EPA Will Soon Weigh In on Weed Killer That May Cause Parkinson’s Disease

The U.S. Environmental Protection Agency (EPA) is expected to publish by Friday a document reassessing its position on the risks of paraquat, an herbicide suspected of causing Parkinson’s disease. Use of the weed killer has recently soared in the United States, but it is banned in more than 70 other countries.

UPDATE: SEE BELOW

The document would inform the agency’s final decision on whether to allow continued use of a cheap, broad-spectrum weed killer, which is extremely toxic if ingested or inhaled. 

Scientists and patient advocacy groups say a mass of evidence from lab dishes, animals, and people—including a rigorous epidemiological study published last year—suggests exposure to paraquat very likely causes Parkinson’s, the fastest growing neurodegenerative disorder.

At stake in the agency’s ultimate decision is “whether thousands of people will be unknowingly and needlessly exposed to a neurotoxic weed killer that almost certainly causes Parkinson’s disease,” says Tim Greenamyre, a neuroscientist and Parkinson’s researcher at the University of Pittsburgh. Greenamyre and other scientists filed an amicus brief supporting plaintiffs who sued EPA in 2021, when it provisionally approved paraquat for another 15 years.

Paraquat’s makers—led by agrobusiness giant Syngenta—say it is safe when used properly. “Despite decades of investigation and more than 1200 epidemiological and laboratory studies of paraquat, no scientist or doctor has ever concluded in a peer-reviewed scientific analysis that paraquat causes Parkinson’s disease,” a Syngenta spokesperson says.

The company contends that greedy plaintiffs’ lawyers are falsely representing the dangers. And it points to a new, preliminary report published last month by the California Department of Pesticide Regulation that examined the most recent populations studies and concluded the evidence doesn’t support a link to the disease.

The compound, which has been marketed since 1961, kills weeds on contact by inhibiting photosynthesis. EPA describes it as a “burn-down herbicide,” used to clear fields before planting. A 2020 assessment by the agency found U.S. use of paraquat rose from 1.9 million hectares in 2006 to 7.5 million in 2017, a surge it attributed to weeds’ increasing resistance to glyphosate, another widely used herbicide used in products including Roundup. The greatest volume of paraquat is used before planting soybeans, cotton, and corn. It’s also used on peanuts, wine grapes, and table grapes. Syngenta argues the weed killer supports sustainable agriculture because it wipes out weeds without disturbing the underlying soil and doesn’t leach into groundwater.

But neuroscientists who study paraquat say the compound that is lethal to weeds is also highly toxic to other living things. “We know from animal work—and this is convincing and consistent—that paraquat isn’t safe,” says Bas Bloem, a neurologist at Radboud University Medical Center. The compound can pass from the bloodstream to the brain, he notes, and kills dopaminergic neurons, the loss of which drives Parkinson’s; indeed, paraquat is used to create laboratory animals with Parkinson-like disease. It also enhances the buildup of a misfolded, toxic version of the protein alpha-synuclein that’s a hallmark of the disease.

Safety concerns prompted the European Union to ban paraquat in 2007, and China and Brazil followed suit in 2017 and 2020. Although Canada nominally allows its use, Syngenta, the only marketer there, pulled its product in 2023. Earlier this week, Australia delayed a reregistration decision to the end of this year. New Zealand and Japan continue to allow its use.

By law, EPA must review and reapprove a pesticide every 15 years to ensure it meets safety standards and bring use requirements up to date with the latest literature. A 2006 decision on paraquat allowed its use by certified applicators with safety measures; in 2021, EPA provisionally reapproved the weed killer with additional restrictions such as requiring buffer zones around areas of aerial application to prevent drift. 

The agency also looked at concerns that paraquat causes Parkinson’s and, based on a literature review it completed in 2019, concluded the evidence was weak. The literature review noted, for example, that just one study in mice linked Parkinson-like behavior to pathological brain changes consistent with Parkinson’s.

Two months later, the Michael J. Fox Foundation for Parkinson’s Research, along with farmworker groups and others, sued EPA, asking a federal appeals court in California to block the agency from moving forward with the reregistration. EPA asked the court to pause the lawsuit while it reconsidered its decision, including its balancing of the weed killer’s risks and harms. But in January 2024, the agency published a document that reached the same conclusion as the 2021 decision—although EPA acknowledged it hadn’t yet reviewed 90 new studies that the Fox foundation had submitted.

“I’m not holding my breath” that EPA regulators will “follow the weight of the science,” says Kimberly Paul, an epidemiologist at the University of California, Los Angeles, because, she says, the agency scrutinizes one study at a time for weaknesses, rather than looking at the global weight of the evidence.

Paul was first author on a paper published in February 2024 that added new grist to advocates’ arguments against paraquat. She and her colleagues used a California pesticide application registry to trace residential and workplace paraquat exposures stretching back to 1974 among more than 800 people with physician-confirmed Parkinson’s and a similar number of matched controls. Paul says the team structured the study to address weaknesses EPA had pointed out in their preceding epidemiological work—including by documenting duration and time of exposure.

They found that the Parkinson’s patients both lived and worked near agricultural facilities applying greater amounts of the herbicide than controls, and people working within 500 meters of such facilities had up to twice the risk of developing the disease. (Two of Paul’s co-authors are serving as paid expert consultants in lawsuits against Syngenta. More than 5000 people claiming paraquat caused their Parkinson’s are suing Syngenta in an Illinois court.)

Those results have shifted the burden onto industry to prove paraquat is safe, Bloem and colleagues argued in an August 2024 commentary.

In defense of the pesticide’s safety, Syngenta and others have invoked a 2020 study that analyzed data from more than 38,000 pesticide applicators and nearly 28,000 of their spouses. It did not find an elevated risk of developing Parkinson’s, except in a subset of people with preexisting head injuries, who were at more than three times the risk of developing the disease. But Paul notes that patients self-reported their disease, and the study did not distinguish between different amounts of exposure.

Her own paper, meanwhile, has also come under attack. A review of recent studies published in October 2024 lambasted the work, arguing it didn’t truly measure paraquat exposure. (The review’s author, Douglas Weed, has long worked as a paid consultant to industry.)

The California appeals court case can resume once EPA publishes its latest thinking, notes Jonathan Kalmuss-Katz of Earthjustice, lead attorney for the Fox foundation and other plaintiffs in that case. If EPA fails to “protect the public from paraquat’s unreasonable risks,” he says, “we will continue to fight in court for the protections that the law requires.”

UPDATE: On 17 January, EPA asked a U.S. appeals court’s permission to withdraw its 2021 interim decision so it can undertake a new analysis of the risks of paraquat volatilization, which it said would take at least four years to complete. The plaintiffs in the lawsuit challenging the 2021 decision vowed to contest EPA’s move, which would end that lawsuit if the court permits it.