What’s the real risk of Monsanto’s controversial weed killer?

The latest government report on glyphosate contradicts the findings of the World Health Organization’s cancer group.

By Willy Blackmore


Over the past 30 years, the Environmental Protection Agency has considered the health risks posed by the herbicide glyphosate—best known by the Monsanto brand name Roundup—on five occasions. When it first looked at the issue in 1985, the agency determined that glyphosate was a “possible human carcinogen.” A year later, a third-party panel called into question the study that first assessment was based on, and the EPA declared glyphosate “not classifiable as to human carcinogenicity”—the jury’s still out, essentially—while promising to continue to examine the issue as new research came out. Then, in 1991, the EPA took another look and backed off further from its initial assessment, saying there was “evidence of non-carcinogenicity for humans.” In 2015, that assessment was updated to the safer-sounding “not likely to be carcinogenic to humans.” On Friday, a new review of research published by the EPA again found that the herbicide, which is now the most widely used in agricultural history, does not cause cancer.

But there’s a bit of a problem with the agency’s three-decade drift toward declaring the herbicide as safe in stronger and stronger terms: Last year, the World Health Organization’s International Agency for Research on Cancer declared that glyphosate was “probably carcinogenic to humans.” The announcement was major news: The declaration has put glyphosate into regulatory limbo in the European Union, where its safety approval may not be renewed, and has led to calls to ban the chemical outright in the United States. In the wake of the IARC announcement, tests of food products paid for by consumer groups have found trace levels of glyphosate in everything from beer to eggs to oatmeal.

The new EPA report—part of its ongoing risk assessment of glyphosate that is now years behind schedule and may not be completed until next spring—follows another that was published briefly on the agency’s website before being taken down. (The EPA said it was published prematurely, but the pages were marked “final.”) It too determined that glyphosate is not a carcinogen.

It has all led to growing consumer concern over glyphosate—and distrust of both Monsanto, its major producer, and the processes by which its safety is determined.

So does glyphosate cause cancer? Consumers tend to see things like carcinogenicity in black-and-white terms: something either gives people cancer or it does not. Just look at the calls for the herbicide to be banned: Armed with the “probably carcinogenic to humans” claim, petitions like one from Care2, which garnered more than 128,000 signatures, argue, “Glyphosate should not be in our consumer products in any amount. It is not safe as previously claimed.”

Reviews of scientific literature like those conducted by the IARC and the EPA are anything but black and white. The process involves sifting through piles of research data, determining what qualifies as a sound result, and making a case—carcinogenic or not carcinogenic—based on the evidence that the bulk of the data, especially the sound data, supports.

Should a weed killer that might cause cancer be banned?

In the IARC report from last year, the authors wrote, “There was limited evidence in humans for the carcinogenicity of glyphosate.” The case-control studies the IARC scientists looked at “reported increased risks for non-Hodgkin lymphoma that persisted after adjustment for other pesticides.”

But those studies predominantly looked at white men in the United States, Canada, and Sweden. Sure, in recent history in those places, white men have done a lot of farming, and what with farmers encountering glyphosate in far higher amounts than anyone ingests by eating oatmeal or honey, that seems like a reasonable place to start investigating whether exposure to an agricultural chemical might give someone cancer. But it’s not a representative sample of humanity—no women, a single ethnicity, and in a limited geographical area. The IARC authors looked at some studies on lab animals and at research done on increased blood levels of a compound associated with glyphosate, but the non-Hodgkin’s lymphoma studies provided the bulk of the evidence.

In compiling the EPA review, a much broader swath of research was reviewed, and the authors also considered the studies included in the IARC report that looked at farmers exposed to glyphosate. The EPA found fault with all the studies, determining that they showed a statistically insignificant increase in risk or did not properly control for other pesticide exposures. In discussing the non-Hodgkin’s lymphoma studies, the authors wrote, “There is clearly a strong potential for confounding by co-exposures to other pesticides since many are highly correlated and have been reported to be risk factors for NHL.” This means that these men may have developed cancer because they worked with pesticides, but glyphosate might not have been the culprit.

With regard to non-Hodgkin’s lymphoma, the EPA authors concluded, “Due to study limitations and contradictory results across studies of at least equal quality, a conclusion regarding the association between glyphosate exposure and risk of NHL cannot be determined based on the available data.”

So does glyphosate cause cancer? The IARC said it’s probable and largely based that assertion on a series of studies that found “limited evidence” that it is carcinogenic to humans. That’s the group’s mandate: to determine if a chemical might, even in rare circumstances, cause cancer. The IARC review by no means says that eating foods that contain trace amounts of glyphosate is a cancer risk. While the EPA arrived at a different conclusion with regard to the non-Hodgkin’s lymphoma studies, it did determine that a link between glyphosate and that type of cancer couldn’t be determined based on the existing data. Though it noted that many of the studies were conducted before 1996, when Roundup Ready crops were first introduced and glyphosate use began to skyrocket.

While it’s likely that more tests finding trace amounts of glyphosate in food products will grab headlines and that there will be continued calls for new regulations, the focus on risk to consumers may be a case of missing the forest for the trees. As Paul Towers of the Pesticide Action Network told TakePart in May, there may be health concerns, “but I think the bigger problem—and potential solution—lies with the USDA supporting a better agricultural system that doesn’t rely on these chemicals. I think our biggest concern remains the kind of system that agriculture gets stuck in—the pesticide treadmill of overuse, misuse, developing or encouraging invasive pests or invasive weeds that we then need to bring in the next chemical in order to deal with.”

Solving that problem involves asking a question more complicated than whether or not glyphosate causes cancer.