Glyphosate and Cancer: What does the data say?

A little over a week ago, the International Agency for Research on Cancer (IARC) announced that glyphosate would be added to their list of agents that are “probably carcinogenic to humans.” Glyphosate wasn’t the only pesticide added to the list, but as Nathanael Johnson noted at Grist, glyphosate tends to be something of a lightning rod due to its association with genetically engineered (Roundup Ready) crops. Let me start by pointing out I’m pretty late to the party writing about this. The IARC is a well-respected agency within the World Health Organization, so this announcement has been widely reported. And it should surprise nobody that Monsanto is vehemently denying any health concerns, while the usual suspects who oppose GMOs and pesticides are using it to advance their agendas. I think the aforementioned piece by Nathanael Johnson at Grist and a piece by Dan Charles at NPR do a good job of putting this new classification into context. Grist also posted a really cool video that explains what the IARC group 2A classification (“probably carcinogenic to humans”) actually means.

Rather than simply re-state what others have said on the topic, I wanted to actually take a thorough look at the evidence supporting this classification. I work with pesticides (especially glyphosate) on a regular basis, so I take this classification very seriously. If glyphosate is indeed likely to cause cancer, I am in the group of people who is most likely to be affected. As most of the reasonable write-ups have previously noted, IARC group 2A agents are problematic mostly for occupational exposure; that is, people who work with (or around) the chemical on a regular basis over a long period of time. The general public is highly unlikely to see any ill effects from any agent with this classification based on available evidence. I’m disappointed that IARC decided to announce the classification about a year before they plan to release the full monograph that details their reason for the decision. Having their list of references sure would have been useful to determine which data they’re using to come to that conclusion. So I did a literature search for studies that included glyphosate and cancer. A recent review article by Pamela Mink et al. (2012) provided a nice starting point. It should be noted that the review article was funded by Monsanto; however, I didn’t actually rely on the conclusions of the Mink paper, so that potential conflict is mostly irrelevant. I simply used the Mink article as a starting point to find research articles that investigated the link between glyphosate and cancer.

Recently, Vox presented a very nice figure that summarized why you shouldn’t put too much faith in any single study about things that cause or cure cancer. I used that as a model to create this figure, which summarizes all of the information I could find relating glyphosate exposure to cancer.

AllCancer_Allcites

In the figure, each point represents the relative risk of developing cancer between people who had been exposed to glyphosate and those who hadn’t. To interpret the figure, any points on the left side of the blue line (less than 1) means that, on average, people who were exposed to glyphosate were less likely to get that type of cancer. Points to the right of the blue line mean that people exposed to glyphosate were more likely to get that type of cancer. There are two important things to note about this figure. First, this is an obvious over-simplification of the data. Presenting the data this way excludes the uncertainty of the relative risk estimates. When a study presents these estimates, they usually also present 95% confidence intervals. Those intervals are critical to determining whether we should put much faith in the estimate. Generally speaking, if the confidence interval spans across 1, then we would conclude that the evidence is too weak to suggest any causal link. Even so, if we have similar numbers of points to the left and right of 1, or the points are all clustered very close to 1, we can safely conclude there is little evidence of a link.

The second point about the figure above is that there appear to be many points on the right side for non-Hodgkin lymphoma. This is important, because that is the type of cancer specifically called out in the Lancet Oncology article that the IARC used to officially announce their new classification. The table on the first page of the Lancet paper states that “Evidence in humans” is “Limited”, with the cancer site listed as “non-Hodgkin lymphoma.” The Lancet Oncology paper lists only 16 references, and as far as I could tell, only 3 of those references actually contained information on glyphosate and non-Hodgkin lymphoma (henceforth referred to as NHL). And those 3 references do seem to suggest a link between glyphosate exposure and NHL.

NHL_iarccites

All three of the studies in this figure are “case control” studies. This type of study takes a large number of ‘cases’ of the disease of interest, finds a similar group of people without the disease, and then tries to find differences in risk factors between the groups. Any factors that are more prevalent in the ‘case’ group (the group with the disease) are viewed as possible risk factors for the disease. Case control studies can be very useful, as Vox points out here. In the three case control studies referenced in the IARC Lancet paper, all of the point estimates are to the right of 1. But the confidence interval from McDuffie et al. (2001) paper includes 1, indicating that the evidence for a link in that study wasn’t very strong. Similarly, DeRoos et al. (2003) used 2 different models, and the confidence interval for one of those models contained 1. As I looked through a variety of case control studies, multiple models were common. The authors would sometimes evaluate 2 or even 3 different models comparing glyphosate-exposed and non-exposed people. More on this later. I was able to find several more studies (in addition to the 3 that IARC referenced) that investigated links between glyphosate and NHL. All of those studies are summarized in the figure below:

NHL_casecontrol

Although many of the confidence intervals contain 1, all of the point estimates are greater than 1. So although there is a lot of variability in the data, the association of glyphosate exposure and NHL does seem to be reasonably consistent across studies. Perhaps this is what the IARC panel saw when they arrived at their conclusion. Similar to DeRoos (2003), both Hardell studies employed more than 1 model. In the studies I read, the difference between models was usually an attempt to adjust for confounding variables. The most common confounding variable in the NHL studies was exposure to other pesticides. A very large percentage of people who are exposed to glyphosate for long periods are also exposed to many other types of pesticides. This is a very important limitation of case control studies. Most people who use glyphosate a lot (like farmers, commercial pesticide applicators, and weed scientists) tend to be exposed to many compounds that are much more rare among the general public. We certainly tend to use a variety of pesticides, but probably also inhale more dust and fertilizers. We are out in the sun a lot. We probably also get exposed to more hydraulic fluid and wake up earlier than the general population. These things are extremely difficult to control for in a case control study.

Additionally, in the case control studies I read, a very small minority of NHL cases were actually exposed to glyphosate. For example, only 97 people (3.8% of the study population) had been exposed to glyphosate in the DeRoos (2003) study. Similarly, only 47 people (2.4% of the study population) had been exposed to glyphosate in the Eriksson (2008) study. These are very small numbers. To look at it another way, only about 3% of the NHL cases in most of the case control studies had actually been exposed to glyphosate. So even if glyphosate does increase the risk, it certainly is not a major contributor to NHL cases in the general population.

But case control studies aren’t the only types of studies that have been used to investigate the link between glyphosate and cancer. DeRoos et al. conducted a follow up to their 2003 study using a different, and arguably better methodology. Cohort studies follow a group of people during some portion (or all, depending on the study) of their lives, and track many risk factors and health outcomes. DeRoos et al. (2005) looked at a group of 54,315 agricultural workers. Once again they used two different models in their analysis, but the results of this study were contrary to what was observed in the case control studies.

NHL_cohort

The point estimates were actually less than 1.0, with confidence intervals that contain 1. These results suggest there is no discernible link between glyphosate and non-Hodgkin lymphoma among a population where glyphosate use is the most common. Over 41,000 of the 54,315 study participants had been exposed to glyphosate in this study. And 99.82% of them did not get non-Hodgkin lymphoma during the course of the study.

So what does this all mean? I may change my mind when the IARC’s full monograph is published later, but based on the data I could find, I don’t see any evidence for alarm.  And I say that as someone who is exposed to more glyphosate than a vast majority of the population. There is nothing here that I think can tarnish glyphosate’s reputation as a very safe pesticide. But that doesn’t mean that we should throw caution to the wind and douse ourselves in it. And for god’s sake, please stop saying things like “glyphosate is safe enough to drink.” Drinking Roundup doesn’t prove anything, anyway. I could smoke a cigarette and drink a beer in front of a crowd, but that doesn’t make alcohol and cigarettes any less responsible for causing cancers. Glyphosate is still a pesticide, after all. Proper protective equipment and procedures should be followed when any pesticide is used. But when used according to label directions, I think there is no reason to be scared whether you’re a homeowner trying to get rid of weeds in your sidewalk, or a commercial applicator spraying 1,000 acres of Roundup Ready corn.


Find this post helpful? Please consider taking a survey about science blog readership!


Suggested links:

So Roundup “probably” causes cancer. This means what, exactly? by Nathanael Johnson, Grist

A top weed killer could cause cancer. Should we be scared? by Dan Charles, NPR

He’s not a Monsanto lobbyist, and weed killer isn’t safe to drink. by Matthew Herper, Forbes

Watch stick figures explain what “probably causes cancer” even means. by Suzanne Jacobs, Grist

Expert reaction to carcinogenicity classification of five pesticides by the International Agency for Research on Cancer (IARC). Science Media Centre

Weed killer, long cleared, is doubted. Andrew Pollack, New York Times

Monsanto’s statement on the IARC classification of glyphosate. Monsanto

Glyphosate as a carcinogen, explained. Farmers Daughter USA

Roundup and Risk Assessment. Michael Specter, The New Yorker

Comments

  1. Andrew. Thanks for a good wrap up and explanation. Let me also throw in (code for soap box) some thoughts as a statistician and analyst.

    Relative risk (RR) is the ratio of the probability of an effect in treated groups divided by the probability of the same effect in the control groups. As a ratio, it is inherently nonlinear and a bugger to interpret on its own. This can be seen in your plots (and those of Vox) where the scale is logarithmic. Equal distances above and below 1.0 on the graph are vastly different changes in RR. That is just a caution when looking at such plots.

    More problematic is the “relative” nature of these metrics. If, for example, we have a RR of 1.25, we can see it is greater than 1.0, implying increased risk relative to the control, but the real question should be what was the control probability? There are an infinite number of probability combinations that will give a ratio of 1.25 (or any other RR). As an example we might have Control: 0.0001 & Treated: 0.000125 where the treated increases the risk probability by 0.000025. Alternatively, we could have Control: 0.40 and Treated: 0.50, an increase in risk of 0.10. Both give a RR=1.25, but the changes in risk are vastly different. In the first case, we might note the control probability is so low that we think “Meh!”, but in the second case take more note with a 10% change from a already likely 40% occurrence. Without knowing the underlying probabilities, we really can not interpret RR alone well. Unfortunately, studies rarely report the underlying probabilities. (This is essentially the same as the problem of only reporting p-values and not the associated effect sizes).

    This also matters when the control probability is very low. It is easier to show a big RR with small control probabilities than large. For example, if our Control group above at 0.0001 changed to 0.0101 in the treated group, we would get a RR of 101! If, however, the control were a value of 0.40 and the treated changed to 0.41, the RR would only be 1.01. Both cases were a change of 1%.

    Take home message: Always look for, request reporting of, interpret RR (or odds ratios) in light of the actual underlying probabilities.

    1. Great points Bill. I tried to hint at some of this in the last few paragraphs, when I talked about the actual numbers involved in some of the studies. But in hindsight, putting the risk ratios into perspective should have probably been a bigger part of the analysis. I’ve put the actual numbers from the three case control studies IARC references into a file here: http://weedcontrolfreaks.com/wp-content/uploads/2015/03/IARC_CaseControlStudyNumbers.png and the numbers for the DeRoos (2005) cohort study here: http://weedcontrolfreaks.com/wp-content/uploads/2015/03/CohortNumbers.png Sorry in advance for not putting them in-line, our blog comment system apparently doesn’t support that very well…

      It is difficult to do this type of math on the case control studies, since they start with a large number of NHL cases, then find similar control groups, then look for risk factors that may play a role. The numbers below aren’t adjusted for age or other risk factors, as they would be in the model. These are just the raw percentages of NHL cases & controls who had used or been exposed to glyphosate in the studies:

      DeRoos 2003:
      Controls: 3.3% used glyphosate
      Cases: 5.9% used glyphosate

      Eriksson 2008:
      Controls: 1.8% used glyphosate
      Cases: 3.3% used glyphosate

      McDuffie 2001:
      Controls: 9.7% used glyphosate
      Cases: 10.9% used glyphosate

      Cohort studies make it a little easier to put the actual risk into context. Several sources online state that the chances of someone in the U.S. getting non-Hodgkin lymphoma during their lifetime is somewhere around 2% (or about 2,000 out of 100,000). The rate of NHL in the DeRoos (2005) cohort study of agricultural workers was much lower than that (0.17%, or only about 170 people out of 100,000). This is probably due to age of the study participants. Just by chance (and not due to glyphosate), quite a few more of the participants in that study are likely to get NHL before they die. [that’s probably the darkest sentence I’ve ever written in a blog comment]

      If the risk of NHL is approximately doubled by using glyphosate (based on the case control studies), the DeRoos (2005) study should have found that occupational exposure to glyphosate would increase the new diagnosis rate for this population from 0.17% to 0.34%. But their results actually showed almost the exact same rate of NHL; 0.16% of the non-glyphosate compared to 0.17% of the glyphosate group (again, not adjusted for other factors such as age). A follow up of the participants in this study may indeed find an increased rate of NHL in the cohort among glyphosate users. But as of now, the evidence just doesn’t seem to be there.

      1. Art Preachuk

        The Fear Mongering continues…based on a probable finding from a probable study? It makes me wonder if some one is being paid to produce probable results? I guess they have gotten the media coverage they wanted.On all the social media and local rags…I first head it on the radio. The station, in its lead up to the Glyphosate segment stated 3 times that “a new study was out that proved Glyphosate causes cancer. At the start of the segment they added “probably”
        So the seed has been sown and we in Ag / and Veg Control are forced to fight the Ghosts!!
        Good article and analysis…too bad they got the Front page and you are in Control Freeks…keep up the resistance!!

  2. Pesticide is the general term. subcategories include herbicides, insecticides, fungicides, etc.

  3. Many thanks for a great piece. It is a pity that it only appears in a blog and does not get into media available to the general public.

    There is in my view (based upon the article in the Lancet) a major flaw in the IARC evaluation: the case control studies did not evaluate risks in relation to glyphosate, but in relation to RoundUp. Co-formulants are a confounding factor that cannot be eliminated from such studies.

    Glyphosate also comes in different forms (acid and various salts). There is a possibility that those forms have an influence on the health effects.

    Furthermore, the studies are rather old (2001, 2003 and 2008). A conclusion has thus been drawn in 2015 on the basis of studies that are more than 10 years old for two of them, and exposures that are even older and are likely to go back to the first years of RoundUp. In the meantime, the glyphosate form used and the co-formulants might have changed. This in turn might (contribute to) explain why more recent studies do not find an effect.

    A large cohort study (180,000 people) is underway in France. The latest results give a non-significant 6 to 22% increase in the incidence of NHL for the male agricultural and rural population relative to the general population, and 3 to 32% for the female population. This is based on 605 new cases (395 male and 210 female).

    .

    You wrote: « The IARC is a well-respected agency within the World Health Organization ». Well, this a judgment that needs to be reevaluated. It is not confined to IARC.

    Fact is that expert groups are more and more influenced by pressure groups and media hysteria. Has this group succumbed ? Possible.

    Fact is also that the press services of such organisations (are tempted to) respond to media and public expectations.

    For carcinogenicity in animals, the reference in the Lancet is a joint FAO/WHO meeting report :

    WHO/FAO. Glyphosate. Pesticides residues in food 2004 Joint FAO/WHO Meeting on Pesticides Residues. Part II Toxicological. IPCS/WHO 2004; 95–162.
    http://www.who.int/foodsafety/areas_work/chemical-risks/jmpr/en/

    Unless I fetched the wrong document from the FAO website, the report states :

    « Long-term studies of toxicity and carcinogenicity were conducted in mice and rats. In the study of carcinogenicity in mice, no toxic effects were observed at up to the highest dose tested (1000 mg/kg bw per day), and there was no evidence of carcinogenicity. »

    http://www.fao.org/fileadmin/templates/agphome/documents/Pests_Pesticides/JMPR/Reports_1991-2006/report2004jmpr.pdf

    There we are… until we get the full monograph… some time next year. This begs the question of why the haste. As far as I am concerned, I am more than disappointed. The monograph will eventually show that there is merit in the classification, but the information currently available does not speak well about this IARC decision.

    So why the decision and the timing? An attempt to influence the reevaluation processes, particularly in Europe? Cannot be excluded… hence the respect for IARC goes down.

    But this is not all.

    IARC’s press release (http://www.iarc.fr/en/media-centre/iarcnews/pdf/MonographVolume112.pdf) also refers to the first EPA classifications:

    «  On the basis of tumours in mice, the United States Environmental Protection Agency (US EPA) originally classified glyphosate as possibly carcinogenic to humans (Group C) in 1985. After a re-evaluation of that mouse study, the US EPA changed its classification to evidence of non-carcinogenicity in humans (Group E) in 1991. »

    The incontrovertible conclusion I draw from the cited document is that:

    1.  the original classification was based on an increased incidence of renal tubular adenomas in males;

    2.  the data were reviewed in 1986, the original finding was found to be erroneous and it was proposed to rate glyphosate D (inadequate animal evidence of carcinogenic potential);

    3.  the proposal was put aside in favor of further studies, which led to the 1991 rating.

    The description of this process by IARC is thus flawed. In any event, renal tubular adenomas are unrelated to NHL, which raises the question why this whole story has been included in the paper.

    .

    On another front, IARC classified radiofrequency electromagnetic fields as possibly carcinogenic to humans (Group 2B), based on an increased risk for glioma, a malignant type of brain cancer , associated with wireless phone use.

    http://www.iarc.fr/en/media-centre/pr/2011/pdfs/pr208_E.pdf

    The basis for the decision was extremely thin. The press release states:

    « The evidence was reviewed critically, and overall evaluated as being limited2 among users of wireless telephones for glioma and acoustic neuroma, and inadequate3 to draw conclusions for other types of cancers. The evidence from the occupational and environmental exposures mentioned above was similarly judged inadequate. The Working Group did not quantitate the risk; however, one study of past cell phone use (up to the year 2004), showed a 40% increased risk for gliomas in the highest category of heavy users (reported average: 30 minutes per day over a 10‐year period). »

    It also states:

    « Dr Jonathan Samet (University of Southern California, USA), overall Chairman of the Working Group, indicated that “the evidence, while still accumulating, is strong enough to support a conclusion and the 2B classification. The conclusion means that there could be some risk, and therefore we need to keep a close watch for a link between cell phones and cancer risk.” »

    The monograph (http://monographs.iarc.fr/ENG/Monographs/vol102/mono102-005-006.pdf) states for its part:

    « There was, however, a minority opinion that current evidence in humans was inadequate, therefore permitting no conclusion about a causal association. »

    Fairness and transparency would have required that such a minority opinion be recorded in the material released to the public.

    And what does the Geneva-based WHO headquarters have to say on this :

    http://www.who.int/mediacentre/factsheets/fs193/en/

    « A large number of studies have been performed over the last two decades to assess whether mobile phones pose a potential health risk. To date, no adverse health effects have been established as being caused by mobile phone use. »

    And:

    « The international pooled analysis of data gathered from 13 participating countries found no increased risk of glioma or meningioma with mobile phone use of more than 10 years. There are some indications of an increased risk of glioma for those who reported the highest 10% of cumulative hours of cell phone use, although there was no consistent trend of increasing risk with greater duration of use. The researchers concluded that biases and errors limit the strength of these conclusions and prevent a causal interpretation. »

    1. Andre, you quoted the FAO site as saying « Long-term studies of toxicity and carcinogenicity were conducted in mice and rats. In the study of carcinogenicity in mice, no toxic effects were observed at up to the highest dose tested (1000 mg/kg bw per day), and there was no evidence of carcinogenicity. » But where EDCs (endocrine-disrupting chemicals) are concerned, this information is totally irrelevant.

      What were the lowest doses tested? You should be aware that when it comes to the endocrine system, “The does makes the poison” becomes irrelevant (I wonder why people who pride themselves on their science cling to a dictum coined by a sixteenth century astrologer and alchemist). According to the Endocrine Society, and they should know, “EDCs have effects at extremely low doses (typically in the part-per-trillion to part-per-billion range) to regulate bodily functions.” (see http://www.endocrine.org/~/media/endosociety/Files/Advocacy%20and%20Outreach/Important%20 Documents/Introduction%20to%20Endocrine%20Disrupting%20Chemicals.pdf).

      Disruption of these functions causes a wide variety of disorders, including (but not limited to) leukemia and brain, breast, testicular, prostate and uterine cancers.
      Glyphosate is an EDC (see Table 2 9n the Endocrine Society statement) and the amount of glyphosate used on crops has increased tenfold in the last two decades. Glyphosate residues well within the endocrine-disrupting range have been found in glyphosate-resistant soybeans (Arregui et al. 2004), in the urine of cows fed with glyphosate-resistant (Kruger at al. 2013), and in human urine (Kruger et al, 2014. In other words, transmission of an EDC is documented through every link in the food chain. Data on the increases in cancer and other disorders attributable to EDCs over the same two decades is provided in Swanson et al. 2014.

      I’d be interested to see what you and Andrew make of all this.

      References:
      Arregui, M. C., Lenardón, A., Sanchez, D., Maitre, M. I., Scotta, R., & Enrique, S. (2004). Monitoring glyphosate residues in transgenic glyphosate‐resistant soybean. Pest Management Science, 60(2), 163-166.
      Krüger, M., Schrödl, W., Neuhaus, J., & Shehata, A. A. (2013). Field investigations of glyphosate in urine of Danish dairy cows. Journal of Environmental and Analytical Toxicology, 3(5): 186-192.
      Kruger, M., Schledorn, P., Schrödl, W., Hoppe, H. W., Lutz, W., & Shehata, A. A. (2014). Detection of glyphosate residues in animals and humans. J Environ Anal Toxicol, 4(210), 2161-0525.
      Swanson, N. L., Leu, A., Abrahamson, J., & Wallet, B. (2014). Genetically engineered crops, glyphosate and the deterioration of health in the United States of America. Journal of Organic Systems, 9(2), 6-37.

      1. This post is about IARC’s decision to classify glyphosate as « probably carcinogenic to humans. Mr. Kniss has provided a piece of information based upon the scarce explanations at hand (the paper in The Lancet) and an advice :

        1.  The three studies referred to in the Lancet paper by a group from among the Working Group are very weak as regards our subject matter.

        2.  Applicators, protect yourself.

        Coming now to your comment, if you want to know what were the lowest doses tested on animals, you will have to go through the whole FAO/WHO paper (once found) and probably also those referred to therein. Since you are on the EDC trail, you would also have to look for the precise periods of the life of the test animals in which glyphosate was administered.

        « You should be aware… » ? Many thanks for the refresher. Although I have not yet seen any convincing paper which demonstrates that “The does makes the poison” is irrelevant in the case of EDCs. The dose-response may not be linear, but the dose is not altogether irrelevant.

        Your parenthesis is also welcome. It opens avenues for scientific and philosophical thoughts. I may turn it round, with two additions and a correction – lawyer’s fashion : I wonder why people who do not pride themselves on their science – and nevertheless pretend to teach scientists – reject a dictum coined by a sixteenth century physician, botanist, astrologer and alchemist, which dictum can be checked by everyone in daily life.

        Glyphosate would be an EDC according to the Endocrine Society ? Their paper is rather confused. Table 2 is preceded by: «  There are over 85,000 manufactured chemicals, of which thousands may be EDCs. A short list of representative EDCs and their applications is provided in Table 2 ». And later on : «  Herbicides in widespread use such as atrazine, 2,4-D, and glyphosate, are considered EDCs, and the fungicide vinclozolin is a known EDC. » So is it, or is it not an EDC for sure (for the ES)? Or is it just because some activist scientists unable to demonstrate an adverse effect settle for an EDC effect?

        I take it that standard studies have failed to show any effects indicative of endocrine modulation.

        http://www.ask-force.org/web/HerbizideTol/Williams-Safety-Evaluation-Risk-Assessment-RR-2000.pdf

        http://npic.orst.edu/factsheets/glyphotech.html#references

        Finally you quoted four papers.

        Two are from Monika Krüger. The Seralini rule – if you favorably cite the 2012 Séralini rats fed on Roundup ready maize study, you just lost the argument – applies mutatis mutandis :

        http://skeptico.blogs.com/skeptico/2013/06/the-seralini-rule-gmo-bogus-study.html

        As for Swanson et al., give me a break…

      2. Dear (I am guessing)Dr. Bickerton: Thank you for some intelligent evaluations and for the solid references you provided. When you add up the total diets of children and adults and at every meal they are ingesting several foods that have glycophosphate in them through their growing and harvesting I support the issue that this most definitely is a serious cause for concern.I will follow your example and write a well researched article for my local paper. Farmers must begin to realize there is more to providing food than the ease of growing and harvesting, aiming for quantity and profits.Sincerely, A retired nurse from Illinois

  4. I’m so glad you looked at this seriously. You are right that this has a direct effect on you, and your family, and your students–in ways that are only theoretical to most people.

    I’m glad you can be the voice of the front line on this. I’m sure it matters to you.

    The other thing that worries me about this drama, though, is that like the Seralini-style trial-by-newspaper-headline, some countries will opt for bad decisions as a result. Like Chipotle, maybe they’ll move to older herbicides with worse characteristics. And kids in fields will be exposed to worse things. That kind of side-effect of this is a real possible consequence of misbranding glyphosate.

  5. This is very helpful. I will continue to use small amounts of glyphosate in my home garden, limiting weed control in vegetable gardens to manual weed removal where possible. We live in a world filled with chemicals; all one can do is stay abreast of the evidence and minimize exposure through hand-washing etc.

  6. I’m surprised that in all this commentary and the article, nobody has mentioned the review level article that establishes a positive association of a 2.0 risk factor between glyphosate exposure and B-cell lymphoma.

    Schinasi, L (April 23, 2014). “Non-Hodgkin lymphoma and occupational exposure to agricultural pesticide chemical groups and active ingredients: a systematic review and meta-analysis.”. International Journal of Environmental Research and Public Health 11 (4): 4449. doi:10.3390/ijerph110404449. PMID 24762670. Retrieved 31 August 2015.

  7. The IARC report said that glyphosate causes cancer in lab animals. Full stop.

    It *causes* cancer in lab animals.

    There is no “probably” in this — the “probably” is in relation to humans because we do not have the same kinds of data on humans as is possible in lab animals. We have only wide epidemiology studies and those are not adequate to date to answer the questions we need to answer.

    The IARC made this assessment using many sources of data, including Monsanto’s own internal studies that are referred to in the 1991 EPA memo on the subject of glyphosate and cancer. That memo is quite interesting, for even while it declared that the data does not show tumorigenesis because there is not a monotonic positive trend across three dose groups, three of the EPA toxicologists refused to sign and wrote things like “DO NOT CONCUR” where their signature would be, obviously not wanting to endorse the weasely non-scientific excuse that the memo contains, even while showing data that clearly shows tumorigenesis being correlated strongly to glyphosate exposure. We cannot base our food supply on weasel argument.

  8. How much did Monsanto pay you to write this?
    No, we shall oppose Monsanto and their death chemicals and Frankenstein food.

  9. I am late to the discussion, but the lawsuits are mounting against Monsanto in relation to the link between NHL (Non Hodgkin Lymphoma) and glyphosate and other reasons as well. As a medical interpreter I see cases of workers ill from excessive exposure to glyphosate and other herbicides/pesticides. Whenever possible I ask workers using glyphosate if they have been advised of proper use of the chemical to protect themselves from overexposure. The answer is always the same- No, I have been told it is safe.
    In addition, I wonder if the countries that have banned it have inferior science compared to ours or is it something else. Do they have higher standards for protecting their people? What do they see that we don’t? This matter is far from being resolved and I for one think that the people who have to use this chemical day in and day out should be better protected.

  10. Most of the intelligent arguments are wasted on me. You might argue that they “Go over my head”. You could be right. We all understand how such products help to increase crops and feed the world etc. etc. All I know is that I have M.C.S. (multiple Chemical Sensitivities.” Before you ask,—-Yes properly tested and diagnosed by one of the country’s leading allergy specialists. That has made an indescribable effect on my very existence for over twenty years now.
    Whilst the argument goes on I sit here wondering what life will be like in about a hundred years from now? Will anyone have listened?. Will another way be found?
    I doubt it. No, heads will remain ‘Stuck in the sand’. Let us just hope that more and more money will be found to pour into the pot of health care. One day it could even become impossible to avoid health systems from collapsing under the uneven weight of the scales of FIT and SICK people. Could the Chicken and Egg question come into play then, or is there still time for it to come into play right now?
    Should more emphasis be put into cutting down on chemical uses wherever possible,
    Or remain as it is now on finding cures for the illnesses which might possibly be direct results of the chemical use? Wool and cotton used to be fine. You probably can’t even smell anything strange about your clothes made with man made fibres.
    However that content in the fabric of your clothes ‘gasses off’ for the duration of it’s lifetime. It can effect a chemically sensitive person to a serious degree by just breathing it in. In my own case causing breathing difficulties. I’ve got the T shirt. (A cotton one now of course.)
    That is just one millionth of problems from chemicals.
    If only we could have foreseen what awful outcomes there can be from chemicals would we have ‘stopped progress’? Of course not. There is always someone out there who knows what is best for all of us. Even now those people will probably just try to find some other chemical cure for the ills of the previous chemical catastrophes. Thank goodness I am 84 years old and won’t live long enough to have to listen to the amazed reactions to what will have happened to the world when that 100 years arrives. (If it does)

    Ena Bowles

Comments are closed