The Right Chemistry: Putting Concerns About Weed Killer Glyphosate Into Perspective
This is just one of the thousands of chemicals to which we are regularly exposed. The chemical glyphosate is used in some products, such as Roundup.
Seeking and Finding in Chemical Testing
I don't often quote the Bible, but I'll snip a phrase from Matthew 7:7: "Seek and you shall find." Matthew did not have chemistry in mind, but of course I do. My take is that if you test a chemical with enough doses and endpoints in rodents, you will often find some statistically significant effect. The key question, though, is whether that effect is biologically meaningful and relevant to human exposure. That question repeatedly comes up in discussions of glyphosate, the most widely-used weed killer in the world.
The History of Glyphosate Controversy
I have followed the glyphosate story closely since the 1990s when the first crops that were genetically engineered to resist this herbicide were introduced. Almost immediately there was a knee-jerk reaction against genetic engineering, as is commonly seen with the introduction of any novel technology be it vaccination, microwave ovens or cellphones.
Farmers, however, welcomed the new technology because it meant that fields could be sprayed with glyphosate to eliminate weeds without causing harm to crops. But this also meant greater use of glyphosate and that precipitated literally hundreds of studies about the risks of occupational exposure to glyphosate as well as of the consequences of consuming trace residues in food.
Scientific Assessment of Risks
Obviously, I haven't read all the papers that have been published about this chemical, but I think I have read enough to form an opinion. I've looked at numerous rodent feeding studies, tests for glyphosate in people's urine and epidemiological investigations of cancer rates among applicators. My conclusion? Dietary residues present no risk to the consumer, but a small risk of non-Hodgkin lymphoma in applicators who do not use proper protective equipment cannot be ruled out.
In 2012, the media fawned all over a paper published by Gilles-Éric Séralini and his group that showed pictures of large tumours on rats that had been fed corn genetically modified to resist glyphosate. The study was widely criticized within the scientific community for using too few animals and using a type of rat prone to spontaneous tumours.
There were also criticisms of the researchers' statistics with some scientists pointing out that there was actually no statistical difference in tumour incidence in the experimental and control groups. The paper was eventually retracted because the evidence was deemed to be "inconclusive." Nevertheless, the idea that glyphosate is linked to cancer was implanted in the public mind.
Understanding Hazard Versus Risk
The controversy picked up steam in 2015 when the International Agency for Research on Cancer (IARC) placed glyphosate in its category 2A, reserved for substances that are "probably carcinogenic to humans." There is much misunderstanding about this classification, which is based on hazard, not risk. Hazard means that a substance has the capability of causing cancer under some condition. That may be the fostering the growth of cancer cells in a Petri dish or triggering a tumour by feeding extremely high doses to test animals. Risk is a determination of whether a substance can cause cancer under real-world conditions.
Regulatory Standards and Daily Exposure
Any novel pharmaceutical, food additive or pesticide that is introduced requires a determination of the acceptable daily intake (ADI), expressed as mg/kg body weight/day. This is the amount that can be consumed every day, over a lifetime, without appreciable health risk. This is determined by toxicity, reproductive and genotoxic studies in rodents fed different doses of the substance. An important determination is that of the no observed adverse effect level (NOAEL), the maximum amount that has no effect on the test animals.
After everything has been factored in, the ADI for glyphosate, as determined by the most stringent regulatory agencies, is 0.5 mg/kg/day. Based on residues in food, a typical daily dietary exposure for people has been found to be 0.001 mg/kg, which is 1/500th of the ADI. Furthermore, carcinogenicity in rodents begins at a dose that is roughly 100,000 times greater than the estimated daily glyphosate residue in the human diet. That is why I don't think there is a need to be concerned about food residues.
Occupational Exposure Considerations
Occupational exposure is a different story. There are a few case-control studies in which patients diagnosed with non-Hodgkin lymphoma were compared with people free of this disease and were also surveyed about their use of glyphosate. The natural incidence of this cancer is about two per cent, and among the heaviest applicators it was found to be 2.6 per cent, which means roughly an extra six cases in 1,000 people over a lifetime. These would be people who use glyphosate occupationally at exposures never encountered by the general public. And, of course, an association can never prove a cause-and-effect relationship. Workers occupationally exposed to glyphosate may be exposed to various other agrochemicals as well.
The problem with case-control studies is that they are based on human memory, which is not very reliable. People diagnosed with cancer are more likely to dredge their memory for substances that they think may have caused the disease, and because they may have heard of a link to glyphosate, they may report having used it while healthy people are less likely to report its use.
Far more reliable are prospective studies in which subjects record their use of a substance over years and researchers then look at their documented health records to see if there is a link. The most famous such study is the Agricultural Health Study in the U.S. that followed tens of thousands of pesticide applicators, and found no statistically significant association between glyphosate use and overall cancer or non-Hodgkin lymphoma.
Legal Settlements and Scientific Evidence
Given that the evidence of glyphosate being a health concern for humans ranges from scant for occupational exposure to none for consumer exposure, how do we account for the fact that Bayer — the marketer of Roundup, the most popular glyphosate formulation — has paid out billions of dollars to settle lawsuits that claimed injury due to the chemical?
Unfortunately, courts do not make judgements solely on the basis of science. Juries can be swayed by compelling anecdotes, emotional narratives, cherry-picked data, and especially by what is seen as misconduct by big companies.
A major paper published in 2000 that was widely quoted to demonstrate the safety of glyphosate was recently retracted because of potential ghostwriting by employees of Monsanto, the producer of Roundup at the time. There were also concerns about undisclosed financial compensation to the listed authors and too much reliance on unpublished data supplied by the company. The paper was retracted because of these ethical concerns, not because of falsified data as some have claimed.
Broader Context of Chemical Exposure
Finally, consider that glyphosate is just one of the thousands of chemicals to which we are regularly exposed and I suspect that if each of these were tested with the same rigour as glyphosate, some adverse effect at some dose in some species would be detected.
"Seek and you shall find."
