Toms River (68 page)

“A good working definition of a public health catastrophe,” a well-known Boston University environmental epidemiologist named David Ozonoff likes to say, “is a health effect so large even an epidemiological study can detect it.” By that arch description, the Toms River childhood cancer cluster was a manmade catastrophe, and a preventable one.

An ordinary town in so many ways, Toms River owed its unwanted status to an extraordinary combination of human folly and perseverance, along with a great deal of luck—both bad and good. The confirmation of the cluster and the identification of its likely causes were the highly improbable results of a decades-long process in which things went terribly wrong and then spectacularly right. First came the unprotected wells, inattentive officials, supercharged growth, and obscure poisons discharged by the ton into a shallow river and sandy soil. Then came people like Lisa Boornazian, Linda Gillick, Floyd Genicola, and Jerry Fagliano, all of whom went far beyond what anyone could have expected in forcing an investigation and driving it forward. In order for Toms River to be identified as a “true” nonrandom cluster, there had to be an extreme number of cases—enough to be noticeable to the people who lived there and also enough to clear
the very high bar of statistical significance. For the cluster to then be linked credibly to local pollution, the environmental damage had to be severe and the epidemiology sophisticated and extremely expensive—and therefore backed by extraordinary political support.

It was hard to imagine a similar set of circumstances occurring elsewhere, and even harder to imagine that a government agency would ever again willingly embark on an investigative process that in Toms River took almost six years, cost well over $10 million, and embarrassed a boatload of public officials on the way to its deeply unsettling conclusion. In that sense, lawyer Jan Schlichtmann’s bold prediction that the Toms River epidemiological studies and legal settlement would “impact public health and environmental policy-making for a very, very long time” has turned out to be correct, but not in the way he expected or wanted. The chief legacy of Toms River instead has been to solidify governmental opposition to conducting any more Toms River–style investigations. And because only governments have unimpeded access to cancer registry information (due to privacy concerns), if public agencies do not investigate clusters, then no one will.

A few health agencies—including even the biggest cluster-buster of all, the Centers for Disease Control and Prevention—have reluctantly undertaken residential cancer cluster studies in recent years, though there have been only a handful of comprehensive investigations. Almost invariably, their studies have followed the Toms River/Woburn template: They were launched only after intense pressure from Linda Gillick–like civic activists and from politicians who rushed in amid widespread media coverage. At least three of those investigations, all in the United States, ended with a disturbing conclusion: The cluster was almost certainly
not
a chance occurrence. The most famous was in the Nevada desert town of Fallon and surrounding Churchill County, where sixteen children were diagnosed with leukemia between 1997 and 2002—more than eight times the expected number.
¹
(By one credible estimate, the odds that the Fallon cluster could have occurred randomly were just one in 232 million. Stated another way, a cluster that extreme would occur by chance in the United States only once every twenty-two thousand years.)
²
A second was in Sierra Vista, Arizona, where at least eleven children were diagnosed
with leukemia between 1997 and 2003 instead of the expected five.
³
The third was a brain cancer cluster among children in a South Florida area called The Acreage, where four cases were diagnosed between 2005 and 2007 when just one case was expected.
⁴

In all three communities, researchers failed to identify a likely cause, even after conducting case-control studies that included testing local air, water, and soil. In fact,
all
of the residential cancer cluster investigations undertaken since the completion of the Toms River studies in 2001 have failed to identify a likely cause. There are still just two lonely names on the roster of non-occupational cancer clusters that have been scientifically associated with specific chemical exposures: Woburn and Toms River.

Why haven’t there been more? There are at least three possible reasons. The Toms River cluster may have been a statistical aberration, a random one-in-a-million event that actually had nothing to do with local pollution, despite the findings of Jerry Fagliano’s two case-control studies. Or perhaps Toms River really was a nonrandom cluster triggered by toxic exposures, but a vanishingly rare one because industrial chemicals may be a trivial cause of cancer compared to more powerful risk factors like tobacco and family history, as the late Richard Doll and some other prominent epidemiologists have long argued. But there is also a third possibility, the one Boston University’s David Ozonoff was hinting at with his wry description of the feebleness of epidemiological studies: Clusters of rare cancers like the one in Toms River may actually be much more common than we can discern with the crude statistical tools of small-number epidemiology. In other words, many more pollution-induced cancer clusters may be out there, but we don’t see them and we rarely even bother to look.

Public health agencies are shunning cluster studies at a time when researchers need all the help they can get in understanding why and how cancer begins. The most important cancer breakthroughs in recent years have been about treatment, not causation—reflecting the priorities of research funders in government and the pharmaceutical industry. (One of the most successful new cancer drugs, Gleevec, introduced in 2001 by Ciba’s successor company, Novartis, has helped to raise the
five-year survival rate for the most common type of pediatric leukemia to 90 percent.)
⁵
For scientists who are still trying to understand carcinogenesis, meanwhile, there has been a major shift since the 2001 completion of the Toms River studies. Moving beyond DNA mutations, they are now studying all of the steps by which genetic instructions are executed within cells, including protein synthesis regulated by DNA’s single-stranded cousin, RNA. Researchers are also looking at a galaxy of epigenetic changes—modifications that alter the way genes function without changing the underlying genetic sequence—in hopes of finding clues about why cells turn malignant.

This new complexity has made it much more difficult for molecular epidemiologists like Frederica Perera and Barry Finette to identify specific changes—biomarkers—within human cells that are reliable indicators of cancer risk across populations. The list of candidate biomarkers keeps growing. Some appear to alter susceptibility to particular cancers, conferring extra vulnerability or extra protection. Others are clinical indicators that hold out the promise of aiding the early diagnosis and treatment of tumors. A few can even serve as “dosimeters”—rough indicators of the extent to which someone has been exposed to a particular toxic compound.
⁶
What almost all of these biomarkers have in common, however, is that they are helpful only some of the time. In the Fallon, Nevada, cluster investigation, for example, researchers found that all eleven local children with leukemia carried a distinctive variation in a gene called SUOX—but 40 percent of cancer-free children did too.
⁷
An ideal biomarker for childhood leukemia would be so sensitive that every child with the biomarker would have leukemia, yet so specific that every child with leukemia would have the biomarker. After decades of intensive searching, however, it now seems likely that ideal biomarkers may never be found for leukemia and many other cancers because there are just too many possible routes to carcinogenesis. Even now, more than 160 years after Rudolf Virchow first saw malignant white blood cells through his microscope, the biochemical pathways to cancer are still mostly unmapped.

The research that Barry Finette undertook for the Toms River families is a telling example. Working intermittently for more than
ten years, whenever he had enough staff and money available to pursue it, Finette and a rotating cast of junior colleagues kept reanalyzing the Toms River blood samples that still filled two freezer shelves in his Vermont laboratory. Searching, as ever, for biomarkers that correlated to chemical exposure and cancer risk, they looked again at the HPRT gene—this time tracking the types and locations of mutations within the gene’s region of the X chromosome. They also tracked alterations in eighteen other genes involved in repairing damaged DNA or expelling toxic chemicals that invade cells. Finally, they scrutinized the children’s blood cells for biomarkers that were much larger than individual snips of DNA—so large, in fact, that they could be easily seen through an optical microscope. They were chunks of malformed chromosomes, floating free in the nucleus or reattached in the wrong place or even upside down.

In the end, none of it worked. Finette’s biomarker studies revealed no significant differences between the DNA of Toms River children and that of children who lived elsewhere. He struggled to come up with reasons for the multiple failures. Perhaps, Finette thought, he had searched for the wrong biomarkers. Or maybe he found nothing because the chemicals in the air and water of Toms River were too insignificant to have a measurable impact on the DNA of the town’s children, even though the state’s case-control studies suggested otherwise. Or possibly those genetic abnormalities had once been present in the children’s blood cells, but were gone by the time he collected his samples in 2000, many years after local pollution was at its worst. It was even possible that the DNA defects were still there but were subtle and thus too difficult to differentiate from random variation within such a small study population. Finette had samples from just forty-three Toms River children with cancer, plus their healthy siblings, which meant that his study was only slightly larger than Jerry Fagliano’s interview study and smaller than Fagliano’s birth record study—both of which had been plagued by small case numbers.

Whatever the reason, Finette’s ten-year search for evidence that chemical exposures in Toms River had altered the genes of the town’s children and raised their cancer risk had come up empty.
⁸
Even the chromosome malformation study, the one Finette had been the most
optimistic about, failed to turn up any associations.
⁹
In a staggering feat of ocular stamina that ended only when the grant money for her salary ran out, a lab technician named Heather Galick had squinted into her microscope and inspected exactly 201,844 white blood cells from eighty-one children. Because each cell was on the verge of dividing and had been stained, Galick could see the chromosomes inside. She counted 466 cells—roughly one out of every four hundred she examined—that had aberrant chromosomes: translocations, loose fragments, or other signs of damage. That seemed like a lot of malformations to Finette, but the ratio of aberrant-to-healthy chromosomes in the Toms River children turned out to be almost identical to the ratio in nonresidents. In fact, the out-of-towners were slightly
more
likely to have malformed chromosomes, a perplexing result Finette could not explain. After a decade of work, there was very little about any of his results that he could explain.

Linda Gillick and a few other Toms River parents got the disappointing news from Finette on a stifling July afternoon in 2010, in the same Ocean of Love office where their needle-wary children had lined up to give blood samples ten years earlier. Back in 2000, dozens of families were involved; by 2010, just three were represented around the conference table. Besides Gillick, there were Bruce and Melanie Anderson (they lived in Pennsylvania now but had made the long drive) and their former neighbor Joseph Kotran, whose daughter Lauren had survived neuroblastoma as a toddler. Most of the other families had not been in touch since the legal settlement nine years earlier. “They have other things in their lives now,” Bruce Anderson would later explain with a shrug.

There was one other big study the Toms River families were waiting for, and many of them cared a great deal about it. Styrene acrylonitrile trimer, the obscure plastics byproduct that Floyd Genicola had discovered in the Parkway wells back in 1996, had acquired outsized importance in the Toms River drama because it was so exotic and apparently unique to the town (it had not been identified as a pollutant anywhere else). To some of the families, SAN trimer felt like an answer—maybe even
the
answer—to their most burning questions
about the cancer cluster: Why here? Why us? Even so, there were reasons to doubt SAN trimer’s true importance: The concentrations in Parkway water were low, the compound’s health risks were unknown, and there were many other toxic exposures in town, especially back in the 1960s and 1970s. Still, “the trimer,” as they called it, retained its almost mystical hold on the families as time passed. “Really, in our hearts, we all feel it’s the cause,” explained Kim Pascarella.

The answers on SAN trimer were supposed to come from the federal government’s National Toxicology Program, but they were painfully slow to emerge. The NTP’s first attempt at its two-year, multigenerational rat study flopped because too few rats were willing to breed, but a second try, in 2005, succeeded. After two years eating trimer-laced chow, the surviving pups were asphyxiated. Pathologists pulled thirty-two organs out of each rat, dunked the organs in formaldehyde, embedded them in wax, and cut them into more than ten thousand slides, each just five microns wide—thin enough to reveal a single layer of cells under a microscope. The slides were then scrutinized for tumors by panels of experts convened by the NTP, in a laborious process that took an additional three years and was quite contentious at times. The debates over the most ambiguous slides—the ones in which it was very hard to tell whether a dot was a tumor or a noncancerous lesion—at times sounded like a roomful of highly opinionated art critics arguing about the meaning of an abstract painting.