Critical endpoints of PFOA and PFOS exposure for regulatory risk assessment in drinking water: Parameter choices impacting estimates of safe exposure levels

Abstract

USEPA issued drinking water interim health advisories (iHA) for PFOA and PFOS. The Agency's choice for critical effect, toxic point-of-departure (PoD), benchmark dose (BMD), pharmacokinetic (PK) model extrapolation to ingested dose, and use of uncertainty factors, resulted in the iHA for PFOS and PFOA being lowered from 70 ppt to 0.04-0.2 ppt. This review addresses key steps in the iHA derivation that influence changes in iHA values and suggests analysis and modeling changes for higher confidence in the iHA derivation, and re-evaluation of critical endpoint data for immunotoxicity and associated BMD modeling to derive a serum antibody PoD in the clinically adverse range. Movement from empirical PK modeling of ingested human dose to a platform that captures biological realism will more accurately reflect PFAS elimination, which impacts model-optimized ingested dose. The uncertainty factor (UF) for human variability should be reconsidered, as in utero and neonate exposures used to derive the iHA represent the likely susceptible populations. Although not part of the iHA derivation, cancer was considered in the drinking water maximum contaminant level goal (MCLG) technical evaluation. We discuss weaknesses in the cancer epidemiological data that require re-evaluation as the drinking water regulation process proceeds to a national standard.