April 5, 2016
In 2015, prominent international health agencies reached conflicting conclusions on the potential carcinogenicity of glyphosate, the most commonly used herbicide in the world.
In particular, the International Agency for Research on Cancer (IARC) classified glyphosate as "probably carcinogenic to humans," based on "sufficient" evidence of carcinogenicity in animals and supportive mechanistic evidence, and "limited" evidence in humans, including a positive association with non-Hodgkin lymphoma.
By contrast, the German Federal Institute for Risk Assessment and, earlier, the U.S. Environmental Protection Agency, the Joint Meeting of Pesticide Residues sponsored by the Food and Agriculture Organization of the United Nations and the World Health Organization, and other agencies classified glyphosate as unlikely to be carcinogenic to humans.
To more closely examine the science underlying this controversial issue, Exponent epidemiologist Dr. Ellen Chang and co-author have recently published a systematic literature review and meta-analysis of all epidemiologic studies of the association between glyphosate exposure and risk of lymphohematopoietic cancers, including non-Hodgkin lymphoma, Hodgkin lymphoma, multiple myeloma, and leukemia.
The article will be published in a forthcoming issue of the Journal of Environmental Science and Health, Part B, and it is now available online. This is the first article to synthesize the complete epidemiologic literature on glyphosate exposure in relation to risk of all major types of lymphohematopoietic cancers. It includes both a quantitative meta-analysis and a qualitative assessment of the weight of epidemiologic evidence that accounts for study quality, potential for bias, and overall evidence for a causal association.
In particular, following standard guidelines for assessing the weight of epidemiologic evidence, this article documents the literature search and selection criteria used to identify all relevant epidemiologic studies, the considerations (including potential for exposure misclassification, selection bias, confounding, and other types of bias) used to evaluate study quality, and an evaluation of the evidence for causality according to the framework of Sir Austin Bradford Hill.
Based on the available epidemiologic evidence, we reach the conclusion that a causal effect of glyphosate exposure on non-Hodgkin lymphoma, Hodgkin lymphoma, multiple myeloma, and/or leukemia has not been established.