What is a cancer cluster?

Cancer clusters have gotten a lot of attention in the media recent years, both in the news and in Hollywood movies. More than 1,000 suspected cancer clusters are reported to state health departments each year. But just what is a cancer cluster?

Many people tend to think of a cancer cluster as a higher than normal number of cases of cancer in a defined community. Cancer clusters are usually thought to be caused by some type of pollutant or carcinogen (cancer-causing agent).

But scientists usually define a cancer cluster differently. They think of them as a greater number of cancer cases than would be expected in a defined geographic area or group of people over a certain period of time. These cases of cancer may not seem random and need to be looked at more closely. A cancer cluster almost always involves only one type of cancer. Studying cancer clusters allows scientists to identify areas of increased cancer risk as well as try to figure out what is causing the increase in risk.

What are the characteristics of a true cancer cluster?

Cancer is a group of more than 100 different diseases. Each type of cancer has its own risk factors and causes. This is why true cancer clusters very rarely involve more than one type of cancer. For it to be considered a true cluster usually means it has one of the following characteristics:

• there are several cases of a rare type of cancer
• there are larger than expected numbers of a more common type of cancer
• it is a type of cancer that is not usually seen in a certain group of people (for example, children getting a cancer usually seen in adults)

If the cluster includes many different types of cancer over a period of many years, it is very unlikely to be a true cluster caused by a single environmental factor.

It's also important to remember that cancer is a common disease. Nearly half of all men and a little over one-third of all women in the United States will develop cancer during their lifetimes. It is also a leading cause of death -- cancer causes about 1 out of every 5 deaths in the United States. So, it isn't uncommon for several people in a relatively small area to develop cancer around the same time.

Most well-documented cancer clusters have not been found in the community. Rather, they have been seen in the workplace, where exposures to certain chemicals or other factors tend to be higher and last longer. Also, the group of people being exposed is much better defined in workplace groups. In fact, the links between cancer and many cancer-causing agents (called carcinogens) were first figured out from studies of workers in certain jobs.

Lung, skin, and bladder cancers are the types of cancer most often linked with high level exposure to workplace carcinogens. Other cancers such as leukemia, lymphoma, testicular, and brain cancer sometimes occur in clusters, too.

Statistics can usually help figure out if a cancer cluster is strictly due to chance. But even statistics can't tell the whole story. The excess number of cases reported in a cancer cluster may look significant based on statistics, but this does not mean that they are caused by something unique to that area. Some clustering of cancer cases can still happen by chance, but people tend to notice and report situations where rates seem to be above average.

Who investigates cancer clusters, and how is this done?

People concerned about a possible cancer cluster often report it to a local or state health department. Procedures vary by state, but most health departments will first ask for information, such as:

• the type(s) and number of cancers involved
• any suspected exposure(s) that might cause cancer
• the area and time period in which the cases occurred
• specific information about each person thought to be affected
• specific information about the cancers themselves

After determining this information, as many as 4 of 5 suspected clusters are found to NOT be true clusters. For example, the group of people may have different types of cancers, or the number of cases isn't greater than what would be expected given the ages of the people in the group and other risk factors.

If the health department feels that the potential cluster should be studied further, more information will be collected. Attempts may be made to contact patients or relatives to check information. Sometimes investigators will ask to see medical records. The health department may do a more in-depth statistical analysis on the number of cases in the affected area and nearby regions. Scientists in the health department may also look at reports in the medical literature to see if other clusters like this have been noted in the past.

If needed, the health department may ask federal agencies, which tend to have more resources for help. The Centers for Disease Control and Prevention (CDC) is the agency most often involved. Other agencies, such as the Environmental Protection Agency (EPA) and the National Institute of Environmental and Health Sciences (NIEHS), may also help in the investigation. Federal agencies may do a more in-depth investigation, including getting more thorough medical histories. These agencies may also take and test samples from the environment (air, soil, drinking water, etc.), and test blood or other body fluid samples from both affected and non-affected people.

Suspected cancer clusters can cause a great deal of concern and confusion in a community. It is very important that government agencies keep members of the community informed from the start of and throughout the investigation. This should include giving people a realistic idea of what may or may not be found.

What are the possible outcomes of a cluster investigation?

It may not be a true cluster

In many cases, investigators can determine that a "cluster" of cancers isn't a true cluster. For example, the number of cases may not really be higher than expected when other factors (such as age, gender, and tobacco use) that could explain the increase are taken into account.

A cancer cluster and its cause may be found

In some cases, especially when there may be an obvious potential cause, a true cancer cluster may be confirmed. At that point, steps can be taken to address it.

A cancer cluster may be found, but no cause can be identified

Even if investigators believe that it may be a true cluster, it's important to know that a cause for it is found very rarely. There are several reasons why this is the case, for example:

Random patterns can look like a cluster

Even if the number of cases in an area is higher than expected, it still may not be a true cancer cluster. More cancer cases in the US are expected in large population centers or in places where the population tends to be older. But even so, for the most part, cancer cases in the United States are spread randomly across the country. As with any random pattern, there will be more cases than expected in some spots, and fewer cases than expected in others. The areas with more cases than expected are more likely to be noticed. But many of these will be due to the "bull's-eye effect" (which is something like drawing a target on the wall after the darts have been thrown). Suppose you took a map of the United States and started drawing circles of different sizes in different locations. You would find that some of the circles would contain more cancers than expected, and some would have less. Some of the circles with more cases might be true clusters, but most would be due to chance.

It is very hard to figure out which of many exposures might be the cause

With rare exceptions, scientists don't have a way of telling what trigger (if any) may have caused cancer in any one person, whether it's part of a cluster or not. Humans are not like lab animals -- their environments are not strictly controlled. People are exposed to many natural and man-made substances during their lifetimes. Think about how hard it would be to test for everything you've been exposed to, even if you knew where to start. Investigators may have a few clear leads or starting points, but they need to look at all of the possibilities. Finding the one exposure that may be the cause can be like looking for the needle in the haystack.

There is often a long delay between exposure and cancer

In clusters where the cause is known, mostly from workplace cases, the time between exposure to the substance and the development of cancer has been anywhere from a few years to several decades. Exposures are not likely to cause cancers right away. And again, it's not easy to study people and their environments. Suppose a group of people live in a community that has a higher than expected number of cases of a certain type of cancer. If there is a potential cause, investigators first have to figure out when these people were exposed to it. Was it a single event or has it been ongoing? Was it 5 years ago, or 10, or 20? And what did the people with cancer have in common during that time? Added to this, some people will have moved into the community, while others may have moved away since then. Should the cases of people who moved into town in recent years be included? And can the people who moved away be found?

The boundaries of the cluster area can be hard to define

Defining the geographic cluster area is not always as clear-cut as one might think it would be. Just how big should the "bull's eye" be? Should it include only the local neighborhood where most of the cases were found? Or should it also include the larger community, or even nearby communities? These areas may have cases that may or may not be related to the others.

Not everyone who is exposed is likely to develop cancer

To make things even more tricky, people may be more or less prone to getting cancer based on their genes. It's unlikely that everyone exposed will develop cancer. And there may be people who were not exposed who develop the same cancer by chance.

Questions may still remain

Scientists do their best to piece the puzzle together, but more times than not, they don't find a likely link. This doesn't mean that there isn't one; it may just be that one can't be teased out with the methods scientists have at the time. Sometimes there really isn't a link, but it's hard for everyone to feel sure of that. This may be an unsatisfactory answer for people in a community being affected, but it may be the case.

What should I do if I suspect a cancer cluster?

Concern about cancer clusters most often comes up in schools, in workplaces, and in certain areas of a community. If you are concerned about a possible cancer cluster, you may want to contact your local or state health department. If you don't know who to call in your area, the CDC has a list of Web links to local and state contacts at www.cdc.gov/nceh/clusters/statelocal.htm. (Also see the "Additional resources" section.)

Some of your concerns may be relieved by making a phone call. If not, the health department will probably look into doing a brief investigation and, if needed, will do a more complete review of the situation. For more complex or urgent situations, the department may also call in experts from the CDC or other agencies to do a more in-depth investigation.

The investigators may be able to figure out with a fair amount of certainty that there is no true cluster, or that there is a true cluster and it is probably caused by a certain exposure. But in many cases, a single cause cannot be found.

Suspected cancer clusters often greatly distress those involved. It is very important that these situations be handled openly, promptly, and professionally. Even though thorough investigations of potential cancer clusters rarely give clear-cut answers, it is important that communities keep reporting suspected cancer clusters, and that health organizations respond to those concerns.

Additional resources

More information from your American Cancer Society

The following information may also be helpful to you. These materials may be ordered from our toll-free number.

• Environmental and Occupational Cancer Risk Factors: Overview

• Known and Probable Human Carcinogens

• What is Cancer? (also available in Spanish)

National organizations and Web sites*

Along with the American Cancer Society, other sources of patient information and support include:

National Cancer Institute (NCI)
Toll-free number: 1-800-4-CANCER (1-800-422-6237)
Web site: www.cancer.gov

Centers for Disease Control and Prevention (CDC)
Toll-free number: 1-800-232-4636
Web site: www.cdc.gov
Area of Web site with cancer cluster information: www.cdc.gov/nceh/clusters/
Links to state and local health departments: www.cdc.gov/nceh/clusters/statelocal.htm

National Institute of Environmental Health Sciences (NIEHS)
Telephone: 919-541-3345
Web site: www.niehs.nih.gov

NCI and NIEHS wrote Cancer and the Environment: What You Need to Know, What You Can Do, about links between cancer and toxic substances. It can be found online at: www.niehs.nih.gov/health/scied/documents/CancerEnvironment.pdf

*Inclusion on this list does not imply endorsement by the American Cancer Society.

No matter who you are, we can help. Contact us anytime, day or night, for information and support. Call us at 1-800-ACS-2345 or visit www.cancer.org.

References

Centers for Disease Control and Prevention. Cancer Clusters. Accessed at www.cdc.gov/nceh/clusters/default.htm on January 21, 2009.

Kingsley BS, Schmeichel KL, Rubin CH. An Update on Cancer Cluster Activities at the Centers for Disease Control and Prevention. Environ Health Perspect. 2007;115:165-171. Accessed at www.ehponline.org/members/2006/9021/9021.pdf on January 22, 2009.

National Cancer Institute. Cancer Clusters Fact Sheet. 2006. Accessed at www.cancer.gov/cancertopics/factsheet/Risk/clusters on January 21, 2009..

Robinson D. Cancer Clusters: Findings vs feelings. MedGenMed. 2002;4(4):16.

Thun MJ, Sinks T. Understanding Cancer Clusters. CA Cancer J Clin. 2004;54:273-280.

Wheeler DC. A comparison of spatial clustering and cluster detection techniques for childhood leukemia incidence in Ohio, 1996 – 2003. Int J of Health Geographics. 2007;6:13. Accessed at www.ij-healthgeographics.com/content/6/1/13 on January 21, 2009.

Last Medical Review: 01/29/2009
Last Revised: 01/29/2009