Modeling Mechanistic Processes from Source to Outcome to Support Evidence Integration and Inform Risk Assessment

&EPA

United States
Environmental Protection
Agency

Modeling mechanistic processes from source to outcome to support
evidence integration and inform risk assessment

David E. Hines1, Rory B. Conolly1, Annie M. Jarabek2

U.S. Environmental Protection Agency, Office of Research and Development; Research Triangle Park, NC;1 National Health and Environmental Effects Research Laboratory; 2 National Center for Environmental Assessment

David E. Hines 1 hines.david@eoa.aov 1 919-541-1469

Introduction

Quantitative Case Study

Discussion

Evidence integration in current IRIS assessments considers the contributions of human health,
animal, and mechanistic data streams according to PECO criteria in a hierarchical and parallel
approach. (Fig. I)

Exposure Estimation

Mild Moderate High

Human

Animal

Mechanistic

Develop
Protocols for
Systematic
Reviews

I I ~~I I I

l Identify I || Evaluate Ipv
'I Evidence |^l| Studies I'

Evidence
Integration

Hazard
Identification

Dose-
Response
Assessment

Scoping

D—[

Problem Formulation

Broad Literature Search

Fig. I: Overview of the IRIS process.
Adapted from NRC (2014)

The NAS has emphasized the use of mechanistic process models of pathogenesis to evaluate
relationships among biomarkers (exposure/effect/susceptibility) as well as modernizing risk
predictions using exposure science and computational models.

We propose mechanistic data should serve as a scaffold for the use of process models when
integrating evidence across human health and ecological endpoints. (Fig. 2)

V©*

Med

99%

Med

99%

Med

99%

t ,.«

99% confidence
interval.

Key Events (see Fig. 3)

= Y A

£-ii=iALi

le+07
le+06-
le+05
10000
. 1000

I 100

12 3 4 5 6 7

2 3 4 5 6 7

EQ I: i is each exposure source, E is the
exposure level, and AL is the acceptable
limit of exposure. AL was the lowest
reported LOAEL for each species

fiiji

¦i 1 T T i

-*r

-jlj-

—

HI = 0.01 to 0.06

U.S. Environmental Protection Agency

Office of Research and Development

Fig. 6: Predicted
internal doses at the
hypothetical site under
different contamination
scenarios compared
with dose-response
toxicity data. See KE
Hines et al. (2018) for
KE details.

Y-axis shows toxicant
dose in pg/kg/d on a
log scale

Fig. 3: Joint AEP-AOP construct for the CIO/ case study.
Detailed description of AOP network in Hines el al. (2018).

HI = 0.3 to 4.0

HI = 10.5 to 46.8

HI =43.7 to 953.7

LOEL/LOAEL

Point of

departure

<•>

Reference dose

Model prediction

Model

extrapolation

NOEL/NOAEL

BMD

BMDL

Median internal

dose

Internal dose

range prediction

The source to outcome case study demonstrates how a workflow for using a mechanistic
scaffold can facilitate evidence integration. (Fig. 7)

Mechanistic
Understanding

• Clarify context for interpretation

• Use exposures to drive risk assessment

• Characterize key events

• Quantify uncertainties using process models

• Facilitate integration of human health and ecological endpoints

Toxicokinetics &
Toxieodynamics

Fig. 7: Benefits of using a mechanistic scaffoldfor evidence integration in risk

• Assembly across
system

• Increased
transparency

• Inform data gaps

• Tailor specific source to
outcome risk
characterization

• Leverage data s<

Integrated Risk
Assessment

The AEP and AOP frameworks facilitate exposure driven risk assessments in support of
assessments required by the new TSCA

Mechanistic approaches to data integration can act as an organizing framework to inform
ontologies or evidence maps, leverage data sources, and facilitate quantitative
characterization of key events in pathogenesis.

Explicit elucidation of key events and parameters supports transparency in risk assessments.

Risk assessments based on exposure use cases and toxicity pathways involved in
pathogenesis allow for more targeted assessment and increased confidence.

Conclusions

A mechanistic scaffold informs problem formulation, aids evaluation of
study quality criteria, and facilitates evidence integration to support
source-to-outcome risk assessments that are:

1) Exposure driven to target specific use-cases

2) Quantitative for key events in relevant AOPs

3) Capable of characterizing human health and
ecological endpoints

Literature Cited & Abbreviations

Ankley, G. T.; Bennett, R. $.; Erickson, R. J.; Hoff, D. J.; Hornung, M. W.; Johnson, R. D.; Mount, D. R.; Nichols, J. W.; Russom, C. L.; Schmieder,
P K.: Serrrano, J. A.. 2010. Adverse outcome pathways; A conceptual framework to support ecotoxicology research and risk assessment. Environ.
Toxicol. Chem. 29 (3), 730-741

Fath, B.D. and Pauen, B.C., 1999. Review of the foundations of network environ analysis. Ecosystems, 2(2), pp. 167-179.

Hines, D.E.; Edwards, S.W.; Conolly, R.B.; Jarabek, A.M., 2018. A case study application of the Aggregate Exposure Pathway (AEP) and Adverse
Outcome Pathway (AOP) frameworks to facilitate the integration of human health and ecological endpoints for Cumulative Risk Assessment
(CRA). Environ. Sci. Technol. 52, 839-849.

Lumen, A., Mattie, D.R., Fisher, J.W., 2013. Evaluation of perturbations in serum thyroid hormones during human pregnancy due to dietary iodide
and perchlorate exposure using a biologically based dose-response model. Toxicological Sciences 133(2), 320-341.

Merrill, E.A., Clewell, R.A., Gearhart, J.M., Robinson. P.J., Sterner, T.R., Yu, K.O., Mattie, D.R. and Fisher, J.W., 2003. PBPK predictions of
perchlorate distribution and its effect on thyroid uptake of radioiodide in the male rat. Toxicological Sciences, 73(2), pp.256-269.

NRC (National Research Council), 2014. Review ofEPA's integrated risk information system (IRIS) process. National Academies Press.

Teeguarden, J.G., Tan, Y„ Edwards, S,W, Leonard, J.A., Anderson, K.A., Corley, R.A., Kile, M.L, Simonich, S.M., Stone, D., Tanquay, R.L., Waters,
K.M., Harper, S.L., Williams, D.E., 2016. Completing the link between exposure science and toxicology for improved environmental health decision
making: The aggregate exposure pathway framework. Environmental Science & Technology 50,4579-4586.

Abbreviations: ADME. Absorption. Distribution. Metabolism and Elimination: AEP. Aggregate Exposure Pathway: AOP. Adverse Outcome
Pathway; BMD, Benchmark Dose; BMDL, Benchmark Dose confidence interval; HI, Hazard Index; IRIS, Integrated Risk Information System;
KE, Key Event; KES, Key Exposure State; LO[A]EL, Lowest Observed [Adverse] Effect Level; NAS; National Academy of Sciences; NIS,
Sodium Iodide Symporter; NO[A]EL, No Observed [Adverse] Effect Level; PBPK, Physiologically Based Pharmacokinetic: PECO,

Population, Exposure, Comparators, Outcomes; TH, Thyroid Hormone; TSE Target Site Exposure; TSCA, Toxic Substances Control Act

Units: pg/kg/d

0% Printed on 100% recycled/recyclable paper

£ with a minimum 50% post-consumer
<¦ W fiber using vegetable-based ink.

Disclaimer: The views expressed in this poster are those of the authors and do not necessarily
represent the views or policies of the U.S. Environmental Protection Agency.

-------