You are here:
Modeling mechanistic processes from source to outcome to support evidence integration and inform risk assessment
Citation:
Hines, D., R. Conolly, AND Ann Jarabek. Modeling mechanistic processes from source to outcome to support evidence integration and inform risk assessment. NAS Evidence Integration Workshop, Washington, DC, June 03 - 04, 2019.
Impact/Purpose:
This poster develops approaches for using mechanistic models to support evidence integration and inform risk assessment. It applies the Aggregate Exposure Pathway (AEP) and Adverse Outcome Pathway (AOP) frameworks, along with Physiologically Base Pharmacokinetic (PBPK) models, to a case study of a chemical at a hypothetical contaminated site to demonstrate how an AEP-PBPK-AOP construct can inform quantitative source-to-outcome analyses. We use this case study to show how mechanistic approaches can organize data and identify data gaps while facilitating simultaneous evaluation of risk in human health and ecological endpoints. Furthermore, we discuss how mechanistic models can inform the construction of process models, the assembly and integration of data from systematic review, and the use of ontologies in support of risk assessment.
Description:
Mechanistic models can inform ontologies and evidence maps by providing an organizing framework that describes the causal relationships among key events, biological endpoints, and Adverse Outcomes (AOs). A mechanistic understanding of exposure pathways, behavior, physicochemical properties, Absorption, Distribution, Metabolism, and Elimination (ADME), and toxicity pathways can provide key advantages for risk assessors because it highlights knowledge gaps, informs inferences, and supports mechanistic evidence integration along a source-to-outcome continuum, thereby increasing the confidence in risk assessment results. This work develops a quantitative approach for integrating mechanistic exposure and toxicity data for human health and ecological endpoints using the Aggregate Exposure Pathway (AEP) and Adverse Outcome Pathway (AOP) frameworks. We demonstrate this approach using a case study of a hypothetical site contaminated by a molecule that competitively inhibits iodide uptake into the thyroid at the sodium-iodide symporter (NIS), and thus affects established AOPs for developmental neurotoxicity. External exposure pathways were quantified in an AEP fate-and-transport model describing chemical movement through the site. This model was used to predict NIS-inhibitor exposure and source apportionment for humans, fishes, and small herbivorous mammals under three contamination scenarios. External exposures were linked to a previously published multi-species AOP network for NIS inhibition using physiologically based pharmacokinetic models, then combined with mechanistic dose-response data to calculate a hazard index (HI) for each potential AO in each species. Thus, we demonstrate how a source-to-outcome mechanistic model allows for the display of exposure pathways and key events into a process model of pathways to characterize pathogenesis. The analysis predicted that surface water contamination was the largest contributor to source apportionment of exposure in fishes, while groundwater contamination was the largest contributor in humans and small herbivorous mammals, and quantified changes in these apportionments across scenarios. HI results showed how quantitative evaluation of mechanistic exposure and toxicity pathways facilitated the evaluation of relative risk of AOs in each species across scenarios. This work demonstrates how the AEP-AOP construct can link environmental transport and transformation, exposure, toxicokinetics, and toxicodynamics; as well as inform cumulative risk assessment, by 1) organizing mechanistic data, 2) identifying data gaps, 3) quantifying uncertainties, and 4) facilitating simultaneous evaluation of risk in human health and ecological endpoints. We show how mechanistic models can inform the construction of process models, the assembly and integration of data from systematic review, and the use of ontologies. The views expressed in this poster are those of the authors and do not necessarily reflect the views or policies of the U.S. Environmental Protection Agency.