Why EPA and IARC Classifications of Glyphosate Oncogenicity Differ

Abstract

In March 2015 IARC classified glyphosate (GLY), and glyphosate-based herbicides (GBHs) as “probable human carcinogens.” The U.S. EPA’s September 2016 “Glyphosate Issue Paper: Evaluation of Carcinogenic Potential” concluded that “The strongest support is for ‘not likely to be carcinogenic to humans’ at doses relevant to human health risk assessment.” This presentation explains why the IARC Working Group and EPA reached such different conclusions. Reasons include: (1) the EPA analysis is based on risks arising from GLY residues in food, while IARC assessed data on all routes of exposure to formulated GBHs, including high-exposure scenarios (e.g., mixer-loaders, applicators using handheld wands); (2) the EPA discounted dozens of tumors in 15 animal bioassays, including several that should have been deemed treatment related had the EPA followed its 2005 Cancer Risk Assessment guidelines; (3) the EPA essentially ignored dozens of positive, formulated GBH genotoxicity assays, including several in vivo studies in humans, whereas IARC placed considerable weight on them in concluding there is “strong mechanistic evidence” linking GBH exposure to cancer; (4) the EPA relied predominantly on animal bioassay and genotoxicity data generated by GBH registrants, whereas IARC placed heavy weight on studies published in peer-reviewed journals; and (5) EPA’s decision to limit its genotoxicity analysis to relatively low dietary exposures to pure GLY. These decisions led the agency to conclude “there is no evidence that glyphosate induces mutations in vivo through the oral route.” This conclusion is largely irrelevant to the evaluation of GBH cancer risk to applicators and others exposed to GBHs via routes other than diet. Had EPA included in its analysis the dozens of positive genotoxicity studies on formulated GBHs, the agency would not have been able to support its judgements regarding the absence of GLY and GBH genotoxicity and cancer risk.