Citation:

Conley, J., C. Lambright, N. Evans, A. Dixon, J. Ford, D. Jenkins-Hill, E. Hines, J. McCord, E. Medlock Kakaley, AND L. Gray. Dose addition-based approaches accurately predict maternal and neonatal effects in Sprague-Dawley rats from three different PFAS mixtures studies. Society of Toxicology 2023 Annual Meeting, Nashville, TN, March 19 - 23, 2023.

Impact/Purpose:

Due to the pervasive detection of multiple PFAS in human and environmental matrices, many health-based agencies are employing or pursuing cumulative assessment and regulatory approaches.  However, the literature is lacking mammalian in vivo studies that test additivity by comparing observed mixture responses to mixture model predictions.  We have conducted three different PFAS mixture studies investigating maternal and neonatal effects from maternal exposure during pregnancy in a rat model.  Experimental data were compared to predictions using established mixtures toxicology approaches and effects were well predicted using dose addition methods.  One of the mixture studies discussed here (PFOA+PFOS mixture) was the result of a request from the Office of Water for technical support of cumulative effects of exposure to multiple PFAS.    The studies described here are some of the only mammalian in vivo mixture toxicity studies of PFAS that incorporate mixture modeling and analyses to characterize the type of mixture effects and the accuracy of predictions using well-established approaches.  The results directly support the proposed approaches by the Office of Water, as well as state and international health-based agencies that have adopted mixture-based risk assessment and regulatory approaches for exposure to multiple PFAS.

Description:

Due to the pervasive detection of multiple PFAS in human and environmental matrices, many health-based agencies are employing or pursuing cumulative assessment and regulatory approaches.  However, the literature is lacking mammalian in vivo studies that test additivity by comparing observed mixture responses to mixture model predictions.  We have conducted three developmental toxicity studies of PFAS mixtures including a 2 PFAS mixture (perfluorooctanoate (PFOA) + perfluorooctanesulfonate (PFOS)), a 3 PFAS mixture (hexafluoropropylene oxide-dimer acid (HFPO-DA or ‘GenX chemicals’) + Nafion byproduct 2 (NBP2) + PFOS), and a 6 PFAS mixture (perfluoromethoxyacetic acid (PFMOAA) + HFPO-DA + PFOA + perfluorohexanesulfonate (PFHxS) + NBP2 + PFOS).  For all individual compounds and mixtures, we exposed Sprague-Dawley rat dams via oral gavage from gestation day 8 (GD8) to postnatal day 2 (PND2) and evaluated a range of maternal and neonatal endpoints.  Dose-response data from individual PFAS studies were modeled to obtain curve parameters, and mixture analyses were used to evaluate observed mixture effects including the relative potency factor (RPF) approach and equations for dose addition and response addition.  The carboxylate PFAS (PFOA, HFPO-DA, PFMOAA) and sulfonate PFAS (NBP2, PFOS, PFHxS) produced multiple common effects across all compounds, however the two groups were not toxicologically identical.  Effects common across all compounds included reduced pup and maternal bodyweights, reduced pup survival, increased maternal and pup relative liver weights, reduced serum thyroid hormones, and increased liver expression of PPAR signaling pathway genes.  With limited exception across studies, these endpoints were modelled with equivalent or better fits (based on corrected Akaike’s Information Criteria (AIC) using dose addition compared to response addition equations.  In the 2 and 3 PFAS mixture studies, reductions in maternal bodyweight and weight gain during pregnancy were less-than-additive.  Thus far, no endpoints have shown greater-than-additive effects that differed by >2-fold from dose addition predictions.  Most endpoints had statistically similar dose response curve slopes across the six PFAS and mixture effects were well predicted by RPF and dose addition equations.  Importantly, RPFs for the six PFAS we have studied vary by >20-fold across the range of endpoints modeled and it was not possible to model all mixture effects with a single set of RPFs.  Endpoints with non-congruent slopes across component PFAS were well-predicted using a dose addition equation that does not assume equal slopes.  Combined exposure to mixtures of PFAS  produced cumulative effects on multiple endpoints and these effects were generally well predicted by dose addition-based approaches, particularly the RPF.  The views expressed in this abstract are those of the author(s) and do not necessarily represent the views or policies of the U.S. Environmental Protection Agency.             

Record Details: