Stalking Woburn’s mystery killer

For 25 years Woburn’s children have been dying of leukemia. Scientists at MIT are trying to find out why.

The following article appeared in the Winter 1989-’90 issue of MetroNorth Magazine. Copyright © 1989-1997 by Daniel D. Kennedy.

By Dan Kennedy

It was July 1988. William Thilly and a student were flying home from a meeting in Stockholm. Thilly, a research scientist at the Massachusetts Institute of Technology, and the student were talking about the difficulty of finding genetic mutations in human DNA molecules. DNA carries the chromosomes that determine whether we will have brown eyes or blue, black skin or white — or will be predisposed to develop cancer later in our lives.

The problem facing Thilly was that mutations occur naturally in only one out of every 1,000,000 cells. Mutations had to occur much more frequently than that — one in every 10,000 cells — for scientists to be able to find them. Thus, detecting mutations required huge, expensive incubators to grow vast quantities of living human cells and sophisticated techniques to concentrate the number of mutated cells.

Thilly mused: What if we changed the order of steps we take in the laboratory before analyzing the mutations? When he returned to Cambridge, he and Phouthone Keohavong, a fellow researcher, talked it over and gave it a try. It worked. Suddenly, Thilly had a technique by which he could detect mutations even if they occurred in only one out of every 100,000,000 cells. “There was no wild lab party,” he says. “It was just, ‘Goddamn, it works.’ What we’ve done is make a leap forward in the sensitivity. It’s a technical accomplishment which has yet to become an epidemiological accomplishment.”

The next step for Thilly and his MIT colleagues is to attempt to turn their discovery into an epidemiological accomplishment in the Aberjona River watershed — and especially in Woburn, where the rate of childhood leukemia is four times what would be expected for a community of its size. Twenty-eight cases have been reported since 1964, making it, according to the federal Centers for Disease Control, the most persistent leukemia cluster in the United States. Sixteen of those cases have ended in death.

The MIT study is a two-tiered project. A team of civil engineers, headed by Harold Hemond, a Lexington resident affiliated with MIT’s Parsons Laboratory, will try to determine exactly what chemicals the 50,000 residents of the area may have been exposed to through the air they breathe and the water they drink. Then Thilly, a Winchester resident who is director of MIT’s Center for Environmental Health Sciences, will examine blood samples from residents of the area to learn whether they have suffered mutations associated with the chemicals identified by Hemond.

If it works, says Richard Clapp, former director of the Massachusetts Cancer Registry and now a Boston University professor and private consultant, “I think it will be a major link in the proof that we need.”

The technique Thilly uses to find mutations was developed several years ago by scientists at the State University of New York at Albany. DNA molecules are placed on a substance known as a “gradient denaturing gel” and subjected to a high dose of electricity, which causes the molecules to migrate. The DNA is exposed to a chemical that separates the two strands — the “double helix” that high-school and college textbooks refer to — and stops the migration. Each sample of DNA molecules forms a band on the gel that looks remarkably similar to the spectrum emitted by a source of light bands or to the bar code on a magazine or a can of string beans. Dark bands in a genetic spectrum indicate mutations. Thilly hopes that the pattern of bands will tell him what chemicals caused the mutations, just as the spectrum from a distant star tells an astronomer what elements the star contains.

While this method represents an exciting new approach to studying health problems in humans, it is also somewhat frightening. What would a person do if he were told he was suffering from genetic mutations associated with exposure to, say, trichloroethylene (TCE) — the principal industrial solvent discovered in East Woburn’s drinking water 10 years ago? Thilly says scientists are a generation away from being able to prove that specific genetic mutations can cause specific illnesses, such as leukemia. But he acknowledges that the “weight of the evidence” would be that TCE can cause leukemia if he learns that members of families stricken by that disease show signs of genetic mutations caused by TCE. Even though Thilly believes no amount of mutation will cause disease in an individual who doesn’t already possess a predisposition to that disease, it would only be natural for a person who is told he has suffered from such mutations to consider himself a walking time bomb.

It is a concern that is never far from Thilly’s mind. During a late-afternoon seminar, he explains the project to a group of undergraduate students and says: “We may add to the anxiety of a community without being able to solve any of their problems.” But Gretchen Latowsky, who is program administrator of FACE (For a Cleaner Environment), the citizens group formed in response to Woburn’s toxic-waste problems, believes people have a right to be told if they have been affected by chemical exposure. “That would be very important information to have, although it would be pretty frightening. But if people had genetic changes, I would think they would want to know,” says Latowsky, a Reading resident.

The project is expensive — about $6.5 million a year, most of it federal money from the National Institute of Environmental Health Sciences. And unless the MIT scientists run into some sort of massive failure, it is likely the project will last for many years. “We’re going to do it for the rest of our lives,” Thilly says. “We’re either going to succeed or not succeed before we croak.”

EAST WOBURN, that portion of Woburn east of Main Street (Route 38), is a low, swampy area that has been home to many industries over the past 150 years. Numerous tanneries and chemical plants have dotted the landscape, bringing prosperity to a community that is in transition from a working-class town to an affluent suburb.

For decades Woburn’s water had been obtained from six wells (known simply as A through F) drilled into the groundwater aquifer surrounding Horn Pond, located in the south-central portion of the city. But in the 1950s, as water became increasingly scarce, city officials began considering drilling in groundwater-rich East Woburn.

Some officials warned that the water obtained from such wells would be of poor quality. But the city moved ahead and drilled well G in 1964, near the east bank of the Aberjona River. In 1968 the city drilled well H, about 500 feet north of well G.

Almost from the moment the new wells went on line, residents of East Woburn complained the water smelled and tasted bad. One neighborhood resident, Anne Anderson, went so far as to question whether the water might have caused the leukemia that her son, Jimmy, was suffering from. James Anderson, born July 16, 1968, was diagnosed in January 1972. By the time he died, on January 18, 1981, many observers believed Anne Anderson had been right. But during the 1970s most people dismissed her as a distraught mother groping for answers.

Repeated tests of wells G and H by local and state health officials showed the water was unpleasant but safe. Then, in 1976, a state health official nearly sutmbled on the truth. While testing an experimental instrument designed to detect extremely small quantities of organic solvent chemicals, he came across inexplicably high readings from wells G and H. But rather than explore the matter further, he assumed the readings were wrong, and used them to calibrate the device.

The truth was finally learned in May 1979. Officials discovered that someone had ditched some barrels several thousand feet north of the wells. The Massachusetts Department of Environmental Quality Engineering (now the Department of Environmental Protection) tested the wells to determine wether the water had been contaminated. They found that the barrels had not leached their contents into the wells — but that the well water contained several solvents used by industry, including TCE and tetrachloroethylene, also known as perchloroethylene (PCE).

The wells were shut down within days and have not been used as a source of drinking water since. The Metropolitan District Commission (now the Massachusetts Water Resources Authority) agreed to replace the lost water by connecting East Woburn to its regional distribution system.

The discovery of the contamination led to a number of studies, several of which are still going on, and to efforts to clean up Woburn’s polluted environment. The Massachusetts Department of Public Health concluded that the city’s rate of childhood leukemia (defined as leukemia diagnosed in people up to the age of 19) was four times higher than would be statistically expected in a community of its size. In the early 1980s the Harvard School of Public Health correlated leukemia cases with the distribution pattern of water from wells G and H to show that leukemia was most highly concentrated in those neighborhoods that had received most of their water from the wells. That study has been criticized, though, because the Harvard researchers — Marvin Zelen and Stephen Lagakos — used FACE volunteers to interview residents in order to save money. A number of scientists have charged that could have introduced bias into the study, although these critics have never been able to show that Zelen’s and Lagakos’s conclusions were wrong.

Largely because of the Love Canal and Woburn problems, Congress enacted the “Superfund” law in the waning days of the Carter administration. The law mandated the US Environmental Protection Agency to investigate sites where hazardous wastes may be located. The EPA soon found there were two sites in Woburn — a 100-acre property in North Woburn that had been home to several chemical plants and tanneries, and the groundwater aquifer that supplied wells G and H, in East Woburn.

Soil at the North Woburn property, known as the Industri-Plex 128, was contaminated with chromium, arsenic, and lead. Old hide piles emitted malodorous hydrogen-sulfide gas. Groundwater was contaminated with benzene, a known cancer-causing substance, and toluene. But the Industri-Plex has never been linked to any health problems. Currently the EPA and several businesses it determined to have liability for the site are getting ready to being a multimillion-dollar project to intercept and treat the groundwater, vent and filter the hide piles, and cover the contaminated soil with an impermeable layer of clay and high-tech plastic to prevent leaching.

East Woburn proved to be a much more difficult problem. Early on the EPA focused on three properties as the most likely sources of contamination to the aquifer:

• W.R. Grace & Company’s Cryovac manufacturing plant, at 369 Washington Street, northeast of the wells.
• UniFirst Corporation, 15 Olympia Avenue, an industrial dry-cleaning operation north of the wells.
• An undeveloped 15-acre property west of the wells owned by the John J. Riley Company tannery, at 228 Salem Street.

The Cryovac plant, which manufactured food-packaging equipment, used TCE to clean tools and thin paint. UniFirst used PCE, a common dry-cleaning chemical.

It has never been clear how chemicals ended up in groundwater at the tannery’s 15-acre property, although a lawsuit brought by eight Woburn families charged that the owners negligently allowed chemicals to be dumped there. Recently the EPA announced a plan to clean the groundwater in East Woburn and restore it to drinking-water quality; it is expected to take between 30 and 50 years and cost as much as $70 million, to be paid by the responsible businesses.

While the families affected by leukemia have been frustrated in their attempts to have government officials identify the cause and solve the problem, they have been equally frustrated by the legal system. In 1982 six families — later joined by two additional families — sued Grace, UniFirst, and Beatrice Foods Company, a Chicago-based conglomerate that had purchased the Riley tannery and the adjacent 15 acre-property in 1978. (Beatrice sold the properties back to Riley in 1983, but retained environmental liability. Riley sold the tannery to a group of longtime employees in 1985; the business closed its doors permanently on January 1, 1989. The Grace and UniFirst plants ceased operation several years ago.)

The families charged that the three companies polluted wells G and H, causing eight cases of leukemia — six of them fatal — and other illnesses, including cardiac arrhythmias and disorders of the liver, nervous system, and immune system.

The results of the lawsuit, though, have been inconclusive. In November 1985 UniFirst, which had been sued separately in Middlesex Superior Court, settled with the families for $1.05 million without admitting responsibility.

The Grace-Beatrice trial was divided into phases by US District Court Judge Walter Jay Skinner, who ruled that the families’ lawyers would have to persuade a jury that one or both of the defendants was responsible for polluting the wells before presenting evidence that the contaminated water caused leukemia and other diseases.

Following a 78-day trial, a jury found in July 1986 that Grace, which had admitted to dumping waste chemicals on company property, had negligently polluted the wells. But because of confusion concerning when the jury believed the wells had become contaminated, Skinner was compelled to throw out the verdict and order a new trial. Rather than repeat that ordeal, Grace and the families settled out of court for an estimated $8 million, with Grace continuing to insist publicly that chemicals dumped on its property had not flowed into the wells.

The jury dismissed the suit against Beatrice. But the case continues to wend its way through a tortuous appeals process. As recently as October 1989 new accusations were made that lawyers for Beatrice and Riley may have conspired to withhold documents from the families’ attorneys during the pre-trial discovery period. Thus, it is possible that the families will win a new trial against Beatrice. Such a ruling would likely lead to a settlement, since the families’ lawyers have exhausted their financial resources. [Note: The families’ appeals were ultimately rejected.]

JOHN DURANT pulls up to the pumphouse at Horn Pond in his rusty white Honda Civic. Several city workers are sitting around. One of them, pointing to the green waders Durant is wearing, calls out, “Are those L.L. Bean specials?” Durant laughs, and continues down a dirt road toward Fowle Brook, where he has a flow meter set up near well F.

Durant, a graduate student in civil engineering at MIT, walks under a foot bridge over the brook and plugs a Toshiba portable computer into a microprocessor inside a stainless-steel can. The microprocessor records the rate at which the water in Fowle Brook flows by; Durant downloads the data into his computer. The equipment is powered by a line running to the nearby wellhouse. But at two other sites, both along the Aberjona River, Sears Die Hard batteries are bolted to the walls of the bridges, and Durant has to change them once a month. Data are collected weekly. The meters were installed in May 1989, and Durant would like to get two years’ worth of data from them. He and other graduate students have also installed wells to measure groundwater, and are collecting data on precipitation as well.

It’s all part of the water-flow model Harry Hemond is putting together as principal investigator for the portion of the MIT project aimed at determining how water and chemicals move through the aquifer. Hemond’s goal is to measure all the water entering the watershed through precipitation, surface water, and groundwater; ideally, it should equal the amount of water measured entering the Upper Mystic Lake, in Winchester, at the mouth of the Aberjona River. Hemond calls it the “global water balance.” The Aberjona watershed takes in 23 square miles and all or part of seven communities: Woburn, Reading, Winchester, Wilmington, Arlington, Lexington, and Medford.

Hemond and his students are also looking for chemicals in water and sediments at the bottom of the river and the Upper and Lower Mystic Lakes. What they have found so far basically confirms what has previously been found by state and federal investigators: solvents and degreasers such as TCE and PCE, polynuclear aromatic hydrocarbons (PAHs), which are waste products produced by gasoline and fuel oil, and some heavy metals, including chromium (used in the tanning of leather), arsenic and lead (used in insecticides), mecury, and nickel. The discovery of chromium could be particularly important, since several dozen tanneries operated in the watershed from 1837 to 1988. The manner in which chemicals were deposited in sediments over the decades allows scientists to learn when the chemicals were being used, Hemond adds. Hemond has also built an instrument that can detect the presence of volatile chemicals, such as solvents and PAHs, to a depth of 30 feet.

Hemond and his colleague Bill Thilly are an interesting contrast. Thilly, blond-haired and clean-shaven, looks dapper in his oval-shaped glasses, suits, and suspenders; Hemond is bearded and dressed casually. The powerfully built Thilly was a collegiate wrestler, and he coaches young wrestlers in a recreation program in Winchester; Hemond is thin and wiry. Thilly enjoys talking about his work; Hemond seems slightly uncomfortable. But they share a commitment to helping people — a commitment that may be intensified in this instance by the fact that the people they are studying are fellow residents of the area. “I think the people who work in this field are driven by the desire to help their neighbors. It’s a social obligation that’s deeply felt,” says Thilly. Adds Hemond: “A lot of it is the motivation to do something that’s of some positive benefit to society.”

While Hemond continues to study the watershed to learn how it behaves and what chemicals are present, Thilly is testing his genetic-analysis technique on volunteers at MIT. If all goes well, Thilly will be ready to analyze the genetic spectra of about 100 volunteers in late 1990 or 1991. Hemond will tell Thilly what chemicals the residents may have been exposed to, so that Thilly can see whether mutations caused by specific chemicals under laboratory conditions show up in living subjects. Thilly will attempt to correlate that data with information concerning the kinds of foods the residents likely ate, since chemicals in food can cause mutations as well.

At the same time, two other MIT scientists, Klaus Biemann, an organic chemist, and Steven Tannenbaum, a toxicologist, will apply technology they have developed that can pinpoint chemicals in the bloodstream at the cellular level. That will provide an excellent reality check for Thilly’s work, since Thilly will be looking for genetic damage caused by the specific chemicals that Hemond has found in the environment and that Biemann and Tannenbaum have found in the blood.

What the MIT scientists may be closing in on is something very close to proof. For instance, let’s return to our consideration of TCE for a moment. Investigators already know that residents of East Woburn were exposed to TCE through their drinking water. Hemond’s work presumably will confirm that and explain how and when residents were exposed. If Biemann and Tannenbaum find TCE in the bloodstream, and if Thilly finds that residents have suffered genetic mutations in a pattern associated with TCE, then it would be difficult for anyone to argue with a statement that TCE was a cause of the city’s high childhood-leukemia rate.

This tool is potentially much moe powerful than the statistical correlations between TCE and leukemia demonstrated by the Harvard study. In fact, at a community meeting in Woburn a year and a half ago at which Thilly outlined his goals, Stephen Lagakos, one of the Harvard investigators, told him, “If you can accomplish what you’ve set out to accomplish, you’ll have advanced the field a century or so.” Thilly and Hemond know that if their work is successful, they could get dragged into court — especially if the Woburn families are awarded a new trial against Beatrice Foods. If that happens, Thilly expects to be vigorously contested by scientists retained by Beatrice who disagree with his findings. As he once quipped to a group of Woburn citizens: “For every PhD, there is an equal and opposite anti-PhD.”

Once Thilly begins testing residents, he says the “simplest possible outcome” would be to find that residents of a particular neighborhood or family have similar genetic spectra associated with exposure to chemcals that Hemond has identified. But he knows it is likely to be more complicated than that. For instance, in experiments in which human cells are exposed to ultraviolet radiation, the spectra produced by different groups of cells are nearly identical — but some contain dark bands that others don’t. “The signficance is that they shouldn’t have occurred at all,” Thilly says, adding he is developing possible explanations for this.

“Sometimes a project reaches a dead end,” he adds candidly. If, for instance, he finds that “virtually everybody has a spectrum that’s unique to themselves,” it would be unlikely that analyzing genetic spectra would be useful in investigations of environmental contaminants and their effect on health. Thilly says that’s one of the risks of science: “I could find that it doesn’t work. That’s research. The potential drives us forward. But we’re not stupid — we know reality could stop us at any point.”

So far, though, everything has gone his way. The technology is in place, the early experiments are promising, and tests on local residents should be ready to proceed in a year or so.

William Thilly acts like a man who believes reality is on his side.