You are here:
Improving Computational Derivation of PFAS Toxicokinetic Half-lives
Citation:
Tague, M., V. Correa, Beth Horton, M. Huse, E. Rowan, K. Wolf, C. Ring, T. Wall, J. Wambaugh, AND R. Sayre. Improving Computational Derivation of PFAS Toxicokinetic Half-lives. SOT, Nashville, TN, March 19 - 23, 2023. https://doi.org/10.23645/epacomptox.22554895
Impact/Purpose:
The toxicokinetic (TK) half-life - the amount of time required for 50% of the chemical analyte to be eliminated from the body - is a critical metric for characterizing potential adverse health effects. Here we provide newly curated in vivo PFAS TK data and estimations of their half-lives according to a newly updated method.
Description:
Per- and polyfluoroalkyl substances (PFAS) are a diverse class of long-lasting, man-made chemicals that have been used for a variety of industrial and consumer purposes. Long-term exposure to PFAS has been linked to a range of adverse health effects, including immunosuppression and liver damage. PFAS have been found in the environment and in drinking water across the US, and in the blood stream of humans and other animals globally. The toxicokinetic (TK) half-life - the amount of time required for 50% of the chemical analyte to be eliminated from the body - is a critical metric for characterizing potential adverse health effects. We present a cheminformatics-ready process for computationally deriving half-lives of PFAS utilizing the existing EPA/ORD developed invivoPKfit R package. InvivoPKfit is available as an open-source package on GitHub: USEPA/CompTox-ExpoCast-invivoPKfit. Concentration versus time (CvT) experimental data for various PFAS were extracted and harmonized from a variety of literature, including both peer-reviewed studies and grey literature. When applied to the PFAS CvT dataset, invivoPKfit systematically and consistently estimates TK parameters for each combination of chemical and species in the dataset. InvivoPKfit obtains initial estimates of TK parameters, such as volume of distribution (Vd) and elimination rate (kelim), by using non-compartmental heuristics. The package then calculates and optimizes the log-likelihood of the model by varying the TK parameters. InvivoPKfit repeats the process for every unique combination of chemical and species in the dataset. Akaike Information Criterion (AIC), used to estimate prediction error, is calculated for both 1-compartment and 2-compartment models as well as a null (no-time response) model. The model with the lowest AIC is chosen as the best model. The estimated kelim values from the chosen model allow for the direct calculation of chemical half-lives. Derived TK parameters are then made publicly available via the USEPA/CompTox-PK-CvTdb GitHub, and the U.S. EPA Computational Toxicology (CompTox) Chemicals Dashboard. Toxicokinetic half-lives of PFAS provide critical information to stakeholders and decision makers as they seek to understand and mitigate the public health impacts from this widespread and persistent class of chemicals. This abstract does not necessarily reflect U.S. EPA policy.
URLs/Downloads:
DOI: Improving Computational Derivation of PFAS Toxicokinetic Half-lives
POSTER.PDF (PDF, NA pp, 460.094 KB, about PDF)