|
|
 |
2000 SOT Meeting
Reinventing MIXTOX: Priority Chemical Mixtures and the
Interactions-Adjusted Hazard Index
J. Colman1,
R. Hertzberg2, P. McClure1, M. Odin1,
F. Llados1, W. Stiteler1 and D. A. Gray1
1Syracuse
Research Corporation, North Syracuse, NY, USA; 2National
Center for Environmental Assessment (NCEA), U.S. EPA, Cincinnati, OH,
USA. Sponsor P. McGinnis.
Abstract
Feedback from Regional Superfund risk assessors indicated that EPA’s
original Mixtox interactions data base was only marginally useful; it
provided no indication of the numerical change in the Hazard Index due
to interactions, and thus no quantitative alterations were being made in
the site assessments. This pilot project initiates the transformation of
EPA’s Mixtox data base into a tool directly useful for Superfund
baseline risk assessment. Binary chemical mixtures (chemical pairs) were
prioritized by screening analytical results from final National Priority
List (NPL) sites using RfD-based screening concentrations, followed by
application of an algorithm to rank the frequency of toxicologically
significant binary mixtures at NPL sites. Priority inorganic and organic
chemical pairs were selected from the top-ranking pairs.
Trichloroethylene and lead were the predominant chemicals in the organic
and inorganic priority pairs, respectively. Binary weight-of-evidence
(WOE) determinations for interactions were then developed for 12
priority pairs using the current EPA mixture risk methodology. Of the 24
WOE classifications of joint toxic action for these 12 pairs, three were
"greater than additive," four were "less than additive," one was
"adequate evidence of additivity," and the rest were default
classifications of additivity (due to conflicting or inadequate
information). The WOE classifications, converted to numerical scores,
are used to estimate an interactions-adjusted hazard index, as
demonstrated in a software mock-up. (This abstract does not necessarily
reflect EPA policy.)
Validation of a Cadmium Dietary Exposure Model (CDEM) for use in Risk
Assessment
G.L. Diamond1,
W.C. Thayer1,2, H. Choudhury3, T.F. Lockwood1,
W.M. Stiteler1, P.E. Goodrum1 and J.M. Hassett2
1Syracuse
Research Corporation, Syracuse, NY; 2State University of New
York College of Environmental Science and Forestry, Syracuse NY; 3U.S.
EPA, NCEA, Cincinnati, OH.
Abstract
The Cadmium Dietary Exposure Model (CDEM), utilizes national survey data
on food cadmium concentrations and food consumption patterns to estimate
dietary intakes in the U.S. population. The CDEM has been linked to a
modification of the cadmium biokinetic model of Kjellström and Nordberg
(KNM) to derive estimates of kidney and urinary cadmium that reflect
U.S. dietary cadmium intake and related variability. Variability in
dietary cadmium intake was propagated through the KNM using a Monte
Carlo approach. The model predicts a mean peak kidney cadmium burden of
approximately 3.5 mg and a 5th-95th percentile range of 2.2-5.1 mg in
males. The corresponding peak renal cortex cadmium concentration in
males is 15 (10-22, 5th-95th % ) m g/g wet cortex. Predicted kidney
cadmium levels in females were higher than males: 5.1 (3.3-7.6) mg total
kidney; 29 (19-43) m g/g wet cortex. Predicted urinary cadmium in males
and females agreed with empirical estimates based on the NHANES III,
with females predicted and observed to excrete approximately twice the
amount of cadmium in urine than males. The predicted 95th percentile
values for peak kidney cadmium burden are approximately 60% of the peak
kidney burden predicted for a chronic intake at the U.S. EPA Reference
Dose of 1 m g/kg-day (8.2 mg). (Supported in part by EPA Cooperative
Agreement CR822761 and EPA Contract 68-C6-0024. Statements in this
report do not reflect opinions or policies of the U.S. EPA).
Evaluation of
Lead Biokinetic Models for Adults
M.A. Maddaloni1,
M.A. Ballew1, M.D. Johnson1, G.A. Khoury1,
K.P. Koporec1, P.A. Van Leeuwen1, M.L. Stifleman1,
R. Troast1, P.D. White1, L.J. Zaragosa1,
G.L. Diamond2
1USEPA,
Washington, DC., 2SRC, Syracuse, NY
Abstract
Responding to a need for a scientifically defensible approach for
assessing human health lead risks at non-residential Superfund sites
(where adults rather than children are the primary receptors), the U.S.
EPA’s Technical Review Workgroup for Lead (TRW) developed the Adult Lead
Model (ALM) in 1996. Rather than incorporating a complex compartmental
or PBPK type kinetic component, the ALM employs a simplified slope
factor approach that relates lead uptake to blood lead. The need for
immediate guidance limited the scope of the approach to a narrowly
defined receptor population (i.e., adult worker at a
commercial/industrial setting) and specific medium (i.e., soil/dust).
Subsequently, the TRW committed to a more exhaustive effort to identify
the most scientifically defensible methodology currently available to
model non-residential lead exposures and risks. The TRW identified seven
lead biokinetic models that have been used to assess the relationship
between environmental lead exposures and blood lead concentration in
adults. The models were evaluated and compared to the ALM based on the
following general evaluation criteria: (1) completeness of exposure
module, (2) kinetic performance, (3) utility of model output, and (4)
ease of use and flexibility. Possible outcomes of the evaluation were:
replace ALM with a superior model, modify ALM, or retain the existing
ALM. Although no single model was judged to be a significant improvement
over the ALM, various components from the different models were
determined to be refinements in adult lead modeling. Specific model
improvements included multimedia exposure modules, and highly versatile
kinetic modules. Rather than invest in integrating beneficial components
into a hybrid model, the TRW recommends retaining the ALM - while
appreciating that certain site-specific risk assessments (e.g.,
short-term lead modeling) may benefit from analysis by alternative
models - and supporting a research initiative to develop an all-ages
biokinetic model that may adopt some of the more attractive features of
existing models. It is noteworthy that the kinetic performance of the
models evaluated produced similar estimates of quasi-steady-state blood
lead when exposure parameters were normalized across models (i.e., all
were set to approximate ALM inputs).
Proposed New
Risk Assessment for Acetaldehyde
M. Osier1,
M. Odin1, P. McGinnis2, and B. Boutin3
1Syracuse
Research Corp., North Syracuse, NY; 2Philadelphia, PA; 2National
Center for Environmental Assessment, U.S. EPA, Cincinnati, OH.
Abstract
EPA is in the process of updating the risk assessment for acetaldehyde.
Animal studies have demonstrated irritant effects of acetaldehyde
following both oral and inhalation exposure, with the critical effect
for both routes of exposure being hyper- and metaplasia of the tissues
along the portal of entry. While oral carcinogenicity data are
unavailable, chronic inhalation studies have demonstrated tumor
formation in the nasal cavity, with tumors being located in the same
areas that the noncancer effects are seen. The RfD (Oral Reference Dose)
is based on a published NOAEL of 125 mg/kg-day for hyper- and
metaplastic responses in the forestomach of male Wistar rats. An
uncertainty factor (UF) of 3000, (10 for intraspecies variation; 10 for
interspecies extrapolation; 10 for use of a subchronic study; 3 for
database insufficiencies), was applied to derive an RfD of 4x10_2
mg/kg-day. The RfC (Inhalation Reference Concentration) is based on a
published NOAEL of 390 ppm for hyper- and metaplastic responses in the
posterior nasal region of male and female Syrian golden hamsters.
Following conversion to a NOAEL[HEC] of 5.45 mg/m3, an UF of 1000 (3 for
interspecies variation; 10 for intraspecies variation; 10 for the use of
a subchronic study; 3 for database insufficiencies) was applied to
derive an RfC of 5x10-3 mg/m3. Oral carcinogenicity data are
unavailable. The inhalation unit risk was based on the incidence of
upper respiratory tract tumors in male Wistar rats following a 28-month
exposure to acetaldehyde. An LEC10 of 7.43 mg/m3 was calculated from the
incidence data using a polynomial curve-fitting program (Global86), and
used to derive an inhalation unit risk of 1.3x10-5 (:g/m3)-1. Under
EPA’s 1986 guidelines for carcinogen risk assessment, acetaldehyde is
classified as group B2. Under the proposed 1996 and 1999 guidelines,
acetaldehyde is considered likely to be carcinogenic to humans following
inhalation exposure; data are inadequate for evaluation following oral
exposure. (This abstract does not necessarily reflect EPA policy). |
