Mar 16 2017

Does Glyphosate Cause Cancer?

glyphosate-effects-fbGlyphosate, the active ingredient in Roundup, is the most popular herbicide used in the world. It has gained particular attention because several of the more commonly used GMOs are glyphosate tolerant, and therefore are intended to be used with the herbicide. Glyphosate is also manufactured by Monsanto (although it is off patent and there are generic versions available).

The question of the safety of glyphosate is in the news again after the New York Times did an article about a recent court case against Monsanto and the documents revealed through discovery and made public by the judge. Unfortunately, in my opinion the NYTs article is poorly done, and reveals significant bias – anti biotech bias is nothing new for the NYTs or the author of this article, Danny Hakim.

Last year I wrote about another article that Hakim wrote in the NYTs about GMOs, concluding:

In my opinion Hakim’s article in the Times was a hack piece with a biased narrative that is nothing more than a rehash of tired anti-GMO tropes that have already been widely deconstructed. He is entering this conversation late, and isn’t up to speed.

There are essentially two questions raised by Hakim’s latest article. The first concerns the behavior of Monsanto. Hakim alleges that they ghostwrote scientific articles for academics and used political pressure to shut down EPA reviews of glyphosate’s safety. I would not assume this assessment is true, and certainly don’t trust Hakim’s journalism given his history. The academics in question deny the allegations, and Monsanto claims these e-mails are taken out of context. We have certainly seen that before.

I don’t have enough info at this time to form my own opinion, but let’s assume that the worst interpretation of the allegations are true. This means that Monsanto tried to put its thumb on the scale to garner favorable reviews for its products. If true, this is clearly wrong. Corporations can (and often must) fund research, but they need to let the chips fall where they may. Academics should resist the temptation to accept too much “help” from corporations. This behavior is, unfortunately, very common and not limited to Monsanto.

But these allegations need to be kept in proper context. Not to excuse them in any way, but Hakim implies that this behavior raises fresh concerns about the safety of glyphosate, which is the second core question raised by the article. I don’t think that it does, however. There is nothing in the revealed e-mails that calls into question the large body of independent scientific research into the safety of glyphosate, any more than the infamous e-mails called into question the large body of scientific research establishing man-made global warming.

Let’s now do something which Hakim did not do in his article – take a look at the actual scientific evidence and see what it says.

When researching the safety of a substance there are several kinds of data. In vitro data looks at the substance in question when exposed to cells in culture. This kind of research is for plausibility only. Because the cells are being directly exposed, which is not analogous to what happens in a living organism, you cannot draw any final conclusion from this type of data, which tends to dramatically overcall the potential for toxicity.

There is also animal data, which is more relevant but still not definitive. Typically several animal models are used, and exposure levels are high. Animal models are useful in determining safety limits for dosing. Remember, everything is a toxin at high enough dose. The question is not, is a substance toxic, but at what dose does it become toxic. That is what animal studies are for.

They are also useful for determining what type of toxicity occurs – which organ systems are affected and by what kind of damage (cancer vs organ failure, for example).

In vitro and animal data are useful in informing safety limits for substances and to inform what kind of toxicity we should be looking for. They can also determine if it is safe to conduct human trials, when dealing with things like pharmaceuticals.

Human data, however, is the most important for definitively answering the question. Human data is mostly either directly experimental (as with drugs) or epidemiological (for chemicals that are used in industry but not intended for human use).

Regarding human epidemiological studies of glyphosate, a 2016 systematic review and meta-analysis concluded:

Meta-analysis is constrained by few studies and a crude exposure metric, while the overall body of literature is methodologically limited and findings are not strong or consistent. Thus, a causal relationship has not been established between glyphosate exposure and risk of any type of LHC.

A 2012 systematic review of glyphosate and cancer concluded:

Our review found no consistent pattern of positive associations indicating a causal relationship between total cancer (in adults or children) or any site-specific cancer and exposure to glyphosate.

They also recommended that future studies use biomonitoring to measure actual exposure levels.

Despite this the International Agency for Research on Cancer (IARC) in 2015 categorized glyphosate as a probable carcinogen, going against many other expert panel reviews. This garnered a lot of press attention, and became a standard talking point of anti-GMO activists. The IARC, however, has a reputation of calling almost anything a carcinogen.

A 2016 expert panel review conducted their own review of the evidence.

The Expert Panel concluded that glyphosate, glyphosate formulations, and AMPA do not pose a genotoxic hazard and the data do not support the IARC Monograph genotoxicity evaluation. With respect to carcinogenicity classification and mechanism, the Expert Panel concluded that evidence relating to an oxidative stress mechanism of carcinogenicity was largely unconvincing and that the data profiles were not consistent with the characteristics of genotoxic carcinogens.

In fact, four independent expert panels reviewed the IARC conclusion, and found:

The overall weight of evidence from the genetic toxicology data supports a conclusion that glyphosate (including GBFs and AMPA) does not pose a genotoxic hazard and therefore, should not be considered support for the classification of glyphosate as a genotoxic carcinogen. The assessment of the epidemiological data found that the data do not support a causal relationship between glyphosate exposure and non-Hodgkin’s lymphoma while the data were judged to be too sparse to assess a potential relationship between glyphosate exposure and multiple myeloma. As a result, following the review of the totality of the evidence, the Panels concluded that the data do not support IARC’s conclusion that glyphosate is a “probable human carcinogen” and, consistent with previous regulatory assessments, further concluded that glyphosate is unlikely to pose a carcinogenic risk to humans.

An independent 2016 expert review found:

Only the Agricultural Health (cohort) Study met our a priori quality standards and this study found no evidence of an association between glyphosate and NHL. For MM, the case control studies shared the same limitations as noted for the NHL case-control studies and, in aggregate, the data were too sparse to enable an informed causal judgment. Overall, our review did not find support in the epidemiologic literature for a causal association between glyphosate and NHL or MM.

For non-cancer outcomes the epidemiology is similar. A 2012 review of developmental outcomes concluded:

In conclusion, the available literature shows no solid evidence linking glyphosate exposure to adverse developmental or reproductive effects at environmentally realistic exposure concentrations.

A 2011 review of glyphosate and all non-cancer outcomes found:

Our review found no evidence of a consistent pattern of positive associations indicating a causal relationship between any disease and exposure to glyphosate.


I am simply searching through PubMed to find reviews of the safety of glyphosate, and this is what I find. You can do the same, it’s a user-friendly searchable database. There is a remarkable consistency to the reviews – they all agree that the evidence does not support an association between glyphosate exposure and any adverse health outcome. The IARC are the only outliers, and yet their flawed and quirky conclusion is the one that garnered the most attention.

There is also a theme in the reviews that we could use more and better quality studies. To put that into context, however, that is almost always the conclusion of such reviews. It is difficult to prove a negative – a lack of a correlation. Such a negative conclusion is only as good as the data supporting it, and therefore the more and more rigorous the data the better the conclusion.

We can always use more and better data when it comes to safety, but the existing data is robust, consistent, and independently replicated, and includes both glyphosate and formulations with glyphosate.

Glyphosate, in fact, is one of the safer pesticides in use (including many organic pesticides). It has replaced far more toxic herbicides. Opposing glyphosate because of unwarranted fears of toxicity is likely to cause harm due to whatever replaces it. Tilling is bad for the soil and releases CO2 into the atmosphere, and we cannot feed the world through hand weeding. Herbicides have to be part of the equation, and glyphosate is one of the safest out there.

31 responses so far

31 thoughts on “Does Glyphosate Cause Cancer?”

  1. Atlantean Idol says:

    Mass Pig gonna be mad you dissed his man-crush Hack-im again!

  2. Npsychdoc says:

    You write “Tilling is bad for the soil and releases CO2 into the atmosphere, and we cannot feed the world through hand weeding.”

    A necessary question is whether there is any detriment to soil ecology with use of glyphosate.

  3. Npsychdoc says:

    For example: Correia, F.V. & Moreira, J.C. Bull Environ Contam Toxicol (2010) 85: 264. doi:10.1007/s00128-010-0089-7

  4. Npsychdoc says:


    Lots of media attention on glyphosate and humans. Its a fair question, and one that is answered sufficiently enough for me. More attention should now be given to its long-term impact on the soil and sustained productivity.

  5. Np – I agree this is an important and complex question. I don’t think the studies you link to do much to address the real questions, however. They are the equivalent in medicine of in vitro studies which show that “stuff happens.” How many outcomes did they look at, how many were significant? It is also extremely difficult to extrapolate from this kind of data to net effects in the real world.

    Further, how glyphosate is used (is it part of integrated pest management?) is as or more important, and must be compared to all other viable options. Every approach will have strengths and weaknesses.

    The fact is, massive agriculture to feed over 7 billion people is tough on the ecosystem and hard to sustain. We need to take a sophisticated and evidence-based approach with an eye toward sustainability (no reasonable person disagrees with this) with every option on the table.

    Every option will have negative effects. They need to be put into a thorough context.

    Glyphosate deserves attention because of its widespread use. But a lot of the attention is ideologically or politically motivated, and amounts to more of a hit job than a balanced view of a complex topic. An ideological approach is bound to have net negative consequences.

  6. BBBlue says:

    Npsychdoc- You might be interested in this review of your references. And don’t forget, Scientific Reports is not Nature

    It is certainly plausible that herbicides (glyphosate or others) might have some direct effect on earthworms, so it seems like a reasonable study. However, because of the study design and methods used in this, none of the effects they observed can be attributed directly to glyphosate.

    I would be surprised if pesticides that find their way into the soil don’t have some effect on living organisms there, but then, that would include both synthetic and natural, approved-for-organic-use compounds like copper.

  7. Npsychdoc says:

    I don’t think the articles I cited address the real questions either, and that wasn’t my point. If you have more satisfying articles related to the issue I would love to see them.

    This issue, as you identify well, is “net effects in the real world.” Glyphosate is already being used in widespread fashion, and we’re not quite sure of the long-term net effect on soil health and productivity. And the majority of research I see are these vitro-study equivalents…

    How about glyphosate versus intercropping with crop rotation?

    Also, you discuss an agricultural approach to feed 7 billion people, and you point out that its tough on the ecosystem and hard to sustain. Perhaps there is a politically motivated ideology at work there…cant help thinking of Earl Butz. Not saying you are advocating this approach, just that it is interesting that an agriculture to feed the world is historically an abnormal approach, is indeed hard to sustain, so why are we not shifting our agricultural policies to support a number of agricultures unique to localities, watersheds, and communities.

  8. Fair Persuasion says:

    Do we have any develop of disease information for multiple myeloma or non-Hodgkins type lymphoma?
    Are there any causal relationships known for these diseases? The fears of glyphosate rests on human inhalation and physical contact with the product.

  9. Npsychdoc says:

    BB – that’s a nice review, thanks for posting.

  10. BBBlue says:

    We can always use more and better data when it comes to safety, but the existing data is robust, consistent, and independently replicated, and includes both glyphosate and formulations with glyphosate.

    At what point does a call for more studies become a disingenuous demand designed to obstruct or move the goal posts rather than to inform? Glyphosate has been studied since before its release 1974 and as Steven notes: “…existing data is robust, consistent, and independently replicated.”

    Antis are bound to a narrative, and if credible studies don’t support that narrative, they turn to advocacy research à la Seralini, or they make unfounded inferences from existing information as in the case of Samsel and Seneff. In other words, they make stuff up or resort to reading tea leaves.

  11. SteveA says:

    BBBlue: “You might be interested in this review of your references. And don’t forget, Scientific Reports is not Nature…

    Nice critique. The lack of a ‘no-plant’ control is odd. I’m not the sharpest knife in the drawer, but even I’d have thought of including one. I’d have wilted the leaves and stems with a flame (after shielding the soil surface to stop anything else getting roasted).

  12. BBBlue says:

    Fair Persuasion- I think Steven’s “Expert panel” reference discusses your question. Also this:

    It is certainly reasonable to be concerned about pesticides, all pesticides, in the context of handler and applicator occupational health, but exposures there are orders of magnitude greater than anything associated with what is actually found on our food.

    Pesticides are divided into four general categories; Cat I (Danger), Cat II (Warning), Cat III (Caution), Cat IV (Caution signal word optional). The older commercial Cat II Roundup formulations I am familiar with were given that designation because their surfactants were irritants capable of causing damage to eyes and mucous membranes. Many (most?) newer formulations have been formulated to be less hazardous in that regard and are Cat III and IV.

  13. Npsychdoc says:

    Ok, first article was very weak, I agree. How about the second one I posted which is getting no attention?

    Look folks, I’m ok with glyphosate, I’m NOT anti-GMO, and I think much of “Organic” farming is ideologically driven and ultimately hurtful to many farmers to appropriately blend conventional and organic methods. There I’ve said it.

    My original point is that I’m done with the question of glyphosate and humans. IMO, its safe enough for me, my kids and my dog.

    But I still don’t have an answer for the long-term effects of glyphosate on soil health and production versus other methods of weed control. We don’t have an answer, yet we’re moving forward in its use. The motivation behind why we’re moving forward without data is certainly not altruistic or to feed the planet, Steven.

    Finally, I like the weedcontrolfreaks. Particularly this post:

  14. Np – seems like we’re on the same page regaring organic, sustainability, and the health effect of glyphosate.

    We are probably also not far off on the soil effects of glyphosate. I have not researched it enough to express a specific opinion, I would just caution against drawing conclusions from flawed preliminary studies.

    There may be effects of glyphosate that are inadequately researched at present. OK. We should do the research and go wherever it leads.

    In the meantime we know that glyphosate is much less toxic than the herbicides it replaces. So far there is no ecological reason to suspect that the current use of glyphosate is harmful.

    You are suggesting looking for more subtle long term effects. Again, sure, let’s do that. That doesn’t mean we have to stop using it in the meantime, especially given the alternatives.

    Finally, as I have written many times, I favor integrated pest management (IPM) which seeks to use evidence and all available methods to control pests as sustainably as possible and minimize resistance.

  15. DickK says:

    I am unfamiliar with jargon used in many professions other than my own, especially the medically oriented ones, and am continually frustrated when acronyms are used without defining each. This article was one of the most egregious I have seen in recent years. No definitions were given for LHC, AMPA, NHL, or MM. A very good definition of one acronym, International Agency for Research on Cancer (IARC), was given, but not so for others.

    My only takeaway was that Steve Novella didn’t think there was evidence linking glyphosate to some sort of health effect(s). I have no idea what health effects were mention since only acronyms were used.

    I am a long time reader and fan of neurologica, and will remain so, but I would certainly appreciate it if acronyms were defined in the article in they are being used. I don’t recall having that problem in any other neurologica posts over the years. I hope this is an anomaly that won’t be repeated.

    Thanks for listening to my rant,
    Dick K.

  16. BBBlue says:


    But I still don’t have an answer for the long-term effects of glyphosate on soil health and production versus other methods of weed control.

    And neither do I. In fact, I would venture a guess that there are no definitive studies on the subject. However, that may be an almost meaningless question considering the number of other pesticides like fungicides and insecticides that wind up in the soil and which are more likely to have a significant effect on the life found there. It’s not just herbicides or soil-applied pesticides that wind up in the soil. In my opinion, the media focus on glyphosate is foolish and counterproductive.

    It has been my experience that most farmers farm the soil to the extent it contributes to the amount and quality of the end product now and into the future. Preserving soil life is not an end in itself; a soil full of earthworms may make some people feel good about what they are doing, but earthworms and many other soil organisms are not necessary for the maintenance of a productive soil.

    Farming, all farming, is not a natural system, it involves human intervention and environmental modification. Proponents of organic production and agroecology think they have a superior system because it is closer to a natural system, but that’s all part of an appeal to nature. If the long term productivity of a soil can be maintained in a heavily manipulated system with limited effects on the surrounding environment and the living things in it, what’s the problem?

    I’ve been tracking soil conditions of conventionally-farmed orchards and vineyards for decades, and we have not only been able to maintain the productivity of those soils, we have consistently improved it.

  17. ScubaSharky says:


    The acronyms you mention are all in quotes from the referenced papers. Clicking the imbedded links will lead you to the definitions you seek.

    LHC – lymphohematopoietic cancer
    AMPA – aminomethylphosphonic acid
    NHL – non-Hodgkin’s lymphoma MM – multiple myeloma

  18. Npsychdoc says:

    BB – you seem to be inferring that I prefer so-called “natural” methods. You are preaching to the choir.

    Perhaps I’m being unreasonably conservative about glyphosate.

    You say: “I’ve been tracking soil conditions of conventionally-farmed orchards and vineyards for decades, and we have not only been able to maintain the productivity of those soils, we have consistently improved it.”

    Can you provide more info on this? I’m honestly curious. Also curious as to what degree orchards and vineyards are a different animal from a monoculture of annuals in terms of the conventional vs other approaches, outlook on soil fertility, etc. If you have any resources our good journals to be aware of I’m all ears.

    Appreciate the discussion.

  19. MaryM says:

    In an interesting comparison of media drama vs science–the same day as the Hakim piece, a European scientific review by their Chemicals Agency: “Glyphosate not classified as a carcinogen by ECHA”.

    Sadly, my local NPR station chose to say this: “Plus: new safety concerns about Monsanto’s weed killer.” And they used the Hakim story to talk about that. No, it didn’t raise new safety concerns. But how about covering the science instead of courtroom drama?

    This is why we can’t get people to listen–they hear the court stuff, and nobody bothered to tell them the science.

    What can we do when even media we’d like to support does this?

  20. > # BBBlueon 16 Mar 2017 at 12:05 pm

    The reference you give:

    doesn’t actually review the article that Npsychdoc refers.
    Npsychdoc referred to:

    and the reference above reviews

    You need to check better.

    Anyway, both of these articles by the same group, and they are the same crap,
    and the fact that both of these appear in “Scientific Reports” suggests very very
    strongly that “Scientific Reports” should be ignored.

  21. BBBlue says:

    Npsychdoc- Don’t mean to infer anything about your preferences, just using your comment/question as a jumping off point for my own comments.

    Biggest improvement has been due to change from frequent tillage of middles between tree and vine rows to no till. Also, prunings are shred and left as mulch on the orchard and vineyard floors. Soil amendments like lime, gypsum, sulfur and composted manures are applied depending on circumstances and soil chemistry. Deep tillage is performed between plantings.

    Formerly, standard practice was to grow native vegetation or planted cover crop between rows and either mow or till them under as needed. Prunings have always been left in the field, but when tilled, their organic matter disappears quickly in the arid soils of California. Very tough to maintain much more than about 1% organic matter in our surface soils under any circumstances.

    Over the course of 3-4 years, an undisturbed soil with a persistent, woody mulch on top is fairly free of weeds. If one does not let what weeds do occur go to seed, the “seed bank” in the soil surface is depleted and the end result can be something on the order of 80-90% weed control without herbicides. Herbicides are still used, but significantly less than the old system required. This sort of system to reduce weeds is something that I often see duplicated in residential landscapes and gardens as well.

    Another benefit of our no-till system is its resistance to soil structure destroying compaction. Also, pesticide adsorption occurs to a greater degree in soils rich in organic matter than in coarse, mineral soils, so a few inches of undisturbed mulch and soil that is relatively high in organic matter can create a barrier of sorts where pesticides degrade and are prevented from leaching deeper into the soil profile than they otherwise would be.

    Finally, nutrient cycling is when plants, either native or cultivated, forage for nutrients from the soil and transport them towards the surface where they are incorporated into plant tissues. Plants go to seed and die leaving their organic matter and nutrients behind near or on the surface. Many important nutrients like phosphorous and potassium are not very soluble, so they naturally wind up accumulating near the surface in both organic and inorganic forms. For those that are soluble, like nitrate, adsorption on organic matter minimizes the degree to which they leach out of the root zone so they can be accessed by plants. Organic forms of nitrogen also remain in the root zone where they slowly release available N as decomposition occurs. Over millennia, as in the Great Plains, that process can result in topsoils rich in nutrients and organic matter that are a couple feet thick or more. In arid soils like ours, a relatively poor topsoil layer may be just a precious few inches thick. By not disturbing the topsoil we do have, we allow roots to develop and forage for nutrients in that thin layer.

    By every objective measure we have such as plant and soil nutrient status, soil chemistry and water infiltration rates, in addition to yield and quality parameters, we have maintained or improved the productivity of our soils over the past 30 years. That is also reflected in more subjective observations of young plant growth and overall health of mature trees and vines.

    Building up or conserving organic matter is possible to one degree or another in every agricultural system I am familiar with. No-till or minimum-till is practiced in many annual crop scenarios, and in order to maintain good soil tilth and nutrient holding capabilities, animal manures or green manures can be top dressed or incorporated. Crop rotation is practiced in annual crops to maintain soil health and productivity. That can be done in permanent crops too, but on a scale of 15 or 20 years or more rather than months. The principles for managing soils are much the same for permanent and annual cropping systems, how they are implemented is different. One must ensure that soil structure allows for root penetration, oxygenation (roots respire) and internal drainage, and that it acts as a suitable matrix for applied nutrients. Soils must also be “alive”, but that does not necessarily mean that one must preserve all native soil organisms, just those that benefit the cropping system; things like nitrifying bacteria, bacteria and fungi responsible for decomposition of organic matter and mycorrhizae that help facilitate nutrient uptake.

    Most modern pesticides have a very short half life in soil and there is usually more than enough life in soils associated with all types of agricultural systems unless someone really goes out of their way to sterilize it, but I have no doubt that the life that is present is not the same as it would be had the soil not been farmed.,

  22. BBBlue says:

    Yehouda Harpaz- You are right, I was a bit careless, but I too saw that they are the same crap and failed to make a distinction.

  23. Npsychdoc says:

    Appreciate the reply, BBBlue. I’ve learned a lot through this discussion.

  24. cmyounger says:

    While not recent, some research concerning the effects of glyphosate on soil organisms is referenced here:

    _Do glyphosate based herbicides harm soil organism such as worms?_

  25. birkett83 says:

    It might help to put the IARC findings in context by looking at the list of IARC “probable carcinogens” that glyphosate is in. Notable highlights include:

    • Red meat
    • Shift work
    • Hairdressing
    • Several chemotherapy drugs used to *treat* cancer

  26. jss367 says:

    Dr. Novella,

    Please correct me if I’m mistating you, but you seem to say that for chemicals used in industry but not intended for human use, there is no human data prior to its introduction. Epidemiological studies take years (or decades) of data to be meaningful. How can this method be trusted to prevent the introduction of another DDT, which was widely used for about three decades until it was banned?

  27. jss – that is what animal data is for. You cannot do a study in which you give people possible toxins to see what happens. That would be unethical. So we study them as best we can with in vitro and animal data, with a wide margin of apparent safety. Then, once they are in use, we monitor the effects epidemiologically.

    Do you have an alternative?

  28. M Hewett says:

    The answer to does glyphosate cause cancer has been answered and it would be great if more research could go into other long term impacts such as soil quality and food sustainability.

  29. chikoppi says:

    I’m not familiar with “chlorpyrifos,” but expect it will enter the public discussion soon…

    WASHINGTON — Scott Pruitt, the head of the Environmental Protection Agency, moved late on Wednesday to reject the scientific conclusion of the agency’s own chemical safety experts who under the Obama administration recommended that one of the nation’s most widely used insecticides be permanently banned at farms nationwide because of the harm it potentially causes children and farm workers.

    The ruling by Mr. Pruitt, in one of his first formal actions as the nation’s top environmental official, rejected a petition filed a decade ago by two environmental groups that had asked that the agency ban all uses of chlorpyrifos. The chemical was banned in 2000 for use in most household settings, but still today is used at about 40,000 farms on about 50 different types of crops, ranging from almonds to apples.

  30. ashap says:

    Fun fact for your 2016 expert panel paper, make sure to check out the declarations of interest before saying all things are equal:

    “Funding for this evaluation was provided by the Monsanto Company which is a primary producer of glyphosate and products containing this active ingredient. Neither any Monsanto company employees nor any attorney reviewed any of the Expert Panel?s manuscripts prior to submission to the journal.”

    So they provided funding for this review? Do you think they might look for people who might side favorably? I probably would.

  31. garyrees says:

    I am confused Steve. In your closing paragraph you appear to use the words pesticide and herbicide interchangeably. Is glyphosate a herbicide or a pesticide or both?

Leave a Reply