Intended for healthcare professionals


Probable carcinogenicity of glyphosate

BMJ 2019; 365 doi: (Published 08 April 2019) Cite this as: BMJ 2019;365:l1613
  1. Manolis Kogevinas, researcher
  1. Barcelona Institute for Global Health (ISGlobal), Barcelona, Spain
  1. manolis.kogevinas{at}

Independent investigation must continue

In 2015, the World Health Organization’s International Agency for Research on Cancer (IARC) identified glyphosate, the world’s most commonly used herbicide, as a probable human carcinogen.12 It quickly became evident that separating science from politics and economic interests would be difficult for glyphosate.

IARC’s assessment prompted a major controversy between health evaluation agencies, led to unprecedented lobbying by Monsanto (the primary manufacturer of glyphosate and genetically modified products resistant to glyphosate), and resulted in high profile court cases in the US.3

Glyphosate typifies the problems associated with research, evaluation, and regulation of pesticides. These include serious difficulties in the conduct of human research; important gaps in post-market research into exposure and risk assessment, particularly in low and middle income countries; lack of information on environmental effects; extensive industry involvement in evaluation and regulatory processes; and the major legal implications of these evaluations.

IARC’s 2A grouping of glyphosate as probable human carcinogen is the second strongest category of evidence in a four tier scale. The 2A is an uncomfortable grouping, falling between “regulation is needed” for a group 1 “human carcinogen” and “more research is needed” for a 2B “possible carcinogen” classification. Scientific evidence is not always clear, hence the need for sometimes troublesome intermediate classifications.

Changing circumstances

The strength of the evidence against agents in IARC’s group 2A varies; some agents lie close to a group 1 designation, but the evidence is weaker for others, including glyphosate. However, substantial published evidence from human, animal, and mechanistic studies at the time of the IARC evaluation indicated that adverse effects from exposure to glyphosate could be classified as probable. One aspect of the controversy is that formulations of glyphosate and exposure patterns have changed over time, and the many positive case-control studies cited by the IARC now reflect older exposure circumstances.

The IARC working group on pesticides identified non-Hodgkin lymphoma as a leading risk, and in 2018 a man from California became the first person to be compensated for non-Hodgkin lymphoma linked to glyphosate exposure.3 More recent evidence, including an update of data that were available before the IARC’s evaluation, showed no association between non-Hodgkin lymphoma and glyphosate but instead suggested an increased risk of a rarer leukaemia subtype.45 And a new analysis pooling the three largest cohort studies on pesticides showed an association between glyphosate and one non-Hodgkin subtype.6 These new analyses do not modify the IARC’s evaluation, but they do have implications for quantifying the risk from glyphosate.

The IARC’s evaluation of glyphosate in 2015 was followed by other evaluations, many of which reached different conclusions, including one by the European Food and Safety Authority (EFSA). These evaluations followed different protocols and some considered different evidence, including unpublished and non-peer reviewed data from industry funded studies.

Undisclosed industry involvement has emerged in evaluations by EFSA, the UN’s Joint Meeting on Pesticide Residues, and the US Environmental Protection Agency.7 Lobbying by industry was widespread and well documented.8 Although it is difficult to quantify, lobbying probably influenced the outcome of some of these evaluations and media coverage and led US government officials to question supporting the IARC/WHO.

Unanswered questions

What next? “More research is needed” is still (unfortunately) an inescapable recommendation. Although existing evidence could justify the classification of glyphosate as a probable human carcinogen, substantial gaps remain, including information about the relative effects of different product formulations, the timing of exposure, and additional data on mechanism of action in humans.

Although the focus so far has been on glyphosate and cancer, potential non-cancer health risks should also be investigated further.9 The few existing epidemiological studies with robust assessment of exposure to pesticides including glyphosate (such as the Agricultural Health Study4 and the French AGRICAN study10) were not conducted in low or middle income countries such as India or Brazil, where use is highest.

This lack of evidence and difficulty in replication will inevitably lead to more controversies when other pesticides are evaluated. Pesticides have hundreds of active ingredients, and less than a dozen have been formally evaluated by the IARC as group 1 or 2A carcinogens. Some of these, such as DDT (dichlorodiphenyltrichloroethane) and lindane, are already banned in most countries. But responsible companies marketing products that may pose a risk to human health and the environment should be collaborating internationally to gather evidence of possible harm, rather than attacking authoritative science.


  • Competing interests: I have read and understood BMJ policy on declaration of interests and declare that I have participated and chaired several IARC monographs but did not participate in the glyphosate monograph.

  • Provenance and peer review: Commissioned; not externally peer reviewed.


View Abstract