The carcinogenic nature of glyphosate, the most widely used pesticide in the world, has opened up a deep division among US federal experts

The US EPA is divided over the safety of glyphosate, according to an article by Stéphane Foucart in Le Monde.

While the World Health Organisation’s cancer agency IARC says glyphosate is a probable human carcinogen, other agencies, such as the German Federal Institute for Risk Assessment (BfR) and the European Food Safety Authority (EFSA) disagree. The next agency to weigh in on the argument is the European Chemicals Agency (ECHA), which is due to deliver its opinion on Wednesday 15 March in the framework of the re-authorization of the product in Europe.

But now it seems that the re-evaluation of the chemical is also opening up fractures within individual institutions.

This is shown by a confidential document from the US Environmental Protection Agency (EPA) obtained by Le Monde. According to this internal memo, the EPA's Office of Research and Development (ORD) clearly disagrees with the preliminary opinion of another EPA department, the Office of Pesticide Programs (OPP).

This preliminary opinion of the OPP, published in error at the end of April 2016 on the EPA’s site, judged glyphosate as ”unlikely to be carcinogenic" – thus agreeing with EFSA and disagreeing with IARC.

The EPA's internal memo seen by Le Monde reveals that members of the scientific department of the US Federal Agency - the ORD - are closely aligned with the view of IARC and in opposition to their colleagues in the pesticides department - the OPP. Scientists from the ORD criticize those of the OPP for not respecting international conventions in the analysis of the carcinogenicity of glyphosate.

"Frameworks for data analysis and determination of causality [between exposure to a substance and cancer] that are currently in effect at the EPA and in the risk assessment community include gradations,” wrote the ORD in its memo. “The preliminary opinion of the OPP appears not to follow this approach,” adds the latter. To read the document, the OPP appears to have abandoned the rules for the evaluation of the dangers of a substance to "use a yes/no approach which could only lead to describe the substances only as "carcinogenic" or “unlikely to be carcinogenic" for humans.

The criticism is very technical, but nonetheless crucial. The EPA classification actually includes, in theory, five different grades:

* Carcinogenic;
* Probably carcinogenic;
* Showing suggestive signs of carcinogenicity;
* Insufficiently documented for carcinogenicity analysis; and
* Unlikely to be carcinogenic.

Thus there are many nuances that, according to the memorandum of the ORD, disappeared from the preliminary analysis conducted by the OPP.

Studies in rats

On the epidemiological aspect, the ORD agrees with the IARC that there is "limited" evidence of an association between exposure to glyphosate and the appearance of cancer in humans. But regarding the laboratory studies, the ORD recalled that "glyphosate has been tested in several two-year studies in rats or mice". "A wide range of tumours have been observed in these studies," adds the memo. “Tumours have been observed in the thyroid, liver, skin, pancreas, lymph, testis, mammary gland, kidneys and lung."

However, the ORD adds that according to the statistical methods used, the incidence of these tumours may be considered significant or not. That is the main point of contention. According to the statistical “mill” that deals with the results of these animal studies, the conclusions may be different: either the cancers that occur may be due to chance or can only be explained by the exposure to the substance tested.

Again, scientists at the ORD strongly criticized the work of their colleagues in the OPP, who claimed they did not find significant effects after using some statistical tests, but "without specifying which tests they used".

In the end, the ORD does not rule on a classification of glyphosate, but states that “the category “‘unlikely to be carcinogenic to humans’ can without doubt be thrown out”. "We can discuss whether the level of evidence is sufficiently high for the category "probable carcinogen”, adds the memo. “But this classification cannot be rejected."

Several questions remain: how will the ORD memo, dated 14 December 2015, be taken into account in the EPA's final opinion? Why was the OPP's preliminary notice published on the agency's website before being removed a few hours later? The EPA states that a discussion paper jointly drafted by the two departments was forwarded in September 2016 to a third group of federal experts, whose opinion is expected to be published on 16 March. This new opinion will in turn be revised prior to publication of the EPA's final official notice.


Source: Le Monde (French): http://www.lemonde.fr/planete/article/2017/03/14/glyphosate-discorde-a-l-agence-de-protection-de-l-environnement-americaine_5094158_3244.html