The Toxicologist - Society of Toxicology

showed incidences of lung and nasal cancer, respectively. These bioassay findings
have raised questions of human relevance and cancer risk. To address this question,
studies of industries with naphthalene-containing streams having the most extensive
epidemiology data available were reviewed; this included studies involving petroleum
refining, asphalt (paving and roofing), and creosote production, as well as
jet fuel handlers. The review focused on respiratory cancer (nasal and lung cancer)
as these are the tumors types seen in the animal studies. The review identified no
epidemiology studies of workers exposed only to naphthalene. There are limited
case reports of laryngeal and colorectal cancer related to naphthalene exposure, but
expert bodies have concluded these data are inadequate for evaluating human cancer
risk. Notably, there are no case reports of nasal tumors for workers in industries
with naphthalene-containing streams. The lack of case reports for a rare tumor like
nasal cancer is important given that case reports (not formal cohort studies) have
historically identified several occupational carcinogens that cause rare tumor types
(e.g., vinyl chloride and angiosarcoma of the liver). When combined with results of
Magee et al. (2010) indicating that the cancer potency estimates based on the rat
NTP nasal tumor findings significantly overestimate the predicted total number of
nasal tumors in the U.S. population, the relevance of rat nasal tumors to humans is
questionable. For lung cancer, the lack of case reports is not informative due to the
high background rate of this tumor and its relation to smoking. Further interpretation
of the human data for lung cancer is complicated by limitations of cohort
and/or case-control studies (e.g., low exposure, confounding by smoking), although
no occupationally-related risks were identified in the industries evaluated
1957 NAPHTHALENE RESEARCH: THE RELEVANCE OF
TUMORS IN RODENTS TO HUMAN RISK
ASSESSMENT.
A. LeHuray 1 and K. Wise 2 . 1 Naphthalene Council, Alexandria, VA and 2 American
Petroleum Institute, Washington, DC.
Industry associations and individual companies formed the Naphthalene Research
Committee (NRC) to co-sponsor research that strives to improve naphthalene risk
assessments. The NRC’s objective is to reduce the use of default assumptions in assessing
cancer risks potentially posed by exposure to naphthalene. To further its objective,
the NRC funds research that is published in the peer-reviewed scientific literature.
Results of several completed NRC sponsored studies could significantly
change a quantitative risk assessment by replacing default values and identifying
species-specific modes of action. Acute (single 6-hour) and 5 day exposure studies
in rats demonstrated that high concentrations used in NTP two-year studies caused
severe damage to the olfactory epithelium - a target tissue for subsequent tumor formation.
Respiratory uptake studies of naphthalene in rats demonstrate that use of
the Category 1 gas model is an inappropriate default assumption in assessing naphthalene
risk. Results of a NRC 90-day exposure study establish a naphthalene
NOEL in the rat, and demonstrate a remarkable degree of post-exposure recovery.
Tissues collected following that 90-day exposure have been used in a p53 mutation
study, genomics analysis and detailed metabolism mass balance and biomarker
studies. NRC dose-response in vitro studies in various tissue types from rats, mice
and humans demonstrate species and organ specific effects. A PBPK model incorporating
naphthalene-specific data rather than regulatory default assumptions is
nearing completion. Physiologic and metabolic differences between rodents and
primates exposed to naphthalene introduce significant uncertainty into the applicability
of studies conducted in rats and mice to determine human risk assessment.
NRC is using the key events framework and the Hypothesis-Based Weight of
Evidence paradigm to evaluate assumptions, alternative hypotheses, and guide development
of data needed for a robust mode of action hypothesis.
1958 PUTATIVE MECHANISMS OF ENVIRONMENTAL
CHEMICAL-INDUCED STEATOSIS.
J. Kaiser. Oak Ridge Institute for Science and Education, Oak Ridge, TN.
Non-alcoholic fatty liver disease (NAFLD) is the most common chronic liver disease
in the U.S. occurring in nearly 30% of the population. NAFLD is characterized
by several pathological changes with steatosis (or fatty liver) representing the
initial step in its pathogenesis. Steatosis is of critical importance because the prevention
of fatty liver can not only obviate the downstream pathologies of NAFLD
(i.e. steatohepatitis, fibrosis, and cirrhosis) but also reduces the cardiometabolic risk
associated with this form of liver disease. One of the major causes of NAFLD is
thought to be obesity, but recent studies have shown a strong correlation between
chemical exposures and fatty liver induction unrelated to being overweight. The
U.S. EPA’s Integrated Risk Information System (IRIS) is a human health assessment
program that evaluates risk information on effects that may result from exposure
to environmental contaminants. The goal of our work was to distinguish IRIS
chemicals that induce steatosis and investigate putative mechanism(s) by which
these chemicals may contribute to this hepatic pathological condition.
Interestingly, a majority of steatosis-causing IRIS chemicals were chlorinated compounds
and we hypothesize that they induce fatty liver by hepatic mitochondrial
function impairment. Furthermore, the most toxic chemicals with respect to fatty
liver induction (e.g., mirex, chlordane, chlordecone) had the greatest degree of
chlorination. We also found that although it is likely important, mitochondrial dysfunction
may not be the sole mediator of chemical-induced steatosis. Insulin resistance,
impaired hepatic lipid secretion, and enhanced cytokine production were
other potential mechanisms by which IRIS chemicals such as vinyl chloride, carbon
tetrachloride, and alachlor, respectively, could contribute to hepatic steatosis. Taken
together, the work described here is significant because it identifies multiple mechanisms
by which specific IRIS chemicals may cause fatty liver and expands upon
our knowledge of the possible role of environmental chemical exposure in the early
progression of NAFLD.
1959 POLYCHLORINATED BIPHENYLS ARE NOT
“DIOXINS”, AND OTHER INCONVENIENT TRUTHS
RELEVANT TO THE USE OF TOXIC EQUIVALENCY
FACTORS (TEFS) IN HUMAN HEALTH RISK
ASSESSMENT.
J. B. Silkworth and E. A. Carlson. General Electric, Niskayuna, NY.
Various health effects have been documented following exposure to polychlorinated
dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs). The
same cannot be said for similar exposures to polychlorinated biphenyls (PCBs),
where any associated health effects have been attributed to non-PCB contaminants.
So why are some PCB congeners referred to as “dioxin equivalents” in risk assessment?
The answer lies in the perpetual use, for over seventy years, of ultra-sensitive
animal models and PCB preparations of unknown purity to predict PCB toxicity to
humans. This body of scientific literature is the basis of the current TEFs. There are
a variety of reasons why the TEF scheme should not be employed to assess the risk
of PCB mixtures to human health. First, PCBs are not “structural equivalents” of
dioxins, cannot achieve rigid planar conformations, and interact within the aryl hydrocarbon
receptor (AHR) ligand-binding domain (LBD) in a manner dissimilar to
that of potent AHR agonists. This results in comparatively weaker binding affinities
for PCBs and relegates most congeners as only partial agonists. In addition, differential
recruitment of co-factors has been documented between PCBs and dioxins,
possibly affecting AHR transactivation in a ligand-specific manner. Second, the
human AHR differs significantly from those of sensitive animal species. An amino
acid substitution in the human AHR LBD pushes its affinity for the most potent
PCB (PCB 126) into the micromolar range. This genetic difference, along with
even greater AHR protein degeneracy among species outside the LBD, appear to result
in severely reduced efficacies for all PCB congeners in humans. Lastly, relative
potencies for AHR ligands differ significantly between responsive species and humans
in a congener-specific manner. This final fact has been proven for several PCB
congeners by multiple investigators, using different cell types. Overall, congenerand
species-specific variation in AHR activity clearly limits the applicability of
TEFs for human health risk assessment.
1960 APPLICATION OF LINEAR AND NON-LINEAR DOSE-
RESPONSE MODELS FOR POLYCHLORINATED
BIPHENYLS (PCBs) BASED ON A RECEPTOR-
MEDIATED MODE OF ACTION (MOA) FOR TUMOR
PROMOTION.
R. E. Keenan 1 , P. O. Gwinn 1 , J. M. Hamblen 1 , J. Shoenfelt 2 and J. B.
Silkworth 3 . 1 Integral Consulting Inc., Portland, ME, 2 Integral Consulting Inc.,
Annapolis, MD and 3 General Electric Global Research, Niskayuna, NY.
A nonlinear dose-response model for polychlorinated biphenyls (PCBs) was developed,
based on a receptor-mediated mode-of-action (MOA) for PCB-promoted rat
liver tumors. Tolerable Daily Intakes (TDIs) for several PCB mixtures and congeners,
which had been tested in suitable lifetime rodent bioassays, were derived
using the nonlinear model. The TDIs were then compared to risk-specific doses calculated
using the linear default extrapolation approach that is currently used by
EPA. The statistical dose-response analysis was conducted using EPA Benchmark
Dose Software to model dose response in the observable region of the data in order
to extrapolate to a point-of-departure (POD) at or near the fringe of the observable
region. A set of conservative toxicological uncertainty factors (UFs) was used for extrapolating
below the POD to the threshold dose anticipated by the MOA. This
analysis led to the determination of TDIs that are substantially different than the
corresponding risk-specific doses (10 -4 -to-10 -6 risk range) for each PCB mixture or
congener. The nonlinear dose-response assessment indicates that much higher levels
of PCBs can be tolerated without subjecting individuals to unreasonable excess
cancer risks. These results have important regulatory policy implications for setting
environmental tolerances and for determining acceptable cleanup levels of PCBs in
soil, sediment, and water.
SOT 2011 ANNUAL MEETING 419