DRAFT
DO NOT CITE OR QUOTE
EPA/600/R-10/038D
Agency/Interagency Review Draft
www.epa.gov/iris
EPA's Reanalysis of Key Issues Related to
Dioxin Toxicity and Response to
NAS Comments, Volume 1
October 2011
NOTICE
THIS DOCUMENT IS AN AGENCY/INTERAGENCY REVIEW DRAFT. It has not been
formally released by the U.S. Environmental Protection Agency and should not at this stage be
construed to represent Agency policy. It is being circulated for comment on its technical
accuracy and policy implications.
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH
-------
DISCLAIMER
This document is distributed solely for the purpose of predissemination peer review under
applicable information quality guidelines. It has not been formally disseminated by EPA. It
does not represent and should not be construed to represent Agency determination or policy.
Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
ABSTRACT
This document comprises the first of two EPA reports (U.S. EPA's Reanalysis of Key
Issues Related to Dioxin Toxicity and Response to NAS Comments Volumes 1 and 2 [Reanalysis
Volumes 1 and 2J) that, together, will respond to the recommendations and comments on 2,3,7,8-
Tetrachlorodibenzo-p-Dioxin (TCDD) dose-response assessment included in the 2006 NAS
report, Health Risks from Dioxin and Related Compounds: Evaluation of the EPA Reassessment.
This document, Reanalysis Volume 1, includes (1) a systematic evaluation of the peer-reviewed
epidemiologic studies and rodent bioassays relevant to TCDD dose-response analysis;
(2) dose-response analyses using a TCDD physiologically-based pharmacokinetic model that
simulates TCDD blood concentrations following oral intake; and (3) an oral reference dose
(RfD) for TCDD. An RfD of 7 x 10 10 mg/kg-day is derived based on two epidemiologic
studies: (a) a study that associated TCDD exposures with decreased sperm concentration and
sperm motility in men who were exposed during childhood and (b) a study that associated
increased thyroid-stimulating hormone levels in newborn infants born to mothers who were
exposed to TCDD. A qualitative discussion of uncertainties in the RfD and a focused
quantitative uncertainty analysis of the choices made in the development of points of departure
for RfD derivation are also provided.
li
DRAFT - DO NOT CITE OR QUOTE
-------
CONTENTS
LIST OF TABLES vii
LIST OF FIGURES ix
LIST OF ABBREVIATIONS AND ACRONYMS xii
PREFACE xvi
AUTHORS, CONTRIBUTORS, AND REVIEWERS xviii
EXECUTIVE SUMMARY xxiii
1. INTRODUCTION Error! Bookmark not defined.
1.1. SUMMARY OF KEY NAS (2006) COMMENTS ON DOSE-RESPONSE
MODELING IN THE 2003 REASSESSMENT.. .Error! Bookmark not defined.
1.2. EPA's SCIENCE PLAN Error! Bookmark not defined.
1.3. SAB REVIEW OF EPA'S DRAFT REANALYSISError! Bookmark not defined.
1.4. SCOPE OF EPA'S REANALYSIS VOLUMES 1 AND 2Error! Bookmark not defined.
1.5. OVERVIEW OF EPA'S RESPONSE TO NAS (2006)Error! Bookmark not defined.
1.5.1. TCDD Literature Update Error! Bookmark not defined.
1.5.2. EPA's 2009 Workshop on TCDD Dose ResponseError! Bookmark not defined.
1.5.3. Organization of EPA's Response to NAS Recommendations
(Reanalysis Volume 1) Error! Bookmark not defined.
2. TRANSPARENCY AND CLARITY IN THE SELECTION OF KEY DATA
SETS FOR DOSE-RESPONSE ANALYSIS Error! Bookmark not defined.
2.1. SUMMARY OF NAS COMMENTS ON TRANSPARENCY AND
CLARITY IN THE SELECTION OF KEY DATA SETS FOR
DOSE-RESPONSE ANALYSIS Error! Bookmark not defined.
2.2. EPA'S RESPONSE TO NAS COMMENTS ON TRANSPARENCY
AND CLARITY IN THE SELECTION OF KEY DATA SETS FOR
DOSE-RESPONSE ANALYSIS Error! Bookmark not defined.
2.3. STUDY SELECTION PROCESS FOR TCDD DOSE-RESPONSE
ANALYSIS Error! Bookmark not defined.
2.3.1. Study Inclusion Criteria for TCDD Epidemiologic StudiesError! Bookmark not defined
2.3.2. Study Inclusion Criteria for TCDD In Vivo Mammalian BioassaysError! Bookmark not
2.4. SUMMARY OF KEY DATA SET SELECTION FOR TCDD
DOSE-RESPONSE MODELING Error! Bookmark not defined.
2.4.1. Key Epidemiologic Data Sets Error! Bookmark not defined.
2.4.2. Key Animal Bioassay Data Sets Error! Bookmark not defined.
3. THE USE OF TOXICOKINETICS IN THE DOSE-RESPONSE MODELING
FOR CANCER AND NONCANCER ENDPOINTS Error! Bookmark not defined.
3.1. SUMMARY OF NAS COMMENTS ON THE USE OF
TOXICOKINETICS IN DOSE-RESPONSE MODELING
APPROACHES FOR TCDD Error! Bookmark not defined.
3.2. OVERVIEW OF EPA'S RESPONSE TO THE NAS COMMENTS ON
THE USE OF TOXICOKENTICS IN DOSE-RESPONSE MODELING
APPROACHES FOR TCDD Error! Bookmark not defined.
in
DRAFT - DO NOT CITE OR QUOTE
-------
CONTENTS (continued)
3.3. PHARMACOKINETICS (PK) AND PK MODELINGError! Bookmark not defined.
3.3.1. PK Data and Models in TCDD Dose-Response Modeling:
Bookmark not defined.
Bookmark not defined.
Bookmark not defined.
Bookmark not defined.
Bookmark not defined.
Bookmark not defined.
Overview and Scope Error!
3.3.2. PK of TCDD in Animals and Humans Error!
3.3.2.1. Absorption and Bioavailability....Error!
3.3.2.2. Distribution Error!
3.3.2.3. Metabolism and Protein Binding.Error!
3.3.2.4. Elimination Error!
3.3.2.5. Interspecies Differences and SimilaritiesError! Bookmark not defined.
3.3.3. PK of TCDD in Humans: Interindividual VariabilityError! Bookmark not defined.
3.3.3.1. Life Stage and Gender Error! Bookmark not defined.
3.3.3.2. Physiological States: Pregnancy and LactationError! Bookmark not defined.
3.3.3.3. Lifestyle and Habits Error! Bookmark not defined.
3.3.3.4. Genetic Traits and PolymorphismError! Bookmark not defined.
3.3.4. Dose Metrics and Pharmacokinetic Models for TCDDError! Bookmark not defined.
3.3.4.1. Dose Metrics for Dose-Response ModelingError! Bookmark not defined.
3.3.4.2. First-Order Kinetic Modeling Error! Bookmark not defined.
3.3.4.3. Biologically Based Kinetic ModelsError! Bookmark not defined.
3.3.4.4. Applicability of PK Models to Derive Dose Metrics for
Dose-Response Modeling of TCDD: Confidence and
Limitations Error! Bookmark not defined.
3.3.4.5. Recommended Dose Metrics for Key StudiesError! Bookmark not defined.
3.3.5. Uncertainty in Dose Estimates Error! Bookmark not defined.
3.3.5.1. Sources of Uncertainty in Dose Metric PredictionsError! Bookmark not define
3.3.5.2. Qualitative Discussion of Uncertainty in Dose MetricsError! Bookmark not del
3.3.6. Use of the Emond PBPK Models for Dose Extrapolation from
Rodents to Humans Error! Bookmark not defined.
4. CHRONIC ORAL REFERENCE DOSE Error! Bookmark not defined.
4.1. NAS COMMENTS AND EPA'S RESPONSE ON IDENTIFYING
NONCANCER EFFECTS OBSERVED AT LOWEST DOSESError! Bookmark not defined.
4.2. NONCANCER DOSE-RESPONSE ASSESSMENT OF TCDDError! Bookmark not defined.
4.2.1. Determination of Toxicologically Relevant EndpointsError! Bookmark not defined.
4.2.2. Use of Toxicokinetic Modeling for TCDD Dose-Response
Assessment Error! Bookmark not defined.
4.2.3. Noncancer Dose-Response Assessment of Epidemiological DataError! Bookmark not d
4.2.3.1. Baccarelli et al. (2008) Error! Bookmark not defined.
4.2.3.2. Mocarelli et al. (2008) Error! Bookmark not defined.
4.2.3.3. Alaluusua et al. (2004) Error! Bookmark not defined.
4.2.3.4. Eskenazi et al. (2002b) Error! Bookmark not defined.
4.2.4. Noncancer Dose-Response Assessment of Animal Bioassay DataError! Bookmark not d
4.2.4.1. Use of Kinetic Modeling for Animal Bioassay DataError! Bookmark not defin
4.2.4.2. Benchmark Dose Modeling of the Animal Bioassay DataError! Bookmark not
iv DRAFT - DO NOT CITE OR QUOTE
-------
CONTENTS (continued)
4.2.4.3. POD Candidates from Animal Bioassays Based on HED
and BMD Modeling Results Error! Bookmark not defined.
4.3. RfD DERIVATION Error! Bookmark not defined.
4.3.1. Toxicological Endpoints Error! Bookmark not defined.
4.3 .2. Exposure Protocols of Candidate PODs . .. Error! Bookmark not defined.
4.3.3. Uncertainty Factors (UFs) Error! Bookmark not defined.
4.3.4. Choice of Human Studies for RfD DerivationError! Bookmark not defined.
4.3.4.1. Identification of POD from Baccarelli et al. (2008)Error! Bookmark not define
4.3.4.2. Identification of POD from Mocarelli et al. (2008)Error! Bookmark not define
4.3.4.3. Identification of POD from Alaluusua et al. (2004)Error! Bookmark not define
4.3.5. Derivation of the RfD Error! Bookmark not defined.
4.3.6. Studies Reporting Outcomes Comparable to the Principal Studies
Used to Derive the RfD Error! Bookmark not defined.
4.3.6.1. Dysregulation of Thyroid Hormone Metabolism
Associated with Dioxin Exposure in NeonatesError! Bookmark not defined.
4.3.6.2. Male Reproductive Effects associated with Dioxin
Exposures Error! Bookmark not defined.
4.4. QUALITATIVE UNCERTAINTIES IN THE RfDError! Bookmark not defined.
4.5. QUANTITATIVE UNCERTAINTY IN THE RfDError! Bookmark not defined.
4.5.1. Epidemiological Sensitivity Analyses Error! Bookmark not defined.
4.5.1.1. Mocarelli et al. (2008) Error! Bookmark not defined.
4.5.1.2. Baccarelli et al. (2008) Error! Bookmark not defined.
4.5.2. Sensitivity Analysis of the Candidate RfD Based on NTP (2006a)Error! Bookmark not <
4.5.3. Evaluation of Range of Alternative PODs for Additional
Epidemiological Endpoints Error! Bookmark not defined.
5. References Error! Bookmark not defined.
APPENDIX A: SUMMARY OF EXTERNAL PEER REVIEW AND PUBLIC
COMMENTS AND DISPOSITION A-l
APPENDIX B: DIOXIN WORKSHOP B-l
APPENDIX C: SUMMARIES AND EVALUATIONS OF CANCER AND
NONCANER EPIDEMIOLOGICAL STUDIES FOR INCLUSION IN
TCDD DOSE-RESPONSE ASSESSMENT C-l
APPENDIX D: SUMMARIES AND EVALUATIONS OF CANCER AND
NONCANCER IN VIVO ANIMAL BIO AS SAY STUDIES FOR
INCLUSION IN TCDD DOSE-RESPONSE ASSESSMENT D-l
v
DRAFT - DO NOT CITE OR QUOTE
-------
CONTENTS (continued)
APPENDIX E: RODENT BIOASSAY KINETIC MODELING E-l
APPENDIX F: EPIDEMIOLOGICAL KINETIC MODELING F-l
APPENDIX G: NONCANCER BENCHMARK DOSE MODELING G-l
APPENDIX H: ENDPOINTS EXCLUDED FROM REFERENCE DOSE DERIVATION
BASED ON TOXICOLOGICAL RELEVANCE II-l
APPENDIX I: LITERATURE SEARCH TERMS 1-1
vi
DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF TABLES
2-1. Epidemiologic studies selected for TCDD cancer dose-response modelingEiror! Bookmark not definec
2-2. Epidemiologic studies selected for TCDD noncancer dose-response modelingEiror! Bookmark not def
2-3. Animal bioassays selected for cancer dose-response modelingError! Bookmark not defined.
2-4. Animal bioassay studies selected for noncancer dose-response modelingError! Bookmark not defined.
3-1. Partition coefficients, tissue volumes, and volume of distribution for TCDD in
humans Error! Bookmark not defined.
3-2. Blood flows, permeability factors, and resulting half lives (VA) for perfusion
losses for humans as represented by the TCDD PBPK model of Emond et al.
(2006; 2005) Error! Bookmark not defined.
3-3. Toxicokinetic conversion factors for calculating human equivalent doses from
rodent bioassays based on first-order kinetics Error! Bookmark not defined.
3-4. Equations used in the concentration and age-dependent model (CADM; Aylward
et al., 2005b) Error! Bookmark not defined.
3-5. Parameters of the concentration and age-dependent model (CADM; Aylward et
al., 2005b) Error! Bookmark not defined.
3-6. Confidence in the CADMa model simulations of TCDD dose metricsbError! Bookmark not defined.
3-7. Equations used in the TCDD PBPK model of Emond et al. (2006)Error! Bookmark not defined.
3-8. Parameters of the PBPK model for TCDD Error! Bookmark not defined.
3-9. Regression analysis results for the relationship between logio serum TCDD at the
midpoint of observations and the logio of the rate constant for decline of TCDD
levels using Ranch Hand data Error! Bookmark not defined.
3-10. Dosing protocols for human and animal models Error! Bookmark not defined.
3-11. Most sensitive variables for the rat and mouse nongestational and gestational
models Error! Bookmark not defined.
3-12. Most sensitive variables for the human nongestational and gestational modelsError! Bookmark not defi
3-13. TCDD serum measurements over time for two Austrian women exposed to
TCDD in 1997 Error! Bookmark not defined.
3-14. TCDD serum measurements over time for two Seveso males exposed to TCDD in
1976 Error! Bookmark not defined.
3-15. Results of Hill coefficient sensitivity analysis simulations with Emond human
PBPK model Error! Bookmark not defined.
3-16. Alternative CYP1A2 parameter estimates for sensitivity analysis of Emond
human PBPK model Error! Bookmark not defined.
3-17. Results of CYP1A2 parameter sensitivity analysis simulations with Emond
human PBPK model Error! Bookmark not defined.
vii DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF TABLES (continued)
3-18. Results of Emond human PBPK model parameter sensitivity analysis simulations.
Comparison of modeled human oral intakes for a range of lifetime average TCDD
serum concentrations for alternative parameter values Error! Bookmark not defined.
3-19. Confidence in the PBPK model simulations of TCDD dose metricsError! Bookmark not defined.
3-20. Overall confidence associated with alternative dose metrics for noncancer dose-
response modeling for TCDD using rat PBPK model Error! Bookmark not defined.
3-21. Overall confidence associated with alternative dose metrics for noncancer dose-
response modeling for TCDD using mouse PBPK model .Error! Bookmark not defined.
3-22. Contributors to the overall confidence in the selection and use of dose metrics in
the dose-response modeling of TCDD based on rat and human PBPK modelsError! Bookmark not defi
3-23. Contributors to the overall uncertainty in the selection and use of dose metrics in
the dose-response modeling of TCDD based on mouse and human PBPK modelsError! Bookmark not
3-24. Comparison of human equivalent doses from the Emond human PBPK model for
the 45-year-old and 25-year-old gestational exposure scenariosError! Bookmark not defined.
3-25. Impact of toxicokinetic modeling on the extrapolation of administered dose to
HED, comparing the Emond PBPK and first-order body burden models
(administered dose = 1 ng/kg-day) Error! Bookmark not defined.
4-1. PODs for epidemiologic studies of TCDD Error! Bookmark not defined.
4-2. Models run for each study/endpoint combination in the animal bioassay
benchmark dose modeling Error! Bookmark not defined.
4-3. Summary of key animal study PODs (ng/kg-day) based on three different dose
metrics: administered dose, first-order body burden HED, and blood
concentration Error! Bookmark not defined.
4-4. TCDD BMDL analysis (NOAEL, LOAEL, BMD, and BMDL values given as
animal whole blood concentrations in ng/kg) Error! Bookmark not defined.
4-5. Candidate PODs for the TCDD RfD using blood-concentration-based human
equivalent doses Error! Bookmark not defined.
4-6. Qualitative analysis of the strengths and limitations/uncertainties associated with
animal bioassays providing PODs for the TCDD RfD Error! Bookmark not defined.
4-7. Basis and derivation of the TCDD reference dose Error! Bookmark not defined.
4-8. Alternative PODs for the impact of TCDD exposure during gestation and nursing
on semen quality of male offspring (Mocarelli et al., 201 l)Error! Bookmark not defined.
4-9. Alternative PODs for developmental endpoints other than increased neonatal TSH
and semen quality Error! Bookmark not defined.
4-10. Alternative PODs for adult endpoints for which critical exposure windows are
undefined Error! Bookmark not defined.
viii DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF FIGURES
2-1. EPA's process to select and identify in vivo mammalian and epidemiologic
studies for use in the dose-response analysis of TCDD Error! Bookmark not defined.
2-2. EPA's selection process to evaluate available epidemiologic studies using study
inclusion criteria and other epidemiologic considerations for use in the dose-
response analysis of TCDD Error! Bookmark not defined.
2-3. EPA's process to evaluate available animal bioassay studies using study inclusion
criteria for use in the dose-response analysis of TCDD. ...Error! Bookmark not defined.
2-4. Results of EPA's process to select and identify in vivo mammalian and
epidemiologic studies for use in the dose-response analysis of TCDD.Error! Bookmark not defined.
3-1. Liver/fat concentration ratios in relation to TCDD dose at various times after oral
administration of TCDD to mice Error! Bookmark not defined.
3-2. First-order elimination rate fits to 36 sets of serial TCDD sampling data from
Seveso patients as function of initial serum lipid TCDD. .Error! Bookmark not defined.
3-3. Observed relationship of fecal 2,3,7,8-TCDD clearance and estimated percent
body fat Error! Bookmark not defined.
3-4. Unweighted empirical relationship between percent body fat estimated from body
mass index and TCDD elimination half-life—combined Ranch Hand and Seveso
observations Error! Bookmark not defined.
3-5. Relevance of candidate dose metrics for dose-response modeling, based on mode
of action and target organ toxicity of TCDD Error! Bookmark not defined.
3-6. Process of estimating a human-equivalent TCDD lifetime average daily oral
exposure (dn) from an experimental animal average daily oral exposure (d\) based
on the body-burden dose metric Error! Bookmark not defined.
3-7. Human body burden time profiles for achieving a target body burden for different
exposure duration scenarios Error! Bookmark not defined.
3-8. Schematic of the CADM structure Error! Bookmark not defined.
3-9. Comparison of observed and simulated fractions of the body burden contained in
the liver and adipose tissues in rats Error! Bookmark not defined.
3-10. Conceptual representation of PBPK model for rat exposed to TCDD.Error! Bookmark not defined.
3-11. Conceptual representation of PBPK model for rat developmental exposure to
TCDD Error! Bookmark not defined.
3-12. TCDD distribution in the liver tissue Error! Bookmark not defined.
3-13. Growth rates for physiological changes occurring during gestation.Error! Bookmark not defined.
3-14. Comparisons of model predictions to experimental data using a fixed elimination
rate model with hepatic sequestration (A) and an inducible elimination rate model
with (B) and without (C) hepatic sequestration Error! Bookmark not defined.
IX
DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF FIGURES (continued)
3-15. PBPK model simulation of hepatic TCDD concentration (ppb) during chronic
exposure to TCDD at 50, 150, 500, or 1,750 ng TCDD/BW using the inducible
elimination rate model compared with the experimental data measured at the end
of exposure Error! Bookmark not defined.
3-16. Model predictions of TCDD blood concentration in 10 veterans (A-J) from
Ranch Hand Cohort Error! Bookmark not defined.
3-17. Time course of TCDD in blood (pg/g lipid adjusted) for two highly exposed
Austrian women (patients 1 and 2) Error! Bookmark not defined.
3-18. Observed vs. Emond et al. (2005) model simulated serum TCDD concentrations
(pg/g lipid) over time (In = natural log) in two Austrian women.Error! Bookmark not defined.
3-19. Comparison of the dose dependency of TCDD elimination in the Emond model
vs. observations of nine Ranch Hand veterans and two highly exposed Austrian
patients Error! Bookmark not defined.
3-20. Elasticities in the nongestational human model, POD dose.Error! Bookmark not defined.
3-21. Elasticities in the nongestational human model, RfD dose.Error! Bookmark not defined.
3-22. Hill coefficient sensitivity analysis Error! Bookmark not defined.
3-23. CYP1A2 parameter sensitivity analysis Error! Bookmark not defined.
3-24. Experimental data (symbols) and model simulations (solid lines) of (A) blood, (B)
liver, and (C) adipose tissue concentrations of TCDD after oral exposure to 150
ng/kg-day, 5 days/week, for 17 weeks in mice Error! Bookmark not defined.
3-25. Comparison of PBPK model simulations with experimental data on liver
concentrations in mice administered a single oral dose of 0.001-300 jag
TCDD/kg Error! Bookmark not defined.
3-26. Comparison of model simulations (solid lines) with experimental data (symbols)
on the effect of dose on blood (cb), liver (cli), and fat (cf) concentrations
following repetitive exposure to 0.1-450 ng TCDD/kg, 5 days/week, for 13
weeks in mice Error! Bookmark not defined.
3-27. Comparison of experimental data (symbols) and model predictions (solid lines) of
(A) blood, (B) liver, and (C) adipose tissue concentrations of TCDD after oral
exposure to 1.5 ng/kg-day, 5 days/week, for 17 weeks in mice.Error! Bookmark not defined.
3-28. Comparison of experimental data (symbols) and model predictions (solid lines) of
(A) blood concentration, (B) liver concentration, (C) adipose tissue concentration,
(D) feces excretion (% dose), and (E) urinary elimination (% dose) of TCDD after
oral exposure to 1.5 ng/kg-day, 5 days/week, for 13 weeks in mice.Error! Bookmark not defined.
x
DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF FIGURES (continued)
3-29. Comparison of experimental data (symbols) and model predictions (solid lines) of
(A) blood concentration, (B) liver concentration, (C) adipose tissue concentration,
(D) feces excretion (% dose), and (E) urinary elimination (% dose) of TCDD after
oral exposure to 150 ng/kg-day, 5 days/week, for 13 weeks in mice.Error! Bookmark not defined.
3-30. PBPK model simulations (solid lines) vs. experimental data (symbols) on the
distribution of TCDD after a single acute oral exposure to A-B) 0.1, C-D) 1.0,
and E-F) 10 jag of TCDD/kg of body weight in mice Error! Bookmark not defined.
3-31. PBPK model simulation (solid lines) vs. experimental data (symbols) on the
distribution of TCDD after a single dose of 24 (J,g/kg BW on GD 12 in mice.Error! Bookmark not defii
3-32. Comparison of the near-steady-state body burden simulated with CADM and
Emond models for a daily dose ranging from 0 to 10,000 ng/kg-day in rats and
humans Error! Bookmark not defined.
3-33. TCDD serum concentration-time profile for lifetime, less-than-lifetime, and
gestational exposure scenarios, with target concentrations shown for each; profiles
generated with Emond human PBPK model Error! Bookmark not defined.
3-34. TCDD serum concentration-time profile for lifetime, less-than-lifetime. and
gestational exposure scenarios, showing continuous intake levels to fixed target
concentration; profiles generated with Emond human PBPK model.Error! Bookmark not defined.
4-1. EPA's process to identify and estimate PODs from key epidemiologic studies for
use in noncancer dose-response analysis of TCDD Error! Bookmark not defined.
4-2. Disposition of noncancer animal bioassays selected for TCDD dose-response
analysis Error! Bookmark not defined.
4-3. EPA's process to identify and estimate PODs from key animal bioassays for use
in noncancer dose-response analysis of TCDD Error! Bookmark not defined.
4-4. Exposure-response array for ingestion exposures to TCDD.Error! Bookmark not defined.
4-5. Candidate RfD array Error! Bookmark not defined.
4-6. Sensitivity tree showing TCDD exposure-variable uncertainty for Mocarelli et al.
(2008) Error! Bookmark not defined.
4-7. Sensitivity tree showing TCDD exposure-variable uncertainty for Baccarelli et al.
(2008) Error! Bookmark not defined.
4-8. Sensitivity tree showing TCDD exposure-variable uncertainty for NTP (2006a).Error! Bookmark not d
4-9. Alternative POD exposure-response array Error! Bookmark not defined.
XI
DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS
2,4,5-T 2,4,5-trichlorophenoxyacetic acid
2,4-D 2,4-dichlorophenoxyacetic acid
AA ascorbic acid
ACOH acetanilide-4-hydroxylase
AHH aryl hydrocarbon hydroxylase
AhR aryl hydrocarbon receptor
AhR-/- AhR-deficient
AIC Akaike Information Criterion
ANL Argonne National Laboratory
ANOVA analysis of variance
APE airborne particulate extract
ASAT aspartate aminotransferase
AUC area under the curve
bHLH-PAS basic helix-loop-helix, Per-Arnt-Sim
Bmax equilibrium maximum binding capacity
BMD benchmark dose
BMDL benchmark dose lower confidence bound
BMDS Benchmark dose software
BMI body mass index
BMR benchmark response
BPS balanopreputial separation
BROD benzyloxy resoufin-O-deethylase
b-TSH blood thyroid-stimulating hormone
BW body weight
C cerebellum
CADM concentration- and age-dependent elimination model
Cc cerebral cortex
CI confidence interval
CSAF chemical-specific adjustment factor
CSLC cumulative serum lipid concentration
Cx connexin
CYP cytochrome P450
Da:HED ratio of administered dose to HED
DEN diethylnitrosamine
df degrees of freedom
DLC dioxin-like compound
DRE/XRE dioxin/xenobiotic response elements
DRL differential reinforcement of low rate
DSA delayed spatial alteration
E2 17P-estradiol
EDX effective dose eliciting x percent response
EGFR epidermal growth factor receptor
Xll
DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS (continued)
EPA
Environmental Protection Agency
ER
estrogen receptor
EROD
7-ethoxyresorufin-O-deethylase
ERa
estrogen receptor alpha
EU
European Union
FFA
free fatty acid
FR
fixed ratio
FSH
follicle stimulating hormone
FT4
free thyroxine
GD
gestation day
GSH
glutathione stimulating hormone
GSH-Px
glutathione stimulating hormone peroxidase
GST
glutathiones-transferase
H
hippocampus
HCH
hexachlorocyclohexane
HED
human equivalent dose
HQ
hazard quotient
HR
hazard ratio
Hsp90
heat shock protein 90
IARC
International Agency for Research on Cancer
IGF
insulin-like growth factor
IL
interleukin
ILSI
International Life Sciences Institute
i.p.
intraperitoneal
IRIS
Integrated Risk Information System
KABS
oral absorption parameters
LASC
lipid-adjusted serum concentration
LD50
lethal dose eliciting x percent response
LED
lower confidence effective dose
LEDX
lower bound of the 95% confidence interval on the dose that yields an x% effect
LH
luteinizing hormone
LOAEL
lowest-observed-adverse-effect level
LOAELhed
HED estimate based on LOAELs
LOEL
lowest-observed-adverse level
MCH
mean corpuscular hemoglobin
MCMC
Markov Chain Monte Carlo
MCV
mean corpuscular volume
MOA
mode of action
MOE
margin of exposure
MROD
7-methoxyresorufin-O-deethylase
MTD
maximum tolerated dose
NAS
National Academy of Sciences
NIOSH
National Institute for Occupational Safety and Health
Xlll
DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS (continued)
NOAEL
no-observed-adverse-effect level
NOEL
no-ob served-effect level
NRC
National Research Council
NTP
National Toxicology Program
OR
odds ratio
OSF
oral slope factor
PA
permeability x area
PAI2
plasminogen activator inhibitor 2
PBMC
peripheral blood mononuclear cells
PBPK
physiologically based pharmacokinetic
PCB
polychlorinated biphenyl
PCDD
polychlorinated dibenzo-p-dioxin
PCDF
polychlorinated dibenzofuran
PEPCK
phosphoenolpyruvate carboxykinase
PF
adipose tissue:blood partition coefficient
PHAH
polyhalogenated aromatic hydrocarbons
PK
pharmacokinetic
PND
postnatal day
POD
point of departure
PP
phosphotyrosyl protein
PRA
probabilistic risk assessment
PRE
body:blood partition coefficient
PROD
7-pentoxyresorufin-O-deethylase
RAR
retinoic acid receptor
REP
relative potency
RfC
reference concentration
RfD
reference dose
RL
reversal learning
RL
risk level
RR
rate ratios
RR
relative risk
RT-PCR
reverse transcription polymerase chain reaction
RXR
retinoid X receptor
S
saline
SA
superoxide anion
SAhRM
SRM for AhRs
S-D
Sprague-Dawley
SD
standard deviation
SIR
standardized incidence ratio
SMR
standardized mortality ratio
SOD
superoxide dismutase
SRBC
sheep red blood cell
SSB
single-strand break
xiv
DRAFT - DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS (continued)
SWHS
Seveso Women's Health Study
T4
thyroxine
TBARS
thiobarbituric acid-reactive substances
TCB
3,3' ,4,4' -tetrachlorobiphenyl
TCDD
2,3,7,8 -T etrachl orodib enzo-p-di oxin
TCP
2,4,5-trichlorophenol
TEF
toxicity equivalence factor
TEQ
toxicity equivalence
TGFa
transforming growth factor a
TK
toxicokinetic
TNF-a
tumor necrosis factor alpha
TOTTEQ
total toxicity equivalence
TSH
thyroid stimulating hormone
TT4
total thyroxine
TWA
time-weighted average
U.S. NRC
U.S. Nuclear Regulatory Commission
UDP
uridine diphosphate
UDPGT
UDP-glucoronosyl transferase
UED
upper confidence bound for the effective dose
UF
uncertainty factor
UFa
interspecies extrapolation factor
UFd
database factor
UFh
human interindividual variability
UFl
LOAEL-to-NOAEL UF
UFS
subchronic-to-chronic UF
UGT
UDP-glucuronosyltransferase
UGT1
uridine diphosphate glucuronosyltransferase I
Vd
volume of distribution
WHO
World Health Organization
xv
DRAFT - DO NOT CITE OR QUOTE
-------
PREFACE
This draft report was developed by the U.S. Environmental Protection Agency's (EPA)
Office of Research and Development (ORD), National Center for Environmental Assessment
(NCEA).
In 2003, EPA, along with other federal agencies, asked the National Academy of
Sciences (NAS) to review aspects of the science in EPA's draft dioxin reassessment entitled,
Exposure and Human Health Reassessment of 2,3,7,8-Tetrachlorodibenzo-p-Dioxin (TCDD) and
Related Compounds ("2003 Reassessment"), and, in 2004, EPA sent the 2003 draft dioxin
reassessment to the NAS for their review. In 2006, NAS released the report of their review
entitled, Health Risks from Dioxin and Related Compounds: Evaluation of the EPA
Reassessment. NAS identified three areas in EPA's 2003 draft reassessment that required
improvement: (1) justification of approaches to dose-response modeling for cancer and
noncancer endpoints; (2) transparency and clarity in selection of key data sets for analysis; and
(3) transparency, thoroughness, and clarity in quantitative uncertainty analysis. NAS provided
EPA with recommendations to address their key concerns.
In 2008, EPA, in collaboration with the Department of Energy's Argonne National
Laboratory (ANL), developed and published a literature database of peer-reviewed studies on
TCDD toxicity, including in vivo mammalian dose-response studies and epidemiologic studies.
EPA subsequently requested public comment on this database. EPA and ANL then convened a
scientific workshop in 2009. The Workshop goals were to identify and address issues related to
the dose-response assessment of TCDD and to ensure that EPA's response to the NAS focused
on the key issues and reflected the most meaningful science.
In May 2010, EPA released a draft report entitled EPA 's Reanalysis of Key Issues
Related to Dioxin Toxicity and Response to NAS Comments ("Reanalysis") that provided a
technical response to the 2006 NAS report. The draft Reanalysis (1) developed a study selection
process to evaluate studies reporting cancer and noncancer effects; (2) utilized a TCDD
physiologically-based pharmacokinetic (PBPK) model in its development of dose-response
analyses of TCDD toxicological and epidemiological literature; (3) presented new analyses of
both the potential cancer and noncancer human health effects that may result from exposures to
TCDD; (4) developed an oral reference dose (RfD) for TCDD; and (5) developed a new cancer
oral slope factor for TCDD. Federal agencies and White House offices were provided an
opportunity for review and comment on the draft Reanalysis prior to its public release; their
comments are available at www.epa.gov/iris.
The draft Reanalysis received public comments and was provided to EPA's Science
Advisory Board (SAB) for independent external peer review. The SAB convened an expert
panel composed of scientists knowledgeable about technical issues related to dioxins and risk
assessment. For their review, SAB held public meetings in June, July, and October 2010, and in
March and June 2011.
SAB released their final report on August 26, 201 1. In their final report, the SAB panel:
(1) commended the comprehensive and rigorous process that was used to identify and evaluate
the TCDD literature; (2) agreed that EPA's choice of kinetic model provided the best available
basis for the dose metric calculations; (3) supported EPA's selection of two cocritical
epidemiologic studies for the derivation of the RfD for TCDD; and (4) generally agreed with
EPA's characterization of TCDD as carcinogenic to humans in accordance with EPA's 2005
xvi DRAFT - DO NOT CITE OR QUOTE
-------
Guidelines for Carcinogen Risk Assessment and with EPA's selection of the critical study for the
quantitative cancer assessment. However, SAB found that the draft Reanalysis did not respond
adequately to the NAS recommendation to adopt both linear and nonlinear methods of
extrapolation to account for the uncertainty in the cancer dose-response curve for TCDD. Also,
the SAB report conveyed disagreement with EPA's position in the draft Reanalysis that a
comprehensive uncertainty analysis was infeasible and suggested a number of methods that
could be used for this purpose.
Based on the SAB review, EPA decided to separate the dioxin assessment into
two portions, the noncancer assessment (Volume 1) and the cancer assessment and quantitative
uncertainty analysis (Volume 2). This document, Volume 1, comprises the noncancer portion of
EPA 's Reanalysis of Key Issues Related to Dioxin Toxicity and Response to NAS Comments.
After completing the noncancer portion, EPA will complete Volume 2.
This document, Volume 1, responds to the recommendations and comments on
noncancer TCDD dose-response assessment included in the 2006 NAS report, focusing on the
NAS comments regarding TCDD dose-response assessment. Volume 1 systematically evaluates
the epidemiologic studies and rodent bioassays relevant to TCDD dose response. It uses a
TCDD PBPK model to simulate TCDD blood concentrations, the dose metric used in all dose-
response analyses for TCDD. Volume 1 also develops an oral reference dose (RfD) based on
two epidemiologic studies that associated TCDD exposures with adverse health effects. The first
study reports decreased sperm concentration and sperm motility in men who were exposed to
TCDD during childhood during the Seveso accident (Mocarelli et al.. 2008). and the second
reports increased thyroid-stimulating hormone levels in newborns born to mothers who were
exposed to TCDD during the Seveso accident (Baccarelli et al.. 2008). Volume 1 also provides a
focused quantitative uncertainty analysis of the decisions made in the development of points of
departure for TCDD RfD derivation.
In Volume 2, EPA will complete the evaluation of cancer mode-of-action, cancer
dose-response modeling, including justification of the approaches used for dose-response
modeling of the cancer endpoints, and an associated quantitative uncertainty analysis. The
information provided in Volume 1 will be used in three ways: (1) as the first of two reports that
contain EPA's response to the NAS (2006b) report, (2) as the Support Document for the TCDD
noncancer IRIS Summary and TCDD oral RfD, and (3) as technical support for Reanalysis
Volume 2.
xvii
DRAFT - DO NOT CITE OR QUOTE
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
PRIMARY AUTHORS
National Center for Environmental Assessment, U.S. Environmental Protection Agency,
Cincinnati, OH
Belinda Hawkins
Glenn Rice (Project Co-Lead)
Jeff Swartout (Project Co-Lead)
Linda K. Teuschler
CONTRIBUTING AUTHORS
National Center for Environmental Assessment, U.S. Environmental Protection Agency,
Cincinnati, OH
Janet Hess-Wilson (formerly with NCEA, currently with Department of Defense)
Scott Wesselkamper
Michael Wright
Bette Zwayer
National Health and Environmental Effects Research Laboratory, U.S. Environmental Protection
Agency, Research Triangle Park, NC
Hi sham El-Masri
Argonne National Laboratory, Argonne, IL
Margaret MacDonell
University of Montreal; BioSimulation Consulting, Newark, DE
Claude Emond
University of Montreal, Montreal, Canada
Kannan Krishnan
CONTRIBUTORS
National Center for Environmental Assessment, U.S. Environmental Protection Agency,
Washington, DC
Karen Hogan
Leonid Kopylev
Argonne National Laboratory, Argonne, IL
Maryka H. Bhattacharyya Mary E. Finster
Andrew Davidson David P. Peterson
xviii
DRAFT - DO NOT CITE OR QUOTE
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS (continued)
CONTRIBUTORS (continued)
Bruce Allen Consulting, Chapel Hill, NC
Bruce C. Allen
Clark University, Worcester, MA
Dale Hattis
Colorado State University, Fort Collins, CO
Raymond Yang, Retired
Emory University, Atlanta, GA
Kyle Steenland
ICF International, Durham, NC
Robyn Blain
Rebecca Boyles
Patty Chuang
Cara Henning
Baxter Jones
Penelope Kellar
Mark Lee
Nikki Maples-Reynolds
Amalia Marenberg
Penn State University, University Park, PA
Jack P. Vanden Heuvel
Resources for the Future, Washington, DC
Roger M. Cooke
Risk Sciences International, Ottawa, Ontario
Jessica Dennis Salomon Sand
Dan Krewski Natalia Shilnikova
Greg Paoli Paul Villenueve
University of California-Berkeley, Berkeley, CA
Brenda Eskenazi
University of California-Irvine, Irvine, CA
Scott Bartell
Garrett Martin
Margaret McVey
Sara Mishamandani
Chandrika Moudgal
Bill Mendez
Ami Parekh
Andrew Shapiro
Courtney Skuce
Audrey Turley
xix
DRAFT - DO NOT CITE OR QUOTE
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS (continued)
REVIEWERS
This document has been provided for review to EPA scientists and interagency reviewers
from other federal agencies and White House offices.
INTERNAL REVIEWERS
National Center for Environmental Assessment,
Ted Berner, Washington, DC
Glinda Cooper, Washington, DC
Ila Cote, Research Triangle Park, NC
Lynn Flowers, Washington, DC
Martin Gehlhaus, Washington, DC
Kate Guyton, Washington, DC
Samantha Jones, Washington, DC
S. Environmental Protection Agency
Matthew Lorber, Washington, DC
Eva McLanahan, Research Triangle Park, NC
Susan Rieth, Washington, DC
Reeder Sams, Research Triangle Park, NC
Paul Schlosser, Research Triangle Park, NC
Jamie Strong, Washington, DC
John Vandenberg, Research Triangle Park, NC
U.
U.S. ENVIRONMENTAL PROTECTION AGENCY
SCIENCE ADVISORY BOARD
DIOXIN REVIEW PANEL
Chair
Timothy Buckley, Associate Professor and Chair, Division of Environmental Health Sciences,
College of Public Health, The Ohio State University, Columbus, OH
Members
Harvey Clewell, Director of the Center for Human Health Assessment, The Hamner Institutes for
Health Sciences, Research Triangle Park, NC
Louis Anthony (Tony) Cox, Jr., President, Cox Associates, Denver, CO
Elaine Faustman, Professor and Director, Institute for Risk Analysis and Risk Communication,
School of Public Health, University of Washington, Seattle, WA
Scott Ferson, Senior Scientist, Applied Biomathematics, Setauket, NY
Jeffrey Fisher, Research Toxicologist, National Center for Toxicological Research, U.S. Food
and Drug Administration, Jefferson, AR
Helen Hakansson, Professor of Toxicology, Unit of Environmental Health Risk Assessment,
Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden
Russ Hauser, Frederick Lee Hisaw Professor, Department of Environmental Health, Harvard
School of Public Health, Boston, MA
B. Paige Lawrence, Associate Professor, Departments of Environmental Medicine and
Microbiology and Immunology, School of Medicine and Dentistry, University of Rochester
School of Medicine and Dentistry, Rochester, NY
xx
DRAFT - DO NOT CITE OR QUOTE
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS (continued)
SCIENCE ADVISORY BOARD (continued)
Michael I. Luster, Professor, Department of Community Medicine, West Virginia University
Health Sciences Center, Morgantown, WV
Paolo Mocarelli, Professor of Clinical Biochemistry, Department of Clinical Laboratory,
Hospital of Desio-Nuovo Monoblous, University of Milano Bicocca, Desio-Milano, Italy
Victoria Persky, Professor, Epidemiology and Biostatistics Program, School of Public Health,
University of Illinois at Chicago, Chicago, IL
Sandra L. Petersen, Professor, Associate Graduate Dean, Department of Veterinary and Animal
Sciences, College of Natural Sciences, University of Massachusetts-Amherst, Amherst, MA
Karl Rozman, Professor, Pharmacology, Toxicology and Therapeutics, The University of Kansas
Medical Center, Kansas City, KS
Arnold Schecter, Professor, Environmental and Occupational Health Sciences, School of Public
Health-Dallas Campus, University of Texas, Dallas, TX
Allen E. Silverstone, Professor, Department of Microbiology and Immunology, Health Science
Center, SUNY Upstate Medical University, Syracuse, NY and Adjunct Professor of
Environmental Medicine, University of Rochester School of Medicine and Dentistry,
Rochester, NY
Mitchell J. Small, The H. John Heinz III Professor of Environmental Engineering, Department of
Civil and Environmental Engineering and Engineering and Public Policy, Carnegie Mellon
University, Pittsburgh, PA
Anne Sweeney, Professor of Epidemiology, Department of Epidemiology and Biostatistics,
School of Rural Public Health, Texas A&M Health Science Center, College Station, TX
Mary K. Walker, Professor, Division of Pharmaceutical Sciences, College of Pharmacy,
University of New Mexico, Albuquerque, NM
ACKNOWLEDGMENTS
National Center for Environmental Assessment,
Rebecca Clark, Washington, DC
Jeff Frithsen, Washington, DC
Kathleen Deener, Washington, DC
U.S. Environmental Protection Agency
Annette Gatchett, Cincinnati, OH
Maureen Johnson, Washington, DC
Linda Tux en, Washington, DC, Retired
Immediate Office of the Assistant Administrator of Office of Research and Development,
U.S. Environmental Protection Agency
Peter Preuss, Washington, DC
National Risk Management Research Laboratory, U.S. Environmental Protection Agency
Andrew Gillespie, Cincinnati, OH
xxi
DRAFT - DO NOT CITE OR QUOTE
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS (continued)
ACKNOWLEDGMENTS (continued)
Office of Administrative and Research Support, U.S. Environmental Protection Agency
Marie Nichols-Johnson, Cincinnati, OH
Colorado State University, Fort Collins, CO
William H. Farland
ECFlex, Inc., Fairborn, OH
Dan Heing
Heidi Glick
Debbie Kleiser
IntelliTech Systems, Inc., Fairborn, OH
Cris Broyles
Luella Kessler
Stacey Lewis
National Institute of Environmental Health Sciences, Research Triangle Park, NC
Linda S. Birnbaum
Christopher J. Portier
National Toxicology Program, Research Triangle Park, NC
Nigel Walker
Michael Devito
2009 Dioxin Workshop Participants
Crystal Lewis
Amy Prues
Lana Wood
Kathleen Secor
Linda Tackett
XXll
DRAFT - DO NOT CITE OR QUOTE
-------
EXECUTIVE SUMMARY
OVERVIEW
Dioxins and dioxin-like compounds (DLCs), including polychlorinated dibenzo-dioxins,
polychlorinated dibenzofurans, and polychlorinated biphenyls are structurally and
toxicologically related halogenated dicyclic aromatic hydrocarbons.1 Dioxins and DLCs are
released into the environment from several industrial sources such as chemical manufacturing,
combustion, and metal processing; from individual activities including the burning of household
waste; and from natural processes such as forest fires. Dioxins and DLCs are widely distributed
throughout the environment and typically occur as chemical mixtures. Additionally, they do not
readily degrade; therefore, levels persist in the environment, build up in the food chain, and
accumulate in the tissues of animals. Human exposure to these compounds occurs primarily
through the ingestion of contaminated foods (Lorber et al.. 2009). although exposures to other
environmental media and by other routes and pathways do occur.
The health effects from exposures to dioxins and DLCs have been documented
extensively in epidemiologic and toxicological studies. 2,3,7,8-Tetrachlorodibenzo-p-dioxin
(TCDD) is one of the most toxic members of this class of compounds and has a robust
toxicological database. Characterization of TCDD toxicity is critical to the risk assessment of
mixtures of dioxins and DLCs because it has been selected repeatedly as the "index chemical"
for the dioxin toxicity equivalence factors (TEF) approach. In this approach, the toxicity of
individual components of dioxin and DLC mixtures is scaled to that of TCDD. Then, the
dose-response information for TCDD is used by the U.S. Environmental Protection Agency
(EPA) and other organizations to evaluate risks from exposure to mixtures of DLCs (U.S. EPA.
2010b; Van den Berg et al.. 2006; Van den Berg et al.. 1998).
The EPA is committed to the development of health assessment information of the
highest scientific integrity for use in protecting human health and the environment. Scientific
peer review is an integral component of the process EPA uses to generate high quality toxicity
and exposure assessments of environmental contaminants. To this end, EPA asked the National
Academy of Sciences (NAS) to review its comprehensive human health assessment external
1 For further information on the chemical structures of these compounds, see U.S. EPA (U.S. EPA. 2010b. 2008b.
2003).
xxiii DRAFT - DO NOT CITE OR QUOTE
-------
review draft entitled, Exposure and Human Health Reassessment of 2,3,7,8-Tetrachlorodibenzo-
p-Dioxin (TCDD) and Related Compounds ("2003 Reassessment") (U.S. EPA. 2003).
In 2006, NAS released their report titled, Health Risks from Dioxin and Related
Compounds: Evaluation of the EPA Reassessment (NAS. 2006a). In this review, the NAS
identified three key recommendations requiring improvement to support a scientifically robust
characterization of human responses to exposures to TCDD. These three key areas are
(1) improved transparency and clarity in the selection of key data sets for dose-response analysis,
(2) further justification of approaches to dose-response modeling for cancer and noncancer
endpoints, and (3) improved transparency, thoroughness, and clarity in quantitative uncertainty
analysis. NAS also encouraged EPA to calculate an oral noncancer reference dose (RfD), and
provided specific comments on various aspects of EPA's 2003 Reassessment.
In May 2009, EPA Administrator Lisa P. Jackson announced the Science Plan for
Activities Related to Dioxins in the Environment ("Science Plan") that addressed the need to
finish EPA's dioxin reassessment and provide a completed health assessment on this high profile
chemical to the American public. The Science Plan states that EPA will release a draft report
that responds to the recommendations and comments included in the NAS review of EPA's 2003
Reassessment, and that, in this draft report, EPA's National Center for Environmental
Assessment, Office of Research and Development, will provide a limited response to key
comments and recommendations in the NAS report.
As required in the Science Plan, in 2009, EPA developed a draft report titled EPA's
Reanalysis of Key Issues Related to Dioxin Toxicity and Response to NAS Comments ("draft
Reanalysis") that responded to the key comments and recommendations in the NAS report (U.S.
EPA. 2010a). The draft Reanalysis focused on TCDD dose-response issues and included
analyses of relevant new studies and the derivation of an oral noncancer RfD and an oral slope
factor (OSF) for cancer. The draft Reanalysis was reviewed internally by EPA scientists and was
provided for review to other federal agencies and White House Offices. On May 21, 2010, the
draft Reanalysis was released for public review and comment and independent external peer
review by EPA's Science Advisory Board (SAB).
Available online at http://www.epa.gov/dioxin/scienceplan.
xxiv DRAFT - DO NOT CITE OR QUOTE
-------
For their review, the SAB held public meetings in June, July, and October 2010, and in
March and June 2011. They released their final report reviewing the draft Reanalysis on August
26, 2011 (SAB. 2011).3 In their report, the SAB communicated the following overarching
observations:
• They found that the draft Reanalysis was clear, logical, and responsive to many—but not
all—of the NAS recommendations; they were impressed with the comprehensive and
rigorous study selection process that was used to identify, review and evaluate the
scientific literature on TCDD dose response;
• They agreed with the choice of the Emond physiologically based pharmacokinetic
(PBPK) model for dose metric calculations and with the selection of whole blood as the
dose metric;
• They agreed with the choice of two epidemiologic studies as cocritical studies whose
developmental toxicity data were used to derive the RfD for TCDD;
• They agreed with EPA's evaluation of TCDD carcinogenicity (with the exception of
one panelist with a dissenting view);
The SAB also identified two deficiencies in EPA's draft Reanalysis with respect to the
completeness of the consideration of two critical elements:
• Nonlinear dose response for TCDD carcinogenicity; and
• Uncertainty analysis
The SAB recommended that EPA fully evaluate both linear and nonlinear dose-response
approaches to TCDD cancer dose-response assessment—including a discussion of carcinogenic
mode of action. The SAB also recommended a number of approaches to quantitative uncertainty
analysis that could be implemented by EPA, including the use of sensitivity analyses and
probability trees.
3 Available online at http://vosemite.epa.gov/sab/sabproduct.nsf/2A45B492EBAA8553852578F9003ECBC5/$File/
EPA-SAB-1 l-014-unsigned.pdf.
xxv DRAFT - DO NOT CITE OR QUOTE
-------
In August 2011, EPA announced a plan for moving forward to complete the draft
Reanalysis.4 This plan included the completion and posting to the IRIS database of the
noncancer portion of the draft Reanalysis separately followed soon thereafter by the completion
and posting to the IRIS database of the cancer portion of the draft Reanalysis. As such, this
current document is the first of two EPA reports (U.S. EPA 's Reanalysis of Key Issues Related to
Dioxin Toxicity and Response to NAS Comments Volumes 1 and 2 [Reanalysis Volumes 1 and
2]) that, together, will respond to the recommendations and comments on TCDD dose-response
assessment included in the NAS review of EPA's 2003 Reassessment. Both Volumes focus on
TCDD only. This report, Reanalysis Volume 1, completes and publishes EPA's study selection
criteria and study selection results for both noncancer and cancer TCDD dose-response
assessment; choice of kinetic model; noncancer RfD for TCDD; and a qualitative discussion of
uncertainties in the RfD with a focused quantitative uncertainty analysis. Reanalysis Volume 1
responds to key comments and recommendations pertaining to noncancer TCDD dose-response
assessment published by the NAS in their review (NAS. 2006b).
The information and these analyses have undergone revisions in response to SAB
comments and recommendations as well as comments provided by the public (see Appendix A).
Reanalysis Volume 2 will address the two deficiencies identified by the SAB, i.e., nonlinear dose
response for TCDD carcinogenicity and quantitative uncertainty analysis. In Volume 2, EPA
will complete the evaluation of cancer mode of action, cancer dose-response modeling, including
justification of the approaches used for dose-response modeling of the cancer endpoints, and an
associated quantitative uncertainty analysis. The information provided in Volume 1 will be used
in three ways: (1) as the first of two reports that contain EPA's response to the NAS (2006b)
report, (2) as the Support Document for the TCDD noncancer IRIS Summary and TCDD oral
RfD, and (3) as technical support for Reanalysis Volume 2.
The three key NAS recommendations specifically pertain to dose-response assessment
and uncertainty analysis. Therefore, EPA's response to the NAS in this document is focused on
these issues.
4 Available online at http://cfpub.epa. gov/ncea/cfm/recordisplav.cfm?deid=209690.
xxvi DRAFT - DO NOT CITE OR QUOTE
-------
EPA thoroughly considered the recommendations of the NAS and, in Reanalysis
Volume 1, responds with an evaluation of TCDD hazard identification and dose-response data
via the following:
• An updated literature search that identified new TCDD dose-response studies (see
Section 2);
• A workshop that included the participation of external experts in TCDD health effects,
toxicokinetics, dose-response assessment and quantitative uncertainty analysis; these
experts discussed potential approaches to TCDD dose-response assessment and
considerations for EPA's response to NAS (U.S. EPA. 2009b) (see Appendices B and J);
• Development of a detailed study selection process including criteria and considerations
for the selection of key epidemiologic and animal bioassay studies (see Section 2.3) for
quantitative TCDD dose-response assessment (see Section 2.4.1/Appendix C and Section
2.4.2/Appendix D, respectively);
• Kinetic modeling that quantifies appropriate dose metrics for use in TCDD dose-response
assessment (see Section 3 and Appendices E and F);
• A sensitivity analysis performed on each of the Emond animal and human PBPK models
that identify the most sensitive variables in each model (see Section 3.3.4);
• Dose-response modeling for all appropriate noncancer data sets (see
Section 4.2/Appendix G);
• A thorough and transparent evaluation of the selected TCDD data for use in the
derivation of an RfD, including justification of approaches used for dose-response
modeling of noncancer endpoints (see Section 4.2 and Appendix H);
• The development of an RfD (see Section 4.3);
• A qualitative discussion of the uncertainty in the RfD and a focused quantitative
uncertainty analyses of the RfD (see Sections 4.4 and 4.5, respectively); and
• Responses to the comments and recommendations made by the SAB in their final report
(SAB. 2011) (see Appendix A).
Those activities and analyses are briefly described in this Executive Summary, and they
are described in detail in the related sections of this document.
xxvii
DRAFT - DO NOT CITE OR QUOTE
-------
In addition to this document, several additional EPA activities address other TCDD
issues, specifically related to the application of dioxin TEFs and to TCDD and DLC background
exposure levels. Information on the application of the dioxin TEFs is published elsewhere by
EPA for both ecological (U.S. EPA. 2008b) and human health assessment (U.S. EPA. 2010b).
As a consequence, EPA does not directly address TEFs herein but makes use of the concept of
toxicity equivalence as applicable to the analysis of exposure dose uncertainty in epidemiologic
studies and an animal bioassay. Furthermore, this document does not address the NAS
recommendations pertaining to the assessment of human exposures to TCDD and other dioxins.
Information on updated background levels of dioxin in the U.S. population has been recently
reported (Lorber et al.. 2009). In 2006, EPA also released a report entitled An Inventory of
Sources and Environmental Releases of Dioxin-Like Compounds in the United States for the
Years 1987, 1995, and 2000, which presents an evaluation of sources and emissions of dioxins,
dibenzofurans, and coplanar polychlorinated biphenyls (PCBs) to the air, land and water of the
United States (U.S. EPA. 2006).
PRELIMINARY ACTIVITIES UNDERTAKEN BY EPA TO ENSURE THAT THE
REANLAYSIS VOLUMES 1 AND 2 REFLECTS THE CURRENT STATE-OF-THE-
SCIENCE
As part of the development of this document, EPA undertook two activities that involved
the public: an updated literature search and a scientific expert workshop. The adverse health
effects associated with TCDD exposures are documented extensively in epidemiologic and
toxicologic studies. As such, the database of relevant information pertaining to the
dose-response assessment of TCDD is vast and constantly expanding. Responding directly to the
NAS recommendation to use the most current and up-to-date scientific information related to
TCDD, EPA, in collaboration with the Department of Energy's Argonne National Laboratory
(ANL), developed an updated literature database of peer-reviewed studies on TCDD toxicity,
including in vivo mammalian dose-response studies and epidemiologic studies. An initial
literature search for studies published since the development of the 2003 Reassessment was
conducted to identify studies published between January 1, 2000 and October 31, 2008. EPA
published the initial literature search results in the Federal Register in November 2008 and
invited the public to review the list and submit additional, relevant, peer-reviewed studies.
Additional studies identified by the public and through continued work on this response were
xxviii DRAFT - DO NOT CITE OR QUOTE
-------
incorporated into the final set of studies for TCDD dose-response assessment (updated through
October 2009). Since release of the draft Reanalysis for public comment and external peer
review in 2010, EPA has collected a limited number of additional studies that inform EPA's
derivation of an RfD for TCDD. These studies were identified by EPA scientists, the SAB, and
the public, and they have been used to further evaluate the biological significance of the
endpoints used to derive the RfD and to develop information on uncertainty in the RfD. These
additional studies are cited in the appropriate sections of this document. No data sets collected
since October 2009 were used quantitatively in the noncancer dose-response assessment of
TCDD.
To assist in responding to the NAS, EPA, in collaboration with ANL, convened a
scientific expert workshop ("Dioxin Workshop") in February 2009 that was open to the public.
The primary goals of the Dioxin Workshop were to identify and address issues related to the
dose-response assessment of TCDD and to ensure that EPA's response to the NAS focused on
the key issues, while reflecting the most meaningful science. EPA and ANL assembled expert
scientists and asked them to identify and discuss the technical challenges involved in addressing
the NAS comments, discuss approaches for addressing these key recommendations, and to assist
in the identification of important published and peer-reviewed literature on TCDD. The
workshop was structured into seven scientific topic sessions as follows: (1) quantitative
dose-response modeling issues, (2) immunotoxicity, (3) neurotoxicity and nonreproductive
endocrine effects, (4) cardiovascular toxicity and hepatotoxicity, (5) cancer, (6) reproductive and
developmental toxicity, and (7) quantitative uncertainty analysis of dose response. External
cochairs (i.e., scientists who were not members of EPA or ANL) were asked to facilitate the
sessions and then prepare summaries of discussions occurring in each session. The session
summaries formed the basis of a final workshop report (U.S. EPA. 2009b) (Appendix B). Some
of the key outcomes from the workshop include the following recommendations:
• to further develop study selection criteria for evaluating the suitability of developing
dose-response models based on animal bioassays and human epidemiologic studies;
• to use kinetic modeling to identify relevant dose metrics and dose conversions between
test animal species and humans, and between human internal dose measures and human
intakes;
XXIX
DRAFT - DO NOT CITE OR QUOTE
-------
• to consider newer human or animal bioassay (NT P. 2006a) publications when evaluating
quantitative dose-response models for cancer;
• to consider both linear and nonlinear modeling in the cancer dose-response analysis.
The discussions held during the Dioxin Workshop helped inform, guide, and focus EPA's
response to NAS.
EPA'S APPROACH TO CONSIDERING TRANSPARENCY AND CLARITY IN THE
SELECTION OF KEY STUDIES AND DATA SETS FOR DOSE-RESPONSE
MODELING
One of the key NAS recommendations to EPA was to utilize a clear and transparent
process for the selection of key studies and data sets for dose-response assessment. EPA agrees
with the NAS and believes that clear delineation of the study selection process and decisions
regarding key studies and data sets will facilitate communication of critical decisions made in the
TCDD dose-response assessment. EPA developed detailed processes and TCDD-specific
criteria and considerations for the selection of key dose-response studies. These criteria and
considerations are based on current guidance for point of departure (POD) identification and RfD
and OSF derivation (U.S. EPA. 2005a. b, 2000. 1998. 1996. 1991. 1986a. b); they also consider
issues specifically related to TCDD. These criteria reflect EPA's goal of developing an RfD and
a cancer OSF for TCDD through a transparent study selection process. Following the selection
of key studies, EPA employed additional processes to further select and identify cancer and
noncancer data sets from these key studies for use in dose-response analysis of TCDD.
Figure ES-1 presents EPA's study selection process for the evaluation of the
epidemiologic studies considered for this TCDD dose-response assessment, including
specific study inclusion criteria (see Section 2.3.1). EPA applied its TCDD-specific
epidemiologic study inclusion criteria to all studies published on TCDD. For all peer
reviewed studies, EPA examined whether the exposures were primarily to TCDD and if
the TCDD exposures could be quantified so that dose-response analyses could be
conducted. Then, EPA required that the effective dose and oral exposure be estimable:
(1) for cancer, information is required on long-term exposures, (2) for noncancer,
information is required regarding the appropriate time window of exposure that is
relevant for a specific, nonfatal health endpoint, and (3) for all endpoints, information
XXX
DRAFT - DO NOT CITE OR QUOTE
-------
concerning the latency period between TCDD exposure and the onset of the effect is
needed. Finally, studies were evaluated using five considerations regarded as providing
the most relevant kind of information needed for quantitative human health risk analyses.
Only studies meeting these criteria and adequately satisfying the considerations were
included in EPA's TCDD dose-response analysis.
Figure ES-2 presents EPA's study selection process for the evaluation of
mammalian bioassays considered for TCDD dose-response assessment—including the
specific study inclusion criteria (see Section 2.3.2). EPA evaluated all available in vivo
mammalian bioassay studies on TCDD. Studies had to be published in the peer-reviewed
literature. Studies on genetically altered species were excluded as their direct relevance
to human health is not known. Next, EPA applied dose requirements to each study's
lowest tested average daily dose, with specific requirements for cancer (<1 (j,g/kg-day)
and noncancer (<30 ng/kg-day) studies. EPA also required that the animals were
exposed via the oral route to only TCDD. Finally, the studies were evaluated for quality
and summarized to ensure the most relevant information for quantitative analyses were
provided. Only studies meeting all of the criteria were included in EPA's TCDD dose-
response analysis.
Figure ES-3 shows the results of EPA's process to select and identify in vivo mammalian
bioassays and epidemiologic studies for quantitative TCDD dose-response assessment. A total
of 1,441 studies were examined. Of these, 637 studies were eliminated from consideration as
they were not suitable study types; these included, in vitro bioassays, review articles, PBPK
modeling studies, and studies that evaluated dioxin-like compounds (DLCs) other than TCDD.
Of the remaining studies, 49 were epidemiologic studies (7 studies contained both cancer and
noncancer endpoints), and 755 were animal bioassays (4 studies contained both cancer and
noncancer endpoints). These epidemiologic studies and animal bioassays were then evaluated
using EPA's study inclusion criteria. Appendices C and D detail EPA's study summaries and
evaluations for the epidemiologic studies and animal bioassays, respectively. Final results of the
study selection process for the epidemiologic studies are shown in Tables 2-1 and 2-2 (key
cancer and noncancer studies, respectively) and for the animal bioassays are shown in Tables 2-3
and 2-4 (cancer and noncancer studies, respectively). Through this study selection process, EPA
was able to identify a group of studies for TCDD dose-response evaluation that spanned across
xxxi DRAFT - DO NOT CITE OR QUOTE
-------
the types of adverse health effects associated with TCDD exposures and encompass the range of
doses in the lower end of the dose-response region most relevant to the development of an RfD.
For the selected studies, EPA conducted additional evaluations to determine
which study/endpoint data sets were the most appropriate for development of the RfD for
TCDD. During the study selection process, EPA identified four epidemiologic studies
and 78 animal bioassays that met the study inclusion criteria and adequately satisfied the
considerations for TCDD dose-response analyses. From the epidemiologice studies, one
was eliminated because EPA could not assess the biological significance of the finding
and could not establish a LOAEL; EPA derived three candidate RfDs from the other
studies. Figure ES-4 overviews the disposition of the 78 noncancer animal bioassays
selected for TCDD dose-response. Of these, EPA eliminated those studies that contained
no toxicologically relevant endpoints for RfD derivation (see Appendix H and Section
4.2.1). EPA then identified PODs from the remaining bioassays and eliminated from
further analysis those studies with PODs above specified dose limits. (See additional
details on POD development in the section below on Derivation of an RfD for TCDD.)
These dose limits were imposed to limit the size of the analysis yet ensure representation
of all important health effects associated with TCDD exposure. EPA derived 37
candidate RfDs from the remaining 48 animal studies, with 11 studies presented as
supporting information.
In summary, EPA conducted a transpartent study selection process to select
epidemiologic studies and animal bioassays for TCDD quantitative dose-response analyses.
From these selected studies, EPA identified 40 candidate RfDs, three from the epidemiologic
studies and 37 from the animal bioassays.
USE OF KINETIC MODELING TO ESTIMATE TCDD DOSES
NAS recommended that EPA utilize state-of-the-science approaches to finalize the
2003 Reassessment. Although NAS concurred with EPA's use of first-order body burden
models in the 2003 Reassessment, analyses of recent TCDD literature and comments by experts
at the Dioxin Workshop suggested that the understanding of TCDD kinetics had increased
significantly since the release of EPA's 2003 Reassessment. These advances led to the
development of several pharmacokinetic models for TCDD (Emond et al.. 2006; Avlward et al..
xxxii DRAFT - DO NOT CITE OR QUOTE
-------
2005a; Emond et al.. 2005; Emond et al.. 2004) and resulted in EPA's incorporation of TCDD
pharmacokinetics in the dose-response assessment of TCDD.
The evaluation of internal dose in exposed humans and other species is facilitated by an
understanding of pharmacokinetics (i.e., absorption, distribution, metabolism, and excretion).
TCDD pharmacokinetics are influenced by three distinctive features: (1) TCDD is highly
lipophilic, (2) TCDD is slowly metabolized, and (3) TCDD induces binding proteins in the liver.
The overall impact of these factors results in preferential storage of TCDD in adipose tissue, a
long half-life of TCDD in blood due to slow metabolism, and sequestration in liver tissue when
binding induction becomes significant. As these kinetic features control target tissue levels of
dioxin, they become important in relating toxicity in animals to possible effects in humans.
Consideration of pharmacokinetic mechanisms is critical to the selection of the dose
metrics of relevance to dose-response modeling of TCDD. Earlier assessments for TCDD—
including the 2003 Reassessment—used estimates of body burden as the dose metric for
extrapolation between animals and humans. These body burden calculations used a simple
one-compartment kinetic model based on the assumption of a first-order decrease in the levels of
administered dose as a function of time. However, the assumption of a constant half-life value
for the clearance of TCDD from long-term or chronic exposure is not well-supported
biologically given the dose-dependent elimination observed in rodents and humans. The
dynamic disposition and redistribution of TCDD between blood, fat, and liver as a function of
time and dose is better described using biologically-based models. Additionally, these models
provide estimates for other dose metrics (e.g., serum, whole blood, or tissue levels) that are more
biologically relevant to response than body burden estimated based on an assumption of
first-order elimination over time.
For extrapolation from rodents to humans, EPA considered the following possible dose
metrics for TCDD: administered dose, first-order body burden, lipid-adjusted serum
concentration (LASC), whole blood concentration, tissue concentration, and functional-related
metrics of relevance to the mode of action (MOA) (e.g., receptor occupancy) (see
Section 3.3.4.1). After evaluation of these dose metrics, EPA chose to use TCDD concentration
in whole blood, modeled as a function of administered dose, as the dose metric for assessing
TCDD dose response in this document. LASC is commonly used in the epidemiologic literature
as the metric of choice because TCDD is highly lipid-soluble and LASC accounts for individual
XXXlll
DRAFT - DO NOT CITE OR QUOTE
-------
differences in the size of the serum lipid compartment. However, whole blood concentration was
chosen because of the structure of the Emond PBPK model, in which the liver and other tissue
compartments are connected to the whole blood compartment rather than to the serum
compartment; LASC is estimated only at the end of the model simulations by multiplying whole-
blood concentrations by a constant. EPA used the time-weighted average whole-blood
concentration over the relevant exposure periods for all animal bioassay dosing protocols,
dividing the area under the time-course concentration curve (AUC) by the exposure duration.
Because all of the epidemiologic studies evaluated by EPA reported TCDD exposures as LASC
rather than whole-blood concentrations, oral intakes were modeled using LASC as the dose
metric. In most cases, the reported TCDD LASC was extrapolated both forward and backward
in time to simulate the actual exposure scenario.5
Several biologically-based kinetic models for TCDD exist in the literature. The more
recent pharmacokinetic models explicitly characterize the concentration-dependent elimination
of TCDD (Emond et al.. 2006; Aylward et al.. 2005a; Emond et al.. 2005; Emond et al.. 2004;
Carrier et al.. 1995a. b). The biologically based pharmacokinetic models describing the
concentration-dependent elimination (i.e., the pharmacokinetic models of Aylward et al. (2005a)
and Emond et al. (2006; 2005) are relevant for application to simulate the TCDD dose metrics in
humans and animals exposed via the oral route. The rationale for considering the application of
the Aylward et al. (2005a) and Emond et al. (2006; 2005; 2004) models was largely based on the
fact that both models reflect research results from recent peer-reviewed publications, and both
models are formulated with dose-dependent hepatic elimination consistent with the physiological
understanding of TCDD kinetics. Dose-response modeling based on body burden of TCDD in
adult animals and humans can be conducted with either of the models, provided the duration of
the experiment is at least 1 month, due to limitations in the Aylward et al. (2005a) model. The
predicted slope and body burden over a large dose range are quite comparable between the
two models (generally within a factor of two).
Results of simulations of serum lipid concentrations or liver concentrations vary for the
two models to a larger extent (up to a factor of 7), particularly for simulations of short duration.
5 For the Seveso cohort, which had a high single TCDD exposure followed by low-level background exposures
leading to a gradual decline in the internal TCDD concentrations, EPA estimated both peak and average exposures
over a defined critical exposure window (see Section 4.2.2).
xxxiv DRAFT - DO NOT CITE OR QUOTE
-------
These differences reflect two characteristics of the Emond et al. (2006) model: first,
quasi-steady-state is not assumed in the Emond et al. (2006) model; second, the serum lipid
composition used in the model is not the same as the adipose tissue lipids. The Aylward et al.
(2005a) model does not account for differential solubility of TCDD in serum lipids and adipose
tissue lipids, nor does it account for the diffusion-limited uptake by adipose tissue. Based on this
evaluation, EPA determined that the Emond et al. (2006) provided more applicability than the
Aylward et al. (2005a) model with respect to the ability to simulate serum lipid and tissue
concentrations during exposures that do not lead to the onset of steady-state condition in the
exposed organism. Additionally, of the two selected models, the pharmacokinetic model
developed by Emond et al. (2006) is more physiologically-based, as compared to the Aylward
et al. (2005a) model, and models the blood compartment directly in the rat, mouse, and human;
there are also gestational and life-time nongestational forms of the Emond et al. (2006) model.
In this document, EPA chose the Emond rodent physiologically-based pharmacokinetic (PBPK)
model to estimate blood TCDD concentrations based on administered doses (see Section 3.3.4,
Appendix E).
To enhance the biological basis of the PBPK model of Emond et al. (2006). three minor
modifications were made before its use in the computation of dose metrics for TCDD:
(1) recalculation of the volume of the "rest of the body compartment" after accounting for
volume of the liver and fat compartments; (2) calculation of the rate of TCDD excreted via urine
by multiplying the urinary clearance parameter by blood concentration in the equation instead of
by the concentration in the rest of the body compartment; and (3) recalibration for the human
gastric nonabsorption constant to yield observed oral bioavailability of TCDD (Poiger and
Schlatter. 1986) (see Section 3.3.4.4 for details). The modified PBPK model was evaluated
against all published data used in the original model. EPA assumed that the same blood TCDD
levels that led to effects in animals would also lead to effects in humans; therefore, the Emond
human PBPK model was used to estimate the lifetime average daily oral doses (consistent with
the chronic RfD) that would correspond to the blood TCDD concentrations estimated to have
occurred during the animal bioassays. EPA used the same Emond human PBPK model to
estimate the lifetime average daily doses that would correspond to the TCDD blood or tissue
concentrations reported in the epidemiological studies (see Appendix F). These estimates are the
Human Equivalent Doses (HEDs) that are used to develop candidate RfDs for TCDD.
xxxv
DRAFT - DO NOT CITE OR QUOTE
-------
A sensitivity analysis was performed on each of the animal and human Emond PBPK
models to determine the most sensitive variables (see Section 3.3.4.3.2.5). In each case, all input
variables in each model were included in the analysis; the sensitivity analysis was conducted by
varying each parameter one at a time. For the rat and mouse nongestational models and rat and
mouse gestational models for the low and high doses when variables were increased by +5%,
predicted TCDD blood concentrations were very sensitive to the Hill coefficient (h in Eq. 3-20,
Section 3.3.4.3.2.2). Other sensitive PBPK model variables are associated with the overall
dioxin elimination/sequestration rate, including the CYP1A2 induction rates, the liver weight, the
binding capacity and affinity, and the gastric and intestinal excretion rates. For the gestational
model dosing protocols, the Hill coefficient remains the most sensitive variable but the elasticity
decreases compared with the nongestational analysis. Otherwise, many of the most sensitive
variables remain those associated with elimination. Additional parameters related to the adipose
tissue blood flow and with the adipose diffusional permeability fraction are also relatively
sensitive. For the human gestational and nongestational models, additional variables associated
with the adipose compartment partition coefficient, the body weight, and the fractional adipose
tissue volume are also relatively sensitive variables at the RfD and POD dose compared with the
animal models. For all models, the elasticities are relatively similar across the different doses
evaluated.
For variables which are optimized, a sensitivity analysis which varies each parameter one
at a time may overestimate the model uncertainty associated with the variable. In this analysis,
the most sensitive variable in all the models is the Hill parameter. The elasticity is high in part
because the Hill parameter is an exponent; thus, small changes in the value can lead to larger
changes in the whole blood concentration. The Hill coefficient (as it is used in the PBPK
models) can only be estimated with high confidence when optimized against in vivo hepatic
CYP1A2 induction data in response to TCDD exposure. This type of data exists in animal
experiments only. When this coefficient is optimized against human blood levels of TCDD, it is
influenced by other parameters describing the dose-dependent elimination mechanism of the
chemical; these data cannot be evaluated in vivo in humans.
This analysis highlights several important research needs. While the disposition of
TCDD following high exposures is reasonably understood and simulated in current models, the
current scientific understanding of disposition following TCDD exposures near current
xxxvi
DRAFT - DO NOT CITE OR QUOTE
-------
background dietary intakes, likely the primary source of TCDD exposure for most of the U.S.
population, are not understood as well at present. This uncertainty affects the estimation of
TCDD intake rates corresponding to the lower blood TCDD levels associated with LOAELs and
NOAELs. The disposition of DLCs following exposures at background levels is similarly not
well understood.
DERIVATION OF AN RfD FOR TCDD
The NAS specifically recommended that EPA derive an RfD for TCDD. Through a
transparent study selection process, EPA identified key studies from both epidemiologic studies
and animal bioassays. EPA then identified PODs for RfD derivation from those key human
epidemiologic studies and animal bioassays. Figure ES-5 (exposure-response array) shows the
PODs for TCDD graphically in terms of human-equivalent intake (ng/kg-day). The human study
endpoints are shown at the far left of the figure and, to the right, the rodent endpoints are
arranged by the following study categories: less than 1 year, greater than 1 year, reproductive,
and developmental.
For each noncancer epidemiologic study that EPA selected as key, EPA evaluated the
dose-response information developed by the study authors to determine whether the study
provided noncancer effects and TCDD-relevant exposure data for a toxicologically-relevant
endpoint. If such data were available, EPA identified a NOAEL or LOAEL as a POD. Then,
EPA used the Emond human PBPK model to estimate the continuous oral daily intake
(ng/kg-day) that would lead to the relevant blood TCDD concentrations associated with the
POD. If all of this information was available, then the result was included as a POD.
Through this process, EPA identified adverse health effects from the following
four epidemiologic studies to be considered as the basis for the RfD: Eskenazi et al. (2002b)
(menstrual cycle effects) Alaluusua et al. (2004) (developmental—tooth development), Mocarelli
et al. (2008) (reproductive—decreased sperm concentrations and motility [semen quality]), and
Baccarelli et al. (2008) (developmental—increased thyroid-stimulating hormone levels in
neonates [neonatal TSH]). All four studies are from the Seveso cohort, whose members were
exposed environmentally to high peak concentrations of TCDD as a consequence of an industrial
accident. For each of the menstrual cycle, tooth development, and semen quality endpoints, EPA
calculated a POD for derivation of a candidate RfD by estimating dose as the mean of the peak
xxxvii
DRAFT - DO NOT CITE OR QUOTE
-------
exposure (following the accident) and the average exposure over a defined critical exposure
window for that endpoint. For neonatal TSH, EPA calculated the POD from estimates of
maternal exposure during pregnancy reported by the study authors (Baccarelli et al., (2008) (see
Section 4.2.3). The PODs estimated for both menstrual cycle and tooth development were well
above those estimated for semen quality and neonatal TSH.
Figures ES-4 and ES-6 together present the strategy EPA used to evaluate the
study/endpoint combinations found in the animal bioassays that met EPA's study inclusion
criteria, estimate PODs, and develop a final set of candidate RfDs for TCDD. Figure ES-4
overviews the disposition of the 78 animal noncancer studies selected for TCDD dose-response
analyses. Of these studies, 16 were eliminated because EPA determined that they contained no
toxicologically relevant endpoints that could be used to derive a candidate RfD (see Appendix H
and Section 4.2.1). EPA then identified PODs from the remaining bioassays; at this point,
Figure ES-4 refers to Figure ES-6, which is a flow chart of the iterative process used to estimate
PODs and compare them within and across studies to arrive at a final set of PODs from these
bioassays (see additional details below). From this final set of PODs, Figure ES-4 shows that
EPA then eliminated 13 studies from further analysis with both a human equivalent dose (HED)
LOAELhed >1 ng/kg-day and a NOAELhed/BMDLhed >0.32 ng/kg-day (see Table 4-3); one
additional study was also not carried forward because of the lack of toxicokinetic information for
estimation of an HED.
Figure ES-6 summarizes the strategy employed for identifying and estimating PODs from
the 62 animal bioassays with at least one toxicologically relevant endpoint for RfD derivation.
For the noncancer endpoints within these studies, EPA first evaluated the toxicological relevance
of each endpoint, rejecting those judged not to be relevant for RfD derivation. Next, initial
PODs based on the first-order body burden metric (see Section 3.3.4.2) and expressed as HEDs
(i.e., NOAELhed, LOAELhed, BMDLhed) were determined for all relevant endpoints
(summarized in Table 4-3). Because there were very few NOAELs and BMDL modeling was
largely unsuccessful due to data limitations (see Section 4.2), the next stage of evaluation was
carried out using LOAELs only. Within each study, effects not observed at the LOAEL (i.e.,
reported at higher doses) with BMDLhedS greater than the LOAELhed were eliminated from
further analysis, as they would not be considered as candidates for the final POD on either a
BMDL or NOAEL/LOAEL basis (i.e., the POD would be higher than the PODs of other relevant
xxxviii
DRAFT - DO NOT CITE OR QUOTE
-------
endpoints). In addition, all endpoints with LOAELhed estimates beyond a 100-fold range of the
lowest identified LOAELhed across all studies were (temporarily) eliminated from further
consideration, as they would not be POD candidates either (i.e., the POD would be higher than
the PODs of other relevant endpoints). For the remaining endpoints, EPA then determined final
potential PODs based on TCDD whole-blood concentrations obtained from the Emond rodent
PBPK models. HEDs were then estimated for each of these PODs using the Emond human
PBPK model. At this point, if the PBPK modeling results suggested considering additional
endpoints at higher doses, the process was repeated. From the final set of HEDs, a POD was
selected6 for each study, to which appropriate uncertainty factors (UFs) were applied following
EPA guidance (see Section 4.3.3 following). The resulting candidate RfDs were then considered
in the final selection process for the RfD. Other endpoints occurring at slightly higher doses
representing additional effects associated with TCDD exposure (beyond the 100-fold LOAELhed
n
range) were evaluated, modeled, and included in the final candidate RfD array to examine
endpoints not evaluated by studies with lower PODs. In addition, Benchmark Dose (BMD)
modeling based on administered dose was performed on all endpoints for comparison purposes.
For BMD modeling, EPA used a 10% BMR for dichotomous data for all endpoints; no
developmental studies were identified with designs that incorporate litter effects, for which a
5% BMR would be used (U.S. EPA. 2000). For continuous endpoints in this document, EPA
used a BMR of 1 standard deviation from the control mean whenever a specific
toxicologically-relevant BMR could not be defined. Importantly, the 2003 Reassessment defined
the EDoi as 1% of the maximal response for a given endpoint, not as a 1% change from control.
Because RfD derivation is one goal of this document, the noncancer modeling effort undertaken
here differs substantially from the modeling in the 2003 Reassessment. Evaluation of BMD
modeling performance, goodness-of-fit, dose-response data, and resulting BMD and BMDL
estimates included statistical criteria as well as expert judgment of their statistical and
toxicological properties. EPA has reported and evaluated the BMD results using the standard
6 In the standard order of consideration: BMDL, NOAEL, and LOAEL.
7 However, studies with a lowest dose tested greater than 30 ng/kg-day were not included in the expanded
evaluation.
xxxix DRAFT - DO NOT CITE OR QUOTE
-------
suite of goodness-of-fit measures from the benchmark dose modeling software (BMDS 2.1).
(See Appendix H and Section 4.2 for more information on the BMD modeling criteria and
results.)
For selection of the POD to serve as the basis of the RfD, EPA gave the epidemiologic
studies the highest consideration because human data are preferred in the derivation of an RfD,
given that the underlying epidemiologic and animal bioassay data are of comparable quality.
This preference for epidemiologic study data also is consistent with recommendations of
panelists at the Dioxin Workshop (U.S. EPA. 2009b) (Appendix B). Figure ES-7 arrays the
candidate RfDs from both the human and animal bioassays in units of human-equivalent intake
(mg/kg-day). The human studies included in Figure ES-7 (Baccarelli et al.. 2008; Mocarelli et
al.. 2008; Alaluusua et al.. 2004) each evaluate a segment of the Seveso civilian population (i.e.,
not an occupational cohort) exposed directly to TCDD released from an industrial accident. EPA
designated the (Baccarelli et al.. 2008; Mocarelli et al.. 2008; Alaluusua et al.. 2004) studies as
co-principal in deriving the RfD (see Section 4.3). In the Seveso cohort, exposures were
primarily to TCDD, the chemical of concern, with apparently minimal DLC exposures beyond
those associated with background intake, qualifying these studies for use in the RfD derivation
for TCDD. In addition, by using PODs derived from human data, the uncertainty of interspecies
extrapolation is eliminated. The study subjects included infants (exposed in utero) and adults
who were exposed when they were less than 10 years of age, identifying effects in potentially
vulnerable lifestages, accounting for at least some part of the uncertainty in extrapolation of
effect levels to sensitive human populations and lifestages.
For Baccarelli et al. (2008). EPA defined the LOAEL (in LASC terms) as the maternal
TCDD LASC of 235 ppt corresponding to a neonatal TSH level of 5 |iU/mL, determined by the
regression modeling performed by the study authors. The World Health Organization (1994)
established the 5 |iU/mL standard as a benchmark indicator for medical follow-up for
investigation of potential congenital hypo-thyroidism. This benchmark was intended to address
potential iodine deficiencies, but it is equally applicable to TCDD exposure for evaluating the
equivalent effect. Baccarelli et al. (2008) discounted iodine status in the population as a
confounder. For TCDD, the toxicological concern is not likely to be iodine uptake inhibition,
but rather increased metabolism and clearance of the thyroid hormone, thyroxine (T4). An
increased TSH level is an indicator of a potential decrease in circulating T4 levels, which could
xl
DRAFT - DO NOT CITE OR QUOTE
-------
eventually lead to neurological deficiencies. TCDD has been associated with reductions in T4 in
o
a number of animal studies as discussed in Section 4.3.6.1. Adequate levels of thyroid hormone
are essential in the newborn and young infant as this is a period of active brain development
(Zoeller and Rovet. 2004; Glinoer and Delange. 2000). Thyroid hormone disruption during
pregnancy and in the neonatal period can lead to neurological deficiencies.
Baccarelli et al. (2008) did not provide oral intakes associated with TCDD serum
concentrations. EPA estimated the maternal TCDD intake corresponding to the LASC LOAEL
of 235 ppt (at delivery) by use of the Emond human PBPK model the continuous daily intake
from birth to age 30, the average age of the maternal cohort at delivery, that resulted in a 235 ppt
maternal LASC at delivery. The resulting modeled maternal daily intake rate of 0.020 ng/kg-day
established the LOAEL POD for the RfD. EPA did not define a NOAEL because it is not clear
what maternal intake should be assigned to the group below 5 |iU/mL.
For Mocarelli et al. (2008). EPA defined the LOAEL as the lowest exposed group
(lst-quartile) mean TCDD LASC of 68 ppt, corresponding to decreased sperm concentrations
(20%) and decreased motile sperm counts (11%) in men who were 1-9 years old at the time of
the Seveso accident (initial TCDD exposure event). There is no clear adverse effect level
indicating male fertility problems for either of these sperm effects. As sperm concentration
decreases, the probability of pregnancy from a single ejaculation also decreases; infertile
conditions arise when the number of normal sperm per ejaculate is consistently and sufficiently
low. Previously, the incidence of male infertility was considered increased at sperm
concentrations less than 20 million sperm/mL (WHO. 1980). More recently, Cooper et al. (2010)
suggested that the 5th percentile for sperm concentration (15 million/mL) could be used as a limit
by clinicians to indicate needed follow-up for potential infertility. Skakkeback (2010) suggests
the following two limits for human sperm concentrations: 15 million sperm/mL, based on
Cooper et al. (2010) and 40 million sperm/mL. Skakkeback justifies the upper level of 40
million sperm/mL citing a study by Bonde et al. (1998) of couples planning to become pregnant
for the first time; in the Bonde study, pregnancy rates declined when sperm concentrations were
below 40 million sperm/mL. Skakkeback suggests that 15 million sperm/mL may be too low of
a cut off for normal fertility and that sperm concentrations between 15 million sperm/mL and
8Sewall et al. (19951. Seo et al. (19951. Van Birgelen et al. (1995a: 1995b). Crofton et al. (20051. and NTP (2006a').
xli DRAFT - DO NOT CITE OR QUOTE
-------
40 million sperm/mL may indicate a range of reduced fertility. For fertile men, between 50%
and 60% of sperm are motile (Swan et al.. 2003; Slama et al.. 2002; Wiichman et al.. 2001). Any
impacts on these reported levels could become functionally significant, leading to reduced
fertility. Low sperm counts are typically accompanied by poor sperm quality with respect to
morphology and motility (Slama et al.. 2002).
EPA judged that the impact on sperm concentration and quality reported by Mocarelli
et al. (2008) is biologically significant given the potential for functional impairment. Although a
decrease in sperm concentration of 25% likely would not have clinical significance for a typical
individual, EPA's concern with the reported decreases in sperm concentration and total number
of motile sperm (relative to the comparison group) is that such decreases associated with TCDD
exposures could lead to shifts in the distributions of these measures in the general population.
Because male fertility is susceptible to reductions in both the number and quality of sperm
produced, such shifts in the population could result in decreased fertility in men at the low ends
of these population distributions. Further, in the group exposed due to the Seveso accident,
individuals 1 standard deviation below the mean had sperm concentrations of 21.8 million/mL;
this concentration falls at the low end of the range of reduced fertility (15 million and
40 million sperm/mL) suggested by Skakkebaek (2010).
For Mocarelli et al. (2008). TCDD LASC levels were measured within approximately
1 year of the initial exposure event. Because effects were only observed in men who were under
10 years of age at the time of exposure, EPA assumed a maximum 10-year critical exposure
window for elicitation of these effects. Using the Emond human PBPK model, EPA has
estimated a continuous daily oral intake of 0.020 ng/kg-day associated with the (LASC) LOAEL
of 68 ppt (see Section 4.2.3.2). The reference group is not designated as aNOAEL because there
is no clear zero-exposure measurement for any of these endpoints, particularly considering the
contribution of background exposure to DLCs, which further complicates the interpretation of
the reference group response as a true "control" response (see discussion in Section 4.4).
However, males less than 10 years old can be designated as a being in a sensitive lifestage as
compared to older males who were not affected.
The two PODs based on the Baccarelli et al. (2008) and Mocarelli et al. (2008) studies,
are adjusted LOAELs with the same value of 0.020 ng/kg-day, providing mutual quantitative
support. Because these two studies define the most sensitive endpoints evaluated in the
xlii
DRAFT - DO NOT CITE OR QUOTE
-------
epidemiologic literature, they are designated as co-principal studies for the RfD. Increased TSH
in neonates (Baccarelli et al.. 2008) and male reproductive effects (decreased sperm count and
motility) (Mocarelli et al.. 2008) are designated as cocritical effects. The adjusted LOAEL of
0.020 ng/kg-day is designated as the POD for the RfD. EPA used a composite UF of 30 for the
RfD. A factor of 10 for UFL was applied to account for lack of a NOAEL. A factor of 3 (10°5)
for UFh was applied to account for human interindividual variability because the effects were
elicited in sensitive lifestages. A UF of 1 was not applied because the sample sizes in these
two epidemiologic studies were relatively small, which, combined with uncertainty in exposure
estimation, may not fully capture the range of interindividual variability. In addition, potential
chronic effects are not well defined for humans and could possibly more sensitive. The resulting
RfD for TCDD in standard units is 7 x 10 10 mg/kg-day.
Although the human data are preferred, Figure ES-7 presents a number of candidate RfDs
derived from animal bioassays that are lower than the human RfDs. Two of the rat bioassays
among this group of studies—Bell et al. (2007b) and NTP (2006a)—are of particular note. Both
studies were recently conducted and very well designed and conducted, using 30 or more
animals per dose group; both also are consistent with and, in part, have helped to define the
current state of practice in the field of toxicology. Bell et al. (2007b) evaluated several
reproductive and developmental endpoints, initiating TCDD exposures well before mating and
continuing through gestation. NTP (2006a) is the most comprehensive evaluation of TCDD
chronic toxicity in rodents to date, evaluating dozens of endpoints at several time points in all
major tissues. Thus, proximity of the candidate RfDs derived from these two high quality, recent
studies, provide additional support for the RfD derived from the two principal epidemiologic
studies.
EPA also developed cross-species comparison tables and figures of selected toxicological
endpoints for all the animal and human studies that met the EPA selection criteria (see
Appendix D.3). The endpoints include male and female reproductive effects, thyroid hormone
levels and developmental dental effects, all of which have been reported for humans. In
addition, immunological and neurological effects are shown because they are sensitive effects in
experimental animal studies, although not evident in humans. The analysis presented in
Appendix D.3 supports the conclusion that there is a substantial amount of qualitative
concordance of effects between rodents and humans, but a much lower quantitative concordance.
xliii
DRAFT - DO NOT CITE OR QUOTE
-------
There are several animal bioassay candidate RfDs at the lower end of the RfD range in
Figure ES-7 that are more than 10-fold below the human-based RfDs. Two of these studies
report effects that are analogous to the endpoints reported in the three human studies and support
the RfDs based on human data. Specifically, decreased sperm production in Latchoumydandane
and Mathur (2002) is consistent with the decreased sperm counts and other sperm effects in
Mocarelli et al. (2008). and missing molars in Keller et al. (2008a; 2008b; 2007) are similar to
the dental defects seen in Alaluusua et al. (2004). Thus, because these endpoints have been
associated with TCDD exposures in humans, these animal studies would not be selected for RfD
derivation in preference to human data showing similar effects.
Another characteristic of the remaining studies in the lower end of the candidate RfD
distribution is that they are dominated by mouse studies (comprising 7 of the 9 lowest
rodent-based RfDs). EPA has less confidence in the candidate RfD estimates based on mouse
data than either the rat or human candidate RfD estimates. EPA has less confidence in the
Emond mouse PBPK model than the other Emond PBPK models used to estimate the PODs
because of the lack of key mouse-specific data, particularly for the gestational component (see
Section 3.3.4.3.2.5). The LOAELhfds identified in mouse bioassays are low primarily because
of the large toxicokinetic interspecies extrapolation factors used for mice, for which there is
more potential for error. In addition, each one of the mouse studies has other qualitative
limitations and uncertainties that make them less desirable candidates as the basis for the RfD
than the human studies.
EPA conducted additional sensitivity analyses of two groups of studies. Using variable
sensitivity trees, EPA further analyzed the impacts of some sources of uncertainty encountered in
the development of candidate RfDs based on Baccarelli et al. (2008). Mocarelli et al. (2008) and
NTP (2006a). specifically examining the sensitivity of the POD value to choices made for
estimating possible contributions associated with exposures to DLCs, exposure uncertainties and
PBPK model variables and inputs (see Section 4.5.1). In Section 4.5.2, EPA also evaluated a
number of endpoints presented in seven other Seveso cohort studies to estimate the range of
potential PODs based on uncertainties in exposure duration, exposure averaging protocols and
DLC background exposures. Included among those seven study/endpoint combinations are
two studies that satisfied all the study selection criteria and considerations—developmental
dental effects (Alaluusua et al.. 2004) and duration of menstrual period (Eskenazi et al..
xliv
DRAFT - DO NOT CITE OR QUOTE
-------
2002b)—a new developmental study on semen quality (Mocarelli et al.. 2011) that was
published after the study selection process was completed but is useful in this uncertainty
analysis of the POD ranges, and four studies that did not satisfy all the study inclusion criteria
and considerations.9
Overall, the results of these sensitivity analyses increase the confidence in the TCDD
RfD—both qualitatively and quantitatively. EPA's sensitivity analyses show some POD
estimates that are higher than the POD used to derive the RfD (e.g., those PODs that consider
background DLCs), while other analyses show POD estimates lower than the POD used to
derive the RfD. These sensitivity analyses also highlight several important research needs. They
highlight that the current scientific understanding of disposition following TCDD exposures that
are closer to current background dietary intakes are not understood as well as the disposition of
high TCDD exposures at present. There is also toxicological uncertainty regarding several of the
endpoints; additional studies corroborating these outcomes and their toxicological significance
would further increase their utility in refining the TCDD RfD.
9 Mocarelli (20001. Eskenazi et al. (20051. and Warner et al. (2007: 2004). See Appendix C for study descriptions.
xlv DRAFT - DO NOT CITE OR QUOTE
-------
No
Study
in peer-reviewed
_ literature?
Yes
No
Yes
Long-term \
exposures and
latency information
available forcancer
X^assessment?/^
Exposure
windows and
latency information
available for RfD
^xassessment'?/
No
No
Yes
Yes
No
Yes
Exposure
primarily to TCDD
and quantifiable?
Considerations
adequately
^ satisfied? .
Study excluded
from TCDD
dose-response
assessment
List of available epidemiologic studies on TCDD and DLCs
(All studies summarized.)
Key study included
forTCDD cancerand/or noncancer
dose-response assessment
Evaluate study using five considerations:
• Methods used to ascertain health outcomes are clear and unbiased?
• Confounding and other potential sources of bias are addressed?
• Association/exposure response between TCDD and adverse effect?
• Exposures based on individual-level estimates, uncertainties described?
• Statistical precision, powerand study follow-up are sufficient?
Figure ES-1. EPA's selection process to evaluate available epidemiologic
studies using study inclusion criteria and other epidemiologic considerations
for use in the dose-response analysis of TCDD.
EPA applied its TCDD-specific epidemiologic study inclusion criteria to all studies published on
TCDD and DLCs. For all peer reviewed studies, EPA examined whether the exposures were
primarily to TCDD and if the TCDD exposures could be quantified so that dose-response analyses
could be conducted. Then, EPA required that the effective dose and oral exposure be estimable:
(1) for cancer, information is required on long-term exposures, (2) for noncancer, information is
required regarding the appropriate time window of exposure that is relevant for a specific, nonfatal
health endpoint, and (3) for all endpoints, the latency period between TCDD exposure and the
onset of the health endpoint is needed. Finally, studies were evaluated using five considerations
regarded as providing the most relevant kind of information needed for quantitative human health
risk analyses. Only studies meeting these criteria and adequately satisfying the considerations
were included in EPA's TCDD dose-response analysis.
xlvi
DRAFT - DO NOT CITE OR QUOTE
-------
Study
in peer-reviewed
literature?
No
Yes
Study on a
genetically-altered
. species? .
Yes
No
No
No
Yes
Yes
No
Yes
^ Lowest dose
tested for n on cancer
endpoint<30
\ng/kg-day?
x Lowest
dose tested for
cancer endpoint<1
s. [jg/kg-day? .
^ Oral ^
exposure to TCDD
only?
Study excluded
from TCDD
dose-response
assessment
List of available in vivo mammalian bioassay studies on TCDD
Study summarized; evaluated for
quality andto note adequacy
of data needed for TCDD
dose-response assessment.
Key study included
forTCDD cancerand/or noncancer
dose-response assessment
Figure ES-2. EPA's process to evaluate available animal bioassay studies using
study inclusion criteria for use in the dose-response analysis of TCDD.
EPA evaluated all available in vivo mammalian bioassay studies on TCDD. Studies had to be published in
the peer-reviewed literature. Studies on genetically-altered species were excluded as their relevance to
human health is not known. Next, EPA applied dose requirements to each study's lowest tested average
daily dose, with requirements for cancer (<1 (ig/kg-day) and noncancer (<30 ng/kg-day) studies. EPA also
required that the animals were exposed via the oral route to only TCDD. Finally, the studies were
evaluated for quality and summarized to ensure providing the most relevant information for quantitative
human health risk analyses. Only studies meeting all of the criteria were included in EPA's TCDD
dose-response analysis.
xlvii
DRAFT - DO NOT CITE OR QUOTE
-------
Noncancer
bioassays
751
Epi cancer
studies
24
Epi noncancer
studies
32
Cancer
bioassays
Animal
noncancer
bioassays
included
Animal
cancer
bioassays
Included
noncancer
studies
included
cancer
studies
Included
Epidemiologic (Epi) studies
49
Animal bioassays
755
Right study type for quantitative
TCDD dose-response analysis:
804 considered further
Wrong study type for quantitative
TCDD dose-response analysis:
637 excluded
Studies from literature search and data collection activities
1,441
TDCC only (142)
Genetically-
altered (66)
Dose cutoffs
(370)
Failed > 1 of:
Peer-review (0)
Non-oral (135)
TDCC only (0)
Genetically-
altered (1)
Failed > 1 of:
Peer-review (0)
Dose cutoffs
Non-oral (1)
Primarily TDCC
(10)
Effective
exposure
estimable (11
Considerations'
Failed > 1 of:
Peer-review (0)
Effective
exposure
estimable (26)
Considerations'
Failed > 1 of:
Peer-review (1)
Primarily TDCC
Indicates those studies that passed all three criteria but were not selected based on
study considerations.
Figure ES-3. Results of EPA's process to select and identify in vivo
mammalian and epidemiologic studies for use in the dose-response analysis
of TCDD.
Criteria not met are not mutually exclusive. Four animal studies and seven epidemiologic studies
contained both cancer and noncancer endpoints. Two epidemiologic cancer studies, Steenland
et al. (1999) and Flesch-Janys et al. (1998). passed all criteria, but were still not selected because
they were superseded by other studies on the same cohort for which an improved analysis was
done. One noncancer epidemiologic study, Baccarelli et al. (2005). passed all criteria, but was
excluded because the health endpoint, chloracne, is considered to be an outcome associated with
high TCDD exposures.
xlviii
DRAFT - DO NOT CITE OR QUOTE
-------
NoncancerAnimal Bioassays Selected for
TCDD Dose-Response Assessment (See Tables 2-4 and D-1
78 Studies
Identify and Estimate PODs from the 62 Remaining Animal Bioassays
for use in Non cancer Dose-Response Analysis of TCDD
(See Figure ES-6)
Derive Candidate RfDs from the
48 Remaining Noncancer Animal Bioassays
Final Candidate RfDs from Noncancer Animal Bioassays
(11 Studies Presented as Supporting Information;
See Table 4-5)
37 Candidate RfDs
Burleson etal. (1996)
Hassoun etal. (1998)
Hassoun etal. (2002)
Hong etal. (1989)
Latchoumycandane etal. (2003)
MallyandChipman (2002)
Slezaket al. (2000)
Tritscher et al. (1992)
Eliminate Studies with NoToxicologically Relevant Endpoints for RfD Derivation
(See Appendix H and Section 4.2.1)
16 Studies Eliminated
DeVitoetal. (1994)
Hassoun etal. (2000)
Hassoun etal. (2003)
Kitchin and Woods(1979)
Lucieret al. (1986)
Sewall etal. (1993)
Sugita-Konishi etal. (2003)
Vanden Heuvel etal. (1994)
Eliminate Studies with Both a
LOAELHed>1 ng/kg-d and a NOAELhed/BMDLhed5- 0.32 ng/kg-d* (See Table 4.3)
14 Studies Eliminated
Croutch etal. (2005)
Ikeda et al. (2005)
Maronpotet al. (1993) Noharaetal. (2000,2002)
Simanainen etal. (2002, 2003, 2004a) Smialowiczet. al. (2004)
Smith etal. (1976)
Chu etal. (2001)
Fox etal. (1993)
*Hochstein etal. (2001) is also not carried forward because of the
lackof toxicokinetic information for estimation of an HED
Weber et al. (1995)
Figure ES-4. Disposition of animal noncancer bioassays selected for TCDD
dose-response analysis.
EPA evaluated each noncancer endpoint found in the 78 studies that passed the study inclusion
criteria. From this evaluation, EPA eliminated 16 studies that contained no toxicologically
relevant endpoints for RfD derivation. Then as detailed in Figure 4-3, EPA selected and
identified PODs for use in deriving candidate RfDs. EPA then eliminated 13 studies based on
dose limits for the PODs' HEDs; one study was also not carried forward because of the lack of
toxicokinetic information for estimation of an HED. Of the remaining 48 studies, EPA derived 37
RfD candidates, with 11 studies presented as supporting information.
xlix
DRAFT - DO NOT CITE OR QUOTE
-------
aionO "ao 3ii3 ion oa - uvaa
era
c
re
H
C/5
en
H
M
T3
O
W!
S
"t
re
"t
re
W!
•a
©
s
W!
re
as
"t
"t
as
V!
O5
"t
5'
era
re
VI
rt-
o"
s
re
M
T3
O
W!
S
"t
re
o
H
n
o
o
o
o
o
o
Moccarelli etal., 2008 (H)
Baccarelli etal., 2008 (H)
Alaluusua etal.,2004 (H)
Li etal., 1997 (R)"
DeCaprio etal., 1986 (G)
Smialowicz etal., 2008 (M)
Vosetal., 1973 (G)
White etal., 1986(M)
Chu etal., 2007 (R)
Cantoni etal., 1981 (R)
Crofton etal., 2005 (R)
Sewall etal., 1995 (R)
Francetal., 2001 (R)
Kociba etal., 1976 (R)
VanBirgelen etal., 1995a (R)
Fattoreetal., 2000 (R)
Toth etal., 1979 (M)"
NTP, 1982 (M)
Kociba etal., 1978 (R)
NTP, 2006 (R)
Shi etal., 2007 (R)"
Bowman etal. 1989, etc. (Mk)
Latch, and Mathur, 2002 (R)
Murray etal., 1979 (R)
Ishihara etal., 2007 (M)
Li etal., 2006 (IVl"
Kuchiiwa et al., 2002(M)
Kelleretal., 2007, etc. (M)
Ohsako etal., 2001 (R)
Bell etal., 2007a (R)
Ma rkowski etal., 2001 (R)
Hojo etal., 2002 (R)
Miettinen etal., 2006 (R)
Kattainen etal., 2001 (R)
Seo etal., 1995(R)
Schantz etal., 1996(R)
Am in etal., 2000 (R)
Huttetal.,2008(R)
Franczak etal., 2006 (R)
Sparschu etal.,1971 (R)
Ingestion Rate
(Human-equivalent dose, ng/kg-d)
o
o
o
o
to-
~ M
~
0—o-
~ >-
o
o
£2 1
= i
S: 3
(D 0>
>
3
3
Si.
CO
C
Q.
CD*
U
I-
n
o
o
>
(J)
III
"5
1
(Q
73
0)
_)
(Q
0
-------
No
Yes
No
No
Is the BMDLIess
than the LOAEL?
Yes
Yes
No
Yes
Yes
No
Is the
endpointobserved
jiear the LOAEL?.
[sthe endpointless
than the minimum
^LOAEL x 100? .
Is the
endpoint under consideration
toxicologically
relevant?
Does kinetic modeling
suggest considering additional
~^ervd po i nts at h i gh e r dosesj^.
Exclude endpoint
as a POD
Include NOAEL/LOAEL/BMDL
asa POD
Determine NOAEL, LOAEL, and BMDL (if possible) human equivalent dose
(HED) based on 1st-orderbody burden foreach study/endpointcombination
Estimate a Human Equivalent Dose (HED)
corresponding to each blood concentration NOAEL, LOAEL, or BMDL
usingthe Emond human PBPKmodel
Study/endpoint combinations from key noncancer animal bioassayswith at
leastone toxicologically relevantendpointfor RfD derivation
Determine a NOAEL, LOAEL, and BMDL (if possible)for each
study/endpoint combination, based on blood concentrationsfrom the
Emond rodentPBPKmodel
Figure ES-6. EPA's process to identify and estimate PODs from key animal
bioassays for use in noncancer dose-response analysis of TCDD.
For the studies with at least one toxicologically relevant endpoint, EPA first determined if each
endpoint was toxicologically relevant. If so, EPA determined the NOAEL, LOAEL, and BMDL
Human Equivalent Dose (HED) based on lst-order body burdens for each endpoint. Within each
study, these potential PODs were included when the endpoint was observed near the LOAEL and
if the BMDL was less than the LOAEL. Then if the endpoint was less than the minimum LOAEL
xlOO across all studies, EPA calculated PODs based on blood concentrations from the Emond
rodent PBPK model and, for all of the PODs, HEDs were estimated using the Emond human
PBPK model. If the kinetic modeling results suggested considering additional endpoints at higher
doses, the process was repeated. Finally, the lowest group of the toxicologically relevant PODs
was selected for final use in derivation of candidate RfDs.
li
DRAFT - DO NOT CITE OR QUOTE
-------
o
o
2
o
H
O
HH
H
W
O
V
o
c
o
H
W
era
c
re
M
C/5
<1
H
H
•a
©
VI
e
re
re
W!
T3
©
S
W!
re
as
"t
"t
as
O5
"t
5'
era
re
VI
rt-
©'
s
re
H
•a
©
VI
e
re
H
n
o
o
Oral Exposure
(mg/kg-day)
m
ho
Moccarelli etal. 2008
Baccarelli etal. 2008
Alaluusua etal. 2004
Li et al. 2006
Kuchiiwaetal. 2002
Smialowicz etal. 2008
Bowman etal. 1989, etc.
Keller etal. 2007a, 2008ab
Tothet al. 1979
Latch. & Mathur 2002
NTP1982
White etal. 1986
Li etal. 1997
DeCaprio etal. 1986
Shi etal. 2007
Markowski etal. 2001
Hojoetal. 2002
Cantoni etal. 1981
Voset al. 1973
Miettinenetal. 2006
Kattainenetal. 2001
NTP2006
Am in etal. 2000
Schantzet al. 1996
Huttet al. 2008
Ohsakoetal. 2001
Murray etal. 1979
Franczaket al. 2006
Chuetal. 2007
Bell etal. 2007
Ishiharaetal. 2007
VanBirgelenetal. 1995a
Kocibaetal. 1978
Fattoreet al. 2000
Seoetal. 1995
Croftonetal. 2005
Sewall etal. 1995
Franc et al. 2001
Kocibaetal. 1976
Sparschuetal. 1971
^ Dec. malefertility
I I
Inc.TSH |
Dental defects
Hormonesin preg. dams
I I I
^ Dec. immunoreactive neuron
^ Dec. SRBC resp.
Neuro behav. eff.
I
Dental defects
I
Dermal lesions
I
^ Dec. malefertility
^ Liver lesions
I I
^ Dec.serumcomp.
5^ Dec. BW, organ weights
5S» '
Neuro behav. eff.
I I
^ Operant behav
0 Inc. urinary porphyrins
I
I
Inc. FSH
Dec. serum estradiol
Dec. sens. resp. to tub
WW
I
Liver lesions
sss
I
o
cn
o
Cariogeniclesions
¦
Dental development
I I
Liver/lung lesions
I I
Red. sacc. co nsump.
I I
Faciliatory errors
I I
^ Embryotoxicity
Dev. malformations
I I
Red. Fertility & p up surv.
I i
^ Abn. estro us cycle
esio
Delayed puberty
I
Male/female sex ratio
I I
^ Dec. liver ret. palm.
I I
Liver/lung lesions
Dec. hepaticretinol
Dec. serumT4
Dec. serum T4
I
Dec. serumT4
I
liv.wt.,dec.thy. wt.
I I
Liver/lung lesions
I I
Dec. fetus BW
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
1. INTRODUCTION
Dioxins and dioxin-like compounds (DLCs), including polychlorinated dibenzo-dioxins,
polychlorinated dibenzofurans, and polychlorinated biphenyls are structurally and
toxicologically related halogenated dicyclic aromatic hydrocarbons.1 Dioxins and DLCs are
released into the environment from several industrial sources such as chemical manufacturing,
combustion, and metal processing; from individual activities including the burning of household
waste; and from natural processes such as forest fires. Dioxins and DLCs are widely distributed
throughout the environment and typically occur as chemical mixtures. Additionally, they do not
readily degrade; therefore, levels persist in the environment, build up in the food chain, and
accumulate in the tissues of animals. Human exposure to these compounds occurs primarily
through the ingestion of contaminated foods (Lorber et al.. 2009). although exposures to other
environmental media and by other routes and pathways do occur.
The health effects from exposures to dioxins and DLCs have been documented
extensively in epidemiologic and toxicological studies. 2,3,7,8-Tetrachlorodibenzo-p-dioxin
(TCDD) is one of the most toxic members of this class of compounds and has a robust
toxicological database. Characterization of TCDD toxicity is critical to the risk assessment of
mixtures of dioxins and DLCs because it has been selected repeatedly as the "index chemical"
for the dioxin toxicity equivalence factors (TEF) approach. In this approach, the toxicity of
individual components of dioxin and DLC mixtures is scaled to that of TCDD. Then, the
dose-response information for TCDD is used by the U.S. Environmental Protection Agency
(EPA) and other organizations to evaluate risks from exposure to mixtures of DLCs (U.S. EPA.
2010b; Van den Berg et al.. 2006; 1998) (also see the World Health Organization's Web site for
the dioxin toxicity equivalence factors [TEFs]).
In 2010, EPA completed and published a report entitled, Recommended Toxicity
Equivalence Factors (TEFs) for Human Health Risk Assessments of 2,3,7,8-Tetrachlorodibenzo-
p-dioxin andDioxin-Like Compounds (TEF report) (U.S. EPA. 2010b). The TEF report
describes EPA's updated approach for evaluating the human health risks from exposures to
environmental media containing DLCs. In the TEF report, EPA recommends use of the
1 For further information on the chemical structures of these compounds, see U.S. EPA (2010b. 2008b. 2003).
2 Available online at http://www.who.int/ipcs/assessment/tef_update/en/.
1 -1 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
consensus TEF values for TCDD and DLCs published in 2005 by the World Health Organization
(Van den Berg et al.. 2006) for all cancer and noncancer effects mediated through aryl
hydrocarbon receptor binding. Further, EPA recommends that the TEF methodology, a
component mixture method, be used to evaluate human health risks posed by these mixtures,
using TCDD as the index chemical. The TEFs are factors that scale individual DLC exposures
to toxicity equivalence (TEQ)3 units of TCDD. To assess health risks for a given exposure to a
mixture of DLCs, the TEQ's of those DLCs are summed, and the sum (i.e., total TEQ) is
compared to dose-response information for TCDD. Therefore, it is imperative to correctly assess
the dose response of TCDD and understand the uncertainties and limitations therein.
In 2003, EPA produced an external review draft of the multiyear comprehensive
reassessment of dioxin exposure and human health effects entitled, Exposure and Human Health
Reassessment of 2,3,7,8-Tetrachlorodibenzo-p-Dioxin (TCDD) and Related Compounds (U.S.
EPA. 2003). This draft report, herein called the "2003 Reassessment," consisted of (1) a
scientific review of information relating to sources of and exposures to TCDD, other dioxins, and
DLCs in the environment; (2) detailed reviews of scientific information on the health effects of
TCDD, other dioxins, and DLCs; and (3) an integrated risk characterization for TCDD and
related compounds.
In 2004, EPA asked the National Research Council of the National Academy of Sciences
(NAS) to review the 2003 Reassessment. The NAS Statement of Task was as follows:
3 Toxicity equivalence (TEQ) is the product of the concentration of an individual DLC in an environmental mixture
and the corresponding TCDD TEF for that compound. These products are summed to yield the TEQ of the mixture.
1 -2 DRAFT - DO NOT CITE OR QUOTE
-------
The National Academies' National Research Council will convene an expert committee that will
review EPA's 2003 draft reassessment of the risks of dioxins and dioxin-like compounds to
assess whether EPA's risk estimates are scientifically robust and whether there is a clear
delineation of all substantial uncertainties and variability. To the extent possible, the review will
focus on EPA's modeling assumptions, including those associated with the dose-response curve
and points of departure; dose ranges and associated likelihood estimates for identified human
health outcomes; EPA's quantitative uncertainty analysis; EPA's selection of studies as a basis
for its assessments; and gaps in scientific knowledge. The study will also address the following
aspects of EPA's 2003 Reassessment: (1) the scientific evidence for classifying dioxin as a human
carcinogen; and (2) the validity of the nonthreshold linear dose-response model and the cancer
slope factor calculated by EPA through the use of this model. The committee will also provide
scientific judgment regarding the usefulness of toxicity equivalence factors (TEFs) in the risk
assessment of complex mixtures of dioxins and the uncertainties associated with the use of TEFs.
The committee will also review the uncertainty associated with the 2003 Reassessment's
approach regarding the analysis of food sampling and human dietary intake data, and, therefore,
human exposures, taking into consideration the Institute of Medicine's report Dioxin and Dioxin-
Like Compounds in the Food Supply: Strategies to Decrease Exposure. The committee will focus
particularly on the risk characterization section of EPA's 2003 Reassessment report and will
endeavor to make the uncertainties in such risk assessments more fully understood by decision
makers. The committee will review the breadth of the uncertainty and variability associated with
risk assessment decisions and numerical choices, including, for example, modeling assumptions,
including those associated with the dose-response curve and points of departure. The committee
will also review quantitative uncertainty analyses, as feasible and appropriate. The committee
will identify gaps in scientific knowledge that are critical to understanding dioxin reassessment
(NAS. 2006b. p. 43. Box 1-11.
1
2
3 In 2006, the NAS published its review of EPA's 2003 Reassessment titled Health Risks from
4 Dioxin and Related Compounds: Evaluation of the EPA Reassessment (NAS. 2006b).
5
6 1.1. SUMMARY OF KEY NAS (2006B) COMMENTS ON DOSE-RESPONSE
7 MODELING IN THE 2003 REASSESSMENT
8 While recognizing the effort that EPA expended to prepare the 2003 Reassessment, the
9 NAS committee identified three key areas that they believed required improvement to support a
10 scientifically robust health assessment. These three key areas are
11
12
13
Transparency and clarity in selection of key data sets for analysis;
14
15
• Justification of approaches to dose-response modeling for cancer and noncancer
endpoints; and
16
17
18
Transparency, thoroughness, and clarity in quantitative uncertainty analysis.
1-3
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
In their Public Summary, the NAS made the following overall recommendations to aid
EPA in addressing their key concerns:
• EPA should identify the most important data sets to be used for quantitative risk
assessment for each of the four key end points (cancer, immunotoxicity, reproductive
effects, and developmental effects). EPA should specify inclusion criteria for the studies
(animal and human) used for derivation of the benchmark dose (BMD) for different
noncancer effects and potentially for the development of RfD (reference dose) values and
discuss the strengths and limitations of those key studies; describe and define
(quantitatively to the extent possible) the variability and uncertainty for key assumptions
used for each key end-point-specific risk assessment (choices of data set, POD [point of
departure],4 model, and dose metric); incorporate probabilistic models to the extent
possible to represent the range of plausible values; and assess goodness-of-fit of
dose-response models for data sets and provide both upper and lower bounds on central
estimates for all statistical estimates. When quantitation is not possible, EPA should
clearly state it and explain what would be required to achieve quantitation (NAS. 2006b.
£L_9)-
• EPA should continue to use body burden as the preferred dose metric but should also
consider physiologically based pharmacokinetic modeling as a means to adjust for
differences in body fat composition and for other differences between rodents and
humans (NAS. 2006b. p. 9).
• When selecting a BMD as a POD, EPA should provide justification for selecting a
response level (e.g., at the 10%, 5%, or 1% level). In either case, the effects of this
choice on the final risk assessment values should be illustrated by comparing point
estimates and lower bounds derived from selected PODs (NAS. 2006b. p. 9).
• EPA should compare cancer risks by using nonlinear models consistent with a receptor
mediated mechanism of action and by using epidemiological data and the new National
Toxicology Program (NTP) animal bioassay data (NTP. 2006a). The comparison should
include upper and lower bounds, as well as central estimates of risk. EPA should clearly
communicate this information as part of its risk characterization (NAS. 2006b. p. 9).
• Although EPA addressed many sources of variability and uncertainty qualitatively, the
committee noted that the 2003 Reassessment would be substantially improved if its risk
characterization included more quantitative approaches. Failure to characterize
variability and uncertainty thoroughly can convey a false sense of precision in the
conclusions of the risk assessment (NAS. 2006b. p. 5).
4 Point of departure: The dose-response point that marks the beginning of a low-dose extrapolation. This point can
be the lower bound on dose for an estimated incidence or a change in response level from a dose-response model
(BMD), or a NOAEL or LOAEL for an observed incidence, or change in level of response (available online at
http://www.epa.gov/iris/help gloss.htm#p).
1 -4 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Importantly, the NAS encouraged EPA to calculate an RfD as the 2003 Reassessment
does not contain an RfD derivation. The committee suggested that:
... estimating an RfD would provide useful guidance to risk managers to help
them (1) assess potential health risks in that portion of the population with intakes
above the RfD, (2) assess risks to population subgroups, such as those with
occupational exposures, and (3) estimate the contributions to risk from the major
food sources and other environmental sources of TCDD, other dioxins, and DLCs
for those individuals with high intakes (NAS. 2006b. p. 6).
The NAS made many other thoughtful and specific recommendations throughout their
review; additional NAS recommendations and comments pertaining to the dose-response
assessment of TCDD will be presented and addressed in various sections throughout this
document.
1.2. EPA'S SCIENCE PLAN
In May 2009, EPA Administrator Lisa P. Jackson announced the "Science Plan for
Activities Related to Dioxins in the Environment' ("Science Plan") that addressed the need to
finish EPA's dioxin reassessment and provide a completed health assessment on this high profile
chemical to the American public.5
The Science Plan outlined EPA's interim milestones for addressing several issues related
to dioxins and DLCs. With regard to EPA's response to the NAS comments on the 2003 Dioxin
Reassessment, the Science Plan stated the following:
1. EPA will release a draft report that responds to the recommendations and comments
included in the National Academy of Sciences' (NAS) 2006 review of EPA's 2003
Dioxin Reassessment.
a. EPA's National Center for Environment Assessment (NCEA) in the Office of
Research and Development, will prepare a limited response to key comments and
recommendations in the NAS report.
b. The draft response will focus on dose-response issues raised by the NAS and will
include an analysis of relevant new key studies.
5 Available at http://www.epa.gov/dioxin/scienceplan.
1 -5 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
2. EPA will provide the draft response to comments report for internal and external review.
a. The draft response to comments report will also undergo both internal EPA
review and interagency review.
b. The draft response will be provided for public review and comment and
independent external peer review.
3. The EPA Science Advisory Board (SAB) will review the science content of the response
to comments report.
As required in the Science Plan, in 2009, EPA developed a draft report titled EPA's
Reanalysis of Key Issues Related to Dioxin Toxicity and Response to NAS Comments (draft
Reanalysis) that responded to the key comments and recommendations in the NAS report (U.S.
EPA. 2010a). The draft Reanalysis focused on TCDD dose-response issues and included
analyses of relevant new studies and the derivation of an oral RfD. The draft Reanalysis was
reviewed internally by EPA scientists and externally by other federal agencies and White House
Offices. On May 21, 2010, the draft Reanalysis was released for public review and comment
and independent external peer review by EPA's SAB.
1.3. SAB REVIEW OF EPA'S DRAFT REANALYSIS
For their review, the SAB convened an expert panel composed of scientists
knowledgeable about technical issues related to dioxins and risk assessment. The SAB held
public meetings in June, July, and October 2010 and March and June 2011. They released their
final report reviewing the draft Reanalysis on August 26, 2011 (SAB. 2011).6 In their report, the
SAB made the following overarching observations:
• They found that the draft Reanalysis was clear, logical and responsive to many, but not
all, of the NAS recommendations; they were impressed with the comprehensive and
rigorous study selection process that was used to identify, review and evaluate the
scientific literature on TCDD dose response;
o .. .the SAB finds that the Report is generally clear, logical, and responsive to
many but not all of the recommendations of the NAS. The SAB has, however,
provided many recommendations to further improve the clarity, organization, and
6 Available online at http://vosemite.epa.gov/sab/sabproduct.nsf/2A45B492EBAA8553852578F9003ECBC5/$File/
EPA-SAB-1 l-014-unsigned.pdf.
1 -6 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
responsiveness of various parts of the Report. The SAB was impressed with the
process that EPA used to identify, review, and evaluate the relevant literature.
The SAB finds that EPA's process was comprehensive and rigorous and included
public participation. (SAB, 2011, p. 1)
• They agreed with the choice of the Emond physiologically based pharmacokinetic
(PBPK) model for dose metric calculations and with whole blood as the appropriate dose
metric;
o The SAB agrees with EPA's use of blood TCDD concentration as a surrogate for
tissue exposure to TCDD. Blood TCDD concentration is a better choice than
using body burden (as in the 2003 Reassessment) because it is more closely
related to the biologically relevant dose metric: the free concentration of dioxin in
the target tissues. It is important to recognize, however, that TCDD distribution
within tissues such as the liver can be nonuniform. The SAB further agrees that
the PBPK model developed by Emond et al. (2006; 2005; 2004) provides the best
available basis for the dose metric calculations in the assessment. (SAB, 2011, p.
2)
• They agreed with the choice of two epidemiologic studies as co-critical studies whose
developmental toxicity data were used to derive the RfD for TCDD;
o The SAB supports EPA's selection of the Mocarelli et al. (2008) and Baccarelli
et al. (2008) studies for identifying "cocritical" effects for the derivation of the
reference dose (RfD). These two human epidemiological studies are well
designed and provide sufficient exposure information, including biological
concentrations that could be used to establish acceptable lifetime daily exposure
levels. (SAB. 2011, p. 3)
• They agreed with EPA's evaluation of TCDD carcinogenicity (with the exception of
one panelist with a dissenting view);
o The SAB agrees with EPA's conclusion that TCDD is "Carcinogenic to
Humans." (SAB, 2011, p. 5).
The SAB also noted two deficiencies in EPA's draft Reanalysis with respect to the
completeness of the consideration of two critical elements:
• Nonlinear dose response for TCDD carcinogenicity, and
• Uncertainty analysis
1-7
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
The SAB recommended that EPA fully evaluate both linear and nonlinear dose-response
approaches to TCDD cancer dose-response assessment, including a discussion of carcinogenic
mode of action. The SAB also recommended a number of approaches to quantitative uncertainty
analysis that could be implemented by EPA, including the use of sensitivity analyses and
probability trees.
• The SAB finds that the Report did not respond adequately to the NAS recommendation to
adopt "both linear and nonlinear methods of risk characterization to account for the
uncertainty of dose-response relationship shape below the ED01." EPA should present
both linear and nonlinear risk assessment approaches. In the absence of a definitive
nonlinear mode of action, the linear option results can serve as the baseline for
comparison with other estimates. (SAB. 2011. P. 6)
• .. .the SAB does not agree with EPA's argument that conducting a unified quantitative
uncertainty analysis for TCDD toxicity is unfeasible EPA argues that a complete
quantitative uncertainty analysis would require data and resources not available. The
SAB disagrees with this logic. While EPA may lack an adequate empirical basis for full
Monte-Carlo propagation of input distributions, there are other options available. More
limited evaluations can, and should, be implemented to inform critical issues in the
dioxin reassessment. (SAB. 2011. p. 7)
The SAB made many additional thoughtful comments and specific recommendations throughout
their review pertaining to the dose-response assessment of TCDD (SAB, 2011).
1.4. SCOPE OF EPA'S REANALYSIS VOLUMES 1 AND 2
In August 2011, EPA announced a plan for moving forward to complete the draft
n
Reanalysis. This plan includes the completion and posting to the IRIS database of the
noncancer portion of the draft Reanalysis separately followed soon thereafter by the completion
and posting to the IRIS database of the cancer portion of the draft Reanalysis. As such, this
document comprises the first of two EPA reports (U.S. EPA's Reanalysis of Key Issues Related
to Dioxin Toxicity and Response to NAS Comments Volumes 1 and 2 [Reanalysis Volumes 1
and 2]) that together will respond y to the recommendations and comments on TCDD dose-
response assessment included in the NAS review of EPA's 2003 Reassessment. Both Volumes
focus on TCDD only. This report, Reanalysis Volume 1, completes and publishes EPA's study
7 Available online at http://cfpub.epa. gov/ncea/cfm/recordisplav.cfm?deid=209690.
1 -8 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
selection criteria and results for both noncancer and cancer TCDD dose-response assessment;
choice of kinetic model; noncancer RfD for TCDD; and a qualitative discussion of uncertainties
in the RfD with a focused quantitative uncertainty analysis.
These information and analyses have undergone revisions in response to SAB comments
and recommendations (see Appendix A). Reanalysis Volume 2 will address the two deficiencies
identified by the SAB, i.e., nonlinear dose response for TCDD carcinogenicity and quantitative
uncertainty analysis. In Volume 2, EPA will complete the evaluation of cancer mode-of-action,
cancer dose-response modeling, including justification of the approaches used for dose-response
modeling of the cancer endpoints, and an associated quantitative uncertainty analysis. The
information provided in Volume 1 will be used in three ways: (1) as the first of two reports that
contain EPA's response to the NAS (2006b) report, (2) as the Support Document for the TCDD
noncancer IRIS Summary and TCDD oral RfD, and (3) as technical support for Reanalysis
Volume 2.
1.5. OVERVIEW OF EPA'S RESPONSE TO NAS (2006B)
In their key recommendations, the NAS commented that EPA should thoroughly justify
and communicate approaches to dose-response modeling, increase transparency in the selection
of key data sets, and improve the communication of uncertainty (particularly quantitative
uncertainty). They also encouraged EPA to calculate an RfD. These main areas of improvement
refer to issues specifically related to TCDD dose-response assessment (and uncertainty analysis);
therefore, as noted in the Science Plan, EPA's response to the NAS is particularly focused on
these issues.
EPA thoroughly considered the recommendations of the NAS and, in Reanalysis
Volume 1, responds with scientific and technical evaluation of TCDD dose-response data via the
following:
• An updated literature search that identified new TCDD dose-response studies (see
Section 2/Appendix J);
• A kickoff workshop that included the participation of external experts in TCDD health
effects, toxicokinetics, dose-response assessment and quantitative uncertainty analysis;
these experts discussed potential approaches to TCDD dose-response assessment and
considerations for EPA's response to NAS (U.S. EPA. 2009b) (see Appendix B);
1-9
DRAFT - DO NOT CITE OR QUOTE
-------
1 • Detailed study inclusion criteria and processes for the selection of key studies (see
2 Section 2.3) and epidemiologic and animal bioassay data for quantitative TCDD
3 dose-response assessment (see Section 2.4.1/Appendix C and Section 2.4.2/Appendix D
4 respectively);
5 • Kinetic modeling that quantifies appropriate dose metrics for use in TCDD dose-response
6 assessment (see Section 3 and Appendices E and F);
7 • Sensitivity analyses that were performed on each of the animal and human Emond PBPK
8 models that identify the most sensitive variables in each model (see Section 3.3.4);
9 • Dose-response modeling for all appropriate noncancer data sets (see
10 Section 4.2/Appendix G);
11 • Thorough and transparent evaluation of the selected TCDD data for use in the derivation
12 of an RfD, including justification of approaches used for dose-response modeling of
13 noncancer endpoints (see Section 4.2 and Appendix H);
14 • The development of an RfD (see Section 4.3);
15 • A qualitative discussion of the uncertainty in the RfD and a focused quantitative
16 uncertainty analyses of the RfD (see Sections 4.4 and 4.5, respectively); and
17 • Responses to the comments and recommendations made by the SAB in their final report
18 (SAB. 2011) (see Appendix A).
19
20
21 Each of those activities is described in detail in subsequent sections of this document.
22 In addition to this document, it should be noted that several additional EPA activities
23 address other TCDD issues, specifically related to the application of dioxin TEFs and to TCDD
24 and DLC background exposure levels. Information on the application of the dioxin TEFs is
25 published elsewhere by EPA for both ecological (U.S. EPA. 2008b) and human health risk
26 assessment (U.S. EPA. 2010b). As a consequence, EPA does not directly address TEFs herein,
27 but makes use of the concept of toxicity equivalence as applicable to the analysis of exposure
28 dose in epidemiologic studies. Furthermore, this document does not address the NAS
29 recommendations pertaining to the assessment of human exposures to TCDD and other dioxins.
30 Information on updated background levels of dioxin in the U.S. population has been recently
31 reported (Lorber et al.. 2009). In 2006, EPA also released a report titled An Inventory of Sources
32 and Environmental Releases of Dioxin-Like Compounds in the United States for the Years 1987,
33 1995 and 2000, which presents an evaluation of sources and emissions of dioxins, dibenzofurans,
1-10
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
and coplanar polychlorinated biphenyls (PCBs) to the air, land and water of the United States
(U.S. EPA. 20061
1.5.1. TCDD Literature Update
EPA has developed a literature database of peer-reviewed studies on TCDD toxicity,
including in vivo mammalian dose-response studies and epidemiologic studies for use in
quantitative TCDD dose-response assessment and supporting qualitative discussions. An initial
literature search for studies published since the 2003 Reassessment was conducted by the
U.S. Department of Energy's Argonne National Laboratory (ANL) through an Interagency
Agreement with EPA. ANL used the online National Library of Medicine database (PubMed)
and identified studies published between the year 2000 and October 31, 2008 (see Appendix J).
Supporting references published since the release of the 2003 Reassessment were also identified.
Supporting studies were classified as studies pertaining to TCDD kinetics, TCDD
mode-of-action, in vitro TCDD studies, and TCDD risk assessment approaches. The literature
search strategy explicitly excluded studies addressing: (1) analytical/detection data and cellular
screening assays; (2) environmental fate, transport and concentration data; (3) dioxin-like
compounds and toxic equivalents; (4) nonmammalian dose-response data; (5) human exposure
analyses only, including body burden data; and (6) combustor or incinerator or other
facility-related assessments absent primary dose-response data.
EPA published the initial literature search results in the Federal Register on
November 24, 2008 (73 FR 70999; November 24, 2008) and invited the public to review the list
and submit additional peer-reviewed in vivo mammalian dose-response studies for TCDD,
including epidemiologic studies that were absent from the list (U.S. EPA. 2008a). Submissions
were accepted by the EPA through an electronic docket, email, and hand delivery, and they were
evaluated for use in TCDD dose-response assessment. The literature search results and
subsequent submissions were used during a 2009 scientific workshop, which was open to the
public and featured a panel of experts on TCDD toxicity and dose-response modeling (discussed
below). Additional studies identified during the workshop, and those collected by EPA scientists
during the development of this report through October 2009, have been incorporated into the
final set of studies for TCDD quantitative dose-response assessment.
1-11
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Since release of the draft Reanalysis for public comment and external peer review in
2010, EPA has collected a limited number of additional studies published since October 2009
that also inform EPA's derivation of an RfD for TCDD. These studies were identified by EPA
scientists, the SAB, and the public, and they have been used to further evaluate the biological
significance of the endpoints used to derive the RfD and to develop information on uncertainty in
the RfD. These additional studies are cited in the appropriate sections of this document. No data
sets collected since October 2009 were used quantitatively in the noncancer dose-response
assessment of TCDD.
1.5.2. EPA's 2009 Workshop on TCDD Dose Response
To assist EPA in responding to the NAS, EPA and ANL convened a scientific workshop
(the "Dioxin Workshop") on February 18-20, 2009, in Cincinnati, OH. The goals of the Dioxin
Workshop were to identify and address issues related to the dose-response assessment of TCDD
and to ensure that EPA's response to the NAS focused on the key issues and reflected the most
meaningful science. The Dioxin Workshop included seven scientific sessions: quantitative
dose-response modeling issues, immunotoxicity, neurotoxicity and nonreproductive endocrine
effects, cardiovascular toxicity and hepatotoxicity, cancer, reproductive and developmental
toxicity, and quantitative uncertainty analysis of dose response. During each session, EPA asked
a panel of expert scientists to perform the following tasks:
• Identify and discuss the technical challenges involved in addressing the NAS comments
related to the dose-response issues within each specific session topic and the TCDD
quantitative dose-response assessment.
• Discuss approaches for addressing the key NAS recommendations.
• Identify important published, independently peer-reviewed literature—particularly
studies describing epidemiologic studies and in vivo mammalian bioassays expected to
be most useful for informing EPA's response.
The sessions were followed by open comment periods during which members of the
audience were invited to address the expert panels. The session's Panel Co-chairs were asked to
summarize and present the results of the panel discussions—including the open comment
periods. The summaries were intended to reflect the core of the panel discussions and
1-12 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
incorporated points of agreement as well as minority opinions. Final session summaries were
prepared by the session Panel Co-chairs with input from the panelists, and they formed the basis
of a final workshop report (U.S. EPA. 2009b) (Appendix B of this report). Because the sessions
were not designed to achieve consensus among the panelists, the summaries do not necessarily
represent the opinions of all the scientists that attended the meeting. Some of the key discussion
points from the workshop that influenced EPA's development of this document are listed below
(see Appendix B for detail):
• In the development of study selection criteria, more relevant exposure-level decision
points using tissue concentrations could be defined.
• A linear approach to body-burden estimation, which was utilized in the 2003
Reassessment (U.S. EPA. 2003). does not fully consider key toxicokinetic issues related
to TCDD—e.g., sequestration in the liver and fat, age-dependent elimination, and
changing elimination rates over time. Thus, kinetic/mechanistic modeling could be used
to quantify tissue-based metrics. In considering human data, lipid-adjusted serum levels
may be preferable over body burden, although the assumptions used in the back
calculation of the body burden in epidemiologic cohorts are of concern. In considering
rat bioassay data, lipid-adjusted body-burden estimates may be preferable.
• New epidemiologic studies on noncancer endpoints have been published since the
2003 Reassessment that may need to be considered (e.g., thyroid dysfunction literature
from Wang et al. (2005) and Baccarelli et al. (2008).
• The 1% of maximal response (EDoi) that was utilized in the 2003 Reassessment has not
typically been used in dose-response assessment. Some alternative ideas were as follows:
(1) the POD should depend on the specific endpoint; (2) for continuous measures, the
benchmark response (BMR) could be based on the difference from control and consider
the adversity level; and (3) for incidence data, the BMR should be set to a fixed-risk
level.
• The quantitative dose-response modeling for cancer could be based on human or animal
data. There are new publications in the literature for four epidemiological cohort studies
(Dutch cohort, NIOSH cohort, BASF accident cohort, and Hamburg cohort). The
increase in total cancers could be considered for modeling human cancer data. However,
non-Hodgkin lymphoma and lung tumors are the main TCDD-related cancer types seen
from human exposure. In reviewing the rat data, the NTP (2006a) data sets are new and
can be modeled. Although the liver and lungs are the main target organs, modeling all
cancers, as well as using tumor incidence in lieu of individual rats as a measure, should
be considered.
1-13
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
• Both linear and nonlinear model functions should be considered in the cancer
dose-response analysis because there are data and rationales to support use of either
below the POD.
• For quantitative uncertainty analysis, consider the impacts of choices among plausible
alternative data sets, dose metrics, models, and other more qualitative choices. Issues to
consider include how much difference these choices make and, also, how much relative
credence should be put toward each alternative as a means to gauge and describe the
landscape of imperfect knowledge with respect to possibilities for the true dose response.
This may be difficult to do quantitatively because the factors are not readily expressed as
statistical distributions. However, the rationale for accepting or questioning each
alternative in terms of the available supporting evidence, contrary evidence, and needed
assumptions, can be delineated.
1.5.3. Organization of EPA's Response to NAS Recommendations (Reanalysis Volume 1)
The remainder of this document, Reanalysis Volume 1, is divided into three sections that
address the three primary areas of concern resulting from the NAS (2006b) review. Section 2
describes EPA's approach to the recommendation for transparency and clarity during selection of
key data sets suitable for TCDD dose-response assessment—including criteria for the selection
of key dose-response studies and results of the evaluations of the important epidemiologic
studies and animal bioassays (Appendices C and D contain study summaries and additional
details on study evaluations for the epidemiologic and animal bioassays, respectively).
Sections 3 and 4 present EPA's response to the NAS recommendation to better justify the
approaches used in dose-response modeling of TCDD for noncancer endpoints. Section 3
discusses the toxicokinetic modeling EPA conducted to support the dose-response analyses.
Section 4 presents EPA's noncancer data set selection, the noncancer dose-response modeling
results, the RfD derivation for TCDD, a qualitative discussion of the uncertainties associated
with the RfD, and a focused quantitative uncertainty analysis of the PODs considered for RfD
derivation.
1-14
DRAFT - DO NOT CITE OR QUOTE
-------
2. TRANSPARENCY AND CLARITY IN THE SELECTION OF KEY DATA SETS
FOR DOSE-RESPONSE ANALYSIS
This section addresses transparency and clarity in the study selection process and
identifies key data sets for 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) dose-response analysis.
Section 2.1 summarizes the National Academy of Sciences (NAS) committee's comments
specifically regarding this issue. Section 2.2 presents U.S. Environmental Protection Agency's
(EPA's) response to those comments and describes EPA's approach to ensuring transparency and
clarity in the selection of studies for subsequent dose-response analyses. Section 2.3 describes
the TCDD-specific study inclusion criteria and study quality evaluation process EPA used in this
document for determining the eligibility of both epidemiologic and experimental animal studies
for TCDD dose-response analysis. Section 2.4 summarizes the results of applying the study
inclusion criteria to the epidemiologic studies (see Section 2.4.1, Tables 2-1 and 2-2) and the in
vivo mammalian bioassays (see Section 2.4.2, Tables 2-3 and 2-4). These results present the key
TCDD epidemiologic and animal bioassays that were identified using the study inclusion
criteria. Additional details on this process can be found in Appendices C and D. Appendix C
summarizes all of the available epidemiologic studies, evaluates the suitability of these studies
for TCDD dose-response analyses, and presents the study selection process results. Appendix D
summarizes only the animal bioassay data that have met the study inclusion criteria for TCDD
dose-response assessment and, in Tables D-l and D-2, shows the results of the study selection
process for all of the animal bioassays identified by EPA. Study/endpoint combination data sets
for developing TCDD toxicity values for noncancer effects are further evaluated in Section 4 of
this document. Based on the cancer studies identified in this document, study/endpoint
combination data sets for developing toxicity values for cancer effects will be explored in a
separate document, Volume 2 of this effort.
2.1. SUMMARY OF NAS COMMENTS ON TRANSPARENCY AND CLARITY IN
THE SELECTION OF KEY DATA SETS FOR DOSE-RESPONSE ANALYSIS
The NAS committee proposed that EPA develop a clear and readily understandable
methodology for evaluating and including epidemiologic and animal bioassay data sets in
dose-response evaluations. The NAS committee recommended the development and application
2-1
DRAFT - DO NOT CITE OR QUOTE
-------
of transparent initial criteria to judge whether or not specific epidemiologic or animal bioassay
studies be included in TCDD dose-response analysis.
Specific NAS comments on the topic of study evaluation and inclusion criteria include
the following:
EPA should specify inclusion criteria for the studies (animal and human) used for
derivation of the benchmark dose (BMD) for different noncancer effects and
potentially for the development of RfD values and discuss the strengths and
limitations of those key studies (NAS. 2006b. p. 27).
.. .in its [EPA's] evaluation of the epidemiological literature of carcinogenicity, it
did not outline eligibility requirements or otherwise provide the criteria used to
assess the methodological quality of other included studies (NAS. 2006b, p. 56).
With regard to EPA's review of the animal bioassay data, the committee
recommends that EPA establish clear criteria for the inclusion of different data
sets (NAS. 2006b. p. 191).
.. .the committee expects that EPA could substantially improve its assessment
process if it more rigorously evaluated the quality of each study in the database
(NAS. 2006b. p. 56).
EPA could also substantially improve the clarity and presentation of the risk
assessment process for TCDD.. by using a summary table or a simple summary
graphical representation of the key data sets and assumptions... (NAS. 2006b. p.
56).
2.2. EPA'S RESPONSE TO NAS COMMENTS ON TRANSPARENCY AND CLARITY
IN THE SELECTION OF KEY DATA SETS FOR DOSE-RESPONSE ANALYSIS
EPA agrees with the NAS committee regarding the need for a transparent and clear
process with criteria identified for selecting studies and key data sets for TCDD dose-response
analyses. The delineation of the study selection process and decisions regarding key data sets
will facilitate communication regarding critical decisions made in the TCDD dose-response
assessment. In keeping with the NAS committee's recommendation to use a transparent process
and improve clarity and presentation of the health assessment process for TCDD, Figure 2-1
provides an overview of the approach that EPA has used in this document to develop a final list
of key cancer and noncancer studies for quantitative dose-response analysis of TCDD. The steps
in Figure 2-1 are further explained below.
2-2
DRAFT - DO NOT CITE OR QUOTE
-------
Literature search for in vivo mammalian and epidemiologic TCDD studies
(2000-2008): EPA conducted a literature search to identify peer-reviewed, dose-response
studies for TCDD that have been published since the 2003 Reassessment. This search
included in vivo mammalian and epidemiologic studies of TCDD from 2000 to 2008.
Additional details describing the conduct of this literature search are presented in
Section 1.5.1 of this document.
Federal Register Notice—Web publication of literature search for public comment:
In November 2008, EPA published a list of citations from results of this literature search
(U.S. EPA. 2008a) and invited the public to review this preliminary list of dose-response
citations for use in TCDD dose-response assessment. EPA requested that interested
parties identify and submit peer-reviewed studies for TCDD that were absent from this
list. Two parties identified additional references that were not included in the 2008
Federal Register notice and submitted additional references for EPA to consider. These
references were included in the final TCDD literature database considered by EPA for
TCDD dose-response analysis.
Initial study inclusion criteria development for TCDD in vivo mammalian
bioassays: EPA developed an initial set of draft criteria for evaluating the extensive
TCDD database of in vivo mammalian bioassays. These initial study inclusion criteria
had three purposes. First, they provided a method to transparently and rigorously
evaluate the scientific quality of each study in EPA's database, a deficiency in the 2003
Reassessment identified by the NAS committee. Second, their application provided an
efficient way to initially screen the vast number of TCDD mammalian bioassays for
consideration in TCDD dose-response analyses. Third, they served as a starting point for
discussions of study inclusion criteria by expert panelists who were convened by EPA for
its scientific workshop on TCDD dose-response analysis (the Dioxin Workshop),
described next [also see the workshop report in Appendix B, U.S. EPA (2009bVI.
Dioxin Workshop and expert refinement of TCDD in vivo mammalian study
inclusion criteria: In February 2009, EPA convened "A Scientific Workshop to Inform
EPA's Response to NAS Comments on the Health Effects of Dioxin in EPA's 2003
Dioxin Reassessment" [see workshop details in Section 1.5.2 and Appendix B (U.S.
EPA. 2009bVI. At the workshop, EPA presented the draft set of study inclusion criteria;
the workshop panelists evaluated the study inclusion criteria in relation to the various
toxic endpoints that were discussed and made recommendations for their revision.
Final development of study inclusion criteria for TCDD in vivo mammalian studies:
Based on discussions and recommendations made at the Dioxin Workshop, the initial
draft study inclusion criteria for evaluating the TCDD mammalian bioassay literature
were revised and are presented in Section 2.3.2.
Development of study inclusion criteria for epidemiologic studies: Following the
Dioxin Workshop, EPA determined that an evaluation process was also needed for
selection of epidemiologic studies for TCDD dose-response assessment. These criteria
were developed and are detailed in Section 2.3.1.
Final literature collection (October 2009): Additional literature was collected as it was
identified by EPA following the Dioxin Workshop through October 2009 to ensure the
consideration of all recently published data for this report.
2-3
DRAFT - DO NOT CITE OR QUOTE
-------
Studies screened using study inclusion criteria: The two sets of TCDD-specific study
inclusion criteria for epidemiologic studies and in vivo animal bioassays presented in
Sections 2.3.1 and 2.3.2, respectively, were used to evaluate all studies included in the
2003 Reassessment, studies identified in the 2000-2008 literature search, studies
identified through public comment and submission, and studies collected in 2009 as
identified by EPA during the development of this document. Section 2.4 and
Appendices C and D present results of EPA's evaluation of epidemiologic and
mammalian bioassay literature for both cancer and noncancer endpoints.
Final list of key cancer and noncancer studies for quantitative dose-response
analysis of TCDD: Application of the study inclusion criteria concludes in Section 2.4
with development of a list of key noncancer and cancer studies to be considered for
quantitative dose-response analyses of TCDD. In Section 4, points of departure (PODs)
are developed and evaluated for all biologically relevant noncancer study/endpoint
combinations from these final key study lists, and key data sets and PODs for the
development of TCDD noncancer toxicity values are identified. Similar analyses will be
undertaken in Volume 2 of this effort for TCDD cancer dose-response assessment.
2.3. STUDY SELECTION PROCESS FOR TCDD DOSE-RESPONSE ANALYSIS
In this section, EPA describes the study selection process that includes both TCDD-
specific study selection criteria and methodological considerations that have been developed to
evaluate epidemiologic studies and animal bioassays for quantitative TCDD dose-response
assessment. These criteria and considerations reflect EPA's goal of developing an RfD and a
cancer OSF for TCDD through a transparent study selection process; they are intended to be
used by EPA for TCDD dose-response assessment only. The TCDD in vivo mammalian
literature base differs from most other chemicals in magnitude and comprehensiveness. It
comprises -1,500 studies that evaluate multiple cancer and noncancer endpoints, many species
including humans, and covers an expansive dose range, including doses at and below
1 nanogram per kilogram body weight per day (ng/kg-day). Thus, the study inclusion criteria
and considerations developed in this document are specific to evaluating the TCDD literature
and cannot necessarily be genetically applied to other chemicals. Further, TCDD has a long
half-life in humans (~7 years) and bioaccumulates in fat tissue, resulting in the specification of
study inclusion criteria for estimating exposures during the critical windows for adverse health
effects. In this effort, EPA sought to identify a group of studies for TCDD dose-response
evaluation that would span the types of adverse health effects associated with TCDD exposures
and encompass the range of doses in the lower end of the dose-response region most relevant to
human health protection. Detailed study inclusion criteria have been developed that consider
2-4 DRAFT - DO NOT CITE OR QUOTE
-------
TCDD-specific issues and reflect EPA methods for POD identification, noncancer RfD
derivation, and cancer OSF derivation. (The effort in this document contrasts with EPA's 2003
Reassessment where the focus was on individual endpoints and the goal was to compare dose
response across studies.)
The study inclusion criteria and considerations were applied to each of the studies listed
in the "Preliminary Literature Search Results and Request for Additional Studies on
2,3,7,8-Tetrachlorodibenzo-p-Dioxin (TCDD) Dose-Response Studies" (U.S. EPA. 2008a);
studies identified and submitted by the public and by participants in the Dioxin Workshop (U.S.
EPA. 2009c); studies included in the 2003 Reassessment; and other relevant published studies
collected by EPA scientists through October 2009. In this effort, the goal was to identify the
most relevant studies for TCDD quantitative human health risk analyses. Those that did not
qualify were not used quantitatively, but some of these were still considered relevant to the
qualitative evaluations of the noncancer and cancer assessments. Similarly, some types of
studies were not screened, i.e., studies on dioxin-like compounds (DLCs), mixtures toxicity,
mode of action, in vitro toxicity, nonmammalian toxicology, and risk assessment; however, they
were considered to be important supplemental information to be used as needed, for example, in
discussions of biological significance.
For the study selection process, EPA has focused on TCDD studies and has not included
studies on DLCs or DLC mixtures because inclusion of the DLC literature would likely increase
the uncertainty in TCDD dose response unnecessarily, given that the TCDD database is quite
robust. In addition, EPA believes that using studies evaluating information primarily or
exclusively on TCDD dose response provides the most appropriate data for the risk assessment
of dioxins and DLCs using the TEF approach. Because TCDD is used as the index chemical in
the TEF approach, the most relevant and accurate information that specifically addresses
quantitative dose response of individual TCDD exposures is needed. The WHO expert panel
assigned TEF values from a conservative perspective that was intended to be health protective
(Van den Berg et al.. 2006). In the development of the TEFs, the WHO expert panel considered
data from Haws et al. (2006a. b), who present summary statistics of relative potency values
assembled from selected in vivo and in vitro studies. For each individual DLC, the WHO expert
panel typically assigned TEF values using an in vivo study whose relative potency value was
above the 50th percentile of the ranges presented by Haws et al. (2006a. b). Thus, when these
2-5
DRAFT - DO NOT CITE OR QUOTE
-------
TEFs are used in a dose-response study, they
produce total TEQ estimates that may be
biased high for certain combinations of
DLCs. If a RfD for TCDD were derived
based on TEQ dose-response data, that RfD
would likely also be biased high and, in that
case, would underestimate health risk from
environmental exposures. Thus, using the
TEQ data to estimate TCDD toxicity values
would not accurately reflect TCDD dose
response.
Finally, there is uncertainty in how the underlying data were used to derive the TEF
values that complicates the extrapolation of TEQ dose-response data to inform TCDD dose
response. The kinds of information available for calculating relative potencies within a study are
highly variable across DLCs, 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 the range of relative potencies presented by Haws et al. (2006a. b),
ranging from comparing dose-response curves, to developing ratios of effective doses that cause
an effect in 50% of the test units (EDsos), to estimating values from graphs of dose-response data.
The uncertainty in the TEFs can be a substantial issue for dose-response modeling when effect
levels in a study occur at doses close to background TEQ levels and TCDD is not a dominant
component of the mixture. In this case, the contribution of TCDD dose to the observed toxic
effect may not be feasible to estimate as it is confounded by other TEQ concentrations and
impacted by other TEF uncertainties.
EPA has undertaken different approaches for epidemiologic versus in vivo animal
bioassay study evaluation and key data set selection. The significant differences between animal
and human health effects data and their use in EPA health assessment support development of
separate study inclusion criteria and different approaches to study evaluation. For example,
animal bioassays on TCDD are closely controlled experiments where dose and effect are
precisely measured and causality is readily apparent; thus, the animal criteria contain precise
dose limits and specific limitations on elements of the experimental design. Because
Text Box 2-1. EPA Risk Assessment Guidelines and
Guidance Documents for Toxicity Assessment
Risk Assessment Guidelines of 1986, including chemical
mixtures, mutagenicity, cancer, exposure assessment,
developmental effects (U.S. EPA. 1986a. b)
Guidelines for Developmental Toxicity Risk Assessment
(U.S. EPA. 19911
Guidelines for Reproductive Toxicity Risk Assessment
(U.S. EPA. 1996)
Guidelines for Neurotoxicity Risk Assessment (U.S. EPA.
19981
Benchmark Dose Technical Guidance Document [external
review draft] (U.S. EPA. 20001
Guidelines for Carcinogen Risk Assessment (U.S. EPA.
2005a')
Supplemental Guidance for Assessing Susceptibility from
Early-Life Exposure to Carcinogens (U.S. EPA. 2005b')
2-6
DRAFT - DO NOT CITE OR QUOTE
-------
epidemiologic studies on TCDD are carried out within a population setting, these observational
studies employ statistical and other analytical techniques to estimate exposures/doses, and to
assess dose-response relationships after controlling or accounting for confounding factors and
other potential sources of bias. Thus, the epidemiologic criteria contain requirements for being
able to reasonably quantify the exposure-response relationship for the biologically-relevant
exposure window.1
Section 2.4 and Appendices C and D present the results of the study selection process. In
Appendix C, all of the available epidemiologic studies on TCDD are summarized and evaluated
for suitability for dose-response modeling using the TCDD-specific study inclusion criteria
described in Section 2.3.1 below; only studies meeting the study inclusion criteria and study
quality considerations are presented as key studies in Section 2.4.1 (see Tables 2-1 and 2-2 for
the cancer and noncancer endpoints, respectively). In Appendix D, because summarizing all of
the available animal bioassays on TCDD was prohibitive, only studies first meeting the in vivo
animal bioassays study inclusion criteria described in Section 2.3.2 below are summarized;
Tables D-l and D-2 present the results of the study selection process evaluations for the studies
that met and did not meet the study inclusion criteria, respectively. The selected animal studies
are presented as key studies in Section 2.4.2 (see Tables 2-3 and 2-4 for cancer and noncancer
endpoints, respectively).
2.3.1. Study Inclusion Criteria for TCDD Epidemiologic Studies
This section describes the process EPA used to select epidemiologic studies for
identifying PODs for TCDD quantitative dose-response assessment. This selection process
includes specific criteria based on EPA's approaches for deriving OSFs and RfDs (see Text Box
2-1). Additional considerations used in selecting epidemiologic data for quantitative
dose-response modeling are also necessary, particularly given EPA's preference to use human
studies over animal studies whenever possible (U.S. EPA. 2005a). As described by Hertz-
Picciotto (1995). key components needed for the use of an epidemiologic study as a basis for
1 Critical exposure windows can be identified either through conceptual understanding of the timing of the affected
biological process, such as a susceptible life-stage during which the effect is manifested, or empirically, when such
critical windows are evident from the results of an epidemiological study. Note that the conceptual understanding
can be obtained independently of the epidemiologic study in question.
2 In general, for these epidemiologic studies, EPA is evaluating tissue concentrations of TCDD that have been used
in conjunction with kinetic modeling to estimate previous TCDD exposures.
2-7 DRAFT - DO NOT CITE OR QUOTE
-------
quantitative risk assessment include issues regarding exposure assessment (a well-quantified
exposure assessment with exposures linked to individuals) and study quality ("strong biases,"
e.g., with respect to inclusion criteria for membership in the cohort and follow-up procedures
"ruled out or unlikely" and "confounding controlled or likely to be limited"). The strength of the
association, either within the full study or within a high exposure subgroup, can also be
considered in the evaluation of suitability for dose-response modeling (Hertz-Picciotto. 1995).
Stayner et al. (1999). however, note that even weak associations could be useful in terms of
providing an estimate of a potential upper bound for a quantitative risk estimate.
EPA's study selection process included applying TCDD-specific study inclusion criteria
to epidemiologic data which met the five following considerations (also see Figure 2-2 for more
details):
1. The methods used to ascertain health outcomes are clearly identified and unbiased (e.g.,
outcome classification was made "blinded" to exposure levels of the study participants).
2. The risk estimates generated from the study are not susceptible to important biases
arising from an inability to control or account for confounding factors or other sources of
bias (e.g., selection or information bias) arising from limitations of the study design, data
collection, or statistical analysis.
3. The study demonstrated an association between TCDD and an adverse health endpoint
(assuming minimal misclassification of exposure and absence of important biases) with
some suggestion of an exposure-response relationship.
This consideration in effect rules out the use of a null study (i.e., a study reporting
no association between TCDD and the health endpoint of interest) in the
quantitative dose-response assessment used to derive an RfD. Theoretically, a
no-observed-adverse-effect level (NOAEL) can be identified from a null study
and used to derive an RfD; that is, such a study could provide a "free-standing
NOAEL" that could serve as a basis for an RfD after appropriate uncertainty
factors were applied. However, a "free-standing NOAEL" from a study in which
no adverse effects have been observed is not usually chosen for RfD derivation
when other available studies demonstrate lowest-observed-adverse-effect levels
(LOAELs). The large and comprehensive database available to assess
quantitative TCDD dose response provides many positive studies that are
considered stronger candidates for derivation of an RfD than free-standing
NOAEL studies. [However, null studies are used by EPA to discuss the
biological significance of the critical endpoint(s) used as the basis for deriving an
RfD]
4. The exposure assessment methodology is clearly described and can be expected to
provide adequate characterization of exposure, with assignment of individual-level
2-8
DRAFT - DO NOT CITE OR QUOTE
-------
exposures within a study (e.g., based on biomarker data, or based on a
job-exposure-matrix approach). Limitations and uncertainties in the exposure assessment
are considered.
5. The size and follow-up period of a cohort study are large enough and long enough,
respectively, to yield sufficiently precise estimates for use in development of quantitative
risk estimates and to ensure adequate statistical power to limit the possibility of not
detecting an association that might be present. Similar considerations regarding sample
size and statistical precision and power apply to other study designs such as case-control
studies.
In addition to these five study considerations, three specific study inclusion criteria were
used to select studies for further evaluation and potential TCDD quantitative dose-response
assessment:
1. The study is published in the peer-reviewed scientific literature and provides an
appropriate discussion of data collection and analysis methods, as well as sufficient detail
to allow consideration of its strengths and limitations.
2. The exposure is primarily to TCDD, rather than DLCs, and can be quantified so that
"3
dose-response relationships can be assessed for non-fatal adverse endpoints. Because all
epidemiologic cohorts have background exposures to DLCs, in which TCDD is a minor
component, only those studies for which TCDD exposure is well above background will
qualify for dose-response modeling. To the extent to which background DLC exposure
becomes more significant with respect to TCDD exposure, limited quantitative
assessment of DLC background exposures may be necessary.
3. The effective dose and oral exposure must be quantifiable. The timing of the
measurement of health endpoints (i.e., the response) also must be consistent with current
biological understanding of the endpoint and its progression.
For cancer endpoints, EPA assumes that cumulative TCDD dose estimates are
toxicologically relevant measures. Thus, cancer studies must provide information
about long-term TCDD exposure levels. Further, for measures of cancer occurrence
or death, sufficient follow-up is needed to allow for examination of latency between
the end of effective exposure and cancer detection or death.
For noncancer endpoints, exposure estimates and analysis must allow for examination
of issues of latency and other issues regarding the appropriate time window of
exposure relevant for specific endpoints. That is, there must be sufficient
information, either in the study or elsewhere, to allow for the identification of a
biologically-relevant critical exposure window of susceptibility. A biologically-
relevant critical exposure window of susceptibility ("critical exposure window" or
3 EPA does not base RfDs on frank effects, such as mortality.
2-9 DRAFT - DO NOT CITE OR QUOTE
-------
"critical window") is an exposure period during some specific life stage over which
an individual is particularly susceptible to the agent (e.g., TCDD) for a particular
health endpoint. In utero and early lifetime exposures are often identified as critical
exposure windows for many defects in anatomical and physiological processes under
development during those periods. Critical exposure windows can be identified either
through conceptual understanding of the timing of the affected biological process,
such as a susceptible life-stage during which the effect is manifested, or empirically,
when such critical windows are evident from the results of an epidemiological study.
An example of the latter is the greater effect of early exposure to TCDD for boys
under 10 years of age on later semen quality than on boys aged 10-17 years at the
time of exposure).4 Identifying such critical windows is important for TCDD in the
practical sense of defining a reasonable duration over which to average internal
exposures that vary greatly from an initial high peak exposure to a much lower
terminal exposure, as is the case for virtually all epidemiologic studies under
consideration for TCDD. EPA considers the internal exposures following the actual
TCDD exposure incident to be relevant for averaging because of the relatively slow
elimination of TCDD and the possibility that these concentrations could still be
affecting the processes leading to the adverse health outcome.
Those studies that satisfied these three study inclusion criteria and, in addition,
adequately satisfied the study quality provisions specified in the five considerations were
considered to be suitable for quantitative TCDD dose-response analyses (see results in
Section 2.4.1 and Appendix C).
2.3.2. Study Inclusion Criteria for TCDD In Vivo Mammalian Bioassays
This section identifies the criteria EPA applied to select nonhuman in vivo mammalian
studies for defining PODs for use in TCDD dose-response modeling. These criteria are
specifically developed to evaluate the TCDD literature and are not necessarily generic, however,
they are based on EPA's approaches for deriving OSFs and RfDs from bioassay data (see
Text Box 2-1). EPA agrees with the NAS committee regarding the utility of an oral RfD and the
need for reevaluation of the OSF for TCDD, specifically in light of data that have been published
since the 2003 Reassessment was released. RfDs and OSFs are generally derived using data sets
that demonstrate the occurrence of adverse effects, or their precursors, in the low-dose range for
that chemical. RfDs and OSFs are derived from a health-protective perspective for chronic
4 Mocarelli et al., (20081: for further details of this Seveso cohort study, see the study summary in Appendix C and
RfD derivation in Section 4-3.
2-10 DRAFT - DO NOT CITE OR QUOTE
-------
exposures. Thus, when a group of studies is available on a chemical for which a number of
effects are observed at various doses across those studies, the studies using the lowest doses that
show effects will typically be selected as the basis of the RfD and OSF derivations, all other
considerations being equal. Studies conducted at higher doses relative to other available studies
are used as supporting evidence for the final RfD or OSF because they were conducted at doses
too high to impact the numeric derivations of toxicity values.
EPA expresses RfDs and OSFs in terms of average daily doses, usually as mg/kg-day and
per mg/kg-day, respectively. Thus, the study inclusion criteria for the animal bioassay data
presented in this section include requirements that average daily exposures in the studies are
within a low-dose range where, relative to other studies, they could be considered for
development of a toxicity value. These low-dose requirements do not imply that TCDD studies
conducted at higher doses are of poor quality, simply that they are not quantitatively useful in the
development of toxicity values because other studies with lower exposures will be selected as the
basis of the RfD and OSF derivations under current EPA guidance (see Text Box 2-1). Because
EPA has identified hundreds of in vivo mammalian studies that may be considered for
quantitative TCDD dose-response assessment, the development and application of these study
inclusion criteria have been critical to moving the health assessment process forward.
EPA's method for applying TCDD-specific study inclusion criteria for mammalian
bioassays is detailed below and in Figure 2-3. Four specific study inclusion criteria were used to
select studies for further evaluation and potential TCDD quantitative dose-response analyses and
identification of PODs:
1. The study is published in the peer-reviewed scientific literature.
2. The study was not conducted on a genetically-altered species.
3. The lowest dose level tested is <1 [j,g/kg-day for cancer studies and <30 ng/kg-day for
noncancer studies.
4. The study design consists of orally administered TCDD-only doses.
Those studies that satisfied these four criteria (see results in Section 2.4.2 and
Appendix D) were considered suitable for quantitative TCDD dose-response analysis.
2-11
DRAFT - DO NOT CITE OR QUOTE
-------
In evaluating the selected in vivo animal studies, EPA considered study quality issues to
ensure that the study provided important information needed to assess the relevance of the
study's endpoints and to quantify the dose-response relationship. Each study needed to test a
mammalian species and identify the strain, gender, and age of the tested animals. The study had
to clearly document its testing protocol, including dosing frequency, duration, and timing of dose
administration relative to age of the animals. For example, the control group or groups had to be
well characterized and appropriate, given the testing protocol. Also, clinical and pathological
examinations conducted during the study needed to be endpoint-appropriate, particularly for
negative findings. EPA used the results of these study evaluations in drafting study summaries
for all of the animal bioassays that met the study inclusion criteria (see Appendix D).
The criteria for dose requirements are intended to be reasonable limits that restrict the
number of studies that would need to be considered while ensuring that all study/data set
combinations that could be candidates for the cancer OSF or RfD were analyzed. Thus, the dose
range under consideration allows for liberal ranges of NOAELs, LOAELs, and benchmark dose
lower confidence bounds (BMDLs) for assessment of both cancer and noncancer effects. The
dose requirements for cancer and noncancer studies were set after EPA conducted a brief review
of typical dose levels in studies analyzed in the 2003 Reassessment and in some of the more
recent studies found through EPA's literature search.
For cancer studies, the low-dose limit was selected liberally so as not to exclude a study
that might possibly report a sensitive tumor endpoint. Given that the limit of 1 [j,g/kg-day is
3 orders of magnitude higher than the lowest-tested dose in one of the most sensitive animal
bioassays (Kociba et al.. 1978) evaluated in U.S. EPA (2003). it is virtually impossible that a
slope factor derived from a study with a low dose of 1 [j,g/kg-day would ever be considered for
the OSF reference value. Following identification of new animal cancer bioassays, no studies
were eliminated based on this limit.
For noncancer studies, the identification of a low-dose limit is more complicated because
of the variety of exposure protocols and endpoints and the consequent varied degree of
toxicokinetic extrapolation to human equivalent exposures. However, EPA is confident that the
low-dose limit of 30 ng/kg-day will not exclude any study from which a POD could be derived
that would be low enough to be considered for the RfD. A preliminary screening of the literature
indicated that, for all study types (e.g., acute, developmental, chronic), there are many studies
2-12
DRAFT - DO NOT CITE OR QUOTE
-------
with apparent effect levels well below 30 ng/kg-day. Effects observed above 30 ng/kg-day,
therefore, would have no chance of being considered as the basis for an RfD.
2.4. SUMMARY OF KEY DATA SET SELECTION FOR TCDD DOSE-RESPONSE
MODELING
To meet the NAS' concerns regarding transparency and clarity in the identification of
TCDD studies for dose-response assessment, EPA has developed and applied two sets of criteria
for epidemiologic studies and animal bioassays. EPA collected these studies through October,
2009, including studies from the 2003 Reassessment and newer studies found via literature
searches and through public submissions (see Section 2.2 and Figure 2-1). Based on these
activities, a total of 1,441 studies were examined for their potential to be used in TCDD
quantitative dose-response analysis. Of these, Figure 2-4 shows that 637 studies were eliminated
from consideration as they were not suitable study types; these included, in vitro bioassays,
review articles, PBPK modeling studies, and studies that evaluated PCBs or other dioxin—like
compounds other than TCDD. Of the remaining studies, 49 were epidemiologic studies
(7 studies contained both cancer and noncancer endpoints), and 755 were animal studies
(4 studies contained both cancer and noncancer endpoints). These epidemiologic and animal
studies were then evaluated using EPA's study inclusion criteria.
Detailed results of EPA's evaluations and study summaries are shown in Appendices C
and D for the epidemiologic studies and animal bioassays, respectively. Final results in tabular
form are shown in this section. Tables 2-1 and 2-2 contain the final lists of key cancer and
noncancer studies, respectively, that have met EPA's study inclusion criteria for epidemiologic
data. Tables 2-3 and 2-4 provide the final lists of key studies that have met EPA's study
inclusion criteria for animal bioassay data for cancer and noncancer studies, respectively.
Collectively, these four tables contain the final set of key studies that EPA has selected for
development of noncancer and cancer dose-response assessments for TCDD.
Through this study selection process, EPA has identified a relevant group of studies that
spans the possible risk analytic choices for human health protection. Each study provides
important TCDD dose-response information but also is associated with limitations and
uncertainties that must be considered and characterized during TCDD dose-response evaluations.
EPA has benefited from this effort by greatly reducing the scope of dose-response modeling and
analyses to a manageable size, and by focusing on the most important studies from the
2-13 DRAFT - DO NOT CITE OR QUOTE
-------
perspective of developing cancer and noncancer toxicity values. Results of applying the study
inclusion criteria showed that exposure information was a primary factor in study selection (see
Figure 2-4). In the epidemiologic studies, exposure needed to be primarily to TCDD and
quantifiable on an individual level. In addition, the identification of critical exposure windows
and the availability of latency information in the epidemiologic studies were vital data for
developing human exposure estimates. In the animal studies, dose limits were the most
important criteria.
2.4.1. Key Epidemiologic Data Sets
The studies listed in Tables 2-1 and 2-2, for cancer and noncancer, respectively, are those
studies that have met the epidemiologic TCDD study inclusion criteria (see Section 2.3.1).
Summaries for all of the epidemiologic studies evaluated are also provided in Appendix C and
are organized by epidemiologic cohort. Following a brief summary of each cohort, its associated
studies are then summarized chronologically, assessed for methodological considerations relative
to epidemiologic cohorts and studies, and evaluated for suitability for TCDD dose-response
assessment. Further, Appendix C presents explicit details regarding whether the considerations
and criteria were met (see summary Tables C-2 and C-3, followed by Tables C-4 though C-56,
which provide details for each study).
The cancer epidemiologic studies on TCDD that were subjected to the study selection
process include 24 peer-reviewed publications from 8 cohorts. An evaluation of these against
EPA's study inclusion criteria resulted in selecting 8 studies from the NIOSH, Boehringer,
BASF, Ranch Hand, and Seveso cohorts for further consideration in TCDD quantitative cancer
dose-response assessment (see Table 2-1). All of these studies had serum TCDD measurements
on individual study participants, used kinetic models to refine exposure estimates, and accounted
for latency or appropriate exposure windows in their analyses. As shown in Figure 2-4, most of
the other studies were excluded because exposures were not primarily to TCDD and not
quantifiable on an individual level; many studies also failed to provide information on an
appropriate latency period or window of exposure for cancer (see Table C-2). In addition,
two studies (Steenland et al.. 1999; Flesch-Janvs et al.. 1998) passed all criteria but were not
selected because they were superseded by other studies on the same cohort for which an updated
analysis was done [i.e., Steenland et al. (2001) and Becher et al. (1998). respectively]. The
2-14
DRAFT - DO NOT CITE OR QUOTE
-------
Baccarelli et al. (2006) study also passed all of the criteria but was not selected because of an
issue identified during evaluation of the study considerations (i.e., lack of an obvious adverse
health endpoint). The noncancer epidemiologic studies (see Table C-3) on TCDD that were
subjected to the study selection process include 32 peer-reviewed publications from 10 cohorts.
An evaluation of these against EPA's study inclusion criteria resulted in selecting four studies
from the Seveso cohort for further consideration in TCDD quantitative noncancer dose-response
assessment (see Table 2-2). The 4 Seveso cohort studies passed all criteria primarily because
TCDD serum levels were available for individuals in the studies, and the critical windows of
exposure were identifiable for the endpoints that served as PODs [e.g., the 9 months of
pregnancy for exposed mothers clearly defined the window of exposure for the fetus in
Baccarelli et al. (2008)1. As shown in Figure 2-4, many of the excluded studies failed to provide
enough information on expected latency for the nonfatal endpoints or failed to provide data on
the critical period of exposure to quantitatively estimate an oral human dose. A number of
studies also had exposures that were not primarily to TCDD. One study, Baccarelli et al. (2005).
passed all criteria but was excluded because the health endpoint, chloracne, is considered to be
an outcome associated with high TCDD exposures; thus this study was not considered further in
RfD derivation. The Warner et al. (2004) study also passed all criteria but was not selected
because EPA could not assess the biological significance of this finding and could not establish a
LOAEL for this effect (i.e., it did not satisfy one of the study considerations).
2.4.2. Key Animal Bioassay Data Sets
The studies listed in Tables 2-3 and 2-4, for cancer and noncancer, respectively, are those
studies that have met the in vivo animal bioassay TCDD study inclusion criteria (see
Section 2.3.2 and Figure 2-3). Appendix D provides study summaries, is organized by
reproductive studies, developmental studies, and general toxicity studies (subdivided by
duration), and summarizes the experimental protocol, the results, and the NOAELs and LOAELs
EPA has identified for each study. The doses shown in Tables 2-3 and 2-4 are expressed as
average daily administered intakes in units of nanograms per kilogram body weight per day
2-15
DRAFT - DO NOT CITE OR QUOTE
-------
(ng/kg-day), adjusted for continuous exposure when necessary.5 Tables D-l and D-2 present the
results of the study selection evaluations for the studies that met and did not meet the study
inclusion criteria, respectively.
A total of eight animal cancer bioassays were available for evaluation using EPA's study
inclusion criteria (see Section 2.3.2 and Figure 2-3). Table 2-3 presents the 6 studies that met
these criteria and are considered suitable for quantitative TCDD dose-response modeling. As
shown in Figure 2-4, only 2 of the available cancer bioassays did not meet EPA's study inclusion
criteria (and are not summarized in Appendix D). These include Eastin et al. (1998) (genetically
altered mouse strain) and Rao et al. (1988) (intraperitoneal injection instead of oral route of
exposure).
A total of 751 animal bioassays on a noncancer endpoint were available for evaluation
using EPA's study inclusion criteria (see Section 2.3.2 and Figure 2-3). As shown in Figure 2-4,
673 of the available noncancer studies were excluded based on one or more of the following
reasons: (1) 66 studies used genetically-altered animals; (2) 370 studies had a lowest tested dose
that was too high (i.e., greater than 30 ng/kg-day); (3) 142 studies tested chemicals that were not
TCDD only or used an unspecified TCDD dose; and (4) 135 studies employed a nonoral dosing
method. Table D-2 of Appendix D shows these studies and identifies the study inclusion criteria
that were not met. For many studies, more than one reason for exclusion was found and
identified. Conversely, in some cases, at least one identified criterion was not met, and, given
the study was then excluded based on that one criterion, not all of the other criteria for exclusion
were further evaluated and articulated. Tables 2-4 and D-l of Appendix D present the 78 studies
that were selected as key data sets for TCDD noncancer dose-response analyses.
In Section 4, additional evaluations are made to determine which study/endpoint data sets
are the most appropriate for development of the RfD for TCDD. For further consideration in the
RfD derivation process, only the toxicologically-relevant endpoints from the studies in Table 2-4
are carried forward to Section 4 (see Section 4.2.1 and Appendix H for details on study/endpoint
combinations not used in RfD derivation for this reason). For some entries in Table 2-4, there
are several publications from the peer-reviewed literature shown in the same row of the table. In
these cases, the publications are grouped together because they are based on the same noncancer
5 Standard EPA guidance was applied for adjustment of intermittent gavage protocols and dietary exposures as
indicated in each specific study description in Appendix D.
2-16 DRAFT - DO NOT CITE OR QUOTE
-------
animal bioassay. Additionally, in Table 2-4, the noncancer adverse effects in the animal studies
listed under the heading, "endpoints examined," are presented as general categories of effects,
such as "developmental effects," "liver effects," or "thyroid function." In Section 4, more
detailed descriptors of the specific endpoints associated with such adverse health effects are
articulated and evaluated to develop PODs for the derivation of an oral RfD for TCDD. Final
candidate study/endpoint data sets are selected in Section 4 based on factors such as
toxicological relevance of the endpoints, dose-response modeling results, and POD comparisons
across studies, as illustrated in Figures 4-1 and 4-3 for epidemiologic and toxicological data,
respectively.
2-17
DRAFT - DO NOT CITE OR QUOTE
-------
Table 2-1. Epidemiologic studies selected for TCDD cancer dose-response modeling
to
00
o
o
o
2
o
H
o
HH
H
W
O
o
c
o
H
W
Location,
No. of
Effect measure/
Health
time
Cohort
Exposure
Exposure
cases or
trend tests
Reference
outcome
period
description
assessment
measures
deaths
(/>-value)
Risk factors
Comments
Akhtar et
Mortality
Vietnam
Ranch Hand
Cumulative
CSLC
Adjusted for age at
Used multiplicative
al. (2004)
and
1962-1971
(RH) cohort
serum lipid
(ppt-years)
tour, military
Poisson regression
incidence
including 1,196
concentrations
RH and C <2
occupation,
models to compare
for all
U.S. military
(CSLC) of
yrs in SEA:
smoking, skin
cancer incidence and
cancers and
males exposed
TCDD based on
reaction to sun
cancer mortality with
for site-
by spraying
serum levels
All site
No.,%
RR (95% CI)
exposure, eye
national rates and
specific
Agent Orange
collected from
Comparison
34,5.9
1.0
color, number of
proportional
cancers
during Vietnam
veterans in
<10
28, 9.8
1.44 (0.82-2.53)
years in SEA.
hazards models to
including
war in Southeast
1987, 1992,
Low >10-118.5
22, 14.6
2.23 (1.24-4.00)
contrast cohorts with
prostate
Asia (SEA);
1997, and a
High >118.5
15,8.6
2.02 (1.03-3.95)
Also stratified
regard to cancer
and
comparison (C)
first-order
Continuous
1.24 (1.01-1.53)
analyses by year
incidence.
melanoma
cohort matched
kinetic model
(Log TCDD)
p = 0.04
of tour of duty.
by age, race, and
with a 7.6-year
Restricted to
military
half-life. CSLC
Melanoma
No., %
<2 years in SEA,
occupation.
estimates for
Comparison
3,0.5
1.0
white Air Force
1,009 RH cohort
<10
4, 1.4
2.99 (0.53-16.8)
veterans, 0% and
and 1,429 C
Low >10-118.5
4, 2.7
7.42 (1.34-41.04)
100% time in
cohort veterans.
High >118.5
3, 1.7
7.51 (1.12-50.21)
Vietnam for RH
Continuous
2.24 (1.29-3.89)
and C Cohorts,
(Log TCDD)
p = 0.004
respectively.
Prostate
No., %
Comparison
7, 1.2
1.0
<10
10,3.5
1.5 (0.51-4.40)
Low >10-118.5
6,4.0
2.17(0.68-6.87)
High >118.5
5,2.9
6.04 (1.48-24.61)
Continuous
1.48 (0.93-2.35)*
(Log TCDD)
p = 0.10
-------
Table 2-1. Epidemiologic studies selected for TCDD cancer dose-response modeling (continued)
Location,
No. of
Effect measure/
Health
time
Cohort
Exposure
Exposure
cases or
trend tests
Reference
outcome
period
description
assessment
measures
deaths
(/j-valuc)
Risk factors
Comments
Becher et
Mortality
Hamburg,
Boehringer
CSLC of TCDD
Included in
al. (1998)
from all
Germany,
cohort including
based on area
U.S. EPA (2003).
cancers
production
approximately
under curve (in
A large number of
combined
period was
1,189 workers
Hg/kg years);
Categorical
Available: year of
models were fitted.
1950-1984,
employed in the
back-
exposures (Cox
entry, age of entry,
These included
and
production of
extrapolation to
model)
124
RR (95% CI)
duration of
models for 5
mortality
herbicides.
date of last
0-
-------
Table 2-1. Epidemiologic studies selected for TCDD cancer dose-response modeling (continued)
Location,
No. of
Effect measure/
Health
time
Cohort
Exposure
Exposure
cases or
trend tests
Reference
outcome
period
description
assessment
measures
deaths
(/j-valuc)
Risk factors
Comments
Cheng et al.
Mortality
USA,
NIOSH cohort
CSLC of TCDD
No exposure
256
The slope (P) was
Available: age,
Confounding by
(2006)
from all
1942-1993
including 3,538
based on work
categories
cancer
3.3 x 10-6 for lag
year of birth, and
smoking was
cancers
occupationally
histories, job-
provided
deaths
of 15 years
race.
considered indirectly
exposed male
exposure matrix,
excluding upper
by analysis of
workers at
and
5% of TCDD
Risks adjusted for:
smoking-related and
8 plants in the
concentration
exposures.
year of birth, age,
smoking-unrelated
United States;
and age-
The slopes ranged
and race.
cancers.
256 cancer
dependent two-
two orders of
Other occupational
deaths.
compartment
magnitude
Indirectly
exposures were
model of
depending on
examined other
considered indirectly
elimination
modeling
potential
by repeated analyses
kinetics.
assumption.
confounders such
removing one plant
as smoking and
at a time.
other occupational
Based on indirect
exposures.
evaluation, there was
no clear evidence of
confounding.
-------
Table 2-1. Epidemiologic studies selected for TCDD cancer dose-response modeling (continued)
to
to
o
O
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
ffl
Location,
No. of
Effect measure/
Health
time
Cohort
Exposure
Exposure
cases or
trend tests
Reference
outcome
period
description
assessment
measures
deaths
(/j-valuc)
Risk factors
Comments
Collins et
Mortality
Midland,
Subset of
CSLC of TCDD
Part per billion-
177
The slope of a
Hazard ratios
Confounding by
al. (2009)
from all
MI, USA.
NIOSH cohort
based on work
year
cancer
proportional
adjusted for age,
smoking was not
cancers and
Follow-up
including 1,615
histories,
estimates of
deaths
hazards regression
year of birth, and
considered directly
specific
period:
occupationally
job-exposure
cumulative
model for fatal
hire year.
due to a lack of data.
cancer
1942-2003.
exposed male
matrix, and
TCDD
soft tissue
Stratified analyses
Relatively long
types
Serum
workers at
concentration
exposure
sarcoma was
used to examine
follow-up period
collection
1 plant in the
and age-
0.05872 (95% CI
potential impact of
(average = 36 years).
period:
United States;
dependent two-
not provided but
pentachlorophenol
Potential outcome
2004-2005
177 cancer
compartment
for Chi-square
exposure on
misclassification for
deaths.
model of
p = 0.0060) for
mortality.
soft tissue sarcoma
elimination
every 1-part per
due to potential
kinetics. Serum
billion-year
inaccuracies on
samples were
increase in
death certificates.
obtained from
cumulative
Data analyzed from
280 former
exposure of
one plant reduces
workers
TCDD. Slope
heterogeneity
collected during
estimates for all
associated with
2004-2005.
fatal cancers
multiplant analyses.
(0.00161,
More serum samples
p = 0.78), fatal
(n = 280) analyzed
lung (-0.00173,
than used to derive
p = 0.89), fatal
TCDD estimates for
prostate (0.01294,
other NIOSH cohort
p = 0.30), fatal
analyses.
leukemias
(-0.12822,
p = 0.34), and
fatal non-Hodgkin
lymphomas
(0.01081,
p = 0.68) were not
statistically
significant.
-------
Table 2-1. Epidemiologic studies selected for TCDD cancer dose-response modeling (continued)
Location,
No. of
Effect measure/
Health
time
Cohort
Exposure
Exposure
cases or
trend tests
Reference
outcome
period
description
assessment
measures
deaths
(/j-valuc)
Risk factors
Comments
Michalek
Cancer
Vietnam
RH cohort
CSLC of TCDD
CSLC
Continu
Cox regression
Without
and Pavuk
incidence,
1962-1971
including 1,196
based on serum
(ppt-years)
ous ex-
proportional
stratification, there
(2008)
all sites
U.S. military
levels collected
posure:
hazards models
was no significant
combined
males exposed
from veterans in
Results
Log
adjusted for year
increase in the risk
by spraying
1987, 1992,
stratified by
(TCDD)
of birth, eye color,
of cancer with
Agent Orange
1997, 2002, and
<1968, >30
No.,%
race, smoking,
log(TCDD) in the
during Vietnam
a first-order
days pre-1967,
67, 12.6
1.4(1.1-1.7)
body mass index
combined cohort.
war in Southeast
kinetic model
<2 yrs in SEA:
p = 0.005
at the qualifying
Asia (SEA); C
with a 7.6-year
tour, military
cohort matched
half-life. CSLC
occupation, and
by age, race, and
estimates for
Cate-
skin reaction to
military
986 RH cohort
gorical
sun exposure.
occupation.
and 1,597 C
cohort veterans.
Comparison
<10
Low >10-91
High >91
TCDD
No., %
30, 11.2
10,8.3
12, 24.5
15, 16.1
RR (95% CI)
1.0
0.5 (0.2-1.1)
1.7 (0.8-3.5)
2.2(1.1-4.4).
Also stratified
analyses by years
of service in SEA,
days of herbicide
spraying, calendar
period of service.
-------
Table 2-1. Epidemiologic studies selected for TCDD cancer dose-response modeling (continued)
Reference
Health
outcome
Location,
time
period
Cohort
description
Exposure
assessment
Exposure
measures
No. of
cases or
deaths
Effect measure/
trend tests
(/j-valuc)
Risk factors
Comments
Ott and
Zober
(1996)
Mortality
and
incidence
for all
cancers
combined,
as well as
for specific
cancer sites
Ludwig-
shafen,
Germany,
1954-1992
BASF cohort,
243 men
exposed from
accidental
release that
occurred in 1953
during
production of
trichlorophenol,
or who were
involved in
clean-up
activities.
CSLC of TCDD
expressed in
|ig/kg based on
TCDD half-life
of 5.1-8.9 years,
Cox regression
model.
Internal
comparisons
based on
continuous
measure of
TCDD.
External
comparisons
exposure
categories (for
malignant
neoplasms):
<0.1,
0.1-0.99
1.0-1.99
>2 ng/kg
Internal
cohort
analysis
31 All
cancer
deaths
47 All
incident
cancers
External
cohort
analyses
Deaths
RR (95% CI)
1.22 (95% CI:
1.00-1.50)
1.11 (95% CI:
0.91-1.35)
SMR (95% CI)
0.8 (0.4-1.6)
1.2 (0.5-2.3)
1.4 (0.6-2.7)
2.0 (0.8-4.0)
Available: age,
BMI, smoking
status, and history
of occupational
exposure to
aromatic amines
and asbestos.
Included in
U.S. EPA (2003)
Positive associations
noted for digestive
cancer, but not for
respiratory cancer.
Association between
TCDD and increased
SMRs found only
among current
smokers.
Last published
account of this
cohort.
H
ffl
O
V
o
c
o
H
ffl
-------
Table 2-1. Epidemiologic studies selected for TCDD cancer dose-response modeling (continued)
Location,
No. of
Effect measure/
Health
time
Cohort
Exposure
Exposure
cases or
trend tests
Reference
outcome
period
description
assessment
measures
deaths
(/j-valuc)
Risk factors
Comments
Steenland
Mortality
USA,
NIOSH cohort
CSLC of TCDD
CSLC
Available: date of
Included in
et al.
from all
1942-1993
including 3,538
based on work
(ppt-years)
RR (95% CI)
birth and age.
U.S. EPA (2003)
(2001)
cancers
male workers,
histories, job-
<335
64
1.00
256 cancer
exposure matrix,
335-520
29
1.26 (0.79-2.00)
Adjusted for date
deaths.
and a simple
520-1,212
22
1.02 (0.62-1.65)
of birth, and age
one-
1,212-2,896
30
1.43 (0.91-2.25)
was used as time
compartment,
2,896-7,568
31
1.46 (0.93-2.30)
scale in Cox
first-order
7,568-20,455
32
1.82(1.18-2,82)
model.
pharmacokinetic
>20,455
48
1.62 (1.03-2,56)
elimination
model with
8.7-year half-
life.
-------
Table 2-1. Epidemiologic studies selected for TCDD cancer dose-response modeling (continued)
to
to
o
O
o
2
o
H
o
HH
H
W
O
o
c
o
H
W
Location,
No. of
Effect measure/
Health
time
Cohort
Exposure
Exposure
cases or
trend tests
Reference
outcome
period
description
assessment
measures
deaths
(/j-valuc)
Risk factors
Comments
Warner et
Breast
Italy
981 women
CSLC of TCDD
Categorical
Cases
RR (95% CI)
Available:
Included in
al. (2002)
cancer
1976-1998
from Zones A
(ppt) collected
<20 ppt
1
1.0
gravidity, parity,
U.S. EPA (2003)
incidence
and B with
between 1976
20.1-44 ppt
2
1.0 (0.1-10.8)
age at first
available archive
and 1981. For
44.1-100 ppt
7
4.5 (0.6-36.8)
pregnancy, age at
serum samples,
most samples
>100 ppt
5
3.3 (0.4-28.0)
last pregnancy,
15 breast cancer
collected after
p = 0.07
lactation, family
cases.
1977, serum
Continuous
history of breast
TCDD levels
(LogioTCDD)
15
2.1 (1.0-4.6)
cancer, age at
were back-
menarche, current
extrapolated
body mass index,
using a first-
oral contraceptive
order kinetic
use, menarcheal
model with a
status at explosion,
9-year half-life.
menopause status
at diagnosis,
height, smoking,
alcohol
consumption.
Adjusted for age,
which was used as
time scale in Cox
model; other
covariates were
evaluated but were
not identified as
confounders.
-------
Table 2-2. Epidemiologic studies selected for TCDD noncancer dose-response modeling
to
to
On
O
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
ffl
Effect measure/
Health
Location,
Cohort
Exposure
Exposure
No. of
trend tests
Reference
outcome
time period
description
assessment
measures
cases
(/j-valuc)
Risk factors
Comments
Alaluusua et al.
Dental
Seveso, Italy,
65 subjects
Serum TCDD
Dental defect %
Available: medical
Dose-response
(2004)
defects
Dental
<9.5 years
(ng/kg) from
Non-ABR
10
26%
history, age, sex,
pattern observed
exams
old at time of
1976 samples
Zone
education,
with dental defects
administered
Seveso
for those who
31-226
1
10%
smoking.
in the ABR zone;
in 2001
explosion
resided in
ng/kg
however, the control
among those
and residing
Zones ABR; no
238-592
5
45%
population had a
exposed to
in Zones
serum levels for
ng/kg
much higher
TCDD in
ABR (i.e.,
non-ABR
700-26,000
9
60%
prevalence of dental
1976
the most
heavily
residents
(unexposed).
ng/kg
p-valuc = 0.016
defects (26%) than
those in the lowest
contaminated
TCDD
<5 years of
25
33%
exposure group
area in
exposure
age at time
/?-value = 0.0009
(10%).
decreasing
represent levels
of accident
order); 130
as of 1976
Also assessed
subjects
(after accident).
Odds Ratios
hypodontia and
recruited
(95% CI)
other dental and oral
from the
(among those
aberrations, but
non-ABR
Non-ABR
<5 years of age at
these were too rare
region (i.e.
Zone or
time of accident)
to allow modeling
the
31-226
1.0
by ABR zone.
unexposed).
ng/kg serum
TCDD
238-26,000
ng/kg serum
TCDD
2.4 (1.3-4.5)
p-valuc = 0.007
-------
Table 2-2. Epidemiologic studies selected for TCDD noncancer dose-response modeling (continued)
to
to
-J
o
O
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Effect measure/
Health
Location,
Cohort
Exposure
Exposure
No. of
trend tests
Reference
outcome
time period
description
assessment
measures
cases
(/>-value)
Risk factors
Comments
Baccarelli et
b-TSH
Italy, 1976;
Population-
Based on zone
Population-
Population-based
Available: gender,
An association with
al. (2008)
measured
children,
based study:
of residence,
based study:
study
birth weight, birth
serum TCDD levels
72 hours after
1994-2005
1,041
estimated mean
Geometric Mean
order, maternal age
of mothers was
birth from a
singletons
values from a
b-TSH (log-
at delivery,
found with b-TSH
heel pick
(56 from
previous study.
transformed)
hospital, type of
among the 51 births
(routine
Zone A, 425
Maternal
delivery.
in the plasma dioxin
screening for
from Zone B,
plasma TCDD
Reference
533
Reference:
study.
all newborns in
and 533 from
levels estimated
births
0.98 (95% CI:
There was limited
the region)
reference)
at the date of
0.90-1.08)
evidence of
born between
delivery using a
Zone A
56
Zone B:
confounding, so
Jan. 1, 1994-
first-order
births
1.66 (95% CI:
mean TSH results
June 30,
pharmacokineti
1.19-2.31)
presented here are
2005.
c model and
Zone B
425
Zone A:
unadjusted.
Plasma
elimination rate
births
1.35 (95% CI:
dioxin study:
estimated in
1.22-1.49)
51 children
Seveso women
born to 38
(half-life =
Plasma
Association
women of
9.8 years).
dioxin
between neonatal
fertile age
study:
b-TSH with
who were
Continuous
plasma TCDD:
part of the
maternal
adjusted p = 0.75
Seveso
plasma
(p< 0.001)
Chloracne
TCDD
Study.
-------
Table 2-2. Epidemiologic studies selected for TCDD noncancer dose-response modeling (continued)
Effect measure/
Health
Location,
Cohort
Exposure
Exposure
No. of
trend tests
Reference
outcome
time period
description
assessment
measures
cases
(/>-value)
Risk factors
Comments
Eskenazi et
Menstrual
Seveso, Italy,
Women who
Serum TCDD
Interquartile
Interview data:
A positive
al. (2002b)
cycle
follow-up
were <40
(ng/kg) from
range was
medical history,
association between
characteristics:
interview
years from
1976 samples.
64-322 ppt
personal habits,
menstrual cycle
menstrual
conducted in
Zones A or B
TCDD
work history,
length and serum
cycle length.
1996-1997 of
in 1976.
exposure level
TCDD
Lengthening of
reproductive
TCDD was found
women
was back-
examined as
the menstrual
history, age,
among women who
exposed to
extrapolated to
continuous
cycle by 0.93
smoking, body
were premenarcheal
TCDD in the
1976 using the
measure
days (95% CI:
mass index, alcohol
at the time of
1976
Filser or the
(per 10-fold
-0.01, 1.86)
and coffee
accident (n = 134).
accident
first-order
increase in
consumption,
kinetic models.
serum
exercise, illness,
levels).
abdominal
surgeries.
-------
Table 2-2. Epidemiologic studies selected for TCDD noncancer dose-response modeling (continued)
to
to
VO
o
o
O
2
o
H
O
HH
H
W
O
o
c
o
H
W
Effect measure/
Health
Location,
Cohort
Exposure
Exposure
No. of
trend tests
Reference
outcome
time period
description
assessment
measures
cases
(/>-value)
Risk factors
Comments
Mocarelli et
Sperm conc.
Italy, 1976,
Among the
Serum TCDD
Median
Available: age,
Results stratified by
al. (2008)
(million/mL)
1998
257 exposed
(in ppt) from
serum
Men exposed
abstinence time,
timing of exposure
Progressive
(from Zone
1976-1977
TCDD
between the ages
smoking status,
(1-9 yrs old vs.
motility (%)
A), men
samples (for
levels (in
1-9 had reduced
education, alcohol
10-17 yrs old in
Serum E2
1-26 in 1976
exposed men);
ppt) by
semen quality
use, maternal
1976).
(pmol/L)
with serum
background
quartile for
22 years later.
smoking during
levels <2000
values were
men aged
Reduced sperm
pregnancy,
ppt in 1976,
assumed for
1-9 in 1976
quality included
employment status,
135 (53%)
unexposed men
(68; 142;
decreases in
BMI, chronic
were
based on serum
345; 733
sperm count
exposure to
included.
analysis of
PPt)
(p = 0.025),
solvents and other
Among the
residents in
progressive
toxic substances.
372
uncontaminated
sperm motility
nonexposed
areas.
(p = 0.001), and
Adjusted for
invitees, 184
total number of
smoking status,
(49%) men
motile sperm
organic solvents,
aged 1-26 in
(p = 0.01) relative
age at time of tests,
1976 were
to the comparison
BMI, alcohol use,
included.
group.
education,
employment status,
and abstinence
(days) for sperm
data.
Hormone data not
adjusted for
education level,
employment status,
and abstinence
time.
-------
Table 2-3. Animal bioassays selected for cancer dose-response modeling
to
o
o
o
o
2
o
H
O
HH
H
W
O
o
c
o
H
W
Reference
Species/strain
Sex
exposure
route/duration
n
Average daily
dose levels
(ng/kg-day)
Cancer types
Statistical significant tumors
(pairwise with controls or trend tests)
Delia Porta et
al. (1987)
Mouse/
B6C3FJ
Male/female
Oral gavage once
per week; 52 weeks
~40 to 50 in
each dose
group
including
controls
0,351, and 714
Females and males:
hepatocellular
adenomas and
carcinomas
Liver: adenomas and carcinomas in females
and carcinomas in males (using incidental
tumor statistical test)
Kociba et al.
(1978);
Goodman and
Sauer (1992)
Rat/Sprague-
Dawley
Male/female
Oral-lifetime
feeding; 2 years
50 each
(86 each in
vehicle
control
group)
0, 1, 10, or 100
Females: liver, lung,
oral cavity
Males: adrenal, oral
cavity, tongue
Adrenal cortex: adenoma
Liver: hepatocellular adenoma(s) or
carcinoma(s); hyperplastic nodules
Lung: keratinizing squamous cell carcinoma
Oral cavity: stratified squamous cell
carcinoma of hard palate or nasal turbinates
Tongue: stratified squamous cell carcinoma
NTP (1982c)
Mouse/
B6C3FJ
Male/female
Oral-gavage twice
per week;
104 weeks
50 each
(75 each in
vehicle
control
group)
0, 1.4, 7.1, or 71 for
males;
0,5.7, 28.6, or 286
for females
Females:
hematopoietic system,
liver, subcutaneous
tissue, thyroid
Males: liver, lung
Hematopoietic system: lymphoma or
leukemia
Liver: hepatocellular adenoma or carcinoma
Lung: alveolar/bronchiolar adenoma or
carcinoma
Subcutaneous tissue: fibrosarcoma
Thyroid: follicular-cell adenoma
NTP (1982c)
Rat/Osborne-
Mendel
Male/female
Oral-gavage twice
per week;
104 weeks
50 each
(75 each in
vehicle
control
group)
0, 1.4, 7.1,or71
Females: adrenal, liver,
subcutaneous tissue,
thyroid
Males: adrenal, liver,
thyroid
Adrenal: cortical adenoma, or carcinoma or
adenoma, NOS
Liver: neoplastic nodule or hepatocellular
carcinoma
Subcutaneous tissue: fibrosarcoma
Liver: neoplastic nodule or hepatocellular
carcinoma
Thyroid: follicular-cell adenoma or carcinoma
-------
Table 2-3. Animal bioassays selected for cancer dose-response modeling (continued)
Reference
Species/strain
Sex
exposure
route/duration
n
Average daily
dose levels
(ng/kg-day)
Cancer types
Statistical significant tumors
(pairwise with controls or trend tests)
NTP (2006a)
Rat/Harlan
Sprague-
Dawley
Female
Oral-gavage
5 days per week;
2 years
53 or 54
0,2.14,7.14, 15.7,
32.9, or 71.4
Liver
Lung
Oral mucosa
Pancreas
Liver: hepatocellular adenoma
Liver: cholangiocarcinoma
Lung: cystic keratinizing epithelioma
Oral mucosa: squamous cell carcinoma
Pancreas: adenoma or carcinoma
Toth et al.
(1979)
Mouse/
Outbred
Swiss/H/Riop
Male
Gastric intubation
once per week;
1 year
43 or 44
(vehicle
control
group = 38)
0, 1, 100, or 1,000
Liver
Liver: tumors
-------
Table 2-4. Animal bioassay studies selected for noncancer dose-response modeling
to
LtJ
to
o
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
ffl
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Reproductive toxicity studies
Bowman et
al.(1989a;
1989b"»; Schantz
and Bowman
(1989); Schantz et
al. (1992: 1986)
Monkey/
Rhesus
Daily dietary
exposure in
female
monkeys
(3.5-4 years)
F (F0, Fl,
F2, F3)
3 to 7 (Fl)
0, 0.12, or
0.67
None
0.12
Reproductive
and
developmental
effects
Neurobehavioral
effects (e.g.,
discrimination-
reversal learning
affected)
Franc et al.
(2001)
Rat/Sprague-
Dawley,
Long-Evans,
Han/Wistar
Biweekly oral
gavage
(22 weeks)
Female
8
0, 10, 30 or
100
10
30
Body weight,
relative liver
weight, relative
thymus weight
Increased relative
liver weight in
Sprague-Dawley and
Long-Evans Rats;
Increased relative
thymus weight in
Sprague-Dawley,
Han/Wistar, and
Long-Evans Rats
Hochstein et al
(2001)
Mink
Daily dietary
exposure
(132 days)
F
12
0.03
(control), 0.8,
2.65,9, or 70
None
2.65
Reproductive
effects
Reduced kit survival
Hutt et al. (2008)
Rat/Sprague-
Dawley
Oral gavage
(GDs 14 and
21, postpartum
days 7 and 14),
(Pups: once
per week for
3 months)
Female
(F0 and
Fl)
3 (F0 and
Fl)
0 or 7.14
None
7.14
Developmental
effects
Lower proportion of
morphologically
normal pre-
implantation
embryos during
compaction stage
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
to
o
o
o
2
o
H
o
HH
H
W
o
o
c
o
H
W
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Reproductive toxicity studies (continued)
Ikeda et al. (2005)
Rat/
Holtzman
Corn oil
gavage
(initial
loading dose
followed by
weekly dose
during
mating,
pregnancy,
and lactation-
about
10 weeks)
F (F0)
F andM
(F1 and
F2)
12 (F0)
Not specified
(F1 andF2)
0 or 16.5
None
16.5
(maternal
exposure)
Reproductive
and
developmental
effects
Decreased
development of the
ventral prostrate
(Fl), decreased sex
ratio (percentage of
males) (F2)
Ishihara et al.
(2007)
Mouse/ICR
Sesame oil
gavage
(initial
loading dose
followed by
weekly doses
for 5 weeks)
M (F0)
42 or 43
0, 0.095, or
950
0.1
100
Reproductive
effects
Decreased
male/female sex ratio
(percentage of males)
(Fl)
Latchoumy-
candane and
Mathur (2002)
and related
Latchoumy-
candane et al.
C2003. 2002a:
2002b)
Rat/Wistar
albino
Olive oil
gavage (daily
for 45 days)
M
6
0, 1, 10, or 100
None
1
Reproductive
effects
Reduced sperm
production,
decreased
reproductive organ
weights
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
to
LtJ
o
o
o
2
o
H
o
HH
H
W
o
o
c
o
H
W
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Reproductive toxicity studies (continued)
Murray et al.
(1979)
Rat/Sprague-
Dawley
Daily dietary
exposure
(3 generations
)
F and M,
(F0)
F and M,
(F1 and
F2)
10-32 (F0)
22 (Fl)
28 (F2)
0, 1, 10, or 100
1
10
Reproductive
and
developmental
effects
Decrease in fertility,
decrease in the
number of live pups,
decrease in
gestational survival;
decrease in postnatal
survival, decreased
postnatal body
weight in one or
more generations
Shi et al. (2007)
Rat/Sprague-
Dawley
Maternal corn
oil gavage
(weekly on
GDs 14 and
21; PNDs 7
and 14)
Offspring
corn oil
gavage
(weekly for
11 months)
F (F0)
F (Fl)
3 (F0)
10 (Fl)
0, 0.14,0.71,
7.14, or 28.6
0.14
0.71
Reproductive
effects
Decrease serum
estradiol levels (Fl)
Yans et al. (2000)
Rhesus
monkey/
Cynomolgus
Fed gelatin
capsules
(5 days/week
for
12 months)
F
6 (treatment)
5 (controls)
0, 0.71,3.57,
or 17.86
17.86
None
Endometriosis
effects
Increased
endometrial implant
survival, increased
maximum and
minimum implant
diameters, growth
regulatory cytokine
dysregulation
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
to
o
o
o
2
o
H
o
HH
H
W
O
o
c
o
H
W
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Developmental toxicity studies
Amin et al.
(2000)
Rat/Harlan
Sprague-
Dawley
Corn oil
gavage (GDs
10-16)
F (F0)
80-88 (Fl)
0, 25, or 100
None
25
Developmental
effects
Decreased preference
in the consumption
of 0.25% saccharin
solution (Fl)
Bell et al. (2007b)
Rat/CRL:WI
(Han)
Maternal
daily dietary
exposure for
an estimated
20 weeks
(12 weeks
prior to
mating
through
parturition)
F (F0)
M (Fl)
65 (F0
treatments)
75 (F0
controls) at
study
initiation;
following
interim
sacrifice
~30 animals
were allowed
to litter; Fl
onPND 21
was ~7
0, 2.4, 8, or 46
None
2.4
(maternal
exposure)
Reproductive
and
developmental
effects
Delayed BPS (Fl)
Franczak et al.
(2006)
Rat/Sprague-
Dawley
Maternal corn
oil gavage
(GDs 14 and
21; PNDs 7
and 14)
Offspring
corn oil
gavage
(weekly for
8 months)
F (F0 and
Fl)
2 or 3 (F0)
7 (Fl)
0, 7.14, or 28.6
None
7.14
Developmental
effects
Decreased serum
estradiol levels (Fl)
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
to
G\
a
&
a
o
2
o
H
O
HH
H
W
O
V
o
c
o
H
W
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Developmental toxicity studies (continued)
Hoio et al. (2002)
and related
Zareba et al.
(2002)
Rat/Sprague-
Dawley
Maternal
single corn oil
gavage
(GD 8)
Offspring
exposed
during
gestation and
lactation
(35 days)
F (F0)
F andM
(Fl)
12 (F0)
50 or 60 (Fl)
0, 20, 60, or
180
None
20
(maternal
exposure)
Developmental
effects
Abrogation of
sexually dimorphic
neuro-behavioral
responses (Fl)
Kattainen et al.
(2001)
Rat/
Han/Wistar
and Long-
Evans
Maternal
single corn oil
gavage
(GD 15)
F (F0)
F andM
(Fl)
4 to 8 (F0)
3F/3M per
treatment
group (Fl)
0, 30, 100,
300, or 1,000
None
30
(maternal
exposure)
Developmental
effects
Reduced mesiodistal
length of the lower
third molar (Fl)
Keller et al.
(2008a: 2008b:
2007)
Mouse/
C57BL/6J,
BALB/cByJ,
A/J, CBA/J,
C3H/HeJ, and
C57BL/10J
Maternal
single corn oil
gavage
(GD 13)
F (F0)
F andM
(Fla, b, c)
Dams not
specified
(F0);
23-36 (Fla);
4-5 (Fib);
107-110
(Flc)
0, 10, 100, or
1,000
None
10
(maternal
exposure)
Developmental
effects
Variation in Ml
morphology in
C57BL/10J males
and females (Fla);
decreased mandible
shape and size in
C3H/HeJ males
(Fib); variation in
molar shape in
C3H/HeJ males
(Flc)
(2008a: 2008b: 2007)
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
to
-J
o
o
o
2
o
H
o
HH
H
W
O
o
c
o
H
W
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Developmental toxicity studies (continued)
Kuchiiwa et al.
(2002)
Mouse/ddY
Maternal
olive oil
gavage
(weekly for
8 weeks prior
to mating)
F (F0)
M (Fl)
1 (F0)
3 (Fl
immuno-
cytochemical
analysis)
6 (Fl cell
number
count)
0, 0.7, or 70
None
0.7
(LOEL)
(maternal
exposure)
Neurotoxicity
Decreased serotonin-
immunoreactive
neurons in raphe
nuclei of male
offspring (Fl)
Li et al. (2006)
Mouse/NIH
(pregnant and
pseudo-
pregnant)
Maternal
sesame oil
gavage daily
for 8 days
(GDs 1-8)
F
10
0, 2, 50, or 100
None
2
Developmental
effects
Decreased
progesterone and
increased serum
estradiol levels
Markowski et al.
(2001)
Rat/Holtzman
Maternal
single olive
oil gavage
(GD 18)
F (F0 and
Fl)
4-7 (F0 and
Fl)
0, 20, 60, or
180
None
20
(maternal
exposure)
Behavioral
effects
Decreased training
responses (Fl)
Miettinen et al.
(2006)
Rat/Line C
Maternal
single corn oil
gavage
(GD 15)
F (F0)
F andM
(Fl)
24-32
(treatment)
12-48
(controls)
0, 30, 100,
300, or 1,000
None
30
(maternal
exposure)
Developmental
effects
Increase in dental
caries (Fl)
Nohara et al.
(2000)
Rat/
Holtzman
Maternal
single corn oil
gavage
(GD 15)
F (F0)
M (Fl)
Not specified
(F0)
5 males and
3 females
(Fl)
0, 12.5, 50,
200, or 800
800
(maternal
exposure)
None
Immunotoxicity
Decreased spleen
cellularity (Fl)
Ohsako et al.
(2001)
Rat/
Holtzman
Maternal
single corn oil
gavage
(GD 15)
F (F0)
M (Fl)
6 (F0)
5 males and
3 females
(Fl)
0, 12.5, 50,
200, or 800
12.5
(maternal
exposure)
50
(maternal
exposure)
Developmental
effects
Decreased anogenital
distance (Fl)
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
to
00
o
o
o
2
o
H
o
HH
H
W
O
o
c
o
H
W
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Developmental toxicity studies (continued)
Schantz et al.
(1996)
Rat/Harlan
Sprague-
Dawley
Maternal corn
oil gavage
(GDs 10-16
F(F0)
~4 (FO);
80-88 (Fl)
0, 25, or 100
None
None
Developmental
effects
Facilitatory effect on
radial arm maze
learning (Fl)
Seo et al. Q995N)
Rat/Sprague-
Dawley
Maternal corn
oil gavage
(GDs 10-16)
F andM
(Fl)
-15 (F0);
5-9 (Fl)
0, 25, or 100
25
100
Developmental
effects
Decreased thymus
weight
Simanainen et al.
(2004a)
Rat/TCDD-
resistant
Han/Wistar
bred with
TCDD-
sensitive
Long-Evans
Maternal corn
oil gavage
(GDs 15)
F (FO)
M (Fl)
5-8 (F0)
0, 30, 100,
300, or 1,000
100
300
Reproductive
effects
Reduction in daily
sperm production
and cauda
epididymal sperm
reserves
Sparschu et al.
(1971)
Rat/Sprague-
Dawley
Maternal corn
oil gavage
(GDs 6-15)
F (FO)
31 (controls)
10-14 (F0)
0, 30, 125,
500, 2,000, or
8,000
50
125
Maternal
toxicity;
Developmental
effects
Decreased body
weight in dams and
male fetuses; fetal
intestinal hemorrhage
and subcutaneous
edema
Smith et al.
(1976)
Mouse/CF-1
Maternal corn
oil gavage
(GDs 6-15)
F (FO)
14-41 (F0)
0, 1.0, 10, 100,
1,000, or 3,000
1,000
(maternal)
100
(fetal)
3,000
(maternal)
1,000
(fetal)
Teratogenic and
developmental
effects
Increased relative
liver weight (F0
dams); increased
incidence of cleft
palate (fetuses)
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
to
a
&
a
o
2
o
H
O
HH
H
W
O
V
o
c
o
H
W
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Developmental toxicity studies (continued)
Sugita-Konishi et
al. (2003)
Mouse/C57/6
NCji
Maternal
drinking
water
exposure
(daily for
17-day
lactational
period)
F (F0)
F andM
(Fl)
8 (F0)
Not specified
(Fl)
0, 1.14, or 11.3
1.14
(NOEL)
(maternal
exposure)
11.3
(LOEL)
(maternal
exposure)
Immunotoxicity
Increased
susceptibility to
Listeria (Fl males
and females);
increase in thymic
CD4+ cells
(Fl males);
decreased spleen
weight (Fl males)
Acute toxicity studies
Burleson et al.
(1996)
Mouse/B6C3
Fi
Corn oil
gavage
(single
exposure)
F
20
0, 1, 5, 10, 50,
100, or 6,000
5
10
Immunotoxicity
Increased mortality
from influenza
infection 7 days after
a single TCDD
exposure
Crofton et al.
(2005)
Rat/Long-
Evans
Corn oil
gavage
(4 consecutiv
e days)
F
14, 6, 12, 6,
6, 6, 6, 6, 6,
and 4,
respectively,
in control
and treated
groups
0, 0.1,3, 10,
30, 100, 300,
1,000, 3,000,
or 10,000
30
100
Thyroid effects
Reduction in serum
T4 levels
Kitchin and
Woods (1979)
Rat/Sprague-
Dawley
Corn oil
gavage
(single dose)
F
4 (treated);
9 (control)
0, 0.6, 2, 4, 20,
60, 200, 600,
2,000, 5,000,
or 20,000
0.6
(NOEL)
2
(LOEL)
Enzyme
induction
Increased
benzo(a)pyrene
hydroxylase (BPH)
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
to
o
O
O
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Acute toxicity studies (continued)
Lietal. (1997)
Rat/Sprague-
Dawley
Corn oil dose
via oral
gastric
intubation
(single dose)
F
10
0, 3, 10, 30,
100, 300,
1,000, 3,000,
10,000, or
30,000
3
10
Hormonal
effects
Increased serum FSH
(1997)
Lucier et al.
(1986)
Rat/Sprague-
Dawley
Corn oil
gavage or
TCDD-
contaminated
soil (single
dose)
F
6
0, 15, 40, 100,
200, 500,
1,000, 2,000,
or 5,000 in
corn oil
0, 15, 44, 100,
220, 500,
1,100, 2,000,
or 5,500 in
contaminated
soil
None
15
(LOEL)
Enzyme
induction
Induction of aryl
hydrocarbon
hydroxylase (at low
dose in both
treatment protocols)
Nohara et al.
(2002)
Mouse/
B6C3Fi ,
BALB/c,
C57BL/6N
and DBA2
Corn oil
gavage
(single dose)
M,F
10-40
0, 5, 20, 100,
or 500
500
None
Mortality and
body-weight
changes
No increased
mortality of virus-
infected mice or
treatment-related
changes in body
weight
Simanainen et al.
(2002)
Rat/TCDD-
resistant
Han/Wistar
bred; TCDD-
sensitive
Long-Evans
Corn oil
gavage
(single dose)
M, F
9-11
30-100,000
100
300
General
toxicological
endpoints, organ
weights, dental
defects
Reduction in serum
T4 levels
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Acute toxicity studies (continued)
Simanainen et al.
(2003)
Rat/TCDD-
resistant
Han/Wistar
bred with
TCDD-
sensitive
Long-Evans
Corn oil
gavage
(single dose)
M, F
5-6
Line A:
30-3,000,000
Line B:
30-1,000,000
Line C:
30-100,000
100
300
General
toxicological
endpoints, organ
weights, dental
defects
Decreased thymus
weight
Smialowicz et al.
(2004)
Mouse/
C57BL/6N
CYP1A2
(+/+) wild-
type
Corn oil
gavage
(single dose)
F
Not specified
0, 30, 100,
300, 1,000,
3,000, or
10,000
300
1,000
Immunotoxicity
Decreased antibody
response to SRBCs
Vanden Heuvel et
al. (1994)
Rat/Sprague-
Dawley
Corn oil
gavage
(single dose)
F
5-15
0, 0.05,0.1, 1,
10, 100, 1,000,
or 10,000
0.1
(NOEL)
1
(LOEL)
Liver effects
Increase in hepatic
EROD activity and
CYP1A1 mRNA
levels
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
to
to
o
O
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Acute toxicity studies (continued)
Weber et al.
(1995)
Inbred
Mouse/
C57BL/6
Corn oil
gavage
(single dose
on Day 0)
Sacrificed on
Day 8
M
4-7
0, 30, 100,
300, 1,000,
3,000, 9,400,
37,500,
75,000,
100,000,
133,00, or
235,000
1,000
3,000
Hepatic and
renal enzyme
and hormone
alterations; liver
and kidney
weight
Increased relative
liver weight
Inbred
Mouse/
DBA/2
Corn oil
gavage
(two doses on
Days -1 and
0) Sacrificed
on Day 8
M
4-7
0, 1,000,
10,000,
97,500,
375,000,
1,500,000,
1,950,000, or
3,295,000
10,000
97,500
Subchronic toxicity studies
Chu et al. (2001)
Rat/Sprague-
Dawley
Corn oil
gavage (daily
for 28 days)
F
5
0, 2.5, 25, 250,
or 1,000
250
1,000
Body- and
organ-weight
changes
Decreased body
weight, increased
relative liver weight
and related
biochemical changes,
decreased relative
thymus weight
Chu et al. (2007)
Rat/Sprague-
Dawley
Corn oil
gavage (daily
for 28 days)
F
5
0, 2.5, 25, 250,
or 1,000
2.5
25
Liver effects
Alterations in
thyroid, thymus, and
liver histopathology
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
to
o
o
o
2
o
H
o
HH
H
W
O
o
c
o
H
W
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Subchronic toxicity studies (continued)
DeCaprio et al.
(1986)
Guinea pig/
Hartley
Daily dietary
exposure
(90 days)
M, F
10/sex
0, 0.12,0.61,
4.9, or 26
(males); 0,
0.12,0.68,
4.86, or 31
(females)
0.61
4.9
Body- and
organ-weight
changes
Decreased body
weight (male and
females); increased
relative liver weights
(males); decreased
relative thymus
weight (males)
DeVito et al.
(1994)
Mice/B6C3Fi
Corn oil
gavage
(5 days/week
for 13 weeks)
F
5
0, 1.07, 3.21,
10.7, 32.1, or
107
None
1.07
(LOEL)
Body- and
organ-weight
changes;
enzyme
induction
Increased EROD,
ACOH and
phosphotyrosyl
proteins at all doses
Fattore et al.
(2000)
Rat/Iva:SIV
50-Sprague-
Dawley
Daily dietary
exposure
(13 weeks)
M, F
6
0, 20, 200, or
2,000
None
20
Liver effects
Reduced hepatic
vitamin A levels
Daily dietary
exposure
(13 weeks)
M, F
6
0 or 200
Daily dietary
exposure
(13 weeks)
M, F
6
0, 200, or
1,000
Daily dietary
exposure
(13 weeks,
26, and
39 weeks)
F
6
0 or 100
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
to
O
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Subchronic toxicity studies (continued)
Fox et al. (1993s)
Rat/Sprague-
Dawley
Gavage
loading/
maintenance
doses (every
4 days for
14 days)
M, F
6
0, 0.55, 307, or
1,607
0.57
327
Body- and liver-
weight changes;
hepatic cell
proliferation
Increased absolute
and relative liver
weight
Hassoun et al.
(1998)
Mouse/
B6C3Fi
Corn oil
gavage
(5 days/week
for 13 weeks)
F
Not
specified
0, 0.32, 1.07,
10.7, or 107
None
0.32
(LOEL)
Brain effects
Induction of
biomarkers of
oxidative stress at all
doses
Hassoun et al.
(2000)
Rat/Harlan
Sprague-
Dawley
Corn oil
gavage
(5 days/week
for 13 weeks)
F
6
0, 2.14,7.14,
15.7, 32.9, or
71.4
None
2.14
(LOEL)
Liver and brain
effects
Induction of
biomarkers of
oxidative stress at all
doses in liver and
brain
Hassoun et al.
(2003)
Rat/Harlan
Sprague-
Dawley
Corn oil
gavage
(5 days/week
for 13 weeks)
F
12
0, 7.14, 15.7,
or 32.9
None
7.14
(LOEL)
Brain effects
Induction of
biomarkers of
oxidative stress at all
doses
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
to
O
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Subchronic toxicity studies (continued)
Kociba et al.
(1976)
Rat/Sprague-
Dawley
Corn oil
gavage
(5 days/week
for 13 weeks)
M, F
12
0, 0.71,7.14,
71.4, or 714
7.14
71.4
Liver effects,
body-weight
changes, and
hematologic and
clinical effects
Reduced body
weight and food
consumption, slight
liver degeneration,
lymphoid depletion,
increased urinary
porphyrins and delta
aminolevulinic acid,
increased serum
alkaline phosphatase
and bilirubin
Mally and
Chioman (2002)
Rat/F344
Corn oil
gavage
(2 days/week
for 28 days)
F
3
0, 0.71,7.14,
or71.4
None
0.71
(LOEL)
Clinical signs
and
histopathology
Decreased Cx32
plaque number and
area in the liver
Slezak et al.
(2000)
Mouse/
B6C3FJ
Corn oil
gavage
(5 days/week
for 13 weeks)
F
Not specified
0, 0.11,0.32,
1.07, 10.7, or
107.14
1.07
(NOEL)
10.7
(LOEL)
Liver, lung,
kidney, and
spleen effects
Increased hepatic
superoxide anion
Smialowicz et al.
(2008)
Mouse/
B6C3Fi
Corn oil
gavage
(5 days/week
for 13 weeks)
F
8-15
0, 1.07, 10.7,
107, or 321
None
1.07
Immunotoxicity
and organ
weight
Reduced antibody
response to SRBC,
increased relative
liver weight
Van Birgelen
et al. (1995a:
1995b)
Rat/Sprague-
Dawley
TCDD in diet
(13 weeks)
F
8
0, 14, 26, 47,
320, or 1,024
None
14
Multiple end-
points
Decreased absolute
and relative thymus
weights, decreased
liver retinoid levels
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Subchronic toxicity studies (continued)
Vosetal. (1973)
Guinea pig/
Hartley
Corn oil
gavage
(weekly for
8 weeks)
F
10
0, 1.14, 5.71,
28.6, or 143
1.14
5.71
Immunotoxicity
Decreased total
leukocytes and
lymphocyte count,
decreased absolute
thymus and weight,
increase in primary
serum tetanus
antitoxin
to
White et al.
(1986)
Mouse/
B6C3Fi
Corn oil
gavage (daily
for 14 days)
F
6-8
0, 10, 50, 100,
500, 1,000, or
2,000
None
10
Immunotoxicity
Reduction of serum
complement activity
On
Chronic toxicity studies
O
Cantoni et al.
(1981)
Rat/CD-
COBS
Corn oil
gavage
(weekly for
45 weeks)
F
4
0, 1.43, 14.3,
or 143
None
1.43
Hepatic
porphyria
Increased urinary
porphyrin excretion
V
fe
H
O
o
2
o
H
Croutch et al.
(2005)
Rat/Sprague-
Dawley
Loading/
maintenance
dose (every
3 days for
different
durations up to
128 days)
F
5
0, 0.85, 3.4,
13.6, 54.3, or
217
(28-day
duration)
54.3
(28-day
duration)
217
(28-day
duration)
Body-weight
changes and
changes in
PEPCK activity
and IGF-I levels
Decreased body
weight, decreased
PEPCK activity, and
reduced IGF-I levels
o
HH
H
W
O
Hassoun et al.
(2002)
Rat/Sprague-
Dawley
Corn oil
gavage
(5 days/week
for 30 weeks)
F
6
0, 2.14,7.14,
15.7, 32.9, or
71.4
None
2.14
(LOEL)
Brain effects
Induction of
biomarkers of
oxidative stress at all
doses
O
c
o
H
ffl
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
to
-J
O
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Chronic toxicity studies (continued)
Hons et al. (1989N)
Rhesus
monkeys.
Daily dietary
(4 years)
F
7-8
0, 0.12, or
0.67
None
None
Immunotoxic
effects
None
Kociba et al.
(1978)
Rat/Sprague-
Dawley
Daily dietary
exposure
(2 years)
M, F
50
0, 1, 10, or 100
1
10
Multiple
endpoints
measured
Increased urinary
porphyrins,
hepatocellular
nodules, and focal
alveolar hyperplasia
Maronpot et al.
(1993)
Rat/Sprague-
Dawley
Biweekly
gavage
(30 weeks)
F
9
0, 3.5, 10.7,
35, or 125
10.7
35
Body- and
organ-weight
changes, clinical
chemistry,
hepatocellular
proliferation
Increased relative
liver weight
NTP (1982c)
Mouse/
B6C3Fj;
Rat/Osborne
Mendel
Corn oil
gavage
(2 days/week
for 104 weeks)
M,F
50
0, 1.4, 7.1, or
71 for rats and
male mice; 0,
5.7, 28.6, or
286 for female
mice
None
1.4
Liver and body-
weight changes
Increased incidences
of liver lesions in
mice (males and
females)
NTP (2006a)
Rat/Sprague-
Dawley
Corn oil
gavage
(5 days/week
for 105 weeks)
F
53
0, 2.14,7.14,
15.7, 32.9, or
71.4
None
2.14
Liver and lung
effects
Increased absolute
and relative liver
weights, increased
incidence of
hepatocellular
hypertrophy,
increased incidence
of alveolar to
bronchiolar epithelial
metaplasia
-------
Table 2-4. Animal bioassay studies considered for noncancer dose-response modeling (continued)
Reference
Species/
strain
Exposure
protocol
Sex
(exposure
group)
n
Average daily
dose levels
(ng/kg-day)
NOAEL
(ng/kg-day)
LOAEL
(ng/kg-day)
Endpoint(s)
examined
LOAEL/NOAEL
Endpoint(s)
Chronic toxicity studies (continued)
Sewall et al.
(1993)
Rat/Sprague-
Dawley
Biweekly
gavage
(30 weeks)
F
9
0, 3.5, 10.7,
35, or 125
None
3.5
(LOEL)
EGFR kinetics
and auto-
phosphorylation,
hepatocellular
proliferation
Decrease in EGFR
maximum binding
capacity
Sewall et al.
(1995)
Rat/Sprague-
Dawley
Biweekly
gavage
(30 weeks)
F
9
0, 0.1,0.35, 1,
3.5, 10.7,35,
or 125
10.7
35
Thyroid
function
Decreased serum T4
levels
Toth et al. (1979)
Mouse/Swis
s/ H/Riop
Sunflower oil
gavage
(weekly for
1 year)
M
38-44
0, 1, 100, or
1,000
None
1
Skin effects
Dermal amyloidosis
and skin lesions
Tritscher et al.
(1992)
Rat/Sprague-
Dawley
Initiated with
i.p. injection of
diethylnitrosa
mine (175
mg/kg) or
saline,
followed 2
weeks later by
biweekly
TCDD in corn
oil gavage (30
weeks)
F
At least 9 per
group
3.5, 10.7,35.7,
or 125
None
None
CYP induction
None
H
ffl
O
V
o
c
o
H
W
ND = not determined.
-------
Criteria
met?
No
Yes
Studies excluded
from quantitative
dose-response
analysis of TCDD
Studies Included in final list of
key cancer and noncancer
studies for quantitative dose-
response analysis of TCDD
Final development of two sets of TCDD study inclusion criteria:
For in vivo mammalian bioassays
For epidemiologic studies
Initial TCDD-specific study inclusion criteria
development for in vivo mammalian bioassays
Final literature collection (October, 2009)
Federal Register Notice; Web publication of literature
search for public comment and submissions
Dioxin workshop (2009) and expert refinement of
TCDD study inclusion criteria for in vivo mammalian bioassays
Literature search for in vivo mammalian bioassays and
epidemiologic TCDD studies (2000-2008)
Studies screened using TCDD study inclusion criteria:
Studies cited in 2003 Reassessment
Studies identified via literature search results
Studies submitted by the public
Studies collected by EPA in 2009
Figure 2-1. EPA's process to select and identify in vivo mammalian and
epidemiologic studies for use in the dose-response analysis of TCDD.
EPA first conducted a literature search to identify studies published since the 2003 Reassessment.
Results were published, and additional study submissions were accepted from the public. Next,
EPA developed TCDD-specific study inclusion criteria for in vivo mammalian studies and held a
Dioxin Workshop where these criteria were discussed and refined. Third, EPA developed
two final sets of study inclusion criteria, one for in vivo mammalian studies and another for
epidemiologic studies. Finally, EPA applied these two sets of criteria to all studies from the
literature search public submissions, 2003 Reassessment, and additional studies identified by EPA
after the Dioxin Workshop through October 2009. The studies that met these criteria formed a list
of key studies for EPA's consideration in TCDD dose-response assessment.
2-49
DRAFT - DO NOT CITE OR QUOTE
-------
No
Study
in peer-reviewed
literature?
Yes
No
Yes
Long-term
exposures and
latency information
available forcancer
xjassessment?/^
Exposure
windows and
latency information
available forRfD
\sassessment?_^
No
No
Yes
Yes
No
Considerations
adequately
. satisfied? .
Yes
^ Exposure
primarily to TCDD
and quantifiable?
Study excluded
from TCDD
dose-response
assessment
List of available epidemiologic studies on TCDD and DLCs
(All studies summarized.)
Key study included
forTCDD cancerand/or noncancer
dose-response assessment
Evaluate study using five considerations:
• Methods used to ascertain health outcomes are clear and unbiased?
• Confounding and other potential sources of bias are addressed?
• Association/exposure response between TCDD and adverse effect?
• Exposures based on individual-level estimates, uncertainties described?
• Statistical precision, power and study follow-up are sufficient?
Figure 2-2. EPA's selection process to evaluate available epidemiologic
studies using study inclusion criteria and other epidemiologic considerations
for use in the dose-response analysis of TCDD.
EPA applied its TCDD-specific epidemiologic study inclusion criteria to all studies published on
TCDD and DLCs. For all peer-reviewed studies, EPA examined whether the exposures were
primarily to TCDD and if the TCDD exposures could be quantified so that dose-response analyses
could be conducted. Then, EPA required that the effective dose and oral exposure be estimable:
(1) for cancer, information is required on long-term exposures, (2) for noncancer, information is
required regarding the appropriate window of exposure that is relevant for a specific, nonfatal
health endpoint, and (3) for all endpoints, the latency period between TCDD exposure and the
onset of the health endpoint is needed. Finally, studies were evaluated using five considerations
regarded as providing the most relevant kind of information needed for quantitative human health
risk analyses. Only studies meeting these criteria and adequately satisfying the considerations
were included in EPA's TCDD dose-response analysis.
2-50
DRAFT - DO NOT CITE OR QUOTE
-------
Study
in peer-reviewed
literature?
No
Yes
Study on a
genetically-altered
^ species? .
Yes
No
No
No
Yes
Yes
No
Yes
Lowest dose
tested for n on cancer
endpoint<30
ng/kg-day?
^ Lowest
dose tested for
cancer endpoint<1
v |jg/kg-day? .
^ Oral ^
exposure to TCDD
only?
Study excluded
from TCDD
dose-response
assessment
List of available in vivo mammalian bioassay studies on TCDD
Study summarized; evaluated for
quality and to note adequacy
of data needed for TCDD
dose-response assessment.
Key study included
forTCDD cancer and/or noncancer
dose-response assessment
Figure 2-3. EPA's process to evaluate available animal bioassay studies
using study inclusion criteria for use in the dose-response analysis of TCDD.
EPA evaluated all available in vivo mammalian bioassay studies on TCDD. Studies had to be
published in the peer-reviewed literature. Studies on genetically-altered species were excluded as
their relevance to human health is not known. Next, EPA applied dose requirements to each
study's lowest tested average daily dose, with requirements for cancer (<1 (ig/kg-day) and
noncancer (<30 ng/kg-day) studies. EPA also required that the animals were exposed via the oral
route to only TCDD. Finally, the studies were evaluated for quality and summarized to ensure
providing the most relevant information for quantitative human health risk analyses. Only studies
meeting all of the criteria were included in EPA's TCDD dose-response analysis.
2-51
DRAFT - DO NOT CITE OR QUOTE
-------
Cancer
bioassays
Epi cancer
studies
Epi n on cancer
studies
Noncancer
bioassays
751
cancer
studies
Included
Animal
cancer
bioassays
Included
noncancer
studies
included
Animal
noncancer
bioassays
included
Epidemiologic (Epi) stu dies
Animal bioassays
755
Wrong study type for quantitative
TCDD dose-response analysis:
637 excluded
Rightstudy type for quantitative
TCDD dose-response analysis:
804 considered further
Studies from literature search and data collection activities
1,441
TDCC only (142)
Genetically-
altered (66)
Dose limits
(370)
Failed > 1 of:
Peer-review (0)
Non-oral (135)
TDCC only (0)
Genetically-
altered (1)
Failed > 1 of:
Peer-review (0)
Dose limits
Non-oral (1)
Primarily TDCC
(10)
Effective
exposure
estimable (11 )
Considerations
Failed > 1 of:
Peer-review (0)
Effective
exposure
estimable (26)
Considerations
Failed > 1 of:
Peer-review (1)
Primarily TDCC
indicates those studies that passed all three criteria but were not selected based on study considerations.
Figure 2-4. Results of EPA's process to select and identify in vivo
mammalian and epidemiologic studies for use in the dose-response analysis
of TCDD.
Criteria not met are not mutually exclusive. Four animal studies and seven epidemiologic studies
contained both cancer and noncancer endpoints. Two epidemiologic cancer studies, Steenland
et al. (1999) and Flesch-Janys et al. (1998). passed all criteria, but were still not selected because
they were superseded by other studies on the same cohort for which an improved analysis was
done. One noncancer epidemiologic study, Baccarelli et al. (2005). passed all criteria, but was
excluded because the health endpoint, chloracne, is considered to be an outcome associated with
high TCDD exposures.
2-52
DRAFT - DO NOT CITE OR QUOTE
-------
3. rill USE OF TOXICOKINETICS IN THE DOSE-RESPONSE MODELING FOR
CANCER AND NONCANCER ENDPOINTS
A key recommendation from the National Academy of Sciences (NAS) for improving the
2003 Reassessment was that U.S. Environmental Protection Agency (EPA) should justify its
approaches to dose-response modeling for cancer and noncancer endpoints. Further, the NAS
suggested that EPA incorporate the most up-to-date and relevant state of the science for the
2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) dose-response assessment.
While EPA believes that at the time of its release, the 2003 Reassessment offered a
substantial improvement over the general state-of-the-science regarding dose-response modeling,
EPA agrees with the NAS that the justification of the approaches to dose-response modeling can
be improved and the methodologies updated to reflect the most current EPA guidance (see Text
Box 2-1) and science. In Section 3, EPA describes the use of toxicokinetic (TK)1 information in
the dose-response modeling of TCDD. Section 3.1 summarizes the NAS comments regarding
the use of TK in the dose-response approaches for TCDD. Section 3.2 overviews EPA's
responses to the NAS comments. Section 3.3 discusses TCDD kinetics, including TK models
developed to simulate disposition of this compound in rodents and humans (see Section 3.3.4),
alternative measures of dose that could be used in a TCDD dose-response analysis, and
uncertainties in the TCDD dose estimates (see Section 3.3.5). Section 4 of this document
incorporates the TK information into noncancer dose-response modeling.
3.1. SUMMARY OF NAS COMMENTS ON THE USE OF TOXICOKINETICS IN
DOSE-RESPONSE MODELING APPROACHES FOR TCDD
The NAS commented on the appropriate use of TK models in dose-response modeling
for TCDD. Specifically, the committee requested that EPA consider using such models to
provide refined estimates of dose, for example, as the underlying science and predictive
capabilities of these models improved.
toxicokinetics (TK) is the branch of the pharmacokinetics (PK) that examines the disposition of toxins and
toxicants.
3-1 DRAFT - DO NOT CITE OR QUOTE
-------
[Discussing Kinetic models],. .the committee encourages further development and
use of these models as data become available to validate and further develop them
(NAS. 2006b. p. 59).
Although the NAS agreed with EPA's use of body burden as a dose metric in the 2003
Reassessment (e.g.. see NAS. 2006b. p. 7). the NAS was concerned about the limitations of
first-order kinetic models, such as the one used in the 2003 Reassessment, to estimate TCDD
body burdens.
TCDD, other dioxins, and DLCs act as potent inducers of CYP, a property that
can affect both the hepatic sequestration of these compounds and their half-lives.
Hepatic sequestration of dioxin may influence the quantitative extrapolation of the
rodent liver tumor results because the body-burden distribution pattern in highly
dosed rats would differ from the corresponding distribution in humans subject to
background levels of exposure. EPA should consider the possible quantitative
influence of dose-dependent toxicokinetics on the interpretation of animal
toxicological data (NAS. 2006b. p. 129).
The NAS also asked EPA to evaluate the impact of kinetic uncertainty and variability on
dose-response assessment. The NAS committee asked EPA to use TK models to examine both
interspecies and human interindividual differences in the disposition of TCDD, which would
better justify EPA dose-response modeling choices.
The Reassessment does not adequately consider the use of a PBPK model to
define species differences in tissue distribution in relation to total body burden for
either cancer or noncancer end points (NAS. 2006b. p. 62).
EPA ... should consider physiologically based pharmacokinetic modeling as a
means to adjust for differences in body fat composition and for other differences
between rodents and humans (NAS. 2006b, p. 10).
The Reassessment does not provide details about the magnitudes of the various
uncertainties surrounding the decisions EPA makes in relation to dose metrics
(e.g., the impact of species differences in percentage of body fat on the
steady-state concentrations present in nonadipose tissues). The committee
recommends that EPA use simple PBPK models to define the magnitude of any
differences between humans and rodents in the relationship between total body
burden at steady-state concentrations (as calculated from the intake, half-life,
bioavailability) and tissue concentrations. The same model could be used to
explore human variability in kinetics in relation to elimination half-life. EPA
3-2
DRAFT - DO NOT CITE OR QUOTE
-------
should modify the estimated human equivalent intakes when necessary (NAS.
2006b. p.731
Finally, the NAS asked EPA to use TK considerations to better justify its choice of dose
metric.
EPA makes a number of assumptions about the appropriate dose metric and
mathematical functions to use in the Reassessment's dose-response analysis but
does not adequately comment on the extent to which each of these assumptions
could affect the resulting risk estimates.. .EPA did not quantitatively describe how
this particular selection affected its estimates of exposure and therefore provided
no overall quantitative perspective on the relative importance of the selection
(NAS. 2006b. p. 51).
3.2. OVERVIEW OF EPA'S RESPONSE TO THE NAS COMMENTS ON THE USE OF
TOXICOKINETICS IN DOSE-RESPONSE MODELING APPROACHES FOR
TCDD
In response to the NAS recommendations regarding TCDD kinetics and choice of dose
metrics, this document presents an in-depth evaluation of TCDD TK models, exploring their
differences and commonalities and their possible application for the derivation of dose metrics
relevant to TCDD. Initially, EPA discusses the application of first-order kinetics to estimate
body burden as a dose metric for TCDD. This first-order kinetic model is used to predict TCDD
body burden for all of the studies identified as Key Studies (see Section 2.4); this model uses a
constant half-life to simulate the elimination of TCDD from the body. However, given the
observed data indicating early influence of cytochrome P450 1A2 (CYP1A2) induction and
binding to TCDD in the liver and later redistribution of TCDD to fat tissue, the use of a constant
half-life for TCDD clearance following long-term or chronic TCDD exposure is not biologically
supported. Therefore, using half-life estimates based on observed terminal steady state levels of
TCDD will not account for the possibility of an accelerated dose-dependent clearance of this
chemical during early stages following elevated TCDD exposures. The biological processes
leading to dose-dependent TCDD excretion are better described using physiologically based
pharmacokinetic (PBPK) models than by simple first-order kinetic models. Additionally, as part
of its preparation for developing this document, EPA evaluated recent TCDD kinetic studies as
NAS advocated. Although the NAS agreed with continued use of body burden metric as the
dose metric of choice, EPA believes that the state-of-the-practice has advanced sufficiently to
3-3 DRAFT - DO NOT CITE OR QUOTE
-------
justify the consideration of alternative dose metrics (other than administered dose) based on an
application of a physiologically based TK model.
EPA identified a number of advances in the overall scientific understanding of TCDD
disposition; many of these are documented in a summary discussion introducing the section on
TCDD kinetics (see Section 3.3). The increased understanding warranted an evaluation of
current kinetic modeling of TCDD to determine if the use of such models would improve the
dose-response assessment for TCDD. Justification of the final PBPK model choice is detailed in
Section 3.3. Through the choice of a published PBPK model to estimate dose metrics for dioxin,
EPA has addressed several of the NAS concerns. The PBPK model can be applied to estimate
dose metrics other than body burden that may be more directly related to response, e.g., tissue
levels, serum levels, blood concentrations, or dose metrics related to TCDD-protein receptor
binding. The selected PBPK model included an explicit description of physiological and
biochemical parameters; therefore, it can also provide an excellent tool for investigating
differences in species uptake and disposition of TCDD. One of the criteria used to select a
PBPK model for TCDD kinetics was the availability of both human and animal models so that
differences in species uptake and disposition of TCDD can be investigated. Additionally, the
PBPK model includes quantitative information that is suitable for addressing the impact of
physiological (e.g., body weight [BW] or fat tissue volume), or biochemical (e.g., induction of
CYP1A2) variability on overall risk of TCDD between species, in response to another area of
concern in the NAS report. The sensitivity analysis and uncertainty in dose metrics derived for
the health assessment of TCDD are also presented in Section 3.3. A detailed discussion on the
uncertainty in choice of PBPK model-driven dose metrics is also provided in Section 3.3.
3.3. PHARMACOKINETICS (PK) AND PK MODELING
3.3.1. PK Data and Models in TCDD Dose-Response Modeling: Overview and Scope
In general, the use of measures of internal dose in dose-response modeling is considered
to be superior to that of administered dose (or uptake) because the former is more closely related
to the response. The evaluation of internal dose, or dose metric, in exposed humans and other
animals is facilitated by an understanding of pharmacokinetics (i.e., absorption, distribution,
metabolism, and excretion). When measurements of internal dose (e.g., blood concentration,
tissue concentration) are not available in animals and humans, pharmacokinetic models can be
3-4
DRAFT - DO NOT CITE OR QUOTE
-------
used to estimate them. The available data on the pharmacokinetics of TCDD in animals and
humans have been reviewed (NAS. 2006b; U.S. EPA. 2003; van Birgelen and van den Berg.
2000V
It is evident based on these reviews and other analyses that three distinctive features of
TCDD play important roles in determining its pharmacokinetic behavior, as discussed below:
• TCDD is very highly lipophilic and thus is more soluble in fat or other relatively
nonpolar organic media than in water. The //-octanol/water partition coefficient is a
commonly used measure of lipophilicity equal to the equilibrium ratio of a substance's
concentration in //-octanol (a surrogate for biotic lipid) to the substance's concentration
in water (Leo et al.. 1971). For TCDD, this coefficient is on the order of 10,000,000 or
more (ATSDR. 1998). It follows that the solubility of TCDD in the body's lipid fraction,
i.e., the fatty portions of various tissues, including adipose, organs, and blood, is
extremely high.
• TCDD is very slowly metabolized compared to many other organic compounds, with an
elimination half-life in humans on the order of years following an initial period of
distribution in the body (Michalek and Pavuk. 2008; Carrier et al., 1995a). Most
laboratory animals used for toxicological testing tend to eliminate TCDD much more
quickly than humans, although even in animals, TCDD is eliminated much more slowly
than most other chemicals.
• TCDD induces binding proteins in the liver that have the effect of sequestering some
of the TCDD. The ability of TCDD to alter gene expression and the demonstration that
the induction of CYP1A2 is responsible for hepatic TCDD sequestration suggest that
both pharmacokinetic and pharmacodynamic events must be incorporated for a
quantitative description of TCDD disposition (Santostefano et al.. 1998). The induction
of these proteins implies that TCDD tends to be eliminated more rapidly in the early
years following short-term, high-level exposures than it is after those initial levels have
declined. Leung et al. (1988) and Andersen et al. (1993). in their PBPK modeling, have
taken into consideration the issue of liver protein binding. Recent efforts of
pharmacokinetic modeling have supported the concentration-dependent elimination of
TCDD in animals and humans (Emond et al.. 2006; Avlward et al.. 2005b).
Sections 3.3.2 and 3.3.3 present the salient features of TCDD pharmacokinetics in
animals and humans, with particular focus on mechanisms and data of relevance to interspecies
and intraspecies variability. Section 3.3.4 describes the various dose metrics for the
dose-response modeling of TCDD and the characteristics of pharmacokinetic models potentially
useful for estimating these metrics. Finally, Sections 3.3.5 and 3.3.6 summarize the results of
application of pharmacokinetic models to derive dose metrics as well as the uncertainty
3-5
DRAFT - DO NOT CITE OR QUOTE
-------
associated with the predictions of dose metrics used in dose-response modeling. Dose metrics
derived via PBPK modeling approaches are utilized in Section 4 of this document for noncancer
TCDD dose-response modeling.
3.3.2. PK of TCDD in Animals and Humans
3.3.2.1. Absorption and Bioavailability
When administered via the oral route in the dissolved form, TCDD appears to be well
absorbed. Animal studies indicate that oral exposure to TCDD in the diet or in an oil vehicle
results in the absorption of >50% of the administered dose (Olson et al.. 1980; Nolan et al..
1979). Human data from Poiger and Schlatter (1986) indicate that >87% of the oral dose (after
ingestion of 105 ng [3H]-2,3,7,8-TCDD [1.14 ng/kg BW] in 6 mL corn oil) was absorbed from
the gastrointestinal tract. Lakshmanan et al. (1986). investigating the oral absorption of TCDD,
suggested that it is absorbed primarily by the lymphatic route and transported predominantly by
chylomicrons.
Oral absorption is generally less efficient when TCDD is more tightly bound in soil
matrices. Based on experiments in miniature swine, Wittsiepe et al. (2007) reported an
approximately 70% reduction in bioavailability when TCDD was administered in the form of
contaminated soil, relative to TCDD after extraction from the same soil matrix with solvents.
Working with soil from the prominent contamination site at Times Beach, Missouri, Shu et al.
(1988) reported an oral bioavailability of approximately 43% based on experiments in rats.
Percent dose absorbed by the dermal route is reported to be less than the oral route, whereas
absorption of TCDD by the transpulmonary route appears to be efficient (Banks and Birnbaum.
1991) (see for example; Roy et al.. 2008; U.S. EPA. 2003; Diliberto et al.. 1996; Nessel et al..
1992; Banks et al.. 1990).
3.3.2.2. Distribution
TCDD in systemic circulation equilibrates and partitions into the tissues where it is then
accumulated, bound, or eliminated. Whereas the bulk of the body tissues are expected to
equilibrate in a matter of hours, the adipose tissue will approach equilibrium concentrations with
blood much more slowly. Consistent with these assertions, a number of experimental and
modeling studies in rats and humans have shown that TCDD has a large volume of distribution
3-6
DRAFT - DO NOT CITE OR QUOTE
-------
(Vd), i.e., the apparent volume in which it is distributed. The Vd corresponds to the volume of
blood plus the product of internal tissue volumes and the corresponding tissue:blood partition
coefficients. This parameter is a key determinant of the elimination rate of TCDD in exposed
organisms. The tissue:blood partition coefficients of TCDD, in turn, are determined by the
relative solubility of TCDD in tissue and blood components (including neutral lipids,
phospholipids, and water).
Column 1 in Table 3-1 presents the tissue:blood partition coefficients for TCDD (Emond
et al.. 2005; Wang et al.. 1997). Column 3 of this table lists the physical volume of each tissue,
scaled to a person weighing 60 kg. The last column shows the implications of the tissue volumes
and tissue:blood partition coefficients for the effective volumes of distribution for each tissue
and for the body as a whole. It can be seen that, purely on the basis of solubility space, the fat
should be expected to contain about 94% of the TCDD in the body, and that the body as a whole
behaves as if it is about 1,200 L in terms of blood-equivalents (i.e., approximately 22-fold larger
than its physical volume).
Maruyama et al. (2002) have published another set of tissue:blood partition coefficients
for TCDD and other dioxin congeners based in part on observations of tissue concentrations
measured in autopsy specimens from eight Japanese people without known unusual exposures to
TCDD. Their estimates of TCDD partition coefficients seem to be rather large and variable,
with a fat:blood value of 247 ± 78 (standard deviation [SD]), a livenblood value of 9.8 ± 5.7, and
a muscle:blood value of 18 ± 10.6. Depending on time of autopsy, tissue samples may not be an
accurate source of information on observed, in vivo partition coefficients because weight loss is
likely to occur pre and post mortem. In particular, a decline in the fat stores volume could lead
to an increased concentration of dioxin in fat in autopsy specimens relative to what would be
observed in vivo.
The calculations shown in Table 3-1 do not include the additional amount that will be
bound to induced proteins in the liver. That induction and binding will tend to increase the
contribution of the liver on the effective volume of distribution (Birnbaum. 1986).
It is also of interest to point out some basic implications of the data in Table 3-1 for the
expected rates of perfusion-mediated transfer of TCDD between blood and each of the
organ/tissues. The rate of loss from a tissue (occurring primarily via blood flow) and the
corresponding half-life can be calculated using the following equations:
3-7
DRAFT - DO NOT CITE OR QUOTE
-------
Rate constant for loss (hour"')
Blood flow (liters / hour)
(Eq. 3-1)
Tissue volume (liters) x Tissue / Blood Partition Coefficent
t1/2 for tissue perfusion loss
ln(2)
Rate constant for loss
(Eq. 3-2)
ln(2) x Tissue volume (liters) x Tissue/Blood Partition Coefficent
Blood flow (liters/hour)
Because TCDD is highly lipophilic, its concentration in the aqueous portion of the blood
is very small, and TCDD tends to partition from blood components into cellular membranes and
tissues, probably in large part via diffusion. As a result, full equilibrium concentrations of
TCDD are not attained by the end of the transit time through organs from the arterial to venous
blood. For organs in which this occurs, diffusion coefficients or "permeability factors" have
been estimated to assess the fractional attainment of equilibrium concentration that occurs by the
time the blood leaving each organ reaches the venous circulation. Table 3-2 presents the
permeability factors and implications for perfusion half-lives for TCDD, per Emond et al. (2006;
2005).
Despite the high lipid bioconcentration potential of TCDD, the adipose tissue does not
always have the highest concentration (Abraham et al.. 1988; Gever et al.. 1986; Poiger and
Schlatter. 1986). Further, the ratios of tissue:tissue concentrations of TCDD and related
compounds (e.g., the livenadipose ratio) may not remain constant during nonsteady-state
conditions. TCDD concentrations have been observed to decrease more rapidly in the liver than
in adipose tissue. For example, Abraham et al. (1988) found that the livenadipose tissue
concentration ratio in female Wistar rats exposed to a subcutaneous TCDD dose of 300 ng/kg
decreased from 10.3 at 1 day postexposure to 0.5 at 91 days postexposure. It should be noted
that even at a ratio of 0.5, the amount of TCDD in the liver is greater than that based on lipid
content of the tissue alone, consistent with the presence of hepatic TCDD-binding proteins. The
livenadipose tissue concentration ratio also was dose-dependent, such that the liver TCDD
burden increased from ~11% of the administered dose at low doses (i.e., 1-10 ng/kg) to -37% of
the dose at an exposure level of 300 ng/kg. The increase in TCDD levels in liver, accompanied
by a decrease in concentration in the adipose tissue, is a particular behavior to be considered in
3-8
DRAFT - DO NOT CITE OR QUOTE
-------
high-dose to low-dose extrapolations. This behavior is essentially a result of dose-dependent
hepatic processes, as described below.
3.3.2.3. Metabolism and Protein Binding
The metabolism of TCDD is slow, particularly in humans, and it is thought to be
mediated by the CYP1A2 enzyme that is inducible by TCDD (Weber et al.. 1997; Olson et al..
1994; Wendling et al.. 1990; Ramsey et al.. 1982). The low rate of metabolism in combination
with sequestration appear to account for the retention of TCDD in liver, and these processes
collectively contribute to the long half-life for elimination of TCDD from the body.
Dynamic changes in TCDD binding in liver and partitioning to adipose tissues have been
studied extensively in rats and mice (Diliberto et al.. 2001; Diliberto et al.. 1995). Figure 3-1
shows observations by Diliberto et al. (1995) of the ratio of liver concentrations to adipose tissue
concentrations for mice given doses spread over a 100-fold range and studied at four different
times following exposure. It can be seen that even for the lowest dose studied, the liver:adipose
concentration ratio is higher than would be expected based on the lipid contents of the tissues
(i.e., 0.06:1, corresponding to the ratio of human livenblood and adipose:blood partition
coefficients; see Table 3-1). Moreover, the relative concentration in the liver consistently rises
with dose, with the steepest rise observed during the first 2 weeks after dosing. If the
distribution of TCDD were governed solely by passive partitioning into adipose, there should be
no such change in relative concentrations with dose. However, data presented in Figure 3-1
illustrate that at longer time points, the ratio of TCDD in the liver to TCDD in adipose decreases,
indicating that a redistribution of the chemical occurs as time goes on for each applied dose. The
redistribution of TCDD tissue levels from liver to adipose with increasing time suggests that
binding of the chemical in the liver (including via induction of CYP1A2) is an important kinetic
consideration at early exposure points with relatively high applied doses.
Experiments with CYP1A2 "knock-out" mice (i.e., congenic strains differing in only a
single gene that is "knocked out" in one of the strains) indicate that the inducible binding of
TCDD is attributable to CYP1A2 (Diliberto et al.. 1999. 1997). As noted previously, this
enzyme is believed to make an important contribution to metabolism of TCDD. Given the
critical role of CYP1A2 induction in the kinetics of TCDD, dose-and time-dependent induction
of this protein in rats has been examined and modeled (Emond et al.. 2006. 2004; Santostefano et
3-9
DRAFT - DO NOT CITE OR QUOTE
-------
al.. 1998; Wang et al.. 1997). Accordingly, the amount of CYP1A2 in the liver can be computed
as the time-integrated product of inducible production and a simple first-order loss process
(Wang et al.. 1997):
dCYl\
dt
2A1 = S(t)K0 - K2CA2t (Eq. 3-3)
where CYPui is the concentration of the enzyme, K2 is the rate constant for the first-order loss,
CA2t is the concentration of CYP1A2 in the liver, K() is the basal rate of production of CYP1A2 in
the liver, and S(t) is a multiplicative stimulation factor for CYP1A2 production in the form of a
Hill-type function:
^nA2 (CAh-TCDD )
(ICA2?+(CAh_TCDDf
Sit) = 1 + ^'mV~ah-tcdd) h (Eq 3_4)
where ICa2 corresponds to the concentration of the aryl hydrocarbon (Ah)-TCDD complex at
which half of the maximum fold stimulation of CYP2A production is reached, and /?, the Hill
exponent, determines the curvature of the stimulation in relation to concentration of the
Ah-TCDD complex at relatively low doses. A value of 0.6 as the Hill exponent has been used by
Wang et al. (2000: 1997) and Emond et al. (2006: 2005; 2004). indicative of a negative
cooperation, i.e., the curve is convex-upward (supralinear), depicting a faster increase in the
low-dose region compared to a straight line. Additional parameters in this expression include
IriA2, the maximum fold increase in the CYP1A2 synthesis rate over the basal rate that can occur
at high levels of TCDD, and (Cah-tcdd), the concentration of TCDD bound to the aryl
hydrocarbon receptor (AhR). This concentration in turn depends on the concentration of TCDD
in the liver (( 7,,/), the concentration of the AhR (AhLl) in liver, and the dissociation constant for
the Ah-TCDD receptor complex, Kdm'-
AhTj x Cnf
C.UCDD = „ ' (Eq. 3-5)
K-DAh + CLif
3-10 DRAFT - DO NOT CITE OR QUOTE
-------
3.3.2.4. Elimination
Estimated elimination half-lives (i.e., the time taken for the concentration to be reduced
to one-half of its initial level) of TCDD range from 11 days in the hamster to 2,120 days in
humans (U.S. EPA. 2003). Hepatic metabolism and binding processes, fecal excretion, and
accumulation in adipose tissue collectively determine the dose-dependent elimination half-lives
in various species. Aylward et al. (2005a) depicted the relationship between the elimination rate
versus initial level of lipid-corrected TCDD in serum for 36 people (see Figure 3-2). Even
though this analysis was done using the initial TCDD level, rather than the geometric mean or
midpoint level in the decline for each person, it indicated a concentration-dependency of the
half-life and elimination of TCDD in exposed individuals.
3.3.2.5. Interspecies Differences and Similarities
Among the pharmacokinetic determinants of TCDD, some are known to vary markedly
among species whereas others are not characterized sufficiently in this regard. Overall, the
qualitative determinants of the body burden and elimination half-lives appear to be similar across
species. Based on empirical observations for TCDD as well as with other PCDFs, Carrier et al.
(1995a. b) argued that in rats, monkeys, and humans, the dose-dependent changes in the fraction
contained in liver and adipose tissue follow a similar pattern across species. The authors
suggested that the half-saturation body burden is around 100 ng/kg, and the plateau of liver dose
(as fraction of body burden) appears to occur around 1,000 ng/kg. Literature also indicates that
AhR is conserved phylogenetically (Harper et al.. 2002; Fuiii-Kurivama et al.. 1995; Nebert et
al.. 1991) and is present in mammalian species, including experimental animals and humans
(Okev et al.. 1994; Lorenzen and Okev. 1991; Manchester et al.. 1987; Roberts et al.. 1986;
Roberts et al.. 1985). These qualitative similarities in pharmacokinetic determinants and
outcome support the use of animal data to infer general patterns of the pharmacokinetic behavior
of TCDD in humans. However, quantitative differences in determinants, including
physiological, physicochemical, and biochemical, need to be taken into account. Even though
species-specific physiological parameters can be obtained from the literature, key data on
species-specific biochemical parameters (particularly binding constants, maximal capacity,
induction rates, and other parameters) are not available for humans at this time. However, these
can be inferred by using a pharmacokinetic model fit to in vivo data on the rate of TCDD
3-11
DRAFT - DO NOT CITE OR QUOTE
-------
elimination from specific compartments in humans (Emond et al.. 2006; Avlward et al.. 2005b;
Emond et al.. 2005; Emond et al.. 2004; Carrier et al.. 1995a. b).
3.3.3. PK of TCDD in Humans: Interindividual Variability
TCDD pharmacokinetics and tissue doses vary across the human population as a function
of the interindividual variability of the key kinetic determinants. Because the NAS comments
focused on health effects associated with chronic, lifetime exposure, the key kinetic determinants
for such exposures include clearance, binding, and temporal changes in volume of distribution.
When considering the interindividual variability in pharmacokinetics and dose metrics of TCDD,
it is important to recognize that the elevated lipid-corrected serum concentrations in highly
exposed persons are associated with greater elimination rates, probably due to greater degrees of
induction of CYP1A2 in the liver and possibly other related metabolic enzymes (Emond et al..
2006; Avlvvard et al.. 2005b; Abraham et al.. 2002; Grassman et al.. 2000).
The interindividual variability in adipose content is a critical parameter in
pharmacokinetic models given the characteristics of TCDD (see Section 3.3.2). Both metabolic
elimination and elimination via the GI tract depend on the fraction of TCDD in the body that is
available outside of adipose tissue. As body fat content rises, a smaller portion of the total body
TCDD will be contained in the relatively available fraction outside of the adipose tissue.
Because elimination of TCDD by both metabolism and fecal excretion depends on the small
proportion of TCDD that exists outside of fat tissue, people with larger proportions of body fat—
including many older people—will tend to require longer times to reduce TCDD levels by a
given proportion than leaner people (Emond et al.. 2006; Rohde et al.. 1999; Van der Molen et
al.. 1998; Van der Molen et al.. 1996).
The sections that follow highlight key aspects of interindividual variability in TCDD
pharmacokinetics, with an emphasis on the available data related to elimination half-lives and
volume of distribution.
3.3.3.1. Life Stage and Gender
The influence of the variability of fat content in human population on the distribution and
clearance of TCDD has been evaluated by several investigators. There are data showing an
inverse dependency of TCDD elimination rate on percent body fat. Figure 3-3 shows this
3-12
DRAFT - DO NOT CITE OR QUOTE
-------
relationship in a study in which TCDD elimination via feces was measured in six people in
relation to their body fat content (Rohde et al.. 1999). Observations of TCDD elimination rates
in a small number of men and women in the Seveso cohort (Avlward et al.. 2005a) provide a
modest opportunity to compare TCDD elimination rates with actual human data. Based on the
partition coefficients reported by Emond et al. (2006). the elimination rates for the men in the
sampled group are expected to be greater than the elimination rates in the women. Taking into
consideration calculations similar to those shown in Table 3-2, and fat proportions inferred from
body mass indices using the equations of Lean et al. (1996). the Seveso men studied are expected
to have an overall average of about 3.92% of their TCDD body burden outside of fat, whereas
the women are expected to have an average of only 2.36% outside of fat. On this basis, the
TCDD elimination rates in the men are expected to be 3.92/2.36 = 1.66 times faster than the
elimination rates in the women. By comparison, Michalek et al. (2002) reported observed
elimination rates in men and women that result in a slightly lower ratio:
men:0.Ill year1 ±0.010 (std.error) , ^
: — = 1.56 (Eq. 3-6)
women:0.071 year ± 0.010 (std.error)
The central estimates for the elimination rates correspond to half lives of 6.5 and 9.6 years for
men and women, respectively.
A further point of comparison can be derived using the observed body mass index (BMI)
and TCDD elimination rate of each of the male Ranch Hand military veterans, whose TCDD
elimination rates were observed between 9 and 33 years after their time in Vietnam. The average
BMI over that time was 29.44 (based on 287 measurements for the 97 veterans, tabulated in three
periods by Michalek et al.. 2002). and their average age was about 44.5 for the measurements.
Based on these data, the corresponding average estimated percent body fat is 29.7% using the
Lean et al. (1996) formula for men. The observed average TCDD elimination rate constant for
these men for the period was 0.092 year-1 ± 0.004 (standard error), corresponding to a half-life of
7.5 years. This half-life is slightly longer than the central estimate of the half-life of 6.2 years
2The body mass index, or BMI, is calculated as the body weight in kilograms divided by the square of the height in
meters.
3-13 DRAFT - DO NOT CITE OR QUOTE
-------
(i.e., ln(2)/0. Ill) for the smaller group of Seveso males with their slightly smaller estimated
percent body fat. Figure 3-4 shows a simple plot of these data and a fitted unweighted regression
line characterizing the relationship between estimated fat content and TCDD elimination rates.
Variation in metabolic enzyme activities and other routes of loss is also likely to be important,
but there is little human quantitative information available on these issues.
More recently, Kerger et al. (2006) estimated the slope of the relationship between
half-life and age to be 0.12 years (95% confidence interval, 0.10-0.14), which corresponds to the
rate of increase in TCDD half-life for each year of age. The authors speculated that although age
explained most of the variance in the individual half-life trends, it was also correlated with
TCDD concentration, BMI, and body fat mass. The regression model developed by these
authors discriminated between the high and low TCDD exposures or concentrations. Thus, after
accounting for the TCDD (concentration x age) term's effect on the slope of age, the final model
for TCDD concentration <700 ppt was
t\ / 2 = 0.35 + 0.12 x Age (Eq. 3-7)
For TCDD concentration >700 ppt, the final model was
t\ / 2 = 0.35 + 0.088 x Age (Eq. 3-8)
where tm is the half-life and Age is the age at time of subsequent sampling. Pharmacokinetic
information relevant to specific age groups is presented in the sections that follow.
3.3.3.1.1. Prenatal period
Data to estimate TCDD elimination rates for fetuses are not available. Levels of TCDD
in fetal tissues for rats were experimentally estimated at different gestational periods and utilized
in a developmental model by Emond et al. (2004). There is information on body composition
that is relevant to prediction of TCDD dose to fetus. These data, summarized as part of the
radiation dosimetry model of the International Commission on Radiological Protection, are
consistent with the idea that early fetuses are nearly all water and less than 1% lipid, and lipid
levels rise toward parity with protein near the time of normal delivery.
3-14 DRAFT - DO NOT CITE OR QUOTE
-------
Bell et al. (2007a) reported that the disposition of TCDD into the fetus shows dose
dependency, with a greater proportion of the dose reaching the fetus at lower doses of TCDD.
Further, both CYP1A1 and CYP1A2 are highly inducible (~103-fold) in fetal liver, whereas
CYP1A2 shows much lower induction (10-fold) in maternal liver. It has been speculated that
this is due to the lower basal levels of CYP1A2 in fetal liver, as compared to maternal liver (Bell
et al.. 2007a). The greater relative disposition to the fetus at low doses may be the result of
higher bioavailability due to less hepatic sequestration and elimination in the mother.
3.3.3.1.2. Infancy and childhood
Hattis et al. (2003) describe the general pattern of change of body fat content with age in
children. Central tendency values for percent body fat begin at about 12% at birth and rise
steeply to reach about 26% near the middle of the first year of life. Fat content then falls to reach
a minimum of approximately 15% at 5-8 years of age, followed by a sex-dependent "adiposity
rebound" that takes females to about 26% body fat while the males remain near 16—17% on
average by age 20. The interindividual variability distributions about these central values are
complex, as some children experience the "adiposity rebound" earlier than others, and this
creates patterns that are not simply interpretable as unimodal normal distributions. Hattis et al.
(2003) did find it possible to fit distributions of body fat content inferred from NHANES skin
fold measures to mixtures of two normal distributions for children between age 5 and 18.
At least two groups of authors have published PBPK modeling results indicating
generally more rapid clearance of TCDD in children than in adults, a trend that is consistent with
the generally lower fat content of children (Leung et al.. 2006; Van der Molen et al.. 2000;
Kreuzer et al.. 1997). The rapid expansion of the adipose tissue compartment can contribute, in
part, to the reduced apparent half-life in children (Clewell et al.. 2004). This reduction may also
be due to varying rates of metabolism and/or fecal lipid excretion (Kerger et al.. 2007; Abraham
et al.. 1996).
Furthermore, very young children have different modes and quantities of TCDD exposure
compared to adults. Lakind et al. (2000) characterize distributions of milk intake for nursing
infants to characterize distributions of TCDD exposure. This is also a corresponding route of
loss of TCDD stores for lactating women, as described in Section 3.3.3.2 below.
3-15
DRAFT - DO NOT CITE OR QUOTE
-------
3.3.3.1.3. Adulthood and old age
The fraction of fat in relation to body weight in adulthood and old age can be computed
as a function of the BMI and age (e.g.. Lean et al.. 1996):
% Body Fat (males) = 1.33 x BMI + 0.236 x Age - 20.2
(Eq. 3-9)
% Body Fat (females) =1.21 x BMI + 0.262 x Age - 6.7
(Eq. 3-10)
The above equations are the result of analysis of data based on underwater weighing of
63 men and 84 women (age range 16.8-65.4). The salient observation with respect to TCDD for
these data is that age and BMI-dependent variability in fat content have implications for the
variability in TCDD elimination rates and internal dose among adults.
3.3.3.2. Physiological States: Pregnancy and Lactation
Data on body fat content in pregnant women at various stages of gestation (Pipe et al..
1979) have potential implications for TCDD elimination rates during pregnancy, even though the
relationship between these parameters has not been formally analyzed.
Lactation is viewed as an additional route of elimination for some chemicals such as
TCDD. According to a recent study, a breast-feeding woman expels through lactation an
estimated 8.76 kg fat per year [^/(kg/day), 0.8 kg milk/day with an average 3% lipid], and the
partition coefficient between blood lipid and milk fat (£bm) for TCDD is 0.92 (Milbrath et al..
2009; Wittsiepe et al.. 2007). The estimated rate of elimination of TCDD due to breast-feeding
(kbfed) can then be computed as follows (Milbrath et al.. 2009):
h
'bfed
(Eq. 3-11)
3-16 DRAFT - DO NOT CITE OR QUOTE
-------
where
ktbfed (unitless) = the fraction of the year during which the woman was actively
breast-feeding;
Assuming no interaction between breast-feeding and other half-life determinants
Milbrath et al. (2009). the authors predicted a half-life of 4.3 years for TCDD in a 30-year-old,
nonsmoking woman with 30% body fat if she did not breast-feed that year, and a half-life of
1.8 years if she breast fed for 6 months.
3.3.3.3. Lifestyle and Habits
One of the factors related to lifestyle and habits that could influence TCDD kinetics is
smoking. Smoking has been reported to enhance the elimination of dioxin and dioxin-like
compounds (Ferribv et al.. 2007; Flesch-Janvs et al.. 1996). Milbrath et al. (2009) accounted for
interindividual variation in body composition as well as smoking habits in an empirical model.
The predicted half-life (years) for an individual i as a function of age, smoking status, and
percent body fat i was as follows
Pbfi
woman's percent body fat; and
woman's body weight in kg.
BW
tl/2(age, smoke, pbf)t = [p(Qage) + p{age) x age J x SFt x
ref(age<)
(Eq. 3-12)
where
(0 age)
intercept constant derived from regressed data;
slope constant derived from regressed data;
age,
specific age i (years);
pbfi
individual percent body fat;
ref(agei)
reference percent body fat; and
the unitless, multiplicative smoking factor.
SFt
3-17
DRAFT - DO NOT CITE OR QUOTE
-------
3.3.3.4. Genetic Traits and Polymorphism
One particular genetic locus that is potentially related to TCDD pharmacokinetics and
tissue dose is the gene for the AhR. Eight candidate AhR polymorphisms have been identified to
date (Connor and Avlward. 2006; Harper et al.. 2002). Given the role of AhR in regulating the
induction of CYP1 isozymes (Connor and Avlward. 2006; Toide et al.. 2003; Baron et al.. 1998).
the polymorphism might lead to interindividual differences in metabolic clearance, the
significance of which would depend upon the dose, fat content, and exposure scenario. In this
regard, it should be noted that the inducibility of aromatic hydrocarbon hydroxylase in human
tissues has been reported to be highly variable, up to 100-fold (Connor and Avlward. 2006;
Smart and Daly. 2000; Wong et al.. 1986).
The scientific literature contains values of K& (the dissociation constant of the
TCDD-AhR complex) ranging from about 1 to much higher values (corresponding to lower
binding affinity) (reviewed in Connor and Avlward. 2006). This provides suggestive evidence
for a heterogeneous human AhR, with functionally important polymorphisms (Micka et al..
1997; Roberts et al.. 1986). even though some of the range may be attributed to experimental
procedural differences and to other factors (Connor and Avlward. 2006; Harper et al.. 2002;
Lorenzen and Okey. 1991; Manchester et al.. 1987).
The various pharmacokinetic processes and determinants (see Sections 3.3.2 and 3.3.3),
individually or together, might influence the dose metrics of relevance to the dose-response
modeling of TCDD.
3.3.4. Dose Metrics and Pharmacokinetic Models for TCDD
3.3.4.1. Dose Metrics for Dose-Response Modeling
The dose metric related to a toxicological endpoint can range from the maximal
concentration, the area under a time-course curve (AUC), or the time-averaged concentration of
the toxic moiety in the body, blood, or target tissue, to an appropriate measure of the resulting
interactions in the target tissue (e.g., receptor occupancy or functional biomarkers related to
specific effects). A single dose metric, however, is unlikely to be sufficient for all endpoints and
exposure durations. Further, the ideal dose metric chosen on the basis of the mode of action
(MOA) may not be the dose metric for which model predictions can be obtained with a high
3-18
DRAFT - DO NOT CITE OR QUOTE
-------
level of confidence. Consideration of these issues is critical to the selection of the dose metrics
of relevance to dose-response modeling of TCDD.
Figure 3-5 lists a range of alternative dose metrics for TCDD in terms of their relevance
based on considerations of pharmacokinetic mechanisms and MOA. The administered dose or
daily intake (ng/kg-day) is the least relevant dose metric for dose-response modeling of TCDD.
This dose adjusts only for body-weight differences between species. The administered dose,
when used with an uncertainty factor for kinetics (or kinetic adjustment factor, such as BW3 4)
and an uncertainty factor for dynamics, can also account for allometrically predicted
pharmacokinetic (clearance) and pharmacodynamic differences between species in deriving the
human equivalent dose (HED). In effect, the use of kinetic and dynamic adjustment or
uncertainty factors facilitates the computation of HED. Such a calculation of HED is associated
with the steady-state blood concentration of parent chemical in rats by accounting for species
differences in metabolic clearance. This is generally done by relating to body surface area or
metabolic rates, with no corresponding temporal changes in the volume of distribution (see, for
example. Krishnan and Andersen. 1991). Such calculations of HED for TCDD may not be
appropriate given that (1) steady-state was not attained in all critical toxicological studies chosen
for the assessment, (2) the clearance is mainly due to enzyme(s) and processes whose levels/rates
do not necessarily vary across species or life stages as a function of body surface differences, and
(3) there is a likelihood of change in volume of distribution over time. Furthermore, the use of
administered dose does not explicitly account for the dose-dependent elimination of TCDD from
tissues as demonstrated in multiple studies (reviewed in Sections 3.3.2 and 3.3.4). The use of
administered dose in TCDD dose-response modeling is unlikely to facilitate the characterization
of the true relationship between the response and the relevant measures of internal dose that are
influenced by dose-dependent elimination and binding processes. Additionally, the use of
administered dose to extrapolate across species or life stages would not effectively take into
account the differences in fat content or the demonstrated dose-dependent and species-dependent
differences in elimination half-life of TCDD.
Dose metrics for TCDD may include absorbed dose, body burden, serum or whole blood
concentration, tissue concentration, and possibly functional-related metrics of relevance to the
MOA (e.g., receptor occupancy, change in protein levels). These measures can be calculated as
a current (terminal), average (over a defined period), or integral quantity. The applicability of
3-19
DRAFT - DO NOT CITE OR QUOTE
-------
the integral measures, such as the AUC (i.e., the area under the curve of a plot of blood or
plasma concentration vs. time), traditionally used for analyzing chronic toxicity data, is
questionable in the case of TCDD. This is because of differences in lifespan and uncertainties
regarding the appropriateness of the duration to be specified for averaging the AUC in
experimental animals and humans for certain critical effects (NAS. 2006b).
Among the alternative dose metrics, the absorbed dose accounts for differences in body
weight as well as species-specific differences in bioavailability. Thus, the absorbed dose is
equivalent to body burden. Body burden, or more appropriately, the body concentration,
represents the amount of TCDD per kg body weight. TCDD body burdens, like other dose
measures, can be determined as the peak, the average over the period of the bioassays, or the
level at the end of the experiments. Thus, the terminal or average body burdens can be obtained
either using data or pharmacokinetic models and used in dose-response modeling. The body
burden is a measure of TCDD dose that reflects the net impact of bioavailability, uptake,
distribution, and elimination processes in the organism. It is essentially a function of the volume
of distribution and clearance processes, and as such, it does take into account the temporal
changes in volume of distribution as well as the concentration-dependent clearance. These are
phenomena that are critical to the understanding of TCDD dose to the target. However, the body
burden may not accurately reflect the tissue dose (NAS. 2006b). and as such, does not allow for
analysis of species-specific differences in target organ sensitivity to TCDD. In essence, the body
burden represents only an "overall average" of TCDD concentration in the body, without regard
to the differential partitioning and accumulation in specific tissues, including the target tissue(s).
Serum (or blood) concentration of TCDD is a dose metric that reflects both the body
burden and the dose-to-target tissues. Serum or blood concentration, at steady-state, would be
reflective of the impact of clearance processes and expected to be directly proportional to the
tissue concentrations of TCDD (NAS. 2006b). This dose metric for lipophilic chemicals such as
TCDD is often expressed as a lipid-normalized value, to adjust for varying serum lipid content
(Niskar et al.. 2009; Patterson et al.. 2009; DeKoning and Karmaus. 2000). particularly in human
biomonitoring studies, thus of relevance to dose-response modeling; however, the serum
lipid-normalized concentrations of TCDD are not routinely collected and reported in animal
toxicological studies. Serum lipid-adjusted TCDD concentration is calculated as the ratio of
serum TCDD content over serum lipid content per unit volume. Alternatively, TCDD serum
3-20
DRAFT - DO NOT CITE OR QUOTE
-------
lipid-normalized calculation can be estimated by using the formula TL = (2.27 x TC) + TG +
62.3 mg/dL where the total lipid (TL) content of each sample is estimated from its total
cholesterol (TC) and triglyceride (TG) (Patterson et al.. 2009). The lipid-adjusted serum
concentration, however, would be reflective of the lipid-adjusted concentration of TCDD in other
organs (reviewed in Avlward et al.. 2008) depending upon the extent of steady-state attained and
the similarity of lipid composition across tissues in each species. In essence, the serum
lipid-normalized measure is representative of the amount of TCDD per specified volume of total
lipids, whereas the whole blood measure will be reflective of the ensemble of free, lipid-bound
and protein-bound TCDD in plasma and erythrocytes, which may be species-specific. Even
though these dose metrics are thought to be more closely and directly related to the tissue
concentrations associated with an effect, a less direct association might occur at increasing doses
when nonlinear processes dominate the kinetics and distribution of TCDD into organs such as
the liver.
Tissue concentration of TCDD, as free, bound, or total TCDD, is a more relevant
pharmacokinetic measure of dose, given that it provides a measure of exposure of the target cells
to the chemical. In this regard, the CYPlA2-bound fraction may be considered as a relevant
dose metric for certain toxic effects; however, the available data contain mixed results regarding
the mechanistic linkage of this dose metric to toxicity and carcinogenicity (reviewed in Budinskv
et al.. 2006). In such cases, the use of alternative dose metrics (e.g., bound concentration as well
as the serum concentration) in dose-response modeling could be considered. Other
function-related biomarkers and dose metrics could facilitate the additional consideration of
pharmacodynamic aspects reflecting tissue- and species-specific sensitivity. These metrics may
represent the most relevant measures of tissue exposure and sensitivity to TCDD. For example,
receptor occupancy and functional biomarkers as dose metrics for TCDD require a clear
understanding of mode of action of TCDD and availability of relevant data. In the absence of
such information, these possible dose metrics cannot be utilized at the present time.
Empirical time-course data on the alternative dose metrics of TCDD associated with
epidemiologic and experimental (animal) studies are not available, requiring the use of
pharmacokinetic models to obtain estimates of these dose metrics. These models may be simple,
based on first-order kinetics (see Section 3.3.4.2), or more complex based on physiochemical,
3-21
DRAFT - DO NOT CITE OR QUOTE
-------
biochemical, and physiological parameters for simulating uptake, distribution (including
sequestration to proteins), and clearance of TCDD (see Section 3.3.4.3).
3.3.4.2. First-Order Kinetic Modeling
Figure 3-6 illustrates the process of estimating a human-equivalent TCDD oral exposure
from an experimental animal-administered dose, based on the assumption that body burden is the
effective dose metric for TK equivalence across species. The primary assumption is that the
time-weighted average (TWA) TCDD body burden over some critical time period is the
"3
proximate toxicokinetically effective dose eliciting a toxicological effect. The process consists
of estimating the effective average body burden in the experimental animal over some time tA
(generally the experimental duration) using a TK model, then "back-calculating" a daily human
exposure level that would result in that average body burden over some time tH (the human
equivalent to /j).
The following closed-form equation is the general formula used to calculate a TCDD
terminal body burden in an experimental animal or human at time (t).
BB(t) = BB(0) + d(l 6 (Eq. 3-13)
where
BB(t)
= the
BB( 0)
= the
d
= the
k
= the
t
= the
fa
= the
3The conversion depicted in Figure 3-6 does not account for toxicodynamic differences between species.
3-22 DRAFT - DO NOT CITE OR QUOTE
-------
7 d a(\-e~^AtA\fa a
For the experimental animal, BB(t) is BB [ (I) = BB \ (0)e A ¦ ' + — —
kA
i t dzt(1 - e~kRtH )fau
and for humans, this parameter is BBH(t) = BBH (0)e H H + —— —.
kH
Setting BBn{t) = BBa($) obtains the following expression:
BB„(+ d"('-)faH = BBa(d)e-k*'A (Eq. 3-14)
kH k,\
Rearranging yields the general solution for dn.
dH = dA tS- f*A 0 g kA'A) +BBA(0y-k-t'A-BBH(0)e-k"'n (Eq 3-15)
kA faH (1 - e~ H H )
Assuming that initial body burdens are very small compared to BB(t) and that the fraction of
TCDD absorbed is the same for humans and experimental animals, and using the relationship
ln(2)
k = ¦
'' 2 , where /u is the whole-body half-life, a simplified solution for dn is obtained.
t (\ — e k '!'' )
dH = dA (Eq. 3-16)
W(1-^)
The term 1-e fais the daily fraction eliminated. Therefore, dn can be seen to be the
average daily administered dose to the experimental animal times the ratio of the animal :human
half-life times the ratio of the animal :human daily fraction eliminated over the respective times,
tA and tn. For both species at (theoretical) steady state (I —~ co; daily fraction eliminated —>¦ 1),
the latter ratio approaches unity, reducing the animal :human conversion factor to the ratio of the
half-lives.
However, for less-than-lifetime exposures eliciting noncancer effects, specific values for
tA and tn must be considered. Furthermore, Eq. 3-16 computes dn on the basis of terminal body
3-23
DRAFT - DO NOT CITE OR QUOTE
-------
burdens at times ^ and tH. The more representative metric for toxicokinetic equivalence based
on average body burden over the respective time periods is given in Eq. 3-17.
Eq. 3-17 is transformed again assuming minimal initial body burden (BB(0) ~ 0) to:
1 (l-e~kAtA)
kAtA
(Eq. 3-18)
1 tm jf-bH'HO -e-kHtH)
- e
fH kHfH
kHfH
where tno is the initial human exposure time.
The value of tA is the duration of the experimental exposure period. For some gestational
exposures, if a critical exposure window is defined, tA will be the duration of the critical
exposure window. The value of tn is the human-equivalent duration corresponding to I a.
However, for ^ less than lifetime (less than 2 years in rodents) and no defined susceptible life
stage, tH cannot begin at 0 (because typically animal experiments do not begin at age 0), but must
end at 25,550 days (70 years) to include the terminal (pseudo) steady-state level, at which the
BBff(t): dn ratio is highest. Otherwise, starting In at 0 would not be protective for
less-than-lifetime effects that could be manifest at any age in humans; the average is determined
from the terminal end of the human exposure period because the daily exposure achieving the
target blood concentration is smaller than for the same exposure period beginning at birth (i.e.,
dn would be higher for earlier exposure periods) and is health protective for effects occurring
after shorter-term exposure.4 Figure 3-7 depicts the relationship of daily dose to TWA body
burden graphically for several exposure duration scenarios. For shorter durations occurring later
in life, the average body burden over the exposure period does not differ substantially from the
steady-state value. Even for half-lifetime exposures, the deviation of the average from steady
4See the following (Section 3.3.4.3) for a more detailed discussion of this concept.
3-24 DRAFT - DO NOT CITE OR QUOTE
-------
state is minimal. Only for lifetime exposures does the difference become more marked, but only
by about 15%. Note that in the 2003 Reassessment, a constant value of 3,000 was used for
BBffft): da, based on the relationship of continuous exposure to theoretical steady-state body
burden (t = lifetime, l<2 = 2,593 days); this approach, while conservative, does not account for
exposure scenarios of different durations and does not strictly reflect the average body burden
dose metric.
The simulation in Figure 3-7 is based on a unit daily exposure to humans, such that the
target body burden represents BBii(ln).dn as a general scalar for calculating tin from any given
d.A. Table 3-3 shows the resulting TK conversion factors for the rodent species and strains
comprising the bulk of the experimental animals in TCDD studies. Monkey and mink values are
not shown in this table because, for the former, only chronic exposures were evaluated and, for
the latter, no TCDD half-life information is available. Monkey (Rhesus) half-life estimates
range from about 200-500 days. A representative value of 365 days is used for this TCDD
assessment. The dA to dH conversion factor for the chronic monkey exposures (3.5-4 years) in
TCDD studies is 9.2-9.7 (BBA\dA = 279-263).
Application of first-order kinetics for the health assessment of TCDD can only be used to
estimate total body burdens or back-calculate administered dose from experimental data. Body
burden calculations using first-order kinetics is based on the assumption of a first-order decrease
in the levels of administered dose as function of time. In that sense, any loss of TCDD from the
body is described by using a rate constant that is not specific to any biological process. This
constant is usually estimated from estimates of half-life of TCDD. Assuming a constant half-life
value for the clearance for long-term or chronic TCDD exposure is not biologically supported
given the observed data indicating early influence of CYP1A2 induction and binding to TCDD
and later redistribution of TCDD to fat tissue. Abraham et al. (1988) found that the liver:adipose
tissue concentration ratio in female Wistar rats exposed to a subcutaneous TCDD dose of
300 ng/kg decreased from 10.3 at 1 day postexposure to 0.5 at 91 days postexposure.
Consequently, using half-life estimates based on observed steady-state levels of TCDD will not
account for the possibility of accelerated dose-dependent clearance of the chemical at the early
stages and, thus, would result in estimation of lower administered levels of the chemical. The
dynamic change in half-life due to dose-dependent elimination at the early stages of TCDD
exposure and its later redistribution to fat tissues for steady-state levels is better described using
3-25
DRAFT - DO NOT CITE OR QUOTE
-------
biologically based models, such as the PBPK models and concentration- and age-dependent
elimination (CADM) models (Emond et al.. 2006; Avlward et al.. 2005b; Emond et al.. 2005;
Emond et al.. 2004; Carrier et al.. 1995a. b). Additionally, these models provide estimates for
other dose metrics (e.g., serum or tissue levels) that are more biologically relevant to response
than administered dose or total body burden (see Section 3.3.4.3).
3.3.4.3. Biologically Based Kinetic Models
The development and evolution of biologically based kinetic models for TCDD have
been reviewed by EPA (2003) and Reddy et al. (2005). The initial PBPK model of Leung et al.
(1988) was developed with the consideration of TCDD binding to CYP1A2 in the liver. The
next level of PBPK models by Andersen et al. (1993) and Wang et al. (1997) used
diffusion-limited uptake and described protein induction by interaction of DNA-binding sites.
The models of Kohn et al. (1993) and Andersen et al. (1997) further incorporated extensive
hepatic biochemistry and described zonal induction of CYP by TCDD. TCDD PBPK models
have evolved to include detailed descriptions of gastrointestinal uptake, lipoprotein transport,
and mobilization of fat, as well as biochemical interactions of relevance to organ-level effects
(Kohn et al.. 1996; Roth et al.. 1994). Subsequently, developed PBPK models either used
constant hepatic clearance rate (Maruvama et al.. 2002; Wang et al.. 2000; Wang et al.. 1997) or
implemented varying elimination rates as an empirical function of body composition or dose
(Van der Molen et al.. 2000; Van der Molen et al.. 1998; Andersen et al.. 1997; Kohn et al..
1996; Andersen et al.. 1993). The more recent pharmacokinetic models explicitly characterize
the concentration-dependent elimination of TCDD (Emond et al.. 2006; Avlward et al.. 2005b;
Emond et al.. 2005; Emond et al.. 2004; Carrier et al.. 1995a. b). The biologically based
pharmacokinetic models describing the concentration-dependent elimination (i.e., the
pharmacokinetic models of Emond et al.. 2006; Avlward et al.. 2005b; Emond et al.. 2005) are
relevant for application to simulate the TCDD dose metrics in humans and animals exposed via
the oral route. The rationale for considering the Aylward et al. (2005b) and Emond et al. (2006;
2005; 2004) models for estimating dose metrics for possible application to TCDD health
assessment is based on the following considerations.
3-26
DRAFT - DO NOT CITE OR QUOTE
-------
• Both models were developed and calibrated using research results from the more recent
peer-reviewed publications.
• Both models are relatively simple and less parameterized than earlier kinetic models for
TCDD. The Aylward et al. (2005b) model is based on two-time scale TCDD kinetics
described by Carrier et al. (1995a). and the Emond et al. (2006; 2005; 2004) PBPK
models are reduced versions of earlier complex PBPK models. Although simple, both
the Aylward et al. (2005b) and Emond et al. (2006; 2005; 2004) models are inclusive of
important kinetic determinants of TCDD disposition.
• Both models are uniquely formulated with dose-dependent hepatic elimination consistent
with current understanding of TCDD toxicokinetics.
• Both models and extrapolated human versions were tested against human data collected
in a variety of human exposure scenarios (Aylward et al.. 2005b; Emond et al.. 2005).
• Both models are capable of deriving one or more of the candidate dose metrics that may
be of interest to EPA's dose-response assessment of TCDD.
3.3.4.3.1. Concentration- and age-dependent model (CADM)
3.3.4.3.1.1. Model structure
The pharmacokinetic model of Aylward et al. (2005b). referred to as CADM in this
report, is based on an earlier model developed by Carrier et al. (1995a. b) that describes the
dose-dependent elimination and half-lives of polychlorinated dibenzo-p-dioxins and furans. This
model describes the TCDD levels in blood (body), liver, and adipose tissue. Blood itself is not
characterized physically as a separate compartment within the model, and the distribution of
TCDD to tissues other than adipose tissue and liver (usually less than 4%) is not accounted for
by the model. The original structure of the Carrier et al. (1995a. b) model was modified by
Aylward et al. (2005b) to include TCDD elimination through partitioning from circulating lipids
across the lumen of the large intestine into the fecal content (see Figure 3-8). The most recent
version of the Carrier model (2008; Aylward et al.. 2005b) includes fecal excretion of TCDD
from two routes: (1) elimination from circulating blood lipid through partitioning into the
intestinal lumen; and (2) elimination of unabsorbed TCDD from dietary intake.
A basic assumption of this model is that metabolic elimination of TCDD is a function of
its current concentration in the liver. The current concentration of TCDD in the liver increases
with increasing body burden in a nonlinear fashion as a result of the induction of (and binding of
TCDD to) specific proteins (i.e., CYP1A2). Consequently, the fraction of TCDD body burden
contained in the liver increases nonlinearly (with a corresponding decrease in the fraction
3-27
DRAFT - DO NOT CITE OR QUOTE
-------
contained in adipose tissues) with increasing body burden of TCDD (Avlward et al.. 2005a;
Carrier et al.. 1995a).
Of particular note is that the adipose tissue compartment of the model is considered to
represent the lipid contained throughout the body. It then assumes that the concentrations of
TCDD in lipids of plasma and various organs are essentially equivalent to that of adipose tissue,
and as such, these concentrations are included in the adipose compartment of the model. Even
though this approximation is fairly reasonable given the available data, there is some concern
that the adipose compartment of this model also includes the lipid content of the liver to some
unknown extent. Because the equilibrium balance between free and bound TCDD in the liver is
dependent on the adipose content of the tissue, removal of lipid volume from the liver would
mathematically alter total hepatic concentration and, therefore, would affect the estimated levels
of the chemical available for binding to proteins.
Distribution in the body is modeled to occur between hepatic and adipose/lipid
compartments, with the fraction of body burden in liver increasing according to a function that
parallels the induction of the binding protein CYP1A2. Elimination is modeled to occur through
hepatic metabolism (represented as a first-order process with rate constant K that decreases with
age) and through lipid-based partitioning of unmetabolized TCDD across the intestinal lumen
into the gut, which is also modeled as a first-order process. As the body burden increases, the
amount of TCDD in the liver increases nonlinearly, resulting in an increased overall elimination
rate.
3.3.4.3.1.2. Mathematical representation
The CADM model describes the distribution to tissues (including liver and adipose
tissue) based on exchange from blood at time intervals of 1 month. The model is based on
quasi-steady-state-approximation, and, thus, it is also based on the consideration that the
intertissue processes reach their equilibrium values "quasi-instantaneously." In this regard,
absorption and internal distribution reflective of kinetics at the cellular level (e.g., diffusion,
receptor binding, and enzyme induction) likely occur on a relatively fast time scale (a few hours
to a few days). However, the overall body concentration (i.e., body burden) varies slowly with
time such that it remains virtually unchanged during short time intervals.
3-28
DRAFT - DO NOT CITE OR QUOTE
-------
The CADM model does not differentiate between binding to AhR and CYP1A2, and it
lacks explicit descriptions of CYP1A2 induction, a key determinant of TCDD kinetics.
However, the empirical equation in the CADM model is based on five parameters (i.e., fmin, fmax,
K, Wa, and Wi; see Tables 3-4 and 3-5) that allow the successful description of the behavior of
TCDD in liver and adipose tissue (i.e., TCDD half-lives in each compartment increase with
decreasing body burden). This observation implies that the model adequately accounts for the
ensemble of the processes. Essentially, the CADM model describes the rate of change in tissue
concentrations of TCDD as a function of total body burden such that the global elimination rate
decreases with decreasing body burden or administered dose.
3.3.4.3.1.3. Parameter estimation
The CADM model is characterized by its simplicity and fewer parameters compared to
physiologically based models. Reflecting this simplicity, hepatic extraction is computed with a
unified empirical equation that accounts for all relevant processes (i.e., protein induction and
binding).
The key parameters (fmi„, fmax, K, and ke) were all obtained by fitting to species-specific
pharmacokinetic data. The physiological parameters (such as tissue weights) used in the model
are within ranges documented in the literature. The fat content is described to vary as a function
of age, sex, and BMI. However, the BMI of the model is not allowed to change during an
individual simulation (which can range from 20 years to 70+ years), when in reality, the
percentage of fat in humans changes over time. None of the TCDD-specific parameters were
estimated a priori or independent of the data set simulated by the model.
3.3.4.3.1.4. Model performance and degree of evaluation
The CADM model was not evaluated for its capabilities in predicting data sets not used
in its parameterization. In other words, one or more of the key input parameters (fhmin, fhmax, ke,
K) was obtained essentially by fitting to the species-specific pharmacokinetic data, such that
there was no "external" evaluation data set to which the model was applied. Despite the lack of
emphasis on the "external" evaluation aspect, the authors (Avlward et al.. 2005a; Carrier et al..
1995a. b) have demonstrated the ability of the model to describe multiple data sets covering a
range of doses and species.
3-29
DRAFT - DO NOT CITE OR QUOTE
-------
The visual comparison of the simulated data to experimental values suggests that the
model could, to an approximate degree, correctly reproduce the whole set of data (e.g.,
pharmacokinetic [PK] profile over a range of dose and time) and not just part of the PK curve,
essentially with the use of a single set of equations and parameters.
The pharmacokinetic data sets for TCDD that were used to calibrate the CADM model by
Aylward et al. (2005a; Carrier et al.. 1995a. b) included the following:
• Adipose tissue and liver concentrations of TCDD following a single oral dose of 1 |ig/kg
in monkeys (McNultv et al.. 1982);
• Percent dose retained in liver for a total dose of 14 ng in hamsters (Van den Berg et al..
1986);
• Elimination kinetics of TCDD in female Wistar rats following a single subcutaneous dose
of 300 ng/kg (data from Abraham et al.. 1988);
• Liver and adipose tissue concentrations (terminal measurements) in Sprague-Dawley rats
given 1, 10, or 100 ng TCDD/kg bw per day for 2 years (Kociba et al.. 1978); and
• Serum lipid concentrations of TCDD over a period of several years in 54 adults (29 men
and 25 women) from Seveso and in three Austrian patients (Aylward et al.. 2005a).
For illustration purposes, Figure 3-9 shows model simulations of rat data from Carrier
et al. (1995a). Figure 3-2 (see Section 3.3.2.4) depicts the human data that were used by the
authors to support the concentration-dependent elimination concept; the model was
parameterized to provide adequate fit to these data (Aylward et al„ 2005a).
The authors did not report any specialized analyses that quantitatively evaluated the
uncertainty, sensitivity, and/or variability of CADM model parameters and structure.
3.3.4.3.1.5. Confidence in CADM model predictions of dose metrics
Using professional judgment, EPA ranked its confidence in the CADM model as low,
medium, or high (or not applicable) based on model simulations of administered dose, absorbed
dose, body burden, serum lipid concentration, total tissue (liver) concentration, and receptor
occupancy. A qualitative level of confidence associated with the predictability and reliability of
absorbed dose and body burden for oral exposures in humans (as well as several animal species)
by this model can be ranked as high (see Table 3-6). This model, however, does not account for
3-30
DRAFT - DO NOT CITE OR QUOTE
-------
the differential solubility of TCDD in serum lipids and adipose tissue lipids, nor does it account
for the diffusion-limited uptake by adipose tissue. Due to these limitations, the confidence
associated with the predictions of the serum lipid concentration of TCDD is considered medium,
particularly when it is not documented that steady-state is reached during the critical
toxicological studies and human exposures. Furthermore, the CADM model does not facilitate
the computation of TCDD concentrations in specific internal organs (other than liver and adipose
tissue). The reliability of this model for simulating the liver concentration (free, bound, or total)
of TCDD at low doses is considered to be low. This low confidence level is a result of the
uncertainty associated with the key parameter, /hmin- This parameter needs to be re-calibrated for
each study/species/population to effectively represent the free fraction of TCDD in liver and the
amount of TCDD contained in the hepatic lipids and bound to the liver proteins (whose levels
might be reflective of background exposures of various sources; see Carrier et al.. 1995a). The
uncertainty related to the numerical value of this parameter in animals and humans—particularly
at very low exposures—raises concern regarding the use of this model to predict TCDD
concentration (free, bound, or total) in liver as the dose metric for dose-response modeling.
Although the use of the parameter /hmax permits the prediction of the dose to liver at high doses,
it does not specifically facilitate the simulation of the amount bound to the protein or level of
induction in liver. Because the CADM model is not capable of simulating enzyme induction
based on biologically relevant parameters, its reliability for predicting the concentration of
TCDD bound specifically to the AhR is not known. Finally, due to the lack of parameterization
or verification with kinetic data in pregnant, lactating, or developing animals or humans, the
CADM model is unlikely to be reliable in the current form for use in predicting potential dose
metrics for these lifestages or study groups that might form the basis of points of departure
(PODs) for the assessment.
3.3.4.3.2. PBPK model
3.3.4.3.2.1. Model structure
Emond et al. (2006. 2004) simplified the eight-compartment rat model of Wang et al.
(1997) to a four-compartmental developmental model (liver, fat, rest of body, and placenta with
fetal transfer) (Emond et al.. 2004). and later to a three-compartment adult model (liver, fat, rest
of the body) (Emond et al.. 2006) (see Figures 3-10 and 3-11). Their rationale for simplification
3-31
DRAFT - DO NOT CITE OR QUOTE
-------
of the model was based on evaluating, critiquing, and improving all earlier PBPK models by
Wang et al. (1997). In general, the main reason for the simplification was that extrapolation of a
PBPK model to humans with these many (i.e., eight compartments) compartments would be
problematic due to the limited availability of relevant human data for validation (Emond et al..
2004). One major difference from earlier models, repeatedly emphasized by Emond et al. (2006;
2005). was their description (included in their simplified PBPK models) of the dose-dependent,
inducible elimination of TCDD. The rationale for including TCDD binding and induction of
CYP1A2 into the model was earlier described by Santostefano et al. (1998).
The most recent version of the rat and human PBPK models developed by Emond et al.
(2006) describes the organism as a set of three compartments corresponding to physiological
tissues—liver, fat, and rest of the body—interconnected by systemic circulation (see
Figure 3-10). The liver compartment includes descriptions of CYP1A2 induction, which is
critical for simulating TCDD sequestration in liver and dose-dependent elimination of TCDD. In
this model, the oral absorption of TCDD from the GI tract accounts for both the lymphatic (70%)
and portal (30%) systems.
The biological relationship between TCDD "sequestration" by liver protein and its
"elimination" by the liver is not entirely clear. TCDD is metabolized slowly by unidentified
enzymes. CYP1A2 is known to metabolize TCDD based on studies in CYP1A2 knockout (KO)
mice (Diliberto et al.. 1999. 1997). in which the metabolic profile is different compared to
wild-type mice. However, because several metabolites appear in the feces of CYP1A2 knock out
mice, it is assumed that there are other enzymes involved in TCDD metabolism. TCDD binds to
AhR and induces not only CYP1A2, but also CYP1A1, CYP1B1, and several UGTs and
transporters (Gasiewicz et al.. 2008). Both hydroxylated and glucuronidated hydroxyl
metabolites are found in the feces of animals treated with TCDD (Hakk et al.. 2009). Because
the exact enzymes involved with TCDD are unknown and yet the metabolism is induced by
TCDD, an assumption of increased elimination rate of TCDD in proportion to the induction of
CYP1A2 is made. In the PBPK model, CYP1A2 is also needed because TCDD binds to rat,
mouse, and human CYP1A2 (Staskal et al.. 2005; Diliberto et al.. 1999). Thus, CYP1A2
induction is necessary to describe TCDD pharmacokinetics due to TCDD binding. Hence,
CYP1A2 can be used as a marker of Ah-receptor induction of "TCDD metabolizing enzymes."
3-32
DRAFT - DO NOT CITE OR QUOTE
-------
Other models use AhR occupancy as a marker of induction of "TCDD metabolizing enzymes"
(Kohn et al.. 2001; Andersen et al.. 1997).
Figure 3-11 depicts the structure of the rat developmental-exposure PBPK model (Emond
et al.. 2004). This model was developed to describe the relationship between maternal TCDD
exposure and fetal TCDD concentration during critical windows of susceptibility in the rat. In
formulating this PBPK model, Emond et al. (2004) reduced the original 8-compartment model
for TCDD in adult rats by Wang et al. (1997) to a 4-compartment (i.e., liver, fat, placenta, and
rest of the body) model for maternal rat. Activation of the placental compartment and a separate
fetal compartment occurs during gestation (Emond et al.. 2004).
3.3.4.3.2.2. Mathematical representation
The key equations of the PBPK model of Emond et al. (2004) are reproduced in
Text Boxes 3-1 and 3-2, whereas those from Emond et al. (2006; 2005) are listed in Table 3-7.
The rate of change of TCDD in the various tissue compartments is modeled on the basis of
diffusion limitation considerations. Accordingly, mass balance equations are used to compute
the rate of change in the tissue (i.e., intracellular compartment) and tissue blood (i.e.,
extracellular compartment). The membrane transfer of TCDD is computed using a permeation
coefficient-surface area cross product (PA) for each tissue. Metabolism and binding of TCDD to
the AhR and inducible hepatic protein (CYP1A2) are described in the liver. The total mass in
the liver is then apportioned between free dioxin (Qf) and bound forms of TCDD (see
Figure 3-12). The dose- and time-dependent induction of hepatic CYP1A2 in the liver is
described per Wang et al. (1997) and Santostefano et al. (1998). Accordingly, the amount of
CYP1A2 in the liver was computed as the time-integrated product of inducible production and a
simple first-order loss process (Wang et al.. 1997):
dCYP
^f^ = S(t)K0 - K2CA2t (Eq. 3-19)
at
In this expression, CYPja2 is the concentration of the enzyme (nmol/g), K2 is the rate constant for
the first-order loss (hour-1), Ca2i is the concentration of CYP1A2 in the liver (nmol/g), Kq is the
basal rate of production of CYP1A2 in the liver (nmol/g/hr), and S(t) (unitless) is a multiplicative
3-33 DRAFT - DO NOT CITE OR QUOTE
-------
stimulation factor for CYP1A2 production in the form of a Hill-type function (see
Section 3.3.2.3):
(CAh-TCDD )
(ICA2f+(CAh_TCDD)h
S(t) = 1 + A2hK Ah-TCDDJ—- (Eq. 3-20)
where, S(t) is the stimulation function, IriA2 is the maximum fold of CYP1A2 synthesis rate over
the basal rate, Cah-tcdd is the concentration of AhR occupied by TCDD, and ICa2 is the
Michaelis-Menten constant of CYP1A2 induction (nM). The dose-dependent or variable
elimination of TCDD was described using the relationship:
KBILE LI =
CYP\A2induced-CYP\A2basal
CYPXA2
basal
x Kelv
(Eq. 3-21)
where CYPlA2indUCedis the concentration of induced CYP1A2 (nmol/mL), CYPlA2baSai is the
basal concentration of CYP1A2 (nmol/mL), and kelv is the interspecies constant adjustment for
the elimination rate (hour-1).
There are various ways of formulating the dose-dependent elimination as a function of
the level of CYP1A2, and the above equation (used by the authors) can be viewed as one means
of describing this behavior quantitatively. The numerator in the equation above will always be
greater than zero when there is TCDD in the system (including TCDD derived from either
background exposures or defined external sources). Consequently, the rate of elimination will
correspond to a nonzero value for situations involving TCDD exposures.
It should be noted that CYP\ h2injucaj should always be greater than CYP\ A2/,t(Vt,/for any
CYP1A2-mediated elimination to take place in Eq. 3-21. This will always be the case whenever
TCDD is present in the liver because the induced levels of CYP1A2 are an estimate of "total"
enzyme content at any time point including basal levels. Furthermore, Eq. 3-21 is a
mathematical representation of the induced elimination rate of TCDD by the liver that is
numerically influenced by the scalable parameter kelv. Hence, the mathematical description for
the elimination of TCDD by the liver is dominated by the level of CYP1A2 induction (as
mathematically influenced by the Hill coefficient in Eq. 3-20) and the numerical estimation of
the kelv constant. The interrelationship between the induction Hill coefficient (h in Eq. 3-20)
3-34 DRAFT - DO NOT CITE OR QUOTE
-------
and kelv becomes a critical consideration when data are used to fit both parameters as will be
illustrated in the sensitivity analysis of the PBPK model.
The gestational model included mathematical descriptions for the changes in physiological
parameters such as body weight, cardiac output, and tissue volumes consistent with experimental
observations in pregnant rats. Additionally, this model included a fetal compartment and
considered the transfer of TCDD between the placental and fetal compartments as a
diffusion-limited process (rather than a perfusion-limited) process (see Text Boxes 3-1 and 3-2).5
Text Box 3-1.
Variation of Body Weight with Age: BWTjme(g) = BWinitial x
' OAlxTime ^
v 1402.5 + Time y
( /•>IVilli>llh'r
C ardiac Output: Oc(mL h) = Occ x 601 ^ 1
A factor of 60 corresponds to the conversion of minutes to hours, and 1,000 is the conversion of
body weight from g to kg.
Blood Compartment:
Cb(nmol ml.) =
((Of x m + (Qre x Creb) + (Oli x Clib) + (Opla x Cplcib) + Lymph)) - (Cb x Clru)
Oc
5Diffusion limited, sometimes also known as "membrane limited," means a chemical's movement from one side of
the membrane to the other is limited by the membrane. Thus, the membrane, in this case, is a limiting factor for
uptake. Perfusion limited, also known as "flow limited" indicates that a chemical is so rapidly taken up (e.g., by the
tissue from the blood) that the flow rate is the only limiting factor.
3-35 DRAFT - DO NOT CITE OR QUOTE
-------
Text Box 3-2.
Placenta Tissue Compartment
(a) Tissue-blood subcompartment
dAplab
(nmol h) = Opla(Ca - Cplab) + PApla(Cplab - Cplajree)
dt
Cplab = 4^
Wplab
(b) Tissue cellular matrices
*(m./ h) = PApla(Cplab - Cplafi-ee) - Mp,a ~f" + dAf" ~pla
dt v ; dt dt
Cpla(nmol / mL) =
Wpla
Free TCDD Concentration in Placenta
Cplajree(nmol / mL) = Clpla -
(iCplajree x Ppla +
^ Plabm&x x Cplajree ^
Kdpla + Cplajree
Dioxin Transfer from Placenta to Fetuses
dAPla jet
= (nmol / n) = C lPla fet x Lpla
dt
Dioxin Transfer from Fetuses to Placenta
cf ' nm()l ! h)= Clpla fet X ( je,^r
Fetal Dioxin Concentration (Fetuses 5 = Per Litter)
dAfet, , ,dAPla fet dAfet Pla
—-— (nmol / h) = — —=
dt dt dt
Ajet
Cjet(nmol /h) =
CjetV (nmol / mL) =
Wjet
Cjet
Pjet
3-36
DRAFT - DO NOT CITE OR QUOTE
-------
3.3.4.3.2.3. Parameter estimation
Table 3-8 lists the numerical values of the adult rat and human PBPK models of Emond
et al. (2006; 2005). The values for key input parameters of the rat gestational model are
summarized in Table 3-8 as well as Figure 3-13.
The parameters for the rat model were obtained primarily from Wang et al. (1997) except
that the value of the affinity constant for CYP1A2 was slightly changed from 0.03 to
0.04 nmol/mL to get a better fit to experimental data (Emond et al.. 2004). and the variable
elimination parameter (kelv) was obtained by optimization of model fit to kinetic data from
Santostefano et al. (1998) and others (Emond et al.. 2006; Emond et al.. 2005; Wang et al..
1997). Wang et al. (1997) used measured tissue weights whereas the tissue blood flows and
tissue blood weights were obtained from International Life Sciences Institute (1LS1. 1994). The
partition coefficients (which were similar to those of Leung et al.. 1990; Leung et al.. 1988). the
permeability x area (PA) value for tissues, the dissociation constant for binding to CYP1A2
(ICa2), and the Hill coefficient (h) were estimated using a two-stage process of fitting to
dose-response and time-course data on TCDD tissue distribution (Wang et al.. 1997). In the
initial stage, the experimental data of arterial blood concentrations were used as input to the
individual compartment to estimate the parameters; then, with the values obtained during stage
one as initial estimates, those unknown parameters were re-estimated by solving the entire model
at once using an optimization route (Wang et al.. 1997). The receptor concentrations and
dissociation constant of TCDD bound to AhR were obtained by fitting the model to TCDD tissue
concentration combined with enzyme data reported by Santostefano et al. (1998) whereas the
basal CYP1A2 in liver was based on literature data (Wang et al.. 1997).
The parameters for the human PBPK model were primarily based on the rat model
(Emond et al.. 2006; Emond et al.. 2005; Wang et al.. 1997). Specifically, the blood fraction in
the tissues, the tissue:blood partition coefficients, tissue permeability coefficient, the binding
affinity of TCDD to AhR and CYP, and the maximum binding capacity in the liver for AhR were
all set equal to the values used in the rat model. The species-specific elimination constant, fe/v,
was estimated by fitting to human data (Emond et al.. 2005).
For the gestational rat model, the parameters describing the growth of the placental and
fetal compartments as well as temporal change in blood flow during gestation were incorporated
based on existing data. Exponential equations for the growing compartments were used (see
3-37
DRAFT - DO NOT CITE OR QUOTE
-------
Figure 3-13), except for adipose tissue, for which a linear growth increment based on literature
data was specified. All relevant physiological parameters for the pregnant rat were obtained
from the literature while remaining input parameters were set equal to that of the nonpregnant rat
(obtained from Wang et al.. 1997); see Tables 3-7 and 3-8. The current version of the rat
gestational model contains parameters for variable elimination from Emond et al. (2006; Table 3-
8) and still provides essentially the same predictions as the original publication (Emond et al..
2004V
3.3.4.3.2.4. Model performance and degree of evaluation
The PBPK model of Emond et al. (2006; 2005; 2004) had parameters estimated by fitting
to dose and time-course data, so that the resulting model consistently reproduced available
kinetic data. The same model structure with a single set of species-specific parameters could
reproduce the kinetics of TCDD following various doses and exposure scenarios not only in the
rat but also in humans. The simulations of the PBPK model of Emond et al. (2006) have been
compared with two sets of previously published rat data: blood pharmacokinetics following a
single dose of 10 |ig/kg (the dose corresponding to the mean effective dose for induction of
CYP1A2) (Santostefano et al.. 1998) (see Figure 3-14); and hepatic TCDD concentrations
following chronic exposure to average daily exposures of 3.5 to 125 ng/kg (Walker et al.. 1999)
(see Figure 3-15). It is relevant to note that the PBPK model of Emond et al. (2006. 2004) is
essentially a reduced version of the Wang et al. (1997) model, and it, therefore, provides
simulations of liver and fat concentrations of TCDD that deviated by not more than 10-15% of
those of Wang et al. (1997). The nongestational model of Emond et al. (2004) was calibrated
against kinetic data in liver, fat, blood, and rest of body of female Sprague-Dawley rats given a
single dose of 10 jug TCDD/kg (data from Santostefano et al.. 1996) and in liver and fat of male
Wistar rats treated with a loading dose of 25 ng/kg followed by a weekly maintenance dose of
5 ng TCDDkg by gavage (data from Krowke et al.. 1989).
The gestational rat PBPK model was calibrated against the following kinetic data sets
(Emond et al.. 2004):
3-38
DRAFT - DO NOT CITE OR QUOTE
-------
• TCDD concentration in blood, fat, liver, placenta, and fetus of female Long-Evans rats
given 1, 10, or 30 rigkg, 5 daysAveek, for 13 weeks prior to mating followed by daily
exposure through parturition (Hurst et al.. 2000b);
• TCDD concentration in tissues (liver, fat), blood, placenta and fetus determined on
gestation day (GD) 16 and GD 21 following a single dose of 0.05, 0.8, or 1 (J,g/kg given
on GD 15 to pregnant Long-Evans rat (Hurst et al.. 2000a);
• Maternal and fetal tissue concentrations on GD 9, GD 16, and GD 21 after a single dose
of 1.15 |.ig TCDD/kg given to Long-Evans rats on GD 9 or GD 15 (Hurst et al.. 1998);
and
• Fetal TCDD concentrations determined on GD 19 and GD 21 in rats exposed to
5.6 (.ig TCDD/kg on GD 18 (Li et al.. 2006).
Furthermore, the scaled rat model was shown to be capable of simulating human data
from the Austrian and Seveso subjects (see Figures 3-16 and 3-17). In this regard, it is useful to
note that the computational version of the PBPK model of Emond et al. (2006; 2005) also
contained the necessary equation to transform the model output of blood concentration into
serum lipid-adjusted concentration of TCDD. This conversion is calculated by dividing the
estimated total blood TCDD levels with the product of two constants, the serum portion of total
blood and the lipid content in serum. The human model of Emond et al. (2005; Emond model)
has advantages for improving the TCDD dosimetry used in existing human epidemiological
studies because the model predicts the redistribution of TCDD within the body (to stores in fat
and liver) based on physiological principles. However, because the dose-dependency of
metabolic elimination in the Emond model was not calibrated to human data, it is important to
review the predictions of this model using a database of human observations that is as extensive
as possible and a spread of internal TCDD concentrations that is as wide as possible. Thus,
presented below is a juxtaposition of modeled elimination rates from the Emond model with
observations for two highly exposed Austrian patients (severe intoxication of "unknown origin"
(Geusau et al.. 2001) and 9 of 10 Ranch Hand veterans6 used for the original "validation"
comparisons presented in the Emond et al. (2005)).
Figure 3-18 shows the time course of the declines in TCDD serum concentrations in
two highly exposed Austrian subjects compared with the Emond model results. The comparison
6In preliminary comparisons, the simulation run for the 10th Ranch Hand veteran appeared anomalous and was,
therefore, excluded from this summary.
3-39 DRAFT - DO NOT CITE OR QUOTE
-------
in Figures 3-17 and 3-18 indicates that the Emond model adequately describes the rate of TCDD
elimination for the more highly exposed Austrian patients but predicts a somewhat faster rate of
decline than that observed for the less heavily exposed patient.
Figure 3-19 shows the results of combining the simulated and observed rates of loss for a
group of Austrian and Ranch Hand subjects evaluated by Emond et al. (2005). counting only
one data point per person. The X-axis in this figure is the TCDD serum concentration at the
midpoint of the observations for each subject. The error bars in the figure represent ±1 standard
error. The results of this figure illustrate two points: (1) the Emond model simulation (open
squares) are generally very close to the actual data (solid circles) for the nine Ranch Hand
subjects (clustered toward lower left corner) and one of the two Austrian patients (upper right
corner); and (2) both the Emond model simulation results and the actual data show a linear trend,
and linear regression lines were plotted, respectively, as shown in Figure 3-19.
Table 3-9 presents the results of regression analyses of the observed rates of decline in
relation to the estimated TCDD serum levels at the midpoint of the observations for each subject
in the Ranch Hand study (see Figure 3-19). These results indicate that some appreciable dose
dependency of TCDD elimination is unequivocally supported. However, the central estimate of
the slope of the relationship between the log of the TCDD elimination rate and the log of the
TCDD level is only about 75% of that expected under the Emond et al. PBPK model
(i.e., 0.092-0.123 =0.748).
Overall, the conclusion from the above analysis is that the Emond model is reasonable to
use, but the model might be improved by (1) including the two dose-independent pathways of
elimination documented in the Geusau papers (GI elimination via the feces and loss via the
sloughing of skin cells), and (2) reducing the extent of loss via the dose-dependent metabolism
pathway from the liver (Harrad et al.. 2003; Geusau et al.. 2002) so that overall loss rates for the
average elimination rates from the Ranch Hand veterans are maintained.
3.3.4.3.2.5. Sensitivity analysis of the PBPK model
A sensitivity analysis was performed on each of the animal and human Emond PBPK
models to determine the most sensitive variables. In each case, all input variables in each model
were included in the analysis. For equations where the parameter value varies with age
according to an equation (body weight in all models, liver and adipose tissue fractions in the
3-40
DRAFT - DO NOT CITE OR QUOTE
-------
human models, and fetal weight, placental weight, and placental perfusion in the gestational
models), a constant multiplier of 1.0 was included in each equation; then, for the sensitivity
analysis, this value was varied by a fixed percentage to determine the relative effect of changing
the compartmental weight fractions.
To perform the analysis, a representative dosing protocol was selected for each model to
ensure the analysis was performed in dose ranges that were applicable to the overall health
assessment. For each study modeled, multiple doses were used to investigate model sensitivity
across a dosing range. Table 3-10 shows the dosing protocols selected for each model. For the
human models, doses in the range of the identified reference dose and POD dose discussed in
Section 4 were used in the analysis.
To perform the sensitivity analysis, variable values were varied by fixed percentages one
at a time to determine the associated change in the average whole blood concentration. The
blood concentration averages were calculated in each study in the same manner as in the main
health assessment, as detailed in Appendix E and repeated for convenience in Table 3-10. To
determine the local sensitivity of the whole blood concentration to each variable, the variable
values were increased and decreased from the standard model configuration by 5%. This local
analysis shows the effects of changing the variables by relatively small amounts to account for a
theoretical level of uncertainty in the input parameters. To determine a more global sensitivity of
the whole blood concentrations to each variable, the variable values were increased and
decreased by 50%. In some cases, such a wide change may overestimate the actual uncertainty
in the variable value in the literature; however, such a change is useful in helping to determine
how the model sensitivity may change across large portions of the variable parameter space.
For each percentage change in the variable, the associated percentage change in the
average whole blood concentration was recorded. Then, the elasticity was calculated as the
percent change in the average whole blood concentration divided by the percent change in the
variable value. Thus, variables where the magnitude of the elasticity is greater than 1 will induce
a change of greater than 5% in the whole blood concentration when the variable value is changed
by 5%. The sign of the elasticity indicates whether the whole blood concentration is positively
or negatively correlated with the variable. The elasticities were examined, and a value of 0.1
was selected as a threshold to determine the most sensitive variables in each model. This value
tended to represent a limit, with a cluster of variables having higher magnitude elasticities and
3-41
DRAFT - DO NOT CITE OR QUOTE
-------
the remaining variables having much lower elasticities. Variables were then ranked according to
the magnitude of the elasticity in the case where the variables were increased by 5% for
presentation.
Table 3-11 shows the most sensitive variables for the rat and mouse nongestational
models and rat and mouse gestational models for the low and high doses when variables were
increased by +5%. The associated elasticities are shown in each case. The only variable with
elasticity above one is the Hill coefficient (h in Eq. 3-20). The other most sensitive variables are
associated with the overall dioxin elimination/sequestration rate, including the CYP1A2
induction rates, the liver weight, the binding capacity and affinity, and the gastric and intestinal
excretion rates. For the gestational model dosing protocols, the Hill coefficient remains the most
sensitive variable, but the elasticity decreases compared with the nongestational analysis.
Otherwise, many of the most sensitive variables remain those associated with elimination.
Additional parameters related to the adipose tissue blood flow and with the adipose diffusional
permeability fraction are also relatively sensitive.
Table 3-12 shows the most sensitive variables for the human gestational and
nongestational models. The additional variables associated with the adipose compartment
partition coefficient, the body weight, and the fractional adipose tissue volume are also relatively
sensitive variables at the reference dose and POD dose compared with the animal models. For
all models, the elasticities are relatively similar across the different doses evaluated.
In order to observe the difference between the local and global elasticities, Figures 3-20
and 3-21 show the elasticities for the most sensitive variables in the human nongestational model
for the POD dose and reference dose, respectively. In general, the elasticities are similar across
the different percentage changes in variable values that were tested. Changes in variables by
-50% tend to lead to the greatest elasticities. Changing the variable values by +5% and -5%
lead to almost the same elasticities for nearly all the variables. These same conclusions hold for
all the other models and doses as well.
Of the variables to which the blood concentrations are most sensitive, most of the
variables are either derived from Wang et al. (1997) or are optimized (see Table 3-8). For the
human model, parameters set equal to values in the rat model may be subject to particular
uncertainty. In particular, the AhR and CYP1A2 induction parameters typically were based on
the rat model parameters. The exception is CYP1A21EMAX, the maximum induction of
3-42
DRAFT - DO NOT CITE OR QUOTE
-------
CYP1A2, which is an optimized parameter. The variable elimination rate, kelv, and the intestinal
excretion, KST, are also both optimized against data. For variables that are optimized, a
sensitivity analysis that varies each parameter one at a time may overestimate the associated
model uncertainty associated with the variable. A change in KST, for example, would
necessitate a commensurate change in the other optimized variables in order to suitably capture
the comparison data, and the overall changes in the blood concentrations might be small.
The most sensitive variable in all the models is the Hill parameter. The elasticity is high
in part because the Hill parameter is an exponent; thus, small changes in the value can lead to
larger changes in the whole blood concentration. However, as stated above, any change in the
Hill parameter would also necessitate changes in optimized variables in order to maintain an
adequate fit with the data. The next section explores the effect of changing the Hill parameter
and the effect of changing the CYP1A2 induction parameters on the model fits to literature data.
3.3.4.3.2.6. Further uncertainty analysis of the Hill coefficient and CYP1A2 induction
parameters
As illustrated by the sensitivity analysis of the PBPK model, the predicted TCDD blood
concentrations are very sensitive to the Hill coefficient (h) as described in Eq. 3-20. This
parameter is included in the mathematical description for the induction of the CYP1A2.
Therefore, the best type of data needed to estimate an in vivo value for this constant would be
time-course levels of hepatic CYP1A2 in response to TCDD exposure. This type of data is only
available in experiments conducted in animals. The PBPK model adopted a value of 0.6 for this
parameter based on the earlier reported models by Wang et al. (2000) and Santostefano et al.
(1998). In both cases, the value of 0.6 used for the Hill coefficient (the model parameter Hill) in
the model was fit to describe the temporal relationship between TCDD exposure and
CYP1A2-induction levels in animals. Note that the value of 0.6 for Hill indicates supralinear
behavior at low exposure levels, which translates to a supralinear relationship between oral
intake and blood TCDD concentrations.
For humans, the only data available to calibrate the in vivo model parameters are blood
levels of TCDD. Predicted TCDD blood levels are influenced by the Hill coefficient when it is
implicitly included in the description for the hepatic elimination of TCDD by induced levels of
CYP1A2 as described in Eq. 3-21. However, as was illustrated earlier, the elimination of TCDD
by the liver is also influenced by the numerical optimization of the kelv constant in the same
3-43 DRAFT - DO NOT CITE OR QUOTE
-------
equation. Therefore, estimation of the Hill coefficient using human blood data is highly
dependent on the simultaneous estimation of kelv.
In order to estimate the interdependence of Hill and kelv and to investigate the behavior
of the Emond human PBPK model in the absence of supralinearity, EPA calibrated the model to
several human data sets after setting Hill to 1 and varying kelv. A Hill coefficient of 1 results in
low-dose linearity, where supralinear behavior is first eliminated. However, EPA does not
consider a Hill value of 1 necessarily to be a plausible replacement for the model variable of 0.6;
it is just being used to investigate the behavior of the model as a sensitivity analysis. The data
sets are TCDD serum concentrations (LASC) over time for four individuals: two Austrian adult
females (Geusau et al.. 2002) and two Italian (Seveso) males—a 6-year-old and a 50-year-old
(Needham et al.. 1998); the data are presented in Tables 3-13 and 3-14. The results of the
simulations are shown in Figure 3-22 and Table 3-15. For each data set, the simulation was run
four times—once with the default model parameters {Hill = 0.6, kelv = 0.0011), once with
Hill =1.0 and kelv unchanged, once with Hill = 0.6 and kelv optimized for best fit to the data,
and once with Hil =1.0 and kelv optimized. In each case, the initial dose (model parameter
doseiv), assuming a single instantaneous exposure at the time of first serum measurement, was
optimized for best fit; the exposure in this case would be a simulation of the body burden at the
time, as the actual exposure scenario is unknown. In all cases, simply changing the value of Hill
resulted in poor fits. Optimizing kelv with Hill set to either to 0.6 or 1 yields much better fits, as
would be expected, with both values fitting the data equally well when the inter-related
parameter, kelv, is optimized.
EPA also investigated the impact of alternate values for other model parameters related to
the CYP1A2 induction algorithm. Budinsky et al. (2010) reported an in vitro temporal
relationship between CYP1A2 induction and TCDD levels in human and rat primary
hepatocytes. Budinsky et al. (2010) used the CYP1A2 induction data to estimate Hill function
constants, such as baseline, fold, and maximal CYP1A2 mRNA inductions. Using their data, an
estimate for the human in vivo baseline, fold, and maximal response of CYP1A2 induction can
be approximated as illustrated in Eq. 3-22 and 3-23:
CYP1A2,
¦basalhumaninvitro
x CYP1A2
basal animai
CYP1A2,
U 7
¦ basal animalinvitro
invivo
(Eq. 3-22)
3-44 DRAFT - DO NOT CITE OR QUOTE
-------
and
= CYP1A2 basal, . , . .
— human equivalentinvivo
fcYPiA2Maxhumanm\ x CYP1A2 (Eq. 3-23)
\CYPlA2MaXammalimntro) ammalinvivo V 4 '
= CYP1A2 Max, . , . .
— human equivalent invivo
The values used in these equations are shown in Table 3-16.
The calculated in vivo human CYP1A2 baseline, fold, and maximal induction response,
with their corresponding minimum and maximum values, are then used in the PBPK model to
estimate mean, minimum, and maximum blood levels in comparison to data for two Austrian
cases, and the Seveso cohort. This analysis was done with Hill set to 0.6 and optimizing kelv and
doseiv for the data sets in Tables 3-13 and 3-14. Results of the simulations are shown in
Figure 3-23 and Table 3-17.
An attempt to directly use the in vitro values of the Hill function estimated in the
Budinsky et al. (2010) in the PBPK model was not successful in simulating blood levels in
Figure 3-23. The failure in using these values directly may be a result of the usual in vitro-to-in
vivo extrapolation complications such as in vitro cellular competency to exhibit toxicological
response comparable to the in vivo ones, and TCDD media to cell sequestration. It is also
important to note that the in vitro preparations in the Budinsky et al. (2010) came from a limited
set of five female subjects. Average and standard variation levels obtained from this set of
human subjects cannot be representative of overall human population.
It is clear from the results shown in Figures 3-22 and 3-23, that several different
combinations of CYP1A2 induction parameters can be used to simulate the data well. This
process illustrates the interdependencies of these parameters when in vivo blood levels in
humans are the only source of data to estimate them.
The impact of varying these parameters on model predictions of human oral intakes
corresponding to a range of lifetime average serum concentrations is shown in Table 3-18. The
range of concentrations was chosen to be representative of human intakes of interest for the RfD
derivation in Section 4. Comparing the optimized simulations for the alternative Hill values
shows that, for these data sets, changing Hill to 1 decreases the modeled intakes for the TCDD
serum concentrations in this range by about 70-85%. Using the alternative parameters estimated
3-45
DRAFT - DO NOT CITE OR QUOTE
-------
from Budinsky et al. (2010) results in 40-60% lower intakes than for the standard parameters
(optimized kelv). Thus, it would appear that, although the Hill value of 0.6 results in a
supralinear relationship between TCDD intake and serum concentrations in the Emond model,
eliminating the supralinear behavior does not result in higher predicted intakes for lower TCDD
serum concentrations, as might be expected. However, strong conclusions cannot be made from
these results because the data used for the optimization are not ideal in at least two respects:
(1) they only address CYP1A2 dynamics indirectly, and (2) there are only four data sets, and
they are not necessarily representative of the entire population.
3.3.4.3.2.7. Confidence in PBPK model predictions of dose metrics
The PBPK model facilitates prediction of absorbed dose, body burden, and blood
concentration of TCDD for oral exposures in adult humans and rats (adult and developing) with
high confidence (see Table 3-19). The model output of blood concentration can be normalized to
lipid content representative of the study group (species, sex, age, lifestage, and diet). However,
the PBPK model of Emond et al. (2006; 2005; 2004) does not simulate plasma and erythrocyte
TCDD concentrations separately, and it predicts tissue concentrations on the basis of
tissue:whole blood partition coefficients and not on the basis of serum lipid-normalized values.
The reliability of this model for simulating the liver concentration of TCDD in rats is
considered to be high, but it is considered to be medium for humans. Although empirical data on
bound or free concentrations were not used to evaluate model performance in humans, the
biological phenomena (consistent with available data) related to the hepatic sequestration,
enzyme induction, and dose-dependent elimination are described in the model. This is one of the
situations where PBPK models are uniquely useful; that is, they permit the prediction of system
behavior based on understanding of the mechanistic determinants, even though the required data
cannot be directly obtained in the system (e.g., bound concentrations in the liver of exposed
humans). For these dose measures (i.e., bound concentration and total liver concentration), the
level of confidence can be further improved or diminished by the outcome of sensitivity analysis.
In this regard, the results of a focused sensitivity analysis indicate that the most sensitive
parameters of the human model are among the most uncertain (i.e., those parameters for which
estimates were not obtained in humans) with respect to prediction of liver TCDD concentration,
contrary to the animal model (see Section 3.3.6).
3-46
DRAFT - DO NOT CITE OR QUOTE
-------
With respect to the mouse model, however, the level of confidence is low to medium,
given that it has not been verified extensively with blood, body burden, or tissue concentration,
time-course, or dose-response data. However, the mouse PBPK model, based on the rat model
that has been evaluated with several PK data sets, has been shown to reproduce well the limited
mouse liver kinetic data (see Figures 3-24 through 3-31; Boverhoff et al.. 2005). The same
model structure has been used for simulating kinetics of TCDD in humans successfully. Overall,
the adult mouse model, given its biological basis combined with its ability to simulate TCDD
kinetics in multiple species, is considered to exhibit a medium level of confidence for simulating
dose metrics for use in high to low dose extrapolation and interspecies (mouse to human)
extrapolation. Even though similar considerations are applicable to gestational model in mice,
the confidence level is considered to be low because very limited comparison with empirical data
has been conducted (see Figure 3-31). Despite the uncertainty in these predictions, the scaled rat
gestational model, given its biological and mechanistic basis, might be of use in predicting dose
metrics in these groups that might form the basis of PODs in certain key studies.
3.3.4.4. Applicability of PK Models to Derive Dose Metrics for Dose-Response Modeling of
TCDD: Confidence and Limitations
Both the CADM and PBPK models describe the kinetics of TCDD following oral
exposure to adult animals and humans by accounting for the key processes affecting kinetics,
including hepatic sequestration phenomena, induction, and nonlinearity in elimination, and
distribution in adipose tissue and liver. Both models can be used for estimating body burdens
and serum lipid adjusted concentrations of TCDD. However, there are several differences
between these two models. The PBPK model calculates the free and bound concentrations of
TCDD in the intracellular subcompartment of tissues. The total or receptor-bound
concentrations in liver are unambiguous and more easily interpretable with the PBPK model than
with the CADM model. In addition, the PBPK model computes bound and total concentrations
as a function of the free concentration in the intracellular compartment of the tissue. By contrast,
the CADM model simulates the total concentration based on empirical consideration of hepatic
processes. Consequently, the amount of TCDD bound to AhR or CYP1A2 cannot be simulated
with the CADM model. The CADM model computes only the total TCDD concentration in liver
and describes TCDD elimination through partitioning from circulating lipids across the lumen of
the large intestine into the feces, while the PBPK model accounts for this process empirically
3-47 DRAFT - DO NOT CITE OR QUOTE
-------
within its hepatic elimination constant. Elimination of TCDD via skin, a minor process, is not
described by either model. Thus, dose-response modeling based on body burden of TCDD in
adult animals and humans can be conducted with either of the models, provided the duration of
the experiment is at least 1 month, due to limitations in the CADM model. As shown in Figure
3-32, the predicted slope and body burden over a large dose range are quite comparable
(generally within a factor of two).
Results of simulations of serum lipid concentrations or liver concentrations vary for the
two models to a larger extent (up to a factor of 7), particularly for simulations of short duration.
These differences reflect two characteristics of the PBPK model: first, quasi-steady-state is not
assumed in the PBPK model; second, the serum lipid composition used in the model is not the
same as the adipose tissue lipids. The CADM model does not account for differential solubility
of TCDD in serum lipids and adipose tissue lipids, nor does it account for the diffusion-limited
uptake by adipose tissue. Therefore, the PBPK model would appear to be superior to the CADM
model with respect to the ability to simulate serum lipid and tissue concentrations during
exposures that do not lead to the onset of steady-state condition in the exposed organism.
The CADM model is less complex than the PBPK model and has fewer parameters.
Because the CADM model is constructed by fitting to data, its performance is likely to be
reliable for the range of exposure doses, species, and life stages from which the parameter
estimates were obtained. On the other hand, the PBPK model structure and parameters are
biologically based and can be adapted for each species and life stage. Accordingly, the PBPK
model has been adapted to simulate the kinetics of TCDD in the human fetus and in pregnant
rats, as well as in adult humans and rats (Emond et al.. 2006; Emond et al.. 2005; Emond et al..
2004). The time step for calculation and dosing in the CADM model corresponds to 1 month.
This requirement represents a constraint in terms of the use of this model to simulate a variety of
dosing protocols used in animal toxicity studies. This requirement, however, is not a constraint
with the PBPK models. So, either model would appear to be useful when simulating the body
burden and serum lipid concentrations following a longer duration of exposure,; but the PBPK
model would be preferred for simulating alternative dose metrics of TCDD (e.g., blood
concentration, total tissue concentration, bound concentration) for various exposure scenarios
(including single dose studies), routes, and life stages in the species of relevance, to TCDD
dose-response assessment, particularly, mice, rats, and humans.
3-48
DRAFT - DO NOT CITE OR QUOTE
-------
Two minor modifications, to enhance the biological basis, were made to the PBPK model
of Emond et al. (2006). before its use in the computation of dose metrics for TCDD. The first
one involved the recalculation of the volume of the rest of the body as follows:
WREO = (0.91 - (WLIBO x WLIO + WFBO x WFO + WLIO + WFO)/(I + WREBO)) (3-24)
where
WREO = weight of cellular component of rest of body compartment (as fraction of
body weight);
WLIO = weight of cellular component of liver compartment (as fraction of body
weight);
WFO = weight of cellular component of fat compartment (as fraction of body
weight);
WREBO = weight of the tissue blood component of the rest of body compartment (as
fraction of body weight);
WLIBO = weight of the tissue blood component of the liver compartment (as fraction
of body weight); and
WFBO = weight of the tissue blood component of the fat compartment (as fraction of
body weight).
In the original code, the weight of the rest of body compartment was calculated as the
difference between 91% of body weight and the sum total of the fractional volumes of blood,
liver tissue (intracellular component), and adipose tissue (intracellular component). The blood
compartment in the PBPK model is not explicitly characterized with a volume; as a result, the
total volume of the compartments is less than 91%. The recalculations shown above were used
to address this problem. Given the very low affinity of TCDD for blood and rest of the body,
reparameterizing the model resulted in less than a 1% change in output compared to the
published version of the PBPK model for chronic exposure scenarios (Emond et al.. 2006).
The second minor modification related to the calculation of the rate of TCDD excreted
via urine. The original model code computed the rate of excretion by multiplying the urinary
clearance parameter with the concentration in the rest of the body compartment. Instead, the
code was modified to use the blood concentration in this equation. This resulted in the
3-49
DRAFT - DO NOT CITE OR QUOTE
-------
re-estimation of the urinary clearance value in the rat and human models, but it did not result in
any significant change in the fit and performance of the original model.
The revised parameter estimates of the rat, mouse, and human models are captured in
Table 3-8 with a footnote.
3.3.4.5. Recommended Dose Metrics for Key Studies
The selection of dose metrics for the dose-response modeling of key studies is largely the
result of (1) the relevance of a dose metric on the basis of current knowledge of TCDD's
mechanism of action for critical endpoints and (2) the feasibility and reliability of obtaining the
dose metric with available PK models. Secondarily, the goodness-of-fit of the dose-response
models (which reflects the relationship of the selected internal dose measures to the response)
can be used to inform selection of the most appropriate dose metric for use in deriving TCDD
toxicity values.
Body burden—even though this metric is based on mechanistic considerations—is a
somewhat distant measure of dose with respect to target tissue dose, and this metric represents
the "overall" average concentration of TCDD in the body. However, a benefit of body burden is
that this metric represents a dose measure for which the available PK models can provide highly
certain estimates. Thus, the overall confidence associated with the use of body burden in TCDD
assessment is categorized as medium.
The confidence in the ability of PK models to simulate blood concentration as a dose
metric is high, given that the models have been shown to consistently reproduce whole blood (or
serum lipid-normalized) TCDD concentration profiles in both humans and rats. Considering the
facts that the PBPK models simulate whole blood rather than the serum lipid-normalized
concentrations of TCDD and that the study-specific values of serum lipid content are not known
with certainty, it is preferable to rely on TCDD blood concentrations as the dose metric. The
blood concentrations, if intended, can be normalized on the basis of appropriate total lipid levels.
However, based on mechanistic considerations, the confidence in their use would be somewhat
lower for hepatic effects. This conclusion reflects the concern regarding the inconsistent
relationship between the two variables with increasing dose levels and the fraction of
steady-state attained at the time of observation. For other systemic effects related to tissue
concentrations, the confidence in the use of TCDD serum or blood concentration is high,
3-50
DRAFT - DO NOT CITE OR QUOTE
-------
particularly for chronic exposures, given the absence of data on organ-specific nonlinear
mechanisms. In general, the tissue concentration typically cannot be calculated as a reliable dose
metric with either the CADM or the Emond models. One exception is the use of the Emond
PBPK models to estimate levels in liver, a metric that is relevant based on MOA considerations.
However, it is noted that the hepatic TCDD level encompasses free and bound TCDD, and it is a
highly complex entity for dose metric considerations. Finally, the AhR-bound concentration
may be evaluated for receptor-mediated effects. This dose metric can be obtained by PBPK
models, although uncertainties associated with the lack of data for this dose metric render it to be
of low confidence (see Table 3-19). The alternative dose metrics for dose-response modeling of
TCDD selected on the basis of MOA and PK modeling considerations are summarized in
Tables 3-20 and 3-21.
These measures of internal dose can be obtained as peak, average, integral (AUC), or
terminal values. For chronic exposures in rodents (ca. 2 years), the terminal and average values
would be fairly comparable under steady-state conditions. For less-than lifetime exposures,
however, the terminal and average values will differ, and, therefore, an overall average or
integrated value (AUC) would be more appropriate. Similarly, for developmental exposures,
these alternative dose metrics can be obtained with reference to the known or hypothesized
exposure window of susceptibility.
3.3.5. Uncertainty in Dose Estimates
3.3.5.1. Sources of Uncertainty in Dose Metric Predictions
3.3.5.1.1. Limitations of available PK data
3.3.5.1.1.1. Animal data
The available animal data relate to blood, liver, and adipose tissue concentrations for
certain exposure doses and scenarios. Although these data are informative regarding the dose-
and time-dependency of TCDD kinetics for the range covered by the specific studies (see
Section 3.3.2), they do not provide the peak, average, terminal, or lipid-normalized values of
dose metrics associated with the key studies selected for this assessment. The limited available
animal PK data are useful, however, in the evaluation of the pharmacokinetic models (see
Section 3.3.4).
3-51
DRAFT - DO NOT CITE OR QUOTE
-------
3.3.5.1.1.2. Human data
The human data on potential dose metrics are restricted to the serum lipid-adjusted
TCDD concentrations associated with mostly uncharacterized exposures (see Sections 3.3.2 and
3.3.3). While these data are useful in estimating half-lives in exposed human individuals, they
do not provide estimates of hepatic clearance or reflect target organ exposure. Some autopsy
data have been used to infer the partition coefficients; however, these data were collected
without quantification of the temporal nature of TCDD uptake (see Section 3.2). Despite the
limitations associated with the available human data, there has been some success in using these
data to infer the half-lives and elimination rates in humans using pharmacokinetic models
(Emond et al.. 2006; Avlward et al.. 2005b; Carrier et al.. 1995a).
3.3.5.1.2. Uncertainties associated with model specification
Uncertainty associated with model specification should be viewed as a function of the
specific application, such as interspecies extrapolation, intraspecies variability, or
high-dose-to-low-dose extrapolation. Because the use of pharmacokinetic models in this
assessment is limited to interspecies extrapolation and high-dose-to-low-dose extrapolation, it is
essential to evaluate the confidence in predicted dose metrics for these specific purposes. For
interspecies extrapolation, the PBPK and CADM models calculate differences in dose metric
between an average adult animal and an average adult human. Both models have a biologically
and mechanistically relevant structure along with a set of parameters with reasonable biological
basis and reproduce a variety of pharmacokinetic data on TCDD in both rodents and humans.
These models possess low uncertainty with respect to body burden, blood, and TCDD/serum
(lipid) concentration for the purpose of conducting rat to human extrapolation. However, for
other dose metrics, such as free, total, or bound hepatic concentrations, the uncertainty is higher
in the CADM model compared to the PBPK model due to model specification differences related
to the mechanisms of sequestration and induction in the liver (see Section 3.3.3).
For the purpose of high-dose-to-low-dose extrapolation in experimental animals,
confidence in both models is high with respect to a variety of dose metrics (see previous
discussion). The high confidence results from the use of the PBPK models to reproduce a
number of data sets covering a wide range of dose levels in rodents (i.e., rats, mice) including the
dose ranges of most of the key toxicological studies. Given that the TCDD levels during and at
3-52
DRAFT - DO NOT CITE OR QUOTE
-------
the end of exposures were not measured in most of the key studies, use of the PBPK models is
preferred because these models account for dose-dependent elimination, induction, and
sequestration. Despite the empirical nature of the specification of these key processes in PBPK
models, they essentially reproduce the dose-dependent behavior in rodents, supporting their use
in deriving dose metrics for dose-response modeling of TCDD. Overall, the confidence in the
use of the alternative dose metrics (identified in Table 3-19) is greater than the confidence in the
use of administered dose for TCDD, for relating to the concentration within tissues to produce an
effect. The administered dose does not take into account interspecies differences in the volume
of distribution and clearance or the complex nonlinear processes determining the internal dose.
The PBPK model of Emond et al. (2006) could benefit from further refinement and
validation, including a more explicit consideration of dose-independent elimination pathways.
As indicated in Section 4, there is some uncertainty associated with the way the elimination of
TCDD is described in the existing human PBPK model. The current model essentially treats all
TCDD elimination as related to dose-dependent metabolism in the liver. In this regard, the
classical and more recent PK data on TCDD may be useful in further improving the confidence
in their predictions. However, it is likely that there is dose-independent elimination of TCDD
via feces and, to a lesser extent, skin; juxtaposition of available elimination rate data with the
PBPK model predictions suggests that the current PBPK model modestly overestimates the
dose-dependency of overall TCDD elimination. (The central estimate of the slope of the
relationship between the log of the TCDD elimination rate and the log of the TCDD level is only
about three-fourths of that expected using the unmodified PBPK model). Emond et al. (2005)
acknowledge that the model did not describe the elimination of TCDD from the blood into the
intestines, but it indirectly accounted for this phenomenon with the use of the optimized
elimination rate.
3.3.5.1.3. Impact of human interindividual variability
The sources and extent of human variability suggested by the available data are presented
in Section 3.3.3, although there is some discussion of the impact of individual differences in
body fat content. The CADM model facilitates the simulation of body burden and serum lipid
concentrations on the basis of BMI and tissue weights of people, and the PBPK model simulates
alternative dose metrics in the fetus and in pregnant animals in addition to adult animals and
3-53
DRAFT - DO NOT CITE OR QUOTE
-------
humans. However, neither of these models has been parameterized for simulation of population
kinetics and distribution of TCDD dose metrics. Therefore, at the present time, a quantitative
evaluation of the impact of human variability on the dose metrics of TCDD is not feasible, and
dose metric-based replacement of the default interindividual factor has not been attempted.
3.3.5.2. Qualitative Discussion of Uncertainty in Dose Metrics
The usefulness of the CADM and PBPK models for conducting dose-response modeling
(rodent bioassays), interspecies (rodent to human) and intraspecies (high-dose to low-dose)
extrapolations is determined by their reliability in predicting the desired dose metrics. The
confidence in the model predictions of dose metrics is dictated by the extent to which the model
has been verified with empirical data relevant to the dose metric, supplemented by sensitivity
and uncertainty analyses. Analysis of sensitivity or uncertainty has not been conducted with the
CADM model. For the PBPK model, Emond et al. (2006) published the initial results from
sensitivity analyses of acute exposure modeling (see Section 3.3.3). One of the objectives of a
sensitivity analysis that is of highest relevance to this assessment is the identification of the most
critical model parameters with respect to the model output (i.e., dose metric).
If the model simulations have only been compared to entities that do not correspond to
the moiety representing the dose metric, or if the comparisons have only been done for some but
not all relevant dose levels, routes, and species, then the reliability in the predictions of dose
metric can be an issue. The extent to which model results are uncertain will depend largely upon
the extent to which the dose metric is measurable (e.g., serum concentrations of TCDD) or
inferred (e.g., AhR-bound TCDD concentration).
With respect to TCDD body burden, whole-liver and blood concentration predictions in
the rat model, which are well-calibrated with measured data, uncertainty is relatively low.
Therefore, the need for sensitivity and uncertainty analysis is less critical, and confidence in
these dose metrics is high. For those dose metrics that are not directly measurable or are less
easily verified by available calibration methods, such as free-liver and AhR-bound
concentrations, sensitivity and uncertainty analyses are crucial for assessing the reliability of
model predictions, and confidence is low. For the human model, calibration is largely dependent
on blood (LASC) TCDD measurements, which are much less extensive than for the rat model.
Because the blood measurements are reported as LASC, uncertainty and variability in
3-54
DRAFT - DO NOT CITE OR QUOTE
-------
serum:blood and fat:serum ratios also are a factor when evaluating the adequacy of the
whole-blood TCDD metric. Furthermore, the human data are mostly representative of much
higher exposures than the environmental exposures of interest to the EPA. Because of these
additional uncertainties, only medium confidence can be held in the human model whole-blood
TCDD concentration predictions at higher exposures (observed effect range) and low-to-medium
confidence at lower exposures (background exposure range).
Sensitivity analysis for the Emond rat PBPK model predictions of liver TCDD
concentration indicated that hepatic CYP1A2 concentration is the most sensitive parameter
(Emond et al.. 2006). For the Emond human PBPK model, the absorption parameters, basal
concentration of CYP1A2, and adipose tissue:blood partition coefficients were identified as
highly sensitive parameters.
Confidence in the Emond rat and human PBPK models at high exposures is medium for
the purpose of rat-to-human extrapolation based on blood concentrations, given that the key
human model parameters are both sensitive and uncertain; confidence is low for lower
exposures. Conversely, confidence in the use of AhR-bound TCDD is low because of the large
uncertainty in the fraction of AhR-bound TCDD in the liver.
With regard to the predictability of body burden, the absorption and excretion parameters
were among the sensitive parameters in the rat. Several other parameters were also identified as
being sensitive in humans. Despite the sensitivity to these parameters and the uncertainty
associated with individual parameter estimates, the overall confidence in the model predictions
of body burden appears to be high given the reproducibility of empirical data on tissue burdens
and blood concentrations of TCDD in various experiments by both models. Similar conclusions
can be drawn for blood concentration of TCDD predicted by the PBPK model, except that the
assigned value of blood (serum) lipid content will have additional impact on this dose metric to
the extent that the calibration data were in terms of LASC. Variability of total lipid levels and
variability of the contribution of phospholipids and neutral lipids to the total lipid pool across
species, lifestage, and study groups is to be expected (Bernert et al.. 2007; Poulin and Theil.
2001).
Both conceptual (biological) relevance and prediction uncertainty are important in the
choice of dose metric for dose-response modeling and interspecies extrapolation. Conceptual
relevance has to do with how "close" the metric is to the observed effect, taking into account
3-55
DRAFT - DO NOT CITE OR QUOTE
-------
both the target tissue and the MOA. In this context, a greater degree of confidence is held for
dose metrics that are more proximate to the event (i.e., specific effect). Prediction uncertainty
reflects the lack of confidence in the model predictions of dose metrics. Tables 3-22 and 3-23
provide a qualitative ranking of the importance and magnitude of each dose metric with respect
to these two sources of uncertainty. Conceptual relevance is low for the use of administered
dose in dose-response modeling because known (nonlinear) physiological processes are ignored;
conversely, conceptual uncertainty is much lower for use of internal dose metrics more proximal
to the affected organs.
Table 3-22 presents a cross-walk of relevance, uncertainty, and overall confidence
associated with the use of various dose metrics for dose-response modeling of TCDD. Using
professional judgment, EPA ranked its confidence in PBPK models as low, medium, or high (or
not applicable) based on model simulations of administered dose, absorbed dose, body burden,
serum lipid concentration, total tissue (liver) concentration, and receptor occupancy. As shown
in Table 3-22, blood/serum levels have the highest overall confidence (medium), followed by
body burden (medium to low) for application in dose-response modeling. When using the mouse
PBPK model along with the human model (see Table 3-23), the contribution of the prediction
uncertainty to the overall uncertainty increases due to the limited comparison of the mouse
model simulations with empirical data.
3.3.6. Use of the Emond PBPK Models for Dose Extrapolation from Rodents to Humans
EPA has selected the Emond et al. (2006; 2005; 2004) PBPK models, as modified by
EPA for this assessment, for establishing toxicokinetically equivalent exposures in rodents and
n
humans. The 2003 Reassessment (U.S. EPA. 2003) presented a strong argument for using the
relevant tissue concentration as the effective dose metric. However, no models exist for
estimation of all relevant tissue concentrations. Therefore, EPA has decided to use the
concentration of TCDD in blood as a surrogate for tissue concentrations, assuming that tissue
concentrations are proportional to blood concentrations. Furthermore, because the RfD is
necessarily expressed in terms of average daily exposure, the blood concentrations are expressed
as averages over the relevant period of exposure for each endpoint. Specifically, blood
7The models will be referred to hereafter as the "Emond human PBPK model" and the "Emond rodent PBPK
model," with variations when referring to individual species or components (e.g., gestational).
3-56 DRAFT - DO NOT CITE OR QUOTE
-------
concentrations in the model simulations are averaged from the administration of the first dose to
the administration of the last dose plus one dosing interval (time) unit in order to capture the
peaks and valleys for each administered dose. That is, for daily dosing, 24 hours of TCDD
elimination following the last dose is included in the average (the modeling time interval is
1 hour); for a weekly dosing protocol, a full week is included. In addition, because of the
accumulation of TCDD in fat and the large differences in elimination kinetics between rodent
species and humans, exposure duration plays a much larger role in TK extrapolation across
species than for rapidly eliminated compounds. Because of these factors, EPA is using discrete
exposure scenarios that relate human and rodent exposure durations. The use of discrete
exposure scenarios was introduced previously in Section 3.4.4.2 describing first-order kinetic
modeling and is further described in the following paragraphs. This section concludes with a
quantitative evaluation of the impact of exposure duration on the rodent-to-human TK
extrapolation from both the human and rodent "ends" of the process.
Figure 3-33 shows the TCDD blood concentration-time profile for continuous exposure
at 0.01 ng/kg-day, as predicted by the Emond human PBPK model, and the target TCDD
concentrations corresponding to the three discrete exposure scenarios used by EPA in this
document. The target concentrations are those that would be identified in the animal bioassay
studies that correspond to a particular POD (no-observed-adverse-effect level,
lowest-observed-adverse-effect level, or benchmark dose lower confidence bound) established
for that bioassay. That is, the target concentrations represent the toxicokinetically equivalent
internal exposure to be translated into an equivalent human intake (or HED).
For the lifetime exposure scenario, the HED is "matched" to the lifetime average TCDD
blood concentration from a lifetime animal bioassay result by determining the continuous daily
intake that would result in that average blood concentration for humans over 70 years. A table
for converting lifetime-average blood concentrations and other internal dose metrics to human
intake is presented in Appendix C.4.
For the gestational exposure scenario, the effective TCDD blood concentration (usually
the peak) determined for the particular POD in a particular developmental study is matched to
the average TCDD blood concentration over the gestational portion of the human gestational
exposure scenario. The HED is determined as the continuous daily intake, starting from birth
that would result in that average blood concentration over the 9-month gestational period for a
3-57
DRAFT - DO NOT CITE OR QUOTE
-------
pregnancy beginning at 45 years of age. The choice of 45 years as the beginning age of
pregnancy is conservative in that the daily exposure achieving the target blood concentration is
smaller than for pregnancies occurring earlier in life (e.g., a pregnancy beginning at 30 years of
age). A table for converting average gestational blood concentrations and other internal dose
metrics to human intake for the 45-year-old pregnancy scenario is presented in Appendix C.4.
Also, a comparison of the 45-year old pregnancy scenario to one beginning at age 25 is presented
in Table 3-24. Using the 25 year-old pregnancy scenario increases the HED by 30 to 60% for
typical animal bioassay PODs (3 to 30 ng/kg).
For a less-than-lifetime exposure, the average TCDD blood concentration over the
exposure period in the animal bioassay associated with the POD is matched to the average over
the 5-year period that includes the peak concentration (58 years for an intake of 0.01 ng/kg-day).
The HED is determined as the continuous daily intake that would result in the target
concentration over peak 5-year period. The use of the peak is analogous to the approach in the
2003 Reassessment, where the terminal steady-state body burden played the same role. The
5-year average over the peak is taken to smooth out sharp peaks and more closely approximate a
plateau. The choice of peak is health protective because humans of any age must be protected
for short-term exposures, and the daily intake achieving a given TCDD blood concentration is
smallest when matched to the peak exposure as opposed to an average over shorter durations.
Thus, target concentrations for any exposure duration of less-than-lifetime must be averaged
backwards from the end of the lifetime scenario, rather than from the beginning. The only
exception would be if the short-term endpoints evaluated in the animal bioassay were associated
with a specific life stage (such as for the gestational scenario). Note that this scenario lumps all
exposures from 1 day to over 1 year in rodents into the same less-than-lifetime category.
Conceptually, duration-specific scenarios could be constructed by defining equivalent rodent and
human exposure durations. However, for the most part, defining duration equivalents across
species is a somewhat arbitrary exercise, not generally based on physiologic or toxicological
processes, but relying primarily on fraction-of-lifetime conversions. EPA defines "lifetime"
exposure as 2 years and 70 years for rodents and humans, respectively. So, a half-lifetime
equivalence of 1 year in rodents and 35 years in humans is defined easily. Also, considering a
subchronic exposure to be 10-15% of lifetime, leads to an equivalence of 90 days in rodents and
7-10 years in humans. However, in the practical sense with respect to the Emond human PBPK
3-58
DRAFT - DO NOT CITE OR QUOTE
-------
model predictions, the differences in the dose-to-target-concentration ratios are not significantly
dissimilar from the peak 5-year average scenario, differing by less than 5%. A table for
converting less-than-lifetime average blood concentrations and other internal dose metrics to
human intake is presented in Appendix C.4.
The net effect of using three different scenarios for estimating the HED from rodent
exposures is that, for the same target concentration, the ratio of administered dose (to the rodent)
to HED will be larger for short-term exposures than for chronic exposures. Figure 3-34 is
similar to Figure 3-33, except that it shows the relationship of daily intake to a fixed target
TCDD blood concentration level. Figure 3-34 shows that, for human intakes of approximately
0.01 ng/kg-day, the difference in the defined scenarios is 40% or less, with a lifetime-scenario
daily intake of 0.014 ng/kg-day required to reach the same target concentration for a shorter-term
exposure of 0.01 ng/kg-day. The corresponding daily intake for the gestational scenario is
0.011 ng/kg-day. Because of the nonlinearities in the Emond human PBPK model, the
magnitude of the difference between the lifetime and less-than-lifetime exposure scenarios
increases at lower intake levels, but not to a substantial degree.
The differential effect of short- and long-term exposures is much more accentuated at the
rodent end of the exposure kinetic modeling. Analogous to the processes described in the
previous section for first-order body burden (see Section 3.4.2.2), the TCDD blood concentration
for single exposures is essentially the immediate absorbed fraction of the administered dose,
which will be somewhat lower than the administered dose, while for chronic exposure, the
TCDD blood concentration will reflect the long-term accumulation from daily exposure, which
will be very much larger than the administered dose (expressed as a daily intake). Table 3-25
shows the overall impact of TK modeling on the extrapolation of administered dose to HED,
comparing the Emond PBPK and first-order body burden models. For comparison purposes, the
administered dose is fixed at 1 ng/kg-day for all model runs. Large animal-to-human TK
extrapolation factors (TKEF) are evident for short-term mouse studies, decreasing in magnitude
with increasing exposure duration. The only exception is the slightly lower extrapolation factor
for the mouse 1-day exposure, which is the result of the relatively short TCDD half-life (10 days)
in mice and the use of the peak TCDD blood concentration as representative of single exposures,
compared to the average TCDD blood concentration over the exposure period used for multiple
exposures. The TKefs are lower for rats because of the slower elimination of TCDD in rats
3-59
DRAFT - DO NOT CITE OR QUOTE
-------
compared to mice. Also, because of the nonlinear kinetics inherent in the Emond PBPK model,
the span of the HED (13-fold for mice) across these exposure durations is greater than the span
of the LASC (fourfold for mice). Because of the dose-dependence of TCDD elimination in the
Emond model, the TKef becomes smaller with decreasing intake. The result of this nonlinearity
is that, although Table 3-25 shows much lower TKEFs for the Emond PBPK model than for the
first-order body burden metric, at much lower HED levels, the predictions of the two models are
much closer.
3-60
DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-1. Partition coefficients, tissue volumes, and volume of distribution
for TCDD in humans
Tissue
Tissuerblood
partition
coefficient
Tissue volume
(liters, for a
60-kg person)
Effective volume of
distribution (Vd—liters of
blood equivalent)
Percent
total Vd
Blood
1
3
3
0.25
Fat
100
11.4
1.140
94.19
Liver
6
1.56
9
0.77
Rest of the body
1.5
38.64
58
4.79
Total
54.6*
1.210
100.00
*The total tissue volume presented here represents only 91% of body weight because some of the weight and
volume of the body is occupied by bone and other structures where TCDD uptake and accumulation do not occur to
a significant extent.
Source: Wang et al. (1997). Emond et al. (2006: 20051.
Table 3-2. Blood flows, permeability factors, and resulting half lives (t1^) for
perfusion losses for humans as represented by the TCDD PBPK model of
Emond et al. (2006: 2005)
Tissue
Permeability (fraction of
compartment blood flow)
Rate constant for
compartmental
elimination (hour-1)
tVi (hrs)
Fat
0.12
0.0049
143
Liver
0.03
0.77
0.90
Rest of the body
0.35
3.84
0.18
3-61
DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-3. Toxicokinetic conversion factors for calculating human equivalent
doses from rodent bioassays based on first-order kinetics
Half-life (days)"
Mouse
Rat (Wistar)
Rat (other)
Guinea pig
10
20
25
40
Exposure
duration (days)
Conversion factor (CF)b55^(ti):^ given in parentheses
1
3,882 (0.77)
3,815 (0.79)
3,802 (0.79)
3,783 (0.79)
7
1,107 (2.71)
1,020 (2.94)
1,004 (2.99)
979 (3.07)
14
681 (4.41)
587 (5.11)
569 (5.27)
543 (5.53)
28
453 (6.62)
350 (8.56)
331 (9.06)
303 (9.90)
90
307 (9.76)
186 (16.1)
163 (18.4)
130 (23.0)
180
282(10.6)
154 (19.5)
129 (23.2)
93 (32.1)
365
270(11.1)
141 (21.3)
115 (26.0)
77 (38.9)
730
226(11.3)
115 (22.2)
93 (27.4)
60 (42.5)
aHalf-life for humans = 2,593 days (7.1 years).
hdH = dJCF- BBH(tH):dH = 2,185 (1-180 days), 2,202 (365 days), 2,555 (730 days).
3-62 DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-4. Equations used in the concentration and age-dependent model
(CADMs Avlward et a!.. 2005bV'
Parameter
Equation
Hepatic
Concentration
O (f f )*C
.r-f *^body if. / f w max J nan.' 'body
mill
W\
(ng/kg)
Fat
Concentration
(ng/kg)
Qbody s| j Cj^rnax ^ ^'body ¦>.
adipose V VJ mm ^ ^ /)
Wa ^ body
Hepatic
Elimination
Exr hepatic = ke*Qbody*(1 (fmm + (/-^/-)*C^))
& +^body
Excretion via
gut of
Unchanged
TCDD
Exr gut = ka*Qa
(Exsorption)
Change of
TCDD due to
bodyweight
change
Change TCDD _BW =
Amount in
body as a
function of
Qbody 0 1 <&) Qtody (0 lixf hepatic + Exr gut + Change TCDD BW
time
Adipose tissue
growth
„r 1.2* BMI + (0.23* Age) 10.8* sex
rr =
100
Change of
hepatic
elimination
k,, — k„Q ~~ ke.slnpv * Age
constant with
age
aFor abbreviations and parameter descriptions, see Table 3-5.
3-63 DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-5. Parameters of the concentration and age-dependent model
CCADM; Aylward et al.. 2005b)
Parameter
Value
Units
Comments/sources
f a
^-hrnin
0.01
unitless
Minimum body burden fraction in liver
f a
J-hmax
0.7
unitless
Maximum body burden fraction in liver
Ka
100
ng/kg
Body burden at half-maximum of fraction
liver
ke
Calculated
per year
ke = ke0 - ke siope*(age) with enforced
minimum of ke mm
keO
0.85
per year
CADM-mean hepatic elimination base rate at
age 0
ke slope
0.011
per year
Change in ke per year of age
lxc mm
0.2
per year
Minimum hepatic elimination rate
wa (adipose weight fraction)
Calculated
unitless
wa = [(1.2*BMI)+0.23*Age-10.8*sex]/100
wh (liver body weight fraction)
0.03
unitless
Assumed constant
ka (adipose clearance factor)
0.0025
per month
Passive elimination rate from intestinal tract
Monthly dose
0.15507069
ng
per month
Estimated absorption fraction
0.97
unitless
From Moser and McLaehlan (2001)
Body weight
70
kg
Standard male weight
Sex
1
unitless
1 = male; 0 = female
Time of administration
840
months
Initial Cbody
0.2
ng/kg
Estimated background young adults UMDES
sampling
Absorbed monthly dose 1
0.150418569
ng
per month
"The values of fhmm, fhmax, and K were obtained by best fit of the model simulations to the experimental data with the
method of least squares (Aylward et al.. 2005a: Carrier et al.. 1995a).
3-64 DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-6. Confidence in the CADMa model simulations of TCDD dose
metricsb
Dose metric
Level of confidence
Administered dose
NA
Absorbed dose
H
Body burden
H
Serum lipid concentration
M
Total tissue (liver) concentration
L
Receptor occupancy (bound concentration)
NA
H = high, M = medium, L = low, NA = not applicable.
'Concentration and age-dependent model (Avlward et al.. 2005b').
bUsing professional judgment, EPA ranked its confidence in the CADM model as low, medium,
or high (or not applicable) based on model simulations of administered dose, absorbed dose,
body burden, serum lipid concentration, total tissue (liver) concentration, and receptor
occupancy.
3-65
DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-7. Equations used in the TCDD PBPK model of Emond et al. (2006)
Aspect
Equation
Body-weight
growth with age
BWtime(g) = BW roxf 0Alxtime )
{1402.5 + time J
Cardiac output
( bw Y75
Qc(mL/hr) = QCCARx60\
A factor of 60 corresponds to the conversion of minutes to hours, and 1,000 is used for the
conversion of BW from grams to kilograms.
Blood
compartment
t^ \_{Qf x CJb) + [Qre x Creb) + [Qli x Clib) + lymph^ (CbxCLURI)
Tissue compartment (fat, rest of the body)
Tissue blood
subcompartment
dAth (nmollmL) = Qt(Ca Ctb) PAt^Ctb
Ath
Ctbinmol / mL) =
Wtb
Tissue cellular
matrices
dAt
{nmol / mL) = PAt
At
Ctinmol / mL) = —
Wt
QA-^1
Pi)
Liver tissue compartment
Tissue blood
subcompartment
dAlib ^nm()j J mj^ _ Qij((
-------
Table 3-7. Equations used in the TCDD PBPK model of Emond et al. (2006)
(continued)
Aspect
Equation
Gastrointestinal absorption and distribution of TCDD to the portal lymphatic circulation
Amount of
TCDD
remaining in
lumen cavity
^^^-(nmol / hr) = [( KS'I' + KAIiS ) x lumen J + intake
Lumen in the amount of TCDD remaining in the GI tract (nmol); intake is the rate of intake of
TCDD during a subchronic exposure (nmol/hr).
Amount of
TCDD
eliminated in the
feces
dFeces ^nmQ^ j ^r^ x iumen
dt
Absorption rate
of TCDD to the
blood via the
lymphatic
circulation
dLymph ^nmQ^ j x iumen x 0.7
dt
Absorption rate
of TCDD by the
liver via portal
circulation
dPortal , , ,, ^ „
(nmol / hr) = KABS x lumen x 0.3
dt
Note: Key parameters and abbreviations are defined in Table 3-8.
3-67 DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-8. Parameters of the PBPK model for TCDD
On
00
o
o
o
2
o
H
o
HH
H
W
O
o
c
o
H
W
Parameter
description
Symbol
Parameter values
Human
nongestational"
Human
gestational"
Mouse
nongestational
Mouse
gestational
Rat
nongestational
Rat
gestational
Body weight (g)
BW
Calculated
Calculated
23-28b
23-28
125-250b
85-190b
Cardiac output (mL/hour/kg)
QCCAR
15.36°'d
Calculated
275°
275°
311.4e
311,4e
Tissue (intracellular) volumes (fraction of BW)
Liver
WLIO
Calculated
Calculated
0.0549f
0.0549f
0.036e
0.036e
Fat
WFO
Calculated
Calculated
0.069e
Calculated
0.069e
Calculated
Tissue blood volumes
Liver (fraction of WLIO)
WLIBO
0.266e
0.266e
0.266e
0.266e
0.266e
0.266e
Fat (fraction of WFO)
WFBO
0.05e
0.05e
0.05e
0.05e
0.05e
0.05e
Rest of body (fraction of WREO)
WREBO
0.03e
0.03e
0.03e
0.03 e
0.03e
0.03e
Placenta tissue fraction of tissue blood
weight (unitless)
WPLABO
N/A
0.5s
N/A
0.5e
N/A
0.5e
Tissue blood flow (fraction of cardiac output)
Liver
QLIF
0.26°
0.26°
0.16 lf
0.161f
0.183e
0.183e
Fat
QFF
0.05°
0.05°
0.07h
0.07h
0.069e
0.069e
Placenta
QPLAF
N/A
Calculated
N/A
Calculated
N/A
Calculated
Tissue permeability (fraction of tissue blood flow)
Liver
PALIF
0.35e
0.35e
0.35e
0.35e
0.35e
0.35e
Fat
PAFF
0.121
0.121
0.121
0.121
0.091e
0.091e
Placenta diffusional permeability fraction
(unitless)
PAPLAF
N/A
0.3s
N/A
0.03s
N/A
0.3s
Rest of body
PAREF
0.03e
0.03e
0.03e
0.03e
0.0298e
0.0298e
-------
Table 3-8. Parameters of the PBPK model for TCDD (continued)
On
VO
O
o
o
2
o
H
o
HH
H
W
O
o
c
o
H
W
Parameter
description
Symbol
Parameter values
Human
nongestational"
Human
gestational"
Mouse
nongestational
Mouse
gestational
Rat
nongestational
Rat
gestational
Partition coefficient
Liver
PLI
6e
6e
6e
6e
6e
6e
Fetus/blood partition coefficient (unitless)
PFETUS
N/A
4J
N/A
4J
N/A
4J
Placenta/blood partition coefficient (unitless)
PPLA
N/A
1.5J
N/A
3s
N/A
1.5J
Fat
PF
o
o
o
o
4001
4001
o
o
o
o
Rest of body
PRE
1.5e
1.5e
3k
3k
1.5e
1.5e
Metabolism constants
Urinary clearance elimination (mL/hour)
CLURI
4.17E-081
4.17E-081
0.091
0.091
0.01J
0.01J
Clearance—transfer from mother to fetus
(mL/hour)
CLPLAFET
N/A
16e
N/A
0.171
N/A
0.171
Liver (biliary elimination and metabolism;
hour"1)
KBILELI
Inducible
Inducible
Inducible
Inducible
Inducible
Inducible
Interspecies constant (hour1)
KELV
0.00111
0.00111
0.41
0.41
0.15e
0.15e
AhR
Affinity constant in liver (nmol/mL)
KDLI
0.1e
0.1e
o.ooor
o.ooor
o.ooor
o.ooor
Binding capacity in liver (nmol/mL)
LIBMAX
0.35e
0.35e
0.00035e
0.00035e
0.00035e
0.00035e
Placenta binding capacity (nmol/mL)
PLABMAX
N/A
0.2J
N/A
0.0002J
N/A
0.0002J
Affinity constant protein (AhR) in placenta
(nmol/mL)
KDPLA
N/A
0.1J
N/A
0.0001J
N/A
0.0001J
-------
Table 3-8. Parameters of the PBPK model for TCDD (continued)
Parameter
description
Symbol
Parameter values
Human
nongestational"
Human
gestational"
Mouse
nongestational
Mouse
gestational
Rat
nongestational
Rat
gestational
CYP1A2 induction parameters
Dissociation constant CYP1A2 (nmol/mL)
KDLI2
401
401
0.02'
0.02'
0.04J
0.04J
Degradation process CYP1A2 (nmol/mL)
CYP1A210UTZ
l,600e
l,600e
1.6e
1.6e
1.6e
1.6e
Dissociation constant during induction
(nmol/mL)
CYP1A21EC50
130e
130e
0.13s
0.13e
0.13e
0.13e
Basal concentration of CYP1A2 (nmol/mL)
CYP1A21A2
l,600e
l,600e
1.5k
1.5k
1.6e
1.6e
First-order rate of degradation (hour1)
CYP1A21KOUT
o.r
o.r
o.r
o.r
o.r
o.r
Time delay before induction process (hour)
CYP1A21TAU
0.25e
0.25e
1.5k
1.5k
0.25e
0.25e
Maximal induction of CYP1A2 (unitless)
CYP1A21EMAX
9,300'
9,300'
600e
600e
600e
600e
Other constants
Oral absorption constant (hour1)
KABS
0.061
0.06'
0.48'
0.48'
0.48e
0.48e
Gastric nonabsorption constant (hour1)
KST
o.or
o.or
0.30'
0.30'
0.36e
0.36e
^ aUnits for human nongestational parameters are L rather than mL and kg rather than g where applicable.
H bBody weight varies by study (Emond et al.. 20041.
' °Krishnan and Andersen (19911.
O dUnits are L/kg/hr.
^ eWang et al. (1997).
O fILSI (19941.
"j 8Fixed.
hh hLeung et al. (1990).
trt 'Optimized.
O JEmond et al. (2004).
F kWang et al. (20001.
O 'Lawrence and Gobas (1997).
g ""Calculated to estimate 87% bioavailability of TCDD in humans (Poiger and Schlatter. 1986).
H
ffl
-------
1 Table 3-9. Regression analysis results for the relationship between logio
2 serum TCDD at the midpoint of observations and the logio of the rate
3 constant for decline of TCDD levels using Ranch Hand data
4
Item
Aspect
Value
Summary of fit
RSquare
0.894
RsquareAdj
0.871
Root mean square error
0.044
Mean responses
0.130
Observations (or sum weights)
11
Parameter estimates
Intercept
Estimate
-0.054
Standard deviation
0.026
t ratio
-2.07
Prob>t
0.0679
Log (TCDDpg/g)
Estimate
0.092
Standard error
0.011
t ratio
8.28
Prob>t
<0.0001
5
6
3-71
DRAFT - DO NOT CITE OR QUOTE
-------
1 Table 3-10. Dosing protocols for human and animal models
Model
Study
Low dose
High dose
Averaging
period
Rat
NTP (2006b); 105 weeks
3 ng/kg 5 days per week
(2.14 ng/kg-day adjusted
dose)
100 ng/kg 5 days per
week (71.4 ng/kg-day
adjusted dose)
105 weeks
Mouse
NTP (1982a); male
mouse, 2-year duration
5 ng/kg biweekly (1.4
ng/kg-day adjusted dose)
200 ng/kg biweekly
(71 ng/kg-day
adjusted dose)
2 years
Rat
gestational
Markowski et al. (2001)
20 ng/kg, single dose
180 ng/kg, single
dose
Single day
Mouse
gestational
Li et al. (2006)
2 ng/kg-day for GDs 1-3
100 ng/kg-day for
GDs 1-3
3 days
Human
Standard lifetime scenario
(daily intake for 70 years)
7 x 10"4 ng/kg-day
0.02 ng/kg-day
70 years
Human
gestational
Standard gestational
scenario (daily intake,
pregnancy at age 45)
7 x 10"4 ng/kg-day
0.02 ng/kg-day
9 months of
pregnancy
3
4
3-72
DRAFT - DO NOT CITE OR QUOTE
-------
1 Table 3-11. Most sensitive variables for the rat and mouse nongestational and
2 gestational models
3
Variable
Variable description
Rat, low
dose, +5%
elasticity
Rat, high
dose, +5%
elasticity
Mouse, low
dose, +5%
elasticity
Mouse, high
dose, +5%
elasticity
Nongestational
HILL
Hill coefficient
3.3
3.0
3.4
2.8
CYP1A2 10UTZ
Induction concentration
in degradation process
(nmol/L)
-0.8
-0.8
-0.8
-0.7
CYP1A21A2
Induction basal
concentration of 1A2
(nmol/L)
0.8
0.8
0.9
0.7
WLIO
Fractional liver weight
(unitless)
-0.6
-0.7
-0.6
-0.6
CYP1A21EMAX
Maximum induction over
basal effect (unitless)
-0.5
-0.7
-0.5
-0.6
KELV
Interspecies constant
(hrA-l)
-0.3
-0.7
-0.5
-0.6
LIBMAX
Liver binding capacity
(nmol/1)
-0.4
-0.4
-0.3
-0.3
CYP1A21EC50
Induction disassociation
constant for 1A2
(nmol/L)
0.4
0.4
0.3
0.4
KDLI
Liver affinity proteins
AhR (nmol/L)
0.3
0.2
0.3
0.3
KABS
Intestinal excretion and
absorption constant
(hrA-l)
0.3
0.3
0.3
0.3
KST
Gastric excretion and
absorption constant
(hrA-l)
-0.3
-0.3
-0.3
-0.3
Gestational
HILL
Hill coefficient
1.2
1.4
0.6
1.4
WLIO
Fractional liver weight
(unitless)
-0.4
-0.4
-0.2
-0.4
KABS
Intestinal excretion and
absorption constant
(hrA-l)
0.4
0.4
0.4
0.3
4
3-73
DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-11. Most sensitive variables for the rat and mouse nongestational and
gestational models (continued)
Variable
Variable description
Rat, low
dose, +5%
elasticity
Rat, high
dose, +5%
elasticity
Mouse, low
dose, +5%
elasticity
Mouse, high
dose, +5%
Elasticity
CYP1A2 10UTZ
Induction concentration
in degradation process
(nmol/L)
-0.4
-0.4
-0.3
-0.4
KDLI2
Liver affinity proteins
1A2 (nmol/L)
0.4
0.4
0.2
0.3
KST
Gastric excretion and
absorption constant
(hrA-l)
-0.4
-0.3
-0.3
-0.3
QCCAR
Cardiac output (1/kg-hr)
-0.3
-0.3
-0.4
-0.3
QFF
Adipose tissue blood
flow fraction of cardiac
output (unitless)
-0.2
-0.2
-0.4
-0.2
CYP1A21EMAX
Maximum induction over
basal effect (unitless)
-0.2
-0.3
-0.1
-0.3
PAFF
Adipose diffusional
permeability fraction
(unitless)
-0.2
-0.2
-0.4
-0.2
LIBMAX
Liver binding capacity
(nmol/L)
-0.1
-0.2
-0.1
-0.2
KDLI
Liver affinity proteins
AhR (nmol/L)
0.1
0.1
0.1
0.2
CYP1A21EC50
Induction disassociation
constant for 1A2
(nmol/L)
0.1
0.2
0.1
0.2
CYP1A21KOUT
Induction first-order rate
of degradation (hrA-l)
-0.1
-0.2
0.0
0.0
3-74 DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-12. Most sensitive variables for the human nongestational and gestational models
-j
O
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Variable
Variable description
Human
nongestational, RfD
dose +5% elasticity
Human
nongestational,
POD dose +5%
elasticity
Human
gestational, RfD
dose +5%
elasticity
Human
gestational,
POD dose +5%
elasticity
HILL
Hill coefficient
3.4
3.6
3.5
3.7
CYP1 A2_ 1OUTZ
Induction concentration in degradation process
(nmol/L)
-0.5
-0.6
-0.5
-0.6
CYP1A21A2
Induction basal concentration of 1A2 (nmol/L)
0.4
0.5
0.5
0.6
CYP1A21EMAX
Maximum induction over basal effect (unitless)
-0.4
-0.6
-0.5
-0.6
SACHNGELI
Fraction liver-weight multiplier for sensitivity
analysis (unitless)
-0.5
-0.6
-0.4
-0.6
KF.I.V
Interspecies constant (hrA-l)
-0.4
-0.5
-0.4
-0.6
CYP1A21EC50
Induction disassociation constant for 1A2 (nmol/L)
0.2
0.3
0.3
0.4
KDLI
Liver affinity proteins AhR (nmol/L)
0.2
0.3
0.3
0.4
LIBMAX
Liver binding capacity (nmol/L)
-0.3
-0.3
-0.3
-0.3
SACHNGEBW
Body-weight multiplier for sensitivity analysis
(unitless)
-0.3
0.0
-0.3
0.1
PF
Adipose tissue :blood partition coefficient (unitless)
-0.2
-0.1
-0.2
0.0
SACHNGEF
Fraction adipose-weight multiplier for sensitivity
analysis (unitless)
-0.2
-0.1
-0.2
0.0
KABS
Intestinal excretion and absorption constant (hrA-l)
-0.1
0.1
0.1
0.1
KST
Gastric excretion and absorption constant (hrA-l)
-0.1
-0.1
-0.1
-0.1
KDLI2
Liver affinity proteins 1A2 (nmol/L)
-0.1
0.1
0.1
0.0
CYP1A21K0UT
Induction first-order rate of degradation (hrA-l)
-0.2
0.0
0.0
0.0
-------
1 Table 3-13. TCDD serum measurements over time for two Austrian women
2 exposed to TCDD in 1997a
3
Austrian Woman 1
Austrian Woman 2
Day
TCDD LASC (ppt)
Day
TCDD LASC (ppt)
0
144,000
0
26,000
63
111,000
53
20,500
116
85,600
63
16,100
126
80,900
77
15,900
135
72,200
84
14,300
147
70,200
98
13,200
161
87,700
105
18,500
168
89,900
140
13,300
203
62,100
177
13,700
240
65,100
207
19,300
270
68,300
238
15,700
295
64,900
267
15,200
309
68,100
326
15,700
316
72,600
437
17,700
323
73,700
533
14,100
330
72,500
637
10,500
366
60,300
718
11,000
389
73,900
841
10,100
466
85,600
998
9,500
500
68,100
596
47,100
700
39,300
781
27,400
904
30,300
1,054
35,900
'Source of data: (Geiisau et al.. 2001).
4
5
3-76
DRAFT - DO NOT CITE OR QUOTE
-------
1 Table 3-14. TCDD serum measurements over time for two Seveso males
2 exposed to TCDD in 1976a
3
Seveso Male (6 years old)
Seveso Male (50 years old)
Day
TCDD LASC (ppt)
Day
TCDD LASC (ppt)
0
15,900
0
1,770
826
4,350
92
807
1,522
2,269
981
1,069
2,193
580
1,218
809
5,867
324
1,921
680
6,011
807
"Source of data: Needham et al. (19981
4
5
6 Table 3-15. Results of Hill coefficient sensitivity analysis simulations with
7 Emond human PBPK model
8
Hill = 0.6
kelv = default
doseiv optimized
Hill = 1
kelv = default
doseiv optimized
Hill = 0.6
kelv and doseiv
optimized
Hill = 1
kelv and doseiv
optimized
Hill
0.6
1.0
0.6
1.0
kelv
Austrian 1
0.0011
0.0011
1.73E-03
5.74E-03
Austrian 2
1.79E-03
4.89E-03
Seveso 6
0.00300
0.00490
Seveso 50
2.94E-04
4.79E-03
doseiv
Austrian 1
7.00E+04
1.20E+04
8.00E+04
1.98E+04
Austrian 2
1.30E+04
2.40E+03
1.80E+04
3.40E+03
Seveso 6
1.10E+04
3.48E+02
1.10E+04
9.98E+02
Seveso 50
4.98E+02
9.76E+01
2.98E+02
1.37E+02
9
10
3-77
DRAFT - DO NOT CITE OR QUOTE
-------
1 Table 3-16. Alternative CYP1A2 parameter estimates for sensitivity analysis
2 of Emond human PBPK model
3
Budinskv et al. (2010)
values
Emond model value
Alternative
scaled value3
Human
Rat
Human
Rat
Human
CYP1A2 Basal
11.6
22.4
1,600
1.6
829
CYP1A2 Max
12,900
322
9,300
600
24,037
EC50/KDLI
0.329
0.0628
130
0.04
209
aEmond model rat value multiplied by the ratio of the corresponding human:rat parameter values
from Budinsky et al. (20101.
4
5
6 Table 3-17. Results of CYP1A2 parameter sensitivity analysis simulations
7 with Emond human PBPK model
8
Hill = 0.6
kelv = default
doseiv optimized
Hill = 0.6
kelv and doseiv
optimized
Hill = 0.6,
Alternative
parameters,3
kelv and doseiv
optimized
kelv
Austrian 1
0.0011
1.73E-03
4.36E-04
Austrian 2
1.79E-03
1.67E-04
Seveso 6
0.00300
0.00030
Seveso 50
2.94E-04
9.68E-06
doseiv
Austrian 1
7.00E+04
8.00E+04
6.98E+04
Austrian 2
1.30E+04
1.80E+04
8.00E+03
Seveso 6
1.10E+04
1.10E+04
5.98E+03
Seveso 50
4.98E+02
2.98E+02
1.97E+02
aAlternative scaled values from Table 3-16.
9
3-78
DRAFT - DO NOT CITE OR QUOTE
-------
Table 3-18. Results of Emond human PBPK model parameter sensitivity analysis simulations. Comparison of
modeled human oral intakes for a range of lifetime average TCDD serum concentrations for alternative parameter
values.
Lifetime
average
TCDD
LASCa
(PPt)
Standard model
configuration
Alternative Hill
Standard Hill, optimized
elimination
Alternative Hill,
optimized elimination
Alternative induction
parametersb
optimized elimination
Hill = 0.6
kelv = 0.0011
CYP1A2 1A1 = 1,600
CYP1A2 1EMAX = 9,300
CYP1A2 1EC50 = 130
PF = 100
Hill = 1
kelv = 0.0011
CYP1A2 1A1 = 1,600
CYP1A2 1EMAX = 9,300
CYP1A2 1EC50 = 130
PF = 100
Hill = 0.6
kelv = 0.0017
Hill = 1
kelv = 0.0050
Hill = 0.6
kelv = 0.0002
CYP1A2 1A1 = 1,600
CYP1A2 1EMAX = 9,300
CYP1A2 1EC50 = 130
PF = 100
CYP1A2 1A1 = 1,600
CYP1A2 1EMAX = 9,300
CYP1A2 1EC50 = 130
PF = 100
CYP1A2 1A1 = 829
CYP1A2 1EMAX = 24.037
CYP1A2 1EC50 = 209
PF = 100
30
1.0E-03
3.8E-04
1.3E-03
3.9E-04
7.7E-04
100
5.7E-03
1.3E-03
8.0E-03
1.5E-03
4.1E-03
300
3.0E-02
4.2E-03
4.3E-02
5.9E-03
1.9E-02
1,000
1.9E-01
1.8E-02
2.8E-01
3.7E-02
1.2E-01
3,000
9.6E-01
8.1E-02
1.4E+00
2.3E-01
5.8E-01
aFrom lifetime female model.
'Estimated from Budinksy et al. (2010).
-------
1 Table 3-19. Confidence in the PBPK model simulations of TCDD dose
2 metrics
3
Dose metric
Human model
Rat model
Mouse model
Administered dose
N/A
N/A
N/A
Absorbed dose
H
H
M
Body burden
H
H
M
Serum (blood) concentration
H
H
M
Total liver concentration
M/L
H
M
Receptor occupancy (bound concentration)
L
L
L
4
5 H = high, M = medium, L = low, N/A = not applicable.
6
7
8 Table 3-20. Overall confidence associated with alternative dose metrics for
9 noncancer dose-response modeling for TCDD using rat PBPK model
10
End point
Body
burden
Blood or serum
concentration
Liver
concentration
Bound
concentration in
liver
Liver effects
M
H
M/L
Nonhepatic effects
M
H
M/L
11
12 H = high, M = medium, L = low.
13
14
15 Table 3-21. Overall confidence associated with alternative dose metrics for
16 noncancer dose-response modeling for TCDD using mouse PBPK model
17
End point
Body
burden
Blood or serum
concentration
Liver
concentration
Bound
concentration in
liver
Liver effects
M
M
L
Nonhepatic effects
M
M
L
18
19 H = high, M = medium, L = low.
20
3-80
DRAFT - DO NOT CITE OR QUOTE
-------
1 Table 3-22. Contributors to the overall confidence in the selection and use of
2 dose metrics in the dose-response modeling of TCDD based on rat and
3 human PBPK models"
4
Dose metric
Conceptual relevance
Prediction uncertainty
Overall confidence
Administered dose
L
NA
L
Body burden
M
M
M-L
Blood concentration
M
L
M
Liver concentration
L
M
L
Receptor (AhR)
occupancy
H
H
L
5
6 aUsing professional judgment, EPA ranked its confidence in the CADM model as low, medium, or high (or not
7 applicable) based on model simulations of administered dose, absorbed dose, body burden, serum lipid
8 concentration, total tissue (liver) concentration, and receptor occupancy.
9 H = high, M = medium, L = low, NA = not applicable.
10
11 Table 3-23. Contributors to the overall uncertainty in the selection and use
12 of dose metrics in the dose-response modeling of TCDD based on mouse and
13 human PBPK models
14
Dose metric
Conceptual uncertainty
Prediction uncertainty
Administered dose
H
NA
Absorbed dose
H
L
Body burden
M
M
Blood or serum concentration
M
M
Tissue concentration
L
MM
Receptor occupancy
L
H
15
16 H = high, M = medium, L = low, NA = not applicable
3-81
DRAFT - DO NOT CITE OR QUOTE
-------
1 Table 3-24. Comparison of human equivalent doses from the Emond human
2 PBPK model for the 45-year-old and 25-year-old gestational exposure
3 scenarios
4
Animal
bioassay POD
(ng/kg-day)
Species
TCDD
blood
concentration3
HED
45 year-old
HED
25 year-old
25-yr:45-yr
ratio
3
Mouse
8.800E-02
6.79E-04
1.03E-03
1.5
Rat
1.815E-01
1.87E-03
2.98E-03
1.6
30
Mouse
7.115E-01
1.51E-02
2.07E-02
1.4
Rat
1.367E+00
4.22E-02
5.41E-02
1.3
5
6 'Determined from the Emond rodent PBPK models assuming a single exposure on GD 13.
7
8
9 Table 3-25. Impact of toxicokinetic modeling on the extrapolation of
10 administered dose to HED, comparing the Emond PBPK and first-order
11 body burden models (administered dose = 1 ng/kg-day)
12
Exposure
duration (days)
lst-order BB
Emond PBPK
HEDa
(ng/kg-day)
TKEFb
LASCc
(ng/kg)
HED
(ng/kg-day)
TKef
Mouse
1
2.57E-4
3,882
75.5
9.49E-4
1,054
14
1.47E-3
681
64.4
8.17E-4
1,224
90
3.25E-3
307
173
3.83E-3
261
365
3.70E-3
270
248
6.66E-3
150
730
4.43E-3
226
263
1.08E-2
93
Rat
1
2.63E-4
3,802
110
1.87E-3
535
14
1.76E-3
569
208
5.22E-3
192
90
6.13E-3
163
599
2.81E-2
36
365
8.68E-3
115
811
4.52E-2
22
730
1.07E-2
93
853
6.47E-2
15
13
14 "Human-equivalent doses.
15 bRodent-to-human toxicokinetic extrapolation factor.
16 °Lipid-adjusted serum concentration.
3-82
DRAFT - DO NOT CITE OR QUOTE
-------
4
o
D 7 Day Livei/Fat
9 14 Day Liver/Fat
A 21 Day Liver/Fat
I 35 Day Liver/Fat
0)
J
•J-l
s
.o
K
•-
=
0)
u
c
o
u
o
K
Ps
Dose fig/Kg
1
2 Figure 3-1. Liver/fat concentration ratios in relation to TCDD dose at
3 various times after oral administration of TCDD to mice.
4
5 Source: Dilberto et al. (1995).
3-83
DRAFT - DO NOT CITE OR QUOTE
-------
0.2"
LU
0
INITIAL SERUM LIPID TCDD LEVEL
-------
0.05
y = 0.09735 - 0.00282x RA2 = 0.752
% Body Fat
Figure 3-3. Observed relationship of fecal 2,3,7,8-TCDD clearance and
estimated percent body fat.
Source: Rohde et al. (1999).
3-85
DRAFT - DO NOT CITE OR QUOTE
-------
5C
u
C5
4>
b
sa
-
C5
a
S3
C5
S3
W
—
—
u
H
11.0
10.0
y = - 1.89 + 0.314x RA2 = 0.998
Seveso Females
Ranch H and M ales
Seveso M ales
Error bars are ± 1 standard error
1
2
3
4
5
% Body Fat
Figure 3-4. Unweighted empirical relationship between percent body fat
estimated from body mass index and TCDD elimination half-life—combined
Ranch Hand and Seveso observations.
3-86
DRAFT - DO NOT CITE OR QUOTE
-------
u
c
>
m
m
Ihr
60
,E
CO
m
Imp
u
c
Functional biomarkers
Receptor occupancy
Total tissue concentration
Blood or serum concentration
Body burden
Absorbed dose
Intake
Figure 3-5. Relevance of candidate dose metrics for dose-response modeling,
based on mode of action and target organ toxicity of TCDD.
3-87
DRAFT - DO NOT CITE OR QUOTE
-------
Experimental Applied Dose
Body BurdenRat{t) = BB{0)e +
, d(l-e^)fa
h,2A (l-e~kjA)
Human
drj = d
H ~~
hfiH (1 - e~kHtH)
Estimated ^
Exposure
Body BurdenRat (/ )
A
V
Body BurdenHyman{t)
Figure 3-6. Process of estimating a human-equivalent TCDD lifetime average daily oral exposure {(In) from an
experimental animal average daily oral exposure (d,\) based on the body-burden dose metric.
The arrows represent mathematical conversions based on toxicokinetic modeling. BBA (TWA animal body burden) and BBH (TWA human
body burden) are assumed to be toxicokinetically equivalent. See text for further explanation.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
10000 15000
20000
25000
target body burden
chronic exposure (BB:d = 2555)
half-chronic exposure (BB:d = 2202)
shorter exposures (BB:d = 2185)
o
o
m
Exposure days
Figure 3-7. Human body burden time profiles for achieving a target body
burden for different exposure duration scenarios.
BB:d is BBH(tH):dH in Figure 3-6. The curve depicted using the solid line illustrates the increase in
the human body burden over time for a hypothetical human administered a daily TCDD dose
where the time-weighted average human body burden estimate over the lifetime is equal to the
target body burden attained in a rodent bioassay. When compared to shorter durations (dashed
lines), a higher average daily TCDD dose is required to yield a time-weighted average human
body burden over a lifetime that is equal to the target body burden attained in a rodent bioassay.
The half-chronic exposure scenario (depicted using a dashed line) is equivalent to a 1-year
exposure in rodents. When compared to a chronic BBH, a lower value of dH is needed to attain the
target body burden in a rodent bioassay when the time-weighted average is over the last 35 years
of life; the dose-to-plateau ratio is also smaller (i.e., dHjC < dH:SC to attain the target body burden in
a rodent bioassay). The shorter exposure scenario is equivalent to most other shorter rodent
exposure durations, from 1 day to subchronic, which are indistinguishable with respect to the
BB:d ratio (subchronic shown).
3-89
DRAFT - DO NOT CITE OR QUOTE
-------
DISTRIBUTION ELIMINATION
Tissue-Specific
Concentration-Dependant
Hepatic metabolism
with first-order rate
^ constant
ABSORPTION
Fecal excretion
with the first-order
^ rate constant ka^
TCDD
Blood
Liver Burden
a(o=a(o*/*(Q)
Adipose Burden
e.c)=ac)»[i-/i(c4|]
Figure 3-8. Schematic of the CADM structure.
Source: Aylward et al. (2005b').
3-90
DRAFT - DO NOT CITE OR QUOTE
-------
c
*
TJ
O
0 .5
- -0u - V - - <£_
4—
o
c
o
o
03
L_
ll
1 10 100 1000
Cb
1
2 Figure 3-9. Comparison of observed and simulated fractions of the body
3 burden contained in the liver and adipose tissues in rats.
4 fraction contained in liver (observation) (~):./,-,un- fraction contained in liver (simulation) (—);
5 /at, fraction contained in the adipose tissue (observation) CO); /,,-,,,,,,, fraction contained in the
6 adipose tissue (simulation) (—); and Cb body concentration in ng TCDD/kg body wt.
7
8 Source: Carrier et al. (1995a): data from Abraham et al. (1988) measured 7 days after dosing.
9
3-91
DRAFT - DO NOT CITE OR QUOTE
-------
Blood systemic circulation
Urinary
excretion
GI tract
elimination
Oral
absorption
Liver
Cellular matrices
Cellular matrices
Blood tissue
Fat
Cellular matrices
Cellular matrices
Blood tissue
Rest of body
Cellular matrices
Cellular matrices
Blood tissue
1
2 Figure 3-10. Conceptual representation of PBPK model for rat exposed to
3 TCDD.
4
5 Source: Emond et al. (2006).
6
3-92
DRAFT - DO NOT CITE OR QUOTE
-------
Urinary
excretion
1
2
3
4
5
6
Elimination Gl tract
Oral absorption
Portal vein
Liver (AhR and CYP1A2 induction)
Rest of body
Placenta (AhR)
Fetus
Figure 3-11. Conceptual representation of PBPK model for rat
developmental exposure to TCDD.
Source: Emond et al. (20041.
3-93
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
Tissue blood
-AB
Bf
TB
i V
PA
Water
fraction
Lipid fraction
Ks
Lip
Kn
TCDD-Ah
CYP1A2-TCDD
Tissue
Bb
Plasma
proteins
Tnb
Nonspecific
bound
Figure 3-12. TCDD distribution in the liver tissue.
Source: Wang et al. (19971.
3-94
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
a
» 2
10 15
Time (days*
20
10 15
Time ichiys)
aDO
700
_ 600
f
— 500
I
«=
1 400
©
I 300
u
_2
a 200
100
0
25
C 50
1, 40
!
130
£i
w
I 20
10
10 15
Time (days*
10 15
Time 4d
-------
0.0001
200
400
600
800
1,000
0.0001
200
400
600
800
1,000
0.0001
Time (hr)
Time (hr)
400 600
Time (hr)
1,000
Figure 3-14. Comparisons of model predictions to experimental data using a fixed elimination rate model with
hepatic sequestration (A) and an inducible elimination rate model with (B) and without (C) hepatic
sequestration.
EXBL, experimental blood levels. Model predictions were compared with the data of Santostefano et al. (1998). where female rats were
exposed to a single oral dose of 10 (ig of TCDD/kg B W. Error bars are ± SD.
Source: Edmond et al. (2006).
-------
100.00
1,750 ng TCDD/kg BW
500 ng TCDD/kg BW
150ng1XDD^
50 ng TCDD/kg BW
10.00
o
>
1.00
o
Q
O
I—
0.10
0.01
0
5
10
15
20
25
30
35
Time (week)
1
2 Figure 3-15. PBPK model simulation of hepatic TCDD concentration (ppb)
3 during chronic exposure to TCDD at 50,150, 500, or 1,750 ng TCDD/BW
4 using the inducible elimination rate model compared with the experimental
5 data measured at the end of exposure.
6
7 Source: Emond et al. (2006).
8
3-97
DRAFT - DO NOT CITE OR QUOTE
-------
.
CBFPTRH
V40922
1000 -
Time (year)
.
CBPPTRH
V33457
Time (Year)
1000000
100000
10000
1000
100
10
1
m
*
CBPPTRH
V30208
10 15 20 25 30
Time (year)
:
CBPPTRH
V49044
15 20 25
Time (year)
-CBPPTRH
V31731
20
Time (year)
40
10000
1000
q 100
o
o
0
CBPPTRH
¦ V46286
10 15 20
Time (year)
25
30
10000
1000
100
10
CBPPTRH
Time (year)
CbPHTRH
Time year)
100000
CBPPTRH
...
V30172
Time (year)
000000
CBPPTRH
00000
V30991
Time (year)
Figure 3-16. Model predictions of TCDD blood concentration in 10 veterans
(A-J) from Ranch Hand Cohort.
Source: Emond et al. (2005).
3-98
DRAFT - DO NOT CITE OR QUOTE
-------
1,000,000
10,000
1,000
200 300
I *
Blood model predictions (pt#l) @|
Patient 1 Vienna women
1,000
_l
1,100
1,200
1,000,000
100,000
1,000 -I—I—I—I—|—I—I—I—|—I—I—I—|—I—I—I—|—I—I—I—|—I—I—I—I—I—I—I—|—I—I—I—|—I—I—I—|—I—I—I—|—I—I—I—|—I—I—I—|—r
0 100 200 300 400 500 600 700 800 900 1,000 1,100 1,200
| [g Blood model predictions (pt #2) [g Patient 2 Vienna women |
2
3 Figure 3-17. Time course of TCDD in blood (pg/g lipid adjusted) for two
4 highly exposed Austrian women (patients 1 and 2).
5 Symbols represent measured concentrations, and lines represent model predictions. These data
6 were used as part of the model evaluation (Geusau et al.. 2002).
7
8 Source: Emond et al. (2005).
9
3-99
DRAFT - DO NOT CITE OR QUOTE
-------
y = 11.8 - 0.48x R"2 = 0.999 Dln(ptl Em Mod Sim pg/g TCDD)
y = 11.8 - 0.45xR"2 = 0.772 Bln(ptl Obs pg/g TCDD)
y = 10.5 - 0.43x RA2 = 1.000 dn(pt2 Em Mod Sim pg/g TCDD)
y = 10.0 - (1.2 KR 2 = 0.612 «n(pt2 Obs pg/g TCDD)
Ye are After Exposure
1
2 Figure 3-18. Observed vs. Emond et al. (2005) model simulated serum
3 TCDD concentrations (pg/g lipid) over time (In = natural log) in
4 two Austrian women.
5
6 Data from Geusau et al. (2002).
3-100
DRAFT - DO NOT CITE OR QUOTE
-------
n Emond Mod Sim Ln(Decline/Yr)
® Observed Ln(Decline/Yr)
0.101+ 0.123x R"2 = 0.995
0.054 + 0.092x R"2 = 0.884
Log(TCDD pg/g at Midpoint Obs)
1
2 Figure 3-19. Comparison of the dose dependency of TCDD elimination in the
3 Emond model vs. observations of nine Ranch Hand veterans and two highly
4 exposed Austrian patients.
5 Circles are observed data.
6
3-101
DRAFT - DO NOT CITE OR QUOTE
-------
o
to
o
O
O
o
H
O
HH
H
W
O
V
o
c
o
H
W
HILL
CYP1A2 10UTZ
SA CHNGELI
CYP1A2 1EMAX
KELV
CYP1A2 1A2
LIBMAX
CYP1A2 1EC50
KDLI
SA CHNGEBW
-6.0
-4.0
-2.0
0.0
Elasticity
~ Elasticity, POD Dose, +50%
~ Elasticity, POD Dose, +5%
~ Elasticity, POD Dose, -5%
~ Elasticity, POD Dose, -50%
2.0
4.0
6.0
Figure 3-20. Elasticities in the nongestational human model, POD dose.
-------
o
LtJ
O
o
o
2
o
H
O
HH
H
W
O
V
o
c
o
H
W
HILL
CYP1 A2_10UTZ
CYP1A2_1EMAX
CYP1 A2_1A2
SA_CHNGELI
KELV
LIBMAX
SA_CHNGEBW
CYP1A2_1EC50
KDLI
PF
SA CHNGEF
-4.0
-3.0
-2.0
E
E
~ Elasticity, RfD Dose, +50%
~ Elasticity, RfD Dose, +5%
~ Elasticity, RfD Dose, -5%
~ Elasticity, RfD Dose, -50%
-1.0 0.0 1.0
Elasticity
2.0
3.0
4.0
Figure 3-21. Elasticities in the nongestational human model, RfD dose.
-------
1
2
3
4
5
6
o
_
24,000
s. -
20,000
~S V
16,000
12,000
w
O.UUU
4.000
•Default parameters, no optimization ^
•Default parameters, kelv optimized
•Alternative parameters, kelv optimizeu
300
900
1,200
1,800,
1,500
1.200
900
600
300
0
-Default parameters, no optimization
•Default parameters, kelv optimized
•Alternative parameters, kelv optimized
v- - - -.!
0 1,000 2,000 3,000 4.000 5,(
Time (days)
Figure 3-22. Hill coefficient sensitivity analysis.
Calibration of Emond human PBPK model for 2 values of Hill for four human data sets:
(a) Austrian Woman 1, (b) Austrian Woman 2, (c) Seveso 6-year-old male, (d) Seveso 50-year-old
male; see text for source of data. Values for kelv other than the standard model value of 0.0011
are optimized.
3-104
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
o
to
<
J
Q
0
0
h
140,000
120,000
100,000
80.000
60,000
40,000
20,000
—Default parameters, no optimization
— -Default parameters, kelv optimized
\ — 'Alternative parameters, kelv optimized
300
600
900
1,200
-Default parameters, no optimization
•Default parameters, kelv optimized
•Alternative parameters, kelv optimized
u
10,000
<
j
Q
0
u
h
3.000 4.000 5,(
Time (days)
28.000
<
~ _
24.000
¦S -
20,000
16,000
12.000
P flflft
w
o.UUU
4.000
•Default parameters, no optimization
•Default parameters, kelv optimized
•Alternative parameters, kelv optimized
b
300
900
1,200
1,800
1,500
1,200
900
600
300
0
-Default parameters, no optimization
•Default parameters, kelv optimized
¦Alternative parameters, kelv optimized
0 1,000 2.000 3,000 4.000 5,<
Time (days)
Figure 3-23. CYP1A2 parameter sensitivity analysis.
Calibration of Emond human PBPK model for alternate values of CYP1A2 parameters other than
Hill for four human data sets: (a) Austrian Woman L (b) Austrian Woman 2, (c) Seveso 6-year-
old male, (d) Seveso 50-year-old male; see text for source of data. Alternate parameters were
estimated from data presented in Budirisky et al. (2010).
3-105
DRAFT - DO NOT CITE OR QUOTE
-------
1,000
100
10
1
0.1
0.01
0.001
|l.L iLWJu..jULiOI.lULpU.. lU.
LWOJLI
JUJJJJl
fcU ljiU. ,jiU
JUIJL5.
[7 —
CB (pg/g) I
Cb measured 1
10 20 30 40 50 60 70 80 90 100 110 120
10,000
1,000
100
10 -
1 -
0 Cli (pgg) Sim |
80 90 100 110 120
10,000
1,000
100
@ Cf (pg/g) Simulated I
@ Cf (pg/g) measured I
90 100 110 120
Figure 3-24. Experimental data (symbols) and model simulations (solid lines)
of (A) blood, (B) liver, and (C) adipose tissue concentrations of TCDD after
oral exposure to 150 ng/kg-day, 5 days/week, for 17 weeks in mice.
Y-axis represents concentration in pg/g, and X-axis represents time in days.
Source: Experimental data were obtained from Diliberto et al. (2001).
3-106
DRAFT - DO NOT CITE OR QUOTE
-------
Q.
Q.
£
O
c
0)
o
c
o
o
L_
0)
>
0000000
1000000
100000
10000
¦ Measured
~ Simulated
0.001 0.01 0.1 1 10
Dose (ug/kg)
100
300
1
2 Figure 3-25. Comparison of PBPK model simulations with experimental
3 data on liver concentrations in mice administered a single oral dose of
4 0.001-300 jig TCDD/kg.
5 The simulations and experimental data were obtained 24 hour post-exposure.
6
7 Source: Data obtained from Boverhoff et al. (2005).
3-107
DRAFT - DO NOT CITE OR QUOTE
-------
100000
10000
1000
100
10
1
cb sim
¦ cb measured
cli sim
¦ cli measured
0
cf sim
¦ cf measured
0
0.
10.0
Dose (ng/kg)
100.0
1000.0
1
2 Figure 3-26. Comparison of model simulations (solid lines) with
3 experimental data (symbols) on the effect of dose on blood (cb), liver (cli),
4 and fat (cf) concentrations following repetitive exposure to 0.1-450 ng
5 TCDD/kg, 5 days/week, for 13 weeks in mice.
6
7 Source: Data obtained from Diliberto et al. (2001).
3-108
DRAFT - DO NOT CITE OR QUOTE
-------
A
R —
CB (pg/g) |
Z ¦
Cb measured |
10 20 30 40 50 60 70 80 90 100 110 120
B
30 40
@
Cli (pgg) Sim 1
Cli (pgg) measured |
90 100 110 120
C
• Cf (pg/g) Simulated I
| Cf (pg/g) measured I
1
2
3
4
5
6
7
Figure 3-27. Comparison of experimental data (symbols) and model
predictions (solid lines) of (A) blood, (B) liver, and (C) adipose tissue
concentrations of TCDD after oral exposure to 1.5 ng/kg-day, 5 days/week,
for 17 weeks in mice.
Y-axis represents concentration in pg/g, and X-axis represents time in days.
Source: Experimental data were obtained from Diliberto et al. (2001).
3-109
DRAFT - DO NOT CITE OR QUOTE
-------
A
f — Cb (riiji'kg) I
> B mriHUff-il |
0 200 4QU 600
1,000 1,200 1,400 1,600 1,000 2,000 2,200
B
M
I V — CU (pfl/gj I
I
0 200 400
800 1,000 1,200 1,400 1,600 1,800 2,003 2,200
C
0 280 400 (>00
1,000 1,200 1,400 1,600 1,800 2,000 2,200
D
^^losej
200 400 600 800 1,000 1,200 1,400 1,609 1,800 2,000 2,200
1
2
3
4
5
6
7
¦
-V
* Urwi4»» (S £
j Hi inured
UOO 1,000 1.200 1,400 1,600 1,800 2,000 2,200
Figure 3-28. Comparison of experimental data (symbols) and model
predictions (solid lines) of (A) blood concentration, (B) liver concentration,
(C) adipose tissue concentration, (D) feces excretion (% dose), and (E)
urinary elimination (% dose) of TCDD after oral exposure to 1.5 ng/kg-day,
5 days/week, for 13 weeks in mice.
Y-axis represents concentration in pg/g, and X-axis represents time in days.
Source: Experimental data were obtained from Diliberto et at (2001).
3-110
DRAFT - DO NOT CITE OR QUOTE
-------
A
jnj
JJ^
~
to
^vJJJ
11
V- — Cb (pg/g) |
¦/ H measured I
200 400 600
1,000 1,200 1,400
2,000 2,200
B
y — Cli (pg/g) I
y J measured |
0 200 400 600 800 1,000 1,200 1,400 1,600 1,800 2,000 2,200
C
10,000
1,000
100
' — Cf (pg/g) I
' ¦ measured I
0 200 400 600 800 1,000 1,200 1,400 1,600 1,800 2,000 2,200
D
- Feces (%dose) I
I measured
200 400 600 800 1,000 1,200 1,400 1,600
2,000 2,200
1
2
3
4
5
6
7
y Urinary (%dose) I
0 200 400 600
1,000 1,200 1,400 1,600 1,800 2,000 2,200
Figure 3-29. Comparison of experimental data (symbols) and model
predictions (solid lines) of (A) blood concentration, (B) liver concentration,
(C) adipose tissue concentration, (D) feces excretion (% dose), and (E)
urinary elimination (% dose) of TCDD after oral exposure to 150 ng/kg-day,
5 days/week, for 13 weeks in mice.
Y-axis represents concentration in pg/g, and X-axis represents time in days.
Source: Experimental data were obtained from Diliberto et al. (2001).
3-111
DRAFT - DO NOT CITE OR QUOTE
-------
5 10 15 JO 3S SO 35 40 iS SO S5 HO «5 TO 75
Figure 3-30. PBPK model simulations (solid lines) vs. experimental data
(symbols) on the distribution of TCDD after a single acute oral exposure to
A-B) 0.1, C-D) 1.0, and E-F) 10 fig of TCDD/kg of body weight in mice.
Liver and adipose concentration for each dose was measured after 72 hours. Y-axis represents
the concentration in tissues (ng/g); insets A, C. and E represent liver tissue, whereas B, D, and F
correspond to adipose tissue. X-axis represents the time in hours.
Source: Experimental data were obtained from Santostefano et al. (1996).
3-112
DRAFT - DO NOT CITE OR QUOTE
-------
280 282 284 286 288 290 292 294 296 298 300 302 304 306 308 310 312 314 316 318 320
,51 Cli (ng/g)
@ experimental |
280 282 284 286 288 290 292 294 296 298 300 302 304 306 308 310 312 314 316 318 320
0 Cf (ng/g)
@ IB Experimental |
280 282 284 286 288 290 292 294 296 298 300 302 304 306 308 310 312 314 316 318 320
Figure 3-31. PBPK model simulation (solid lines) vs. experimental data
(symbols) on the distribution of TCDD after a single dose of 24 jig/kg BW on
GD 12 in mice.
Concentrations expressed as ng TCDD/g tissue. (A) maternal blood, (B) maternal liver, and (C)
maternal adipose tissue. Y-axis represents the tissue concentration whereas X-axis represents the
time in hours.
Source: Experimental data were obtained from Abbott et al. (1996).
3-113
DRAFT - DO NOT CITE OR QUOTE
-------
¦S 1.E+07
1
2
3
4
5
6
7
.c
0)
©
>
TJ
0
DQ
i.
©
a
c
©
TJ
i.
3
DQ
>
TJ
0
DQ
1.E+06
1.E+05
ro 1.E+04
O)
£ 1.E+03
1.E+02
1.E+01
1.E+00
CADM(rat)
Emond (rat)
CADM (human)
Emond (human)
2000
4000 6000 8000
Intake (ng/kg-day)
10000
12000
Figure 3-32. Comparison of the near-steady-state body burden simulated
with CADM and Emond models for a daily dose ranging from 0 to
10,000 ng/kg-day in rats and humans.
The rat model was ran for 13 weeks, and the human model was ran from ages 20 to 30. The time-
averaged concentration was used for each.
3-114
DRAFT - DO NOT CITE OR QUOTE
-------
o
o
CM
peak 5-year average concentration
gestational averagej;oncentration
o
LO
Iifetime average concentration
o
o
o
LO
Concentration-time profile
Less-than-lifetime scenario
Gestational scenario
Lifetime scenario
o
0
20
40
60
Year
Figure 3-33. TCDD serum concentration-time profile for lifetime, less-than-
lifetime, and gestational exposure scenarios, with target concentrations
shown for each; profiles generated with Emond human PBPK model.
3-115
DRAFT - DO NOT CITE OR QUOTE
-------
o
LO -
CM
Lifetime scenario
Less-than-lifetime scenario
Gestational scenario
Target concentration
o
o -
CM
O
LO
o
o -
//
o
LO
o -
0
20
40
60
Year
Figure 3-34. TCDD serum concentration-time profile for lifetime, less-than-
lifetime. and gestational exposure scenarios, showing continuous intake levels
to fixed target concentration; profiles generated with Emond human PBPK
model.
3-116
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
4. CHRONIC ORAL REFERENCE DOSE
This section presents U.S. Environmental Protection Agency (EPA)'s response to the
National Academy of Sciences (NAS) recommendations that EPA discuss more explicitly the
modeling of noncancer endpoints and develop a Reference Dose (RfD) to address noncancer
effects associated with oral 2,3,7,8-tetrachl orodi benzo-/>dioxin (TCDD) exposures. Section 2
details the selection of the animal bioassays with the lowest TCDD doses associated with the
development of adverse noncancer effects and the selection of relevant epidemiologic studies of
adverse noncancer health effects. Section 3 discusses the kinetic modeling and estimation of
human equivalent daily oral doses that are used in TCDD RfD development in this section. This
section discusses the modeling of noncancer health effects data associated with TCDD exposure
and the derivation of an RfD. Specifically, Section 4.1 summarizes the NAS comments on
TCDD dose-response modeling and EPA's response, including justification of selected
noncancer effects and statistical characterization of modeling results. Section 4.2 presents the
TCDD dose-response modeling undertaken for identification of candidate points of departure
(PODs) for derivation of an RfD. In Section 4.3, EPA derives an RfD for TCDD. Section 4.4
describes the qualitative uncertainties in the RfD. Finally, Section 4.5 presents two separate
quantitative analyses of uncertainty for the TCDD RfD. The first focuses on three data sets
(from two epidemiologic studies and one animal bioassay) and quantifies the consequences of
alternative decisions in the development of PODs based on these studies. The second develops
POD estimates for several Seveso cohort studies that did not qualify for consideration for RfD
derivation in the study selection process, but could be considered in the context of investigating
uncertainty limits for the RfD.
4.1. NAS COMMENTS AND EPA'S RESPONSE ON IDENTIFYING NONCANCER
EFFECTS OBSERVED AT LOWEST DOSES
The NAS recommended that EPA identify the noncancer effects associated with
low-dose TCDD exposures and discuss its strategy for identifying and selecting PODs for
noncancer endpoints, including biological significance of the effects.
4-1
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
With respect to noncancer end points, the committee notes that EPA does not use
a rigorous approach for evaluating evidence from studies... (p. 47 NAS. 2006b)
The Reassessment should describe clearly the following aspects:
1. The effects seen at the lowest body burdens that are the primary focus for
any risk assessment—the "critical effects."
2. The modeling strategy used for each noncancer effect, paying particular
attention to the critical effects, and the selection of a point of comparison
based on the biological significance of the effect; if the EDoi is retained,
then the biological significance of the response should be defined and the
precision of the estimate given... (p. 187. NAS. 2006b).
In this document, EPA has developed a strategy for identifying the noncancer data sets
and PODs that represent the most sensitive and toxicologically-relevant endpoints for derivation
of an RfD for TCDD. EPA began this process by using the animal bioassays and epidemiologic
studies that met its study inclusion criteria as sources of these data sets.
For all epidemiologic studies that were identified as suitable for further quantitative
dose-response analyses in Section 2.4.1, EPA has chosen to use NOAELs and LOAELs to
identify PODs; benchmark dose (BMD) modeling was not feasible given the nature of the data
presented in these studies. Figure 4-1 shows EPA's process for determination of PODs from
these key epidemiologic studies. EPA first evaluated the dose-response information in the study
to determine whether it provided an estimate of TCDD exposure and an observed health outcome
that was toxicologically relevant1 for RfD derivation. If such data were available, EPA
identified a NOAEL or LOAEL as a POD. For each of these, EPA applied a toxicokinetic model
to estimate the continuous oral daily intake associated with the POD that could be used in the
derivation of an RfD (see Section 4.2). If all of this information was available, the result was
included as a POD for derivation of a candidate RfD.
Figures 4-2 and 4-3 together present the strategy EPA used to evaluate the study/endpoint
combinations found in the animal bioassays that met EPA's study inclusion criteria, estimate
PODs, and develop a final set of candidate RfDs for TCDD. Figure 4-2 summarizes the
1 RfDs are based on health endpoints that are inherently adverse or clearly linked to downstream functional or
pathological alterations (U.S. EPA. 2002).
4-2 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
disposition of the 78 animal noncancer studies selected for TCDD dose-response analyses. Of
these studies, 16 were eliminated because EPA determined that they contained no
toxicologically-relevant endpoints that could be used to derive a candidate RfD (discussed
further in Section 4.2.1). EPA then identified PODs from the remaining bioassays; at this point,
Figure 4-2 refers to Figure 4-3, which is a flow chart of the iterative process used to estimate
PODs and compare them within and across the remaining studies to arrive at a final set of PODs
from these bioassays (see additional details below). From this final set of PODs, Figure 4-2
shows that EPA then eliminated 13 studies from further analysis with both a human equivalent
dose (FLED) LOAELhed >1 ng/kg-day and a NOAELhed or BMDLni n >0.32 ng/kg-day (see
Table 4-3); one additional study was also not carried forward because of the lack of toxicokinetic
information for estimation of an FLED. These dose limits were imposed to limit the size of the
analysis yet ensure representation of all important health effects associated with TCDD
exposure. From the final list of 48 studies, EPA derived 37 candidate RfDs, with 11 studies
presented as supporting information.
Figure 4-3 summarizes the strategy employed for identifying and estimating PODs from
the 62 animal bioassays with at least one toxicologically relevant endpoint for RfD derivation.
For the noncancer endpoints within these studies, EPA first evaluated the toxicological relevance
of each endpoint, rejecting those judged not to be relevant for RfD derivation. Next, initial
PODs based on the first-order body burden metric (see Section 3.3.4.2) and expressed as HEDs
(i.e., NOAELhed, LOAELhed, BMDLni n) were determined for all relevant endpoints
(summarized in Table 4-3). Because there were very few NOAELs and BMDL modeling was
largely unsuccessful due to data limitations (see Section 4.2), the next stage of evaluation was
carried out using LOAELs only. Within each study, effects not observed at the LOAEL (i.e.,
reported at higher doses) with BMDLheds greater than the LOAELhed were eliminated from
further analysis, as they would not be considered as candidates for the final POD on either a
BMDL or NOAEL/LOAEL basis (i.e., the POD would be higher than the PODs of other relevant
endpoints). In addition, all endpoints with LOAELhed estimates beyond a 100-fold range of the
lowest identified LOAELhed across all studies were (temporarily) eliminated from further
consideration, as they would not be POD candidates either (i.e., the POD would be higher than
the PODs of other relevant endpoints). For the remaining endpoints, EPA then determined final
potential PODs based on TCDD whole-blood concentrations obtained from the Emond rodent
4-3
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
PBPK models. HEDs were then estimated for each of these PODs using the Emond human
PBPK model. At this point, if the PBPK modeling results suggested considering additional
endpoints at higher doses, the process was repeated. From the final set of HEDs, a POD was
selected for each study, to which appropriate uncertainty factors (UFs) were applied following
EPA guidance (see Section 4.3.3 following). The resulting candidate RfDs were then considered
in the final selection process for the RfD. Other endpoints occurring at slightly higher doses
representing additional effects associated with TCDD exposure (beyond the 100-fold LOAELhed
"3
range) were evaluated, modeled, and included in the final candidate RfD array to examine
endpoints not evaluated by studies with lower PODs. In addition, Benchmark Dose (BMD)
modeling based on administered dose was performed on all endpoints for comparison purposes.
The final array of selected endpoints is shown in Table 4-4 (summary of BMD analysis) and
Table 4-5 (candidate RfDs).
The NAS recommended that EPA better justify the selection of response levels for
endpoints used to develop risk estimates. The NAS commented on EPA's decision to estimate
an EDoi (effective dose eliciting a 1% response) for noncancer bioassay/data set combinations as
a comparative tool across studies, suggesting that EPA identify and evaluate the levels of change
associated with adverse effects to define the benchmark response (BMR) level for continuous
noncancer endpoints.
The committee notes that the choice of the 1% response level as the POD
substantially affects ... the noncancer analyses.... The committee recommends
that the Reassessment use levels of change that represent clinical adverse effects
to define the BMR level for noncancer continuous end points as the basis for an
appropriate POD in the assessment of noncancer effects (p. 72, NAS. 2006b).
The committee concludes that EPA did not adequately justify the use of the
1% response level (the EDoi) as the POD for analyzing epidemiological or animal
bioassay data for ... noncancer effects (p. 18. NAS. 2006b).
2 In the standard order of consideration: BMDL, NOAEL, and LOAEL.
3 However, studies with a lowest dose tested greater than 30 ng/kg-day were not included in the expanded
evaluation.
DRAFT - DO NOT CITE OR QUOTE
4-4
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
In the 2003 Reassessment (U.S. EPA. 2003). EPA was not attempting to derive an RfD
when it conducted TCDD dose-response modeling. The 2003 Reassessment developed ED0i
estimates for noncancer effects in an attempt to compare disparate endpoints on a consistent
response scale. Importantly, the 2003 Reassessment defined the EDoi as 1% of the maximal
response for a given endpoint, not as a 1% change from control. Because RfD derivation is the
primary goal of noncancer health effects assessment in this document, the noncancer modeling
effort undertaken here differs substantially from the modeling in the 2003 Reassessment.
The NAS committee was concerned with the statistical power to determine the shape of
the dose-response curve at doses far below observed dose-response information. EPA agrees
that the shape of the dose-response curve in the low-dose region cannot be determined
confidently when based on higher-dose information. An observed response above background
near (or below) the BMR level is needed for discrimination of the shape of the curve and for
accurate estimation of an EDX or BMDL. Although many of the ED0iS presented in the 2003
Reassessment were near the lowest dose tested, responses at the lowest doses were often high
and much greater than a 1% response (i.e., 1% of the maximum response). The lack of an
observed response near the BMR level is often a problem in interpretation of BMD modeling
results.
In this document, EPA has used a 10% BMR for dichotomous data for all endpoints;
there were no developmental studies that accounted for litter effects, for which a 5% BMR would
be used (U.S. EPA. 2000). For continuous endpoints in this document, EPA has used a BMR of
1 standard deviation from the control mean whenever a specific toxicologically-relevant BMR
could not be defined. For the vast majority of continuous endpoints, EPA could not establish
unambiguous levels of change representative of adversity, which EPA defines as "a biochemical
change, functional impairment, or pathologic lesion that affects the performance of the whole
organism, or reduces an organism's ability to respond to an additional environmental challenge"
(U.S. EPA. 2009a). For body and organ weight change, EPA has previously established a BMR
of 10% change, which also is used in this document.
The NAS commented on EPA's development of EDoi estimates for numerous study/data
set combinations in the 2003 Reassessment, suggesting that EPA had not appropriately
characterized the statistical confidence around such model predictions in the low-response region
of the model.
4-5
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
It is critical that the model used for determining a POD fits the data well,
especially at the lower end of the observed responses. Whenever feasible,
mechanistic and statistical information should be used to estimate the shape of the
dose-response curve at lower doses. At a minimum, EPA should use rigorous
statistical methods to assess model fit and to control and reduce the uncertainty of
the POD caused by a poorly fitted model. The overall quality of the study design
is also a critical element in deciding which data sets to use for quantitative
modeling (NAS. 2006b, p. 18).
EPA should ... assess goodness-of-fit of dose-response models for data sets and
provide both upper and lower bounds on central estimates for all statistical
estimates. When quantitation is not possible, EPA should clearly state it and
explain what would be required to achieve quantitation (NAS. 2006b. p. 10).
The NAS also commented that EPA report information describing the adequacy of
dose-response model fits, particularly in the low response region. For those cases where
biostatistical modeling was not possible, NAS recommended that EPA identify the reasons.
The Reassessment should also explicitly address the importance of statistical
assessment of model fit at the lower end and the difficulties in such assessments,
particularly when using summary data from the literature instead of the raw data,
although estimates of the impacts of different choices of models would provide
valuable information about the role of this uncertainty in driving the risk estimates
(NAS. 2006b. p. 73).
To address this concern, in this document EPA has reported the standard suite of
goodness-of-fit measures from the benchmark dose modeling software (BMDS 2.1). These
include chi-square/^-values, Akaike's Information Criterion (AIC), scaled residuals at each dose
level, and plots of the fitted models. For the multistage model, when restricted lower-order
coefficients hit the lower bound (zero), EPA used likelihood ratio tests to evaluate whether the
improvement in fit afforded by estimating successively higher-order coefficients could be
justified. Goodness-of-fit measures are reported for all key data sets in Appendix G.
(Section 4.2.4.2 discusses the BMD modeling criteria for model evaluation.)
4-6
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
4.2. NONCANCER DOSE-RESPONSE ASSESSMENT OF TCDD
This section describes EPA's evaluation of TCDD dose response for noncancer endpoints
from studies that met the study inclusion criteria. Discussions include BMD modeling
procedures, kinetic modeling, and development of PODs for derivation of the RfD. Section 4.2.1
discusses the types of endpoints that are considered relevant by EPA for derivation of toxicity
values (U.S. EPA. 2005a. b, 1998. 1996. 1994. 1991) and lists the study/endpoint combinations
that were not considered for the TCDD RfD derivation, with supporting text in Appendix H.
Section 4.2.2 describes how EPA has used PBPK modeling to estimate effective internal
exposures as an alternative to using administered doses or body burdens based on first-order
kinetics. Section 4.2.3 details the dose-response analysis of the epidemiologic data, with
supporting information on kinetic modeling in Appendix F. Section 4.2.4 details the
dose-response analysis for the animal bioassay data; Appendix G provides the BMDS input
tables (Section G.l) and output for all modeling, including blood concentrations (Section G.2)
and administered dose (Section G.3).
4.2.1. Determination of Toxicologically Relevant Endpoints
The NAS committee commented on the low-dose model predictions and the need to
discuss the biological significance of the noncancer health effects modeled in the 2003
Reassessment. In selecting POD candidates from the animal bioassays for derivation of the
candidate RfDs, EPA considered the toxicological relevance of the identified endpoint(s) from
any given study. Some endpoints/effects may be sensitive, but lack general toxicological
significance because of lack of inherent adversity4, being an adaptive response, or not being
clearly linked to downstream functional or pathological alterations. Endpoints not considered to
be toxicologically relevant for TCDD include Cytochrome P-450 (CYP) induction, oxidative
stress measures, mRNA induction, protein phosphorylation, certain immune system responses,
gap junction disruption, and epidermal growth factor signaling. As an example, CYP induction
alone is not considered a significant toxicological effect given that CYPs are induced as part of
the normal hepatic metabolism of xenobiotic agents. Additionally, the role of CYP induction in
4 An adverse effect is defined in EPA's Integrated Risk Information System glossary as "a biochemical change,
functional impairment, or pathologic lesion that affects the performance of the whole organism, or reduces an
organism's ability to respond to an additional environmental challenge" (U.S. EPA. 2009a').
4-7 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
the noncancer toxicity of TCDD is unknown, thus, due to the lack of obvious pathological
significance, TCDD-induced CYP induction is not considered a relevant endpoint for RfD
derivation. Another example is oxidative stress. As an example, TCDD has been shown to
induce changes in oxidative stress markers, but no other indicators of brain pathology were
assessed (Hassoun et al.. 2003; Hassoun et al.. 2000; Hassoun et al.. 1998). In this case, it is
impracticable to link the markers of oxidative stress to a toxicological outcome in the brain; thus,
this endpoint is not considered relevant for RfD derivation. Studies otherwise meeting the study
inclusion criteria, but with no toxicologically-relevant endpoints that were considered suitable
for derivation of a candidate RfD are described and discussed in Appendix H.
4.2.2. Use of Toxicokinetic Modeling for TCDD Dose-Response Assessment
Because relevant toxicokinetic models for TCDD disposition in rodents and humans are
available, EPA has not applied the standard uncertainty factor approach in the derivation of the
TCDD RfD. In addition, because of the much slower elimination of TCDD in rodents than in
humans, EPA has determined that the standard uncertainty factor approach can underestimate the
interspecies toxicokinetic extrapolation factor by an order of magnitude or more (U.S. EPA.
2003). The toxicokinetic models chosen by EPA are the rodent and human PBPK models
described by Emond et al. (2006; 2005; 2004)5 and modified by EPA for this assessment as
described in Section 3.3.4 (hereafter referred to as the "Emond [rodent or human] PBPK
model"). Both the rodent and human models have a gestational component, which allow for
more relevant exposure comparisons between general adult exposures and the numerous
gestational exposure studies. Ideally, a relevant tissue concentration for each effect would be
estimated. However, at present, no models exist for estimation of all relevant tissue
concentrations. As virtually all TCDD is found in the adipose fraction of tissues, or bound to
specific proteins, a preferred approach to developing a dose metric would be to account for the
fat fraction of each tissue and protein binding; however, EPA has decided that the modeling of
such estimates is too uncertain, and EPA has not found sufficient data to implement this
approach. Therefore, EPA has decided to use the concentration of TCDD in whole blood as a
surrogate for tissue concentrations, assuming that tissue concentrations are proportional to
5The Emond PBPK models are three-compartment dynamic models.
4-8 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
whole-blood concentrations. Furthermore, because the RfD is necessarily expressed in terms of
average daily exposure, the blood concentrations are expressed as averages over the relevant
period of exposure for each endpoint. For the animal bioassays, the relevant period of exposure
is the duration of dosing, starting at the age of the animals at the beginning of the study. For
humans, the relevant period of exposure is generally a lifetime, which is defined as 70 years.
However, EPA varied the averaging time for the equivalent human blood concentrations to
correspond to the test-animal exposure duration in the following manner.
• For correspondence with animal chronic exposures,6 the human-equivalent TCDD
blood concentration is assumed to be the 70-year average.
• For correspondence with animal gestational exposures, the human-equivalent
TCDD blood concentration is assumed to be the average over 45 years for a
n
female, beginning at birth, plus 9 months of gestational exposure. Forty five
years of age is considered here as an upper limit on the age at which a typical
human female can conceive and bear a child.
• For correspondence with any other animal exposure duration, the
human-equivalent TCDD blood concentration is assumed to be the average over
the equivalent human exposure duration calculated backward from the peak
exposure plateau at or near the end of the 70-year scenario. The average is
determined from the terminal end of the human exposure period to be protective
of less-than-lifetime exposures occurring at any time in a lifetime; the daily oral
intake achieving the target blood concentration is smaller than for the same
exposure period beginning at birth. The determination of equivalent exposure
durations across species is problematic and somewhat arbitrary, so EPA uses the
average peak blood concentration as the human equivalent for all
o
less-than-chronic animal exposures (other than gestational). For the first-order
kinetics model, the average peak exposure is close to the theoretical steady-state
asymptote (see Section 3.3.4.2). However, for the Emond human PBPK model
used by EPA in this assessment, the timing of the peak exposure is
dose-dependent and tends to decline after 60 years in some cases. Therefore, the
5-year average TCDD blood concentration that includes the peak ("5-year peak")
is used as the relevant dose-metric for the PBPK model applications (see Section
3.3.6 and Figure 3-33).
6Assumed to be >75% of nominal lifetime, or about 550 days in rodents.
7 See Section 3.3.4.2 for a discussion of this issue, including a comparison of the 45-year old pregnancy scenario to
one beginning at age 25 in Table 3-24.
8By comparison to a half-lifetime equivalent (1 year in rodents, 35 years in humans), in the lst-order kinetic model
the ratio of body burden to oral intake does not differ significantly from the average-peak scenario; all shorter-term
scenarios differ even less (see Section 3.3.4.2). These relationships, with respect to the 5-year peak, hold for the
PBPK model results, as well (see Section 3).
4-9
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
4.2.3. Noncancer Dose-Response Assessment of Epidemiological Data
The following four epidemiologic studies describing noncancer endpoints were identified
in Section 2.4.1 as studies to be evaluated for development of PODs for derivation of candidate
RfDs: Baccarelli et al. (2008). Mocarelli et al. (2008), Alaluusua et al. (2004). and Eskenazi et al.
(2002b). Each of these studies described effects observed in the Seveso cohort (see detailed
study summaries in Appendix C and Table 2-2). Each study reported individual-level human
exposure measures and provided information from which EPA could determine a critical
exposure window of susceptibility over which the effective TCDD exposures could be quantified
for dose-response assessment. For studies that reported grouped data by TCDD exposure ranges,
the representative values for the ranges were determined by taking the geometric mean of the
range limits, assuming that the TCDD concentration distribution in the population is more likely
to be skewed (e.g., lognormal) than symmetrical (e.g., normal or uniform). A sufficient number
of significant digits are carried through intermediate results to avoid round-off error in the final
value. EPA used toxicokinetic modeling (Emond human PBPK model) to estimate daily TCDD
intake rates for the exposure groups presented in these studies (see Appendix F for details). The
exposure scenario in all of these studies, except Baccarelli et al. (2008). entailed an initial high
pulse exposure at the time of the plant explosion followed by low-level background exposure
over a period of several years across the critical exposure window, resulting in internal exposure
profiles characterized by a 5 to 10-fold difference in initial and final TCDD serum concentrations
(as lipid adjusted serum concentrations [LASC]). For these scenarios, EPA modeled both the
peak TCDD LASC and the average LASC over the critical window, then estimated daily average
continuous TCDD intakes over the critical-window duration corresponding to each of the peak
and critical-window average serum concentrations. Estimation of LASC and intakes was
accomplished using the Emond human PBPK model. EPA considered the critical-window
average exposures to be important, although they were much lower than the peak exposures,
because the relatively slow elimination of TCDD engenders concerns for an ongoing
contribution of residual TCDD body burdens to the adverse health outcomes during the period of
susceptibility. However, the overall average exposure does not reflect the influence of the much
4-10
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
higher peak exposure, which may be a significant factor in TCDD toxicity (Kim et al.. 2003).9
That is, EPA is uncertain as to whether the health outcomes, often observed many years beyond
the period of susceptibility, are a result of permanent damage from the initial high exposure or
more gradual impairment from longer-term ongoing exposure. For these reasons, EPA derived
the PODs for RfD consideration by averaging the TCDD intakes for the peak exposure and
critical-window exposure average, essentially treating each as equally likely. EPA focused on
identifying NOAELs and LOAELs for these studies. EPA did not conduct BMD modeling
because the covariates identified by the study authors could not be incorporated by modeling the
grouped response data. EPA's development of PODs for these studies is described in this section
and Appendix F; the results are shown in Table 4-1.
4.2.3.1. Baccarelli et al. (2008)
For Baccarelli et al. (2008). EPA was able to define a LOAEL in terms of the maternal
TCDD serum levels corresponding to neonatal thyroid stimulating hormone (TSH) level above
5 |i-Units TSH per mL of serum (5 |iU/mL) (See Appendix C, Section C. 1.2.1.5.7, and Table 2-2
for study details). The adversity benchmark of 5 |iU/mL is based on the WHO (1994) indicator
for follow up examination for potential hypothyroidism (see following discussion in Section
4.3.4.1). Baccarelli et al. (2008) performed regression modeling of neonatal TSH against
maternal TCDD LASC but did not estimate the equivalent oral intake. The regression model
related the level of TSH in 3-day-old neonates to TCDD concentrations in maternal plasma at
birth (given as LASC). The authors extrapolated maternal plasma concentrations from previous
measurements using a simple first-order pharmacokinetic model. The study authors also
reported group average neonatal TCDD serum levels for infants above and below the 5 |iU/mL
limit. However, because there is limited information regarding the relationship between
maternal and neonatal serum TCDD levels, EPA determined that there was too much uncertainty
in estimating maternal intake from neonatal TCDD serum concentrations directly. Therefore,
EPA determined the maternal intake at the LOAEL from the maternal serum-TCDD/TSH
regression model by finding the maternal TCDD LASC at which neonatal TSH exceeded
5 |iU/mL. EPA then used the Emond PBPK model under the human gestational scenario (see
9 Kim et al. (20031 found a significantly higher fraction of altered hepatic foci in rats treated with a single high
TCDD dose than those administered a continuous dose over 15 weeks, yielding similar terminal liver TCDD
concentrations.
4-11 DRAFT - DO NOT CITE OR QUOTE
-------
1 Section 4.2.2) to estimate the continuous daily oral TCDD intake that would result in a TCDD
2 LASC corresponding to a neonatal TSH of 5 |iU/mL at the end of gestation; EPA established the
3 resulting maternal intake (0.020 ng/kg-day) as the LOAEL, shown in Table 4-1 as a POD for
4 derivation of candidate RfDs (PBPK modeling details are shown in Appendix F).
5
6 4.2.3.2. Mocarelli et al. (2008)
7 Mocarelli et al. (2008) reported decreased sperm concentrations (20%) and decreased
8 motile sperm counts (11%) in men who were 1-9 years old in 1976 at the time of the accident
9 (initial TCDD exposure event) (see Appendix C, Section C. 1.2.1.5.8, and Table 2-2 for study
10 details). Men who were 10-17 years old in 1976 were not adversely affected. Serum (LASC)
11 TCDD levels were measured within 1 year of the initial exposure. Serum TCDD levels and
12 corresponding responses were reported by quartile, with a reference group of less-exposed
13 individuals assigned a TCDD LASC value of 15 ppt (which was the mean of individuals outside
14 the contaminated area). The lowest exposed group mean was 68 ppt (1st quartile). Because
15 effects were detected only among boys under the age of 10, EPA assumes there is a maximum
16 10-year critical exposure window for elicitation of these effects. However, for the exposure
17 profile, with a high initial pulse followed by an extended period of elimination with only
18 background exposure, the estimation of an average exposure resulting in the effect is somewhat
19 problematic. Therefore, EPA implemented a procedure for the estimation of the continuous
20 daily TCDD intake associated with the LOAEL in the Mocarelli et al. (2008) study using the
21 following 5-step process:
22
23
24
25
26
27
28
29
1. Using the Emond human PBPK model, the initial (peak) serum TCDD concentrations
(LASC) associated with the accident were back-calculated based on the time that had
elapsed between the explosion and the serum collection. As serum measurements were
taken within 1 year after the event, a lag time to measurement of 0.5 years was assumed.
The group average peak serum concentration for the 1st quartile was estimated to be
249 ppt.
30
31
2. The oral exposure associated with the peak serum TCDD concentration (peak exposure)
was calculated using the Emond PBPK model.
32
33
34
35
3. Starting with the peak exposure and accounting for background TCDD intake, the
average daily serum TCDD concentration experienced by a representative individual in
the susceptible lifestage (boys under 10 years old) was estimated using the Emond PBPK
model. The average subject age at the time of the event was 6.2 years. Consequently, a
4-12
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
critical exposure window for the cohort was estimated to be, on average, 3.8 years (i.e., a
boy aged 6.2 years would remain in this exposure window for 3.8 more years until he was
10 years of age). The critical window average serum concentration for the 1st quartile
group was estimated to be 57.7 ppt (45 ppt at 10 years).
4. Using the Emond PBPK model, the average daily TCDD intake rate needed to attain the
3.8-year average serum TCDD concentration in a boy 10 years old was calculated.
5. The LOAEL POD was calculated as the average of the peak exposure intake
(0.032 ng/kg-day) and the 3.8-year average exposure intake (0.0080 ng/kg-day), resulting
in LOAEL of 0.020 ng/kg-day, shown in Table 4-1 as a POD for derivation of a
candidate RfD.
The PBPK modeling details are shown in Appendix F.
4.2.3.3. Alaluusua et al. (2004)
For Alaluusua et al. (2004). the approach for estimation of daily oral TCDD intake is
virtually identical to the approach used for the Mocarelli et al. (2008) data. (See Appendix C,
Section C. 1.2.1.5.5, and Table 2-2 for study details.) Alaluusua et al. (2004) reported dental
effects in male and female adults who were less than 5 years of age at the time of the initial
exposure (1976). For the 75 boys and girls who were less than 5 years old at the time of the
accident, 25 (33%) were subsequently diagnosed with some form of dental enamel defect. For
the 38 individuals who were older than 5, only 2 (5.3%) suffered dental enamel defects at a later
date. In addition, the incidence of missing permanent teeth (lateral incisors and second
premolars) was 3 times as prevalent in zone ABR subjects compared with zone non-ABR
residents. A window of susceptibility of approximately 5 years is assumed. Serum
measurements for this cohort were taken within a year of the accident. Serum TCDD levels and
corresponding responses were reported by tertile, with a reference group of less-exposed
individuals assigned a TCDD LASC value of 15 ppt (ng/kg); the tertile group geometric means
were 72.1, 365.4, and 4,266 ppt. The incidence of dental effects for the reference group was
26%) (10/39). The incidence of dental effects in the 1st, 2nd, and 3rd tertile exposure groups was
10%o (1/10), 45%o (5/11), and 60%> (9/15), respectively. EPA judged that the NOAEL and
LOAEL were 72.1 and 365.4 ppt TCDD in serum (LASC), in the 1st tertile and 2nd tertile,
respectively. Following the same procedure used for the Mocarelli et al. (2008) study (see
Section 4.2.3.2), EPA estimated the continuous daily human oral TCDD intake associated with
each of the tertiles for both peak and average exposure across the critical exposure window,
4-13 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
assuming that the average age of the susceptible cohort at the time of the accident was 2.5 years.
Separate estimates for boys and girls were developed based on both the peak intake and average
intake across the critical exposure window (PBPK modeling details are shown in Appendix F).
The estimated averaged daily oral intakes for the tertiles, averaged for boys and girls, are 0.0655,
1.65, and 111 ng/kg-day for the peak exposure and 0.0156, 0.149, and 4.81 ng/kg-day for the
critical exposure window average. The LOAEL for this study was determined to be
0.897 ng/kg-day, which is the average of the peak exposure and window average exposure for
the second tertile. A study NOAEL of 0.0406 ng/kg-day for the first tertile was determined
similarly and serves as a POD for derivation of a candidate RfD in Table 4-1.
4.2.3.4. Eskenazi et al. (2002b)
The approach used to estimate daily TCDD intake in Eskenazi et al. (2002b) combines
the approaches EPA used for Baccarelli et al. (2008). Mocarelli et al. (2008). and Alaluusua et al.
(2004). Eskenazi et al. (2002b) reported menstrual effects in female adults who were
premenarcheal in 1976 at the time of the initial exposure (see Appendix C, Section C. 1.2.1.4.1
and Table 2-2 for study details). In Rigon et al. (2010), the median age at menarche was shown
to be 12.4 in Italian females with intergenerational decreases in age at menarche. Thus, EPA
established a window of susceptibility of approximately 13 years for this analysis. The average
age of the premenarcheal girls at the time of the initial exposure in 1976 was 6.8 years,
establishing an average critical-window exposure duration of 6.2 years for this cohort. Serum
samples were collected within a year of the accident from this cohort. However, serum TCDD
levels and corresponding responses were not reported by percentile, and no internal reference
group was identified. As for Baccarelli et al. (2008). Eskenazi et al. (2002b) developed a
regression model relating menstrual cycle length to plasma TCDD concentrations (LASC)
measured in 1976. The model estimated that menstrual cycle length was increased 0.93 days for
each 10-fold increase in TCDD LASC, with a 95% confidence interval of-0.01 to 1.86 days.
The determination of a LOAEL is somewhat arbitrary, with no independent measure of an
adversity threshold to establish the toxicological significance of a given increase in menstrual
cycle length. The study authors did not present data for unexposed premenarcheal girls (in
1976), so an appropriate reference population is not available. EPA did not conduct BMD
modeling because of the lack of a reference population and the inability to include the covariates
4-14
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
considered by the study authors in their analysis. However, an approximate LOAEL can be
estimated from Figure 1 in Eskenazi et al. (2002b). noting that both the length of the menstrual
cycle and its variance increases above TCDD concentrations of about 1,000 ppt. The highest
measured concentration is 16,500 ppt. Consistent with the previously established method for
determining representative values for age limits (Sections 4.2.3.2 and 4.2.3.3), the geometric
mean of 4,060 ppt for this range is assigned as a LOAEL. The lower range of TCDD
concentrations is too large to treat as a single group for estimating a NOAEL, but using the study
authors' regression model, a TCDD LASC of about 50 corresponds to a menstrual cycle length
of 28 days, generally considered to be the average normal length. These two (1976) serum levels
were then modeled by EPA using the Emond human PBPK model in the same manner as for
Mocarelli et al. (2008) and Alaluusua et al. (2004). but with a 6.2-year exposure window for the
premenarcheal girls. The resulting peak and window-average TCDD intakes for the 50 ppt
exposure are 0.0168 and 0.00364 ng/kg-day, respectively; the average of the two intakes is
0.0102 ng/kg-day. The peak and window-average TCDD intakes for the LOAEL exposure
(4,060 ppt) are 60.0 and 1.52 ng/kg-day, respectively; the average of the two intakes of
30.8 ng/kg-day defines the LOAEL POD. Further details of the PBPK modeling can be found in
Appendix F. Although 0.0102 ng/kg-day could be considered to be a NOAEL, there is too much
uncertainty in the upper end of the NOAEL range, given the very large (3,000-fold) difference
between it and the LOAEL, for using it as a NOAEL POD. The LOAEL of 30.8 ng/kg-day, also
uncertain in magnitude and toxicological significance, is 1,540-fold higher than the LOAEL
PODs for Mocarelli et al. (2008) and Baccarelli et al. (2008). and will not be a factor in the
derivation of the RfD. Therefore, the LOAEL for this study is not considered further in this
assessment except in the context of the RfD uncertainty analysis presented in Section 4.5.
4.2.4. Noncancer Dose-Response Assessment of Animal Bioassay Data
EPA followed the strategy illustrated in Figures 4-2 and 4-3 to evaluate the animal
bioassay data for TCDD dose response. For the administered average daily doses (ng/kg-day) in
each animal bioassay, EPA identified NOAELs and/or LOAELs based on the original data
presented by the study author. Section 2.4.2 identifies these values in Table 2-4 and in the study
summaries found in Appendix D. These became PODs for consideration in the derivation of an
RfD for TCDD. The candidate RfD values associated with these PODs are presented in
4-15
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Table 4-5. All PODs were converted to HEDs using the Emond PBPK models, with
whole-blood TCDD concentration as the effective dose metric. The remainder of this section
describes the steps in this process and concludes with the PODs from the animal bioassay data
that were considered for derivation of the RfD.
4.2.4.1. Use of Kinetic Modeling for Animal Bioassay Data
Whole-blood TCDD concentrations corresponding to the administered doses in each
mouse or rat bioassay qualifying as a final RfD POD were estimated using the appropriate
Emond rodent PBPK model. In each case, the simulation was performed using the exposure
durations, body weights, and average daily doses from the original studies. For all
multiple-exposure protocols, the time-weighted average blood TCDD concentrations over the
exposure period were used as the relevant dose metric. For single (gestational and
nongestational) exposures, the initial peak blood TCDD concentrations were considered to be the
most relevant exposure metric. Gestational exposures were modeled using the species-specific
gestational component of the Emond rodent PBPK model. Bioassays employing exposure
protocols spanning gestational and postpartum life stages were modeled by sequential
application of the gestational and nongestational models.
The Emond PBPK models do not contain a lactation component, so exposure during
lactation was not modeled explicitly. Only one bioassay (Shi et al.. 2007) considered as a POD
for RfD derivation included exposure during lactation. In Shi et al. (2007). pregnant animals
were exposed weekly to TCDD throughout gestation and lactation. Exposure was continued in
the offspring following weaning for 10 months. For assessment of maternal effects, the Emond
gestational model was used, terminating at parturition. For assessment of long-term exposure in
the offspring, the Emond nongestational model was used, ignoring prior gestational and
lactational exposure, with the assumption that the total exposure during these periods was small
relative to exposure in the following 10 months. The assumption is conservative in that effects
observed in the offspring would be attributed entirely to adult exposure, which is somewhat less
than the actual total exposure.
The model code, input files, and PBPK modeling results for each bioassay are reported in
Appendix E. The modeled TCDD blood concentrations were used for BMD modeling of
bioassay response data and determination of NOAELs and LOAELs. BMD modeling was
4-16
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
performed, as described in Section 3.5.2.2.1, by substituting the modeled blood concentrations
for the administered doses and calculating the corresponding BMDL. For each of these LOAEL,
NOAEL, or BMDL blood-concentration equivalents, corresponding HEDs were estimated using
the Emond human PBPK model for the appropriate gestational or nongestational scenario as
described previously (see Section 4.2.2).
4.2.4.2. Benchmark Dose Modeling of the Animal Bioassay Data
BMD modeling was performed for each study/endpoint combination using BMDS 2.1 to
determine BMDs and BMDLs. The input data tables for these noncancer studies are shown in
Appendix G, Section G.l, including both administered doses (ng/kg-day) and blood
concentrations (ng/kg [ppt]) and either incidence data for the dichotomous endpoints or mean
and standard deviations for the continuous endpoints (See Section 4.2.4.1 and Sections 3.3.4 and
3.3.5 for a description of the development of TCDD blood concentrations using kinetic
modeling).
Evaluation of BMD modeling performance, goodness-of-fit, dose-response data, and
resulting BMD and BMDL estimates included statistical criteria as well as professional judgment
of their statistical and toxicological properties. For the continuous endpoints, all available
models were run separately using both the assumption of constant variance and the assumption
of modeled variance. Saturated (0 degrees of freedom) model fits were rejected from
consideration. Parameters in models with power or slope parameters were constrained to prevent
supralinear fits, which EPA considers not to be biologically plausible and which often have
undesirable statistical properties (i.e., the BMDL converges on zero). Table 4-2 shows each
model and any restrictions imposed.
For the quantal/dichotomous endpoints, all primary BMDS dichotomous models were
run. The alternative dichotomous models were fit to several data sets, but the results were very
sensitive to the assumed independent background response and the fits were not accepted. The
confidence level was set to 95%, and all initial parameter values were set to their defaults in
BMDS. For the continuous endpoints, 1 standard deviation was chosen as the default for the
4-17
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
BMR when a specific toxicologically-relevant BMR could not be defined. For the dichotomous
endpoints, a BMR of 10% extra risk was used for all endpoints.10
The model output tables in Appendix G show all of the models that were run, both
restricted and unrestricted, goodness-of-fit statistics, BMD and BMDL estimates, and whether
bounds were hit for constrained parameters. After all models were run, the one giving the best
fit was selected using the selection criteria in the draft BMD Technical Guidance (U.S. EPA.
2000). Acceptable model fits were those with chi-square goodness-of-fit ^-values greater than
0.1. For continuous endpoints, the preference was for models with an asymptote term (plateau
for high-dose-response) because continuous measures do not continue to rise (or fall) with dose
forever; this phenomenon is particularly evident for TCDD. Unbounded models, such as the
power model, must account for the plateauing effect entirely in the shape parameter, generally
resulting in a supralinear fit. Also, for the continuous endpoints, thep-walue for the homogenous
variance test (Test 2) was used to determine whether constant variance (p > 0.1) or modeled
variance (p < 0.1) should be used. As BMDS offers only one variance model, model fits for
modeled variance models were not necessarily rejected if the variance model did not fit well
(Test 3 p-w alue < 0.05). Within the group of models with acceptable fits, the selected model was
generally the one with the lowest AIC. If the AICs were similar, the model with the lowest
BMDL was selected. However, particularly for continuous models, the fit of the model to the
control-group response and in the lower response range was assessed. Models with higher
BMDLs or AICs but much better fit to the lower response data were often chosen over the
nominally best-fitting model.
For many data sets, no models satisfied the acceptance criteria, and no clear
BMD/BMDL selection could be made. In these cases, model fits were examined on an
individual basis to determine the reasons for the poor fits. On occasion, high doses were
dropped, and the models were refit. Also, if a poor fit to the control mean was evident, the
model was refit to the data after fixing the control mean by specifying the relevant parameter in
BMDS. However, these techniques rarely resulted in better fits. If the fit was still not
acceptable, the NOAEL/LOAEL approach was applied to the study/data set combination. Most
of the problems with BMD modeling were a consequence of lack of response data near the
10There were no developmental studies that accounted for litter effects, for which a 5% BMR would be used.
4-18 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
BMR; many of the TCDD data sets failed to show a response near the BMR, whether it was a
10% dichotomous relative change or a continuous 1 standard deviation change. Responses at the
lowest doses were generally much higher than the BMR, resulting in a lack of "anchoring" at the
critical response levels of interest, resulting in insufficient information for precise numerical
estimation of BMDLs.
4.2.4.3. PODs from Animal Bioassays Based on HED and BMD Modeling Results
Table 4-3 summarizes the PODs that EPA estimated for each key animal study included
for TCDD noncancer dose-response modeling that also contained toxicologically relevant
endpoints (see Section 4.2.1 and Appendix H for excluded studies). After estimating the blood
TCDD concentration associated with a particular toxicity measure (NOAEL, LOAEL, or
BMDL) obtained from a rodent bioassay, EPA estimated a corresponding HED using the Emond
human PBPK model (described in Section 3). Table 4-3 summarizes the NOAEL, LOAEL, or
BMDL based on the administered animal doses for each key bioassay/data set combination.
Table 4-3 also summarizes the continuous daily HED corresponding to these administered doses
as 1st order body burdens and as whole-blood concentrations. The doses in Table 4-3 are defined
as follows, all in units of ng/kg-day:
• Administered Dose NOAEL: Average daily dose defining the NOAEL for the test species
in the animal bioassay
• Administered Dose LOAEL: Average daily dose defining the LOAEL for the test species
in the animal bioassay
• Administered Dose BMDL: BMDL for the test species based on modeling of the
administered doses from the animal bioassay
• First-Order Body Burden HED NOAEL: Average daily dose defining the NOAEL for
humans derived from the animal bioassay using the first-order kinetics body-burden
model
• First-Order Body Burden HED LOAEL: Average daily dose defining the LOAEL for
humans derived from the animal bioassay using the first-order kinetics body-burden
model
• First-Order Body Burden HED BMDL: Human-equivalent BMDL from BMD modeling
of the animal bioassay data using first-order body burdens
• Blood Concentration HED NOAEL: Average daily dose defining the NOAEL for
humans derived from the animal bioassay using the Emond human PBPK model
4-19 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
• Blood Concentration HED LOAEL: Average daily dose defining the LOAEL for humans
derived from the animal bioassay using the Emond human PBPK model
• Blood Concentration HED BMDL: Human-equivalent BMDL from BMD modeling of
the animal bioassay data using the Emond human PBPK model
An evaluation of key BMD analyses is presented in Table 4-4. Tables showing the best
model fit for each study/endpoint combination and the associated BMD/BMDL are shown in
Appendix G. As described in Section 4.2.4.2, the BMD modeling was largely unsuccessful,
primarily because of a lack of response data near the BMR, poor modeled representation of
control values, or nonmonotonic responses yielding poor fits. The comments column in
Table 4-4 lists reasons for poor results.
4.3. RFD DERIVATION
Table 4-5 lists all the studies and endpoints considered for derivation of the RfD in order
of candidate RfD from lowest to highest (The selection process was previously described in
Section 4.1). The range of studies includes three of the four human studies.11 Figure 4-4
(exposure-response array) shows all of the endpoints listed in Table 4-5 graphically in terms of
PODs in human-equivalent intake units (ng/kg-day). The human study endpoints are shown at
the far left of the figure, and the animal bioassay endpoints are arranged by category to the right.
Figure 4-5 demonstrates the same endpoints, arrayed by RfD value, showing the POD, applicable
UFs, and candidate RfD.
Table 4-5 illustrates the study, species, strain and sex, study protocol, and toxicological
endpoints observed at the lowest TCDD doses. The table also identifies the human-equivalent
BMDLs (when applicable), NOAELs, and LOAELs, as well as the composite UF that applies to
12
the specific endpoint and the corresponding candidate RfD. The NOAELS, LOAELs, and
BMDLs are presented as HEDs, based on the assumption that whole-blood concentration is the
toxicokinetically equivalent TCDD dose metric across species and serves as a surrogate for
13
tissue concentration. For rats and mice, these estimates relied on the two Emond PBPK
nThe RfD derived from the study of Eskenazi et al. (2002b') was outside the RfD range presented in Table 4-5.
12Extra digits are retained for transparency and comparison prior to rounding to one significant digit for the final
RfD.
13The procedures for estimating HEDs based on TCDD blood concentration are described in the preceding section.
4-20 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
models—one for the relevant rodent species and one for the human—as described previously
(see Section 3.3.4.3). The guinea pig and monkey studies that are included in Table 4-5 are
given in HED units based on the first-order body burden model (described in Section 3.3.4.2)
because there are no published PBPK models to estimate TCDD disposition in guinea pigs and
monkeys. The values listed for guinea pigs and monkeys are not directly comparable to those for
rats and mice but are probably biased low, as first-order body burden HED estimates for rats and
mice are generally 2- to 5-fold lower than the corresponding PBPK model estimates. The
LOAELs for the human studies also rely on the Emond PBPK model, as described in
Sections 4.2.2 and 4.2.3.
As is evident from Table 4-5, very few NOAELs and even fewer BMDLs have been
established for low-dose TCDD studies. BMD modeling was unsuccessful for all of the
endpoints without a NOAEL, primarily because of the lack of dose-response data near the BMR
(see discussion in Section 4.2). Therefore, the RfD assessment rests largely on evaluation of
LOAELs to determine the POD.
4.3.1. Toxicological Endpoints
As can be seen in Table 4-5, a wide array of toxicological endpoints has been observed
following TCDD exposure, ranging from subtle developmental effects to overt toxicity.
Developmental effects in rodents include embryotoxicity, neonatal mortality, dental defects,
delayed puberty in males, and several neurobehavioral effects. Reproductive effects reported in
rodents include altered hormone levels in females and decreased sperm production in males.
Immunotoxicity endpoints, such as decreased response to SRBC challenge in mice and decreased
delayed-type hypersensitivity response in guinea pigs, are also observed. Longer durations of
TCDD exposure in rodents are associated with organ and body weight changes, renal toxicity,
hepatotoxicity, and lung lesions. Adverse effects in human studies are also observed, which
include both male and female reproductive effects, increased TSH in neonates, and dental defects
in children. Other outcomes including diabetes (Michalek and Pavuk. 2008) and hepatic effects
(Michalek et al.. 2001b) have also been associated with adult human TCDD exposures, but EPA
was unable to quantify the exposure-response relationship (see Appendix C). All but three of the
study/endpoint combinations from animal bioassays listed in Table 4-5 are on TCDD-induced
toxicity observed in mice and rats; the other three study/endpoint combinations are effects in
4-21
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
guinea pigs and monkeys. Although the effects of TCDD also have been investigated in
hamsters and mink, those studies were not included for final POD consideration because the
effect levels were greater than those in Table 4-5, or because effective oral intakes could not be
estimated.
Three human studies were also included for final POD consideration in the derivation of
an RfD and are presented in Table 4-5 as candidate RfDs. All three human study/endpoint
combinations are from studies on the Seveso cohort. The developmental effects observed in
these studies were associated with TCDD exposures either in utero or in early childhood between
1 and 10 years of age. Baccarelli et al. (2008) reported increased levels of TSH in newborns
exposed to TCDD in utero, indicating a possible dysregulation of thyroid hormone metabolism.
Mocarelli et al. (2008) reported decreased sperm concentrations and decreased motile sperm
counts in men who were 1-9 years of age in 1976 at the time of the Seveso accident (initial
TCDD exposure event). Alaluusua et al. (2004) reported dental effects in adults who were less
than 5 years of age at the time of the initial exposure (1976).
4.3.2. Exposure Protocols of PODs
The studies in Table 4-5 represent a wide variety of exposure protocols, involving
different methods of administration and exposure patterns across virtually all exposure durations
and life stages. Both dietary and gavage administration have been used in rodent studies, with
gavage being the predominant method. Gavage dosing protocols vary quite widely and include
single gestational exposures, multiple daily exposures (for up to 2 weeks, intermittent schedules
that include 5 days/week, once weekly, or once every 2 weeks), and loading/maintenance dose
protocols, in which a relatively high dose is initially administered followed by lower weekly
doses. The intermittent dosing schedules require dose-averaging over time periods as long as
2 weeks, which introduces uncertainty in the effective exposures. In other words, the high unit
dose may be more of a factor in eliciting the effect than the average TCDD tissue levels over
time. Although the loading/maintenance dose protocols are designed to maintain a constant
internal exposure, these protocols are somewhat inconsistent with the constant daily TCDD
dietary exposures associated with human ingestion patterns.
The epidemiologic studies conducted in the Seveso cohort represent exposures over
different life stages including gestation, childhood, and young adulthood. The Seveso exposure
4-22
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
profile is essentially a high initial pulse TCDD exposure followed by a 20-30 year period of
elimination with only background exposures to TCDD and other dioxin-like compounds (DLCs).
While the exposures were measured soon after the initial pulse, health outcomes were realized,
or measured, 10-20 years following the initial exposure; the biologically-relevant critical
exposure window for susceptibility varies with effect and may be unknown. Therefore, the
effective exposure profiles for the Seveso cohort studies vary considerably. For the Mocarelli et
al. (2008) and Alaluusua et al. (2004) studies, where early childhood exposures proximate to the
initial event are associated with the outcomes, there is some uncertainty as to the magnitude of
the effective doses. Although the effects are associated with TCDD exposure in the first
10 years of life, it is not clear to what extent the initial peak exposure is primarily responsible for
the effects. It is also not clear if averaging exposure over the critical window is appropriate
given the fairly large (sixfold) difference between initial TCDD body burden and body burden at
the end of the critical exposure window. Because of the uncertainty in the influence of the peak
exposure relative to the average exposure over the entire window of susceptibility, the LOAELs
for both Mocarelli et al. (2008) and Alaluusua et al. (2004) are calculated as the average of the
peak exposure and average exposure across the critical exposure window (see Section 4.2 for
details).
For the gestational exposure study (Baccarelli et al.. 2008). the critical exposure window
is strictly defined and relatively short (9 months) and occurs long after the initial maternal
exposure (20-30 years). The maternal serum TCDD concentrations were measured 16-22 years
after the initial exposure when internal exposures were falling off less steeply; consequently,
there is less uncertainty in the toxicokinetic extrapolation between time of measurement and time
of birth. The narrow critical exposure window at a much later time than the initial exposure
(where the TCDD elimination curve is flattening) is assumed to lead to a relatively steady-state
exposure over the critical time period with much less uncertainty in the magnitude of the
effective dose. With the exception of Eskenazi et al. (2002b) (see Section 4.2.4), the effective
exposures for other effects reported for the Seveso cohort (see Section 2.4.1.1.1.4) have not been
quantified for consideration as an RfD POD and are not represented in Table 4-5 because either
critical exposure windows cannot be identified, unequivocal adverse effect levels cannot be
determined, or individual exposure estimates were not reported. Several of these studies,
however, are included in the uncertainty analysis presented in Section 4.5.
4-23
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
4.3.3. Uncertainty Factors (UFs)
Based on U.S. EPA (2002), UFs address five areas of uncertainty. Table 4-5 summarizes the
composite (total) UF applied to the POD for each endpoint.
For the PODs based on animal bioassays, the following UFs were applied:
• Interspecies extrapolation (UFa). A factor of 3 (10°'5) was applied for interspecies
extrapolation. The factor of 3 represents the residual uncertainty for toxicodynamics after
accounting for toxicokinetic differences with kinetic modeling. Although there are in
vitro studies (Budinsky et al.. 2010; Silkworth et al„ 2005) that report higher rodent
sensitivities than humans for AhR-dependent enzyme induction, EPA believes that there
is insufficient information on subsequent toxicological processes to conclude that rodents
are more sensitive than humans for downstream adverse effects.
• Human interindividual variability (UFh). A factor of 10 was applied to account for
human interindividual variability in susceptibility to TCDD because there is insufficient
information on sensitive populations to justify a lower value.
• LOAEL-to-NOAEL (UFi). For all PODs based on the animal bioassay endpoints lacking
a NOAEL, a factor of 10 was applied to account for LOAEL-to-NOAEL uncertainty.
The factor of 10 is the standard value in the absence of information suggesting a lower
value; the magnitude of the effects for most of the LOAELs is relatively high compared
to controls.
• Subchronic-to-chronic (UFs). A UF for study duration was not applied, because chronic
effects for animal bioassays are well represented in the database.
• Database factor (UFd). A UF for database deficiencies was not applied because the
database for TCDD contains an extensive range of human and animal studies that
examine a comprehensive set of endpoints. There is no evidence to suggest that
additional data would result in a lower RfD.
For the PODs based on epidemiologic studies, the following UFs were applied:
• Interspecies extrapolation (UFa). A UF for interspecies extrapolation was not applied
because human data were utilized for derivation of the RfD.
• Human interindividual variability (UFh). A factor of 3 was selected for interindividual
variability to account for human-to-human variability in susceptibility. The individuals
evaluated in the two principal studies included infants (exposed in utero) and adults who
were exposed when they were less than 10 years of age, groups that are considered to
represent sensitive lifestages. These studies considered together associate TCDD
exposures with health effects in potentially vulnerable lifestage subgroups. A UF of 1
4-24
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
was not applied because the sample sizes for the lifestages studied were relatively small,
which, combined with uncertainty in exposure estimation, may not fully capture the range
of interindividual variability. In addition, potential chronic effects were not fully
elucidated for humans and could possibly be more sensitive.
• LOAEL-to-NOAEL (UFl). A factor of 10 was applied to account for
LOAEL-to-NOAEL uncertainty. The factor of 10 for UFl is the standard value in the
absence of information suggesting a lower value.
• Subchronic-to-chronic (UFs). A UF for study duration was not applied, because,
although chronic effect levels are not well defined for humans, animal bioassays indicate
that duration of exposure is not likely to be a determining factor in toxicological
outcomes. Developmental effects and other short-term effects occur at doses similar to
effects noted in chronic studies.
• Database factor (UFd). A UF for database deficiencies was not applied because the
database for TCDD contains an extensive range of human and animal studies that
examine a comprehensive set of endpoints. There is no evidence to suggest that
additional data would result in a lower RfD.
4.3.4. Choice of Human Studies for RfD Derivation
For selection of the POD, the human studies are preferred, as EPA favors human data
over animal data of comparable quality. The human studies included in Table 4-5 (Baccarelli et
al.. 2008; Mocarelli et al.. 2008; Alaluusua et al.. 2004) each evaluate a segment of the Seveso
civilian population (i.e., not an occupational cohort) exposed directly to TCDD released from an
industrial accident. (The identification of PODs from these studies is detailed in
Sections 4.3.4.1, 4.3.4.2, and 4.3.4.3.) Thus, exposures were primarily to TCDD, with
apparently minimal DLC exposures beyond those associated with background intake,14
qualifying these studies for use in RfD derivation for TCDD. In addition, health effects
associated with TCDD exposures were observed in humans, eliminating the uncertainty
associated with interspecies extrapolation. The cohort members who were evaluated included
infants (exposed in utero) and adults who were exposed when they were less than 10 years of
age. These studies considered together associate TCDD exposures with health effects in
potentially vulnerable lifestages. Finally, the two virtually identical RfDs from different
14As an example, note the lack of statistically significant effects reported by Baccarelli et al. (20081 (Figures 2 C and
D) in regression models based on either maternal plasma levels of noncoplaner PCBs or total TEQ on neonatal TSH
levels.
DRAFT - DO NOT CITE OR QUOTE
4-25
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
endpoints in different studies provide an additional level of confidence in the use of these data
for derivation of the RfD for TCDD.
Although the human data are preferred, Table 4-5 presents a number of animal studies
with RfDs that are lower than the human RfDs. Two of the rat bioassays among this group of
studies—Bell et al. (2007b) (RfD = 1.4E-9 mg/kg-day based on delay in the onset of puberty)
and NTP (RfD = 4.6E-10 mg/kg-day based on liver and lung lesions) (2006a)—are of particular
note. Both studies were recently conducted. Both were very well designed and conducted, using
30 or more animals per dose group (see Table 4-6 for a discussion of these studies' strengths and
weaknesses); both also are consistent with and, in part, have helped to define the current state of
practice in the field. Bell et al. (2007b) evaluated several reproductive and developmental
endpoints, initiating TCDD exposures well before mating and continuing through gestation.
NTP (2006a) is the most comprehensive evaluation of TCDD chronic toxicity in rodents to date,
evaluating dozens of endpoints at several time points in all major tissues. Thus, proximity of the
RfDs derived from these two recent high-quality studies provides additional support for the use
of the human data for RfD derivation.
There are several animal bioassay candidate RfDs at the lower end of the RfD range in
Table 4-5 that are more than 10-fold below the human-based RfDs. Two of these studies report
effects that are analogous to the endpoints reported in the three human studies and support the
RfDs based on human data. Specifically, decreased sperm production in Latchoumycandane and
Mathur (2002) is consistent with the decreased sperm counts and other sperm effects in
Baccarelli et al. (2008). and missing molars in Keller et al. (2008a; 2008b; 2007) are similar to
the dental defects seen in Alaluusua et al. (2004). Thus, because these endpoints have been
associated with TCDD exposures in humans, these animal studies were not selected for RfD
derivation in preference to human data showing the same effects.
Another characteristic of the remaining studies in the lower end of the candidate RfD
distribution is that they are dominated by mouse studies (comprising 7 of the 9 lowest candidate
RfDs). EPA has less confidence in the candidate RfD estimates based on mouse data than those
based on either the rat or human data. EPA has less confidence in the use of the Emond mouse
PBPK model to estimate the PODs because of the lack of key mouse-specific data, particularly
for the gestational component (see Section 3.3.4.3.2.5). The toxicokinetic interspecies
extrapolation factors used for mice are very large, introducing a potential for large errors. The
4-26
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
ratio of administered dose to HED (Da:HED) ranges from 65 to 1,227 depending on the duration
of exposure. The Da:HED for mice is, on average, about four times larger than that used for rats.
In addition, each one of the mouse studies has other qualitative limitations and uncertainties
(discussed above and in Table 4-6) that further reduce confidence in using them as the basis for
theRfD.
4.3.4.1. Identification of POD from Baccarelli et al (2008)
Baccarelli et al. (2008) reported increased levels of TSH in newborns exposed to TCDD
in utero, indicating a possible dysregulation of thyroid hormone metabolism. The study authors
related TCDD concentrations in maternal plasma to neonatal TSH levels using a multivariate
linear regression model adjusting for a number of covariates (gender, birth weight, birth order,
maternal age, hospital, and type of delivery). Based on this regression modeling, EPA has
defined the LOAEL for Baccarelli et al. (2008) to be the maternal TCDD LASC of 235 ppt (at
delivery) corresponding to a neonatal TSH level of 5 |iU/mL.
The World Health Organization (WHO/UN1CEF/1CC1DD. 1994) established the
5 |iU/mL standard as an indicator of potential iodine deficiency and potential thyroid problems
in neonates. Increased TSH levels are indicative of decreased thyroid hormone (T4 and/or T3)
levels. The 5 |iU/mL limit for TSH measurements in neonates was recommended by WHO
(1994) for use in population surveillance programs as an indicator of iodine deficiency disease
(IDD). In explaining this recommendation, WHO (1994) stated that
While further study of iodine replete populations is needed, a limit of 5|iU/ml
whole blood... may be appropriate for epidemiological studies of IDD [iodine
deficiency disease.] Populations with a substantial number of newborns with
TSH levels above the limit could indicate a significant IDD problem.
For TCDD, the toxicological concern is not likely to be iodine uptake inhibition, but
rather increased metabolism and clearance of T4, as evidenced in a number of animal studies
(see discussion in Section 4.3.6.1). Baccarelli et al. (2008) discount iodine status in the
population as a confounder, as exposed and referent populations all lived in a relatively small
geographical area. It is unlikely that there was iodine deficiency in one population and not in the
other population based on iodine levels in the soil.
4-27 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Clinically, a TSH level of >4 |iU/mL in a pregnant woman is followed up by an
assessment of free T4, and treatment with L-thyroxine is prescribed if T4 levels are low (Glinoer
and Delange. 2000). This is to ensure a sufficient supply of T4 for the fetus, which relies on
maternal T4 exclusively during the 1st half of pregnancy (Chan et al.. 2005); (Calvo et al.. 2002;
Morreale de Escobar et al.. 2000). Adequate levels of thyroid hormone also are essential in the
newborn and young infant as this is a period of active brain development (Zoeller and Rovet.
2004; Glinoer and Delange. 2000). Smaller reserves, higher demand, and shorter half-life of
thyroid hormones in newborns and young infants also could make this lifestage more susceptible
to the impact of insufficient levels of T4 (Savin et al.. 2003; Greer et al.. 2002; Van Den Hove et
al.. 1999). Thyroid hormone disruption during pregnancy and in the neonatal period can lead to
neurological deficiencies, particularly in the attention and memory domains (Oerbeck et al..
2005). While such altered hormone levels are associated with decreased IQ scores, (e.g.. 2009).
report such associations among adolescents), the exact relationship between TSH increases and
adverse neurodevelopmental outcome is not well defined. A TSH level above 20 [jU/L in a
newborn infant is cause for immediate intervention to prevent mental retardation, often caused
by a malformed or ectopic thyroid gland in the newborn (WHO. 2007; Rovet. 2002; Glinoer and
Delange. 2000). Recent epidemiological data indicate concern for even lower level thyroid
hormone perturbations during pregnancy. For example, Haddow et al. (1999) reported that
women with subclinical hypothyroidism, with a mean TSH of 13.2 [J.U/L had children with IQ
deficits of up to 4 IQ points on the Wechsler IQ scale. Neonatal TSH within the first 72 hours of
birth [as was evaluated by Baccarelli et al. (2008)1 is a sensitive indicator of both neonatal and
maternal thyroid status (Delange et al.. 1983). Animal models have recently indicated that very
modest perturbations in thyroid status for even a relatively short period of time can lead to
altered brain development (Sharlin et al.. 2010; Rovland et al.. 2008; Sharlin et al.. 2008; Auso et
al.. 2004; Lavado-Autric et al.. 2003). Rodent bioassay results also suggest that elevated TSH
levels in neonates can affect sperm development as adults (Anbalagan et al.. 2010); this study
also reported reduced fertility among adult males and females with increased neonatal TSH
levels.
EPA has defined the LOAEL for Baccarelli et al. (2008) to be the maternal TCDD LASC
of 235 ppt corresponding to a neonatal TSH level of 5 |iU/mL, determined by the regression
modeling performed by the study authors. Using the Emond human PBPK model, the daily oral
4-28
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
intake at the LOAEL is estimated to be 0.020 ng/kg day (see Section 4.2.3.1). A NOAEL is not
defined because it is not clear what maternal intake should be assigned to the group below
5 |iU/mL,
4.3.4.2. Identification of POD from Mocarelli et al (2008)
Mocarelli et al. (2008) reported decreased sperm concentrations and decreased motile
sperm counts in men who were 1-9 years old in 1976 at the time of the Seveso accident (initial
TCDD exposure event). The sperm concentrations and motile sperm counts of men who were
10-17 years old in 1976 were not decreased. Serum (LASC) TCDD levels were measured in
samples collected within 1 year of the initial exposure. Serum TCDD levels and corresponding
responses were reported by quartile, with a reference group of less-exposed individuals assigned
a TCDD LASC value of 15 ppt (which was the mean of the TCDD LASC reported in individuals
outside the contaminated area). In the reference group, mean sperm concentrations and motile
sperm counts were approximately 74 million sperm/mL and 41%, respectively6. The lowest
exposed group (lst-quartile) TCDD LASC mean was 68 ppt. In the 1st quartile, mean sperm
concentrations of approximately 55 million sperm/mL15 and motile sperm counts of
approximately 36% were reduced about 25 and 12%, respectively, from the reference group.
Further decrease in these measures in the groups exposed to more than 68 ppt was minimal.
Relative to the reference population, the percent decreases in sperm concentrations were
approximately 25, 21, and 33% in the 2nd, 3rd, and 4th quartiles, respectively, and the percent
decreases in progressive sperm motility were approximately 20, 25, and 22% in the 2nd, 3rd, and
4th quartiles, respectively.
Mocarelli et al. (2008) also conducted a separate analysis of all the 22-31 year-old men
(combining all quartiles of the men exposed when they were 1-9 years of age). In the exposed
men, the mean total sperm concentration was reported by Mocarelli et al. (2008) to be
53.6 million/mL, with a value of 21.8 million/mL at 1 standard deviation below the mean. In the
comparison group that consisted of men not exposed to TCDD by the Seveso explosion and of
the same age as the exposed men, the mean total sperm concentration was 72.5 million/mL
(31.7 million/mL at 1 standard deviation below the mean).
15 This estimate is based on Figure 3 in Mocarelli et al. (2008).
4-29 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
There is no clear adverse effect level indicating male fertility problems for either of these
sperm effects. As sperm concentration decreases, the probability of pregnancy from a single
ejaculation also decreases; infertile conditions arise when the number of normal sperm per
ejaculate is consistently and sufficiently low. Previously, the incidence of male infertility was
considered increased at sperm concentrations less than 20 million sperm/mL (WHO. 1980).
More recently, Cooper et al. (2010) suggested that the 5th percentile for sperm concentration
(15 million/mL) could be used as a limit by clinicians to indicate needed follow-up for potential
infertility. Skakkeback (2010) suggests the following two limits for human sperm
concentrations: 15 million sperm/mL, based on Cooper et al. (2010) and 40 million sperm/mL.
Skakkeback justifies the upper level of 40 million sperm/mL citing a study by Bonde et al.
(1998) of couples planning to become pregnant for the first time; in the Bonde study, pregnancy
rates declined when sperm concentrations were below 40 million sperm/mL. Skakkeback
suggests that 15 million sperm/mL may be too low of a limit off for normal fertility and that
sperm concentrations between 15 million sperm/mL and 40 million sperm/mL may indicate a
range of reduced fertility. For fertile men, between 50% and 60% of sperm are motile (Swan et
al.. 2003; Slama et al.. 2002; Wiichman et al.. 2001). Any impacts on these reported levels could
become functionally significant, leading to reduced fertility. Low sperm counts are typically
accompanied by poor sperm quality with respect to morphology and motility (Slama et al..
2002).
EPA judged that the impact on sperm concentration and quality reported by Mocarelli
et al. (2008) is biologically significant given the potential for functional impairment. Although a
decrease in sperm concentration of 25% likely would not have clinical significance for a typical
individual, EPA's concern with the reported decreases in sperm concentration and total number
of motile sperm (relative to the comparison group) is that such decreases associated with TCDD
exposures could lead to shifts in the distributions of these measures in the general population.
Because male fertility is susceptible to reductions in both the number and quality of sperm
produced, such shifts in the population could result in decreased fertility in men at the low ends
of these population distributions. Further, in the group exposed due to the Seveso accident,
individuals 1 standard deviation below the mean had sperm concentrations of 21.8 million/mL;
this concentration falls at the low end of the range of reduced fertility (15 million and
40 million sperm/mL) suggested by (Skakkebaek. 2010).
4-30
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
EPA has designated the lowest exposure group (68 ppt) as a LOAEL, which translates to
a continuous daily oral intake of 0.020 ng/kg-day (see Section 4.2.3.2). The reference group is
not designated as a NOAEL because the serum levels were not measured for this group, directly,
and background exposures to dioxin-like compounds (DLCs) are relatively large by comparison
to TCDD in this group, introducing too much uncertainty in quantifying the full NOAEL
exposure (see discussion in Section 4.5). Also, there is no clear zero-exposure measurement for
any of these endpoints, complicating the interpretation of the reference group response as a true
"control" response (see discussion in Section 4.4). However, males less than 10 years old can be
designated as a sensitive lifestage as compared to older males who were not affected.
4.3.4.3. Identification of POD from Alaluusua et al. (2004)
Alaluusua et al. (2004) reported dental enamel defects and missing permanent teeth in
male and female adults who were less than 5 years of age, but not older, at the time of the initial
exposure (1976) in Seveso. EPA used the same approach to estimate daily TCDD intake as was
used for the Mocarelli et al. (2008) data; a window of susceptibility of about 5 years was
established. Serum measurements for this cohort were taken within a year of the accident.
Serum TCDD levels and corresponding responses were reported by tertile, with a reference
group of less-exposed individuals assigned a TCDD LASC value of 15 ppt (ng/kg); the tertile
group means were 130, 383, and 1,830 ppt. Both a NOAEL and LOAEL can be defined for this
study. The NOAEL is 0.12 ng/kg-day, corresponding to the TCDD LASC of 130 ppt at the first
tertile. The LOAEL is 0.93 ng/kg-day at the second tertile. The children in this cohort less than
5 years old can be designated as a sensitive lifestage as compared to older individuals who were
not affected relative to the reference group.
4.3.5. Derivation of the RfD
The two human studies, Baccarelli et al. (2008) and Mocarelli et al. (2008). have identical
LOAELs of 0.020 ng/kg-day. Together, these two studies define the most sensitive health
effects in the epidemiologic literature and constitute the best foundation for establishing a POD
for the RfD, and are designated as co-principal studies. Therefore, increased neonatal TSH
levels in Baccarelli et al. (2008) and male reproductive effects (decreased sperm count and
motility) in Mocarelli et al. (2008) are designated as co-critical effects. A composite UF of 30 is
4-31
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
applied to the LOAEL of 0.020 ng/kg-day to account for lack of a NOAEL (UFL =10) and
human interindividual variability (UFH = 3); the resulting RfD in standard units is
7 x 10~10 mg/kg-day. Table 4-7 presents the details of the RfD derivation.
4.3.6. Studies Reporting Outcomes Comparable to the Principal Studies Used to Derive
the RfD
Other animal and human epidemiological studies report associations between TCDD
exposures and effects similar to those reported by Baccarelli et al. (2008) and Mocarelli et al.
(20081
4.3.6.1. Dysregulation of Thyroid Hormone Metabolism Associated with Dioxin Exposure in
Neonates
One of the principal studies for the dioxin noncancer RfD, Baccarelli et al. (2008),
reported increased levels of TSH in newborns exposed to TCDD in utero, indicating a possible
dysregulation of thyroid hormone metabolism. No other human studies that met the selection
criteria of this analysis reported similar effects.
However, based on an analysis of over 20 epidemiology studies, Goodman et al. (2010)
concluded that DLC exposures were not clearly or consistently correlated with differences in
thyroid hormone levels in neonates and children less than 12 years of age. Focusing on neonatal
TSH for direct comparison to Baccarelli et al. (2008). Goodman et al. (2010). in Table 3 of their
analysis, identify 13 different studies, including Baccarelli et al. (2008), which measured infant
TSH levels within 1 week of birth. Of these studies, only Baccarelli et al. (2008) was
TCDD-specific and evaluated exposures well above ambient exposure levels. The other studies
examined total TEQ or individual DLCs near background exposure levels. The LOAEL derived
by EPA from Baccarelli et al. (2008) is approximately sixfold higher than the ambient total TEQ
exposure levels at the time of the exposures for the general Seveso population16 and more than
30-fold above an estimate of current TEQ levels (Lorber et al.. 2009). In the other studies, the
exposures appear to have been largely to DLCs, with TCDD as a minor component. Because the
equivalent TCDD exposure for DLCs is derived from TEF methodology, which is conservative
in nature (TEFs are higher than the median), the total TEQ concentrations would likely be over-
16 3
Estimated by EPA to be 3.5 x 10" ng/kg-day on a total TEQ basis (see Section 4.5.1.1 and Appendix F).
4-32 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
estimated (relative to TCDD) and uncertain. In addition, only 2 of the other 12 studies evaluated
by Goodman et al. (2010) reported TSH measures 3 days after birth, which is an international
standard and would be most comparable to those in Baccarelli et al. (2008). TSH levels
generally peak about 2 hours after birth then decline rapidly to typical long-term levels over the
next few days (Steinmaus et al.. 2010). Several of the studies included in Table 3 of Goodman et
al. (2010) evaluated cord-blood TSH measurements, which represent early high TSH
concentrations and are not directly comparable to 3-day measurements. Given these
considerations, particularly the relatively low ambient exposures and differences in the timing of
TSH measures, it would be unlikely that any consistent pattern would be detected across these
studies.
Several animal studies that met the selection criteria evaluated the effects of TCDD on
the thyroid or thyroid hormone levels. Overall, this set of studies show that TCDD affects
thyroid hormone levels and the thyroid gland. The studies of Sewall et al. (1995). Seo et al.
(1995). Van Birgelen et al. (1995a; 1995b). Crofton et al. (2005). and NTP (2006a) each reported
decreases in T4 levels. In response to TCDD treatment, NTP (2006a) reported increases in total
T3 concentrations, and both NTP (2006a) and Sewall et al. (1995) reported increased TSH
concentrations. Sewall et al. (1995) and Chu et al. (2007) reported reductions in thyroid
follicles, with Chu et al. (2007) noting that, of the health effects observed in their study, thyroid
effects were the most sensitive to TCDD exposures. Although none of these studies address in
utero or neonatal exposure, they show that TCDD can affect the level of thyroid hormones and
the thyroid organ in adult animals.
4.3.6.2. Male Reproductive Effects associated with Dioxin Exposures
The other principal study for the dioxin noncancer RfD, Mocarelli et al. (2008). reported
decreased sperm concentrations and decreased motile sperm counts in men who were aged
1-9 years at the time of the Seveso accident (initial TCDD exposure event). The sperm
concentrations and motile sperm counts of men who were 10-17 years old in 1976 were not
adversely affected. While no other human studies that met the selection criteria of this analysis
reported similar effects, a newly published study, Mocarelli et al. (2011). also reports male
reproductive effects. Several animal studies that met the study selection criteria also reported
male reproductive effects.
4-33
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Mocarelli et al. (2011) examined the relationship between maternal serum TCDD levels
and semen quality in male offspring. Analyses were based on 39 of the 78 men aged
18-26 years born to women residing in the areas most heavily polluted by dioxin after the
explosion in Seveso, Italy, in 1976 and age-matched controls (58 out of 123 recruited) born to
women residing in noncontaminated areas of Italy. In the exposed group of women, pregnancies
occurred between 9 months and 6 years after the accident (March 1977-January 1984). The
male offspring of these women were categorized based on whether they were breastfed (n = 21,
born to 20 mothers) or formula-fed (n = 18, born to 17 mothers) as infants. In the comparison
group, 36 were breastfed, and 22 were formula-fed. Sons born to dioxin-exposed women whose
spouses were also exposed to TCDD, as well as all men with reported diseases, were excluded.
TCDD exposures were based on estimated maternal serum concentration at conception.
To estimate these levels in the exposed group, the authors relied on maternal serum measures,
all of which were collected shortly after the accident in 1976-1977, and a biokinetic model
(Kreuzer et al.. 1997) that estimated TCDD elimination from the time of the accident to
conception for individual women (average half-life = 4 years). Mothers of sons in the
comparison group were assumed to be exposed to average background TCDD levels of 10 ppt
based on measurements reported in Eskenazi et al. (2004).
Semen samples were collected from all participants. These samples were maintained at
37°C and examined within an hour of ejaculation. For serum inhibin B and FSH analyses,
fasting blood samples were obtained the morning of semen collection. Statistical analyses were
performed on sperm properties, serum hormone levels, and TCDD levels using a "general linear
model" (Mocarelli et al.. 2011). Model covariates included age, duration of abstinence prior to
semen collection, smoking status, exposures to organic solvents, adhesives or paints, BMI,
alcohol use, educational level, and employment status.
Relative to the comparison group, men born to exposed mothers had decreased sperm
concentration (46 million vs. 81 million sperm/mL; p = 0.01), total sperm count (144 million vs.
231 million sperm; p = 0.03), and total number of motile sperm (51 million vs. 91 million;
p = 0.05). Relative to the breastfed comparison group, breastfed sons born to exposed mothers
exhibited decreased sperm concentrations (36 million vs. 86 million sperm/mL; p = 0.002), total
sperm counts (117 million vs. 231 million sperm; p = 0.02), and motile sperm counts (39 million
vs. 98 million; p = 0.01). Relative to the breastfed comparison group, breastfed sons born to
4-34
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
exposed mothers also exhibited increased FSH concentrations (4.1 vs. 2.6 IU/L; p = 0.03) and
decreased inhibin B levels (70.2 million vs. 101.8 pg/mL; p = 0.01). The formula-fed exposed
and comparison groups were not significantly different by any of these measures.
This study was well-designed with well-characterized exposures (for the exposed group),
which relied on measured sera TCDD concentrations and a peer-reviewed TCDD elimination
model to estimate maternal serum TCDD levels at the time of conception. Exposures in the
comparison group relied on estimates from other studies. The study excluded sons of fathers that
were likely highly exposed to TCDD, to limit potential influences from highly exposed fathers.
The study relies on self-reported recollection of infant feeding (i.e., breastfed vs. formula-fed),
which may lead to some misclassification based on recall error. Statistically significant
associations were evident for both the exposed men and their comparison group and breastfed
men and the breastfed comparison group.
In this study, elevated TCDD exposures during and after pregnancy (via breast-feeding)
led to long-term decrements in male reproductive endpoints. These effects included changes in
levels of hormones that affect spermatogenesis; they also include decreases in sperm
concentration and sperm motility.
In addition, two rodent bioassays also report sperm effects associated with dioxin
treatment. Latchoumycandane and Mathur (2002) reported decreased daily sperm production
and decreased reproductive organ weights in male albino Wistar rats given daily oral doses of
TCDD for 45 days. The LOAEL was 1.0 ng/kg-day, which corresponds to a LOAELhed of
0.016 ng/kg-day (see Table 4-5); a NOAEL was not identified. Simanainen et al. (2004b)
reported a reduction in daily sperm production and cauda epididymal sperm reserves in male rat
pups born to dams exposed to 300 ng/kg TCDD or higher on GD 15 by oral gavage. In this case
a NOAEL of 100 ng/kg was identified, which corresponds to a NOAELhed of 0.433 ng/kg-day,
with a LOAELhed of 1.7 ng/kg-day (see Table 4-5). Detailed descriptions of these studies can
be found in Appendix D.
4.4. QUALITATIVE UNCERTAINTIES IN THE RFD
Exposure assessment is a key limitation of the epidemiologic studies (of the Seveso
cohort) used to derive the RfD. The Seveso cohort exposure profile consists of an initial high
4-35
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
17
TCDD exposure followed by a drop in body burden to background levels over a period of
about 20 years, at which time the effects were observed. This exposure scenario is inconsistent
with the constant daily intake scenario addressed by the RfD methodology. The determination of
an effective average daily dose from the Seveso exposure scenario requires a consideration of the
biologically-relevant critical time-window of susceptibility and the influence of the peak
exposure on the occurrence of the observed effects, particularly when the peak exposure is high
relative to the average exposure over the critical exposure window. For one of the principal
studies (Mocarelli et al.. 2008). a maximum susceptibility exposure window can be identified
based on the age of the population at risk. However, the influence of the peak exposure on the
effects observed 20 years later is unknown, and the biological significance of averaging the
exposure over several years, with internal exposure measures spanning a 5.5-fold range, is
unknown. EPA has not developed guidance for large interval averaging. Furthermore, because
there is an assumption of a threshold level of exposure below which noncancer effects are not
expected to occur, averaging over large intervals could include exposures that are below a
threshold. The process used by EPA to estimate the LOAEL exposure for the Mocarelli et al.
(2008) study is a compromise between the most- and least-conservative alternatives; as such,
there is some uncertainty in the estimate, perhaps in the range of 3- to 10-fold in either direction.
This uncertainty also applies to the LOAEL determined for the developmental dental effects
reported in Alaluusua et al. (2004) and the increased menstrual cycle length reported in Eskenazi
et al. (2002b) (see Section 4.2.3.4); in both of those studies, the uncertainty is greater, as the
difference between peak and average internal exposures is an order of magnitude or more. The
LOAEL for increased TSH in neonates (Baccarelli et al.. 2008). however, is less uncertain
because the critical exposure window is much narrower (9 months), and the developmental
exposures occurred 20 to 30 years after the initial exposure, when internal TCDD concentrations
for the pregnant women likely were leveling off; that is, exposure over the critical window was
more constant and estimation of the relevant exposures was less uncertain. However, there is
some uncertainty in the magnitude of the exposures because they were estimated from
1'Mocarelli (20011 reported the release from the Seveso plant to contain a mixture of TCDD, ethylene glycol, and
sodium hydroxide. Because these chemicals are not thought to persist in the environment or in the body, coexposure
to these additional contaminants along with TCDD would not have a significant impact on longer-term TCDD
dose-response. For acute exposure, male reproductive or thyroid hormone effects are not evident for ethylene glycol
(U.S. EPA. 2009a'). It is unlikely that sodium hydroxide, being primarily a caustic agent, would cause these effects.
4-36 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
measurements in sera taken several years prior to pregnancy and do not take into account
changing patterns of exposure during pregnancy.
Another source of uncertainty using human epidemiologic data is the lack of completely
unexposed populations. The available TCDD epidemiologic data were obtained by comparing
populations that experienced elevated TCDD exposures to populations that experienced lower
exposures, rather than to a population with no TCDD exposure. An additional complicating
factor is coexposure to DLCs, which can act toxicologically in the same way as TCDD.
Although the accidental exposure to the Seveso women's cohort was primarily to TCDD,
background exposure was largely to DLCs. Eskenazi et al. (2004) reported that TCDD
comprised only 20% of the total toxicity equivalence (TEQ) in the serum of the reference group
that was not exposed as a result of the Seveso factory explosion, which implies that the effective
background TEQ exposure was approximately fivefold higher than exposure to TCDD. WHO
(1998) estimated that TCDD may comprise only 5-20% of background exposures to dioxin and
DLCs. The higher background exposure could be significant at the lower TCDD exposure
levels, with the effect diminishing as TCDD exposure increased. For dose-response modeling,
the effect of a higher background dose (i.e., total TEQ), if included, would be to shift the
response curve to the right, with responses now being associated with higher exposures. Adding
a constant to all exposures would also reduce the proportional spread of the exposures, which
would tend to alter the shape of the dose-response curve towards sublinear. Both the right shift
and the more sublinear shape would result in higher POD estimates. In addition, the response in
the reference population is not a true zero-exposure (TEQ-free) response. The actual magnitude
of the impact of the DLC background exposure is impossible to assess without knowing the zero-
exposure background response. The (TEQ-free) background response cannot be assessed as no
TEQ-free population exists. Ideally, an independent absolute measure of adversity in terms of
the response variable, such as the 5 nU/mL neonatal TSH benchmark, is needed for
dose-response modeling.
As part of the uncertainty analysis for the TCDD RfD, the possible influence of different
background DLC exposure assumptions on the POD estimates derived from the two principal
studies, Baccarelli et al. (2008) and Mocarelli et al. (2008). and one comprehensive animal
bioassay, NTP (2006a). is examined quantitatively in Section 4.5. In addition, the range of
4-37
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
possible PODs for other epidemiologic studies that did not pass all the selection criteria in
comparison to the principal studies is presented in Section 4.5.
A primary strength of the TCDD database is that analogous effects have been observed in
animal bioassays for most of the human endpoints, increasing the overall confidence in the
relevance to humans of the effects reported in rodents and the association of TCDD exposure
with the health outcomes reported in humans. Table 4-5 shows that low-dose TCDD exposures
are associated with a wide array of toxicological endpoints in rodents including developmental
effects, reproductive effects, immunotoxicity, and chronic toxicity. Effects reported in human
studies are similar, including male reproductive effects, increased TSH in neonates, and dental
defects in children; other human health effects such as female reproductive effects and chloracne
have been observed at higher exposures (see Appendix C). Severe liver toxicity, which is a
consistently reported effect in rodents, has not been observed in humans; Michalek et al.(2001c),
however, reported slightly elevated liver enzyme levels in serum and other nonspecific liver
effects for the Ranch Hand cohort, suggestive of mild liver toxicity. Overt immunological
endpoints, reported in the rodent bioassays, also have not been reported in human studies.
However, with respect to immunological effects, Baccarelli et al. (2004; 2002) evaluated
immunoglobin and complement levels in the sera of TCDD-exposed individuals from the Seveso
cohort and found reduced immunoglobulin in the highest exposure groups but no effect on other
immunoglobulins or on C3 or C4 complement levels and no indication of compromised immune
response. The latter finding indicates that at least one immunological measure in humans is not a
sensitive endpoint, as it is for mice, with large reductions in serum complement at low exposure
levels (White et al.. 1986).
Although there is a substantial amount of qualitative concordance of effects between
rodents and humans, quantitative concordance is not as strong, with reference to Table 4-5. The
differential sensitivity of mice and humans for the serum complement endpoint is one example.
Other examples of differential sensitivity are developmental dental effects and thyroid hormonal
dysregulation. Developmental dental defects are relatively sensitive effects in rodents, appearing
at exposure levels in mice (Keller et al.. 2008a; Keller et al.. 2008b; Keller et al.. 2007) more
than an order of magnitude lower than effect levels in humans (Alaluusua et al.. 2004). In
contrast, thyroid hormone effects are seen in rats (Croft on et al.. 2005) at 30-fold higher
exposures than for humans (Baccarelli et al.. 2008). Male reproductive effects (sperm
4-38
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
production) occur in rats (Latchournvcandane and Mathur. 2002) and humans (Mocarelli et al..
2008) at about the same dose. To what extent these differential sensitivities depend on specifics
of the comparison, such as species (mouse vs. rat), life-stage (e.g., fetal vs. adult), endpoint
measure (e.g., thyroxine [T4] vs. TSH), or magnitude of the lowest dose tested, cannot be
determined, so strong conclusions about quantitative concordance cannot be made.
A more detailed tabular and graphical presentation of qualitative and quantitative
cross-species comparisons of selected toxicological endpoints for all the animal and human
studies that met the EPA selection criteria is given in Appendix D.3. The endpoints include male
and female reproductive effects, thyroid hormone levels, and developmental dental effects, all of
which have been reported for humans. In addition, immunological and neurological effects are
shown because they are sensitive effects in experimental animal studies, although not evident in
humans. Hepatic effects, which are not shown in Appendix D.3, are evident in virtually all
rodent studies that looked for them and are often severe, but are not severe in humans. The
analysis presented in Appendix D.3 supports the conclusion that there is a substantial amount of
qualitative concordance of effects between rodents and humans, but a much lower quantitative
concordance. However, there are no endpoints in the selected animal bioassays that address
diabetes or glucose metabolism. There may be other animal studies showing effects of interest at
higher doses in those studies that did not meet the dose limit selection criterion.
A number of qualitative strengths and limitations/uncertainties are associated with the
animal bioassays listed in Table 4-5, as articulated in Table 4-6. Considering the issue of lowest
tested dose, the general lack of NOAELs and acceptable BMDLs is a primary weakness of the
rodent bioassay database. None of the eight most sensitive rodent studies in Table 4-5, spanning
an 18-fold range of LOAELs, had defined NOAELs or BMDLs. NOAELs or BMDLs were
established for only 4 of the next 13 rodent studies. In addition, many of these LOAELs are
characterized by relatively high responses with respect to the control population, so it is not
certain that a 10-fold lower dose (based on the application of UFL of 10) would be approximately
equivalent to a NOAEL. A major reason for the failure of BMD modeling was that the responses
were not "anchored" at the low end (i.e., first response levels were far from the BMR [see
Table 4-4]). Another major problem with the animal bioassay data was nonmonotone and flat
response profiles. The small dose-group sizes and large dose intervals probably contributed to
many of these response characteristics that prevented successful BMD modeling. Larger study
4-39
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
sizes with narrower dose intervals at lower doses are still needed to clarify rodent response to
TCDD.
Lower TCDD doses have been tested in rodents but almost entirely for investigation of
18
specialized biochemical endpoints that EPA does not consider to be toxicologically relevant for
the derivation of a noncancer RfD (see Appendix H). There is, however, a fundamental limit to
the lowest dose of TCDD that can be tested meaningfully, as TCDD is present in feed stock and
accumulates in unexposed animals prior to the start of any study. This issue is illustrated by the
presence of TCDD in tissues of unexposed control animals, often at significant levels relative to
the lowest tested dose in low-dose studies (Bell et al.. 2007b; Ohsako et al.. 2001; Vanden
Heuvel et al.. 1994; Vanden Heuvel et al.. 1994 see Text Box 4-1). Some DLCs also have been
measured in animal feeds (Bell et al.. 2007b; NTP. 2006a) and are anticipated to accumulate in
unexposed test animals, further complicating the interpretation of low-dose studies.
18 Enzyme induction, oxidative stress indicators, mRNA levels, etc.
4-40 DRAFT - DO NOT CITE OR QUOTE
-------
Text Box 4-1. Background levels of TCDD in Control Group Animals
TCDD tissue levels in control animals are rarely reported either explicitly or implicitly. Vanden Heuvel et al.
(1994). however, reported TCDD concentrations in livers of control animals (10-week-old female Sprague-Dawley
rats) of 0.43 ppt (ng/kg) compared to 0.49 ppt in the livers of animals given a single oral TCDD dose of 0.1 ng/kg.
Assuming proportionality of liver concentration to total body burden, the body burden of untreated animals was 87.8%
of that of treated animals at the lowest dose. The equivalent (single) administered dose for untreated animals (d0) can
be calculated as equal to 0.878 x (0.1 + do), assuming proportionality of body burden to administered dose and that all
animals started with the same TCDD body burdens. The calculation yields a value of 0.72 ng/kg for d0, which
represents the accumulated TCDD from all sources in these animals prior to being put on and during test. This value
would raise the nominal 0.1 ng/kg TCDD dose 8-fold to 0.82 ng/kg. The next higher dose of 1 ng/kg would be nearly
doubled to 1.72 g/kg. The impact on higher doses would be negligible, because the ratio of treatment dose to apparent
background exposure levels increases with higher treatment levels. Bell et al. (2007b) reported slightly higher levels
(0.66 ppt) in the livers of slightly older untreated pregnant female Sprague-Dawley rats (mated at 16-18 weeks of age
and tested 17 days later).
Ohsako et al. (2001) reported TCDD concentrations in the fat of offspring of untreated pregnant Holtzman rats
that were 46% of the TCDD fat concentrations in animals exposed in utero to 12.5 ng/kg (single exposure on GD 15).
This level of TCDD would imply a very large background exposure, but quantitation based on simple kinetic
assumptions probably would not reflect the more complicated indirect exposure scenario.
Bell et al. (2007b) also reported concentrations of 0.1 and 0.6 ppt TCDD measured in two samples of feed stock.
Assuming that the average of 0.35 ppt is representative of the entire supply of feed stock and a food consumption
factor of 10% of body weight per day, the average daily oral exposure from feed to these animals would be
0.035 ng/kg. Discrimination of outcomes from longer-term repeated exposures might be problematic at exposure
levels around 0.1 ng/kg-day. Background exposure was not much of an issue for Bell et al. (2007b). as the lowest
TCDD exposure level was 2.4 ng/kg-day (28-day dietary exposure).
NTP (2006c) reported TCDD concentrations in the liver and fat of untreated female Sprague-Dawley rats after
2 years on test that were 1% and 2.5% of the levels in the liver and fat of the low-dose TCDD treatment group
(2.14 ng/kg-day) (NTP. 2006a). respectively. Assuming proportionality of fat concentration and oral intake, control
animal exposure would have been approximately 0.05 ng/kg-day, similar to the estimate from Bell et al. (2007b). As
for the latter study, background intake for the NTP (2006a) study animals would not have a large effect on the
dose-response assessment given the lowest exposure level of 2.14 ng/kg-day.
In all of these studies, except the 28-day exposure in Bell et al. (2007b). control animals were gavaged with corn
oil vehicle. TCDD concentrations in corn oil were not reported in any of the studies.
1
2
3 4.5. QUANTITATIVE UNCERTAINTY IN THE RFD
4 The development of each candidate RfD in Sections 4.1 through 4.3 required the analysis
5 of numerous kinetic, toxicologic, and epidemiologic data sets. These analyses included
6 interpretive decisions that were made considering different sources of uncertainty in each study
7 and EPA's methods for developing RfDs. This section quantifies the impacts of some sources of
8 uncertainty encountered in the development of candidate RfDs (Sections 1.1 and 1.3 describe the
9 the NAS and SAB comments pertaining to uncertainty analysis for the RfD). In Section 4.5.1,
10 the impacts of some sources of uncertainty encountered in the development of candidate RfDs
11 based on Baccarelli et al. (2008). Mocarelli et al. (2008) and NTP (2006a) are elucidated using
12 "variable sensitivity" trees depicting the sensitivity of the POD value to choices made for PBPK
13 model variables and inputs. In Section 4.5.2, an additional range of potential PODs is presented
4-41
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
as a bounding analysis considering background DLC exposures and several epidemiologic
studies that did not qualify for RfD consideration, but for which limiting NOAEL and LOAEL
values can be estimated.
4.5.1. Development of Variable Sensitivity Trees for the Principal Epidemiological
Studies that were the basis of the RfD and the NTP (2006a) Rodent Bioassay
In Section 4.5.1, the impacts of some sources of uncertainty encountered in the
development of candidate RfDs based on Baccarelli et al. (2008), Mocarelli et al. (2008) and
NTP (2006a) are elucidated using "variable sensitivity" trees depicting the sensitivity of the POD
value to choices made for PBPK model variables and inputs. Baccarelli et al. (2008) and
Mocarelli et al. (2008) are the principal studies used to develop the RfD. NTP (2006a) is among
the most recent and comprehensive rodent bioassay studies of TCDD. For each of the three
PODs used to develop candidate RfDs from these studies, EPA generated plausible alternative
interpretations of information used to define judgment-based inputs for specific model variables.
The goal of this analysis is to provide quantitative insights on critical uncertainties encountered
in the development of the RfD by illustrating the consequences (quantified as alternative PODs
at the end of each branch in each tree) of plausible alternative interpretations of these key data
sets.
Previously, in their examination of low-dose carcinogenicity associated with
formaldehyde and chloroform exposures, Evans et al. (1994a; 1994b) assigned subjective
weights to each branch of a probability tree and calculated probability masses for population
risks associated with alternate interpretations of toxicological and pharmacokinetic data and
exposure information.19 In the examination of uncertainty undertaken in this report, EPA utilizes
the development of sensitivity trees; subjective probability weights are not developed for any of
the branches, and there is no propagation of probabilities across branches. Further, these trees do
not present a comprehensive analysis of quantitative uncertainty of the three candidate RfDs;
rather, EPA has focused on the impacts of key interpretive decisions largely dealing with
exposure and kinetic modeling uncertainties. However, it should be noted that because POD
values do not vary greatly across each of the three trees (less than threefold in either direction), it
19 Small (20081 discusses other studies of distributional approaches in risk assessment by Sielken and collaborators
that are similar to those of Evans and colleagues. These include the following: Sielken (1993. 19901. Holland and
Sielken (1993), Sielken and Valdez Flores (1999. 19961. and Sielken et al. (19951.
4-42 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
is unlikely that the distribution of probability mass resulting from specific probability
assignments would result in a significant amount of mass away from the chosen PODs. To
extend this analysis further, candidate RfDs can be estimated by dividing the POD values EPA
has generated by the appropriate uncertainty factors. The latter is largely a judgment call and
cannot be modeled, per se. However, the impact of the magnitude of uncertainty factors on the
RfD is proportional and relatively trivial to compute.
In this analysis, the structure of the decisions and the resulting POD estimates are
presented as sensitivity trees in graphical form (see Figures 4-6 through 4-8). In these figures,
the left-hand columns depict the variables considered in the sensitivity analysis. The values used
for these variables were either directly specified in the literature or were based on judgment
using exposure information provided in related papers. Each variable was assessed one at a time,
while fixing all the other variables at the values used in the primary POD estimation that was
used to develop the RfD in Section 4.3, termed hereafter the "standard pathway," and indicated
in Figures 4-6 through 4-8 by the bolded lines. Up to three significant digits are shown for the
PODs that are presented so that differences among the PODs across analytic choices can be
readily discerned.
4.5.1.1. Epidemiological Sensitivity Analyses
In estimating the PODs for the principal studies for the RfD (Baccarelli et al.. 2008;
Mocarelli et al.. 2008). a series of assumptions were made to model the exposure history of the
cohorts and to estimate an intake leading to the observed effect. In this section, the series of
trees highlights the effects of choosing alternative assumptions on the POD estimates.
4.5.1.1.1. Mocarelli et al. (2008)
To examine the impacts of potential uncertainties associated with the assumptions made
in estimating the standard pathway LOAEL POD in Mocarelli et al. (2008) (see Section 4.2.3.2),
EPA evaluated the impact of several alternate exposure assumptions on the oral intakes
associated with the POD, as shown in Figure 4-6. The left side of the figure depicts the variables
of the exposure analysis considered in the sensitivity analysis (i.e., background exposure,
exposure duration, measurement lag, and age at exposure). The values used for these variables
were not directly specified in the literature but were based on judgment of the exposure
4-43
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
information provided in Mocarelli et al. (2008) and related papers. All of these variables are
inputs to the Emond human PBPK model, which was used to estimate the actual exposures to the
affected population and the corresponding continuous intakes for determining the RfD POD; all
modeling for this analysis was carried out using the Emond human PBPK model. Each variable
was assessed one at a time, while fixing all the other variables at the values used in the standard
pathway analysis. The sensitivity analysis begins with the reported LASC of 68 ppt TCDD in
the LOAEL group. The terminal nodes at the bottom of the figure show the daily oral intakes
(ng/kg-day) resulting from each alternative value for the variables examined.
In Figure 4-6 and in the text that follows, the following abbreviations are used:
• "i5" identifies the intake associated with peak LASC exposure estimates.
• "W" identifies the intake associated with the average LASC over the actual exposure
window.
• "AVG" is the average of the intakes associated with "i5" and "W" Intakes associated
with either "i5" or "W" conceivably could have been selected as the primary POD.
Because of the relatively large differences between peak exposures and average
20
exposures decreasing over a relatively long time span, and the uncertainty of the relative
influence of acute high exposures vs. lower longer-term averages on the toxicological outcome,
EPA elected to use the average of the peak exposure intake (P) and the critical-window exposure
average intake (W) as the basis for the POD, giving equal weight to both (see discussion in
Section 4.2.3); these values are labeled as "AVG" across all terminal nodes in the tree.
For Figure 4-6, background exposures in the population (labeled "Background") were
estimated in several ways, taking into account background exposures of TCDD only or the
presence of other DLCs. The Emond human PBPK model was used to estimate all background
intakes by assuming a constant exposure from birth to time of measurement for each scenario
(see Appendix F for modeling details). The background value used in the standard pathway
analysis was based on an LASC of 15 ppt used by Mocarelli et al. (2008) in their analysis as the
TCDD level in the comparison group; this value was reported by Needham et al. (1998) to be the
median TCDD concentration in the unexposed reference adult population (25 years or older)
20 The modeled TCDD LASC decreased by a factor of 5.5 from peak exposure to the terminal value at 10 years.
4-44 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(designated "Needham" in Figure 4-6). EPA estimated a corresponding daily TCDD intake of
3.5 x 10 4 ng/kg-day from birth, assuming that 15 ppt was obtained at age 35. The alternative is
an age-specific background intake based on an average TCDD concentration of 40.5 ppt for girls
21
less than 12 years of age (designated "Eskenazi" in Figure 4-6) (Eskenazi et al.. 2004).
Assuming that background TCDD concentrations were similar for boys and girls in the Seveso
cohort, EPA estimated an average TCDD intake of 3.52 x 10 ng/kg-day corresponding to the
same average 40.5 ppt LASC for boys of similar age. The 10-fold higher value than for the adult
background is likely a result of higher food consumption in children and a higher average
environmental concentration for the relevant childhood exposure period (1964-1976) than for
the adult exposures (ca. 1941-1976) (Lorber. 2002; Pin sky and Lorber. 1998).
The other alternate background scenarios take into account the presence of DLCs (i.e.,
other than TCDD) in the background exposure. Because DLCs are presumed to behave in the
same manner as TCDD (for AhR induction), the magnitude of the background DLC exposure is
an important concern in establishing the POD.
Both the "Needham" and "Eskenazi" background exposure scenarios are evaluated for
DLCs. For this analysis, the total DLC-TEQ, whether reported by the authors or modeled herein,
is assumed to be applicable for estimation of equivalent TCDD intake. However, the reported
TEQ values are based on serum concentrations, while the TEFs on which the TEQ values are
based are largely derived from oral dosing studies conducted in experimental animals. The
outcomes from such studies implicitly account for DLC toxicokinetics (i.e., absorption,
distribution, metabolism, and elimination). Applications of TEFs to DLC tissue concentrations
do not account for toxicokinetics. Whole body half-life estimates for the DLCs vary from about
6 months to 20 years (Ogura et al.. 2004; Flesch-Janvs et al.. 1996). so the equivalence of
internally estimated TEQ with ingested quantities is not strictly valid. Currently, there is no
human PBPK model capable of addressing all the DLC congeners, although both EPA (U.S.
EPA. 2003) and Lorber (2002) have used DLC half-life estimates and tissue concentrations to
estimate intake rates of individual DLCs in humans; however, the dioxin-like PCBs were not
included in either Lorber (2002) or EPA (2003). In addition, the TEF methodology is designed
21 Table 3 in Eskanazi et al. (20041 reports the results of two pools of sera collected from girls aged 0-12 years, who
did not reside in areas affected by the Seveso accident and were presumably exposed only to background levels of
TCDD. EPA estimated the mean of these reported sera concentrations of 47.6 ppt TCDD and 33.4 ppt TCDD.
4-45 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
to be health protective, in that the TEFs are not central tendency estimates but biased high by
design (Van den Berg et al.. 2006). Two different approaches for estimating background DLC
exposures are presented.
The first approach models the exposure directly, by matching the total TEQ (TCDD
included) at the time of measurement with the corresponding intake using the Emond model.
The total TEQ for the Eskenazi background scenario is estimated from Table 3 in Eskenazi et al.
(2004). The average TEFos DLC-TEQ contribution was estimated by multiplying the 0-12 year
old average of 76.05 ppt (based on TEFgg values) by a factor of 0.7. The factor of 0.7 is an
approximation based on a ratio of 0.72 for the TEFos to TEF98 background DLC-TEQ values for
the Ranch Hand cohort (Pavuk et al.. 2007) and a ratio of 0.65 based on serum collected in 1998
for 78 Seveso women (Warner et al.. 2005). The Ranch Hand value was determined by Pavuk
et al. (2007) and reported directly. The 0.65 ratio for the Seveso women was determined by EPA
by calculating the total TEQ using both the 1998 and 2005 TEF values from the median
congener concentrations reported by Warner et al. (2005). Figure 4-6 shows the results of
modeling total TEQ directly under this approach, labeled as "Modeled" under the "Total TEQ"
branches for both the "Needham" and "Eskenazi" background exposure scenarios.
The second approach for estimating DLC background exposure is a simple additive one,
in which an estimate of background DLC-TEQ intake is added to the modeled TCDD intake.
This is accomplished by assuming that TCDD comprises 10% of the total background TEQ,
which is about the proportion of TCDD to total TEQ in serum as estimated by WHO (1998). In
addition, TCDD is about 10% of the total serum TEQ as calculated by EPA from the NHANES
(2001/2002) data reported by Lorber et al. (2009). However, the same qualifier holds here as for
modeling total TEQ directly, in that the TEFs are based on oral exposures. If the proportional
relationship (i.e., TCDD is 10% of total TEQ) is assumed for oral exposure, the modeled TCDD
intake is simply multiplied by nine to get the corresponding DLC-TEQ intake. The TCDD
background exposures for the Needham and Eskenazi background scenarios are
3.5 x 10 4 ng/kg-day and 3.5 x 10 3 ng/kg-day, respectively (see Appendix F for details); the
corresponding DLC-TEQ intakes for the additive background approach are
_3 _2
3.15x10 ng/kg-day and 3.15 x 10 ng/kg-day, respectively. Figure 4-6 shows the additive
approach, labeled as "DLC-TEQ added" under the "Total TEQ" branches for both the
"Needham" and "Eskenazi" background exposure scenarios.
4-46
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
"Exposure duration" refers to the duration of the elevated (external) TCDD exposures
immediately following the Seveso accident, which is not known with certainty. In the standard
pathway analysis, the "exposure duration" of the TCDD exposures due to the Seveso accident
was modeled using the Emond model as a single pulse on 1 day (i.e., 24 hours). The alternative
also uses the Emond model but models the exposures following the Seveso accident using pulse
doses on two consecutive days (i.e., 48 hours).
"Measurement lag" refers to the period of time between TCDD exposure following the
Seveso accident and the collection of blood for future TCDD analyses. Within the Seveso
cohort, serum samples were collected in 1976 and 1977, so in the standard pathway analysis, an
average measurement lag time of 6 months was assumed for exposure to TCDD. The alternative
analyses simulate lag times of 1 month and 1 year.
"Age at exposure" is the average age of the susceptible lifestage (boys, 1-9 years old) at
the time of the Seveso accident. Within the cohort, the average age at exposure was reported to
be 6.2 years, which was used in standard pathway analysis. The alternative analysis considers
individuals who would have been 1 year or 9 years of age at the time of the Seveso accident,
representing the bounds of the susceptible age range. This category is included to show the
potential range of exposures across the cohort rather than to evaluate plausible alternatives to the
mean age of 6.2 years. That is, the intakes associated with ages 1 or 9 would not be considered
as PODs.
Overall, excluding the age-at-exposure variable, the daily intakes (TCDD or total TEQ)
based on the alternative assumptions in this tree vary between 0.007 ng/kg-day (Wfor 1-month
measurement lag) and 0.05 ng/kg-day (P for modeled total TEQ, Needham background). This
range spans the LOAEL for the standard pathway analysis of 0.020 ng/kg-day by less than a
factor of three on each side. The AVG values vary over a smaller range from 0.013 ng/kg-day
(TCDD-only, Eskenazi background) to 0.0335 ng/kg-day (modeled total TEQ, Needham
background), bracketing the LOAEL for the standard pathway by less than a factor of two.
The ratio of peak intake to window-average intake (P: Wratio) is of interest in evaluating
the range of exposures over which an average is taken. The P: W ratio is 4 for the standard
pathway POD. In general, the P: Wratios are greater than three across the terminal nodes.
However, the higher the background exposure, the lower the peak intake and the lower the P: W
ratio and the lower the impact of averaging P and W. The P: Wratio is lowest for all the
4-47
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Eskenazi background scenarios, decreasing to about a factor of 1.3 for the TEQ analyses. The
higher background exposure scenario had the largest impact on the TCDD-only intakes, with a
35% lower AVG than for the standard pathway RfD LOAEL POD. The next largest variation
was for the 48-hour exposure time, with a 24% lower AVG than for the 24-hour scenario.
However, the modeled exposures on each of the 2 days were equal when, in reality, they would
be decreasing with time, such that the peak is somewhat underestimated in this exercise; longer
exposure scenarios assuming constant levels would not be realistic. The largest differences in
the other direction were obtained for the modeled total TEQ scenarios, with a 67% higher AVG
for the Needham background assumption (compared to the standard pathway RfD POD) and a
30%) higher AVG for the Eskenazi background assumption. Note that any DLC background
exposure estimate based on TEQ will be an over-estimate because of the conservative nature of
the TEF methodology. All the other alternative assumptions resulted in a 16% or lower change
in the AVG values. Although not a consideration for defining the POD, the TCDD AVG intakes
across the susceptible age range (1-9 years) were within 5% of the standard pathway RfD POD,
but with a large P: Wratio (10) for 1-year-olds.
In summary, the quantitative uncertainties evaluated here for the RfD POD based on
Mocarelli et al. (2008) span less than a 3-fold range in either direction. The largest differences
are those between peak and window-average exposures, which decrease when considering the
alternative Eskenazi background. Using the latter, the AVG POD is about half of the RfD POD,
but is more impacted by background DLC exposure; considering the TEQ contribution from this
background exposureresults in approximately the same value as the RfD POD with additive
background DLC. Using the directly-modeled approach, background DLC exposure has a larger
impact on the standard RfD POD, increasing it by 67%. At this time, EPA cannot recommend
any approach for incorporating background DLC exposure directly into the POD for the RfD.
Overall, given the bidirectional nature and relatively small magnitude of the uncertainties, EPA
believes that this sensitivity analysis provides support for the magnitude of the RfD.
4.5.1.1.2. Baccarelli et al. (2008)
To examine the impacts of potential uncertainties associated with the assumptions made
in estimating the standard pathway POD for Baccarelli et al. (2008) (see Sections 4.2 and 4.3),
EPA analyzed alternate assumptions about exposure and the level of change in neonatal TSH
4-48
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
levels associated with the designation of a LOAEL or a NOAEL from this study as shown in
Figure 4-7. For the NOAEL in Figure 4-7, the equivalent LOAEL (by multiplying by 1022) is
also shown for direct comparison to the LOAEL estimates. The uncertainty considerations and
the approach presented in Figure 4-7 are similar to those depicted in Figure 4-6, but the variables
are different. There are several ways in which a POD could be derived from the Baccarelli et al.
(2008) study. In the standard pathway RfD analysis, EPA used the study authors' regression
model results from their Figure 2A (designated the "Regression Model") to determine a LOAEL
based on the maternal plasma concentration corresponding to neonatal TSH levels of 5 nU/mL.
The regression model was used to account for covariates that influenced the dose-response
relationship. Three alternative values are examined by selecting specific points or ranges from
the figures in the Baccarelli paper, without consideration of the regression modeling results (the
"graphical method"). The alternative values, therefore, do not account for the covariates. The
first assumes a NOAEL of 40 ppt maternal LASC, which is essentially the highest TCDD
concentration above which neonatal TSH levels are consistently above 5 [j,U/mL [see Figure 2A
in Baccarelli et al. (2008)1. The figure (2A) shows that 5 of the 6 neonates born to women who
had TCDD concentrations above 40 ppt had TSH levels above 5 nU/mL; among the 45 women
who had TCDD concentrations below 40 ppt, only two had babies with TSH levels above
5 [j,U/mL. The second alternative assumes that the 6 neonates born to women with TCDD LASC
above 40 ppt comprise a LOAEL group, with a median maternal LASC of 90 ppt. The
third alternative assumes a LOAEL at the highest neonatal TSH level (8.5 (j,U/mL) shown in
Figure 2A, which corresponds to a maternal TCDD LASC of 312 ppt.
Background exposures in the population were estimated in several ways. The
background TCDD exposure used in the standard pathway RfD analysis was based on
continuous intake necessary to obtain 15 ppt at 30 years for females (the "Needham"
background); the modeled TCDD intake was 3.9 x 10 4 ng/kg-day, slightly higher than that for
males. To examine the maternal TEQ exposures associated with a LOAEL based on a neonatal
TSH level of 5 [jU/mL, EPA relied on the regression results reported in Baccarelli et al. (2008).
Baccarelli et al. (2008) reported maternal plasma TEQ concentrations in the following two ways:
(1) PCDDs, PCDFs, coplanar PCBs, without noncoplanar PCBs (see Figure 2B) and (2) PCDDs,
22 A tenfold factor is used because the LOAEL POD is divided by a UFL of 10 in the RfD derivation. The
"equivalent" LOAEL is not meant to be an alternative LOAEL but is used strictly for comparison.
4-49 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
PCDFs, coplanar PCBs, and noncoplanar PCBs, termed total TEQ (see Figure 2D). The
concentrations in their Figures 2B and 2D are reported as TEQs and were modeled as TCDD for
this analysis. Excluding the noncoplanar PCBs, maternal TEQ levels of 219 ppt in serum are
associated with neonatal TSH level of 5 [j,U/mL. For the total TEQ, maternal TEQ levels of
485 ppt in serum are associated with a neonatal TSH level of 5 nU/mL. Confidence in the total
TEQ estimate is lower than that for the one without the noncoplanar PCBs because of the lower
significance of the total TEQ regression coefficient (p = 0.14) than the one without the
noncoplanar PCBs (p = 0.005).
For the standard pathway RfD analysis, the maternal "age at conception" was set at
30 years, which was the average reported in Baccarelli et al. (2008). The alternative assumes the
maternal age at conception to be 45 years of age; this is the standard gestational scenario used in
estimating the human equivalent doses for the animal bioassays reporting reproductive or
developmental effects and is considered to be a reasonable upper end of female fertility.
The alternative LOAELs based on this analysis of Baccarelli et al. (2008) vary between
0.005 and 0.059 ng/kg-day. These two values are roughly a factor of 4 lower and a factor of
3 larger, respectively, than the LOAEL estimate of 0.020 ng/kg-day that was the basis of the
standard pathway RfD. The TCDD intake of 0.0016 ng/kg-day corresponding to the alternative
NOAEL is slightly more than an order of magnitude lower than the standard pathway RfD
LOAEL POD and would yield a slightly lower RfD estimate than the current RfD after
eliminating the 10-fold UFl factor. EPA has much less confidence in the NOAEL estimate than
in the selected LOAEL because the NOAEL does not take into account the covariates and falls in
a lower concentration range where the background DLC exposures are a much more significant
component. The largest downward impact on the standard pathway LOAEL POD results from
grouping the highest exposures independent of the modeling results (POD = 0.005), which
decreases the LOAEL by a factor of four; however, analogous to the NOAEL alternative, the
approach ignores the contribution of covariates.
The largest upward impact on the standard pathway LOAEL POD is the inclusion of
modeled total TEQ (POD = 0.059), which increases the LOAEL by a factor of three. However,
the model fit is poor, and the result can be compared with an analogous calculation to the
additive DLC approach used for the Mocarelli analysis in Figure 4-6. An additive DLC-TEQ
background of 3.5 x 10 ng/kg-day can be estimated for the women in the Baccarelli analysis by
4-50
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
multiplying the TCDD background intake of 3.9 x 10 4 ng/kg-day by 9 (not shown in
Figure 4-7). Adding the estimated DLC background to the standard pathway RfD LOAEL POD
of 0.020 gives a corresponding total-TEQ intake of 0.023 ng/kg-day. This is 18% higher than
the standard pathway RfD POD but 2.6-fold lower than the modeled total-TEQ POD. Leaving
out the noncoplanar PCBs greatly improves the model fit, which could suggest that the
noncoplanar PCBs do not contribute to the effect as much as the PCDDs and PCDFs or that there
is greater uncertainty in the TEQ estimates for the noncoplanar PCBs. In either case, as for the
Mocarelli analysis, any estimate of background DLC exposure based on TEQ is likely an
over-estimate because of the conservative nature of TEFs. Overall, although background DLC
exposures will effectively increase the POD to some degree, EPA believes that the effect is
relatively small in the range of the estimated standard pathway TCDD LOAEL.
In summary, the quantitative uncertainties evaluated here for the RfD POD based on
Baccarelli et al. (2008) span a 3- to 4-fold range in either direction. The alternative LOAELs at
either extreme are not strong POD candidates; the lowest value (from the graphical method) does
not account for covariates and there is greater uncertainty in the (total TEQ) regression model for
the highest value than for the other regression models. All the other alternative LOAELs are
within a factor of 1.5 of the RfD POD. Overall, as for Mocarelli et al. (2008) analysis, EPA
believes that this sensitivity also supports the magnitude of the RfD.
4.5.1.2. NTP (2006a) Sensitivity Analysis
To examine the impacts of some of the uncertainties associated with estimating the POD
from the NTP (2006a) study (see Section 4.2), EPA analyzed two different approaches for
estimating dose and alternate choices of rodent kinetic model and background. Figure 4-8
depicts this analysis, which relied on an approach similar to those used in characterizing some of
the uncertainties in the RfDs derived from Mocarelli et al. (2008) and Baccarelli et al. (2008).
The lowest administered dose was determined to be the animal LOAEL based on liver and lung
lesions in the rats. In the standard pathway candidate RfD analysis, the LOAELhed was the
POD.
Exposures were estimated either based on a kinetic model of the administered TCDD
dose or on the measured concentrations of TCDD and DLCs in the rat adipose tissue after
terminal sacrifice. NTP reported concentrations of TCDD, 2,3,4,7,8-pentachlorodibenzofuran
4-51
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
(PeCDF), and 3,3N,4,4N,5-pentachlorobiphenyl (PCB-126) in the adipose and liver tissues
obtained from the rats after terminal sacrifice. The 2005 WHO TEF values for PeCDF and
PCB-126 are 0.3 and 0.1, respectively (Van den Berg et al.. 2006).
To predict average tissue concentrations based on the administered TCDD dose, EPA
used both the Emond and CADM kinetic models. EPA also used the first-order body burden
model to predict whole body TCDD concentrations; this model uses a constant half-life to
simulate the elimination of TCDD from the body. Section 3 describes all of these models.
EPA used several alternative dose metrics based on the modeling approach and measured
tissue concentrations. The first-order body burden model estimates the TCDD concentration in
the whole body. When using the Emond model to evaluate the disposition of TCDD, EPA
evaluated both whole-blood TCDD concentrations and LASC. For the CADM model, EPA
simulated TCDD concentrations in the adipose compartment following the administered TCDD
dose. EPA also used the TCDD (see Table 13 in the NTP report) or DLC concentrations (see
Tables 10 and 11 in the NTP (2006c) report) measured in the adipose tissue collected at study
termination.
Using the DLC concentration information, EPA estimated TEQ in two ways. In the first
approach, based on an analysis of DLCs in the adipose tissue that was reported in another NTP
study on DLC mixtures (NTP. 2006c). EPA initially estimated the ratio of the adipose tissue
TEQ concentration to the adipose tissue TCDD concentration, then applied this ratio to the
Emond whole-blood TCDD estimates assuming proportionality (resulting in a LOAEL whole
blood concentration of 2.75 ppt instead of the TCDD-only concentration of 2.56 ppt).
In the second approach, EPA estimated TEQ dose based on adipose tissue TCDD levels
reported by NTP; the reported TCDD concentration in the fat given in the study at the lowest
dose was used to estimate a LOAEL using the Emond model. Finally, using the 2005 WHO TEF
values (Van den Berg et al.. 2006). EPA converted the reported concentrations of TCDD,
PeCDF, and PCB-126 measured in the fat of the control rats in the NTP mixtures study (NTP.
2006c) to TEQ using eq. 4-1.
Chemical ( fat, ,,,) x TEF, ^ ,
Chemical, (B) = — -—- x Doserrnn (Eq. 4-1)
TCDD(fatTCDD)
4-52
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Where
Chemicali(B)
Chemicali(fatMc)
TCDD(fatxcDD)
DosexcDD
TEF;
= estimate of background exposure to Chemical z in ng/kg units of TCDD
blood concentrations at 105 weeks, for i = TCDD, PeCDF, and PCB126.
= mean pg/g of Chemical z in the fat tissues of the control animals at
105 weeks in mixtures study (NTP. 2006c).
= mean pg/g of TCDD in the fat tissues of the 3 ng/kg dose group at
105 weeks in the TCDD study (NTP. 2006a).
= 2.56 ng/kg TCDD blood concentration for the 3 ng/kg dose group in the
TCDD study (NTP. 2006a).
= Toxicity Equivalence Factor for Chemical z [from Van den berg et al.
(2006)1.
Assuming simple proportionality of blood TCDD concentrations between controls and
low-dose (2.14 ng/kg-day) animals, the TEF-adjusted ratio of each congener (Chemical z) in
control animal fat to low-dose-animal fat is multiplied by the modeled TCDD blood
concentration for the low-dose animals to obtain an equivalent background exposure in the dose
metric (ppt whole blood). For total TEQ, the estimates of all three congeners are summed. Total
TEQ estimates likely are biased somewhat high because they are based on terminal (2-year)
measurements rather than representing lifetime averages.
Overall, the alternative LOAEL estimates in this tree (see Figure 4-8) vary between 0.023
and 0.44 ng/kg-day. The LOAEL for the standard pathway RfD was estimated to be
0.14 ng/kg-day and is at the lower end of the range. The alternative LOAEL based on first order
body burden (0.023 ng/kg-day) is the lowest value in the range, approximately 85% lower than
the LOAEL based on the standard pathway approach. The difference between these
two estimates is consistent with the more conservative approach used in modeling first-order
TCDD body burdens. The alternative LOAEL based on the TEQ in whole blood is less than
10% greater than the LOAEL from the standard pathway RfD. The alternative candidate
LOAEL based on the TCDD in lipid-adjusted serum is approximately 120% greater than the
LOAEL for the standard pathway RfD. The use of the CADM model to estimate adipose tissue
concentration based on administered dose resulted in a 35% increase in the LOAEL estimate
relative to the LOAEL based on the standard pathway approach. The LOAELs based on
measured TCDD or TEQ levels in rodent adipose tissue were greater than the LOAEL from the
standard pathway RfD by approximately a factor of three.
4-53
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
4.5.2. Evaluation of Range of Alternative PODs for Additional Epidemiological
Endpoints
In addition to the principal studies depicted in Figures 4-6 and 4-7, EPA evaluated a
number of endpoints presented in seven other Seveso cohort studies to estimate the range of
potential PODs based on uncertainties in exposure duration, exposure averaging protocols, and
DLC background exposures. Included in those study/endpoint combinations are the following:
two that passed all the selection criteria, developmental dental effects (Alaluusua et al.. 2004)
and duration of menstrual period (Eskenazi et al.. 2002b); a new developmental study on semen
quality (Mocarelli et al.. 2011) that was published after the study selection process was
completed but is useful in this uncertainty analysis of the POD ranges; and four studies that did
not pass all the criteria for qualification as POD candidates (Warner et al.. 2007; Eskenazi et al..
2005; Warner et al., 2004; Mocarelli. 2000), but for which limiting NOAEL and LOAEL values
can be estimated. Descriptions and evaluations of each of these studies can be found in
Appendix C. Tables 4-8 through 4-10 and Figure 4-9 present the exposure values modeled using
the Emond human PBPK model for potential POD ranges for 7 additional endpoints studied in
23
the Seveso cohort. For most of the studies that did not pass all the criteria, the major
uncertainties are the definition of the critical exposure window and the corresponding relevant
exposure-averaging time, and the determination of adverse effect levels. Eskenazi et al. (2002b)
passed the selection criteria because a critical exposure window could be identified, but the
determination of an adverse effect level for length of menstrual cycle is somewhat arbitrary. A
critical exposure window can be identified also for Warner et al. (2004) (age at menarche), but
no TCDD-related adverse health outcomes were observed. However, with some additional
assumptions, NOAELs and LOAELs at nominal group-exposure levels can be determined for
each of these studies. The critical exposure window is assumed to be the entire duration from
exposure in 1976 to time of interview (i.e., end of follow-up period) when a critical window
cannot be identified. Tentative NOAELs and LOAELs are designated for those endpoints where
adversity levels are difficult to define. Given these assumptions, TCDD and total TEQ intakes
can be modeled but must be considered to be lower bounds on the effective exposures, given the
23 The details of the kinetic modeling for these endpoints and the corresponding background exposures can be found
in Appendix F.
4-54 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
conservative nature of the assumptions; EPA does not consider these estimates suitable for use in
the derivation of the TCDD RfD.
Additional endpoints reported in the epidemiologic literature were considered in the
context of this uncertainty analysis but were excluded based on large uncertainties in defining
24
adversity or plausible exposure profiles over time. All the Ranch Hand studies were excluded
because of the inability to construct effective exposure profiles with any confidence, given the
20-year lag between the actual TCDD exposures and measurement of serum levels. For the
25
Seveso cohort, several studies were eliminated from consideration because uncertainties in
defining plausible NOAELs or LOAELs were too large.
For modeling of the endpoints in Tables 4-8 to 4-10, grouped exposure ranges were
represented by the geometric mean of the range limits. The average daily intakes for exposures
(LASC) in the background range were estimated as the continuous exposure from birth resulting
in the reported serum concentrations (TCDD or total TEQ) at the average subject age at time of
measurement. Peak and critical-window average exposures (as LASC) were modeled for
measured LASC values greater than background using the actual exposure scenarios. Because
all exposure durations were less than lifetime, average daily intakes for all modeled peak and
window-average LASC were estimated using the terminal 5-year-peak average as described in
Section 3.3.6. Precision is expressed to the nearest 10"5 ng/kg-day for all intake estimates to
"3
avoid rounding errors when adding DLC background intakes. Values less than or equal to 10"
are shown in scientific notation for readability.
Figure 4-9 shows the range of NOAELs and LOAELs and exposures for all of the
endpoints considered in this uncertainty analysis, the endpoints on which they are based, and the
study citation. The study/endpoint combinations are separated into two groups representing
either those chosen for RfD POD consideration ("Candidate RfD") or those not otherwise
qualifying ("Uncertainty Analysis Only"). The NOAELS and LOAELS are indicated for each
study, as appropriate, and the vertical lines through these PODs represent the range of possible
PODs based on Emond PBPK results using alternative exposure scenarios. The
limits—indicated by symbols of the same type—for each POD type (NOAEL or LOAEL) for
24 (Michalek and Pavuk. 2008: Pavuk et al.. 2003: Michalek et al.. 2001a: Michalek et al.. 2001b: Michalek et al..
2001c: Longnecker and Michalek. 20001
25 (Eskenazi et al.. 2007: Baccarelli et al.. 2005: Baccarelli et al.. 2004: Eskenazi et al.. 2003: Landi et al.. 2003:
Baccarelli et al.. 2002: Eskenazi et al.. 2002a)
4-55 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
each endpoint cover the full range of alternative PODs in Tables 4-8 to 4-10, without distinction
of the relative plausibility of each one. That is, all the PODs are treated equally without
considering the relative confidence held in each one, individually. The low end of most of the
ranges is the critical-window average exposure, which does not take into account the influence of
the much higher peak exposure. Conversely, the upper end of the range is generally the peak
exposure, which does not account for the potential effect of longer-term continuous exposure.
On the "uncertainty analysis only" side of Figure 4-9, most of the NOAELs and many of the
LOAELs are somewhat speculative and would not be considered as strong candidates for the
RfD POD. The range limits are themselves uncertain, as constraints were applied to the lower
and upper limits to keep them in the range of the data. The same DLC modeling issues presented
in Section 4.5.1 apply to all the TEQ results here, so the TEQ results are approximations and are
unlikely to be very accurate. Also, the lowest POD estimates are more affected by background
DLC exposure than are the PODs closer to the RfD POD; generally, TCDD is a minor
component of the total TEQ for the lower PODs, subjecting the lowest alternative PODs to the
greatest uncertainty. The RfD LOAEL POD (0.02 ng/kg-day) and its equivalent NOAEL
estimate (0.002 ng/kg-day, with the 10-fold UF), along with the RfD (7 x 10 4 ng/kg-day), are
shown on the figure for comparison to the alternative POD ranges.
The LOAEL ranges for the two principal studies (Baccarelli et al.. 2008; Mocarelli et al..
2008) span the RfD LOAEL POD, whether based on TCDD alone or total TEQ. The single
NOAEL estimate for Baccarelli et al. (2008) is only slightly below the equivalent RfD NOAEL
POD. The NOAEL and the lowest alternative LOAELs for Baccarelli et al. (2008) are not strong
POD candidates because they are based on the raw observations and do not take into account the
covariates that affect the exposure-response relationship, as does the regression model on which
the RfD LOAEL POD is based. In general, background DLC exposure has a small impact on the
LOAEL PODs for the co-principal studies, raising the effective exposure level by 15% for the
Mocarelli et al. (2008) RfD LOAEL POD and yielding essentially the same value for the
Baccarelli et al. (2008) RfD LOAEL POD, if noncoplanar PCBs are excluded (see Figure 4-7).
Including the noncoplanar PCBs from the Baccarelli et al. (2008) regression modeling results has
a much bigger impact, raising the LOAEL by a factor of 3; however, the significance of the
modeled slope is relatively poor (p = 0.14), so EPA does not place much biological significance
on the outcome. The POD ranges for the other candidate RfD endpoints are well above their
4-56
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
respective comparison NOAEL/LOAEL benchmarks. The NOAEL for Eskenazi et al. (2002b) is
somewhat arbitrary, based simply on a continuous average exposure over a 13-year window
corresponding to a normal 28-day menstrual cycle, without considering the possible range of
normal durations.
Of the endpoints that were not selected as RfD POD candidates, there are three whose
LOAEL ranges are wholly or mostly below the RfD LOAEL POD. The sperm effects in men
who were exposed in utero and by lactation reported by Mocarelli et al. (2011) are very similar
to those in men exposed as boys in one of the principal studies (Mocarelli et al.. 2008). The
maternal exposures associated with the effects reported by Mocarelli et al. (2011) are very low
with the TCDD-only LOAEL being 12-fold lower than the RfD LOAEL POD for the 30-year
exposure scenario. For this study, a TCDD-only NOAEL can be established at 2.9 x 10"4 ng/kg-
day (for the reference population), which is sevenfold below the equivalent RfD NOAEL POD.
Both the TCDD-only NOAEL and LOAEL are much lower than the estimated DLC background
exposure; however, assuming a simple TEQ additive model, and with the aforementioned
uncertainties concerning DLC-TEQ estimation, a TEQ NOAEL and LOAEL of 2.9 x 10 3 and
5.4 x 10 3 ng/kg-day can be estimated (Table 4-8). Although the TEQ LOAEL is still well below
that for the RfD POD, the TEQ NOAEL is in the range of the equivalent RfD NOAEL POD.
Given the large amount of uncertainty in the modeled NOAEL and LOAEL for this endpoint,
EPA elected not to consider either as a POD.
The second endpoint with lower LOAELs than the RfD POD is age at menopause
reported by Eskenazi et al. (2005). The figure for this endpoint includes two separate LOAEL
candidates because of uncertainty in determining adversity at the lower exposure level in
rd
question (3 quintile). For that reason, the daily intakes associated with the critical-window
average and peak exposures are labeled ("W" and "P," respectively). The intakes associated
with the peak are in the range of the RfD LOAEL benchmark, while the window-average TCDD
intakes are closer to the NOAEL benchmark. Considering background DLC intake, the
window-average TEQ intakes are considerably higher, the DLC exposures being larger than the
TCDD intakes, themselves, but still below the LOAEL benchmark. The range of the TEQ P/W
average of 0.01-0.031 ng/kg-day (see Table 4-10), however, straddles the RfD LOAEL
benchmark of 0.02 ng/kg-day. Uncertainty in the NOAEL is similar to that for the LOAEL,
depending on whether the 1st or 2nd quintile can be called a NOAEL. Although the response in
4-57
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
the 2nd quintile is not significant compared to the 1st quintile, the NOAEL determination is
complicated by the lack of an absolute measure of "normal." EPA considered the quantitative
and qualitative uncertainties to be too large to consider this endpoint as an RfD POD candidate.
The NOAELs and LOAELs for altered sex ratio reported by Mocarelli et al. (2000) span
their respective RfD POD benchmarks and are above the benchmarks when considering the
peak/window exposure averages or background DLC exposures. The uncertainties for lack of an
identifiable critical exposure window also apply to this endpoint. The other two endpoints, age
at menarche (Warner et al.. 2004) and ovarian function (Warner et al.. 2007). are unbounded
NOAELs at the highest exposures. The ovarian function endpoint also is uncertain for lack of an
identifiable critical exposure window.
Additional uncertainties not covered explicitly in this analysis include exposure to other
AhR agonists, either naturally occurring in food-stuffs (Connor et al.. 2008) or by-products of
combustion or manufacturing processes (e.g., poly-aromatic hydrocarbons), and choice of
uncertainty factor. As a final note on background DLC exposure, the background DLC intake
estimates for the standard scenario (Needham) used in this assessment are somewhat crude, in
that they are simple multiples of modeled TCDD intake based on an approximation of the
proportion of TCDD to total TEQ. TCDD exposures are modeled over durations of up to
35 years (1941-1976) using a single fixed background intake term (a model limitation).
However, background TCDD/TEQ exposures are thought to have varied widely over that time
period, increasing gradually in the United States from the early 20th century to a peak in 1965,
then decreasing rapidly to near current levels in the early 1980s (Lorber. 2002). Based on a
digitization of Figure 6 in Lorber (2002). depicting the estimated TEQ intake over the course of
the 20th century, a time-weighted average total TEQ intake for the period 1941-1976 of
4.6 x 10"3 ng/kg-day can be estimated. Adjusting the TEFgg-based Lorber (2002) TEQ intakes to
TEF05-based values, assuming a 10% TCDD fraction and using the 0.7 TEF05:TEF98 factor
described previously (see Section 4.5.1), yields a DLC-TEQ intake estimate of
3.4 x 10" ng/kg-day for that time period, which is similar to the estimated DLC background
intake of 3.33 x 10" ng/kg-day for the standard scenario using the simple scaling model.
However, the DLC intake estimate based on Lorber (2002) is somewhat of an
underestimate because it does not include dioxin-like PCBs. Pinsky and Lorber (1998) estimated
a TCDD intake of 4 x 10"4 ng/kg-day for the U.S. population in the 1970s, which is almost the
4-58
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
same as the modeled TCDD background intake for the Seveso population. However, there is no
information on comparative environmental exposures for the United States and Italy during this
period, and TCDD exposures before 1970 for these populations were not necessarily the same,
on average. Higher TCDD background exposures have been estimated by others. Pinsky and
Lorber (1998) estimated an average TCDD-only intake of 1.4 x 10"3 to 1.9 x 10"3 ng/kg-day for
the U.S. population in the late 1960s and early 1970s using a lst-order kinetics model with a
variable intake term and a TCDD half-life of 7.1 years. Aylward and Hays (2002) estimated a
"3
TCDD intake of at least 1.3 x 10" ng/kg-day for the United States, Canada, Germany, and
France prior to 1972 using a lst-order kinetics model assuming a TCDD half-life of 7.5 years.
These estimates are 3.5-5 times higher than the background TCDD intake estimated by EPA
using the Emond PBPK model for this assessment. Total TEQ background would increase
proportionally. However, none of these estimates, including EPA's, is based on actual intake
measurements and are all dependent on modeling assumptions. Raising the background DLC
exposure would obviously increase the effective PODs. However, increasing the background
TCDD intake for modeling purposes would decrease the contribution of the actual TCDD
exposures experienced by the Seveso population in 1976, resulting in a lower TCDD POD, as
can be seen in the Eskenazi background scenario for Mocarelli et al. (2008) (see Figure 4-6).
The overall result would be a slightly higher POD (ca., 0.032 ng/kg-day) based on TEQ.
This analysis highlights several important research needs. While the disposition of
TCDD following high exposures is reasonably understood and simulated in current models, the
current scientific understanding of disposition following TCDD exposures that are closer to
current background dietary intakes, likely the primary source of TCDD exposure for most of the
U.S. population, is not understood as well at present. This uncertainty affects the estimation of
TCDD intake rates corresponding to the lower blood TCDD levels associated with LOAELs and
NOAELs. The disposition of DLCs following exposures at background levels is similarly not
well understood. Furthermore, there is uncertainty in the relationship of DLC tissue
concentrations to oral intakes in the current TEF approach. Finally, there is toxicological
uncertainty regarding several of the endpoints. Additional studies corroborating these outcomes
and their toxicological significance would further increase their utility in refining the TCDD
RfD.
4-59
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Overall, EPA believes that the results of this analysis of alternative endpoints and PODs
increase the confidence in the TCDD RfD, both qualitatively and quantitatively. EPA's analyses
of some studies show POD estimates higher than the RfD PODs—primarily those analyses that
consider background DLCs. Other analyses show POD estimates lower than the RfD POD, such
as the use of alternative age-adjusted background TCDD/DLC intake rates and some evaluations
of more uncertain endpoints (e.g., age at menopause endpoint in Eskanazi et al. (2005). The
more extreme values on the lower end are also the most uncertain, particularly with respect to the
contribution of TCDD relative to total TEQ. In addition, except for the male reproductive effects
in Mocarelli et al. (2011). determination of adversity for the lower LOAELs is problematic,
leading to lower confidence in the PODs. The TCDD and TEQ LOAELs for semen quality in
males exposed in utero and by lactation (Mocarelli et al.. 2011) are much lower than the
corresponding LOAELs for males exposed between ages 1 and 10 years (Mocarelli et al.. 2008).
However, the NOAEL established for in utero and lactational exposure is fairly strong in the
qualitative sense; that is, there is fairly clear indication that semen quality is unaffected at the
corresponding dioxin exposure level. Quantitatively, there is more uncertainty, but considering
background DLC exposure, the NOAEL is close to the RfD NOAEL benchmark.
4-60
DRAFT - DO NOT CITE OR QUOTE
-------
1 Table 4-1. PODs for epidemiologic studies of TCDD
2
Study
POD (ng/kg-day)
Critical effects
Alaluusua et al. (2004)
0.04 063 (NOAEL)
Dental effects in adults exposed to TCDD in childhood
Baccarelli et al. (2008)
0.0199b (LOAEL)
Elevated TSH in neonates
Mocarelli et al. (2008)
0.020 lc(LOAEL)
Decreased sperm count and motility in men exposed to
TCDD in childhood
3
4 aMean of peak exposure (0.0655 ng/kg-day) and average exposure over 10-year critical window (0.0156 ng/kg-day).
5 bMaternal exposure corresponding to neonatal TSH concentration exceeding 5 |iU/mL.
6 °Mean of peak exposure (0.032 ng/kg-day) and average exposure over 10-year critical window (0.0080 ng/kg-day).
7
8
9 Table 4-2. Models run for each study/endpoint combination in the animal
10 bioassay benchmark dose modeling
11
Model
Restrictions imposed
Continuous models
Exponential M2-M5,
not grouped
Adverse direction specified according to the response data; power >1
Hill
Adverse direction is automatic; n> 1
Linear
Adverse direction is automatic; degree of polynomial = 1
Polynomial
Adverse direction is automatic; degree of polynomial unrestricted; restrict the
sign of the power to nonnegative or nonpositive, depending on the direction of
the responses
Power
Adverse direction is automatic; power >1
Dichotomous models
Gamma
Power > 1
Logistic
None
Log-Logistic
Slope >1
Log-Probit
None
Multistage
Beta >0, 2nd degree polynomial
Probit
None
Weibull
Power > 1
12
4-61
DRAFT - DO NOT CITE OR QUOTE
-------
Table 4-3. Summary of key animal study PODs (ng/kg-day) based on three different dose metrics: administered
dose, first-order body burden HED, and blood concentration
On
to
o
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Study
Endpoint
Administered dose"
lst-order body burden HEDb
Blood concentration HEDC
NOAEL
LOAEL
BMDLd
NOAEL
LOAEL
BMDLd
NOAEL
LOAEL
BMDLd
Amin et al. (2000)
Saccharin preference ratio,
female
—
2.50E+01
e
—
2.49E-02
e
—
1.71E-01
e
Bell et al. (2007b)
Balano-preputial separation
in male pups
—
2.40E+00
2.87E+00
—
1.26E-02
1.50E-02
—
8.85E-02
4.34E-0
2
Bowman et
al. ("1989a: 1989b):
Schantz and
Bowman (1989):
Schantz et al. (1986):
Schantz et al. (1992)
Neurobehavioral effects
1.20E-01
8.22E-03
Cantoni et al. (1981)
Urinary coproporhyrins
-
1.43E+00
e
-
1.24E-02
e
-
6.37E-02
e
Chu et al. (200 D
Tissue-weight changes
2.50E+02
1.00E+03
-
7.55E-01
3.02E+00
-
7.03E+00
2.96E+01
-
Chu et al. (2007)
Liver lesions
2.50E+00
2.50E+01
-
7.55E-03
7.55E-02
-
3.49E-02
5.63E-01
-
Crofton et al. (2005)
Serum T4
3.00E+01
1.00E+02
e
1.92E-02
6.40E-02
e
1.69E-01
7.43E-01
e
Croutch et al. (2005)
Decreased body weight
5.43E+01
2.17E+02
-
2.22E-01
8.89E-01
-
7.81E-01
3.57E+00
-
DeCaprio et al.
(1986)
Decreased body weight,
organ-weight changes
6.10E-01
4.90E+00
—
4.11E-03
3.30E-02
—
—
—
—
Fattore et al. (2000)
Decreased hepatic retinol
-
2.00E+01
-
-
1.23E-01
-
-
7.82E-01
-
Fox et al. (1993)
Increased liver weight
5.70E-01
3.27E+02
-
1.42E-03
8.12E-01
-
8.08E-04
3.05E+00
-
-------
Table 4-3. Summary of key animal study PODs (ng/kg-day) based on three different dose metrics: administered
dose, lst-order body burden HED and blood concentration HED (continued)
On
o
o
o
2
o
H
O
HH
H
W
O
V
o
c
o
H
W
Study
Endpoint
Administered dose"
lst-order body burden HEDb
Blood concentration HEDC
NOAEL
LOAEL
BMDLd
NOAEL
LOAEL
BMDLd
NOAEL
LOAEL
BMDLd
Franc etal. (2001)
Organ-weight changes
1.00E+01
3.00E+01
1.34E+01
6.62E-02
1.99E-01
8.87E-02
4.49E-01
1.41E+00
2.61E-01
Franczak et al. (2006)
Abnormal estrous cycle
-
7.14E+00
-
-
5.95E-02
-
-
3.18E-01
-
Hoio et al. (2002)f
DRL response per min
-
2.00E+01
e
-
5.26E-03
e
-
5.51E-02
e
Hochstein et al.
(200 ng
Kit mortality at 6 wk
—
2.65E+00
—
—
—
—
—
—
—
Hutt et al. (2008)
Embyrotoxicity
-
7.14E+00
-
-
4.67E-02
-
-
2.52E-01
-
Ikeda et al. ^2005^
Sex ratio
-
1.65E+01
-
-
1.05E-01
-
-
2.75E+00
-
Ishihara et al. (2007)
Sex ratio
1.00E-01
1.00E+02
-
3.18E-04
3.18E-01
-
4.91E-05
4.96E-01
-
Kattainen et al.
(2001)
3rd molar length
-
3.00E+01
e
-
7.89E-03
e
-
9.01E-02
e
Keller et al. (2008a:
2008b:2007)
Missing mandibular molars
-
1.00E+01
e
-
2.58E-03
e
-
9.88E-03
e
Kociba et al. (1976)
Liver and hematologic
effects and body-weight
changes
7.14E+00
7.14E+01
4.53E-02
4.53E-01
2.62E-01
3.03E+00
Kociba et al. (1978)
Liver and lung lesions,
increased urinary
porphyrins
1.00E+00
1.00E+01
e
1.07E-02
1.07E-01
e
6.33E-02
6.34E-01
e
Kuchiiwa et al.
(2002)
Immunoreactive neurons
-
7.00E-01
-
-
3.11E-03
-
-
2.75E-03
e
Latchoumycandane
and Mathur (2002)h
Sperm production
-
1.00E+00
e
-
3.87E-03
e
-
1.62E-02
e
Li etal. (1997)
Increased serum FSH
3.00E+00
1.00E+01
e
7.89E-04
2.63E-03
e
2.90E-03
1.67E-02
e
Li et al. (2006)
Hormone levels (serum
estradiol)
—
2.00E+00
e
—
9.85E-04
e
—
1.58E-03
e
-------
Table 4-3. Summary of key animal study PODs (ng/kg-day) based on three different dose metrics: administered
dose, lst-order body burden HED and blood concentration HED (continued)
On
o
o
o
2
o
H
o
HH
H
W
o
o
c
o
H
W
Study
Endpoint
Administered dose"
lst-order body burden HEDb
Blood concentration HEDC
NOAEL
LOAEL
BMDLd
NOAEL
LOAEL
BMDLd
NOAEL
LOAEL
BMDLd
Markowski et al.
(2001)
FR2 revolutions
—
2.00E+01
e
—
6.25E-03
e
—
5.15E-02
e
Maronpot et al.
(1993)
Increased relative liver
weight
1.07E+01
3.50E+01
-
8.97E-02
2.93E-01
-
5.03E-01
1.71E+00
-
Miettinen et al.
(2006)
Cariogenic lesions in pups
-
3.00E+01
e
-
7.89E-03
e
-
8.95E-02
e
Murray etal. (1979)
Fertility index in F2
generation
1.00E+00
1.00E+01
e
9.43E-03
9.43E-02
e
2.89E-02
3.79E-01
e
NTP (1982b)
Liver lesions
-
1.39E+00
e
-
6.47E-03
e
-
2.16E-02
e
NTP (2006a)
Liver and lung lesions
-
2.14E+00
e
-
2.34E-02
e
-
1.36E-01
e
Nohara et al. (2000)
Decreased spleen
cellularity
8.00E+02
-
-
2.10E-01
-
-
5.34E+00
-
-
Nohara et al. (2002)
Mortality from influenza
virus-A challenge
5.00E+02
-
-
1.29E-01
-
-
1.37E+00
-
-
Ohsako et al. (2001)
Anogenital distance in
pups
1.25E+01
5.00E+01
e
3.29E-03
1.32E-02
e
2.74E-02
1.78E-01
e
Schantz et al. (1996)
Maze errors
-
2.50E+01
e
-
e
4.55E-02
-
1.71E-01
e
Seo et al. (1995)
Decreased thymus weight
2.50E+01
1.00E+02
-
2.49E-02
9.96E-02
-
1.67E-01
9.15E-01
-
Sewall et al. (1995)
Serum T4
1.07E+01
3.50E+01
5.16E+00
8.97E-02
2.93E-01
4.33E-02
5.03E-01
1.71E+00
1.80E-01
Shi et al. (2007)
Serum estradiol in female
pups
1.43E-01
7.14E-01
2.24E-01
1.23E-03
6.13E-03
1.92E-03
4.47E-03
2.69E-02
4.74E-03
-------
Table 4-3. Summary of key animal study PODs (ng/kg-day) based on three different dose metrics: administered
dose, lst-order body burden HED and blood concentration HED (continued)
On
o
o
o
2
o
H
o
HH
H
W
o
o
c
o
H
W
Study
Endpoint
Administered dose"
lst-order body burden HEDb
Blood concentration HEDC
NOAEL
LOAEL
BMDLd
NOAEL
LOAEL
BMDLd
NOAEL
LOAEL
BMDLd
Simanainen et al.
(2002)
Decreased serum T4
1.00E+02
3.00E+02
—
2.63E-02
7.89E-02
—
4.26E-01
1.67E+00
—
Simanainen et al.
(2003)
Decreased thymus weight
and change in EROD
activity
1.00E+02
3.00E+02
2.63E-02
7.89E-02
4.26E-01
1.67E+00
Simanainen et al.
C2004aN)
Decreased daily sperm
production
1.00E+02
3.00E+02
-
2.63E-02
7.89E-02
-
4.26E-01
1.67E+00
-
Smialowicz et al.
(2004)
Decreased antibody
response to SRBCs
3.00E+02
1.00E+03
-
7.73E-02
2.58E-01
-
7.23E-01
3.28E+00
-
Smialowicz et al.
(2008)
PFC per 10A6 cells
-
1.07E+00
e
-
5.00E-03
e
-
6.26E-03
e
Smith etal. (1976)
Cleft palate in pups
1.00E+02
1.00E+03
1.84E+02
1.59E-01
1.59E+00
2.93E-01
5.24E-01
7.61E+00
9.46E-01
Sparschu et al.
(1971)
Decreased fetal body
weight
3.00E+01
1.25E+02
e
5.45E-02
2.27E-01
-
3.18E-01
1.73E+00
e
Toth et al. (19791
Skin lesions
-
1.00E+00
e
-
3.70E-03
e
-
9.91E-03
e
VanBirgelen et al.
ri995aV
Decreased liver retinyl
palmitate
-
1.35E+01
e
-
8.32E-02
e
-
5.14E-01
e
Vosetal. (1973)
Decreased delayed-type
hypersensitivity response
to tuberculin
1.14E+00
5.71E+00
6.43E-03
3.22E-02
Weber etal. M995^
Increased liver weight
1.00E+03
3.00E+03
-
3.51E-01
1.05E+00
-
3.27E+00
1.18E+01
-
White etal. (1986)
Decreased serum
complement
-
1.00E+01
e
-
2.23E-02
e
-
2.77E-02
e
Yans et al. (2000)
Increased endometrial
implant survival
1.79E+01
—
—
6.74E-01
—
—
—
—
—
-------
Table 4-3. Summary of key animal study PODs (ng/kg-day) based on three different dose metrics: administered
dose, lst-order body burden HED and blood concentration HED (continued)
aAverage administered daily dose over the experimental exposure period.
bHED based on lst-orderbody burden model described in Section 3.2.4.4.
°HED based on Emond rodent and human PBPK models described in Section 3.3.6.
dBMR = 0.1 for quantal endpoints and 1 standard deviation control mean for continuous endpoints, except for body and organ weights, where BMR = 10%
relative deviation from control mean.
eBMD modeling unsuccessful (see Table 4-4 and Appendix G for details).
fZareba et al. (20021 is considered to be the same study but report effects at doses above the LOAEL that are not considered further; this study is not carried
forward for determination of an RfD POD but is included in the RfD uncertainty analysis presented in Section 4.4.
8Hochstein et al. (20011 is not carried forward because of the lack of toxicokinetic information for estimation of an HED.
hLatchoumycandane et al. (2002a: 2002b) are considered to be the same study but report effects (not toxicologically relevant) at doses above the LOAEL that
not considered further; these two studies are not carried forward.
'Van Birgelen et al. (1995b) is considered to be the same study but reports effects at doses above the LOAEL that are not considered further; this study in not
carried forward for determination of an RfD POD but is included in the RfD uncertainty analysis presented in Section 4.4.
- value not established or not modeled.
-------
Table 4-4. TCDD BMDL analysis (NOAEL, LOAEL, BMD, and BMDL values given as animal whole blood
concentrations in ng/kg)a
On
-J
o
o
o
2
o
H
o
HH
H
W
o
o
c
o
H
W
Studyb'c
NOAEL/
LOAEL
Endpoint
Control
response
First
response"1
Max
response6
Model fit detail
BMD/
BMDL
Comments
Amin et al.
(2000)
(rat)
3.38E+00
Saccharin
consumed, female,
(0.25%) (n = 10)
22% |
(0.3 SD)
66% |
Continuous linear,
modeled variance
(p =0.55)
9.15E+00
6.09E+00
BMDL > LOAEL; restricted
power model, constrained
parameter hit lower bound
Continuous power,
modeled variance,
unrestricted
(p = NA)
8.37E+00
3.42E+00
Saturated model; supralinear fit
(power = 0.74)
Saccharin
consumed, female
(0.50%) (n = 10)
49% |
(0.7 SD)
80% |
Continuous linear,
modeled variance
(p = 0.06)
1.02E+01
6.57E+00
Restricted power model,
constrained parameter hit lower
bound
Continuous power,
modeled variance,
unrestricted
(p = NA)
6.57E+00
1.15E+00
Saturated model; supralinear fit
(power = 0.40)
Saccharin preference
ratio, female
(0.25%)
(n = 10)
29% |
(1.8 SD)
33% |
Continuous linear,
modeled variance
(p = 0.002)
1.16E+01
5.57E+00
BMDL > LOAEL; no response
near BMR; near maximal
response at LOAEL
Saccharin preference
ratio, female
(0.50%)
(n = 10)
39% |
(1.1 SD)
54% |
Continuous linear,
constant variance
(p = 0.14)
8.14E+00
5.11E+00
BMDL > LOAEL; near maximal
response at LOAEL; restricted
power model, constrained
parameter hit lower bound
Continuous power,
constant variance,
unrestricted
(p = NA)
2.60E+00
1.06E-14
Saturated model; supralinear fit
(power = 0.28)
-------
Table 4-4. TCDD BMDL analysis (NOAEL, LOAEL, BMD, and BMDL values given as animal whole blood
concentrations in ng/kg)a (continued)
On
00
o
o
o
2
o
H
o
HH
H
W
O
o
c
o
H
W
Studyb'c
NOAEL/
LOAEL
Endpoint
Control
response
First
response"1
Max
response6
Model fit detail
BMD/
BMDL
Comments
Bell et al.
(2007b)
(rat)
2.20E+00
Balano-preputial
separation in male
pups
(n = 30 [dams])
1/30
5/30
15/30
Dichotomous log-
logistic, restricted
(p = 0.78)
2.25E+00
1.39E+00
Adequate fit; constrained
parameter bound hit; not litter
based; selected
Dichotomous log-
logistic, unrestricted
(p = 0.50)
2.00E+00
2.80E-01
Supralinear fit
(slope = 0.93); selected
Cantoni et al.
(1981)
(rat)
1.85E+00
Urinary uroporhyrins
(« = 4)
2.4-fold t
(5.7 SD)
87-fold t
Continuous
exponential (M2),
modeled variance
(p = 0.0003)
3.76E+00
2.76E+00
No response near BMR; poor fits
for all modeled variance models;
constant variance poor
representation of control SD;
BMDL > LOAEL
Urinary
coproporhyrins
(n = 4)
2.4-fold t
(3.1 SD)
4.0-fold t
Continuous
exponential (M4),
modeled variance
(p = 0.49)
5.34E-01
1.80E-01
No response near BMR
Continuous power,
modeled variance,
unrestricted
(p =0.61)
2.77E-02
2.03E-05
Supralinear fit (n = 0.30); poor
model choice for plateau effect
Crofton et al.
(2005)
(rat)
3.46E+00
9.26E+00
Serum T4,
(n = 4-14)
29% |
(1.9 SD)
51% |
Continuous
exponential (M4),
constant variance
(p = 0.94)
5.19E+00
3.03E+00
No response near BMR
-------
Table 4-4. TCDD BMDL analysis (NOAEL, LOAEL, BMD, and BMDL values given as animal whole blood
concentrations in ng/kg)a (continued)
On
VO
O
o
o
2
o
H
o
HH
H
W
O
o
c
o
H
W
Studyb'c
NOAEL/
LOAEL
Endpoint
Control
response
First
response"1
Max
response6
Model fit detail
BMD/
BMDL
Comments
Franc et al.
(2001)
(rat)
6.59E+00
1.45E+01
S-D Rats, Relative
Liver Weight
8.1% t
(0.58 SD)
55% t
Continuous power,
constant variance
(p = 0.84)
9.47E+00
4.59E+00
Acceptable fit; selected
L-E Rats, Relative
Liver Weight
6.3% t
(0.63 SD)
22% t
Continuous Hill,
modeled variance,
restricted
(p = 0.83)
7.72E+00
1.22E+00
Constrained parameter hit lower
bound; poor fit for variance model
Continuous Hill,
modeled variance,
unrestricted
(p = N/A)
7.22E+00
1.15E+00
Supralinear fit (power = 0.55)
S-D Rats, Relative
Thymus Weight
9.0% |
(0.11 SD)
77% |
Continuous
exponential (M4),
modeled variance
(p = 0.72)
1.88E+00
9.22E-01
Poor fit for responses in controls
and lowest exposure group
Continuous
polynomial, modeled
variance
(p = 0.40)
4.78E+00
3.89E+00
No response near BMR;
otherwise acceptable fit
L-E Rats, Relative
Thymus Weight
7.7% |
(0.15 SD)
66% |
Continuous
exponential (M4),
constant variance
(p = 0.23)
2.08E+00
5.93E-01
Poor fit for responses in controls
and lowest exposure group;
dose-response relationship not
significant
H/W Rats, Relative
Thymus Weight
3.7% |
(0.10 SD)
51% |
Continuous
exponential (M2),
constant variance
(p = 0.70)
5.09E+00
3.13E+00
No response near BMR;
otherwise acceptable fit
-------
Table 4-4. TCDD BMDL analysis (NOAEL, LOAEL, BMD, and BMDL values given as animal whole blood
concentrations in ng/kg)a (continued)
-j
o
O
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Studyb'c
NOAEL/
LOAEL
Endpoint
Control
response
First
response"1
Max
response6
Model fit detail
BMD/
BMDL
Comments
Hojo et al.
(2002)
(rat)
1.62E+00
DRL reinforce per
min
(n = 12)
55% t
(1.0 SD)
80% t
Continuous
exponential (M4),
constant variance
(p = 0.054)
1.32E+00
2.37E-03
Poor fit; near maximal response at
lowest dose, BMD/BMDL ratio
>100
DRL response per
min
(n = 12)
105% 4
(2.4 SD)
105% |
Continuous
exponential (M4),
constant variance
(p = 0.48)
3.81E-01
1.55E-02
No response data near BMR;
maximal response at lowest dose,
BMD/BMDL ratio »20
Kattainen et al.
(2001)
(rat)
2.23E+00
3rd molar length in
pups
(n = 4-8)
15% 4
(4.2 SD)
27% |
Continuous Hill,
modeled variance,
restricted
(p = 0.02)
3.13E-01
1.68E-01
No response data near BMR;
Constrained parameter lower
bound hit
Continuous Hill,
modeled variance,
unrestricted
{p< 0.001)
1.21E-02
BMDL could not be calculated
3rd molar eruption in
pups
(n = 4-8)
1/16
3/17
13/19
Dichotomous log-
logistic, restricted
(p = 0.98)
2.40E+00
1.33E+00
Constrained parameter lower
bound hit
Dichotomous log-
logistic, unrestricted
(p = 0.95)
1.93E+00
1.84E-01
Supralinear fit (slope = 0.91)
Keller et al.
(2008a: 2008b:
2007)
(mouse)
5.37E-01
Missing molars
(n = 23-36)
0/29
2/23
30/30
Dichotomous 1°
multistage
(p = 0.26)
1.09E+00
7.62E-01
Poor fit at first response level; not
most sensitive endpoint; other
endpoints not amenable to BMD
modeling
-------
Table 4-4. TCDD BMDL analysis (NOAEL, LOAEL, BMD, and BMDL values given as animal whole blood
concentrations in ng/kg)a (continued)
O
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Studyb'c
NOAEL/
LOAEL
Endpoint
Control
response
First
response"1
Max
response6
Model fit detail
BMD/
BMDL
Comments
Kociba et al.
(1978)
(rat)
1.55E+00
7.15E+00
Uroporphyrin per
creatinine, females
(« = 5)
15% t
(0.48 SD)
89% t
Continuous linear,
constant variance
(p = 0.79)
1.31E+01
9.29E+00
BMDL > LOAEL; otherwise
adequate fit
Urinary
coproporphyria,
females
(n = 5)
67% t
(5.1 SD)
78% t
Continuous
exponential (M4),
modeled variance
(p = 0.01)
1.57E+00
7.18E-01
Poor fit; no response near BMR
Liver lesions
(n = 50)
No data presented
Lung lesions
(n = 50)
No data presented
Kuchiiwa et al.
(2002) (mouse)
1.42E+02
Immunoreactive
Neurons in Dorsalis,
males
(n = 6)
42% |
(3.5 SD)
64% |
Continuous linear,
constant variance
(p = NA, insufficient
degrees of freedom)
6.04E-02
4.27E-02
No response near BMR
Immunoreactive
Neurons in
Medianus, males
(n = 6)
63% |
(4.8 SD)
75% |
Continuous linear,
modeled variance
(p = NA, insufficient
degrees of freedom)
4.93E-02
3.23E-02
No response near BMR
Immunoreactive
Neurons inB9,
males
(n = 6)
69% |
(6.6 SD)
87% |
Continuous linear,
constant variance
(p = NA, insufficient
degrees of freedom)
4.17E-02
3.01E-02
No response near BMR
Immunoreactive
Neurons in Magnus,
males
(n = 6)
55% |
(7.0 SD)
75% |
Continuous linear,
modeled variance
(p = NA, insufficient
degrees of freedom)
3.35E-02
2.05E-02
No response near BMR
-------
Table 4-4. TCDD BMDL analysis (NOAEL, LOAEL, BMD, and BMDL values given as animal whole blood
concentrations in ng/kg)a (continued)
to
O
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
ffl
Studyb'c
NOAEL/
LOAEL
Endpoint
Control
response
First
response"1
Max
response6
Model fit detail
BMD/
BMDL
Comments
Latchoumy-
candane and
Mathur (2002)
(rat)
7.85E-01
Daily sperm
production
(n = 6)
29% |
(1.0 SD)
41% |
Continuous Hill,
constant variance,
restricted
(p = 0.96)
1.17E-01
1.32E-02
Near maximal response at
LOAEL; constrained parameter
bound hit; standard deviations
given in paper interpreted as
standard errors
Continuous Hill,
constant variance,
unrestricted
(p = N/A)
9.96E-02
1.23E-09
Slightly supralinear fit (n =0.92)
Lietal. (1997)
(rat)
2.66E-01
7.99E-01
FSH in female rats
(n = 10)
3.6-fold t
(2.0 SD)
19-fold t
Continuous power,
modeled variance,
restricted
(/?<0.01)
2.00E+02
1.36E+02
Power hit lower bound
Continuous power,
modeled variance,
unrestricted
{p = 0.003)
1.96E-01
2.48E-02
Supralinear fit (power = 0.31)
Li et al. (2006)
(mouse)
1.59E-01
Serum estradiol
(n = 10)
2.0-fold t
(0.8 SD)
2.4-fold t
Continuous linear,
constant variance
(p = 0.16)
1.61E+01
5.38E+00
BMDL > LOAEL; high control
CV (1.25); near maximal response
at low dose; nonmonotonic
response; other model fits are
step-function-like
Serum progesterone
(n = 10)
33% |
(2.0 SD)
61% |
Continuous Hill,
modeled variance
(p = 0.39)
9.46E-04
8.01E-11
No response data near BMR;
large CVs (>1) for treatment
groups; poor fit for variance
model; Hill coefficient at lower
bound (step-function)
-------
Table 4-4. TCDD BMDL analysis (NOAEL, LOAEL, BMD, and BMDL values given as animal whole blood
concentrations in ng/kg)a (continued)
-j
o
o
o
2
o
H
o
HH
H
W
o
o
c
o
H
W
Studyb'c
NOAEL/
LOAEL
Endpoint
Control
response
First
response"1
Max
response6
Model fit detail
BMD/
BMDL
Comments
Markowski
etal. (2001)
(rat)
1.56E+00
FR5 run
opportunities
(n = 4-7)
10% |
(0.21 SD)
51% |
Continuous Hill,
constant variance
(p = 0.94)
Continuous power,
constant variance,
unrestricted
(p = 0.13)
1.72E+00
9.08E-01
2.67E+00
1.03E-14
Constrained parameter upper
bound hit
Saturated model; supralinear fit
(power =0.39); BMD/BMDL
ratio »100
FR2 revolutions
(n = 4-7)
9% 4
(0.15 SD)
43% |
Continuous Hill,
constant variance
(p = 0.65)
1.84E+00
5.99E-01
Constrained parameter bound hit
(upper bound)
Continuous power,
constant variance,
unrestricted
(p = 0.16)
5.74E+00
1.03E-14
Supralinear fit (power = 0.32)
FRIO run
opportunities
(n = 4-7)
15% |
(0.24 SD)
57% |
Continuous
exponential (M2),
constant variance
(p = 0.30)
8.57E+00
2.89E+00
BMDL > LOAEL
Miettinen et al.
(2006)
(rat)
2.22E+00
Cariogenic lesions
in pups
(n = 4-8)
25/42
23/29
29/32
Dichotomous log-
logistic, restricted
(p = 0.60)
1.43E+00
5.17E-01
Constrained parameter lower
bound hit; near maximal response
at LOAEL; high control response
Dichotomous log-
logistic, unrestricted
(p = 0.73)
4.94E-02
Supralinear fit (slope = 0.47);
BMDL could not be calculated
Murray et al.
(1979)
(rat)
1.12E+00
5.88E+00
Fertility in F2 gen.
(no litters)
(n = 20)
4/32
0/20
9/20
Dichotomous
multistage
(p = 0.08)
2.73E+00
1.37E+00
Poor fit; nonmonotonic response;
no response data near BMR
-------
Table 4-4. TCDD BMDL analysis (NOAEL, LOAEL, BMD, and BMDL values given as animal whole blood
concentrations in ng/kg)a (continued)
O
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
ffl
Studyb'c
NOAEL/
LOAEL
Endpoint
Control
response
First
responsed
Max
response6
Model fit detail
BMD/
BMDL
Comments
NTP (1982b)
(mouse)
7.67E-01
Toxic hepatitis;
males
(n = 50)
1/73
5/49
44/50
Dichotomous
multistage
(p = 0.04)
2.78E+00
1.34E+00
No acceptable model fits; lowest
BMDL shown
NTP (2006a)
(rat)
2.56E+00
Hepatocyte
hypertrophy
(n = 53-54)
0/53
19/54
52/53
Dichotomous
multistage
(p = 0.02)
9.27E-01
7.91E-01
Poor fits for all models
Alveolar metaplasia
(n = 52-54)
2/53
19/54
46/52
Dichotomous log-
logistic
(P = 0.72)
6.50E-01
3.75E-01
No response near BMR
Oval cell hyperplasia
(n = 53-54)
0/53
4/54
53/53
Dichotomous probit
(p = 0.23)
5.67E+00
4.79E+00
Relatively poor fit for control and
low-dose groups; negative
response intercept (same for
logistic); BMDL > LOAEL
Dichotomous Weibull
(p = 0.08)
5.72E+00
4.09E+00
Marginal fit; BMDL > LOAEL
Gingival hyperplasia
(n = 53-54)
1/53
7/54
16/53
Dichotomous log-
logistic, restricted
(p = 0.06)
5.85E+00
3.73E+00
Poor fit; constrained parameter
bound hit; BMDL > LOAEL
Dichotomous log-
logistic, unrestricted
(p = 0.66)
7.05E-01
1.26E-05
Supralinear fit (slope = 0.37)
Eosinophilic focus,
multiple
(n = 53-54)
3/53
8/54
42/53
Dichotomous probit
(p = 0.46)
5.58E+00
4.86E+00
Relatively poor fit to control
response; BMDL > LOAEL
Liver fatty change,
diffuse
(n = 53-54)
0/53
2/54
48/53
Dichotomous Weibull
(P = 0.72)
3.92E+00
2.86E+00
BMDL > LOAEL; otherwise
adequate fit
-------
Table 4-4. TCDD BMDL analysis (NOAEL, LOAEL, BMD, and BMDL values given as animal whole blood
concentrations in ng/kg)a (continued)
-j
O
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Studyb'c
NOAEL/
LOAEL
Endpoint
Control
response
First
response"1
Max
response6
Model fit detail
BMD/
BMDL
Comments
NTP (2006a)
(rat)
(continued)
2.56E+00
(continued)
Liver necrosis
(n = 53-54)
1/53
4/54
17/53
Dichotomous log-
probit, unrestricted
(p = 0.80)
7.50E+00
3.50E+00
Adequate fit; slightly supralinear;
BMDL > LOAEL
Liver pigmentation
(n = 53-54)
4/53
9/54
53/53
Dichotomous log-
probit
(p = 0.96)
2.46E+00
1.89E+00
Adequate fit
Toxic hepatopathy
(n = 53-54)
0/53
2/54
53/53
Dichotomous
multistage
(p = 0.69)
3.98E+00
3.06E+00
BMDL > LOAEL; otherwise
adequate fit
Ohsako et al.
(2001)
(rat)
1.04E+00
3.47E+00
Anogenital distance
in male pups
(»= 5)
12% |
(1.0 SD)
17% |
Continuous Hill,
constant variance,
restricted
(p = 0.15)
2.88E+00
8.03E-01
Constrained parameter lower
bound hit; near maximal response
at LOAEL
Continuous Hill,
constant variance,
unrestricted
(p = 0.056)
3.49E+00
3.05E-01
Supralinear fit (n = 0.59)
Schantz et al.
(1996)
3.38E+00
Facilitory effect on
radial arm maze
learning
(n = 10)
22% |
(1.2 SD)
34% |
Continuous linear,
constant variance
(p = 0.16)
7.00E+00
4.60E+00
BMDL > LOAEL; otherwise
adequate fit
Sewall et al.
(1995)
(rat)
7.11E+00
1.66E+01
Serum T4
(« = 9)
9.1%|
(0.6 SD)
40% |
Continuous Hill,
constant variance,
restricted
(p = 0.90)
1.03E+01
3.60E+00
Constrained parameter hit lower
bound; otherwise acceptable fit;
selected
Continuous Hill,
constant variance,
unrestricted
(p = 0.86)
9.71E+00
1.97E+00
Supralinear fit (power = 0.57)
-------
Table 4-4. TCDD BMDL analysis (NOAEL, LOAEL, BMD, and BMDL values given as animal whole blood
concentrations in ng/kg)a (continued)
-j
On
O
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Studyb'c
NOAEL/
LOAEL
Endpoint
Control
response
First
response"1
Max
response6
Model fit detail
BMD/
BMDL
Comments
Shi et al. (2007)
(rat)
3.42E-01
1.07E+00
Serum estradiol in
female pups
(n = 10)
38% |
(0.4 SD)
62% |
Continuous
exponential (M4),
modeled variance
(p = 0.69)
8.07E-01
3.54E-01
Adequate fit; selected
Smialowicz
et al. (2008)
(mouse)
4.38E-01
PFC per spleen
(n = 15)
24% |
(0.5 SD)
89% |
Continuous power,
unrestricted, modeled
variance
(p = 0.27)
1.19E+01
3.76E+00
BMDL > LOAEL; fit at control
and low dose inconsistent with
data; constrained parameters in
other models hit lower bounds
PFC per 10A6 cells
(n = 8-15)
24% |
(0.5 SD)
9.3-fold |
Continuous power
unrestricted, constant
variance
(p = 0.48)
1.90E+00
2.16E-01
Constant variance test failed;
observed control variance
underestimated by 35%; poor fits
for all modeled variance models
Smith et al.
(1976)
(mouse)
7.11E+00
5.06E+01
Cleft palate in pups
(n = 14-41)
0/34
2/41
10/14
Dichotomous log-
logistic, restricted
(p = 0.42)
3.52E+01
1.06E+01
Adequate fit; selected
Sparschu et al.
(2008; 1971)
(rats)
5.09E+00
1.63E+01
Male fetus weight
(n =3-117)
2.7% t
(0.1 SD)
33% |
Continuous
exponential (M5),
modeled variance
(p < 0.0001)
5.46E+02
1.30E+02
BMDL > LOAEL; variance not
captured by either variance
model; poor fit in region
surrounding NOAEL and LOAEL
Female fetus weight
(n = 4-129)
2.3% t
(0.06 SD)
30% |
Continuous
exponential (M2),
modeled variance
(p < 0.028)
1.03E+03
6.48E+02
BMDL > LOAEL; variance not
captured by either variance
model; poor fit in region
surrounding NOAEL and LOAEL
-------
Table 4-4. TCDD BMDL analysis (NOAEL, LOAEL, BMD, and BMDL values given as animal whole blood
concentrations in ng/kg)a (continued)
-j
-j
O
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Studyb'c
NOAEL/
LOAEL
Endpoint
Control
response
First
response"1
Max
response6
Model fit detail
BMD/
BMDL
Comments
Toth et al.
(1979)
(mouse)
5.73E-01
Skin lesions
(n = 38-44)
0/38
5/44
25/43
Dichotomous log-
logistic, restricted
(p = 0.08)
6.41E+00
4.02E+00
Constrained parameter lower
bound hit
Dichotomous
log-logistic,
unrestricted
{p = 0.74)
5.97E-01
6.77E-02
Supralinear fit (slope = 0.48)
5.73E-01
(cont.)
Dermal amyloidosis
(n = 38-44)
0/38
5/44
17/43
Dichotomous log-
logistic, restricted
(p = 0.05)
1.50E+01
8.75E+00
Poor fit; constrained parameter
lower bound hit; BMDL >
LOAEL
Dichotomous log-
logistic, unrestricted
(p = 0.90)
4.84E-01
5.31E-03
Supralinear fit (slope = 0.33)
Van Birgelen
et al. (1995a)
(rat)
7.20E+00
Hepatitic retinol
(«= 8)
44% |
(0.74 SD)
96% |
Continuous
exponential (M4),
modeled variance
(/?<0.01)
2.49E+01
3.36E+00
Poor fit
Continuous power,
modeled variance,
unrestricted
(p = 0.01)
3.80E-01
1.39E-02
Poor fit; supralinear fit
(power = 0.14)
Hepatitic retinyl
palmitate (n = 8)
80% |
(1.4 SD)
99% |
Continuous
exponential (M4),
modeled variance
(/?<0.01)
1.42E+02
3.65E+01
Poor fit; no response near BMR
Continuous power,
modeled variance,
unrestricted
(p = 0.24)
5.26E-02
5.89E-05
Supralinear fit (power = 0.06)
-------
Table 4-4. TCDD BMDL analysis (NOAEL, LOAEL, BMD, and BMDL values given as animal whole blood
concentrations in ng/kg)a (continued)
Studyb'c
NOAEL/
LOAEL
Endpoint
Control
response
First
response"1
Max
response6
Model fit detail
BMD/
BMDL
Comments
White et al.
(1986)
(mouse)
1.09E+00
Total hemolytic
complement activity
(CH50)
(» = 8)
41% |
(2.6 SD)
81% |
Continuous
Hill, modeled
variance, restricted
(p = 0.002)
8.63E+00
1.50E+00
Poor fit; no response near BMR;
constrained parameter bound hit;
BMDL > LOAEL
Continuous Hill,
modeled variance,
unrestricted
(p = 0.07)
1.48E-01
4.35E-03
Supralinear fit (n = 0.25)
'Animal whole blood concentrations were used to determine the HEDs in Table 4-3 and Table 4-5.
bThe following studies previously presented in Table 4-3 are not presented in Table 4-4 because toxicokinetic models for guinea pigs, minks, or monkeys, and
were not found: DeCaprio et al. (19861: Hochstein et al (20011: Rier et al. (1995: 19931: Vos et al. (19731: Yang et al. (20001.
°The following studies previously presented in Table 4-3 are not presented in Table 4-4 because the data were not amenable to BMD modeling: Chu et al. (20011:
Chu et al. (20071: Croutch et al. (20051: Fattore et al. (20001: Fox et al. (19931: Franczak et al. (20061: Hutt et al. (20081: Ikeda et al. (20051: Ishihara et al.
(20071: Kociba et al. (19761: Maronpot et al. (19931: Nohara et al. (20001: Nohara et al. (20021 Schantz et al. (19961Seo et al. (19951: Simanainen et al. (20021:
Simanainen et al. (20031: Simanainen et al. (2004a); Smialowicz et al. (20041: Weber et al. (19951.
dMagnitude of response at first dose where response differs from control value (in the adverse direction); continuous response magnitudes given as relative to
control plus change relative to control standard deviation; quantal response given as number affected/total number.
"Magnitude of response maximally differing from control value (in the adverse direction).
SD = standard deviation; S-D = Sprague-Dawley; L-E = Long-Evans; H-W = Han-Wistar.
-------
Table 4-5. Candidate PODs for the TCDD RfD using blood-concentration-based human equivalent doses
vo
O
O
o
2
o
H
o
HH
H
W
O
o
c
o
H
W
Study
Species, strain
(sex, if not
both)
Protocol
Endpoint
NOAELHED(N)or
BMDLhed (B)
(ng/kg-day)
LOAELm.|)
(ng/kg-day)
UFa
RfD
(mg/kg-day)
Li et al. (2006)
Mouse, NIH (F)
Gavage GDs 1-3;
n = 10
Hormone levels in pregnant dams (decreased
progesterone, increased estradiol)
—
1.6E-03
300
5.3E-12
Kuchiiwa et al.
(2002)
Mouse, ddY
Maternal 8 week-
gavage prior to
mating; n = 3
Decreased serotonin-immunoreactive neurons
in raphe nuclei of male offspring (Fl)
2.7E-03
300
9.2E-12
Smialowicz
et al. (2008)
Mouse, B6C3Fi
(F)
90-day gavage;
n = 8-15
Decreased SRBC response
—
6.3E-03
300
2.1E-11
Bowman
et al.(1989a:
1989b): others'3
Rhesus Monkey
(F)
Daily dietary
exposure, 3.5-4
years
n = 3-7
Neurobehavioral effects
8.2E-030
300
2.7E-11
Keller et al.
(2008a: 2008b:
2007)d
Mouse, CBA/J
and C3H/HeJ
Gavage GD 13;
n = 23-36 (pups)
Missing molars, mandibular shape changes in
pups
9.9E-03
300
3.3E-11
Toth et al.
(1979)
Mouse, Swiss/
H/Riop (M)
1-year gavage;
n = 38-44
Dermal amyloidosis, skin lesions
—
9.9E-03
300
3.3E-11
Latchoumy-
candane and
Mathur (2002);
others'5
Rat, Wistar (M)
45-day oral
pipetting; n 6
Decreased sperm production
1.6E-02
300
5.4E-11
NTP (1982b)
Mouse, B6C3FJ
(M)
2-year gavage;
n = 50
Liver lesions
—
2.2E-02
300
7.2E-11
-------
Table 4-5. Candidate points of departure for the TCDD RfD using human equivalent doses (continued)
00
o
O
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Study
Species, strain
(sex, if not
both)
Protocol
Endpoint
NOAELHED(N)or
BMDLhed (B)
(ng/kg-day)
LOAELhed
(ng/kg-day)
UFa
RfD
(mg/kg-day)
White et al.
(1986)
Mouse, B6C3Fi
(F)
14-day gavage;
n = 6-8
Decreased serum complement
—
2.8E-02
300
9.2E-11
Li et al. (1997)
Rat, S-D
(F, 22 day-old)
Single gavage;
n= 10
Increased serum FSH
2.9E-03 (N)
1.7E-02
30f
9.7E-11
DeCaprio et al.
(1986s)
Guinea pig,
Hartley
90-day dietary;
n = 10
Decreased body weight, organ weight
changes (liver, kidney, thymus, brain)
4.1E-03C(N)
3.3E-020
30f
1.4E-10
Shi et al. (2007)
Rat, S-D (F)
11-month gavage;
n = 10
Decreased serum estradiol
4.5E-03 (N)
4.7E-03 (B)
2.7E-02
30f
1.6E-10
Markowski
et al. (200 1)
Rat, Holtzman
Gavage GD 18;
n = 4-7
Neurobehavioral effects in pups (running,
lever press, wheel spinning)
—
5.2E-02
300
1.7E-10
Hojo et al.
(2002); Zareba
et al. (2002)
Rat, S-D
Gavage GD 8;
n = 12
Food-reinforced operant behavior in pups
5.5E-02
300
1.8E-10
Cantoni et al.
(1981)
Rat, CD-COBS
(F)
45-week gavage;
n = 4
Increased urinary porhyrins
—
6.4E-02
300
2.1E-10
Vos et al.
(1973)
Guinea pig,
Hartley (F)
8-week gavage;
n = 10
Decreased delayed-type hypersensitivity
response to tuberculin
6.4E-03C (N)
3.2E-020
30f
2.1E-10
Miettinen et al.
(2006)
Rat, Line C
Gavage GD 15;
n = 3-10
Cariogenic lesions in pups
—
8.9E-02
300
3.0E-10
Kattainen et al.
(2001)
Rat, Line C
Gavage GD 15;
n = 4-8
Inhibited molar development in pups
—
9.0E-02
300
3.0E-10
NTP (2006a)
Rat, S-D (F)
2-year gavage;
n = 53
Liver and lung lesions
—
1.4E-01
300
4.5E-10
Amin et al.
(2000)
Rat, S-D
Gavage GDs 10-16;
n = 10
Reduced saccharin consumption and
preference
—
1.7E-01
300
5.7E-10
Schantz et al.
(1996)
Rat, S-D (F)
Gavage GDs 10-16;
n = 80-88
Maze errors (facilitatory effect)
—
1.7E-01
300
5.7E-10
-------
Table 4-5. Candidate points of departure for the TCDD RfD using human equivalent doses (continued)
00
o
O
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Study
Species, strain
(sex, if not
both)
Protocol
Endpoint
NOAELHED(N)or
BMDLhed (B)
(ng/kg-day)
LOAELhed
(ng/kg-day)
UFa
RfD
(mg/kg-day)
Mocarelli et al.
(2008)
Human (M)
Childhood
exposure; n = 157
Decreased sperm concentration and sperm
motility, as adults
—
2.0E-028
30h
6.7E-10
Baccarelli
et al. (2008)
Human infants
Gestational
exposure; n = 51
Increased TSH in newborn infants
—
2.0E-02'
30h
6.7E-10
Hutt et al.
(2008)
Rat, S-D (F)
13-week dietary;
n = 3
Embryotoxicity
—
2.5E-01
300
8.4E-10
Ohsako et al.
(200 n
Rat, Holtzman
Gavage GD 15;
n = 5
Decreased anogenital distance in male pups
2.7E-02 (N)
1.8E-01
30f
9.1E-10
Murray et al.
(1979)
Rat, S-D
3-generation dietary
Reduced fertility and neonatal survival (F0
andFl)
2.9E-02 (N)
3.8E-01
30f
9.6E-10
Franczak et al.
(2006)
Rat, S-D (F)
Gavage GD 14, 21,
PND 7, 14; n = 7
Abnormal estrous cycle
—
3.2E-01
300
1.1E-09
Chu et al.
(2007)
Rat, S-D (F)
28-day gavage,
n = 5
Liver lesions
3.5E-02 (N)
5.6E-01
30f
1.2E-09
Bell et al.
(2007b)
Rat, CRL:WI
(Han) (M)
17-week dietary;
n = 30
Delay in onset of puberty
4.3E-02 (B)
8.9E-02
30f
1.4E-09
Ishihara et al.,
(2007)
Mouse, ICR (M)
Weekly gavage for
5 weeks; n = 42-43
Decreased male/female sex ratio
_i
5.0E-01
300
1.7E-09
VanBirgelen
et al. (1995a)k
Rat, S-D (F)
13-week dietary;
n = 8
Decreased liver retinyl palmitate
—
5.1E-01
300
1.7E-09
Kociba et al.
(1978)
Rat, S-D (F)
2-year dietary;
n = 50
Liver and lung lesions, increased urinary
porhyrins
6.3E-02 (N)
6.3E-01
30f
2.1E-09
Fattore et al.
(2000)
Rat, S-D
13-week dietary;
n = 6
Decreased hepatic retinol
—
7.8E-01
300
2.6E-09
Seo et al.
(1995)
Rat, S-D
Gavage GDs 10-16;
n = 10
Decreased serum T4 and thymus weight
1.7E-01 (N)
9.1E-01
30f
5.6E-09
Crofton et al.
(2005)
Rat, Long-Evans
(F)
4-day gavage;
n = 4-14
Decreased serum T4
1.7E-01 (N)
7.4E-01
30f
5.6E-09
-------
Table 4-5. Candidate points of departure for the TCDD RfD using human equivalent doses (continued)
Study
Species, strain
(sex, if not
both)
Protocol
Endpoint
NOAELHED(N)or
BMDLhed (B)
(ng/kg-day)
LOAELhed
(ng/kg-day)
UFa
RfD
(mg/kg-day)
Sewall et al.
(1995)
Rat, S-D (F)
30-week gavage;
n = 9
Decreased serum T4
5.0E-01 (N)
1.8E-01 (B)
1.7E+00
30f
6.0E-09
Franc et al.
(2001)
Rat, Long-Evans
(F)
22-week gavage;
n = 8
Increased relative liver weight; decreased
relative thymus weight
4.5E-01 (N)
2.6E-01 (B)
1.4E+00
30f
8.7E-09
Kociba et al.
(1976)
Rat, S-D
5-days/week gavage
for 13 weeks; n = 12
Liver and lung lesions, increased urinary
porphyrins
2.6E-01 (N)
3.0E+00
30f
8.7E-09
Sparschu et al.
(1971)
Rat, S-D (F)
Gavage GD 6-15;
n = 4-129
Decreased fetal body weight
3.2E-01 (N)
1.7E+00
30f
1.1E-08
Alaluusua et al.
(2004)
Human
Childhood exposure;
n = 48
Dental defects
4.1E-021 (N)
9.0E-01m
3n
1.4E-08
10 aExcept where indicated, UFA = 3 (for dynamics), UFH = 10, UFL = 10.
bSchantz and Bowman (^1989^); Schantz et al. (19861: Schantz et al. (19861.
°HED determined from lst-order body burden model; no PBPK model available for guinea pigs or monkeys; Hochstein et al. (20011 was not presented in the
table because no PBPK model exists for minks and lst-order body burden could not be calculated because a TCDD half-life could not be determined.
dResults from three separate studies with identical designs combined.
O eLatchoumycandane et al. (2002a: 2002b).
^ fUFL = 1 (NOAEL or BMDL).
t-ri 8Mean of peak exposure (0.0321 ng/kg-day) and average exposure over 10-year critical window (0.0080 ng/kg-day).
H hUFH = 3, UFl = 10.
q 'Maternal exposure corresponding to neonatal TSH concentration exceeding 5 |iU/mL.
O JThe NOAEL of 4.9E-5 was excluded from consideration because of the large dose spacing in the study.
2 kVan Birgelen et al. (1995b) is considered to be the same study but reports effects at doses above the LOAEL that are not considered further; this study in not
O carried forward for determination of an RfD POD but is included in the RfD uncertainty analysis presented in Section 4.4.
^ 'Mean of peak exposure (0.0655 ng/kg-day) and average exposure over 10-year critical window (0.0156 ng/kg-day).
— mMean of peak exposure (1.65 ng/kg-day) and average exposure over 10-year critical window (0.149 ng/kg-day).
a nuFH=3.
O
P3 S-D = Sprague-Dawley.
O
c
o
H
ffl
-------
Table 4-6. Qualitative analysis of the strengths and limitations/uncertainties associated with animal bioassays
providing PODs for the TCDD RfD
00
O
o
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Study
Strengths
Limitations
Remarks
Bell et al.
(2007b)
• Large sample size of both rat dams and
offspring/dose employed
• Several developmental effects tested
• Batch-to-batch variation of up to 30% in TCDD
concentration in the diet
• Longer-term dosing of dams does not accurately
define gestational period when fetus is especially
sensitive to TCDD-induced toxicity
Study is a significant addition
to a substantial database on the
developmental toxicity of
TCDD in laboratory animals
Cantoni et al.
(1981)
• Experiments were designed to test qualitative
and quantitative composition and the course of
urinary excretion in TCDD-induced porphyria
• Small sample size of rats/dose employed (n = 4)
• Concurrent histological changes with tissue
porphyrin levels were not examined
• TCDD used for dosing was of unknown purity
Early study on porphyrogenic
effects of TCDD
DeCaprio et al.
(1986)
• Subchronic oral dosing duration up to 90 days
• Male and female guinea pigs tested
• Relatively small sample size of guinea pigs/dose
employed (n= 10)
• No histopathological analyses performed
• TCDD used for dosing was of unknown purity
Limited subchronic study;
PBPK model not available for
estimation of HED
Franc et al.
(2001)
• Three different rat strains with varying
sensitivities to TCDD were utilized (Sprague-
Dawley, Long Evans, Han/Wistar)
• Longer-term oral dosing up to 22 weeks
• Relatively small sample size of rats/dose employed
(« = 8)
• Only female rats were tested
• Concurrent liver histopathological changes with
liver-weight changes were not examined
• Gavage exposure was only biweekly
Limited subchronic study
Hoio et al. (2002)
• Low TCDD dose levels used allowed for subtle
behavioral deficits to be identified in rat
offspring
• Preliminary training sessions in operant
chamber apparatuses were extensive
• Neurobehavioral effects are exposure-related
and cannot be attributed to presence of learning
or discrimination deficits
• Relatively small sample size of rat dams/dose
employed (n= 12)
• Small sample size of rat offspring/dose evaluated
(n = 5-6)
• Neurobehavioral effects induced by TCDD at earlier
or later gestational dosing dates are unknown
because of single gavage administration on GD 8
• Although BMD analysis was conducted, the model
parameters were not constrained according to EPA
guidance, so the results cannot be used
One of a few neurobehavioral
toxicity studies; somewhat
limited study size
-------
Table 4-6. Qualitative analysis of the strengths and limitations/uncertainties associated with animal bioassays
possessing candidate points-of-departure for the TCDD RfD (continued)
00
o
O
o
2
o
H
o
HH
H
W
O
V
o
c
o
H
W
Study
Strengths
Limitations
Remarks
Keller et al.
(2008a: 2008b:
2007)
• Six different inbred mouse strains were utilized
• Large sample size of mouse
offspring/dose/strain evaluated
• Low TCDD dose levels used compared to
typical mouse studies allowed for identification
of subtle sensitivity differences in presence of
absence of third molars, variant molar
morphology, and mandible structure in offspring
• Unknown sample size of mouse dams/dose/strain
employed
• All inbred strains possessed sensitive b allele at the
Ahr locus (i.e., a potentially resistant subpopulation
was not evaluated for comparison purposes)
• Morphological dental and mandibular changes
induced by TCDD at earlier or later gestational
dosing dates are unknown because of single gavage
administration on GD 13
• Difficulties breeding A/J mice led to abandonment
of that strain in the analysis (Keller et al.. 2008a:
Keller et al.. 2008b)
Endpoint similar to effects
observed at higher exposure
levels in humans; HED highly
uncertain using mouse PBPK
model
Latchoumy-
candane and
Mathur (2002)
• Compared to epididymal sperm counts, the
testicular spermatid head count provides better
quantitation of acute changes in sperm
production and can indicate pathology
• Small sample size of rats/dose employed (n = 6)
• Oral pipette administration of TCDD may be a less
efficient dosing method than gavage
Endpoint has human relevance,
similar to critical effects in
principal human study for RfD
Li et al. (2006)
• Female reproductive effects (i.e., early embryo
loss and changes in serum progesterone and
estradiol) were tested at multiple exposure
times—early gestation, preimplantation, and
peri- to postimplantation
• Small sample size of dams/dose (n = 10)
• Large dose-spacing interval (25-fold at lowest
2 doses)
Endpoint has human relevance
but HED highly uncertain
using mouse PBPK model
Markowski et al.
(2001)
• Low TCDD dose levels used allowed for subtle
behavioral deficits to be identified in rat
offspring
• Several training sessions on wheel apparatuses
were extensive
• Neurobehavioral effects are exposure-related
and cannot be attributed to motor or sensory
deficits
• Unknown sample size of rat dams/dose employed
• Small sample size of rat offspring/dose evaluated
(n = 4-7)
• TCDD used for dosing was of unknown purity and
origin
• Only two treatment levels
• Neurobehavioral effects induced by TCDD at earlier
or later gestational dosing dates are unknown
because of single gavage administration on GD 18
One of a few neurobehavioral
toxicity studies; somewhat
limited study size
-------
Table 4-6. Qualitative analysis of the strengths and limitations/uncertainties associated with animal bioassays
possessing candidate points-of-departure for the TCDD RfD (continued)
00
o
o
o
2
o
H
o
HH
H
W
o
o
c
o
H
W
Study
Strengths
Limitations
Remarks
NTP (1982b)
• Large sample size of mice and rats/dose
employed
• Comprehensive 2-year bioassay that assessed
body weights, clinical signs, and pathological
changes in multiple tissues and organs
• Elevated background levels of hepatocellular tumors
in untreated male mice
• Gavage exposure was only 2 days/week
• Only two treatment levels
Comprehensive chronic
toxicity evaluations of TCDD
in rodents; HED highly
uncertain using mouse PBPK
model
NTP (2006a)
• Chronic exposure duration with several interim
sacrifices
• Large number of dose groups with close spacing
• Large number of animals per dose group
• Comprehensive suite of endpoints evaluated
• Comprehensive biochemical, clinical, and
histopathological tests and measures
• Detailed reporting of results, with individual
animal data presented as well as group
summaries
• Single species, strain, and sex
• Lowest dose tested too high for establishing
NOAEL
Study is the most
comprehensive chronic TCDD
toxicity evaluation in rats to
date
Shi et al. (2007)
• Study design evaluated TCDD effects on aging
female reproductive system (i.e., exposure
began in utero and spanned across reproductive
lifespan)
• Several female reproductive endpoints were
evaluated, including cyclicity, endocrinology,
serum hormone levels, and follicular reserves
• Relatively small sample size of rats/dose employed
(n = 10)
Endpoint similar to effects
observed at higher exposure
levels in humans
Smialowicz et al.
(2008)
• Sheep red blood cell (SRBC) plaque forming
cell assay is highly sensitive and reproducible
across laboratories when examining TCDD
• Small sample size of animals/dose (n = 8)
• Only female mice were tested
• Thymus and spleen weights were only other
immune response-related endpoints tested
Limited immunotoxicity study
Toth et al. (1979)
• Large sample size of mice/dose employed
• Chronic exposure duration
• Reporting of findings is terse and lacks sufficient
detail (e.g., materials and methods, thorough
description of pathological findings, etc.)
• Limited number of endpoints examined
• Only male mice were tested
Limited chronic study; HED
highly uncertain using mouse
PBPK model
-------
Table 4-6. Qualitative analysis of the strengths and limitations/uncertainties associated with animal bioassays
possessing candidate points-of-departure for the TCDD RfD (continued)
Study
Strengths
Limitations
Remarks
Vosetal. (1973)
• Three different animal species tested (guinea
pigs, mice, and rats)
• Effects of TCDD tested on both cell-mediated
and humoral immunity
• Small sample size of animals/dose employed in each
experiment (n = 5-10)
• Only female guinea pigs and rats were tested, and
only male mice were tested
• Only one experimental assay was utilized to assess
cell-mediated or humoral immunity; humoral
immunity was only investigated in guinea pigs
• TCDD used for dosing was of unknown purity
Endpoints relevant to humans
but study size limited; PBPK
model not available for
estimation of HED
White et al.
(1986)
• Total hemolytic complement (CH50) is
representative functional assay of the complete
complement sequence
• Small sample size of rats/dose employed (n = 6-8)
• Individual complement factors may be significantly
depleted without affecting CH50 activity (only C3 is
measured)
• TCDD used for dosing was of unknown purity
Endpoint similar to effects
observed at higher exposure
levels in humans; HED highly
uncertain using mouse PBPK
model
-------
1 Table 4-7. Basis and derivation of the TCDD reference dose
2
Princi
jal study detail
Study
POD (ng/kg-day)
Critical effects
Mocarelli et al. (2008)
0.020 (LOAEL)
Decreased sperm count (20%) and motility (11%) in
men exposed to TCDD during childhood
Baccarelli et al. (2008)
0.020 (LOAEL)
Elevated TSH (>5 |iU/mL) in neonates
RfD derivation
POD
0.020 ng/kg-day (2.0E-8 mg/kg-day)
UF
30 (UFl = 10, UFh = 3)
RfD
7 x 10~10 (7E-10) mg/kg-day (2.0E-8 - 30)
Uncertainty factors
LOAEL-to-NOAEL
(UFl)
10
No NOAEL established; cannot quantify lower exposure
group in Baccarelli et al. (2008); magnitude of effects at
LOAEL sufficient to require a 10-fold factor.
Human interindividual
variability
(UFh)
3
A factor of 3 (10°'5) is used because the effects were
elicited in sensitive lifestages. A further reduction to 1
was not made because the sample sizes were relatively
small, which, combined with uncertainty in exposure
estimation, may not fully capture the range of
interindividual variability. In addition, chronic effects
are levels are not fully elucidated for humans and could
possibly be more sensitive.
Interspecies extrapolation
(UFa)
1
Human study.
Sub chroni c-to-chroni c
(UFS)
1
Chronic effect levels are not well defined for humans;
however, animal bioassays indicate that duration of
exposure does not seem to be a determining factor in
toxicological outcomes. Developmental effects and
other short-term effects occur at doses similar to effects
noted in chronic studies. Considering that exposure in
the principal studies encompasses the critical window of
susceptibility associated with development, a UF to
account for exposure duration is not used.
Database sufficiency
(UFd)
1
The database for TCDD contains an extensive range of
human and animal studies that examine a
comprehensive set of endpoints. There is no evidence to
suggest that additional data would result in a lower RfD.
4-87
DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Table 4-8. Alternative PODs for the impact of TCDD exposure during
gestation and nursing on semen quality of male offspring (Mocarelli et al..
2011)
POD type
Age-at-conception
scenario
Averaging
protocol3
Maternal intake (ng/kg-day)
TCDD only
TCDD + DLCb
NOAEL
30 years
Cont. avg.
2.9 x 1(T4
2.90 x 10~3
LOAEL
1.64 x 1(T3
5.36 x 10~3
NOAEL
45 years
Cont. avg.
1.9 x 1(T4
1.90 x 10~3
LOAEL
1.10 x 10~3
4.80 x 10~3
aCont. avg. = average continuous exposure over the specified duration.
bAddedDLC = 9 x TCDD intake for NO AEL (in background range), 3.51 x 10 3 ng/kg-day for LOAEL (above
background).
Table 4-9. Alternative PODs for developmental endpoints other than
increased neonatal TSH and semen quality
Population, endpoint
(cite)
POD type
Averaging
protocol3
TCDD only (ng/kg-day)
TCDD + DLC (ng/kg-day)
Needham
Eskenazi
Needhamb
Eskenazi0
Girls, duration of menstrual
cycle as women
(Eskenazi et al.. 2002b)
NOAEL
Cont. avg.
9.52 x 10~3
2.90 x 10~3
0.0130
0.0120
LOAEL
Peak
3.13
2.94
3.13
2.95
Window
0.122
0.126
0.126
0.135
P/W avg.
1.64
1.53
1.64
1.54
Girls and boys,
developmental dental effects
(Alaluusua et al.. 2004)
NO A F.I.
Peak
0.0655
0.0437
0.0688
0.0528
Window
0.0157
0.0175
0.0190
0.0266
P/W avg.
0.0406
0.0306
0.0439
0.0397
LOAEL
Peak
1.65
1.51
1.65
1.52
Window
0.149
0.151
0.152
0.160
P/W avg.
0.897
0.841
0.900
0.841
Girls, age at menarche
(Warner et al.. 2004)
NOAEL
Peak
0.604
0.517
0.607
0.526
Window
0.0394
0.0424
0.0427
0.0515
P/W avg.
0.322
0.280
0.325
0.289
aCont. avg. = average continuous daily intake over the specified duration; Peak = average intake for peak exposure;
Window = average intake for critical-window exposure; PAV avg. = average of "Peak" and "Window" intakes.
bAdded DLC = 3.51 x 10 ' ng/kg-day for girls, 3.33 x 10 3 ng/kg-day for boy/girl average.
°Added DLC = 9.10 x 10 ng/kg-day for all.
4-88
DRAFT - DO NOT CITE OR QUOTE
-------
1 Table 4-10. Alternative PODs for adult endpoints for which critical exposure
2 windows are undefined
3
Population, endpoint
(cite)
POD type
Averaging
protocol"
TCDD only
(ng/kg-day)
TCDD + DLCb
(ng/kg-day)
Men, sex ratio of
offspring
(Mocarclli et al.. 2000)
NOAEL
Peak
0.0341
0.0373
Window
1.58 x 10~3
4.73 x 10~3
PAV avg.
0.0178
0.0210
LOAEL
Peak
0.162
0.165
Window
4.69 x 10~3
7.84 x 10~3
PAV avg.
0.0831
0.0863
Women, age at
menopause
CEskenazi et al.. 2005s)
NOAEL
Peak
1.6 x 10~4-3.4 x 10~3
1.6 x 10~3-6.9 x 10~3
Window
1.6 x 10~4-1.0 x 10~3
1.6 x 10~3-4.5 x 10~3
PAV avg.
1.6 x 10~4-2.2 x 10~3
1.6 x 10~3-5.7 x 10~3
LOAEL
Peak
0.013-0.052
0.016-0.055
Window
1.7 x 10~3-3.4 x 10~3
5.2 x 10~3-7.0 x 10~3
PAV avg.
7.3 x 10~3-0.028
0.011-0.031
Women, ovarian function,
progesterone
(Warner et al.. 2007)
NOAEL
Peak
0.204
0.208
Window
3.00 x 10~3
6.51 x 10~3
PAV avg.
0.104
0.108
4
5 'Cont. avg. = average continuous daily intake over the specified duration; Peak = average intake for peak exposure;
6 Window = average intake for critical-window exposure; PAV avg. = average of "Peak" and "Window" intakes.
7 bAddedDLC = 3.15 x 10 1 ng/kg-day for males. 3.51 x 10 3 ng/kg-day for females. 3.33 x 10 3 ng/kg-day for
8 male/female average.
4-89
DRAFT - DO NOT CITE OR QUOTE
-------
No
Yes
Exclude study from
POD estimation
Does the
study provide data
on n on cancer effects an dTCDD
ixposurefor determining a POD o
Ss\a toxicologically relevant/^
endpoint?
Include as POD
Identify a study NOAELor LOAEL
for use in POD estimation
List of key noncancer epidemiologic studies for
quantitative dose-response analysis of TCDD
Use kinetic model to estimate
continuous oral daily intake (ng/kg-day)
in the affected study population
Figure 4-1. EPA's process to identify and estimate PODs from key
epidemiologic studies for use in noncancer dose-response analysis of TCDD.
For each noncancer study that qualified using the study inclusion criteria, EPA evaluated the dose-
response information developed by the study authors for whether the study provided noncancer
effects and TCDD dose data for a toxicologically relevant endpoint. If such data were available,
EPA identified a NOAEL or LOAEL as a POD. Then, EPA used a human kinetic model to
estimate the continuous oral daily intake (ng/kg-day) for the POD that could be used in the
derivation of a candidate RfD based on the study data. If all of this information was available,
then the result was included as a POD.
4-90
DRAFT - DO NOT CITE OR QUOTE
-------
NoncancerAnimal Bioassays Selected for
TCDD Dose-Response Assessment (See Tables 2-4 and D-1
78 Studies
Identify and Estimate PODs from the 62 Remaining Animal Bioassays
for use in Noncancer Dose-Response Analysis of TCDD
(See Figure 4-3)
Derive Candidate RfDs from the
48 Remaining Noncancer Animal Bioassays
Final Candidate RfDs from Noncancer Animal Bioassays
(11 Studies Presented as Supporting Information;
See Table 4-5)
37 Candidate RfDs
Burleson etal. (1996)
Hassoun etal. (1998)
Hassoun etal. (2002)
Hong etal. (1989)
Latchoumycandaneetal. (2003)
Mallyand Chipman (2002)
Slezaket al. (2000)
Tritscher et al. (1992)
Eliminate Studies with NoToxicologically Relevant Endpoints for RfD Derivation
(See Appendix H and Section 4.2.1)
16Studies Eliminated
DeVitoetal. (1994)
Hassoun etal. (2000)
Hassoun etal. (2003)
Kitchin and Woods (1979)
Lucieret al. (1986)
Sewall etal. (1993)
Sugita-Konishi etal. (2003)
Vanden Heuvel etal. (1994)
Chu etal. (2001)
Fox etal. (1993)
Maronpotet al. (1993)
Simanainen etal. (2002, 2003, 2004a) Smialowicz et. al. (2004)
Weber etal. (1995)
Smith etal. (1976)
*Hochstein etal. (2001) is also not carried forward because of the
lack of toxicokinetic information forestimation of an HED
Eliminate Studies with Both a
LOAELHed>1 ng/kg-dand NOAELhed/BMDLhed5-0.32 ng/kg-d* (See Table 4.3)
14 Studies Eliminated
Croutch etal. (2005)
Ikeda et al. (2005)
Noharaetal. (2000,2002)
Final Candidate RfDs from Noncancer Animal Bioassays
(11 Studies Presented as Supporting Information;
See Table 4-5)
37 Candidate RfDs
1
2 Figure 4-2. Disposition of noncancer animal bioassays selected for TCDD
3 dose-response analysis.
4 EPA evaluated each noncancer endpoint found in the 78 studies that passed the study inclusion
5 criteria. From this evaluation EPA eliminated 16 studies that contained no toxicologically
6 relevant endpoints for RfD derivation. Then as detailed in Figure 4-3, EPA selected and
7 identified PODs for use in deriving candidate RfDs. EPA then eliminated 13 studies based on
8 dose limits for the PODs' HEDs; one study was also not carried forward because of the lack of
9 toxicokinetic information for estimation of an HED. Of the remaining 48 studies, EPA derived 37
10 RfD candidates, with 11 studies presented as supporting information.
4-91 DRAFT - DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
No
Yes
No
No
Is the BMDLIess
than the LOAEL?
Yes
Yes
No
Yes
Yes
No
Is the
endpointobserved
Jiear the LOAEL?.
Is the endpointless
than the minimum
^ LOAEL x100? ^
Is the
endpoint under consideration
toxicologically
relevant?
Does kinetic modeling
suggest considering additional
-^gndpoints at higher doses?^-
Exclude endpoint
asa POD
Include NOAEL/LOAEL/BMDL
asa POD
Determine NOAEL, LOAEL, and BMDL (if possible) human equivalentdose
(HED) based on 1st-orderbody burden foreach study/endpointcombination
Estimate a Human Equivalent Dose (HED)
corresponding to each blood concentration NOAEL, LOAEL, or BMDL
usingthe Emond human PBPKmodel
Study/endpoint combinations from key noncanceranimal bioassayswith at
least one toxicologically relevant endpointfor RfD derivation
Determine a NOAEL, LOAEL, and BMDL (if possible) for each
study/endpoint combination, based on blood concentrationsfrom the
Emond rodent PBPKmodel
Figure 4-3. EPA's process to identify and estimate PODs from key animal
bioassays for use in noncancer dose-response analysis of TCDD.
For the studies with at least one toxicologically relevant endpoint, EPA first determined if each
endpoint was toxicologically relevant. If so, EPA determined the NOAEL, LOAEL, and BMDL
Human Equivalent Dose (HED) based on lst-order body burdens for each endpoint. Within each
study, these potential PODs were included when the endpoint was observed near the LOAEL and
if the BMDL was less than the LOAEL. Then if the endpoint was less than the minimum LOAEL
xlOO across all studies, EPA calculated PODs based on blood concentrations from the Emond
rodent PBPK model and, for all of the PODs, HEDs were estimated using the Emond human
PBPK model. If the kinetic modeling results suggested considering additional endpoints at higher
doses, the process was repeated. Finally, the lowest group of the toxicologically relevant PODs
was selected for final use in derivation of candidate RfDs.
4-92
DRAFT - DO NOT CITE OR QUOTE
-------
aionO "ao 3ii3 ion oa - uvaa
£6-17
Ingestion Rate
(Human-equivalent dose, ng/kg-d)
era
c
re
H
H
T3
O
W!
S
"t
re
"t
re
W!
•a
©
s
W!
re
as
"t
"t
as
V!
O5
"t
5'
era
re
VI
rt-
o"
s
re
M
T3
O
W!
S
"t
re
o
H
n
o
o
o
o
o
o
o
o
o
o
o
Moccarelli etal., 2008 (h
Baccarelli etal., 2008 (h
Alaluusuaetal.,2004 (H
Li etal., 1997 (R
DeCaprio etal., 1986 (G
Sm ialowicz et al., 2008 (M
Vosetal., 1973 (G
White etal., 1986 (M
Chu etal.,2007 (R
Cantoni etal., 1981 (R
Crofton etal., 2005 (R
Sewalletal., 1995(R
Franc etal., 2001 (R
Kociba etal., 1976 (R
VanBirgelenetal., 1995a (R
Fattoreetal., 2000 (R
Toth etal., 1979 (M
NTP, 1982 (M
Kociba etal., 1978 (R
NTP, 2006 (R
Shi etal., 2007 (R
Bowman etal. 1989, etc. (Mk
Latch, and Mathur, 2002 (R
Murrayetal.,1979(R
Ishihara etal., 2007 (M
Li etal., 2006 (M
Kuchiiwa et al., 2002(IV
Kelleretal., 2007, etc. (M
Ohsako etal., 2001 (R
Bell etal., 2007a (R
Markowski etal., 2001 (R
Hojo etal., 2002 (R
Miettinen etal., 2006 (R
Kattainen etal., 2001 (R
Seoetal., 1995(R
Schantzetal., 1996 (R
Aminetal.,2000(R
Hutt etal., 2008 (R
Franczaketal.,2006 (R
Sparschu etal., 1971 (R
~-
l>
tC-
~ M
~
-o-
~-
~ >¦
-o
0-
0-
o
o
£2 1
= i
a 3
o o>
3
>
3
3
SL
CO
C
Q.
O
0
I' O |
CD
Z r~ O
<
O O o
n
> £ -¦
r~
m rn =>
1— 1 CQ
ZD
Q)
3
CQ
(D
-------
aionO "ao 3ii3 iom oa - uvaa
P6~P
Oral Exposure
(mg/kg-day)
era
c
re
J-
en
n
»
s
a
a
§
3
P
-t
22
m
¦
K>
O
CD
O
00
O
^1
O
CD
0
01
O
-N
Moccarelli etal. 2008
Baccarelli etal. 2008
Alaluusua etal. 2004
Li et al. 2006
Kuchiiwa etal. 2002
Smialowicz etal. 2008
Bowman etal. 1989, etc.
Kelleretal. 2007a,
Toth et al. 1979
Latch. & Mathur 2002
NTP1982
White et al. 1986
Li etal. 1997
DeCaprio etal. 1986
Shiet al. 2007
Markowski etal. 2001
Hojo etal. 2002
Cantoni etal. 1981
Vosetal. 1973
Miettinen etal. 2006
Kattainen etal. 2001
NTP2006
Amin etal. 2000
Schantz et al. 1996
Huttet al. 2008
Ohsako et al. 2001
Murray etal. 1979
Franczaket al. 2006
Chu et al. 2007
Bell et al. 2007
Ishihara etal. 2007
VanBirgelen etal. 1995a
Kociba etal. 1978
Fattore et al. 2000
Seo et al. 1995
Crofton etal. 2005
Sewall etal. 1995
Franc et al. 2001
Kociba etal. 1976
Sparschu etal. 1971
Moccarelli etal. 2008
Baccarelli etal. 2008
I I
Alaluusua etal.2004
Li etal. 2006
^ Kuchiiwa etal. 2002
^ Smialowiczetal.2008
I I
Bowman etal. 1989, etc
I I
Kelleretal. 2007a, 2008ab
Toth etal. 1979
^ Latch. & Mathur 2002
* NTP1982
sss
^ White etal. 1986
L eta .1997
DeCaprioetal. 1986
I I
AVWf Shi etal. 2007
Markowski etal. 2001
rn Hojo etal. 2002
W Cantoni etal. 1981
IVVVVf Vosetal. 1973
Miettinen etal. 2006
i i
Kattainen etal. 2001
NTP2006
^ Amin etal. 2000
Schantz etal. 1996
^ Huttetal. 2008
E55S5S
l + Ohsako etal. 2001
\\\\^ Murray etal. 1979
^ Franczaketal.2006
Chue a .2007
Bell etal. 2007
»
shiharaeta .2007
^ VanBirgelen etal. 1995a
WV^ Kociba etal. 1978
vvvl
Fattore etal. 2000
Seo etal. 1995
Crofton etal. 2005
i i
Sewall etal. 1995
I I
Franc etal. 2001
I I
Kociba etal. 1976
iwyv» Sparschu etal. 1971
-------
Reported Exposure
68 ppt TCDD
vo
O
o
o
2
o
H
O
HH
H
W
O
V
o
c
o
H
W
Iskenazi'
Needham
Background
Exposure
Total TEQ
93.7 ppt
0.0112 ng/kg-d
TCDD
40.5 ppt
DLC-TEQ 3.5xl0"3 ng/kg-d
added
3.2xl0~3 ng/kg-d
Total TEQ
TCDD
15 ppt
3.5xl04 ng/kg-d
Modeled
3.5xl0~3 ng/kg-d
40.4 ppt
Modeled
93.7 ppt
0.0112ngtg-d
Exposure
Duration
48 hours
24 hours
Measurement
Lag
1 month
6 months
1 year
Age at
Exposure
6.2 years
1 year 6.2 years 9 years
P = 0.0294
W= 0.0101
AVG = 0.0198
P = 0.0369
W= 0.00872
AVG = 0.0228
P = 0.0225
W= 0.00796
AVG = 0.0152
P = 0.0349
P = 0.0267
W= 0.00709 W= 0.00362
AVG = 0.0169 AVG = 0.0193
P = 0.0512
W= 0.0159
AVG = 0.0335
P = 0.0353
W=0.0111
AVG = 0.0232
P = 0.0134
W= 0.0103
AVG = 0.0118
P = 0.0197 P = 0.0225
W= 0.0246 W= 0.0194
AVG = 0.0221 AVG = 0.0209
6 months
24 hours
6.2 years
P= 0.0321
W = 0.00797
AVG = 0.0201
Figure 4-6. Sensitivity tree showing TCDD exposure-variable uncertainty for Mocarelli et al. (2008).
-------
Effect
Elevated Neonatal TSH Levels
POD
Basis
Background
Exposure
POD Method
Maternal LASC
Age at
Conception
POD (ng/kg-day)
NOAEL
(Equivalnet LOAEL)
TCDD Only
Graphical
Method
40ppt
30
years
Graphical
Method
90 ppt
30
years
0.00161 0.00514
(0.0161)
Graphical
Method
312 ppt
30
years
I
0.0303
LOAEL
TCDD Only
Total TEQ
excludingNon- Total TEQ
CoplanarPCBs
Regression
Model
235 ppt
30 45
years years
i \
0.0196 0.0162
Regression
Model
219 ppt
30
years
I
0.0180
Regression
Model
485 ppt
30
years
I
0.0593
Figure 4-7. Sensitivity tree showing TCDD exposure-variable uncertainty for Baccarelli et al. (2008).
-------
Approach
Rodent
Kinetic
Model
Dose
Metric
Background
Exposure
Human
Kinetic
Model
LOAEL
Modeling
Administered Doses
Measured Tissue
Concentrations
First Order
Body Burden
Whole Body
Concentration
Emond
Whole Blood
Concentration
TCDD Only
57.4
First
Order
Kinetics
TCDD Only
2.56 ng/kg
TotalTEQ
2.75 ng/kg
Lipid-Adjusted
Serum
Concentration
TCDD Only
1408 ng/kg
CADM
Adipose
Concentration
Adip ose Concentration
TCDD Only TCDD Only Total TEQ
302 ng/kg 505 ng/kg 529
Emond Human PBPK Model
0.023
0.14
0.15
0.31
0.19
0.41
0.44
Figure 4-8. Sensitivity tree showing TCDD exposure-variable uncertainty for NTP (2006a).
-------
o
oc
a
&
H
~
O
o
H
O
HH
H
m
o
&
o
a
o
H
m
l.E+03
l.E+02
^ l.E+01
diD
it i E+00
a Neonatal
g' TSH
S l.E-01
M
W
l.E-02
l.E-03
1E-04
l.E-05
O
<> o
Q
cy
Q
cy
p
cy
0
o
P
Q
w
9
w
Q
w
Q
M
P
O
H
H
O
f—1
O
H
O
H
CO
CO
o
H
^r
o
H
r i
n
o
H
o
H
o
o
o
03
O
Id
55
03
o
53
r/5
w
O
O
o
o
W
—1
s
Candidate RfD
TEQ Exposure Range
A NOAEL
RfD NOAEL Equiv alent
RfD
» TCDD Exp osure Range
LOAEL
" RfD LOAEL
P=Peak,W=Window Ave.
-Sex-
Ratio
o
Semen
Quality
Age at
Menopause
Age at
Menaiche
A A
< >
A
< >
A
P
W
AW'
W -
Uncertainty Analysis Only
Figure 4-9. Alternative POD exposure-response array.
W = critical window average, P = peak exposure.
Ovarian
Function
cy
Q
a
Q
a
Q
a
w
Q
w
Q
w
Q
w
H
O
H
O
H
O
H
H
H
*=r
H
r-
o
o
CI
i/~,
O
o
CI
^r
o
o
o
M
r-
o
o
o
o
i
Ol
.—i
c-i
;
CI
j
03
03
cz
CC
Id
03
Id
~CZ
"CD
03
15
Id
¦R
CD
15
CD
Id
«
C3
cc
CD
CD
r-1
(03
a>
o
o
CD
in
£
tcc
1
3
-^1
72
w
W
1
£
-------
5. REFERENCES
Abbott BP; Birnbaum. LS; Diliberto. JJ. (1996). Rapid Distribution of 2,3,7,8-
Tetrachlorodibenzo-p-dioxin (TCDD) to Embryonic Tissues in C57BL/6N Mice and
Correlation with Palatal Uptakein Vitro. Toxicol Appl Pharmacol 141: 256-263.
Abraham. K; Krowke. R; Neubert. D. (1988). Pharmacokinetics and biological activity of
2,3,7,8-tetrachlorodibenzo-p-dioxin. 1. Dose-dependent tissue distribution and induction
of hepatic ethoxyresorufin O-deethylase in rats following a single injection. Arch Toxicol
62: 359-368.
Abraham. K; Knoll. A: Ende. M; Papke. O: Helge. H. (1996). Intake, fecal excretion, and body
burden of polychlorinated dibenzo-p-dioxins and dibenzofurans in breast-fed and
formula-fed infants. Pediatr Res 40: 671-679.
Abraham. K; Geusau. A; Tosun. Y; Helge, H; Bauer. S; Brockmoiler. J. (2002). Severe 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) intoxication: insights into the measurement of
hepatic cytochrome P450 1A2 induction. Clin Pharmacol Ther 72: 163-174.
http://dx.doi.org/10.1067/mcp.2002.126408.
Akhtar. FZ; Garabrant. DH; Ketchum. NS: Michalek. JE. (2004). Cancer in US Air Force
veterans of the Vietnam war. J Occup Environ Med 46: 123.
Alaluusua. S; Calderara. P; Gerthoux. PM; Lukinmaa. PL; Kovero. O; Needham. L; Patterson. J,
r. D. G.; Tuomisto. J: Mocarelli. P. (2004). Developmental dental aberrations after the
dioxin accident in Seveso. Environ Health Perspect 112: 1313-1318.
Am in. S: Moore. RW: Peterson. RE: Schantz. SL. (2000). Gestational and lactational exposure to
TCDD or coplanar PCBs alters adult expression of saccharin preference behavior in
female rats. Neurotoxicol Teratol 22: 675-682. http://dx.doi.org/10.1016/S0892-
0362(00)00094-5.
Anbalagan. J: Sashi. AM: Vengatesh. G: Stanley. J A: Neelamohan. R; Aruldhas. MM. (2010).
Mechanism underlying transient gestational-onset hypothyroidism-induced impairment of
posttesticular sperm maturation in adult rats. Fertil Steril 93: 2491-2497.
Andersen. ME: Mills. JJ: Gargas. ML: Kedderis. L; Birnbaum. LS: Neubert. D; Greenlee. WF.
(1993). Modeling receptor-mediated processes with dioxin: Implications for
pharmacokinetics and risk assessment. Risk Anal 13: 25-36.
http://dx.doi.org/10.1111/i. 1539-6924.1993 ,tb00726.
Andersen. ME; Birnbaum. LS; Barton. HA; Eklund. CR. (1997). Regional hepatic CYP1A1 and
CYP1A2 induction with 2,3,7,8-tetrachlorodibenzo-p-dioxin evaluated with a
multicompartment geometric model of hepatic zonation. Toxicol Appl Pharmacol 144:
145-155. http://dx.doi.org/10.1006/taap. 1996.8067.
ATSDR (Agency for Toxic Substances and Disease Registry). (1998). Toxicological profile for
chlorinated dibenzo-p-dioxins (CDDs). Atlanta, GA.
http://www.atsdr.cdc,gov/toxprofiles/tp 104.pdf.
Auso. E; Lavado-Autric. R; Cuevas. E; Del Rev. FE; Morreale De Escobar. G; Berbel. P. (2004).
A moderate and transient deficiency of maternal thyroid function at the beginning of fetal
neocorticogenesis alters neuronal migration. Endocrinology 145: 4037-4047.
http://dx.doi.org/10.1210/en.2004-0274.
5-1
DRAFT - DO NOT CITE OR QUOTE
-------
Avlward. LL; Havs. SM. (2002). Temporal trends in human TCDD body burden: Decreases over
three decades and implications for exposure levels. J Expo Anal Environ Epidemiol 12:
319-328.
Avlward. LL; Brunei RC; Carrier. G; Hays. SM; Cushing, CA; Needham. LL; Patterson. DG;
Gerthoux. PM; Brambilla. P; Mocarelli. P. (2005a). Concentration-dependent TCDD
elimination kinetics in humans: Toxicokinetic modeling for moderately to highly exposed
adults from Seveso, Italy, and Vienna, Austria, and impact on dose estimates for the
NIOSH cohort. J Expo Anal Environ Epidemiol 15: 51-65.
http://dx.doi.org/10.1038/si .iea.7500370.
Avlward. LL; Brunet, RC; Starr. TB; Carrier. G; Delzell. E; Cheng. H; Beall. C. (2005b).
Exposure reconstruction for the TCDD-exposed NIOSH cohort using a concentration-
and age-dependent model of elimination. Risk Anal 25: 945-956.
http://dx.doi.org/10.1111/i. 1539-6924.2005.00645.x.
Avlward. LL; Goodman. IE; Charnlev. G; Rhomberg. LR. (2008). A margin-of-exposure
approach to assessment of noncancer risks of dioxins based on human exposure and
response data. Environ Health Perspect 116: 1344-1351.
http://dx.doi.org/10.1289/ehp.11514.
Baccarelli. A; Mocarelli. P; Patterson. DG. Jr; Bonzini. M; Pesatori. AC; Caporaso. N; Landi.
MT. (2002). Immunologic effects of dioxin: new results from Seveso and comparison
with other studies. Environ Health Perspect 110: 1169-1173.
Baccarelli. A; Pesatori. AC; Masten. SA; Patterson. DG. Jr; Needham. LL; Mocarelli. P;
Caporaso, NE; Consonni. D; Grassman. J A; Bertazzi. PA; MT. L. (2004). Aryl-
hydrocarbon receptor-dependent pathway and toxic effects of TCDD in humans: a
population-based study in Seveso, Italy. Toxicol Lett 149: 287-293.
http://dx.doi.org/10.1016/i .toxlet.2003.12.062.
Baccarelli. A; Pesatori. AC; Consonni. D; Mocarelli. P; Patterson. DG. Jr; Caporaso. NE;
Bertazzi. PA; Landi. MT. (2005). Health status and plasma dioxin levels in chloracne
cases 20 years after the Seveso, Italy accident. Br J Dermatol 152: 459-465.
http://dx.doi.org/10.1111 /J. 1365-2133.2005.06444. X.
Baccarelli. A; Hirt. C; Pesatori. AC; Consonni. D; Patterson. DG. Jr; Bertazzi. PA; Pol ken. G;
Landi. MT. (2006). t(14; 18) translocations in lymphocytes of healthy dioxin-exposed
individuals from Seveso, Italy. Carcinogenesis 27: 2001-2007.
http://dx.doi.org/10.1093/carcin/bglQ 1 1.
Baccarelli. A; Giacomini. SM; Corbetta. C; Landi. MT; Bonzini. M; Consonni. D; Grillo. P;
Patterson. DG; Pesatori. AC; Bertazzi. PA. (2008). Neonatal thyroid function in Seveso
25 years after maternal exposure to dioxin. PLoS Med 5: el61.
Banks. YB; Brewster. DW; Birnbaum. LS. (1990). Age-related changes in dermal absorption of
2,3,7,8-tetrachlorodibenzo-p-dioxin and 2,3,4,7,8-pentachlorodibenzofuran. Fundam
Appl Toxicol 15: 163-173.
Banks. YB; Birnbaum. LS. (1991). Absorption of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)
after low dose dermal exposure. Toxicol Appl Pharmacol 107: 302-310.
http://dx.doi.org/10.1016/0041 -008X(91 )90210-6.
Baron. JM; Zwadio-Klarwasser. G; Jugert. F; Hamann. W; Riibben. A; Mukhtar. H; Merk. HF.
(1998). Cytochrome P450 1B1: A major P450 isoenzyme in human blood monocytes and
macrophage subsets. Biochem Pharmacol 56: 1105-1110.
http://dx.doi.Org/l 0.1016/S0006-2952(98)00105-1.
5-2
DRAFT - DO NOT CITE OR QUOTE
-------
Becher. H; Flesch-Janvs. D. (1998). Dioxins and furans: Epidemiologic assessment of cancer
risks and other human health effects. Environ Health Perspect 106: 623-624.
Becher. H; Steindorf. K; Flesch-Janvs. D. (1998). Quantitative cancer risk assessment for dioxins
using an occupational cohort. Environ Health Perspect 106: 663-670.
Bell. DR.; CI ode. S: Fan. MO; Fernandes. A: Foster. PM; Jiang. T; Loizou. G: MacNicoll. A:
Miller. BG; Rose. M; Tran. L; White. S. (2007a). Relationships between tissue levels of
2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), mRNAs and toxicity in the developing
male Wistar(Han) rat. Toxicol Sci 99: 591-604. http://dx.doi.oru/10.1093/toxsci/kfm 179.
Bell. DR.; CI ode. S; Fan. MO; Fernandes. A; Foster. PM; Jiang. T; Loizou. G; MacNicoll. A;
Miller. BG; Rose. M; Tran. L; White. S. (2007b). Toxicity of 2,3,7,8-tetrachlorodibenzo-
p-dioxin in the developing male Wistar(Han) rat. II: Chronic dosing causes
developmental delay. Toxicol Sci 99: 224-233. http://dx.doi.org/10.1093/toxsci/kfm 141.
Bernert. JT; Turner. WE; Patterson. DG; Needham. LL. (2007). Calculation of serum total lipid
concentrations for the adjustment of persistent organohalogen toxicant measurements in
human samples. Chemosphere 68: 824-831.
http://dx.doi. org/ 10.1016/j. chemosphere.2007.02.043.
Birnbaum. LS. (1986). Distribution and excretion of 2,3,7,8-tetrachlorodibenzo-p-dioxin in
congenic strains of mice which differ at the Ah locus. Drug Metab Dispos 14: 34-40.
Bonde. J. . P. E. Ernst. E... Jensen. T... K. Hiollund. N... H. I. Kolstad. H... Scheike. T...
Giwercman. A... Skakkebask. N... E. Henriksen. T... B. 01 sen. J... (1998). Relation
between semen quality and fertility: A population-based study of 430 first-pregnancy
planners. 352: 1172-1177. http://dx.doi.org/10.1016/sO 140-6736(97) 105 14-1.
Boverhoff. PR; Burgoon. LP; Tashiro. C; Chittim. B; Harkema. JR; Jump. DB; Zacharewski.
TR. (2005). Temporal and dose-dependent hepatic gene expression patterns in mice
provide new insights into TCDD-mediated hepatotoxicity. Toxicol Sci 85: 1048-1063.
http://dx.doi.org/10.1093/toxsci/kfi 162.
Bowman. RE; Schantz. SL; Gross. ML; SA. F. (1989a). Behavioral effects in monkeys exposed
to 2,3,7,8-TCDD transmitted maternally during gestation and for four months of nursing.
Chemosphere 18: 235-242.
Bowman. RE; Schantz. SL; Weerasinghe. NIC A; Gross. ML; Barsotti. DA. (1989b). Chronic
dietary intake of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) at 5 or 25 parts per trillion
in the monkey: TCDD kinetics and dose-effect estimate of reproductive toxicity.
Chemosphere 18: 243-252.
Budinsky. RA; Paustenbach. D; Fontaine. D; Landenberger. B; Starr. TB. (2006). Recommended
relative potency factors for 2,3,4,7,8 pentachlorodibenzofuran: The impact of different
dose metrics. Toxicol Sci 91: 275-285. http://dx.doi.org/10.1093/toxsci/kfi 125.
Budinsky. RA; LeCluvse. EL; Ferguson. SS; Rowlands. JC; Simon. T. (2010). Human and rat
primary hepatocyte CYP1A1 and 1A2 induction with 2,3,7,8-tetrachlorodibenzo-p-
dioxin, 2,3,7,8-tetrachlorodibenzofuran, and 2,3,4,7,8-pentachlorodibenzofuran. Toxicol
Sci 1 18: 224-235. http://dx.doi.org/10.1093/toxsci/kfci238.
Buelke-Sam. J; Nelson. CJ; Bvrd. RA; Hoi son. JF. (1982a). Blood flow during pregnancy in the
rat: I Flow patterns to maternal organs. Teratology 26: 269-277.
Buelke-Sam. J; Hoi son. JF; Nelson. CJ. (1982b). Blood flow during pregnancy in the rat: II
Dynamics of and litter variability in uterine flow. Teratology 26: 279-288.
5-3
DRAFT - DO NOT CITE OR QUOTE
-------
Burleson. GR; Lebrec. H; Yang. YG; Ibanes. JD; Pennington. KN; Birnbaum. LS. (1996). Effect
of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on influenza vims host resistance in
mice. Fundam Appl Toxicol 29: 40-47.
Calvo. RM; Jauniaux. E; Gulbis. B; Asuncion. M; Gervy. C; Contempre, B; Morreale De
Escobar. G. (2002). Fetal tissues are exposed to biologically relevant free thyroxine
concentrations during early phases of development. J Clin Endocrinol Metab 87: 1768-
1777.
Cantoni. L; Salmona. M; Rizzardini. M. (1981). Porphyrogenic effect of chronic treatment with
2,3,7,8-tetrachlorodibenzo-p-dioxin in female rats. Dose-effect relationship following
urinary excretion of porphyrins. Toxicol Appl Pharmacol 57: 156-163.
Carrier. G: Brunet. RC: Brodeur. J. (1995a). Modeling of the toxicokinetics of polychlorinated
dibenzo-p-dioxins and dibenzofurans in mammalians, including humans. I. Nonlinear
distribution of PCDD/PCDF body burden between liver and adipose tissues. Toxicol
Appl Pharmacol 131: 253-266. http://dx.doi.org/10.1006/taap. 1995.1068.
Carrier. G: Brunet. RC; Brodeur. J. (1995b). Modeling of the toxicokinetics of polychlorinated
dibenzo-p-dioxins and dibenzofurans in mammalians, including humans: II. Kinetics of
absorption and disposition of PCDDs/PCDFs. Toxicol Appl Pharmacol 131: 267-276.
http://dx.doi.org/10.1006/taap. 1995.1069.
Chan. SY; Franklvn. J A; Kilbv. MP. (2005). Maternal thyroid hormones and fetal brain
development. Curr Opinion Endocrinol Diab 12: 23-30.
Cheng. H; Avlward. L; Beall. C: Starr. TB; Brunet. RC; Carrier. G: Delzell. E. (2006). TCDD
exposure-response analysis and risk assessment. Risk Anal 26: 1059-1071.
http://dx.doi.org/10.1111/i. 1539-6924.2006.00800.X.
Chu. I; Lecavalier. P; Hakansson. H; Yagminas. A; Valii. YE; Poon. P; M. F. (2001). Mixture
effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin and polychlorinated biphenyl congeners in
rats. Chemosphere 43: 807-814. http://dx.doi.org/10.1016/S0045-6535(00)00437-9.
Chu. I; Valii. VE; Rousseaux. CG. (2007). Combined effects of 2,3,7,8-tetrachlorodibenzo-p-
dioxin and polychlorinated biphenyl congeners in rats. Toxicol Environ Chem 89: 71-87.
http://dx.doi.org/10.1080/0277224060Q942548.
Clewell. HI; Gentry. PR; Covington. TR; Sarangapani. R; Teeguarden. JG. (2004). Evaluation of
the potential impact of age- and gender-specific pharmacokinetic differences on tissue
dosimetry. Toxicol Sci 79: 381-393. http://dx.doi.org/10.1093/toxsci/kfh 109.
Collins. JJ; Bodner. K; Avlward. LL; Wilken. M; Bodnar. CM. (2009). Mortality rates among
trichlorophenol workers with exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Am J
Epidemiol 170: 501-506. http://dx.doi.org/10.1093/aie/kwp 153.
Connor. KT; Avlward. LL. (2006). Human response to dioxin: Aryl hydrocarbon receptor (AhR)
molecular structure, function, and dose-response data for enzyme induction indicate an
impaired human AhR. J Toxicol Environ Health B Crit Rev 9: 147-171.
http://dx.doi.org/10.1080/1528739050Q196487.
Connor. KT; Harris. MA; Edwards, MR; Budinsky. RA; Clark. GC; Chu. AC; Finlev. BL;
Rowlands. JC. (2008). AH receptor agonist activity in human blood measured with a cell-
based bioassay: evidence for naturally occurring AH receptor ligands in vivo. J Expo Sci
Environ Epidemiol 18: 369-380. http://dx.doi.org/10.1038/si.ies.7500607.
Cooper. TG; Noonan. E; von Eckardstein. S; Auger. J; Baker. HW; Behre. HM; Haugen. TB;
Kruger. T; Wang. C; Mbizvo. MT; Vogelsong. KM. (2010). World Health Organization
5-4
DRAFT - DO NOT CITE OR QUOTE
-------
reference values for human semen characteristics. Hum Reprod Update 16: 231-245.
http://dx.doi.org/10.1093/humupd/dmpQ48.
Crofton. KM: Craft. ES: Hedge. JM; Gennings. C; Simmons. JE; Carchman. RA; Carter. WH, Jr;
DeVito. MJ. (2005). Thyroid-hormone-disrupting chemicals: Evidence for dose-
dependent additivity or synergism. Environ Health Perspect 113: 1549-1554.
Croutch. CR; Lebofskv. M; Schramm. KW: Terranova. PF; Rozman. KK. (2005). 2,3,7,8-
Tetrachlorodibenzo-p-dioxin (TCDD) and 1,2,3,4,7,8-hexachlorodibenzo-p-dioxin
(HxCDD) alter body weight by decreasing insulin-like growth factor I (IGF-I) signaling.
Toxicol Sci 85: 560-571. http://dx.doi.org/10.1093/toxsci/kfi 106.
DeCaprio, AP; McMartin, DN; O'Keefe. PW; Rei. R; Silkworth. IB; Kaminsky. LS. (1986).
Subchronic oral toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the guinea pig:
Comparisons with a PCB-containing transformer fluid pyrolysate. Fundam Appl Toxicol
6: 454-463. http://dx.doi.org/10.1093/toxsci/6.3.454.
DeKoning. EP; Karmaus. W. (2000). PCB exposure in utero and via breast milk. A review
[Review], J Expo Anal Environ Epidemiol 10: 285-293.
Delange. F; Bourdoux, P; Ketelbant-Balasse. P; Humskerken. AY: Glinoer. D; Ermans. AM.
(1983). Transient primary hypothyroidism in the newborn. In JH Dussault; P Walker
(Eds.), Congenital hypothyroidism (pp. 275-301). New York, NY: Marcel Dekker.
Delia Porta. G: Dragani. TA; Sozzi, G. (1987). Carcinogenic effects of infantile and long-term
2,3,7,8-tetrachlorodibenzo-p-dioxin treatment in the mouse. Tumori 73: 99-107.
DeVito. MJ; Ma. X: Babish. JG: Menache. M; Birnbaum. LS. (1994). Dose-response
relationships in mice following subchronic exposure to 2,3,7,8-tetrachlorodibenzo-p-
dioxin: CYP1A1, CYP1A2, estrogen receptor, and protein tyrosine phosphorylation.
Toxicol Appl Pharmacol 124: 82-90.
Diliberto. JJ; Akubue. PI; Luebke. RW: Birnbaum. LS. (1995). Dose-response relationships of
tissue distribution and induction of CYP1A1 and CYP1A2 enzymatic activities following
acute exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in mice. Toxicol Appl
Pharmacol 130: 197-208. http://dx.doi.org/10.1006/taap. 1995.1025.
Diliberto. JJ; Jackson. J A; Birnbaum. LS. (1996). Comparison of 2,3,7,8-Tetrachlorodibenzo-p-
dioxin (TCDD) Disposition Following Pulmonary, Oral, Dermal, and Parenteral
Exposures to Rats. Toxicol Appl Pharmacol 138: 158-168.
Diliberto. JJ; Burgin. DE; Birnbaum. LS. (1997). Role of CYP1A2 in hepatic sequestration of
dioxin: Studies using CYP1A2 knock-out mice. Biochem Biophys Res Commun 236:
431-433. http://dx.doi.org/10.1006/bbrc. 1997.6973.
Diliberto. JJ; Burgin. DE; Birnbaum. LS. (1999). Effects of CYP 1A2 on Disposition of 2,3,7,8-
Tetrachlorodibenzo-p-dioxin, 2,3,4,7,8-Pentachlorodibenzofuran, and 2,2',4,4',5,5'-
Hexachlorobiphenyl in CYP1A2 Knockout and Parental (C57BL/6N and 129/Sv) Strains
of Mice. Toxicol Appl Pharmacol 159: 52-64. http://dx.doi.org/10.1006/taap. 1999.8720.
Diliberto. JJ; DeVito. MJ; Ross. DG; Birnbaum. LS. (2001). Subchronic Exposure of [3H]-
2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in female B6C3F1 mice: Relationship of
steady-state levels to disposition and metabolism. Toxicol Sci 61: 241-255.
Dimitropoulos. A; Molinari. L; Etter. K; Lang-Muritano. M; Jenni. OG; Largo. RH; Latal. B.
(2009). Children with congenital hypothyroidism: Long-term intellectual outcome after
early high-dose treatment. Pediatr Res 65: 242-248.
5-5
DRAFT - DO NOT CITE OR QUOTE
-------
Eastin. WC; Haseman. JK; Mahler. JF; Bucher. JR. (1998). The National Toxicology Program
evaluation of genetically altered mice as predictive models for identifying carcinogens.
Toxicol Pathol 26: 461-473.
Emond. C; Birnbaum, LS; DeVito. MJ. (2004). Physiologically based pharmacokinetic model for
developmental exposures to TCDD in the rat. Toxicol Sci 80: 115-133.
http://dx.doi. org/10.1093/toxsci/kfh 117kfh 117.
Emond. C: Michalek. JE; Birnbaum. LS: DeVito. MJ. (2005). Comparison of the use of a
physiologically based pharmacokinetic model and a classical pharmacokinetic model for
dioxin exposure assessments. Environ Health Perspect 113: 1666-1668.
Emond, C; Birnbaum, LS: DeVito. MJ. (2006). Use of a physiologically based pharmacokinetic
model for rats to study the influence of body fat mass and induction of CYP1A2 on the
pharmacokinetics of TCDD. Environ Health Perspect 114: 1394-1400.
Eskenazi. B; Mocarelli. P; Warner, M; Samuels. S: Vercellini. P; Olive. D; Needham. LL;
Patterson DG. J. r; Brambilla. P; Gavoni. N: Casalini. S: Panazza. S: Turner. W:
Gerthoux. PM. (2002a). Serum dioxin concentrations and endometriosis: A cohort study
in Seveso, Italy. Environ Health Perspect 110: 629-634.
Eskenazi. B; Warner. M; Mocarelli. P; Samuels. S: Needham. LL; Patterson. DG. Jr; Lippman.
S: Vercellini. P; Gerthoux. PM; Brambilla. P; Olive. D. (2002b). Serum dioxin
concentrations and menstrual cycle characteristics. Am J Epidemiol 156: 383-392.
Eskenazi. B; Mocarelli. P; Warner. M; Chee. WY; Gerthoux. PM; Samuels. S; Needham. LL;
Patterson. DG. Jr. (2003). Maternal serum dioxin levels and birth outcomes in women of
Seveso, Italy. Environ Health Perspect 111: 947-953.
Eskenazi. B; Mocarelli. P; Warner. M; Needham. L; Patterson. DG. Jr; Samuels. S; Turner. W;
Gerthoux. PM; Brambilla. P. (2004). Relationship of serum TCDD concentrations and
age at exposure of female residents of Seveso, Italy. Environ Health Perspect 112: 22-27.
http://dx.doi.org/10.1289/ehp.6573.
Eskenazi. B; Warner. M; Marks. AR; Samuels. S; Gerthoux. PM; Vercellini. P; Olive. PL;
Needham. L; Patterson. D. Jr; Mocarelli. P. (2005). Serum dioxin concentrations and age
at menopause. Environ Health Perspect 113: 858-862.
Eskenazi. B; Warner. M; Samuels. S; Young. J; Gerthoux. PM; Needham. L; Patterson. D; Olive.
D; Gavoni. N; Vercellini. P; Mocarelli. P. (2007). Serum dioxin concentrations and risk
of uterine leiomyoma in the Seveso Women's Health Study. Am J Epidemiol 166: 79-87.
http://dx.doi.org/10.1093/aie/kwm048.
Evans. JS; Graham. JD; Gray. GM; Sielken. RL, Jr. (1994a). A distributional approach to
characterizing low-dose cancer risk. Risk Anal 14: 25-34.
Evans. JS; Gray. GM; Sielken. RL; Smith. AE; Valdez-Flores. C; Graham. JD. (1994b). Use of
probabilistic expert judgment in uncertainty analysis of carcinogenic potency. Regul
Toxicol Pharmacol 20: 15-36.
Fattore. E; Trossvik. C; Hakansson. H. (2000). Relative potency values derived from hepatic
vitamin A reduction in male and female Sprague-Dawley rats following subchronic
dietary exposure to individual polychlorinated dibenzo-p-dioxin and dibenzofuran
congeners and a mixture therof. Toxicol Appl Pharmacol 165: 184-194.
http://dx.doi.org/10.1006/taap.2000.8943.
Ferribv. LL; Knut sen. JS; Harris. M; Unice. KM; Scott. P; Nonv. P; Haws. LC; Paustenbach. D.
(2007). Evaluation of PCDD/F and dioxin-like PCB serum concentration data from the
5-6 DRAFT - DO NOT CITE OR QUOTE
-------
2001-2002 National Health and Nutrition Examination Survey of the United States
population. J Expo Sci Environ Epidemiol 17: 358-371.
Fisher. JW; Whittaker. TA; Taylor. DH; Clewell. HJ. Ill; Andersen. ME. (1989). Physiologically
based pharmacokinetic modeling of the pregnant rat: A multiroute exposure model for
trichloroethylene and its metabolite, trichloroacetic acid. Toxicol Appl Pharmacol 99:
395-414. http://dx.doi.org/10.1016/0041 -008X(89)90149-X.
Flesch-Janvs. D; Becher. H; Gurn. P; Jung. D; Konietzko. J: Manz. A: Papke. O. (1996).
Elimination of polychlorinated dibenzo-p-dioxins and dibenzofurans in occupationally
exposed persons. J Toxicol Environ Health 47: 363-378.
Flesch-Janvs. D; Steindorf. K; Gurn. P; Becher. H. (1998). Estimation of the cumulated exposure
to polychlorinated dibenzo-p-dioxins/furans and standardized mortality ratio analysis of
cancer mortality by dose in an occupationally exposed cohort. Environ Health Perspect
106: 655-662.
Fox. TR; Best. LL; Goldsworthv. SM; Mills. JJ; Goldsworthv. TL. (1993). Gene expression and
cell proliferation in rat liver after 2,3,7,8-tetrachlorodibenzo-p-dioxin exposure. Cancer
Res 53: 2265-2271.
Franc. MA; Pohianvirta. R; Tuomisto. J: Okev. AB. (2001). Persistent, low-dose 2,3,7,8-
tetrachlorodibenzo-p-dioxin exposure: effect on aryl hydrocarbon receptor expression in
a dioxin-resistance model. Toxicol Appl Pharmacol 175: 43-53.
http://dx.doi.org/10.1006/taap.2001.9222.
Franczak. A; Nvnca. A; Valdez. KE; Mizinga. KM; Petroff. BK. (2006). Effects of acute and
chronic exposure to the aryl hydrocarbon receptor agonist 2,3,7,8-tetrachlorodibenzo-p-
dioxin on the transition to reproductive senescence in female Sprague-Dawley rats. Biol
Reprod 74: 125-130. http://dx.doi.org/10.1095/biolreprod. 105.044396.
Fuiii-Kurivama. Y; Ema. M; Mimura. J; Matsushita. N; Sogavva. K. (1995). Polymorphic forms
of the Ah receptor and induction of the CYP1A1 gene. Pharmacogenetics 5 (S): 149-153.
Gasiewicz. TA; Henry. EC; Collins. LL. (2008). Expression and activity of aryl hydrocarbon
receptors in development and cancer. Crit Rev Eukaryot Gene Expr 18: 279-321.
Geusau. A; Abraham. K; Geissler. K; Sator. MO; Stingl. G; Tschachler. E. (2001). Severe
2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) intoxication: Clinical and laboratory effects.
Environ Health Perspect 109: 865-869. http://dx.doi.org/10.1067/mcp.2002.126408.
Geusau. A; Schmaldienst. S; Derfler. K. (2002). Severe 2,3,7,8-tetrachlorodibenzo-p-dioxin
(TCDD) intoxication: Kinetics and trials to enhance elimination in two patients. Arch
Toxicol 76: 316-325. http://dx.doi.org/10.1007/s00204-002-Q345-7.
Gever. H; Scheunert. I; Korte. F. (1986). Bioconcentration potential of organic environmental
chemicals in humans. Regul Toxicol Pharmacol 6: 313-347.
http://dx.doi .org/10.1016/0273 -2300(86)90002-4.
Glinoer. D; Delange. F. (2000). The potential repercussions of maternal, fetal, and neonatal
hypothyroxinemia on the progeny. Thyroid 10: 871-887.
http://dx.doi.org/10.1089/thy.2000.10.871.
Goodman. DG; Sauer. RM. (1992). Hepatotoxicity and carcinogenicity in female Sprague-
Dawley rats treated with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD): a pathology
working group reevaluation. Regul Toxicol Pharmacol 15: 245-252.
Goodman. M; Squibb. K; Youngstrom. E; Anthony. LG; Kenworthv. L; Lipkin. PH; Mattison.
DR.; Lakind. JS. (2010). Using systematic reviews and meta-analyses to support
5-7
DRAFT - DO NOT CITE OR QUOTE
-------
regulatory decision making for neurotoxicants: Lessons learned from a case study of
PCBs. Environ Health Perspect 118: 727-734. http://dx.doi.org/10.1289/ehp.09Q1835.
Grassman. J A; Needham. LL; Masten. SA; Patterson. D; Portier. CJ: Lucier. GW: Walker. NJ.
(2000). Evidence of hepatic sequestration of dioxin in humans? An examination of tissue
levels and CYP1A2 expression. Organohalogen Compounds 48: 87-90.
Greer. MA; Goodman. G: Pleus. RC: Greer. SE. (2002). Health effects assessment for
environmental perchlorate contamination: The dose response for inhibition of thyroidal
radioiodine uptake in humans. Environ Health Perspect 110: 927-937.
Haddovv. JE; Palomaki. GE; Allan. WC: Williams. JR.; Knight. GJ; Gagnon. J: O'Heir. CE;
Mitchell. ML; Hermos. RJ; Waisbren. SE; Faix, JD; Klein. RZ. (1999). Maternal thyroid
deficiency during pregnancy and subsequent neuropsychological development of the
child. N Engl J Med 341: 549-555. http://dx.doi.org/10.1056/NEJM 199908193410801.
Hakk. H; Diliberto. JJ; Birnbaum. LS. (2009). The effect of dose on 2,3,7,8-TCDD tissue
distribution, metabolism and elimination in CYP1A2 (-/-) knockout and C57BL/6N
parental strains of mice. Toxicol Appl Pharmacol 241: 119-126.
Harper. PA; Wong, JY; Lam, MS; Okey. AB. (2002). Polymorphisms in the human AH receptor.
Chem Biol Interact 141: 161-187.
Harrad. S; Wang, Y; Sandaradura. S; Leeds. A. (2003). Human dietary intake and excretion of
dioxin-like compounds. J Environ Monit 5: 224-228. http://dx.doi.org/10.1039/b21 1406b.
Hassoun. EA; Wilt. SC; Devi to. MJ; V an Birgelen. A; A1 sharif. NZ; Birnbaum. LS; Stohs. SJ.
(1998). Induction of oxidative stress in brain tissues of mice after subchronic exposure to
2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol Sci 42: 23-27.
http://dx.doi.org/10.1093/toxsci/42.1.23.
Hassoun. EA; Li. F; Abushaban. A; Stohs. SJ. (2000). The relative abilities of TCDD and its
congeners to induce oxidative stress in the hepatic and brain tissues of rats after
subchronic exposure. Toxicology 145: 103-113.
Hassoun. EA; Wang. H; Abushaban. A; Stohs. SJ. (2002). Induction of oxidative stress
following chronic exposure to TCDD, 2,3,4,7,8-pentachlorodibenzofuran, and
2,3',4,4',5-pentachlorobiphenyl. J Toxicol Environ Health A 65: 825-842.
Hassoun. EA; Al-Ghafri. M; Abushaban. A. (2003). The role of antioxidant enzymes in TCDD-
induced oxidative stress in various brain regions of rats after subchronic exposure. Free
Radic Biol Med 35: 1028-1036. http://dx.doi.org/10.1016/S0891 -5849(03)00458-1.
Hattis. D; Ginsberg. G; Sonawane. B; Smolenski. S; Russ. A; Kozlak. M; Goble. R. (2003).
Differences in pharmacokinetics between children and adults: II. Childrens variability in
drug elimination half-lives and in some parameters needed for physiologically-based
pharmacokinetic modeling. Risk Anal 23: 1 17-142. http://dx.doi.Org/10.l 111/1539-
6924.00295.
Haws. LC; Su. SH; Harris. M; Devi to. MJ; Walker. NJ; Farland. WH; Finlev. B; Birnbaum. LS.
(2006a). Development of a refined database of mammalian relative potency estimates for
dioxin-like compounds. Toxicol Sci 89: 4-30. http://dx.doi.org/10.1093/toxsci/kfi294.
Haws. LC; Su. SH; Harris. M; DeVito. MJ; Walker. NJ; Farland. WH; Finlev. B; Birnbaum. LS.
(2006b). Supplementary database II to article: Development of a refined database of
mammalian relative potency estimates for dioxin-like compounds. Toxicol Sci 89: Sl-
S48.
Hertz-Picciotto. I. (1995). Epidemiology and quantitative risk assessment: a bridge from science
to policy. Am J Public Health 85: 484-491.
5-8
DRAFT - DO NOT CITE OR QUOTE
-------
Hochstein. MS. Jr; Render. JA; Bursian. SJ; Aulerich. RJ. (2001). Chronic toxicity of dietary
2,3,7,8-tetrachlorodibenzo-p-dioxin to mink. Vet Hum Toxicol 43: 134-139.
Hoio. R; Stern. S; Zareba. G; Markowski. VP; Cox. C; Kost. JT; Weiss. B. (2002). Sexually
dimorphic behavioral responses to prenatal dioxin exposure. Environ Health Perspect
110: 247-254.
Holland. CD; Sielken. RL. Jr. (1993). Quantitative cancer modeling and risk assessment.
Hong. R; Taylor. K; Abonour. R. (1989). Immune abnormalities associated with chronic TCDD
exposure in rhesus. Chemosphere 18: 3 13-320. http://dx.doi.org/10.1016/0Q45-
6535(89)90136-7.
Hurst. CH; Abbott. BP; DeVito. MJ; Birnbaum, LS. (1998). 2,3,7,8-Tetrachlorodibenzo-p-
dioxin in Pregnant Long Evans Rats: Disposition to Maternal and Embryo/Fetal Tissues.
45: 129-136.
Hurst. CH; DeVito. MJ; Setzer. RW; Birnbaum. LS. (2000a). Acute administration of 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) in pregnant Long Evans rats: association of
measured tissue concentrations with developmental effects. Toxicol Sci 53: 411-420.
Hurst. CH; DeVito. MJ; Birnbaum, LS. (2000b). Tissue disposition of 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) in maternal and developing long-evans rats
following subchronic exposure. Toxicol Sci 57: 275-283.
Hutt. KJ; Shi. Z; Albertini. DF; Petroff. BK. (2008). The environmental toxicant 2,3,7,8-
tetrachlorodibenzo-p-dioxin disrupts morphogenesis of the rat pre-implantation embryo.
BMC Dev Biol 8: 1-12. http://dx.doi.org/10.1 186/1471 -213X-8-1.
Ikeda. M; Tamura. M; Yamashita. J; Suzuki. C; Tomita. T. (2005). Repeated in utero and
lactational 2,3,7,8-tetrachlorodibenzo-p-dioxin exposure affects male gonads in offspring,
leading to sex ratio changes in F2 progeny. Toxicol Appl Pharmacol 206: 351-355.
http://dx.doi.org/10.1016/i .taap.2004.11.019.
1LS1 (International Life Sciences Institute). (1994). Physiological parameter values for PBPK
models. Washington, DC: U.S. Environmental Protection Agency.
Ishihara. K; Warita. K; Tanida. T; Sugawara. T; Kitagawa. H; Hoshi. N. (2007). Does paternal
exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) affect the sex ratio of offspring.
J Vet Med Sci 69: 347-352.
Kattainen. H; Tuukkanen. J; Simanainen. U; Tuomisto. JT; Kovero. O; Lukinmaa. PL;
Alaluusua. S; Tuomisto. J; Viluksela. M. (2001). In Utero/Lactational 2,3,7,8-
Tetrachlorodibenzo-p-dioxin Exposure Impairs Molar Tooth Development in Rats.
Toxicol Appl Pharmacol 174: 216-224. http://dx.doi.org/10.1006/taap.2001.9216.
Keller. JM; Huet-Hudson. YM; Leamv. LJ. (2007). Qualitative effects of dioxin on molars vary
among inbred mouse strains. Arch Oral Biol 52: 450-454.
http://dx.doi.org/10.1016/i .archoralbio.2006.10.017.
Keller. JM; Huet-Hudson. Y; Leamv. LJ. (2008a). Effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin
on molar development among non-resistant inbred strains of mice: A geometric
morphometric analysis. Growth Development and Aging 71: 3-16.
Keller. JM; Zelditch. ML; Huet. YM; Leamv. LJ. (2008b). Genetic differences in sensitivity to
alterations of mandible structure caused by the teratogen 2,3,7,8-tetrachlorodibenzo-p-
dioxin. Toxicol Pathol 36: 1006-1013. http://dx.doi.org/10.1 177/0192623308327409.
Kerger. BP; Leung. HW; Scott. P; Paustenbach. DJ; Needham. LL; Patterson. DG. Jr; Gerthoux.
PM; Mocarelli. P. (2006). Age- and concentration-dependent elimination half-life of
5-9 DRAFT - DO NOT CITE OR QUOTE
-------
2,3,7,8-tetrachlorodibenzo-p-dioxin in Seveso children. Environ Health Perspect 114:
1596-1602. http://dx.doi.org/10.1289/ehp.8884.
Kerger. BP; Leung. HW; Scott. PK; Paustenbach. DJ. (2007). Refinements on the age-dependent
half-life model for estimating child body burdens of polychlorodibenzodioxins and
dibenzofurans. Chemosphere 67: S272-S278.
http://dx.doi.org/10.1016/i.chemosphere.2006.05.108.
Kim. AH: Kohn. MC: Nvska. A: Walker. NJ. (2003). Area under the curve as a dose metric for
promotional responses following 2,3,7,8-tetrachlorodibenzo-p-dioxin exposure. Toxicol
Appl Pharmacol 191: 12-21. http://dx.doi.org/10.1016/S0041 -008X(03)00225-4.
Kitchin. KT; Woods. JS. (1979). 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) effects on hepatic
microsomal cytochrome P-448-mediated enzyme activities. Toxicol Appl Pharmacol 47:
537-546. http://dx.doi.org/10.1016/0041 -008X(79)90524-6.
Kociba. RJ: Keeler. PA; Park. CN: Gehring. PJ. (1976). 2,3,7,8-T etrachl orodi benzo-p-di ox i n
(TCDD): Results of a 13-week oral toxicity study in rats. Toxicol Appl Pharmacol 35:
553-574. http://dx.doi.org/10.1016/0041 -008X(76)90078-8.
Kociba. RJ; Keyes. DG; Beyer. IE; Carreon. RM; Wade, CE; Dittenber. DA; Kalnins. RP:
Frauson. LE; Park. CN; Barnard. SD; Hummel. RA; Humiston. CG. (1978). Results of a
two-year chronic toxicity and oncogenicity study of 2,3,7,8-tetrachlorodibenzo-p-dioxin
in rats. Toxicol Appl Pharmacol 46: 279-303. http://dx.doi.org/10.1016/0041 -
008X( 78)90075-3.
Kohn. MC; Lucier. GW; Clark. GC; Sewall. C; Tritscher. AM; Portier. CJ. (1993). A
mechanistic model of effects of dioxin on gene expression in the rat liver. Toxicol Appl
Pharmacol 120: 138-154. http://dx.doi.org/10.1006/taap. 1993.1096.
Kohn. MC; Sewall. CH; Lucier. GW; Portier. CJ. (1996). A mechanistic model of effects of
dioxin on thyroid hormones in the rat. Toxicol Appl Pharmacol 165: 29-48.
http://dx.doi.org/10.1006/taap. 1996.0004.
Kohn. MC; Walker. NJ; Kim. AH; Portier. CJ. (2001). Physiological modeling of a proposed
mechanism of enzyme induction by TCDD. Toxicology 162: 193-208.
http://dx.doi.org/10.1016/S0300-483X(01 )00363-8.
Kreuzer. PE; Csanadv. GA; Baur. C; Kessler. W; Papke. O; Greim. H; Filser. JG. (1997).
2,3,7,8-Tetrachlorodibenzo-p -dioxin (TCDD) and congeners in infants. A toxicokinetic
model of human lifetime body burden by TCDD with special emphasis on its uptake by
nutrition. Arch Toxicol 71: 383-400.
Krishnan. K; Andersen. ME. (1991). Interspecies scaling in pharmacokinetics. In A Rescingo; A
Thakkur (Eds.), New trends in pharmacokinetics (pp. 203-226). New York, NY: Plenum
Press.
Krowke. R; Chahoud. I; Baumann-Wilschke. I; Neubert. D. (1989). Pharmacokinetics and
biological activity of 2,3,7,8-tetrachlorodibenzo-p-dioxin 2: pharmacokinetics in rats
using a loading-dose/maintenance-dose regime with high doses. Arch Toxicol 63: 356-
360. http://dx.doi.org/10.1007/BF00303 123.
Kuchiiwa. S; Cheng. SB; Nagatomo. I; Akasaki. Y; Uchida. M; Tominaga. M; Hashiguchi. W;
Kuchiiwa. T. (2002). In utero and lactational exposure to 2,3,7,8-tetrachlorodibenzo-p-
dioxin decreases serotonin-immunoreactive neurons in raphe nuclei of male mouse
offspring. Neurosci Lett 317: 73-76. http://dx.doi.org/10.1016/S0304-3940(01 )02434-X.
LaKind. JS; Berlin. CM; Park. CN; Naiman. DO; Gudka. NJ. (2000). Methodology for
characterizing distributions of incremental body burdens of 2,3,7,8-TCDD and DDE from
5-10
DRAFT - DO NOT CITE OR QUOTE
-------
breast milk in North American nursing infants. J Toxicol Environ Health A 59: 605-639.
http://dx.doi.org/10.1080/00984100Q156628.
Lakshmanan. MR: Campbell. BS; Chirtel. SJ: Ekarohita. N: M. E. (1986). Studies on the
mechanism of absorption and distribution of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the
rat. J Pharmacol Exp Ther 239: 673-677.
Landi. MT; Bertazzi. PA: Baccarelli. A: Consonni. D; Masten. S: Lucier. G: Mocarelli. P:
Needham. L: Caporaso. N: Grassman. J. (2003). TCDD-mediated alterations in the Ah In-
dependent pathway in Seveso, Italy, 20 years after the accident. Carcinogenesis 24: 673-
680. http://dx.doi.org/10.1093/carcin/bgg002.
Latchoumycandane. C; Chitra. KC; Mathur. PP. (2002a). The effect of 2,3,7,8-
tetrachlorodibenzo-p-dioxin on the antioxidant system in mitochondrial and microsomal
fractions of rat testis. Toxicology 171: 127-135. http://dx.doi.org/10.1016/S0300-
483X(01)00563-7.
Latchoumycandane. C: Mathur. PP. (2002). Effects of vitamin E on reactive oxygen species-
mediated 2,3,7,8-tetrachlorodibenzo-p-dioxin toxicity in rat testis. J Appl Toxicol 22:
345-35 1. http://dx.doi.org/10.1002/jat.866.
Latchoumycandane. C: Chitra. C: Mathur. P. (2002b). Induction of oxidative stress in rat
epididymal sperm after exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Arch Toxicol
76: 113-118. http://dx.doi.org/10.1007/s00204-001 -0308-4.
Latchoumycandane. C: Chitra. KC; Mathur. PP. (2003). 2,3,7,8-Tetrachlorodibenzo-p-dioxin
(TCDD) induces oxidative stress in the epididymis and epididymal sperm of adult rats.
Arch Toxicol 77: 280-284. http://dx.doi.org/10.1007/s00204-003-0439-x.
Lavado-Autric. R; Auso. E; Garcia-Velasco. IV: Del Carmen Arufe. M; Escobar Del Rev. F;
Berbel. P; Morreale De Escobar. G. (2003). Early maternal hypothyroxinemia alters
histogenesis and cerebral cortex cytoarchitecture of the progeny. J Clin Invest 111: 1073-
1082. http://dx.doi.org/10.1 172/JC116262.
Lawrence. GS: Gobas. FAP. C. (1997). A pharmacokinetic analysis of interspecies extrapolation
in dioxin risk assessment. Chemosphere 35: 427-452. http://dx.doi.org/10.1016/S0045-
6535(97)00108-2.
Lean. MEJ; Han. TS: Deurenberg. P. (1996). Predicting body composition by densitometry from
simple anthropometric measurements. Am J Clin Nutr 63: 4-14.
Leo. A: Hansch. C: El kins. D. (1971). Partition coefficients and their uses [Review], Chem Rev
71: 525-616. http://dx.doi.org/10.102 l/cr60274a001.
Leung. HW; Ku. RH; Paustenbach. DJ; Andersen. ME. (1988). A physiologically based
pharmacokinetic model for 2,3,7,8-tetrachlorodibenzo-p-dioxin in C57BL/6J and DBA/2J
mice. Toxicol Lett 42: 15-28.
Leung. HW: Poland. A: Paustenbach. DJ; Murray. FJ; Andersen. ME. (1990). Pharmacokinetics
of [125I]-2-iodo-3,7,8-trichlorodibenzo-p-dioxin in mice: analysis with a physiological
modeling approach. Toxicol Appl Pharmacol 103: 411-419.
Leung. HW; Kerger. BP; Paustenbach. DJ. (2006). Elimination half-lives of selected
polychlorinated dibenzodioxins and dibenzofurans in breast-fed human infants. J Toxicol
Environ Health A 69: 437-443. http://dx.doi.org/10.1080/1528739050Q246886.
Li. B; Liu. HY; Dai. LJ; Lu. JC; Yang. ZM; Huang. L. (2006). The early embryo loss caused by
2,3,7,8-tetrachlorodibenzo-p-dioxin may be related to the accumulation of this compound
in the uterus. Reprod Toxicol 21: 301-306.
http://dx.doi.org/10.1016/j .reprotox.2005.09.008.
5-11
DRAFT - DO NOT CITE OR QUOTE
-------
Li. X; Johnson. DC; Rozman. KK. (1997). 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD)
increases release of luteinizing hormone and follicle-stimulating hormone from the
pituitary of immature female rats in vivo and in vitro. Toxicol Appl Pharmacol 142: 264-
269. http://dx.doi.org/10.1006/taap.1996.8044.
Longnecker. MP: Michalek. JE. (2000). Serum Dioxin level in relation to Diabetes Mellitus
among Air Force veterans with background levels of exposure. Epidemiology 11: 44-48.
Lorber. M. (2002). A pharmacokinetic model for estimating exposure of Americans to dioxin-
like compounds in the past, present, and future. Sci Total Environ 288: 81-95.
Lorber. M; Patterson. D; Huwe. J: Kahn. H. (2009). Evaluation of background exposures of
Americans to dioxin-like compounds in the 1990s and the 2000s. Chemosphere 77: 640-
65 1. http://dx.doi.Org/10.1016/i.chemosphere.2009.08.016.
Lorenzen. A: Okev. AB. (1991). Detection and characterization of Ah receptor in tissue and cells
from human tonsils. Toxicol Appl Pharmacol 107: 203-214.
Lucier. GW: Rumbaugh. RC: McCoy. Z; Hass. R; Harvan. D; Albro. P. (1986). Ingestion of soil
contaminated with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) alters hepatic enzyme
activities in rats. Fundam Appl Toxicol 6: 364-371.
Mall v. A: Chipman. JK. (2002). Non-genotoxic carcinogens: Early effects on gap junctions, cell
proliferation and apoptosis in the rat. Toxicology 180: 233-248.
Manchester. DK; Gordon. SK; Golas. CL; Roberts. EA; Okev. AB. (1987). Ah receptor in
human placenta: stabilization by molydate and characterization of binding of 2,3,7,8-
tetrachlorodibenzo-p-dioxin, 3-methylcholanthrene, and benzo(a)pyrene. Cancer Res 47:
4861-4868.
Markowski. VP: Zareba. G: Stern. S: Cox. C: Weiss. B. (2001). Altered operant responding for
motor reinforcement and the determination of benchmark doses following perinatal
exposure to low-level 2,3,7,8-tetrachlorodibenzo-p-dioxin. Environ Health Perspect 109:
621-627.
Maronpot. RR; Foley. JF; Takahashi. K; Goldsworthv. T; Clark. G: Tritscher. A: Portier. C:
Lucier. G. (1993). Dose response for TCDD promotion of hepatocarcinogenesis in rats
initiated with DEN: histologic, biochemical, and cell proliferation endpoints. Environ
Health Perspect 101: 643-642.
Maruvama. W: Yoshida. K; Tanaka. T; Nakanishi. J. (2002). Determination of tissue-blood
partition coefficients for a physiological model for humans, and estimation of dioxin
concentration in tissues. Chemosphere 46: 975-985.
McNulty. WP; Nielsen-Smith. KA; Lay. JO. Jr; Lippstreu. PL; Kangas. NL; Lyon. PA; Gross.
ML. (1982). Persistence of TCDD in monkey adipose tissue. Food Chem Toxicol 20:
985-986.
Michalek. JE; Akhtar. FZ; Longnecker. MP; Burton. JE. (2001a). Relation of serum 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) level to hematological examination results in
veterans of Operation Ranch Hand. Arch Environ Health 56: 396-405.
http://dx.doi.org/10.1080/00039890109604474.
Michalek. JE; Ketchum. NS; Longnecker. MP. (200 lb). Serum dioxin and hepatic abnormalities
in veterans of Operation Ranch Hand. Ann Epidemiol 11: 304-311.
http://dx.doi.org/10.1016/S1047-2797(00)00218-0.
Michalek. JE; Akhtar. FZ; Arezzo. JC; Garabrant. DH; Albers. JW. (2001c). Serum dioxin and
peripheral neuropathy in veterans of Operation Ranch Hand. Neurotoxicology 22: 479-
490.
5-12
DRAFT - DO NOT CITE OR QUOTE
-------
Michalak. JE; Pirkle. JL; Needham. LL; Patterson. DG. Jr; Caudill. SP; Tripathi. RC; Mocarelli.
P. (2002). Pharmacokinetics of 2,3,7,8-tetrachlorodibenzo-p-dioxin in Seveso adults and
veterans of operation Ranch Hand. J Expo Anal Environ Epidemiol 12: 44-53.
Michalek. JE; Pavuk. M. (2008). Diabetes and cancer in veterans of Operation Ranch Hand after
adjustment for calendar period, days of spraying, and time spent in Southeast Asia. J
Occup Environ Med 50: 330-340. http://dx.doi.org/10.1097/JQM.0b013e31815f889b.
Micka. J: Milatovich. A: Men on. A: Grabowski. GA; Puga. A: Nebert. DW. (1997). Human Ah
receptor (AHR) gene: Localization to 7pl5 and suggestive correlation of polymorphism
with CYP1A1 inducibility. Pharmacogenetics 7: 95-101.
Miettinen. HM; Sorvari. R; Alaluusua. S; Murtomaa. M; Tuukkanen. J; Viluksela. M. (2006).
The Effect of Perinatal TCDD exposure on caries susceptibility in rats. Toxicol Sci 91:
568-575. http://dx.doi.org/10.1093/toxsci/kf] 158.
Milbrath. MO; Wenger. Y; Chang. CW: Emond. C: Garabrant. D; Gillespie. BW: Jolliet. O.
(2009). Apparent half-lives of dioxins, furans, and polychlorinated biphenyls as a
function of age, body fat, smoking status, and breast-feeding. Environ Health Perspect
1 17: 417-425. http://dx.doi.org/10.1289/ehp. 1 1781.
Mocarelli. P. (2000). Dioxin and human sex ratio: The Seveso case. Milan, Italy: Department of
Laboratory Medicine Hospital of Desio.
Mocarelli. P; Gerthoux. PM; Ferrari. E; Patterson Jr. DG; Kieszak. SM; Brambilla. P; Vincoli.
N; Signorini. S; Tramacere. P; Carreri. V; Sampson. EJ; Turner. WE. (2000). Paternal
concentrations of dioxin and sex ratio of offspring. Lancet 355: 1858-1863.
http://dx.doi. org/10.1016/SO140-673 6(00)02290-X.
Mocarelli. P. (2001). Seveso: A teaching story. Chemosphere 43: 391-402.
http://dx.doi.org/10.1016/S0045-6535(00)00386-6.
Mocarelli. P; Gerthoux. PM; Patterson. DG. Jr; Milani. S; Limonata. G; Bertona. M; Signorini.
S; Tramacere. P; Colombo. L; Crespi. C; Brambilla. P; Sarto. C; Carreri. V; Sampson.
EJ; Turner. WE; Needham. LL. (2008). Dioxin exposure, from infancy through puberty,
produces endocrine disruption and affects human semen quality. Environ Health Perspect
1 16: 70-77. http://dx.doi.org/10.1289/ehp. 10399.
Mocarelli. P; Gerthoux. PM; Needham. LL; Patterson. DG. Jr; Limonta. G; Falbo. R; Signorini.
S; Bertona. M; Crespi. C; Sarto. C; Scott. PK; Turner. WE; Brambilla. P. (201 1).
Perinatal exposure to low doses of dioxin can permanently impair human semen quality.
Environ Health Perspect 119: 713-718.
Morreale de Escobar. G; Obregon. MJ; Escobar del Ray. F. (2000). Is neuropsychological
development related to maternal hypothyroidism or to maternal hypothyroxinemia? J Clin
Endocrinol Metab 85: 3975-3987. http://dx.doi.org/10.1210/ic.85.1 1.3975.
Moser. GA; McLachlan. MS. (2001). The influence of dietary concentration on the absorption
and excretion of persistent lipophilic organic pollutants in the human intestinal tract.
Chemosphere 45: 201-21 1. http://dx.doi.org/10.1016/S0045-6535(00)0055 1 -8.
Murray. FJ; Smith. FA; Nitschke. KD; Humiston. CG; Kociba. RJ; Schwetz. BA. (1979). Three-
generation reproduction study of rats given 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)
in the diet. Toxicol Appl Pharmacol 50: 241-252. http://dx.doi.org/10.1016/0041 -
008X(79)90149-2.
NAS (National Academy of Sciences). (2006a). Health risks from dioxin and related compounds
Retrieved February 9, 2010, from
http://vvvvvv.nap.edu/vvebcast/vvebcast detail.php?webcast id=328
5-13
DRAFT - DO NOT CITE OR QUOTE
-------
NAS (National Academy of Sciences). (2006b). Health risks from dioxin and related
compounds: Evaluation of the EPA reassessment. Washington, DC: National Academy
Press, http://www.nap.edu/catalog.php7record id=l 1688.
Nebert. DW; Peterson. DP; Puua. A. (1991). Human AH locus polymorphism and cancer:
Inducibility of CYP1A1 and other genes by combustion products and dioxin.
Pharmacogenetics 1: 68-78.
Needham. LL; Gerthoux. PM; Patterson. DG; Brambilla. P; Turner. WE; Beretta. C: Pirkle. JL;
Colombo. L; Sampson. EJ; Tramacere. PL; Signorini. S: Meazza. L; Carreri. V: Jackson.
RJ; Mocarelli. P. (1998). Serum dioxin levies in Seveso, Italy, population in 1976.
Teratog Carcinog Mutagen 17: 225-240.
Nessel. CS; Amoruso. MA; Umbreit. TH; Meeker. RJ; Gallo. MA. (1992). Transpulmonary
uptake and bioavailability of 2,3,7,8-TCDD from respirable soil particles. Chemosphere
25: 29-32.
Niskar. A; Needham. LL; Rubin. C; Turner. WE; Martin. CA; Patterson. DG. Jr; Hasty. L;
Wong. LY; Marcus. M. (2009). Serum dioxin, polychlorinated biphenyls, and
endometriosis: A case-control study in Atlanta. Chemosphere 74: 944-949.
http://dx.doi.org/10.1016/i.chemosphere. 2008.10.005.
Nohara. K; Fuiimaki. H; Tsukumo. S; Ushio. H; Mivabara. Y; Kijitna. M; Tohvama. C;
Yonemoto. J. (2000). The effects of perinatal exposure to low doses of 2,3,7,8-
tetrachlorodibenzo-p-dioxin on immune organs in rats. Toxicology 154: 123-133.
http://dx.doi.org/10.1016/S0300-483X(00)00323-1.
Nohara. K; Izumi. H; Tamura. S; Nagata. R; Tohvama. C. (2002). Effect of low-dose 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) on influenza A virus-induced mortality in mice.
Toxicology 170: 131-138.
Nolan. KJ; Smith. FA; Hefner. JG. (1979). Elimination and tissue distribution of 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) in female guinea pigs following a single oral dose
[Abstract], Toxicol Appl Pharmacol 48: A162. http://dx.doi.org/10.1016/0041 -
008X(79)90509-X.
NTP (National Toxicology Program). (1982a). Carcinogenesis bioassay of 2,3,7,8-
tetrachlorodibenzo-p-dioxin in Swiss-Webster mice (dermal study). RTP, NC: U.S.
Department of Health and Human Services.
NTP (National Toxicology Program). (1982b). Carcinogenesis bioassay of BIS(2-chloro-l-
methylethyl) ether ( 70%) (CAS no. 108-60-1) containing 2-chloro-l-methyl ethyl (2-
chloropropyl) ether ( 30%) (CAS no. 83270-31-9) in B6C3F1 mice (gavage study).
(NTP-81-55). Research Triangle Park, NC and Bethesda, MD.
NTP (National Toxicology Program). (1982c). NTP Technical Report on carcinogenesis
bioassay of 2,3,7,8-tetrachlorodibenzo-p-dioxin in Osborne-Mendel rats and B6C3F1
mice (gavage study). Research Triangle Park, NC.
http://ntp.niehs.nih.gov/7obiectidKJ7060172-DEB2-6542-D7CD537BAB5B2ACD.
NTP (National Toxicology Program). (2006a). NTP technical report on the toxicology and
carcinogenesis studies of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) (CAS no. 1746-
01-6) in female harlan Sprague-Dawley rats (gavage studies). (NTP TR 521; NIH
Publication No. 06-4468). RTP, NC.
NTP (National Toxicology Program). (2006b). NTP toxicology and carcinogenesis studies of a
binary mixture of 3,3',4,4',5-Pentachlorobiphenyl (PCB 126) (CAS No. 57465-28-8) and
2,2',4,4',5,5'-Hexachlorobiphenyl (PCB 153) (CAS No. 35065-27-1) in female Harlan
5-14
DRAFT - DO NOT CITE OR QUOTE
-------
Sprague-Dawley rats (Gavage studies). (NTP TR 530). Research Triangle Park, NC.
http://ntp.niehs.nih.gov/files/TR530 Web 1 .pdf.
NTP (National Toxicology Program). (2006c). Toxicology and carcinogenesis studies of a
mixture of 2,3,7,8-tetrachlorodibenzo-p-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). (NTP TR 526). Research Triangle Park, NC.
http://ntp.niehs.nih.gov/index.cfrn?obi ectid=070B7300-0E62-BF 12-
F4C3E3B5B645A92B.
Oerbeck. B; Sundet. K; Kase. BF; Heyerdahl. S. (2005). Congenital hypothyroidism: No adverse
effects of high dose thyroxine treatment on adult memory, attention, and behaviour. Arch
Dis Child 90: 132-137. http://dx.doi.org/10.1136/adc.2003.043935.
Ogura. I: Masunaga. S: Nakanishi. J. (2004). Quantitative source identification of dioxin-like
PCBs in Yokohama, Japan, by temperature dependence of their atmospheric
concentrations. Environ Sci Technol 38: 3279-3285.
http://dx.doi.org/10.102 l/es0354622.
Ohsako. S: Mivabara. Y; Nishimura. N: Kurosawa. S: Sakaue. M; Ishimura. R; Sato. M; Takeda.
K; Aoki. Y; Sone. H; Tohvama. C: Yonemoto. J. (2001). Maternal exposure to a low
dose of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) suppressed the development of
reproductive organs of male rats: Dose-dependent increase of mRNA levels of 5a-
reductase type 2 in contrast to descrease of androgen receptor in the pubertal ventral
prostate. Toxicol Sci 60: 132-143.
Okev. AB; Riddick. PS; Harper. PA. (1994). The Ah receptor: Mediator of the toxicity of
2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and related compounds. Toxicol Lett 70: 1-
22. http://dx.doi.org/10.1016/0378-4274(94)90139-2.
Olson. JR.; Holscher. MA; Neal. RA. (1980). Toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in
the Golden Syrian hamster. Toxicol Appl Pharmacol 55: 67-78.
http://dx.doi.org/10.1016/0041 -008X( 80)90221 -5.
Olson. JR.; McGarrigle. BP; Gigliotti. PJ; Kumar. S; McRevnolds. JH. (1994). Hepatic uptake
and metabolism of 2,3,7,8-tetrachlorodibenzo-p-dioxin and 2,3,7,8-
tetrachlorodibenzofuran. Fundam Appl Toxicol 22: 631-640.
http://dx.doi.org/10.1006/faat. 1994.1069.
Ott. MG; Zober. A. (1996). Morbidity study of extruder personnel with potential exposure to
brominated dioxins and furans. II. Results of clinical laboratory studies. Occup Environ
Med 53: 844-846.
Patterson. DG. Jr; Wong. LY; Turner. WE; Caudill. SP; Dipietro. ES; McClure. PC; Cash. TP;
Osterloh. JD; Pirkle. JL; Sampson. EJ; Needham. LL. (2009). Levels in the U.S.
population of those persistent organic pollutants (2003-2004) included in the Stockholm
Convention or in other long range transboundary air pollution agreements. Environ Sci
Technol 43: 1211-1218. http://dx.doi.org/10.1021 /es801966w.
Pavuk. M; Schecter. AJ; Akhtar. FZ; Michalek. JE. (2003). Serum 2,3,7,8-tetrachlorodibenzo-p-
dioxin (TCDD) levels and thyroid function in Air Force veterans of the Vietnam War.
Ann Epidemiol 13: 335-343. http://dx.doi.org/10.1016/S1047-2797(02)00422-2.
Pavuk. M; Patterson. DG. Jr; Turner. WE; Needham. LL; Ketchum. NS. (2007). Polychlorinated
dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), and dioxin-like
5-15
DRAFT - DO NOT CITE OR QUOTE
-------
polychlorinated biphenyls (PCBs) in the serum of US Air Force veterans in 2002.
Chemosphere 68: 62-68. http://dx.doi.org/ 10.1016/i.cheinosphere.2006.12.039.
Pin sky. PF; Lorber. MN. (1998). A model to evaluate past exposure to 2,3,7,8-TCDD. J Expo
Anal Environ Epidemiol 8: 187-206.
Pipe. NG; Smith. T; Hallidav. D; Edmonds. CJ; Williams. C; Coltart. TM. (1979). Changes in
fat, fat-free mass and body water in human normal pregnancy. Br J Obstet Gynaecol 86:
929-940. http://dx.doi.org/10.1111/i. 1471-0528.1979.tb 1 1240.x.
Poiger. H; Schlatter. C. (1986). Pharmacokinetics of 2,3,7,8-TCDD in man. Chemosphere 15:
1489-1494. http://dx.doi.org/10.1016/0045-6535(86)90429-7.
Poulin. P; Theil. FP. (2001). Prediction of pharmicokinetics prior to in vivo studies. 1.
mechanism-based prediction of volume of distribution. J Pharm Sci 91: 129-156.
http://dx.doi.org/10.1002/jps. 10005.
Ramsey. JC: Hefner. JG; Karbowski. RJ; Braun. WH; Gehring. PJ. (1982). The in vivo
biotransformation of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the rat. Toxicol
Appl Pharmacol 65: 180-184. http://dx.doi.org/10.1016/0041 -008X(82)90377-5.
Rao. MS; Subbarao. V; Prasad. JD; Scarpelli. DG. (1988). Carcinogenicity of 2,3,7,8-
tetrachlorodibenzo-p-dioxin in the Syrian golden hamster. Carcinogenesis 6: 1677-1679.
Reddv. M; Yang. R; Clevvell. HJ; Andersen. ME. (2005). Physiologically based pharmacokinetic
modeling: Science and applications. Hoboken, New Jersey: John Wiley & Sons.
Rier. SE; Martin. DC; Bowman. RE: Dmowski. WP; Becker. JL. (1993). Endometriosis in
Rhesus Monkeys (Macaca mulatta) Following Chronic Exposure to 2,3,7,8-
Tetrachlorodibenzo-p-dioxin. Fundam Appl Toxicol 21: 433-441.
Rier. SE; Martin. DC; Bowman. RE; Becker. JL. (1995). Immunoresponsiveness in
endometriosis: Implications of estrogenic toxicants. Environ Health Perspect 103: 151-
156.
Rigon. F; Bianchin. L; Bernasconi. S; Bona. G; Bozzola. M; Buzi. F; Cicognani. A; De Sanctis.
C; De Sanctis. V; Radetti. G; Tato. L; Tonini. G; Perissinotto. E. (2010). Update on age
at menarche in Italy: Toward the leveling off of the secular trend. J Adolesc Health 46:
238-244. http://dx.doi.org/ 10.1016/i.iadohealth.2009.07.009.
Roberts. EA; Shear. NH; Okev. AB; Manchester. DK. (1985). The Ah receptor and dioxin
toxicity: From rodent to human tissues. Chemosphere 14: 661-674.
http://dx.doi.org/10.1016/0045-6535(85 )90174-2.
Roberts. EA; Golas. CL; Okev. AB. (1986). Ah receptor mediating induction of aryl
hydrocarbon hydroxylase: Detection in human lung by binding of 2,3,7,8-
[H]tetrachlorodibenzo-p-dioxin. Cancer Res 46: 3739-3743.
Rohde. S; Moser. GA; Papke. O; McLachlan. MS. (1999). Clearance of PCDD/Fs via the
gastrointestinal tract in occupationally exposed persons. Chemosphere 38: 3397-3410.
http://dx.doi.org/10.1016/S0045-6535(98)0055 1 -7.
Roth. WL; Ernst. S; Weber. LWD; Kerescen. L; Rozman. KK. (1994). A pharmacodynamically
responsive model of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) transfer between liver
and fat at low and high doses. Toxicol Appl Pharmacol 127: 151-162.
http://dx.doi.org/10.1006/taap. 1994.1 149.
Rovet. JF. (2002). Congenital hypothyroidism: An analysis of persisting deficits and associated
factors. Child Neuropsychol 8: 150-162. http://dx.doi.org/10.1076/chin.8.3.150.13501.
5-16
DRAFT - DO NOT CITE OR QUOTE
-------
Roy. T; Hammerstrom. K; Schaum. J. (2008). Percutaneous absorption of 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) from soil. J Toxicol Environ Health A 71: 1509-
1515. http://dx.doi.org/10.1080/152873908Q2349875.
Royland. JE; Parker. JS; Gilbert. ME. (2008). A genomic analysis of subclinical hypothyroidism
in hippocampus and neocortex of the developing rat brain. J Neuroendocrinol 20: 1319-
1338. http://dx.doi.org/10.1111/i. 1365-2826.2008.01793.x.
SAB (U.S. EPA Science Advisory Board). (2011). SAB review of EPA's reanalysis of key issues
related to dioxin toxicity and response to NAS comments. (EPA-SAB-011-014).
Washington, DC: U.S. Environmental Protection Agency.
http://yosemite.epa. gov/sab/sabproduct.nsf/2A45B492EBAA8553852578F9003ECBC5/$
File/EPA-SAB-1 1-014-unsigned.pdf.
Santostefano. MJ; Johnson. KL; Whisnant. NA; Richardson. VM; Devi to. MJ; Birnbaum. LS.
(1996). Subcellular localization of TCDD differs between the liver, lungs, and kidneys
after acute and subchronic exposure: Species/dose comparison and possible mechanism.
Fundam Appl Toxicol 34: 365-375.
Santostefano. MJ; Wang, X; Richardson. VM; Ross. DG; DeVito. MJ; Birnbaum. LF. (1998). A
pharmacodynamic analysis of TCDD-Induced Cytochrome 450 gene expression in
multiple tissues: Dose and time-dependent effects. Toxicol Appl Pharmacol 151: 294-
310.
Savin. S; Cveiic. D; Nedic. O; Radosavlievic. R. (2003). Thyroid hormone synthesis and storage
in the thyroid gland of human neonates. J Pediatr Endocrinol Metab 16: 521-528.
Schantz. SL; Laughlin. NK; Van Valkenberg. HC; Bowman, RE. (1986). Maternal care by
rhesus monkeys of infant monkeys exposed to either lead or 2,3,7,8-tetrachlorodibenzo-
P-dioxin. Neurotoxicology 7: 637-650.
Schantz. SL; Bowman, RE. (1989). Learning in monkeys exposed perinatally to 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD). Neurotoxicol Teratol 11: 13-19.
http://dx.doi.org/10.1016/0892-0362(89)90080-9.
Schantz. SL; Ferguson. SA; Bowman, RE. (1992). Effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin
on behavior of monkeys in peer groups. Neurotoxicol Teratol 14: 433-446.
Schantz. SL; Seo. BW; Moshtaghian. J; Peterson. RE; Moore. RW. (1996). Effects of gestational
and lactational exposure to TCDD or coplanar PCBs on spatial learning. Neurotoxicol
Teratol 18: 305-313.
Seo. BW; Li. MH; Hansen. LG; Moore. RW; Peterson. RE; Schantz. SL. (1995). Effects of
gestational and lactational exposure to coplanar polychlorinated biphenyl (PCB)
congeners or 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on thyroid hormone
concentrations in weanling rats. Toxicol Lett 78: 253-262.
Sevvall. C; Lucier. G; Tritscher. A; Clark. G. (1993). TCDD-mediated changes in hepatic
epidermal growth factor receptor may be a critical event in the hepatocarcinogenic action
of TCDD. Carcinogenesis 14: 1885-1893.
Sevvall. CH; Flagler. N; Vanden Heuvel. JP; Clark. GC; Tritscher. AM; Maronpot. RM; Lucier.
GW. (1995). Alterations in thyroid function in female Sprague-Dawley rats following
chronic treatment with 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol Appl Pharmacol
132: 237-244.
Sharlin. PS; Tighe. D; Gilbert. ME; Zoeller. RT. (2008). The balance between oligodendrocyte
and astrocyte production in major white matter tracts is linearly related to serum total
thyroxine. Endocrinology 149: 2527-2536. http://dx.doi.org/10.1210/en.2007-143 1.
5-17
DRAFT - DO NOT CITE OR QUOTE
-------
Sharlin. PS; Gilbert. ME; Taylor. MA; Ferguson. DC; Zoeller. RT. (2010). The nature of the
compensatory response to low thyroid hormone in the developing brain. J
Neuroendocrinol 22: 153-165. http://dx.doi.Org/10.llll/i.1365-2826.2009.01947.x.
Shi. Z; Valdez. KE; Ting. AY; Franczak. A; Gum. SL; Petroff, BK. (2007). Ovarian endocrine
disruption underlies premature reproductive senescence following environmentally
relevant chronic exposure to the aryl hydrocarbon receptor agonist 2,3,7,8-
tetrachlorodibenzo-p-dioxin. Biol Reprod 76: 198-202.
http://dx.doi.org/10.1095/biolreprod. 106.053991.
Shu. H; Teitelbaum. P; Webb. AS; Marple. L; Brunck. B; Dei Rossi. D; Murray. FJ;
Paustenbach. D. (1988). Bioavailability of soil-bound TCDD: Dermal bioavailability in
the rat. Fundam Appl Toxicol 2: 335-343. http://dx.doi.org/10.1016/0272-
0590(88)90319-3.
Sielken. RL. Jr. (1993, February 15-17, 1993). Evaluation of chloroform risk to humansThe
Toxicology Forum, 1993 Annual Winter Meeting, The Capitol Hilton, Washington, DC.
Sielken. RL. Jr; Valdez-Flores. C. (1996). Comprehensive realism's weight-of-evidence based
distributional dose response characterization. Hum Ecol Risk Assess 2: 175-193.
http://dx.doi.org/10.1080/10807Q39.1996.10387467.
Sielken. RL. Jr; Valdez-Flores. C. (1999). Probabilistic risk assessment's use of trees and
distributions to reflect uncertainty and variability and to overcome the limitations of
default assumptions. Environ Int 25: 755-772. http://dx.doi.org/10.1016/SO160-
4120(99)00053-7.
Sielken. RL, Jr. (1990). A weight-of-evidence approach to quantitative cancer risk assessment:
Information analysis. In G Schettler; D Schmahl; T Klenner (Eds.), Risk assessment in
chemical carcinogenesis. New York, NY: Springer-Verlag.
Sielken. RL. Jr.; Bretzlaff. RS; Stevenson. DE. (1995). Challenges to default assumptions
stimulate comprehensive realism as a new tier in quantitative cancer risk assessment.
Regul Toxicol Pharmacol 21: 270-280.
Sikov. M. (1970). Radiation biology of the fetal and juvenile mammal. Science 167: 1640-1641.
Silkworth. JB; Koganti. A; Illouz. K; Possolo. A; Zhao. M; Hamilton. SB. (2005). Comparison
of TCDD and PCB CYP1A induction sensitivities in fresh hepatocytes from human
donors, sprague-dawley rats, and rhesus monkeys and HepG2 cells. Toxicol Sci 87: 508-
5 19. http://dx.doi.org/10.1093/toxsci/kfi261.
Simanainen. U; Tuomisto. JT; Tuomisto. J; Viluksela. M. (2002). Structure-Activity
relationships and dose responses of Polychlorinated Dibenzo-p-dioxins for short-term
effects in 2,3,7,8- Tetrachlorodibenzo-p-dioxin-Resistant and sensitive rat strains.
Toxicol Appl Pharmacol 181: 38-47. http://dx.doi.org/10.1006/taap.2002.9386.
Simanainen. U; Tuomisto. JT; Tuomisto. J; Viluksela. M. (2003). Dose-response analysis of
short-term effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin in three differentially
susceptible rat lines. Toxicol Appl Pharmacol 187: 128-136.
http://dx.doi.org/10.1016/S0041 -008X(02)00068-6.
Simanainen. U; Haavisto. T; Tuomisto. JT; Paranko. J; Toppari. J; Tuomisto. J; Peterson. RE;
Viluksela. M. (2004a). Pattern of male reproductive system effects after in utero and
lactational 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) exposure in three differentially
TCDD-sensitive rat lines Pattern of male reproductive system effects after in utero and
lactational 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) exposure in three differentially
5-18
DRAFT - DO NOT CITE OR QUOTE
-------
TCDD-sensitive rat lines. Toxicol Sci 80: 101-108.
http://dx.doi.org/10.1093/toxsci/kfhl42.
Simanainen. U; Tuomisto. JT; Pohianvirta. R; Svriala. P: Tuomisto. J: M. V. (2004b). Postnatal
development of resistance to short-term high-dose toxic effects of 2,3,7,8-
tetrachlorodibenzo-p-dioxin in TCDD-resistant and -semiresistant rats. Toxicol Appl
Pharmacol 196: 11-19. http://dx.doi.Org/10.1016/i.taap.2003.l 1.025.
Skakkebaek. NE. (2010). reference ranges for semen quality and their relations to fecundity.
Asian J Androl 12: 95-98.
Slama. R; Eustache. F; Ducot. B; Jensen. TK; Jorgensen. N: Horte. A: Irvine. S: Suominen. J:
Andersen. AG; Auger. J; Vierula. M; Toppari. J; Andersen. AN; Keiding. N;
Skakkebaek. NE; Spira. A; Jouannet. P. (2002). Time to pregnancy and semen
parameters: A cross-sectional study among fertile couples from four European cities.
Hum Reprod 17: 503-5 15. http://dx.doi.org/10.1093/humrep/17.2.503.
Slezak. BP; Hatch. GE; DeVito. MJ; Diliberto. JJ; Slade. R; Crissman. K; Hassoun. E;
Birnbaum. LS. (2000). Oxidative stress in female B6C3F1 mice following acute and
subchronic exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Toxicol Sci 54:
390-398.
Small. MJ. (2008). Methods for assessing uncertainty in fundamental assumptions and associated
models for cancer risk assessment. Risk Anal 28: 1289 - 1308.
http://dx.doi.org/10.1111/i. 1539-6924.2008.01 134.x.
Smart. J; Daly. A. (2000). Variation in induced C YP1A1 levels: Relationship to CYP1A1, Ah
receptor, and GSTM1 polymorphisms. Pharmacogenetics 10: 11-24.
Smialowicz. RJ; Burgin. DE; Williams. WC; Diliberto. JJ; Setzer. RW; Birnbaum. LS. (2004).
CYP1A2 is not required for 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced
immunosuppression. Toxicology 197: 15-22. http://dx.doi.org/10.1016/i.tox.2003.1 1.016.
Smialowicz. RJ; DeVito. MJ; Williams. WC; Birnbaum. LS. (2008). Relative potency based on
hepatic enzyme induction predicts immunosuppressive effects of a mixture of
PCDDS/PCDFS and PCBS. Toxicol Appl Pharmacol 227: 477-484.
http://dx.doi.org/10.1016/i .taap.2007.11.018.
Smith. FA; Schwetz. BA; Nitschke. KD. (1976). Teratogenicity of 2,3,7,8-tetrachlorodibenzo-p-
dioxin in CF-1 mice. Toxicol Appl Pharmacol 38: 517-523.
http://dx.doi.org/10.1016/0041 -008X(76)90183-6.
Sparschu. G. . L.; Dunn. F. . L.; Rowe. V. . K. (1971). Study of the teratogenicity of 2,3,7,8-
tetrachiorodibenzo-p-dioxin in the rat. Food Cosmet Toxicol 9: 405-412.
http://dx.doi.org/10.1016/0015-6264(71 )90045-9.
Staskal. DF; Diliberto. JJ; DeVito. MJ; Birnbaum. LS. (2005). Inhibition of human and rat
CYP1A2 by TCDD and dioxin-like chemicals. Toxicol Sci 84: 225-231.
http://dx.doi.org/10.1093/toxsci/kfi090.
Stavner. L; Bailer. AJ; Smith. R; Gilbert. S; Rice. F; Kuempel. E. (1999). Sources of uncertainty
in dose-response modeling of epidemiological data for cancer risk assessment. Ann N Y
Acad Sci 895: 212-222. http://dx.doi.org/10.1 111/i. 1749-6632.1999.tb08087.x.
Steenland. K; Piacitelli. L; Deddens. J; Fingerhut. M; Chang. LI. (1999). Cancer, heart disease,
and diabetes in workers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. J Natl Cancer
Inst 91: 779-786.
Steenland. K; Deddens. J; Piacitelli. L. (2001). Risk assessment for 2,3,7,8-tetrachlorodibenzo-p-
dioxin (TCDD) based on an epidemiologic study. Am J Epidemiol 154: 451-458.
5-19
DRAFT - DO NOT CITE OR QUOTE
-------
Steinmaus. C; Miller. MP; Smith. AH. (2010). Perchl orate in drinking water during pregnancy
and neonatal thyroid hormone levels in California. J Occup Environ Med 52: 1217-1524.
http://dx.doi.org/10.1097/JQM.0b013e3181fd6fa7.
Sugita-Konishi, Y; Kobavashi. K; Naito. H; Miura. K; Suzuki. Y. (2003). Effect of lactational
exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin on the susceptibility to Listeria infection.
Biosci Biotechnol Biochem 67: 89-93.
Swan. SH; Brazil. C: Drobnis. EZ; Liu. F; Kruse. RL; Hatch. M; Redmon. JB; Wang. C:
Overstreet. JW. (2003). Geographic differences in semen quality of fertile U.S. males.
Environ Health Perspect 111: 414-420. http://dx.doi.org/10.1289/ehp.5927.
Toide. K; Yamazaki. JH; Nagashima. R; Itoh. K; Ivvano. S; Takahashi. Y; Watanabe. S;
Kamataki. T. (2003). Aryl hydrocarbon hydroxylase represents CYP1B1 and not
CYP1A1, in human freshly isolated white cells: Trimodal distribution of Japanese
population according to induction of CYP1B1 mRNA by environmental dioxins. Cancer
Epidemiol Biomarkers Prev 12: 219-222.
Toth. K; Somfai-Relle. S: Sugar. J: Bence. J. (1979). Carcinogenicity testing of herbicide 2,4,5-
trichlorophenoxyethanol containing dioxin and of pure dioxin in Swiss mice. Nature 278:
548-549.
Tritscher. AM: Goldstein. J A; Portier. CJ: McCoy. Z; Clark. GC; Lucier. GW. (1992). Dose-
response relationships for chronic exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin in a rat
tumor promotion model: Quantification and immunolocalization of CYP1A1 and
CYP1A2 in the liver. Cancer Res 52: 3436-3442.
U.S. EPA (U.S. Environmental Protection Agency). (1986a). Guidelines for mutagenicity risk
assessment. (EPA/630/R-98/003). Washington, DC: U.S. Environmental Protection
Agency, Risk Assessment Forum, http://vvvvvv.epa.gov/iris/backgrd.html.
U.S. EPA (U.S. Environmental Protection Agency). (1986b). Guidelines for the health risk
assessment of chemical mixtures. (EPA/630/R-98/002). Washington, DC: U.S.
Environmental Protection Agency, Risk Assessment Forum.
http://cfpub.epa. gov/ncea/cfm/recordisplav.cfm?deid=22567.
U.S. EPA (U.S. Environmental Protection Agency). (1991). Guidelines for developmental
toxicity risk assessment. (EPA/600/FR-91/001). Washington, DC: U.S. Environmental
Protection Agency, Risk Assessment Forum, http://vvvvvv.epa. gov/i ri s/back grd. htm 1.
U.S. EPA (U.S. Environmental Protection Agency). (1994). Methods for derivation of inhalation
reference concentrations and application of inhalation dosimetry. (EPA/600/8-90/066F).
Research Triangle Park, NC: U.S. Environmental Protection Agency, Office of Research
and Development, Office of Health and Environmental Assessment, Environmental
Criteria and Assessment Office.
http://cfpub.epa. gov/ncea/cfm/recordisplav.cfm?deid=71993.
U.S. EPA (U.S. Environmental Protection Agency). (1996). Guidelines for reproductive toxicity
risk assessment. (EPA/630/R-96/009). Washington, DC: U.S. Environmental Protection
Agency, Risk Assessment Forum.
http://vvvvvv.epa.gov/raf/publications/pdfs/REPR051 PDF.
U.S. EPA (U.S. Environmental Protection Agency). (1998). Guidelines for neurotoxicity risk
assessment. (EPA/630/R-95/001F). Washington, DC: U.S. Environmental Protection
Agency, Risk Assessment Forum.
http://vvvvvv.epa.gov/raf/publications/pdfs/NEUROTQX.PDF.
5-20
DRAFT - DO NOT CITE OR QUOTE
-------
U.S. EPA (U.S. Environmental Protection Agency). (2000). Benchmark dose technical guidance
document [external review draft], (EPA/630/R-00/001). Washington, DC: U.S.
Environmental Protection Agency, Risk Assessment Forum.
http://www.epa.gov/raf/publications/benchmark-dose-doc-draft.htm.
U.S. EPA (U.S. Environmental Protection Agency). (2002). Guidelines for ensuring and
maximizing the quality, objectivity, utility, and integrity, of information disseminated by
the Environmental Protection Agency. (EPA/260/R-02/008). Washington, DC.
http://www.epa.gov/qualitv/informationguidelines/documents/EPA InfoOualitvGuideline
s.pdf.
U.S. EPA (U.S. Environmental Protection Agency). (2003). Exposure and human health
reassessment of 2,3,7,8 tetrachlorodibenzo-p dioxin (TCDD) and related compounds
[NAS review draft], (EPA/600/P-00/001). Washington, DC: U.S. Environmental
Protection Agency, National Center for Environmental Assessment.
http://www.epa.gov/nceawwwl/pdfs/dioxin/nas-review/.
U.S. EPA (U.S. Environmental Protection Agency). (2005a). Guidelines for carcinogen risk
assessment. (EPA/630/P-03/001F). Washington, DC.
http ://www. epa. gov/cancerguidelines/.
U.S. EPA (U.S. Environmental Protection Agency). (2005b). Supplemental guidance for
assessing susceptibility from early-life exposure to carcinogens. (EPA/630/R-03/003F).
Washington, DC: U.S. Environmental Protection Agency, Risk Assessment Forum.
http://www.epa.gov/cancerguidelines/guidelines-carcinogen-supplement.htm.
U.S. EPA (U.S. Environmental Protection Agency). (2006). Approaches for the application of
physiologically based pharmacokinetic (PBPK) models and supporting data in risk
assessment (final report). (EPA/600/R-05/043F). Washington, DC: U.S. Environmental
Protection Agency, Office of Research and Development.
U.S. EPA (U.S. Environmental Protection Agency). (2008a). 2,3,7,8-Tetrachlorodibenzo-p-
dioxin (TCDD) dose-response studies: Preliminary literature search results and request
for additional studies. (EPA/600/R-08/119). Washington, DC: U.S. Environmental
Protection Agency, National Center for Environmental Assessment.
http://cfpub.epa.gov/nceaycfm/recordisplav.cfm7deidH99923.
U.S. EPA (U.S. Environmental Protection Agency). (2008b). Framework for application of the
toxicity equivalence methodology for polychlorinated dioxins, furans, and biphenyls in
ecological risk assessment. (EPA/100/R 08/004). Washington, DC: U.S. Environmental
Protection Agency, Office of Science Advisor,
http ://www. epa. gov/raf/t effram ework/i ndex. htm.
U.S. EPA (U.S. Environmental Protection Agency). (2009a). Integrated risk information system
(IRIS) Retrieved June 24, 2009, from http://www.epa.gov/iris/index.html
U.S. EPA (U.S. Environmental Protection Agency). (2009b). Summary of U.S. EPA dioxin
workshop: February 18-20, 2009, Cincinnati, Ohio. (EPA/600/R-09/027). Cincinnati,
OH: U.S. Environmental Protection Agency, National Center for Environmental
Assessment, http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=205603.
U.S. EPA (U.S. Environmental Protection Agency). (2009c). Using probabilistic methods to
enhance the role of risk analysis in decision-making with case study examples.
(EPA/100/R-09/001). Washington, DC.
5-21
DRAFT - DO NOT CITE OR QUOTE
-------
U.S. EPA (U.S. Environmental Protection Agency). (2010a). EPA's reanalysis of key issues
related to dioxin toxicity and response to NAS comments. (EPA/600/R-10/038A).
Washington, DC.
U.S. EPA (U.S. Environmental Protection Agency). (2010b). Recommended Toxicity
Equivalence Factors (TEFs) for Human Health Risk Assessments of 2,3,7,8-
Tetrachlorodibenzo-p- Dioxin and Dioxin-Like Compounds. (EPA/100/R-10/005).
Washington, DC. http://www.epa.gov/raf/files/tefs-for-dioxin-epa-00-r-10-005-fmal.pdf.
Van den Berg. M; Birnbaum. L; Bosveld. AT; Brunstrom. B; Cook. P; Feelev. M; Giesv. JP;
Hanberg. A; Hasegawa. R; Kennedy. SW; Kubiak. T; Larsen. JC; van Leeuwen. FX;
Li em. AK; Nolt. C; Peterson. RE; Poel linger. L; Safe. S; Schrenk. D; Tillitt. D; Tysklind.
M; Younes. M; Waern. F; Zacharewski. T. (1998). Toxic equivalency factors (TEFs) for
PCBs, PCDDs, PCDFs for humans and wildlife. Environ Health Perspect 106: 775-792.
Van Birgelen. AP; Van der Kolk. J; Fase. KM; Bol. I; Poiger. H; Brouwer. A; Van den Berg. M.
(1995a). Subchronic dose-response study of 2,3,7,8-tetrachlorodibenzo-p-dioxin in
female Sprague-Dawley rats. Toxicol Appl Pharmacol 132: 1-13.
http://dx.doi.org/10.1006/taap. 1995.1080.
Van Birgelen. AP; Smit. EA; Kampen. IM; Groeneveld. CN; Fase. KM; Van der Kolk. J; Poiger.
H; Van den Berg. M; Koeman. JH; Brouwer. A. (1995b). Subchronic effects of 2,3,7,8-
TCDD or PCBs on thyroid hormone metabolism: Use in risk assessment. Eur J
Pharmacol 293: 77-85. http://dx.doi.org/10.1016/0926-6917(95)90021 -7.
van Birgelen. AP; van den Berg. M. (2000). Toxicokinetics. Food Addit Contain 17: 267-273.
Van Den Hove. MF; Beckers. C; Devlieger. H; De Zegher. F; De Nayer. P. (1999). Hormone
synthesis and storage in the thyroid of human preterm and term newborns: effect of
thyroxine treatment. Biochimie 81: 563-570.
Van den Berg. M; de Vroom. E; 01 ie. K; Hutzinger. O. (1986). Bioavailability of PCDDs and
PCDFs of fly ash after semi-chronic oral ingestion by guinea pig and Syrian golden
hamster. Chemosphere 15: 519-533.
Van den Berg. M; Birnbaum, LS; Denison. M; De Vito. M; Farland. W; Feeley. M; Fiedler. H;
Hakansson. H; Hanberg. A; Haws. L; Rose. M; Safe. S; Schrenk. D; Tohvama. C;
Tritscher. A; Tuomisto. J; Tysklind. M; Walker. N; Peterson. RE. (2006). The 2005
World Health Organization reevaluation of human and mammalian toxic equivalency
factors for dioxins and dioxin-like compounds. Toxicol Sci 93: 223-241.
http://dx.doi.org/10.1093/toxsci/kfl055.
Van der Molen. G; Kooijman. A; Slob. W. (1996). A generic toxicokinetic model for persistent
lipophilic compounds in humans: An application to TCDD. Fundam Appl Toxicol 31:
83-94. http://dx.doi.org/10.1093/toxsci/3 1.1.83.
Van der Molen. GW; Kooijman. SAL. M; Michalek. JE; Slob. W. (1998). The estimation of
elimination rates of persistent compounds: A re-analysis of 2,3,7,8-tetrachlorodibenzo-p-
dioxin levels in Vietnam veterans. Chemosphere 37: 1833-1844.
http://dx.doi.Org/l 0.1016/S0045-653 5(98)00249-5.
Van der Molen. GW; Kooijman. BA; Wittsiepe. J; Schrev. P; Flesch-Janvs. D; Slob. W. (2000).
Estimation of dioxin and furan elimination rates with a pharmacokinetic model. J Expo
Anal Environ Epidemiol 10: 579-585.
Vanden Heuvel. JP; Clark. GC; Kohn. MC; Tritscher. AM; Greenlee. WF; Lucier. GW; Bell.
DA. (1994). Dioxin-responsive genes: examination of dose-response relationships using
quantitative reverse transcriptase-polymerase chain reaction. Cancer Res 54: 62-68.
5-22
DRAFT - DO NOT CITE OR QUOTE
-------
Vanden Heuvel. JP; Clark. GC; Tritscher. A; Lucier. GW. (1994). Accumulation of
polychlorinated dibenzo-p-dioxins and dibenzofurans in liver of control laboratory rats.
Fundam Appl Toxicol 23: 465-469. http://dx.doi.Org/10.1093/toxsci/23.3.465.
Vos, JG; Moore. JA; Zinkl, JG. (1973). Effect of 2,3,7,8-tetrachlorodibenzo-p-dioxin on the
immune system of laboratory animals. Environ Health Perspect 5: 149-162.
Walker. NJ: Portier. CJ: Lax. SF; Crofts. FG: Li. Y; Lucier. GW: Sutter. TR. (1999).
Characterization of the dose-response of CYP1B1, CYP1A1, and CYP1A2 in the liver of
female Sprague-Dawley rats following chronic exposure to 2,3,7,8-tetrachlorodibenzo-p-
dioxin. Toxicol Appl Pharmacol 154: 279-286. http://dx.doi.org/10.1006/taap.1998.8595.
Wang, SL; Su, PH; Jong, SB; Guo. YL; Chou. WL; Papke. O. (2005). In utero exposure to
dioxins and polychlorinated biphenyls and its relations to thyroid function and growth
hormone in newborns. Environ Health Perspect 113: 1645-1650.
http://dx.doi.org/10.1289/ehp.7994.
Wang, X: Santostefano. MJ; Evans. MY: Richardson. VM; Diliberto. J J; Birnbaum, LS. (1997).
Determination of parameters responsible for pharmacokinetic behavior of TCDD in
female Sprague-Dawley rats. Toxicol Appl Pharmacol 147: 151-168.
http://dx.doi.org/10.1006/taap. 1997.8242.
Wang. X: Santostefano. MJ; DeVito. MJ; Birnbaum. LS. (2000). Extrapolation of a PBPK model
for dioxins across dosage regimen, gender, strain, and species. Toxicol Sci 56: 49-60.
Warner, M; Eskenazi. B; Mocarelli. P; Gerthoux. PM; Samuels. S; Needham. L; Patterson. D;
Brambilla. P. (2002). Serum dioxin concentrations and breast cancer risk in the seveso
women's health study. Environ Health Perspect 110: 625-628.
Warner. M; Samuels. S; Mocarelli. P; Gerthoux. PM; Needham. L; Patterson. DG. Jr; Eskenazi.
B. (2004). Serum dioxin concentrations and age at menarche. Environ Health Perspect
1 12: 1289-1292. http://dx.doi.org/10.1289/ehp.7004.
Warner. M; Eskenazi. B; Patterson. DG. Jr; Clark. G; Turner. WE; Bon si gnore. L; Mocarelli. P;
Gerthoux. PM. (2005). Dioxin-like TEQ of women from the Seveso, Italy area by ID-
HRGC/HRMS and CALUX. J Expo Anal Environ Epidemiol 15: 310-318.
http://dx.doi. org/10.1038/si.iea.7500407.
Warner, M; Eskenazi. B; Olive. PL; Samuels. S; Quick-Miles. S; Vercellini. P; Gerthoux. PM;
Needham. L; Patterson. DG. Jr; Mocarelli. P. (2007). Serum dioxin concentrations and
quality of ovarian function in women of seveso. Environ Health Perspect 115: 336-340.
http://dx.doi.org/10.1289/ehp.9667.
Weber. LW; Lebofsky. M; Stahl. BU; Smith. S; Roztnan. KK. (1995). Correlation between
toxicity and effects on intermediary metabolism in 2,3,7,8-tetrachlorodibenzo-p-dioxin-
treated male C57BL/6J and DBA/2J mice. Toxicol Appl Pharmacol 131: 155-162.
http://dx.doi.org/10.1006/taap. 1995.1057.
Weber. R; Schmitz. HJ; Schrenk. D; Hagenmaier. H. (1997). Metabolic degradation, inducing
potency, and metabolites of fluorinated and chlorinated-fluorinated dibenzodioxins and
dibenzofurans. Chemosphere 34: 29-40. http://dx.doi.org/10.1016/S0045-6535(96)00365-
7.
Wendling. JM; Orth. RG; Poiger. H. (1990). Determination of [3H]-2,3,7,8-tetrachlorodibenzo-
p-dioxin in human feces to ascertain its relative metabolism in man. Anal Chem 62: 796-
800. http://dx.doi.org/10.102l/ac00207a005.
5-23
DRAFT - DO NOT CITE OR QUOTE
-------
White. KL. Jr; Lvsv. HH; McCav. JA; Anderson. AC. (1986). Modulation of serum complement
levels following exposure to polychlorinated dibenzo-p-dioxins. Toxicol Appl Pharmacol
84: 209-219. http://dx.doi.oru/10.1016/0041 -008X(86)90128-6.
WHO (World Health Organization). (1980). WHO laboratory manual for the examination of
human semen and sperm-cervical mucus interaction. Singapore: Press Concern.
WHO (World Health Organization). (1998). Executive summary: Assessment of the health risk
of dioxins: Re-evaluation of the tolerable daily intake (TDI). Geneva, Switzerland: WHO
European Centre for Environmental Health and International Programme on Chemical
Safety.
WHO (World Health Organization). (2007). Assessment of iodine deficiency disorders and
monitoring their elimination. Geneva, Switzerland: WHO Press.
http://vvvvvv.vvho.int/nutrition/publications/micronutrients/iodine deficiencv/9789241595
827/en/index.html.
WHO/UN1CEF/1CC1DD (World Health Organization/ United Nations Children's Fund/
International Council for the Control of Iodine Deficiency Disorders). (1994). Indicators
for assessing iodine deficiency disorders and their control through salt iodization.
(WHO/NUT/94.6). Geneva: World Health Organization.
Wiebkin. P; Fry. JR: Jones. CA; Lowing. R; Bridges. JW. (1976). The metabolism of biphenyl
by isolated viable rat hepatocytes. Xenobiotica 6: 725-743.
http://dx.doi.org/10.3109/0049825760915139Q.
Wiichman. JG: de Wolf. BT; Graafe. R; Arts. EG. (2001). Variation in semen parameters derived
from computer-aided semen analysis, within donors and between donors. J Androl 22:
773-780.
Wittsiepe. J: Erlenkamper. B; Welge. P; Hack. A: Wilhelm. M. (2007). Bioavailability of
PCDD/F from contaminated soil in young Goettingen minipigs. Chemosphere 67: S355-
S364. http://dx.doi.org/10.1016/i.chemosphere.2006.05.129.
Wong. TK; Domin. BA; Bent. PE; Blanton. TE; Anderson. MW: Phil pot. RM. (1986).
Correlation of placental microsomal activities with protein detected by antibodies to
rabbit cytochrome P-450 isozyme 6 in preparations from humans exposed to
polychlorinated biphenyls, quaterphenyls, and dibenzofurans. Cancer Res 46: 999-1004.
Yang. JZ; Agarvval. SK; Foster. WG. (2000). Sub chronic exposure to 2,3,7,8-tetrachlorodibenzo-
p-dioxin modulates the pathophysiology of endometriosis in the cynomolgus monkey.
Toxicol Sci 56: 374-381.
Zareba. G; Hojo, R; Zareba. KM; Watanabe. C; Markovvski. VP: Baggs. RB; Weiss. B. (2002).
Sexually dimorphic alterations of brain cortical dominance in rats prenatally exposed to
TCDD. J Appl Toxicol 22: 129-137. http://dx.doi.org/10.1002/iat.839.
Zoeller. RT; Rovet. J. (2004). Timing of thyroid hormone action in the developing brain: clinical
observations and experimental findings. J Neuroendocrinol 16: 809-818.
http://dx.doi.org/10.1111/i. 1365-2826.2004.01243.x.
5-24
DRAFT - DO NOT CITE OR QUOTE
------- |