I was very pleased to have the opportunity to testify in Congress to defend the reputation and integrity of the International Agency for Research on Cancer (IARC), the cancer research arm of the World Health Organization. It has been the target of an orchestrated attack campaign from Monsanto Co. and other corporate giants whose toxic products are being linked to cancer.
In Spring, 2015, IARC convened a Working Group of 17 cancer science experts from 11 countries to review the evidence, and they concluded that glyphosate, the main ingredient in the herbicide Roundup, “probably” causes cancer in humans (Group 2A, Volume 112). In immediate response, Monsanto and the agrochemical industry launched a ferocious anti-IARC campaign that IARC Director Dr. Christopher Wild described as, “unprecedented, coordinated efforts to undermine the evaluation, the program and the organization”. The goals of Monsanto’s anti-IARC propaganda campaign are to:
- support glyphosate registration and approval worldwide;
- defend itself against litigation claims by farmers that were once Monsanto customers and are now cancer patients;
- prevent labeling of glyphosate-containing products as containing a carcinogen in the State of California.
The House Congressional Hearing was titled: In Defense of Scientific Integrity: Examining the IARC Monograph Programme and Glyphosate Review (Feb 6, 2018). The House Science Committee in recent years has become known for holding hearings to advance the chemical industry’s agenda of attacking government chemical assessment programs. The full three-hour show trial can be viewed on the Committee website, along with the testimony of all four witnesses—three aligned with Monsanto, and my lone testimony defending the process and conclusions of the IARC Working Group.
Republicans are threatening to cut funding for cancer research after @WHO researchers found that Glyphosate is carcinogenic. @NRDC’s senior scientist Dr. @JBSass testified to Congress to defend science and speak out against this deadly pesticide. pic.twitter.com/3xY1y33lLg
— NRDC 🌎 (@NRDC) February 8, 2018
The industry-led criticisms of the IARC Monographs are part of a documented public relations campaign described by Dr. Jonathan Samet, a prestigious medical professor and frequent Chair of National Academies committees, as, “…archetypical of strategies for creating ‘doubt’ about scientific evidence that has policy implications. Such strategies can be traced to the ‘playbook’ of the tobacco industry for discrediting findings related to active and passive smoking.” (Samet 2015)
It seems that these tobacco industry strategies have also captured the pesticide regulatory agencies on both sides of the Atlantic. Both the US EPA Pesticide Office and the European Food Safety Authority (EFSA) have classified glyphosate as not linked to cancer, breaking with long-standing practice of aligning with the IARC cancer hazard assessments. To quote an internal Monsanto memo, “[Monsanto Regulatory Affairs] is not aware of a situation where a regulatory body took a different position than IARC”. (Monsanto memo, Feb 23, 2015).
So, why are the US and EU pesticide regulatory agencies so closely aligned with the perspective of the big agrochemical companies? We may never know the full story, but we know this much:
- There has been a disturbing level of communication and collaboration over a number of years between Monsanto employees and senior EPA Pesticide Office official Jess Rowland, who headed up the EPA pesticide Cancer Assessment Review Committee. Monsanto internal emails in late 2015 specifically identify that Rowland, “will be retiring from EPA in 5-6 [months] and could be useful as we move forward with ongoing glyphosate defense.” Rowland has retired from EPA, but concerns of collusion sparked an investigation by the EPA Inspector General that is ongoing.
- The European Food Safety Authority (EFSA) assessment of glyphosate was essentially drafted from Monsanto documents. The European Parliament hosted a public hearing and launched an investigation of conflicts of interest at all levels regarding whether the European Commission followed its own regulations when approving glyphosate for another five years. That investigation is also ongoing.
The Science Committee’s hearing and the associated campaign against IARC have featured a series of misleading or made-up claims that target IARC’s glyphosate assessment, while promoting an alternate approach—which has largely been adopted by both the EPA Pesticide Office and EFSA—in which evidence of cancer or other health harms is disregarded.
Industry’s methods are aggressively promoted by corporate chemical trade organizations that include Monsanto among its members; the International Life Sciences Institute (representing food manufacturers); CropLife America International (representing Agrochemical manufacturers); and, the American Chemistry Council (representing chemical manufacturers) that was dubbed The Cancer Lobby by NY Times columnist Nicholas Kristof.
Setting the Record Straight on False Science
I’ve blogged in the past about how the EPA Pesticide Office’s and EFSA’s cancer assessments differ from IARC, and also how the EPA Pesticide Office veered from the Agency’s own Cancer Guidelines in coming to its conclusions. Below I untangle some of the most oft-spouted false or manipulative claims of the chemical industry, using the IARC and EPA glyphosate assessments as a case study.
False: IARC is an extreme organization that says everything causes cancer
The Truth: Only about 20% of substances examined by IARC have been classified as known or probable human carcinogens
To date the IARC Monographs have evaluated over 1,000 chemicals or other agents, all with at least enough cancer data to support a nomination for consideration. Yet, only 120 are classified as known human carcinogens (Group 1) and only about 80, including glyphosate, as probable human carcinogens (Group 2A). About half have too little data to classify at all (Group 3), which is a terrible gap in our chemical regulatory process given that many of them are used in workplaces and in commercial products.
Some examples are below:
Group 1 Known to be carcinogenic to humans
asbestos, benzene, diesel exhaust, formaldehyde, tobacco smoke, Polychlorinated biphenyls (PCBs), outdoor air pollution, TCE
Group 2A Probably carcinogenic to humans
glyphosate, malathion, DDT, methylene chloride, polybrominated biphenyls, perchloroethylene
Group 2B Possibly carcinogenic to humans
1-bromopropane, acrylonitrile, carbon black, carbon tetrachloride, parathion, multiwalled carbon nanotubes (MWCNT-7),
Group 3 Not classifiable, inadequate evidence
prednisone, caffeine, methotrexate, aniline, isopropyl alcohol, single-walled carbon nanotubes
Group 4 Probably not carcinogenic
Caprolactam (used in some plastics)
False: Two Monsanto-sponsored review articles prove glyphosate is safe
The Truth: The review articles available at the time of IARC’s review did not report enough study details to be independently verified—had the detailed data been available, it may have strengthened the link with cancer
The centerpiece of the industry-manufactured criticism of IARC is that it allegedly disregarded two Monsanto-sponsored review articles, Greim et al 2015 and Kier and Kirkland 2013. In both cases, however, the publications didn’t provide sufficient details of the underlying studies “for independent evaluation of the conclusions reached by the Monsanto scientist and other authors,” according to IARC. IARC only considers publicly available information, and no other information or study details were available to the Working Group or the public.
Greim et al (2015) deserves special mention because the study details in the review article have since been made available to the European Food Safety Authority (EFSA), so we now know that there are many more tumors in the animal studies than the authors reported, making the link to cancer even stronger. These are the data sets that the Republican majority witness, industry consultant Dr. Tarone claims in his Hearing testimony were not reviewed by IARC, and which he alleges would have exonerated glyphosate. But rather than exonerating glyphosate, an analysis of those data conducted by Dr. Christopher Portier identifies an excessively high number of malignant lymphomas and hemangiosarcomas in male mice (see report, p. 37, Table 15).
Dr. Helmut Greim is himself of questionable scientific integrity. Dr. Greim chaired a ‘scientific panel’ funded by carmakers to counter the 2012 IARC determination that diesel exhaust is a known human carcinogen. Dr. Greim's panel conducted studies on monkeys—reported as ‘Monkeygate’—exposing them in a chamber to diesel exhaust. However, the studies were rigged because the cars in the chambers were using the “cheating” device that reduced emissions. The study was never published, but the events were reported in the NY Times. Reuters reported that the German government said such studies are unjustifiable.
False Claim: The recently published update of the National Cancer Institute’s Agricultural Health Study does not show a link between glyphosate and non-Hodgkin’s lymphoma (NHL), and therefore IARC is wrong
The Truth: The AHS update and other epidemiology studies show a link with cancer
The IARC Working Group identified epidemiologic studies from the US, Canada and Sweden that reported an elevated risk of non-Hodgkin lymphoma (NLH), a type of blood cancer, associated with exposure to glyphosate, even after adjusting for exposure to other pesticides (see IARC Monograph on glyphosate, p.75-76).
However, the Agricultural Health Study (Ag Health Study), which was included in the IARC Working Group assessment, did not show an excess NHL risk among the study subjects. This study has been ongoing since 1993 by the U.S. National Cancer Institute. It tracks pesticide exposures and health status for almost 90 thousand farmers and their spouses (called a prospective cohort study). The incremental results for many pesticides have been published in dozens of studies.
Lost or buried in much of the reporting of the Ag Health Study is that the 2017 study did find some evidence of a possible association between glyphosate and another type of blood cancer called acute myeloid leukemia (AML). The possible link with this type of leukemia should be very concerning to the public and particularly to pesticide applicators, because AML is a very serious fast-growing blood cancer, with only about one-quarter of the people that have it surviving longer than 5 years. The authors warn, “Given the prevalence of use of this herbicide worldwide, expeditious efforts to replicate these findings are warranted.”
Also published since IARC’s assessment is a 2016 Monsanto-sponsored systematic review and meta-analysis of epidemiologic studies, including the Ag Health Study, which specifically identifies a statistically significant risk of NHL from glyphosate exposure (Chang and Delzell 2016), strengthening the IARC assessment conclusions from a year earlier.
With the epidemiology studies identified in the 2015 IARC review linking glyphosate-based products to blood cancer, the 2016 Monsanto-sponsored systematic review and meta-analysis of studies identifying a link to blood cancer, and the 2017 Ag Health Study report of a possible association with another blood cancer type, there is compelling evidence that glyphosate-based products used out in the fields under real-world conditions pose an elevated risk of blood cancers.
False: The dose makes the poison, so things at low doses are safe even if they are toxic at high doses
The Truth: This statement—a favorite of chemical industry consultants and lobbyists—is about 500 years out of date
The chemical industry mantra is that the dose makes the poison, so that any consideration of hazard without its “context”—by which the industry means the exposure or dose—is outdated. This was the main—actually, the only—thrust of the Hearing witness Dr. Tim Pastoor, who retired from agrochemical giant Syngenta in 2015. This wisdom traces back to Paracelsus, a 16th Century physician, alchemist, and astrologer, who—consistent with medical understanding at the time—treated his syphilis patients with enough mercury to kill some of them, and drained the blood of others.
While the dose is certainly an important part of characterizing the potential risk posed by a substance, additional factors like genetics, co-exposures to other toxic substances, diet and lifestyle effects, health status, age at exposure, and duration or pattern of exposure can all have a strong, or even dominant impact on whether cancer or other adverse effects occur. For example, exposure to mercury or lead for even a short duration may cause severe neurological impacts to children, but have less or even no obvious effects on adults. A 2017 National Academies report emphasizes the importance of these factors when evaluating low-dose toxicity of chemicals.
False Claim: Tumors in rodents in the high-dose treatment group shouldn’t count as evidence of cancer risk
The Truth: Tumors in the high dose group predict cancer risks at lower doses
Laboratory studies in animals are conducted over a wide range of doses, often including high doses far above what the average person is exposed to, in order to have enough statistical power to detect an effect in a small number of animals. For example, if we are testing a chemical at a dose that causes cancer in 1-in-a-thousand people—a very high rate of cancer—we would need at least one thousand rodents at that dose just to see one rodent with tumors. For ethical, practical, and economic reasons, most of the glyphosate tests conducted by Monsanto used about 50 rodents per dose, which is standard practice. So, it is routine practice to have a high-dose treatment group, and then extrapolate from risks at the high dose down to lesser risks at lower doses. According to EPA’ own Cancer Guidelines, tumors at all doses should be considered relevant. (EPA 2005, p. 41).
EPA’s last assessment of glyphosate failed review by the independent expert Scientific Advisory Panel (SAP), with the Panel pointing out that EPA’s numerous ways of discounting tumors in the test animals (see details in my March 28, 2017 blog) was “flawed” and that it had failed to follow its own Cancer Guidelines in numerous critical ways (for example, see SAP March 2016 report, pages 18-21). The majority of SAP members also disagreed with EPA’s classification of glyphosate as ‘not likely’ a carcinogen, and felt that there was enough evidence of cancer risks to “suggest human carcinogenic potential of glyphosate…” (SAP March 2016 report, pages 16-17).
In its most recent revised draft assessment, EPA’s Pesticide Office has still not addressed the Science Advisory Panel’s concerns. The draft ignores most of the concerns raised by the Panel, and instead simply repeats the mantra that it is following a systematic review process for determining which studies to assess and rely upon. But nobody has seen or had an opportunity to review or comment on the Pesticide Office’s review methods. It is a black box in which, it appears, EPA is still relying upon the same approach that ran afoul of the Science Advisory Panel and violates the Agency’s own Cancer Guidelines, including improperly discounting cancer evidence.
False Claim: EPA followed international standards and best practices for study evaluation and data integration (systematic review)
The Truth: EPA and EFSA followed the agrochemical industry recommendations, which led to disregarding evidence that glyphosate may cause cancer
The EPA Pesticide Office and the European EFSA that approves pesticides for food crops both followed Monsanto’s methods for discarding and downplaying evidence of tumors in glyphosate studies (detailed in my blog, March 28, 2017). While EFSA developed their assessment from a draft provided to them by Monsanto directly, the EPA Pesticide Office says that it is using a systematic review process of the EPA’s Office of Chemical Safety and Pollution Prevention (OCSPP). This office, known as the Toxics Office, is now under the management of Nancy Beck, a chemical industry lobbyist prior to her recent political appointment at EPA. Dr. Beck’s previous foray into developing risk assessment guidelines was a failure, as evidenced by the National Academies conclusion that the draft government-wide risk assessment bulletin which she authored while at the Office of Management and Budget (OMB) was “fundamentally flawed” and the unprecedented recommendation for its withdrawal (NAS 2007).
The OCSPP review method used for the glyphosate assessment has not been subjected to public and stakeholder scrutiny, or peer review. Further, it veers from the recommendations of the National Academies and best practices of the EPA chemical testing program, Integrated Risk Information System (IRIS) in a number of critical ways, all of which are promoted by the chemical industry, and favor industry outcomes including:
- preferentially relying on Guideline studies, which are conducted by the regulated industry to support the approval of its products; and,
- preferentially relying on studies following so-called Good Laboratory Practices (GLP), which are required by industry product-testing labs to prevent malfeasance and misconduct.
An internal EPA memo indicates that the Pesticide Office failed to follow EPA Cancer Guidelines, and had it done so its assessment, and conclusions, might look more like IARC’s (ORD memo, Dec 14, 2015).
The problem with relying on Guideline (validated protocol) studies is that they are meant for chemical manufacturers to support the review and approval of their products. For that reason, they are most often designed to identify only major toxic effects (apical effects). However, a focus on only major health endpoints will not be predictive or indicate early-warnings of potential toxicity that may lead to other major adverse health outcomes. Guideline studies don’t necessarily use modern methods for evaluating chemicals and aren’t designed to grapple with the problems of low-dose exposures, endocrine or hormonal effects, behavioral or learning effects, immunotoxicity, cardiotoxicity, or pre-adverse ‘upstream’ effects like reduced sperm count or reduced anogenital distance which are predictors of infertility.
Dr. James Bus—previously a chemist for Dow Chemical—in testimony before the House Science Committee on behalf of the chemical industry, said high quality studies were ones that adhered to ‘Good Laboratory Practices’, i.e. GLP standards. The GLP standards have been required for industry test labs since the 1970s after flagrant violations and fraud were identified (Industrial Bio-Test laboratory was shut down and its directors were imprisoned as a result of the investigations). The chemical industry has used its considerable powers of alchemy to transform the restrictions imposed to prevent rampant fraud into a badge of honor and good housekeeping seal of approval. The industry likes to promote so-called “GLP studies” as higher quality and more reliable, because it favors the industry’s own studies over those of independent scientists in academia and government who aren’t subject to the same standards. Since those independent sources are also more likely to identify concerns over the safety of chemicals, whether or not the studies are “GLP” is a useful way of excluding studies that pose the greatest threat to the industry. However, while GLP standards are useful for ensuring that commercial laboratories don’t commit outright fraud, they are not a reliable indicator of whether a particular study is of low or high quality. A 2014 National Academies of Sciences report noted that GLP guidelines fail to prevent flawed, unreliable or biased-by-design studies (pages 62-63). The same report praised the EPA IRIS program systematic review methods which are not weighted to favor either GLP or Guideline studies.
Bad Science Also Bad for Health
Chemical corporations and their Cancer Lobby trade groups have been shown many times to defend their toxic products using manipulative tactics, false scientific information and phony front groups (see Chicago Tribune series on defending toxic flame retardants in household furniture and other consumer products, for example). As Nicholas Kristof states in his NY Times column, “The larger issue is whether the federal government should be a watchdog for public health, or a lap dog for industry.” That is, when regulated agencies are captured by the economic interests they are supposed to regulate, public health is sacrificed.
IARC Monographs are considered essential for informing cancer prevention strategies and effective public health decision-making around the world. Fundamentally, the recent Congressional Hearing and the larger public debate, which is mostly being fueled by Monsanto’s high-powered public relations and product defense campaign—is about the ability of a public health agency to call a carcinogen a carcinogen, even if the carcinogen makes a huge amount of money for a powerful corporation.
The predictable next step after the House Science Committee Hearing is an effort during the budget and appropriations negotiations this spring and summer to cut off U.S. funding for IARC and other public-funded chemical assessment programs. Of course, even without government chemical hazard assessment programs, the cancers will still occur—with their obvious terrible toll on individuals, families, health care costs, and the economy—but the tumors won’t be counted, and the causes won’t be tracked, making successful prevention more difficult.
We must continue to resist the endless lobby campaign of Monsanto and the chemical industry and protect government scientists and chemical assessment programs, to let them do the important work of generating credible publicly-available chemical hazard assessments on glyphosate, and all the other chemicals to which we are routinely exposed in our food, our drinking water, in household products, in building materials and through so many other every day routes of exposure. Our health depends on it.