Characterizing Risks of Chemicals Exhibiting Extensive Low Dose Clearance in the Gut, Liver, and Blood: Dietary Chlorpyrifos as a Case Example

Paul S. Price M.S.1, Charles Timchalk2, Torka S. Poet2, Paul Hinderliter2, and Michael Bartels1. 1The Dow Chemical Company. 2Battelle, Pacific Northwest Division.

This project sought to improve risk assessment methodology by incorporating technical advances in computer based modeling to better characterize exposure and potential toxicity. The goals of the project were to: 1) provide an illustration of how existing exposure and physiologically based pharmacokinetic/pharmacodynamic (PBPK/PD) models can be linked to produce an integrated “source-to-outcome” model for a real world example; 2) investigate how such linked models can be used to investigate the impact of variation in age, behavior, and ADME (absorption, distribution, metabolism, and excretion) in exposed populations; and 3) provide example strategies for assessing and accommodating uncertainties associated with the available models and data. The assessment of risks posed by exposures to chemicals is currently determined by a system of technical findings and policy based decisions. Risk assessments are improved when safety factors are replaced with empirical data and when the assessment develops quantitative estimates of risk that reflect uncertainty and variability. One method of developing such estimates is to use computer simulations of specific events in the pathway from chemical exposure to health outcome (e.g. a “source-to-outcome” model). Both academia and the U.S. EPA have programs to develop source-to-outcome models.

Selection of the Chemical and Source of Exposure

Chlorpyrifos was selected as the subject of this project and is an insecticide that has been the subject of numerous research publications. Dietary exposures to residues on crops are regulated based on probabilistic models of exposures. A PBPK/PD model has been developed to describe the distribution, metabolism and the effects of its oxon metabolite on the cholinergic system. The effects of chlorpyrifos on the cholinergic system are mediated by AChE inhibition in the central nervous system, specifically brain AChE inhibition; however, direct measurements of such inhibition cannot be made in humans. Therefore, researchers have used measures of AChE inhibition in red blood cells (RBC AChE inhibition) as a measure exposure for organophosphorous pesticides. The PBPK/PD model for chlorpyrifos has been calibrated for changes in RBC AChE and plasma butyrylcholinesterase from controlled human exposures and has been shown to be consistent with observations of changes in plasma butyrylcholinesterase in occupationally exposed populations.

The relationship between RBC AChE inhibition and observed adverse effects has been investigated for a number of AChE inhibitors. Based on these data, certain governmental agencies have established recommendations for acceptable and unacceptable changes in RBC AChE. For example, the States of Washington and California both consider that a decrease of >30 percent in RBC AChE compared to a worker’s baseline is required before reduction in exposure is necessary and that a worker can return to work when RBC levels return to a level within 20 percent of the worker’s baseline value.

While the subject of this case study, chlorpyrifos, was a specific insecticide, the focus of this project was not the assessment of the safety of this compound. The work was intended to be a real-world example of the application of a source-to-outcome model to the assessment of outcomes in a population exposed to an actual source. Therefore, the focus of this project was on the demonstration of how the existing data and models allow a determination of health outcomes without the use of safety factors and how this approach can serve as an example to be applied to other chemicals.


The project used publicly available software programs to characterize the longitudinal dietary exposure and anthropometry of exposed individuals. These predictions were applied to a validated PBPK/PD model to estimate interindividual and longitudinal variation in biomarkers (chlorpyrifos in blood, TCPy in urine, and RBC AChE inhibition). The predicted levels of chlorpyrifos in blood and TCPy in urine were compared with published measurements. RBC AChE predictions were compared to values associated with reported exposure-related effects in humans to determine the potential for the occurrence of adverse outcomes.


The predicted distributions of biomarkers of exposure were consistent with published values, supporting the predictive value of the exposure and PBPK portions of the source-to-outcome model. Key sources of uncertainty in model predictions were investigated and found to have a modest impact on the model predictions. The final predictions of RBC and brain AChE inhibition from estimated dietary exposure were well below the levels that are reported to be associated with health outcomes.


This project demonstrated how to integrate state-of-the-science exposure, physiologically based pharmacokinetic, and pharmacodynamic models. Although many advances have been made in each of these modeling areas, and each type of model has been used in the risk assessment process, no one has linked the three types of models together to make quantitative predictions of the frequency of specific adverse effects in human populations. This project demonstrated how models can be integrated to predict risk directly and provide an example of how the application of more scientifically advanced risk assessment methods can improve the risk assessment process.


PBPK/PD models, cholinesterase, probabilistic, dietary exposure, biomarkers

Project Start and End Dates

2007 – 2009

Project ID


Peer-reviewed Publication(s)

None to date.

Other Publication(s)

“Source-to-outcome modeling physiologically based pharmacokinetic/pharmacodynamic (PBPK/PD) model linked to a dietary exposure model: Chlorpyrifos as a case study.” Science Issues Paper for Science Advisory Panel Meeting on Chlorpyrifos Physiologically Based Pharmacokinetic/Pharmacodynamic (PBPK/PD) Model linked to the Cumulative and Aggregate Risk Evaluation System (CARES), February 15-18, 2011, Washington, D.C.

* This abstract was prepared by the principal investigator for the project.


News & Resources

View our resource center to find press releases, testimonies, infographics and more.