The US E.P.A. Can No Longer Hide Behind Ghost Written Bogus Study on Glyphosate

Williams et al. (2000) has (finally) been retracted, per the journal, "to preserve the scientific integrity of the journal."

James Lyons-Weiler, IPAK-EDU.org

The long‑standing regulatory confidence in glyphosate, the world’s most widely used herbicide, once rested heavily on a single review published in 2000 by Williams, Kroes, and Munro. For decades the paper was treated as an independent, authoritative synthesis of toxicology, repeatedly cited by the Environmental Protection Agency and international regulators as a foundation for the assertion that glyphosate was “not likely to be carcinogenic.”

That façade collapsed in 2025, when the journal Regulatory Toxicology and Pharmacology formally retracted the article after new disclosures revealed that Monsanto employees substantially ghostwrote the manuscript, directed its conclusions, and concealed their financial involvement. The retraction, coming a full twenty-five years after publication, arrived in the midst of a wave of scientific, legal, and policy developments that together render the EPA’s continued reliance on the review indefensible.

The Williams et al. paper was central to EPA’s internal cancer classification documents, including the 2016 Glyphosate Issue Paper, which cited it as a key summary of toxicology studies in support of its “not likely to be carcinogenic to humans” designation (EPA-HQ-OPP-2009-0361-0073). The paper was also repeatedly referenced in the 2017 Draft Human Health Risk Assessment and the 2020 Interim Registration Review Decision. EPA staff used the Williams paper to bolster claims that glyphosate had no oncogenic potential in rodents and minimal risk to humans, often citing its summary tables of Monsanto-conducted studies.

When the Williams paper first appeared, it seemed to provide regulators with precisely what they wanted: a comprehensive rebuttal to mounting concerns about glyphosate’s carcinogenicity. Internally, Monsanto viewed the paper as a strategic asset. Disclosures from the “Monsanto Papers” litigation—including email exhibits released during the Johnson v. Monsanto trial—show that company scientists were involved in the drafting and editing of the manuscript. Although the authors listed were academic toxicologists, none disclosed Monsanto’s editorial or financial role.

In 2025, Regulatory Toxicology and Pharmacology formally retracted the Williams, Kroes, and Munro review after its handling co-Editor-in-Chief conducted an investigation into longstanding concerns about the paper’s authorship, data sources, and undisclosed conflicts of interest. The retraction notice explained that the journal sought an explanation from the sole surviving author, Gary M. Williams, and received no response. The editor documented that the article’s conclusions regarding the carcinogenicity of glyphosate were drawn solely from unpublished Monsanto studies, while multiple long-term carcinogenicity studies available at the time of writing were omitted without explanation. The notice further cited evidence from U.S. litigation showing that Monsanto employees may have contributed to the writing of the article without acknowledgment, raising “serious ethical concerns regarding the independence and accountability of the authors.” It referenced misrepresentation of contributions, undisclosed financial compensation from Monsanto, ambiguity regarding which portions of the article were authored by corporate personnel, and a biased weight-of-evidence approach shaped by unpublished industry studies. Concluding that these issues “undermine the academic integrity of this article and its conclusions,” the co-Editor-in-Chief stated that confidence in the review’s findings had been lost and that retraction was necessary to preserve the integrity of the journal. The full retraction notice is herehttps://www.sciencedirect.com:5037/science/article/pii/S0273230099913715.

This reckoning did not arise in isolation. Earlier in 2025, Environmental Science & Policy published a landmark paper by Alexander Kaurov and Naomi Oreskes titled The afterlife of a ghost-written paper: How corporate authorship shaped two decades of glyphosate safety discourse (DOI: 10.1016/j.envsci.2025.104160). Their bibliometric analysis showed that WKM2000 remained one of the most cited toxicology reviews on glyphosate worldwide—used in over 700 academic papers, EPA filings, and even AI training datasets. They documented how the paper’s influence extended into public knowledge platforms like Wikipedia, where editors referencing the review rarely noted its conflicts of interest or ghostwriting origins, even after legal revelations surfaced. Their work underscored that scientific fraud, when embedded in citation networks and reinforced by regulatory institutions, can influence generations of policy.

Only months after the Oreskes-Kaurov analysis, the journal retracted the Williams paper. The timing implicitly validated their critique and exposed the fragility of EPA’s continued reliance. The agency’s glyphosate cancer determination, last publicly affirmed on its website in September 2025, continued to claim that glyphosate presented “no risks of concern to human health” and was “not likely to be carcinogenic to humans”—echoing language shaped by the very review now withdrawn for ethical and scientific misconduct.

While the Ninth Circuit Court of Appeals vacated the human-health portion of EPA’s 2020 Interim Registration Review Decision for glyphosate in June 2022—holding that EPA’s “not likely to be carcinogenic to humans” conclusion did not satisfy the evidentiary standards required under FIFRA—the agency withdrew the interim decision and, as of its most recent public updates, has not issued a new final human-health determination to replace the vacated analysis.

Congress has extended EPA’s statutory deadline for completing the first round of registration‑review decisions for older pesticides to October 1, 2026, and EPA’s own schedule now anticipates issuing a final registration‑review decision for glyphosate by that date. That clock is now ticking without the Williams–Kroes–Munro review, which the journal has retracted, as a credible scientific shield.

Independent Research Did Not Support The Narrative

In contrast to EPA’s inertia, the independent scientific literature has advanced considerably. In June 2025, the carcinogenicity arm of the Global Glyphosate Study was published in Environmental Health (Panzacchi et al., 2025, Environ Health 24:36, 10.1186/s12940-025-01187-2). This large, long-term toxicology study exposed male and female Sprague–Dawley rats to glyphosate and two glyphosate-based herbicides from gestational day 6 through 104 weeks of age. The test materials were administered in drinking water at 0.5, 5, and 50 mg/kg/day, corresponding to the European Union acceptable daily intake (ADI) and no-observed-adverse-effect level (NOAEL) for glyphosate. The authors reported statistically significant dose-related increases in benign and malignant tumors at multiple anatomic sites, including haemolymphoreticular tissues (leukemia/lymphoid malignancies), mammary gland, liver, thyroid, nervous system, ovary, adrenal glands, kidney, urinary bladder, bone, endocrine pancreas, uterus, and spleen, with early-onset leukemia deaths in treated animals. They concluded that glyphosate and glyphosate-based herbicides “at exposure levels corresponding to the EU ADI and the EU NOAEL” produced robust evidence of carcinogenicity in experimental animals, consistent with IARC’s 2015 classification.

A comprehensive human-health review by Galli et al. (2024, Frontiers in Toxicology, 10.3389/ftox.2024.1474792) synthesizes biomonitoring, epidemiological, and mechanistic data on glyphosate and its main metabolite AMPA. Drawing on multiple population-based studies, the authors report that glyphosate can be detected in urine, blood, and breast milk, with approximately 60–80% of individuals in general populations—including children—testing positive in biomonitoring campaigns, and several cohorts in North America and Europe showing even higher detection frequencies. The review catalogues associations between glyphosate exposure and non-Hodgkin lymphoma in occupationally exposed workers, liver and adipose-tissue changes consistent with steatosis and non-alcoholic fatty liver disease, respiratory disease and asthma-like phenotypes, endocrine-disrupting effects in reproductive tissues, gut dysbiosis, and gut–brain axis–mediated impairment of cognitive function. Galli and colleagues emphasize converging evidence from in vitro, animal, and human studies that glyphosate and glyphosate-based herbicides have mutagenic/carcinogenic potential and act as endocrine disruptors, and they underline the discordance between this evidence and the non-carcinogenic hazard classifications adopted by EFSA, ECHA, and EPA.

Internationally, regulatory assessments have diverged. In 2015, the International Agency for Research on Cancer classified glyphosate as a “probable human carcinogen” (Group 2A) based on limited evidence for non-Hodgkin lymphoma in humans and sufficient evidence of carcinogenicity in experimental animals, together with strong evidence of genotoxicity. France has banned domestic and home-garden uses of glyphosate even while agricultural uses continue. Austria’s parliament voted for a national glyphosate ban in 2019, but procedural missteps and EU law prevented the law from entering into force; the country now applies partial restrictions rather than a full prohibition. Mexico issued a 2020 presidential decree calling for a phase-out of glyphosate by 2024 and has successfully defended the decree against several industry-backed legal challenges, although subsequent policy changes have delayed and weakened the original phase-out timetable. Meanwhile, in November 2023 the European Commission renewed glyphosate’s approval for ten years, until December 2033, after EU member states failed to reach a qualified majority either for or against renewal, triggering intense legal and political opposition from civil society groups and independent scientists.

The European Chemicals Agency’s Committee for Risk Assessment and EFSA’s pesticide panel have repeatedly concluded that glyphosate does not meet EU criteria for classification as carcinogenic, mutagenic, toxic for reproduction, or an endocrine disruptor, and EFSA’s 2023 peer review identified “no critical areas of concern” for human health in the representative uses assessed. These hazard and risk assessments rely primarily on GLP-compliant guideline studies submitted by manufacturers, including numerous long-term animal studies that remain unpublished but are treated as the highest-weight evidence in the regulatory dossier, while independent, peer-reviewed studies are often omitted or discounted. That evidentiary hierarchy parallels EPA’s now-vacated 2020 interim registration review decision, whose human-health section the U.S. Ninth Circuit Court of Appeals set aside in 2022 for failing to follow EPA’s own cancer guidelines and for mischaracterizing the underlying evidence. Critics such as Eva Novotny, in her Toxics review “Glyphosate, Roundup and the Failures of Regulatory Assessment” (2022, 10.3390/toxics10060321), characterise the EU and U.S. glyphosate assessments as cases of regulatory capture, marked by revolving doors between agencies and industry, heavy reliance on confidential manufacturer studies, and systematic sidelining of independent toxicological and epidemiological research.

Political Flip-Flops?

At the political level in the U.S., the first version of the MAHA Commission’s 2025 report raised alarm about environmental toxins and pesticides, identifying chemical exposures—including herbicides—among possible contributors to chronic disease in children. As debate swelled and industry and farm-lobby stakeholders mobilized, the September 9, 2025 MAHA Strategy Report was released with a starkly different tone: the public document omits the earlier language warning about pesticide hazards and instead emphasizes endorsing the existing pesticide-approval process and promoting cooperation with the agriculture and chemical sectors. Observers and public-health advocates, including those writing for The New Lede, characterize the shift as a retreat under pressure from chemical-industry and agribusiness lobbying, from a policy posture that once acknowledged pesticides as a central public-health threat to one that champions streamlined regulatory review.

The broader implications are profound. The Williams review was not merely flawed—it was the keystone in a larger edifice of manufactured certainty. Its removal exposes the extent to which corporate-funded science shaped public policy, not only through the literature itself, but through recursive citation, institutional repetition, and echo chambers in both academic and public platforms.

The U.S. Environmental Protection Agency is now at a crossroads. With the court’s 2026 deadline approaching, the retraction of its most relied-upon safety narrative, and mounting public distrust, it must decide whether to build its final glyphosate assessment on independent evidence—or cling to the remnants of a discredited artifact. The agency can no longer hide behind the Williams paper. The public, the courts, and the scientific record will not allow it.

---

Cited and linked sources:

• Williams, G.M., Kroes, R., & Munro, I.C. (2000). Safety Evaluation and Risk Assessment of the Herbicide Roundup and Its Active Ingredient, Glyphosate, for Humans. Regulatory Toxicology and Pharmacology. PMID: 10854122. (Retracted in 2025)

• Retraction notice: Regulatory Toxicology and Pharmacology, Vol. 146, 2025. https://www.sciencedirect.com/science/article/pii/S0273230024000619

• Kaurov, A.A., & Oreskes, N. (2025). The afterlife of a ghost-written paper: How corporate authorship shaped two decades of glyphosate safety discourse. Environmental Science & Policy. DOI: 10.1016/j.envsci.2025.104160

• Panzacchi, S. et al. (2025). Carcinogenic effects of long-term exposure from prenatal life to glyphosate and glyphosate-based herbicides in Sprague–Dawley rats. Environmental Health, 24, 36. DOI: 10.1186/s12940-025-01187-2

• Galli, C. et al. (2024). Overview of human health effects related to glyphosate exposure. Frontiers in Toxicology, 6, 1474792. DOI: 10.3389/ftox.2024.1474792

• Novotny, E. (2022). Glyphosate, Roundup and the Failures of Regulatory Assessment. Toxics, 10(6), 321. DOI: 10.3390/toxics10060321

• EPA Glyphosate Docket: EPA-HQ-OPP-2009-0361. https://www.regulations.gov/docket/EPA-HQ-OPP-2009-0361

• Ninth Circuit Court of Appeals, NRDC v. EPA, Case No. 20-70787 (2022 ruling vacating EPA’s interim decision on glyphosate)

• IARC Monograph Vol. 112 (2015). Glyphosate classification as probable human carcinogen https://monographs.iarc.who.int/wp-content/uploads/2018/06/mono112-10.pdf

• Civil Eats (Oct 2025). MAHA Action activists confront EPA over pesticide failures

https://civileats.com

• The New Lede (2025). Pesticide language erased from White House MAHA strategy

https://www.thenewlede.org

• Monsanto Papers (U.S. District Court for the Northern District of California, In Re: Roundup Products Liability Litigation) https://www.baumhedlundlaw.com/toxic-tort-law/monsanto-roundup-lawsuit/monsanto-secret-documents/