Human testing faces ethical scrutiny

The pesticide industry's desire to base safety standards on lab trials meets swift resistance from environmentalists and reignites a debate

Wednesday, December 8, 1999

By Brent Walth and Alex Pulaski of The Oregonian staff

On an Ottawa stage in October 1993, Monty Eberhart put his beliefs into words.

Pesticides should be tested not just on lab rats, he said, but on people, too.

By that time, Eberhart had spent his entire 13-year career as a toxicologist with Bayer Corp.'s agriculture division in Kansas City, Mo., and was the company's product safety manager. The company, known for aspirin, has sales of $2.3 billion each year through its crop-protection group.

Eberhart's job -- and his passion -- was making sure that Bayer products designed to kill bugs didn't also harm people.

He knew that some of the 150 scientists from around the world, who had gathered in the Canadian capital to talk about how to protect farm workers from pesticides, might not agree with him. But he was convinced that a rational debate among scientists would persuade the doubters. After all, he reasoned, how could researchers expect to accurately measure how pesticides act in the human body if all their lab test subjects had four legs and a tail?

He knew that some people had qualms, but he thought the concerns could be put to rest using the highest ethical standards.

"We must be prepared to argue," Eberhart said, "that human volunteer studies with pesticides are safe and humane and that they are required to protect the interests of society."

Three years later, Congress passed the Food Quality Protection Act to better protect U.S. children from harmful pesticide residues. The act endangered chemical companies' most potent products, called organophosphates.

One of them, azinphos methyl, better known by Bayer's brand name of Guthion, became the first pesticide examined under the law. The chemical is a cornerstone of the Northwest's apple industry, sprayed on 91 percent of the state's crop.

Bayer turned to human tests to prove Guthion's safety.

Environmental and consumer groups turned on the pesticide industry.

Eberhart was bewildered.

How could a rational tool for better research be held against scientists?

"I would have never envisioned the irony," he said.

"Serious risk concern"

In June, 1998, Guthion was in trouble.

The U.S. Environmental Protection Agency believed the insecticide, after 42 years on the market, posed a "serious risk concern," according to a 1997 agency document.

But the state of California posed an even bigger problem for the chemical's manufacturer.

California's Department of Pesticide Regulation is one of the most aggressive state agencies dealing with agricultural chemicals. It had just slapped severe restrictions on azinphos methyl, the primary chemical in Guthion, to protect farm workers. The restrictions meant many growers would abandon Guthion for other chemicals.

Eberhart, Bayer's chief of product safety, thought the state was overreacting.

He believed Guthion was not nearly as dangerous to people as California officials had concluded it was. Dozens of tests on lab animals had convinced him.

But California's decision gave Bayer an opening.

The state's report suggested that tests on humans could make Guthion look safer.

To Eberhart, it was just plain common sense.

"All kinds of chemicals are tested on people all of the time," Eberhart said. "You add the word 'pesticides' and it becomes more sensitive and more of an emotional issue."

A human test might also help persuade the EPA not to overly restrict Guthion's use.

Normally, the agency bases its pesticide standards on doses that have been found safe for lab rats. Then it reduces that amount by at least 100 times to account for the differences between animals and humans, and among individual people. The EPA calls this additional protection a "safety" or "uncertainty" factor.

Bayer and other pesticide companies say that human tests can eliminate the guesswork by showing that humans often react no differently to pesticides than do lab rats.

One other event propelled him: Faced with the tougher standards of the Food Quality Protection Act, farming groups and pesticide companies, including Bayer, had issued a report in mid-1998 that included a call for more human testing.

Eberhart felt strongly that human tests could be done ethically, particularly by spelling out the risks to volunteers and gaining their full consent. Inside Bayer, Eberhart pressed to launch just such a test.

"There is a value to the public and to society," Eberhart said. "If you're spraying pesticides or eating residues on your food, what would you rather have us relying on to keep you safe, rat data or human data?"

Unwitting subjects

In the past, however, human testing had left ethics behind.

Pesticide companies' widespread use of human trials didn't begin until the 1960s and '70s. In many cases, the subjects had no idea they were being exposed to pesticides for tests.

Between 1965 and 1967, Italian researchers wanted to see how dichlorvos, a nerve agent in the same class as Guthion, affected sick people. The pesticide, also known as DDVP, was used by

Shell in pest strips, once popular for killing flies and other bugs.

Researchers hung the pest strips in a Milan hospital over the beds of men with liver disease, women in labor or who had just given birth, and ill children, to see how the DDVP wafting off the strips would affect them. Many showed signs that the nerve agent was lowering their cholinesterase, the enzyme that helps regulate the nervous system.

In 1968, the researchers repeated the test, this time hanging the strips over the cribs of 22 newborns.

In different cases, tests were not calibrated carefully, and humans got too big a dose. A 1981 trial using male volunteers who agreed to swallow DDVP capsules had to be stopped when subjects suffered dramatic drops in their nerve-protecting enzymes.

The Oregonian's review of scores of clinical reports published since World War II found that the U.S. Public Health Service itself paid for four Guthion studies on humans between 1969 and 1972.

Researchers at the Franklin Hospital Foundation in San Francisco gave inmates Guthion pills. The lead researcher, Dr. J. Alfred Rider, said the inmates at the Vacaville state prison signed consent forms before the tests.

In a 1974 study, the state of California's Department of Public Health helped sponsor the prisoner test that influenced the state's 1998 decision on Guthion. Researchers from the University of California School of Medicine stuck pesticide-laced patches on inmates' arms to see how quickly Guthion soaked through the skin.

Later, the federal government banned the practice of using prisoners.

Environmental group reacts

When he heard the news, Ken Cook uttered an expletive.

Cook sat in the glass-walled conference room of the Environmental Working Group, with a veranda and a view of Dupont Circle and downtown Washington, D.C. As president of the organization, Cook had called a meeting to talk about the EPA's enforcement of the new pesticide law. Someone at the table mentioned human testing.

A researcher from another group, David Wallinga of the Natural Resources Defense Council, had discovered records in an EPA file about the Italian tests of DDVP on hospital patients.

Cook's own staff added that the pesticide industry was on record as calling for more human tests. He knew that pesticide companies could use the results of such tests to undermine the Food Quality Protection Act.

"We've got to make a stink out of this," Cook said.

Along with the Natural Resources Defense Council, the Environmental Working Group is one of the few advocacy organizations that regularly track pesticide issues. It had already called for bans or big cuts in organophosphate use.

With training in agriculture and soil sciences, Cook had led the Environmental Working Group since 1993. He had helped hone an effective media weapon: reports that mixed a bit of science with the group's political agenda, all aimed at getting attention.

"We develop our research with an eye to what an editor and reporter will think is worth writing about," Cook says. "We don't make any bones about that."

Since becoming the assistant EPA administrator overseeing pesticide programs, Lynn Goldman had greatly increased environmentalists' access. Cook called her to warn her of his strategy.

The Environmental Working Group planned to call the EPA to task for not having a policy on the use of human testing. He found her sympathetic.

Inside the EPA, Goldman had been fighting a battle over human testing under the new Food Quality Protection Act. She believed companies such as Bayer were planning to use the human tests to undermine the extra protection for children that the new law afforded.

Less than two weeks later, on July 27, 1998, the Environmental Working Group issued a report raising ethical questions about a rash of new pesticide trials in Britain. Playing off a popular movie, the group titled its report "The English Patients."

Tipped off that the report was coming, EPA officials had an official statement ready to go the same day: "EPA is deeply concerned that some pesticide manufacturers seem to be engaging in health-effects studies on human subjects as a way to avoid more protective results from animal tests under the new Food Quality Protection Act."

In Kansas City, Monty Eberhart read the Environmental Working Group report and EPA statement with growing concern and anger. He thought the report was filled with falsehoods and hype.

"I see this as an attack on the science I have spent my whole life working on," he said.

He was convinced the EPA would realize that the groups were attacking human testing only because such tests proved pesticides were safer than previously believed.

"I guess I was naive," he said.

Yet even as the EPA was raising new doubt about human testing, Bayer was forging ahead, looking for volunteers who would be willing to swallow pesticides for pay.

The Scottish trials

Robert Lonie needed some extra quid, and quickly.

He had a 1,200-pound bank debt coming due. His summer job at a local answering service wasn't cutting it. But then Lonie, a 23-year-old student at Heriot Watt University in Edinburgh, Scotland, jumped on another money-making idea. The answering service's biggest client was Inveresk Clinical Research Ltd., which paid people to try out medications.

The more he took calls for the clinic, the more he wondered, Why couldn't I do this? The clinic offered to pay 460 pounds, about $780.

Easy money.

Lonie signed up and cheerfully reported his new venture to his parents. His mother, a high school science teacher, studied Inveresk's description of the test. She noted that the trial chemical, azinphos methyl, was referred to as both a "drug" and a "pesticide."

She checked her chemistry textbooks and learned that the chemical, the key component of Bayer's Guthion, was an organophosphate. It was a Class I pesticide -- the most toxic.

"She put a stop to me doing it," Lonie said.

He later found out that a hay fever medication he took would have disqualified him anyway from Inveresk Study No. 013219, commissioned by Bayer.

Inveresk officials declined requests for comment. The disclosure documents Inveresk provided to Lonie and other volunteers described what the volunteers were told and how the study worked.

Inveresk even predicted the outcome of the tests: "The results of this study will further confirm that the use of azinphos methyl does not pose an unreasonable risk to either workers or consumers."

But Inveresk also disclosed to volunteers the potential problems of Guthion exposure: "vomiting, diarrhea, abdominal pain, blurred vision, weakness, constricted pupils, excess saliva, slow pulse and involuntary muscle twitching."

"It is highly unlikely that any of these effects will occur," the disclosure form added. The volunteers underwent physicals and blood tests, and their personal physicians were consulted; the clinic especially checked to see whether any female volunteers might be pregnant, excluding them if they were.

Inveresk selected 50 volunteers. The clinic required medical screenings and made sure that the 10 women who were to take part were not pregnant.

On Sept. 14, 1998, the first two volunteers, both men, checked into Inveresk and awoke the next day for breakfast. One ate a placebo; the other washed down a Guthion capsule with 5 ounces of water. The volunteers stayed for three days, and researchers drew their blood eight times to see whether their cholinesterase levels had changed.

Inveresk clinicians said they didn't see an effect.

One week latter, a second group of seven men checked in and took a higher dose.

Again, Inveresk saw no effect.

With the third, fourth and fifth groups, the doses went higher and higher and higher.

Finally, on Nov. 24, seven men were given the biggest dose yet -- 1 milligram of Guthion for every 2.2 pounds they weighed. The dose equaled the amount Bayer believed was the threshold reached in lab rats. If the subjects showed no ill effects, Bayer felt it would prove its point: Guthion affected people no differently than it did rats. The subjects swallowed the pills, and waited.

The money motive

At the same time, across the Atlantic, Lynn Goldman was holding her showdown over human testing.

On Dec. 10 and 11, a collection of American researchers and ethicists gathered at the Sheraton Crystal Hotel in Arlington, Va., just a few blocks from the EPA's Crystal City pesticide offices. The 16-member EPA advisory panel represented a century's worth of experience looking at children's health, the risks posed by chemicals and the ethics of using humans in lab tests. An EPA review had found that the agency accepted human tests at least 18 times in the past to help set pesticide standards, including a 1997 human study involving DDVP.

Some were questionable. One test using an insecticide dosed a single person. In another study in the early 1970s, Union Carbide executives ate aldicarb, an insecticide, to demonstrate their faith in the product's safety.

On the meeting's first day, Goldman set the tone for the debate when she argued that the use of human tests was driven by money. "The inevitable conclusion is that there is a very strong financial incentive to conduct these studies," Goldman said. "You could say that 10 times more of the residues would then be allowable in the diet and potentially 10 times more user sales."

Goldman said the EPA had never set any standards for judging the tests. How many subjects are adequate to prove a pesticide is safer than believed? And should a single test be enough to change standards that protect so many people?

The panel members listened to nearly 91,000 words of testimony from experts and debate among its members. It adjourned without a conclusion, promising a report in two months.

No adverse reactions

A few days later, Bayer had its answer from Scotland. No matter how high a dose the Inveresk subjects received, Guthion didn't affect them.

Bayer paired these results with another human trial the company conducted in the Netherlands, one intended to see how quickly the human skin absorbs Guthion. The Netherlands test confirmed results from the California test on prisoners 24 years earlier. Early this year, Bayer took the results to California. On the basis of the human tests, California officials decided to drop the extra safety factors to protect workers they had applied the previous growing season.

So now that California had accepted the test, what would the EPA say?

But everything hinged on the agency's report on human testing, due in February. Would the panel of scientists allow these tests to be considered? March, April and May passed, but still the answer didn't come.

In the meantime, the EPA was ignoring Bayer's Scottish test, just as a new threat to Guthion was emerging.

And the threat came from a computer.