US Government Scientists Find Glyphosate Herbicides Linked to Blood Cancers in Major NIEHS Review

Posted on Nov 21 2026 - 11:50pm by Sustainable Pulse

A systematic review by the Report on Carcinogens group at the US National Institute of Environmental Health Sciences, carried out at the request of the Department of Health and Human Services, concludes that exposure to glyphosate-based herbicides is positively associated with lymphohematopoietic cancers, including non-Hodgkin lymphoma and leukemia.

Scientists at the US National Institute of Environmental Health Sciences (NIEHS) have found that the human evidence points to a positive association between glyphosate-based herbicides (GBHs) and cancers of the blood and lymphatic system, in what is the most detailed US government assessment of the epidemiology to date.

The review, released as a preprint dated 20 November 2026 and not yet peer-reviewed, was conducted by the Report on Carcinogens (RoC) Group in NIEHS’s Division of Translational Toxicology. HHS asked the group to evaluate the cancer hazard evidence for glyphosate using systematic review methods, citing the “potentially large public health impact” and the divergent conclusions reached by regulators and scientific bodies around the world.

“The current body of epidemiology studies found a positive association between GBH exposure and LHCs,” the authors, led by Sabah M. Quraishi and corresponding author Ruth M. Lunn, write in their abstract.

Every study showed a positive association

The NIEHS team screened more than 800 papers before narrowing the evidence base to eight studies, including the US Agricultural Health Study (AHS), the international AGRICOH cohort consortium, and the InterLymph, North American Pooling Project (NAPP) and Swedish pooled case-control analyses. Together these cover 24 independent study populations reporting on non-Hodgkin lymphoma (NHL), 20 on multiple myeloma, seven on Hodgkin lymphoma, and two on leukemia.

The central finding is stark: every study included had at least one positive association between GBH exposure and a lymphohematopoietic cancer, mostly NHL or an NHL subtype.

The reviewers rated the overall evidence as “some evidence” of an association for:

  • non-Hodgkin lymphoma overall
  • B-cell lymphomas
  • diffuse large B-cell lymphoma (DLBCL), the most common NHL subtype
  • follicular lymphoma

The evidence for multiple myeloma, Hodgkin lymphoma, CLL/SLL and T-cell lymphoma was judged inconclusive.

The strongest single result for NHL came from the Swedish pooled case-control study, which found a more than doubled risk among those exposed (OR 2.2, 95% CI 1.3–3.8), with risk highest among people with the longest time since first exposure. In the AGRICOH cohort consortium, which pools farmers in the US, France and Norway, ever-exposure was linked to a 36% higher risk of DLBCL (meta-HR 1.36, 95% CI 1.00–1.85). In the NAPP study, people with the highest frequency of use had more than twice the risk of DLBCL (HR 2.14, 95% CI 1.07–4.28).

The review also looks closely at InterLymph, the largest pooled analysis, whose headline result has often been cited as evidence of no effect. The NIEHS team found that a single site, the Mayo Clinic study, “substantially attenuated” the pooled estimate. With that study removed, the NHL odds ratio rose from 1.03 to 1.41, and the authors note this could have pulled the overall result toward the null.

A new signal for leukemia, including in children

One of the most significant parts of the review concerns leukemia, which has had far less attention than NHL in the long-running glyphosate debate.

In the Agricultural Health Study, which followed more than 54,000 pesticide applicators in Iowa and North Carolina, the risk of acute myeloid leukemia (AML) rose with increasing lifetime intensity of glyphosate use. The pattern held across several exposure lag periods, which the reviewers said increased their confidence in the result. They rated the study moderate-to-high informativeness and said any bias was unlikely to explain the finding.

A California study of children found that prenatal exposure, measured as glyphosate applied within 400 metres of the family home at birth, was associated with roughly three times the risk of childhood AML and around 2.2 times the risk of acute lymphoblastic leukemia (ALL). The NIEHS reviewers concluded that bias was “unlikely to account for the entirety of the observed results.”

June 2025: International Study Reveals Glyphosate Weed Killers Cause Multiple Types of Cancer

The authors point to supporting biological evidence. Studies nested within the AHS have linked glyphosate exposure to mosaic loss of the Y chromosome, a condition associated with clonal hematopoiesis and leukemia development, and to biomarkers of oxidative stress. In animal studies where exposure began before birth, glyphosate treatment was linked to unusually early deaths from leukemia in rats, a reference to the Ramazzini Institute’s Global Glyphosate Study.

Only two studies examined leukemia, and the authors call for more research, including on myelodysplastic syndromes, a condition that can progress to AML.

Bias does not explain the findings

For years, critics of the epidemiology have argued that positive findings can be explained by recall bias in case-control studies or by farmers’ exposure to other pesticides. The NIEHS review tested these arguments directly, including with quantitative bias analyses.

Its conclusion is that bias did not reasonably explain the positive findings. Most of the potential biases identified would push results toward the null, meaning they would hide a real effect rather than create a false one. Recall bias was expected to be minimal because the studies were mostly carried out before glyphosate’s cancer risks became widely known, and because positive associations were not seen across the board for other pesticides in the same populations. Where studies had not adjusted for other pesticides, the bias-adjusted risk estimates stayed elevated.

The review was checked by subject-matter experts from NIEHS, the National Cancer Institute and the National Institute for Occupational Safety and Health who had no role in the evaluation.

Animal evidence points the same way

The human review is one of three linked NIEHS papers. A parallel review of experimental animal studies (Urich et al., submitted) found treatment-related tumour responses at multiple sites and suggested that glyphosate targets the lymphohematopoietic system in rodents. A third paper (Blake et al., submitted) examines how the two bodies of evidence fit together.

According to the authors, the human and animal evidence together “increases the confidence in an association between GBH exposure and LHC.”

Glyphosate Box

Glyphosate Residue Free Certification for Food Brands – Click Here

Test Your Food and Water at Home for Glyphosate – Click Here

Research gaps: the most exposed are missing

The review highlights how little is known about the people with the most exposure. The authors found no studies of cancer in glyphosate production workers, the most highly exposed group. Every study was carried out in Western, high-income countries, and several North American and European case-control studies were limited to white men. Hispanic farmworkers and landscape workers in the US, who are likely to be heavily exposed, are largely absent from the data.

The authors also note that most of the studies predate the surge in glyphosate use from the late 1990s. Today, exposure is close to universal: in recent US national biomonitoring (NHANES 2013–2018), more than 70% of participants had glyphosate in their urine, mostly through diet. As a result, simply comparing “ever” and “never” users is becoming less meaningful, and the authors call for studies that measure cumulative exposure from all sources.

They also recommend that leukemia, childhood exposure windows and women be studied more closely, and that other cancers, including breast, liver and kidney, be assessed in future reviews.

A widening divide with regulators

The review adds to a long-running split between independent scientists and pesticide regulators. In 2015 the World Health Organization’s International Agency for Research on Cancer (IARC) classified glyphosate as “probably carcinogenic to humans” (Group 2A), citing human evidence for NHL. The US Environmental Protection Agency (EPA) concluded in 2020 that glyphosate is “not likely carcinogenic to humans,” and European regulators, including ECHA and EFSA, have repeatedly found that the evidence does not meet EU criteria for a cancer classification.

The timing matters. In September 2026, HHS and EPA signed a memorandum of understanding bringing HHS scientific agencies, including NIEHS, the National Toxicology Program, NCI and NIOSH, into EPA’s ongoing glyphosate registration review. EPA has said its updated human health risk assessment is expected around the end of 2026. Public comments on EPA’s glyphosate human health open literature search close on 23 November 2026 (docket EPA-HQ-OPP-2009-0361).

With the government’s own cancer-hazard scientists now finding a positive association between glyphosate herbicides and blood cancers, including leukemia in children exposed before birth, pressure on EPA to reconsider its “not likely carcinogenic” conclusion is set to grow.


Study: Quraishi SM, Arroyave WD, Blake BE, Haque E, Urich ML, Mehta SS, Lunn RM. Evaluating Evidence of Glyphosate-Based Herbicides and Lymphohematopoietic Cancers from Epidemiologic Studies: A Systematic Review. Preprint, not yet peer-reviewed, 20 November 2026. National Institute of Environmental Health Sciences.

1 Star2 Stars3 Stars4 Stars5 Stars
(1 votes, average: 5.00 out of 5)
Loading...
About the Author

Sustainable Pulse is a global news outlet covering sustainable agriculture, GMOs and pesticides.

Leave A Response

You must be logged in to post a comment.