In mid-April, news outlets began reporting a possible cancer cluster linked to a high school in New Jersey.
More than 100 former students and staff had been diagnosed with brain tumors over a 30-year period, leaving some local residents calling for an investigation.
Just weeks later, another cancer cluster warning hit the news. A former Philadelphia Phillies baseball pitcher had died of brain cancer in mid-May, making that six retired Phillies players since 2003. The death ignited fears that the cases could be linked.
Searching for the “why me?” is a natural reaction to a cancer diagnosis. And when a possible pattern presents itself, people may be tempted to connect the dots themselves.
About 1000 suspected cancer clusters are reported to state health departments in the United States each year. That’s almost three a day. These reports tend to originate from concerned community members who perceive a higher-than-normal number of cancers in their area.
But cancer clusters are notoriously difficult to confirm. Most cases don’t rise to the level of cluster, and even for those that do, the cause often remains elusive despite lengthy investigations.
Experts reveal what makes cancer clusters so tough to identify and understand and how oncologists can respond to people’s fears about them.
Is It Really a Cluster?
A group of cancers that pop up in a small community may seem alarming at first glance.
“There’s a public perception that cancer is rare,” said Jerry Fagliano, MPH, PhD, an epidemiologist and former manager of the Environmental and Occupational Health Surveillance Program for the New Jersey Department of Health.
When “someone recently diagnosed with a certain cancer or a family member finds out someone down the street or at work had the same type of cancer, it sort of evolves from there,” according to Fagliano, now at Drexel University in Philadelphia.
Often, however, these numbers align with what epidemiologists would expect.
Cancer is the second leading cause of death in the US each year, just behind heart disease. And roughly 40% of people in the US will be diagnosed with cancer in their lifetime, according to the National Cancer Institute.
To rise to the level of a cluster, a greater-than-expected number of cancer cases must occur within a group of people in a defined geographic area over a specific period, according to the US Centers for Disease Control and Prevention (CDC) definition.
State and local public health agencies investigating potential cancer clusters will assess whether a collection of cases fits these criteria, which inevitably will vary by cancer type. The first step may be creating a ratio of observed to expected cases by comparing the group of people who have a particular cancer to a group that shares similar characteristics. That may mean comparing the incidence of cancer in one community to anticipated numbers for the county or the state.
Suspected clusters of one type of cancer, a rare type of cancer, or a type of cancer not typical in a certain group — leukemia in children, for instance — may be more likely than others to meet cancer cluster criteria.
For instance, in the early 2000s, researchers identified a cluster of childhood leukemia in Fallon, Nevada — 16 cases were diagnosed over 5 years. Fewer than two would have been expected over that period in the population.
But, according to the CDC, most cancer cluster investigations end after an initial inquiry.
Even cancers that are considered rare may add up in a population over time, explained Robert Laumbach, MD, MPH, an occupational and environmental medicine physician at the Environmental and Occupational Health Sciences Institute of Rutgers University in New Jersey.
That’s what appears to be the case at the New Jersey high school. An investigation by an environmental engineering firm found no evidence of a cluster or radiation hazard to help explain the brain tumors and cancers. The lifetime risk of developing brain cancer or other nervous system cancers is only about 0.6% for an individual, but that can amount to dozens of cases in a community of several thousand.
Still, a seemingly sizable group of cancer cases in a community can spark concern.
“Frequently, patients and family members ask me whether their personal experience with cancer is out of bounds with what might be expected and if I know of others like them in a specific area,” said oncologist Arif Kamal, MD, chief patient officer for the American Cancer Society.
The question comes up particularly among people diagnosed at earlier ages or among those who have lived or worked in a specific geographic area, he explained. But linking cancer cases geographically can prove challenging as well.
In the past, “we weren’t able to systematically track cancer diagnoses in an area or population,” Kamal said. “Oncologists had no way of coordinating with other oncologists on what they were seeing.”
There’s much more cancer surveillance taking place now. Big advances in data collection have been made since the 1970s with population-based cancer registries, according to Kevin Henry, PhD, a medical geographer at Temple University and Fox Chase Cancer Center in Philadelphia.
Cancer registries allow researchers and health officials to track cancer diagnoses for public health reasons. These data can be used to follow cancer trends over time and identify cancer patterns in certain regions or groups of people.
Ideally, this tracking system would allow experts to see clusters popping up. Yet registries aren’t a perfect tool.
With no single surveillance program in the United States, keeping tabs on cancer cases can prove challenging. Cancer registries typically are tied to a particular hospital system or a state’s public health department, which means cases in certain regions or populations may go underreported.
The Search Is Often a Dead End
Did a toxicant leach into the water supply? Is a pollutant lingering in the air?
Environmental contaminants are frequently the go-to suspect when concerns about a cancer cluster emerge.
In some instances, investigations do uncover a cause.
In the 1970s, researchers found that exposure to vinyl chloride, a chemical used in the production of PVC plastics, at a polymer production plant in Kentucky likely led to a rare liver cancer among some workers.
But even when experts confirm a cancer cluster exists, the hunt for a source is often a dead end, Laumbach said.
According to the CDC, the presence of a cluster “does not necessarily mean that there is any single, external cause or hazard that can be addressed.”
A 2012 study that evaluated 428 investigations of 567 cancers of concern found that only one investigation revealed a clear cause.
“The environment is a big tent,” said James Wurzer, MD, medical director of radiation oncology at Atlanticare Regional Medical Center in Egg Harbor Township, New Jersey.
Potential environmental risk factors for cancer can include a mix of basic lifestyle factors — diet, exercise, and smoking — as well as exposure to UV rays and certain viruses, such as human papillomavirus, Wurzer explained. Cancer may also happen by chance or be caused by a source not yet associated with cancer.
Cancer risk frequently involves the accumulation of lifetime exposures layered on predisposing genetic factors, which can be difficult, if not impossible, to unravel.
That is why even resource-heavy investigations to unearth the reason for a cluster can come up empty.
In the Nevada case, one of the biggest cancer cluster investigations conducted by the CDC, researchers found no associations between the cluster of childhood leukemia cases and any environmental contaminant. Investigators analyzed blood, urine, and cheek cell samples for more than 100 chemicals, viral markers, and genetic abnormalities. They took air, water, soil, and dust samples from dozens of homes. And still no clear culprit emerged. The investigators ultimately concluded that the overall small number of cases — 16 — may have limited their ability to find an association between leukemia and environmental exposures.
Another major roadblock to pinpointing a source: Most cancers take years, even decades, to develop. Mesothelioma, a rare cancer with a strong environmental link to asbestos exposure, may not show up for 30 to 60 years after exposure.
With such long latency periods, the timeline of exposure becomes even more complicated when people move.
“Communities are dynamic,” said Fagliano. “The people living there today may not be the same ones that were living there 20 to 30 years ago. And it becomes very challenging to reconstruct whether and to what degree someone may have been exposed to past environmental conditions.”
Even detailed residential histories don’t necessarily provide the relevant exposure data that researchers would need to prove a probable environmental cause.
“An exposure may not always be where you live right now. It could be where you work or where you used to work. It could be where you were as a child,” Henry said.
Amid Challenges, Solutions Still Possible
In some instances, a cancer cluster investigation does turn up a clear environmental contamination. Still, it is rare to prove that the pollutant caused a person’s cancer.
In Tom’s River, New Jersey, an investigation of a childhood cancer cluster in the late 1990s and early 2000s turned up decades of groundwater pollution from multiple sources. Investigators determined that exposure to the water pollution while in utero likely led to an increased risk for a subset of the cancers — specifically, leukemia in girls. But a causal contaminant was not found.
“Causation can be very complicated to prove, especially when it comes to something like air pollution or water pollution, which can be a mixture of exposure agents,” said Moon Chen, MPH, PhD, a cancer epidemiologist at the University of California, Davis. “Rather than trying to explain it, sometimes the best thing we can do is try to prevent exposure.”
That’s what happened in Tom’s River. The state closed two wells and expanded the water treatment system.
“If an issue with a contaminated water supply or waste site is uncovered, it should be addressed whether or not it is linked to a local cancer cluster,” explained Fagliano, who was involved in the Tom’s River response.
Identifying the cause of a patient’s cancer can help people make sense of something that feels random, chaotic, and scary. But because the cause is frequently unclear, conversations with concerned patients and families may be tricky.
“We have to be very judicious in how we discuss causality,” said Wurzer. “We can say that an environmental exposure increases risk, but I don’t think we can ever 100% say that a certain environmental risk factor caused a particular cancer.”
Wurzer said, “part of the healing process is being able to move on from the ‘why’ so the patient can fully engage in treatment.”