All Sections
Welcome guest, you have 3 views left.  Register| Sign In

Home | Health

Health official: We may never learn what caused Seacoast 'cancer cluster'

New Hampshire Sunday News

April 02. 2016 6:17PM
Paul and Lynne Thomas talk about their son "Sam the Man," a child who lost his life to Rhabdomyosarcoma in North Hampton. He's counted as part of the state's Seacoast cancer cluster. (Thomas Roy/Union Leader)
How it all came about
In March 2014, a Rye resident contacted the state health department with concerns about a small number of children from Rye who were being diagnosed with a rare kind of pediatric cancer.

Rhabdomyosarcoma (RMS) is a type of soft-tissue cancer; two-thirds of cases are diagnosed in children under the age of 6, according to a state report.

The Division of Public Health Services launched an investigation, looking at cancer rates in Rye and four neighboring communities: New Castle, Portsmouth, Greenland and North Hampton. The first step was to gather data from the New Hampshire State Cancer Registry, which collects reports of all state residents diagnosed with cancer.

Investigators then compared the incidence of RMS and other cancers with the expected rates in the rest of Rockingham County.

The number of RMS cases in the five communities was in fact higher than would be expected, although the actual number was “fewer than five,” the state reported.

The rates of other types of cancer in the five communities for both adults and children were in line with expected rates countywide, except for one: a rare type of lung cancer called pleuropulmonary blastoma (PPB).

Again, there were “fewer than five” PPB cases in children from the five communities from 2005 to 2014.

Confidentiality laws prevent officials from revealing just how many cases there are if the number is lower than five; it's too easy to identify patients, officials explained.

That could change, however, if additional RMS or PPB cases are confirmed, pushing the numbers above five.

At a community meeting held two weeks ago in Rye, residents told state epidemiologist Dr. Benjamin Chan that there were other RMS cases in families who previously had lived in the area but had moved out of state.

Those cases would not be in the state Cancer Registry, since it only captures data on New Hampshire residents, according to Dr. Judy Rees, the registry director.

State residents who are diagnosed out of state — at a Boston hospital, for example — will be reported to the database but there's typically a lag time, Rees said.

As part of an ongoing study, public health officials plan to contact families who previously lived in the Rye area and whose children were later diagnosed with RMS.

Rees said the registry is also checking on cases that have been reported to the state since the community meeting, to make sure they are included in the registry.

“If we have missed cases, we will say so and amend the analysis,” she said in an email.

The blunt truth is that Seacoast residents may never learn what - if anything - caused a "cluster" of two rare pediatric cancers over a decade.

But the state epidemiologist says public health investigators will do their best to identify any environmental or behavioral exposures that could be linked to the cancer cases recently identified in the Rye area.

Finding that link won't be easy, Dr. Benjamin Chan warned in a Friday interview.

"I think people should be aware that the majority of times, these types of cancer cluster investigations ... don't give an answer," he said. "And especially when you're dealing with such small numbers, it becomes very difficult to identify connections between exposures and cancers."

Still, he said, the state is broadening its investigation in response to community concerns.

Defining a 'cluster'

The Centers for Disease Control and Prevention (CDC) defines a cancer cluster as "a greater-than-expected number of cancer cases that occurs within a group of people in a geographic area over a period of time." All of the cases must involve the same type of cancer, or types of cancer scientifically proven to have the same cause, according to the CDC website.

In the Seacoast cluster, the state found fewer than five cases each of two rare types of pediatric cancer - rhabdomyosarcoma (RMS) and pleuropulmonary blastoma (PPB), according to a report released in February.

But Chan said additional RMS cases have since been reported to the state by families who previously lived in the Rye area; the state plans to contact those families to confirm the diagnosis and other facts.

Investigation continues

The next phase of the investigation will focus on trying to identify possible causes for these cancers, Chan said. The state plans to send detailed questionnaires to affected families, asking about "behavioral" factors such as smoking or drug use, and environmental risks such as exposure to chemicals.

The state also will look into concerns residents have about possible exposures such as the Superfund site at the former Coakley landfill in North Hampton.

But finding answers could be elusive, in part because there are so few cases to look at for common risk factors, Chan said.

Scientific studies done to date have not identified specific environmental or behavioral risk factors that have been linked to RMS or PPB. And that makes it harder to find a cause for the Seacoast cluster, he said.

Here's why: If you know a certain chemical is linked to a particular kind of cancer, investigators could ask affected families about possible exposure to that chemical. But that's not the case for RMS or PPB, he said.

Some studies have found "a strong genetic and familial connection" to PPB, according to the state report.

An expert in childhood cancer at Dartmouth-Hitchcock Medical Center says the higher-than-expected number of cases on the Seacoast does not necessarily mean that there's a common underlying cause.

Expert in field

Dr. Jack van Hoff is chief of pediatric hematology/oncology at DHMC and an associate professor of pediatrics at Dartmouth's Geisel School of Medicine. He examined the state report on the Seacoast cancer cluster and attended a public meeting in Rye at Chan's request to answer concerned residents' questions.

What struck him most, van Hoff told the Sunday News, is the "very small number of cases" involved.

"Rare events do happen, and events that are randomly distributed sometimes wind up on the same street or in the same neighborhood or in the same town," he said. "That doesn't mean that there's an environmental or regional or other precise cause for it."

From an epidemiological standpoint, he said, the more you look for something, the more likely you are to find it. "Just by chance alone, if you look at enough different places, you're going to come up with more than the expected number of events in a particular town or a particular three or four or five towns."

May be no connection

There's something called the "sharpshooter effect," he said. "That's somebody who stands back, fires three shots at the barn, and then draws a target around it and says what a great sharpshooter he is," he said.

The same could be true here, he said. "The three events happen, you then draw a target around these three events, then you say this is more than what might have been expected. But they could easily have happened by chance."

That's how two seemingly opposite things can be true: A cancer cluster can be both "statistically significant" and random.

Statistically significant in scientific terms "just means that when you did a mathematical analysis, you got a number that says this probably didn't happen by chance - but it could have happened by chance," van Hoff said.

He knows that's of little comfort to the families. He has devoted his career to working with pediatric cancer patients and said the death of a child from cancer is "devastating."

Hard call

But when there are so few cases, it's difficult to determine potential causes, van Hoff said.

For instance, he said, if you learned that all the families used the same lawn-care service, it might be easy to conclude that some chemical applied to lawns could be the culprit. But it could be that they all use the same company "because that's where they live, and there's maybe no connection whatever."

The larger the case numbers, the more comfortable researchers may be in determining a cause, he said. But one previous study of 300 RMS cases did not identify potential environmental factors that might have contributed to the cancers.

"That doesn't mean that environment cannot play any role," he said. But this kind of study "has been attempted and hasn't really identified environmental factors that might have contributed."

The state's report also turned up a "small excess" of another pediatric cancer known as pleuropulmonary blastoma (PPB), a rare type with a strong genetic and familial link. There's a genetic mutation has been identified; only about 10 percent of those with the mutation will get the cancer, van Hoff said.

Van Hoff actually ran into the same problem 20 years ago when he was doing an epidemiological study of pediatric leukemia rates in areas exposed to radiation from the Chernobyl nuclear accident.

Restricting the analysis to contaminated zones didn't produce enough cases to be statistically significant, he said. But broadening the study area reduced the radiation exposure to undetectable levels.

"So there we had a known carcinogen being deposited in a very known area ... and in spite of that we couldn't demonstrate a statistically significant difference in anything except thyroid cancer rates," which was expected, he said.

It shows how difficult it may be to find a cause for the Seacoast cluster, he said.

"I think what we can say based on what's been demonstrated so far is that we can be fairly sure that if there is an exposure to account for these events, that the impact of it is quite small."

As a pediatrician who has devoted his career to caring for young cancer patients, Van Hoff said he understands how devastating the illness is for these families.

At the same time, however, he has some frustration about how many other more pervasive health threats to children are ignored, such as smoking, lack of exercise and diet.

Community involvement

Van Hoff praised the local residents who have pressed the issue. "They're proactively looking at this and getting themselves informed," he said, "and asking the state to look into it in a way that's appropriate, and I think that's the right thing."

The next part of the study will take time, Chan said. Investigators have to decide which cancers to investigate further and whether to expand the number of communities involved.

"The best chance of being successful at identifying a connection between some exposure or some behavior and this cancer is to do it in a very scientifically rigorous process," he said.

Van Hoff commended public health folks for taking residents' concerns seriously despite the small number of cases.

"The state of New Hampshire didn't say, 'Oh this is nothing' and sweep it under the rug," he said. "But they said, 'We will look into this. Our people are concerned, that makes it important to us and we're checking it out.'"

Van Hoff lived and worked in Connecticut for 20 years before coming to Dartmouth eight years ago.

"Personally, I think the state has done a very proactive and thoughtful investigation," he said. "And I don't think that would happen many other places, to be honest."

Health Deaths in the news General News North Hampton

Follow us on Twitter Follow us on Facebook Follow our RSS feed
Union Leader app for Apple iPad or Android *
Click to download from Apple Apps StoreClick to download from Android Marketplace
* e-Edition subscription required