Glyphosate is almost certainly NOT carcinogenic. I say “almost” because I would change my mind if a preponderance of peer-reviewed good-quality studies with adequate evidence were published that demonstrated a risk.
Nothing of the kind has happened. None the less, the International Agency for Research on Cancer (IARC) lists glyphosate as a Group 2A carcinogen. It is worth checking what that means in humans. “An agent may be classified in this group when there is inadequate evidence of carcinogenicity in humans.” There is “inadequate evidence,” indeed little or none, to suggest that glyphosate could be a carcinogen.
That said, I can no more prove that glyphosate is NOT a carcinogen than I can prove that this table is NOT giving me cancer. But the onus for scientific claims lies with the claimant. And so far, there are no good studies showing conclusively that glyphosate could increase your risk of getting cancer. That could change. But given the number of good studies that have not shown any link, if there is one, if it were shown to increase risk at all, the effect would have to be very small to be so difficult to detect.
Glyphosate is a very important and useful agricultural chemical with an excellent safety record and, again, no credible evidence that anyone can show that it causes cancer. So, glyphosate’s evident safety illustrates a problem with IARC.
Group 2A is called “Probable Carcinogens.” And some of the items so classed “probably” do, at least somewhat, increase the risk of cancer. But since the Group explicitly does not require that there be adequate evidence, there is a good chance that at least some of these hazards are “probably” imaginary. For others, it is likely that the increased risk is so small as to be statistically pointless to discuss. No one should much care if they are exposed to an increased risk from 1:10,000 to 1.12:10,000, even though that is a 12% increase. Everyone does stuff every day that is more dangerous than that.
And this ought to be pretty easy to understand. Because in Group 2A you will also find, working nights, cutting hair, fireplaces, hot tea, steak, the smell of fries cooking, urethane, and bizarrely malaria.
“One estimate, which has been published in a 2002 Nature article, claims that malaria may have killed 50-60 billion people throughout history, or about half of all humans that have ever lived.” Malaria killed 619,000 people in 2021. And IARC thinks it is worth considering if it “probably” could cause cancer, without adequate evidence that it does.
Group 2B carcinogens are said by IARC to be “possibly” carcinogenic. So less certain than “probably.” Or to put it another way, like everything else in the world, allowing that new evidence could arise, it is NOT “impossible” that some Group 2B agents are somewhat carcinogenic. That is more than “possibly” not a useful claim. And while this group contains a long list of scary chemicals, it also contains Ginkgo extract, aloe vera, magenta, pickles, carpentry, talc, and sewing.
For due diligence here, it is worth noting that Groups 2A and 2B include agents for which IARC believes there is, “sufficient evidence of carcinogenicity in experimental animals.” Which is why so many flash-in-the-pan claims need to have, “…in mice,” appended to them. Or, as we shall see shortly, “…in rats.”
Group 3 carcinogens according to IARC are agents that, “can not be classified in regard to their carcinogenicity to humans.” Where the “evidence of carcinogenicity is inadequate in humans and [also] inadequate or limited in experimental animals.” So, this is explicitly a grouping of agents for which there is literally no reason why they should be on any list of “carcinogens.” More than half the agents on IARC’s carcinogen list are in Group 3.
IARC used to have another category. Group 4: “‘Probably not carcinogenic to humans’ There is strong evidence that it does not cause cancer in humans.” Problematically this, if comprehensive, would be a list of every last thing in the universe not included in Groups 1–3. But probably also most of what is in Group 3. They put one agent in Group 4, but since have removed that group altogether.
Alcohol is now alphabetically at the top of the Group 1 list among agents including ionizing radiation and asbestos. Which ought to elicit some surprise and clearly illustrates how misleading IARC Groups are for assessing risk. I will happily drink a shot of rye any day, rather than a shot of plutonium. Plutonium is in Group 1 with rye.
And while IARC’s list may be useful for informing research, it is almost useless for public health. Groups 2A–3 in practice appear to be a naughty-list of things IARC would like you to think of as carcinogens, but with “inadequate” or indeed no evidence that they are.
In 2023 IARC put aspartame in Group 2B. “Possibly” causes cancer in humans, except that, again, there is, by definition, “inadequate evidence” for that claim.
Which brings me to saccharine. That sweetener was used extensively for something like a century with no problem in humans. Then it was discovered that because of the nature of their urinary tract, which they do not have in common with humans, massive doses of saccharine could sometimes cause bladder cancer “…in rats.” Subsequently however it was shown that this was not true for humans. Much as I can survive eating chocolate, but your dog might not.
“IARC originally classified saccharin in Group 2B (“possibly carcinogenic to humans”) based on the rat studies, but downgraded it to Group 3 (“not classifiable as to the carcinogenicity to humans”) upon review of the subsequent research.” But let’s be clear. Saccharine is NOT A CARCINOGEN.
But because it is inevitable that the public and governments will interpret IARC group classifications as safety advice, saccharine was assumed to be a carcinogen anyway, and banned in many places. Most of those restrictions were lifted more than 20 years ago, when the claim was debunked. But saccharine is not used as a sweetener today, despite its vindication, because no one would buy a product with it on the ingredients.
And that is what is wrong with using the IARC Group lists as a guide to what is a carcinogen. Clearly there may be hundreds of agents on that list that are not carcinogenic. And there will be hundreds more that may have a nearly undetectable risk. You should not worry much about an exposure that has a one in a thousand chance of causing a cancer every five hundred years. And IARC should, in view of how their reports are misinterpreted, develop a much more nuanced classification system, that reflects relative risk and does not make claims for which, by their own admission, they do not have adequate evidence. Because as things stand, some of the agents they claim “probably” or “possibly” cause cancer, or “probably” don’t, make an important intergovernmental agency forming part of the World Health Organization of the United Nations, look kinda ridiculous.