Cancer Clusters: Findings Vs Feelings

David Robinson

In This Article

Are Clusters Real?

There is no doubt that cancer clusters are a real phenomenon. Scientists have repeatedly used epidemiology to identify clusters that establish surprising and previously unsuspected causes for cancer. Diethylstilbestrol (DES), a drug used during the 1940s and 1950s to prevent premature labor in pregnant women, is one example. Because a prestigious group of Boston physicians advocated the use of DES, many of the women who used it lived in the Boston area. Fifteen to 20 years after these women had used DES, doctors noticed a miniature "epidemic" of vaginal cancers among their daughters. The specific form of cancer observed -- adenocarcinoma -- is so rare that the hospital where 8 of the daughters were treated within the space of a few years had never encountered any cases of it before. Physician concern over the cancer led to an investigation that linked the vaginal cancers to DES.[3] Most observers might not call this a "cluster" because it does not involve supposedly carcinogenic conditions in the environment where people live, but it is considered one by epidemiologists since a large number of rare cancers in one area (Boston) prompted the study.

Other cancer clusters have been found in occupational settings. One example concerns the material polyvinyl chloride (PVC), more commonly known as vinyl. PVC is one of the most widely used synthetic compounds in the world. It is an important material for construction and siding, a common insulator of electrical and other wiring, and a common material for making shower curtains. At factories around the world, workers were exposed to very high concentrations of one of the ingredients used to make PVC, a volatile chemical known as vinyl chloride monomer (VCM). VCM is not used directly by the public, but it is a key part of the PVC manufacturing process. The workers who handled VCM, it was observed, had an elevated prevalence of a rare liver cancer, hepatic angiosarcoma. Rigorous investigations eventually proved that VCM elevated the risk of hepatic angiosarcoma in these factory workers.[4] Both the VCM and DES investigations are examples in which a thorough investigation established an unequivocal link between chemical exposure and human cancer. The public expresses concern over many possible clusters each year. In most cases, the perceived increase in the prevalence of cancer that has generated alarm is not, in fact, an increase in prevalence at all. According to public health officials, 85% or more of the possible clusters reported to them by concerned community members are, in fact, not statistically significant elevations of cancer rates.[5,6] Many of these supposed clusters include whatever types of cancer have been diagnosed in the area. This ignores the reality that different cancers are really different diseases.

As noted, cancer clusters have been reliably linked through controlled studies with specific medicines, like DES, and with occupational exposures to chemicals, like VCM. Scientific findings do not, however, match up with public perceptions about localized elevations in prevalence of many different cancers as a result of industrial pollutants in the environment. To quote from the most recent edition of Cancer Epidemiology and Prevention, "overall, there are a sufficient numbers of point source studies to conclude that excess lung cancer risk was very likely caused by large, relatively uncontrolled sources of community air pollution, particularly arsenic-emitting smelters.... However, the risks for community exposure are likely to be quite small, between 1.5 and 2.0 [times the normal rate] and will require substantial study populations and careful assessment of other known lung cancer risk factors."[7] In other words, ambient air pollution in nonoccupational settings may be linked to lung cancer -- not to the wide range of other cancers for which it is often blamed. At present, lung cancer is the only cancer that has been scientifically tied to ambient air pollution.

As to pollution in drinking water, the evidence is similarly thin compared with public fears. There is a complex mixture of trace elements present in drinking water, usually at levels of 10 or fewer parts per billion. Cancer Epidemiology and Prevention reviews several suggested cancer links -- including possible carcinogenic effects of arsenic and organochloride compounds with bladder, gastrointestinal tract, and other cancers. It does not identify any chemicals about which the body of evidence is large or consistent enough to support the conclusion that they increase cancer risk in the amounts that are present in US drinking water.[8]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.