2
0.01
0.02
0.1
0.0003
0.00004
0.000009
0.0003
0.000007
0.00004
0.06
0.5
0.01
N/A
N/A
0.06
0.05
N/A
0.08
0.03
N/A
0.4
0.02
0.02
0.2
0.0003
0.00008
0.00004
0.0003
0.00002
0.00004
0.1
0.75
0.01
N/A
N/A
0.09
0.07
N/A
0.09
0.03
N/A
0.6
0.08
0.06
0.3
0.0003
0.0006
0.0001
0.0004
0.00005
0.00005
0.2
0.9
0.02
N/A
N/A
0.1
0.1
N/A
0.1
0.03
N/A
0.8
0.1
0.08
0.7
0.0003
0.001
0.001
0.0004
0.001
0.0001
0.4
max
0.04
N/A
N/A
0.4
0.2
N/A
0.2
0.03
N/A
2
1
0.1
4
0.0003
0.002
0.002
0.0005
0.002
0.0001
0.9
2005 TEF
0.01
0.01
0.01
0.1
0.1
0.1
0.1
0.1
0.1
1
0.03
0.1
0.3
0.0003
0.0003
0.00003
0.00003
0.00003
0.00003
0.1
oo
-------
Table 3. Percentiles of in vivo ReP values (continued)
Percentile
Congener
PCB156
PCB157
PCB167
PCB169
PCB189
PCB77
PCB81
TCDF
n
16
2
0
15
3
16
17
min
0.000002
0.0004
N/A
0.000002
0.00004
0.000002
N/A
0.006
0.1
0.000005
0.0006
N/A
0.0004
0.00004
0.000006
N/A
0.008
0.25
0.00003
0.0007
N/A
0.003
0.00005
0.00001
N/A
0.01
0.5
0.00006
0.001
N/A
0.02
0.00006
0.00006
N/A
0.03
0.75
0.0005
0.001
N/A
0.2
0.0001
0.0001
N/A
0.1
0.9
0.09
0.002
N/A
0.6
0.0002
0.02
N/A
0.3
max
0.4
0.002
N/A
0.7
0.0002
0.04
N/A
0.5
2005 TEF
0.00003
0.00003
0.00003
0.03
0.00003
0.0001
0.0003
0.1
Source: Haws et al. (2006) 2004 ReP Database, Figure A-4.
-------
Table 4. Percentiles of combined in vivo and in vitro ReP values
Compound
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
1,2,3,4,7,8-HxCDD
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDD
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDD
1,2,3,7,8,9-HxCDF
1,2,3,7,8-PeCDD
1,2,3,7,8-PeCDF
2,3,4,6,7,8-HxCDF
2,3,4,7,8-PeCDF
OCDD
OCDF
PCB105
PCB114
n
18
2
2
21
13
5
18
6
2
45
28
10
99
6
9
26
8
min
0.001
0.02
0.02
0.01
0.01
0.03
0.003
0.01
0.1
0.04
0.003
0.01
0.01
0.0003
0.000004
0.0000005
0.0001
]
0.1
0.004
0.05
0.02
0.04
0.04
0.03
0.01
0.02
0.1
0.1
0.01
0.01
0.05
0.0003
0.00003
0.000005
0.0002
'ercentile
0.25
0.01
0.1
0.02
0.05
0.04
0.04
0.03
0.03
0.1
0.2
0.01
0.04
0.1
0.0003
0.00004
0.00001
0.0002
0.5
0.01
0.2
0.03
0.08
0.07
0.04
0.07
0.05
0.2
0.4
0.05
0.2
0.2
0.0003
0.001
0.0001
0.001
0.75
0.03
0.2
0.04
0.1
0.3
0.06
0.1
0.06
0.2
0.6
0.1
0.3
0.5
0.002
0.002
0.0003
0.001
0.9
0.04
0.3
0.04
0.4
0.5
0.1
0.1
0.07
0.2
0.8
0.1
0.3
1
0.003
0.002
0.002
0.002
max
0.1
0.3
0.04
0.6
4
0.2
0.2
0.07
0.2
2
1
0.3
4
0.003
0.003
0.07
0.002
2005 TEF
0.01
0.01
0.01
0.1
0.1
0.1
0.1
0.1
0.1
1
0.03
0.1
0.3
0.0003
0.0003
0.00003
0.00003
to
o
-------
Table 4. Percentiles of combined in vivo and in vitro ReP values (continued)
Percentile
Compound
PCB118
PCB123
PCB126
PCB156
PCB157
PCB167
PCB169
PCB189
PCB77
PCB81
TCDF
n
25
6
115
30
9
5
30
5
49
12
30
min
0.0000004
0.000003
0.0001
0.000002
0.00004
0.000002
0.000002
0.000002
0.000002
0.00004
0.01
0.1
0.000002
0.00001
0.01
0.00001
0.0001
0.000005
0.0007
0.000005
0.00002
0.0006
0.01
0.25
0.00001
0.00002
0.05
0.00004
0.0001
0.00001
0.002
0.00001
0.0001
0.004
0.03
0.5
0.00002
0.00004
0.1
0.0001
0.0004
0.00001
0.01
0.00004
0.001
0.01
0.08
0.75
0.0005
0.0001
0.2
0.001
0.001
0.00001
0.06
0.00006
0.02
0.01
0.2
0.9
0.002
0.0004
0.4
0.2
0.002
0.0004
0.5
0.0001
0.1
0.02
0.3
max
0.08
0.0007
0.9
0.5
0.002
0.001
0.8
0.0002
0.5
0.05
0.6
2005 TEF
0.00003
0.00003
0.1
0.00003
0.00003
0.00003
0.03
0.00003
0.0001
0.0003
0.1
Source: Haws et al. (2006) 2004 ReP Database, Figure A-2.
-------
are assigned based on professional judgment using both information from the Haws et al. (2006)
database and from the available toxicology data; thus, these TEFs cannot be evaluated using
statistics relevant to a mean or median value.
Haws et al. (2006) discuss the limitations of the current ReP database for use in
quantitative uncertainty analysis. The RePs were calculated using various approaches, ranging
from comparing dose-response curves, to developing ratios of effective doses that cause an effect
in 50% of the test units (ED50s), to estimating values from graphs of dose-response data. The
RePs also represent a wide variety of study types and endpoints, including biochemical changes,
systemic toxicity and carcinogenicity; some of these data may provide estimates that are more
consistent than others with individual PCDD, PCDF, or PCB compound toxicity at higher levels
of biological organization and such considerations will need to be included in a risk
characterization. Finally, Haws et al. (2006) note a number of issues associated with the
dose-response data (e.g., nonparallel dose-response curves, differences in maximal response
among PCDD, PCDF, or PCB compounds within a study, incomplete dose-response data due to
insufficient dose levels). In addition, the number of RePs available varies widely across the
congeners from n = 2ton= 115 RePs. Thus, the Haws et al. (2006) database provides
"statistical descriptions," not probability distributions, as the RePs in the database are not
unbiased random samples of TEF values.
Although EPA recognizes the limitations associated with the use of the Haws et al.
(2006) database in sensitivity analyses, EPA believes the benefits associated with the conduct of
such an analysis outweigh the limitations. The development of a more refined ReP database and
additional examination of the uncertainties inherent in a TEF process would improve TEF-based
risk assessments.
22
-------
CONCLUSIONS
When whole mixture data or data on a sufficiently similar mixture are not available for
DLC exposures, the EPA recommends use of the consensus mammalian TEF values from
van den Berg et al. (2006) in the assessment of human health risks posed by exposures to
mixtures of TCDD and DLCs (see Table 2), using TCDD as the index chemical. EPA Program
Offices and Regions have historically used TEF values in their risk assessments; this document
recommends the 2005 WHO consensus TEFs, but does not address specific risk assessment
applications of TEFs. Further, while ideally a full quantitative uncertainty analysis is desirable,
currently available ReP data that could be used to characterize the distributions of the TEFs are
not suitable for use in simulation procedures that are typically undertaken. Because limitations
in both the underlying ReP data and in the ability to statistically analyze them preclude conduct
of a full quantitative uncertainty analysis of the TEQs, the EPA recommends that conduct of a
sensitivity analysis be considered when using TEFs in major risk assessments, as determined by
EPA Program Offices or Regions. In conducting a TEF-based risk assessment the EPA suggests
addressing the key risk characterization recommendations that have been discussed in this
document and are summarized in Table 5. The EPA will update all of these recommendations in
the future based on the evaluation of new toxicity data for the DLCs, updates to the ReP database
including statistical summaries of RePs for individual DLCs, and the results of new consensus
processes undertaken to update the TEF approach.
23
-------
Table 5. Summary of risk characterization recommendations for TEF
applications
1) Apply the TEF methodology to situations where exposures are predominantly to mixtures
of dioxins, furans, and PCBs, and the goal of the assessment is to analyze the human health
risks posed by the mixture.
2) Identify the fraction of the TEQ attributable to TCDD, each DLC, and to each chemical
class, i.e., the PCDDs, PCDFs, and dioxin-like PCBs. Alternatively, the analysis of
chemical classes could examine separately the contributions from 2,3,7,8-TCDD alone, all
dioxin congeners, and the dioxin-like compounds (PCBs and PCDFs) to the TEQ.
3) When it is deemed appropriate to apply TEFs to a multiroute exposure as an interim
approach, identify the fractional contributions of oral, dermal, and inhalation route
exposures to the predicted TEQ. Within each route of exposure, identify the fractional
contribution of each congener to the predicted TEQ and identify the fraction of the TEQ
associated with each chemical class.
4) Address the implications of the identified fractional contributions to the TEQ for the risk
assessment being conducted, in particular, their impacts on the overall confidence in the
analytic results.
5) Include occurrence or exposure data, if available, for the following compounds as part of a
qualitative risk characterization:
• PCB37
• Polybrominated dibenzo-p-dioxins and polybrominated dibenzofurans
• Mixed halogenated dibenzo-p-dioxins and mixed halogenated dibenzofurans
• Hexachlorobenzene
• Polychlorinated naphthalenes and polybrominated naphthalenes
• Polybrominated biphenyls
6) For major risk assessments as determined by EPA Program Offices or Regions, EPA
recommends the conduct of a sensitivity analysis be considered to characterize the impact
of TEF variability on the TEQ.
• For the TEQu and TEQL estimates that are generated, identify the fraction of the TEQ
attributable to TCDD, each DLC and each chemical class.
• Identify the TEF; values that are most influential to changing the TEQ estimate.
24
-------
REFERENCES
Ahlborg, UG; Becking, G; Birnbaum, L; et al. (1994) Toxic equivalency factors for dioxin-like PCBs. Report on a
WHO-ECEH and IPCS consultation, December, 1993. Chemosphere 28(6): 1049-1067.
Barnes, D; Alford-Stevens, A; Birnbaum, L; Kutz, F; Wood, W; Fatten, D. (1991) Toxicity equivalency factors for
PCBs? Qua! Assur 1(1):70-81.
Birnbaum, LS. (1994) The mechanism of dioxin toxicity: relationship to risk assessment. Environ Health Perspect
102(Suppl. 9): 157-167.
Birnbaum, LS. (1999) TEFs: a practical approach to a real-world problem. Hum Ecol Risk Assess 5:13-24.
Birnbaum, LS; DeVito, MJ. (1995) Use of toxic equivalency factors for risk assessment for dioxins and related
compounds. Toxicology 105(2-3):391-401.
DeVito, MJ; Birnbaum, LS. (1995) Dioxins: model chemicals for assessing receptor-mediated toxicity. Toxicology
102:115-123.
Eadon, GL; Kaminsky, J; Silkworm, K; et al. (1986) Calculation of 2,3,7,8-TCDD equivalent concentrations of
complex environmental contaminant mixtures. Environ Health Perspect 70:221-227.
Hankinson, O. (1995) The aryl hydrocarbon receptor complex. Ann Rev Pharmacol Toxicol 35:307-340.
Haws, LC; Su, SH; Harris, M; et al. (2006) Development of a refined database of mammalian relative potency
estimates for dioxin-like compounds. Toxicol Sci 89(1):4-30.
NAS (National Academy of Science). (2006) Health risks from dioxin and related compounds: evaluation of the
EPA reassessment. National Academies Press, Washington, DC. Available online at
http://www.nap.edu/catalog.php?record_id=11688.
NATO (North Atlantic Treaty Organization). (1988) Pilot study on international information exchange on dioxins
and related compounds: Emissions of dioxins and related compounds from combustion and incineration sources.
North Atlantic Treaty Organization, Committee on the Challenges of Modern Society. Report #172. August 1988.
NTP (National Toxicology Program). (2006) Toxicology and carcinogenesis studies of a mixture of
2,3,7,8-tetrachlorodibenzo-^-dioxin (TCDD) (CAS No. 1746-01-6), 2,3,4,7,8-pentachlorodibenzofuran (PeCDF)
(CAS No. 57117-31-4), and 3,3',4,4',5-pentachlorobiphenyl (PCB 126) (CAS No. 57465-28-8) in female Harlan
Sprague-Dawley Rats (gavage studies). Public Health Service, U.S. Department of Health and Human Services;
NTP TR 526. Available from the National Institute of Environmental Health Sciences, Research Triangle Park, NC.
Available at: http://ntp.niehs.nih.gov/index.cfm?objectid=070B7300-OE62-BF12-F4C3E3B5B645A92B.
OME (Ontario Ministry of the Environment). (1984) Scientific criteria document for standard development,
No. 4-84. Polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs). Ontario
Ministry of the Environment, Toronto, Canada.
Okey, AB; Riddick, DS; Harper, PA. (1994) The Ah receptor: mediator of the toxicity of 2,3,7,8-tetrachlorodibenzo-
p-dioxm (TCDD) and related compounds. Toxicol Lett 70:1-22.
Safe, S. (1990) Polychlorinated biphenyls (PCBs), dibenzo-p-dioxins (PCDDs), dibenzofurans (PCDFs), and related
compounds: environmental and mechanistic considerations which support the development of toxic equivalency
factors (TEFs). Crit Rev Toxicol 21(l):51-88.
25
-------
Sewall, CH; Lucier, GW. (1995) Receptor-mediated events and the valuation of the Environmental Protection
Agency (EPA) of dioxin risks. Mutat Res 333:111-122.
U.S. EPA (Environmental Protection Agency). (1986) Guidelines for health risk assessment of chemical mixtures.
EPA/630/R-98/002. Federal Register 51:34014-34025. September 24.
U.S. EPA (Environmental Protection Agency). (1987) Interim procedures for estimating risks associated with
exposures to mixtures of chlorinated dibenzo-p-dioxins and -dibenzofurans (CDDs and CDFs). EPA/625/3-87/012.
NTIS PB-90145756. Risk Assessment Forum, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1989) Interim procedures for estimating risks associated with
exposures to mixtures of chlorinated dibenzo-p-dioxins and -dibenzofurans (CDDs and CDFs) and 1989 update.
EPA/625/3-89/016. Risk Assessment Forum, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1991) Workshop report ontoxicity equivalency factors for
polychlorinated biphenyls congeners. EPA/625/3-91/020. U.S. Environmental Protection Agency, Washington,
DC.
U.S. EPA (Environmental Protection Agency). (1996) PCBs: cancer dose-response assessment and application to
environmental mixtures. National Center for Environmental Assessment, Washington, DC. EPA/600/P-96/001F.
September. Available online at http://www.epa.gov/osw/hazard/tsd/pcbs/pubs/pcb.pdf
U.S. EPA (Environmental Protection Agency). (1997) Integrated Risk Information System (IRIS). Polychlorinated
biphenyls (PCBs) (CASRN 1336-36-3). Office of Research and Development, National Center for Environmental
Assessment, Washington, DC. Available online at http://www.epa.gov/iris/subst/0294.htm [accessed 2 November,
2009].
U.S. EPA (Environmental Protection Agency). (2000) Supplementary guidance for conducting health risk
assessment of chemical mixtures. EPA/630/R-00/002. August. U.S. Environmental Protection Agency,
Washington, DC.
U.S. EPA (Environmental Protection Agency). (2002) Guidance on cumulative risk assessment of pesticide
chemicals that have a common mechanism of toxicity. Office of Pesticide Programs, Washington, DC.
Available online at http://www.epa.gov/oppfeadl/trac/science/cumulative_guidance.pdf.
U.S. EPA (Environmental Protection Agency). (2003) Chapter 9. Toxic equivalency factors (TEF) for dioxin and
related compounds in Part II: Exposure and human health reassessment for 2,3,7,8-tetrachlorodibenzo-p-dioxin
(TCDD) and related compounds. NAS review draft. NCEA-I-0836. December. Office of Research and
Development, National Center for Environmental Assessment, Washington, DC.
U.S. EPA (Environmental Protection Agency). (2008) Framework for application of the toxicity equivalence
methodology for polychlorinated dioxins, furans, and biphenyls in ecological risk assessment. EPA/100/R-08/004.
June. Risk Assessment Forum, Washington, DC.
van den Berg, M; Birnbaum, L; Bosveld, AT; et al. (1998) Toxic equivalency factors (TEFs) for PCBs, PCDDs,
PCDFs for humans and wildlife. Environ Health Perspect 106(12):775-792.
van den Berg, M; Birnbaum, LS; Denison, M; et al. (2006) The 2005 World Health Organization re-evaluation of
human and mammalian toxic equivalency factors for dioxins and dioxin-like compounds. Toxicol Sci
93(2):223-241.
Walker, NJ; Crockett, PW; Nyska, A; et al. (2005) Dose-additive carcinogenicity of a defined mixture of "Dioxin
like Compounds." Environ Health Perspect 113(l):43-48.
26
-------
United States
Environmental Protection
Agency
U.S. Environmental Protection Agency
Office of Research and Development
Washington, DC 20460
Official Business
Penalty for Private Use
$300
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT NO. G-35
------- |