by Dan Fagin
Knudson developed a theory that could explain all of retinoblastoma’s quirks. He called it the “two hit” hypothesis, and it was based on the idea that a retinoblast cell required two mutations to become malignant. Children with hereditary retinoblastoma, Knudson believed, were born with a mutation they carried in all of their cells and thus required just one more mutation “hit” in any retinoblast cell to develop a tumor. With about one hundred million cells per eye, there were many opportunities for that second mutation. In fact, it could easily occur in several cells in one or both eyes, triggering a separate tumor each time. The timing was interesting, too, because cell divisions, opportunities for mutation, occurred most often during the last months of fetal growth and the first few months after birth, when the eyes were developing most quickly.
The two-hit hypothesis thus explained why children with hereditary retinoblastoma tended to be younger and more likely to have multiple tumors. Children with sporadic retinoblastoma, on the other hand, got the disease only after an individual retinoblast cell suffered two mutational events—akin to lightning striking twice in the same place. Knudson’s theory even explained why hereditary cases sometimes skipped generations, since the parent in the cancer-free middle generation carried the inherited mutation but luckily avoided the second hit. “It all fit together,” Knudson recalled. “With these retinoblastoma kids, it just seemed obvious that inheriting this gene isn’t enough to make the cancer. Something else has to happen. There has to be a second hit.”
By the time he was ready to test his theory, in 1970, Knudson was working in Texas and had access to the medical records of forty-eight children treated for retinoblastoma. Twenty-three of them had tumors in both eyes, which meant that they were almost certainly hereditary cases. The next step was to set up a statistical test: If two-eyed cases required just one mutation while one-eyed cases needed two mutations, how many of each kind would be expected in a typical group of forty-eight children with retinoblastoma, and how old would each child be at diagnosis? Knudson calculated an expected distribution and then looked to see if they matched the actual results from the forty-eight cases. He was thrilled to discover that he got a hand-in-glove fit for all of his most important predictions. He had predicted, for example, that the average number of tumors among the two-eyed cases would be three—a prediction that fit the hospital data almost perfectly.13 Knudson was right: Retinoblastoma was a two-hit cancer—the first multistep cancer with a fully plotted pathway to malignancy.
It is almost impossible to overstate the impact Knudson’s discovery had on research into carcinogenesis, including the role of pollutants in places like Toms River. He would move on to Fox Chase Cancer Center in Philadelphia in 1976, continuing his cancer work and happily watching researchers around the world expand his findings in sometimes surprising ways.14 The biggest surprise came in 1983, when the inherited “retinoblastoma gene” in the human genome was finally identified. The gene was on Chromosome 13, which is where Knudson had predicted it would be found back in 1976. The surprise was that it turned out not to be a villainous oncogene like the leukemia mutation created by the Philadelphia translocation. Instead, the newly named “Rb” gene was a hero—a new type of gene whose existence had yet again been foretold by Boveri seventy years earlier. It was a tumor-suppressor gene, and it was ultimately found to protect against not only retinoblastoma but also bone cancer and other malignancies. The problem for children with hereditary retinoblastoma was that their cells contained only one functional copy of the Rb gene instead of the normal complement of two, one from each parent. If a second mutation knocked out the remaining copy, then the affected cell would lose its ability to regulate its rate of replication. It would become a cancer cell.
The multi-hit model Knudson laid out for retinoblastoma became the dominant paradigm for carcinogenesis, the default assumption of how cancer begins. A few cancers required just one “hit,” but most were now assumed to be the products of complex sequential processes that began with inherited mutations and continued with additional genetic “hits.” If those subsequent mutations occurred at critical times, especially during fetal development or early childhood when so many cells were dividing, the results could be calamitous. That lesson would not be lost on the families of Toms River.
There was another, even more powerful message for Toms River in Knudson’s work: Most of the chemicals that loomed large in the town’s unhappy environmental history were mutagens—they were capable of altering DNA. As he developed his two-hit theory, Knudson did not speculate on the possible causes of the non-inherited mutations that were essential steps on the road to carcinogenesis. His mathematical model for retinoblastoma assumed that these mutations occurred randomly—due to collisions with cosmic rays, for example, or because of DNA copying errors during cell division. But Knudson and everyone else in the field knew that there were other sources of mutations. Hermann Muller had proved that X-rays were mutagens, and by the late 1970s, the California biochemist Bruce Ames—using the genes of bacteria, which were even easier to work with than fruit flies—had proved that dozens of known carcinogens, including many industrial chemicals, were also capable of mutating DNA.15
Just about any stray sample of soil, groundwater, or air from Ciba-Geigy or Reich Farm was likely to contain at least one mutagenic compound.16 Many coal tar derivatives were mutagens, including benzene, benzo(a)pyrene, and benzidine. Epichlorohydrin, the resin feedstock that had so terrified Ciba-Geigy workers in the 1960s, was mutagenic too. Over at Reich Farm, meanwhile, the underground pollution plume seeping south toward the Parkway wells included several of the same mutagens, plus a few others.
For a town where mutagenic chemicals were now part of the landscape, the implications of Alfred Knudson’s research were clear: If a substance was capable of delivering a “hit” to DNA, it was a cancer risk, and no one was more likely to take a hit—or two or three—in a short period of time than a fast-developing fetus or a young child. When Bruce Anderson and other Ocean of Love parents heard about that, they wondered how many hits their children had endured, and who had delivered the blows. Before long, the State of New Jersey and the United States of America would be spending many millions of dollars to try to find out.
Lisa Boornazian, the oncology nurse at The Children’s Hospital of Philadelphia, was not a boat-rocker. In 1995, she was a twenty-four-year-old who loved her job and respected the hierarchy of the hospital. The doctors had told her it was just a coincidence that so many children on the ward were from Toms River, and she was inclined to take their word for it. They were very good doctors, and they gave their patients excellent medical care, even if they were too busy to get to know them. But she could not shake her uneasy feeling that the doctors might be wrong about Toms River.
She and her husband, Adam, came from large families and often got together on weekends with their siblings, many of whom still lived in the Philadelphia area. They were especially close to Adam’s sister, Laura Janson, and her husband, Eric. A few weeks after Carrie-Anne Carter’s funeral in Toms River in February of 1995, the two couples were having dinner on a Friday night, and somehow the conversation turned to the funeral and to Boornazian’s worries about Toms River. The discussion was not something that she had planned. “It just sort of happened by accident,” she would remember much later.
In their extended family, Laura Janson was an authority figure on environmental matters. She worked in the Philadelphia regional office of the U.S. Environmental Protection Agency, where she specialized in technical assessments of hazardous waste sites, though she had never worked on any in Toms River. After eleven years at the EPA, Janson was a bit jaded about information that came in from the public because it usually turned out to be confused, poorly documented, or otherwise unreliable. But this was different. Her sister-in-law was not an alarmist; she was a medical professional at a major hospital where thousands of children were treated every year. At dinner, when Boornazian started talking about all the sick childre
n she had treated from Toms River, Janson listened. When Boornazian asked her to check to see if anyone at the EPA was looking into the issue, she agreed. Janson would later explain her decision this way: “When you work at EPA, people are always saying something’s wrong with their water, but if it’s your sister-in-law talking, and she’s a nurse at CHOP who has made actual observations of cancer in children, you figure you’d better follow up.”
Janson checked and learned that there were two Superfund sites in Toms River and that neither had been the subject of an EPA health study. Was the agency considering launching such a study in Toms River? No, it was not, she was told. So Janson decided to call another federal agency, one that few members of the public had ever heard of. The Agency for Toxic Substances and Disease Registry was, and still is, a backwater in the federal environmental bureaucracy. Congress created it in 1980, when anxiety about hazardous waste was at its zenith, just seven months after the evacuation of Love Canal. The idea was that the EPA would oversee dumpsite cleanups, while the ATSDR would advise the EPA on the health hazards posed by each waste site. But the ATSDR had very little money to do its job, especially as the number of Superfund sites ballooned in the late 1980s. By 1996, the EPA was spending well over one billion dollars per year (about 20 percent of its budget) on Superfund; the entire ATSDR budget, meanwhile, was just $60 million.17
Steven Jones came to the ATSDR in 1992 from the EPA, where he had worked on Superfund cleanups in the Midwest. His new title, deputy regional director, suggested sweeping authority; the reality was that the ATSDR’s regional office, which occupied a small corner of the EPA’s space in Manhattan, consisted of just two managers: Jones and his boss. Their main responsibility was to make sure that state and local environmental health agencies were doing the work the short-staffed ATSDR could not do itself. Only rarely did Jones field calls from the public about supposed cancer clusters; when he did, they usually ended in mutual frustration. Like Laura Janson, he had been around long enough to know that ordinary citizens rarely understood what constituted a true cancer cluster.
So when Steve Jones’s office phone rang one morning in March of 1995 and the woman on the end of the line started talking about a possible cluster in Toms River, New Jersey, there was no reason to think it would be anything more than another dead-end conversation. As Jones listened, though, he heard some things that caught his attention. For starters, the caller was not a typical anxious citizen. Laura Janson introduced herself as an EPA employee, explaining that she was calling unofficially to relay the concerns of an oncology nurse, her sister-in-law, at The Children’s Hospital of Philadelphia. Unlike so many callers, Janson was not passing along a vague report in which all of the most common types of adult cancer—lung, breast, skin, and all the rest—were lumped together. She spoke instead about much rarer children’s cancers, especially brain tumors. “What she was expressing to me was very specific,” Jones remembered. “I felt like I wouldn’t be doing my job if I didn’t follow up. Someone needed to look at it.”
Like Lisa Boornazian and Laura Janson, Jones knew almost nothing about the sordid history of chemical pollution in Toms River. Also like the two sisters-in-law, he was neither a rebel nor a starry-eyed idealist, and he was anything but a naive civilian. Jones, Boornazian, and Janson all worked inside sprawling, risk-averse bureaucracies. They had been around long enough to be a tad cynical about the outsiders, the amateurs who were always demanding investigations without knowing the facts. And yet, in this case, all three felt that they could not ignore what they were hearing. The numbers were too high to dismiss as coincidence, and each case represented a face, a child, and a family.
Under the circumstances, Steve Jones felt that he could not say no to the EPA employee who called him, just as Laura Janson felt that she could not say no to her sister-in-law, the oncology nurse. As for Lisa Boornazian, she, too, felt that she had no real choice. She had to speak up for the memory of Carrie-Anne Carter and all the other Toms River children who had come through the cancer ward. They deserved an answer.
So on March 13, 1995, Steve Jones called the New Jersey Department of Health and formally requested an investigation of childhood cancer in Toms River.
They sparked everything that came afterward in Toms River, yet Lisa Boornazian and Laura Janson quickly faded from view after Janson’s phone call to Steve Jones in 1995. More accurately, the two women were never in view at all. Until the publication of this book, their identities were unknown to everyone connected with the Toms River saga except Jones and a few of Boornazian’s fellow nurses at CHOP.18 The ATSDR had a strict policy regarding calls from citizens seeking investigations: Callers had the right to remain anonymous. At the time, the two sisters-in-law did not want to be named, though they changed their minds many years later. Janson felt that she had nothing more to contribute, and Boornazian did not want to make waves at the hospital. “I felt like I had done my part, and now it was up to the experts,” Boornazian explained later.
Over the years, the anonymous “CHOP nurse,” as she was widely called, would acquire a sort of mythical standing in Toms River as the person who finally got the state’s full attention. Yet no one ever knew who she was—not even Linda Gillick. In the absence of solid information, the nameless nurse was a blank slate upon which the various players in the Toms River drama drew the characteristics that suited them. Officials at the ATSDR asserted that the nurse had gotten interested in the Toms River cluster after attending an agency workshop on environmentally induced cancer held at CHOP. Some of the Toms River families, meanwhile, had heard that she became involved because she had relatives in town and was a mother herself. There was even a rumor that she had grown up in Toms River. None of it was true. Boornazian was a Philadelphia native with no children of her own yet. She had not attended any special workshops and had no connection to Toms River except through the families she met on the ward.
In the years that followed Janson’s 1995 call to the ATSDR, she and Boornazian occasionally saw or heard snippets of news about the Toms River investigations: a stray newspaper headline here, a few seconds on the six o’clock news there. But life rolled on; they were too busy to pay attention to the extraordinary events they had initiated in Toms River. In November of 1995, Laura Janson gave birth to a son, Kevin. Lisa Boornazian’s daughter, Amanda, was born the following January. Boornazian took a leave from the hospital, and then another one in 1997 when her second child was born. She quit for good in 1999. Boornazian had never stopped loving the job, but balancing the demands of motherhood and the oncology ward was too difficult—especially the emotional demands.
On her last day on the ward, she watched a mother climb into bed with her dying daughter. “She just held her little girl and said, over and over, ‘What am I going to do without you? What am I going to do without you?’ For me, that was it. Now that I had my own kids, it all just felt too real,” Boornazian remembered. “It was time to leave.”
Laura Janson, meanwhile, stayed at the EPA, working faithfully on Superfund and water pollution cases as they moved slowly through the bureaucracy. Thirteen years after her brief call to Steve Jones, she ruminated on its unintended significance: “I’ve been with the EPA for twenty-three years, and that one phone call may have been the biggest contribution I’ve made to the protection of public health and the environment,” she said. “That was my moment, I guess.”
CHAPTER FIFTEEN
Cluster Busting
By the spring of 1995, when Steve Jones called to ask him to look into a possible cluster of childhood cancer in Toms River, Michael Berry had been New Jersey’s chief cluster investigator for almost nine years. It was still just a part-time responsibility—Berry spent most of his time on other tasks at the state health department—but it was now the least enjoyable part of his job. One of the first “incidence analyses” Berry ever attempted was the 1986 study of childhood cancer in Toms River. Its ambiguous results turned out to be a harbinger of dozens of similarly unsatisfying cluster s
tudies he undertook around the state—including another one about Toms River kids in 1991. “After a while, it got frustrating,” he recalled many years later. “I mean, what were we accomplishing?”
A cluster study in New Jersey was like one of those old-fashioned Hollywood movie backdrops that looked fairly impressive until you leaned on it and it toppled over. Berry’s frustrations ran much deeper than just the usual problems with the state cancer registry, which in 1995 was still in poor condition, its records incomplete and arriving three years late or longer. As he accumulated years on the job, Berry came to realize that even if the registry had been up-to-date and reliable, he would not have been able to tell callers what they really wanted to know: whether an environmental problem was causing cancer in their neighborhood. For reasons that were not easy to explain in a phone call, it was a question Berry could not answer. In fact, Berry spent almost as much time explaining the limitations of cluster studies as he did conducting them. Sometimes Berry felt his job was more about being a therapist to anxious callers—he got about thirty cluster calls per year—than investigating what they told him. “It was hard to believe I was really addressing anybody’s concerns,” he recalled.
Most of the people who called him had at least three major misunderstandings about cancer patterns, each of which led them to assume that all clusters had a hidden cause and that Michael Berry could unearth it—if only he tried hard enough. The first misunderstanding was the nature of clustering itself. As Berry knew, everything clustered to a degree, often for no reason other than chance. Nothing that was subject to the complexities of the natural world, whether birds in a flock or sick people in a city, was distributed evenly in space and time. Some clumping was inevitable. In cancer incidence studies, the challenge was not to find the clumps—that was usually pretty easy, thanks to the registry—but to identify which were likely to have an underlying cause other than randomness.
