In 2015, IARC published their review of the evidence for carcinogenicity of glyphosate (vol. 112). They concluded that there was “limited” evidence that glyphosate exposure causes cancer in humans, and “sufficient” evidence that it does in animal studies. They therefore classified glyphosate as “probably carcinogenic to humans”.

In its scrutiny of the IARC analysis, the German national risk assessment agency BfR more-or-less accepted IARC’s take on the human evidence but rejected IARC’s interpretation of the animal evidence, concluding “no hazard classification for carcinogenicity is warranted”. EFSA and experts from EU Member States reviewed the BfR risk assessment and IARC analysis and agreed that glyphosate is “unlikely to pose a carcinogenic hazard to humans”.

This is a pretty big difference: how can IARC look at the animal evidence and be convinced it shows that glyphosate causes cancer, yet EFSA can look at the animal evidence and be convinced it shows that glyphosate does not cause cancer?

One reason which has been offered is that EFSA and IARC looked at different data. EFSA had access to at least three confidential industry studies (let’s call them the “Mysterious Three”) described by José Tarazona, head of EFSA’s Pesticides Unit, as “key” and “pivotal”. IARC either only had access to summaries of these studies which were lacking key information, or no access at all (since they are unpublished, confidential documents). Since IARC excludes from analysis research summaries which lack essential data, they were not taken into account in their final conclusion.

But do the Mysterious Three provide a credible explanation for the difference in opinion between IARC and EFSA?

To answer the question, we need to understand that when IARC says there is “sufficient evidence” that glyphosate causes cancer in animals, they are saying they are convinced by the evidence, which means that there is almost no chance at all that a new animal study could come along and show glyphosate does not cause cancer (i.e. any such study would have to be big enough and robust enough that it would nullify all of the convincing evidence to the contrary).

EFSA, on the other hand, appears to be saying that IARC is wrong, that the existing evidence shows glyphosate “is unlikely” to cause cancer, implying that they think the chance of a new animal study coming along which shows glyphosate causes cancer is in fact very small (i.e. they feel the existing evidence shows glyphosate does not cause cancer, and that a new study would therefore have to be very large and very robust to nullify the current evidence).

The Mysterious Three therefore need to be strong enough evidence that they would persuade IARC that the evidence in fact points away from glyphosate’s carcinogenicity, rather than convincingly towards it. We can only speculate, because we haven’t seen the studies. But they would presumably have to be seriously powerful, otherwise they would mean little against the weight of evidence which convinced IARC that glyphosate is carcinogenic in animal studies.

If the secret studies really are that compelling, it is odd they have not been published: surely they would put to bed the controversy about glyphosate’s carcinogenicity?

The other possibility, of course, is that the studies don’t in fact count for much, and that EFSA has made a mistake in weighing them so heavily in its analysis. According to Kate Guyton, the Senior Toxicologist at the IARC Monographs Programme who followed the file at IARC, the Mysterious Three showed a correlation between glyphosate exposure and a rare type of tumour (statement in interview with Corporate Europe Observatory).

How it is these few, not even obviously negative, studies could make glyphosate seem less likely to be carcinogenic sounds all the more like a question EFSA needs to answer. They can only do this by publishing the data and explaining how they interpreted it. Any further additional research to which EFSA had access but IARC did not (such as mentioned in this letter) also needs to be made public as well, otherwise nobody can double-check EFSA’s workings and see if they got the right result.

In medicine, the AllTrials Campaign is “calling for all past and present clinical trials to be registered and their full methods and summary results reported”. This is an essential step in ensuring that all medical treatments can have the evidence of their efficacy properly assessed, increasing the number of effective treatments and eliminating the ineffective and harmful ones. The same needs to happen with chemical toxicity research, because secret data cannot be properly evaluated, and risk of people and the environment being exposed to excessively harmful chemicals therefore increases.

In this particular case, if the evidence is as strong as EFSA claims, then there is no reason for keeping the data secret. If it is not, then glyphosate may be a chemical which needs to be removed from the market because it poses a cancer hazard. Either way, we need to find out.

Post discussion

Leave a comment

About CPES

The Cancer Prevention and Education Society is a Charity (No. 1089082) & Company Limited by Guarantee (No. 4157505). Registered in England and Wales. Registered Office: The Cancer Prevention & Education Society, Meads House, Leighterton, Tetbury, Gloucestershire, GL8 8UW.


Please help!

To continue our work, we are dependent on donations and volunteers. Please do get in touch to find out how you might be able to help out by contributing money or time by sending an email to Paul:


Sign up for Health & Environment: News and science round-ups, videos and monthly feature articles about how chemicals may be affecting human health.