Harvard Reports on Cancer Prevention
Volume I: Human Causes of Cancer
Cancer Causes & Control:
An International Journal of Studies of Cancer in Human Populations
Official Journal of the International Association of Cancer Registries
Volume 7 Supplement November 1996 ISSN 0957-5243

Environmental Pollution

Introduction
Environmental influences in the broadest sense probably are related causally to the development of most cancers. However, the majority of these environmental causes are thought to be related to behavioral factors (e.g., cigarette smoking, alcohol, and delayed childbearing), viral agents (e.g., human papilloma virus and Epstein-Barr virus), occupational exposures (e.g., benzene, asbestos, and coke oven emissions), or dietary factors (e.g., animal fats and aflatoxins). Other sections of this publication consider specific environmental factors, such as drinking water contaminants, passive exposure to environmental tobacco smoke, nonoccupational exposures to asbestos, indoor radon, ultraviolet radiation, and electric and magnetic fields. This section will focus on two remaining environmental factors that might be related to human cancer: air pollution and hormonally active aromatic organochlorines.

Air pollution
An excess of cancer in urban populations has been the subject of speculation since the European Renaissance. Descriptive studies over the last 20 years have provided support for the hypothesis that urban air pollution is associated with an increased risk of lung cancer. Studies from Europe, Japan, and Australia have reported an increased risk of lung cancer in urban areas with increased incidence of lung cancer in Norway observed to be as much as 60 percent for men and 90 percent for women.¹ These early studies suggested large excess risks, but failed to consider different lifestyle factors of urban residents, such as an early age of initiation of smoking, environmental tobacco smoke, domestic radon, and occupational exposures.

What in air pollution could be responsible for any observed increase in lung cancer? Emission inventories and direct chemical analyses of air samples have documented the presence in urban air of many compounds that are known to cause cancers in animals. Combustion products, principally polycyclic aromatic hydrocarbons, have long been considered as one agent responsible for the carcinogenic potential of outdoor air pollution. Industrial emissions, motor vehicle emissions, and consumer products contain benzene, formaldehyde, and other volatile organic compounds that are known carcinogens in both animal and human studies. Other exposures of concern include metal refineries (arsenic, cadmium, and nickel) and ore processing wastes (asbestos and uranium).

Recent epidemiologic studies of lung cancer and air pollution have controlled for the effects of cigarette smoking and estimated lifetime exposures to air pollution from direct, but short-term, measurements of particles, sulfur dioxide, or polycyclic aromatic hydrocarbons. Among nonsmokers, the risk of lung cancer is low in both urban and rural populations and the urban-rural differences are not consistent across multiple studies. In contrast, six cohort studies and six case-control studies of the general population have found that urban smokers a re at increased risk estimated to be from 10 percent to 80 percent due to the additional effects of urban exposure.^2,3

From these studies, the Environmental Protection Agency⁴ has estimated that the proportion of lung cancers due to air pollution is about one percent or about 1,600 lung cancer deaths per year in the United States population. The vast majority of all lung cancers are due to the effects of active and passive exposure to cigarette smoke, and a small proportion are due to occupational exposures. In summary, most of the epidemiologic studies can be considered as supportive or consistent with a small role for urban air pollution in causing lung cancer.

Aromatic organochlorines
Aromatic organochlorines include chlorophenoxy pesticides (DDT and its metabolite DDE), combustion by-products, such as polychorinated dibenzo-p-dioxins (PCDDs and TCDD) and dibenzofurans (PCDFs), and industrial products, such as polychlorinated biphenyls (PCBs) and polybrominated biphenyls (PBBs). These compounds are chemically stable over many decades, are passed along the food chain, accumulate in fatty tissue, and are eliminated slowly from the body. Animal studies have shown that TCDD, PCBs, and DDT are carcinogens at high doses.

Many aromatic organochlorines are capable of binding to estrogen receptors and are hormonally active in animals causing either estrogenic or antiestrogenic effects. However, the observed hormonal effects are three orders of magnitude less than with naturally occurring estrogens. In human studies, some evidence suggests that women exposed to higher levels of DDE have a decreased duration of lactation. These hormonal properties are of interest since estrogens are growth factors for tissues of the breast and uterus. At lower doses, the aromatic organochlorines may serve as tumor promoters to increase the rate by which a transformed cell grows into a clinically detectable tumor. A recent study has shown that DDT is a weak tumor promoter in male rats.

A recent technical review of the current epidemiologic literature has concluded that epidemiologic studies present conflicting evidence regarding the association of aromatic organochlorines and breast and uterine cancer.⁵ Recent trends and international, racial, and socioeconomic patterns in breast and uterine cancer incidence do not match data on exposure to organochlorines. One large occupational study of TCDD exposure found evidence of an association, but five larger studies (two on TCDD and three on PCB) did not. A study of breast cancer in Long Island, New York, did not find an association with proximity to known hazardous waste sites or contaminated wells but did find an association with residence near industrial facilities among postmenopausal women.⁶ A recent case-control study of 58 breast cancer cases⁷ found a marginal association with measured levels of DDE in blood, while a later case-control study of 150 breast cancer cases⁸ showed a much weaker association with blood levels of DDE. Neither of these breast cancer studies found an association with blood levels of PCB.

Conclusion
Environmental exposures via routes such as air and food contamination may include known or suspected carcinogens. Current epidemiologic research supports an association between urban air pollutants and cancers of the respiratory system with about one percent of lung cancer in the US attributable to air pollution. The evidence for an association of hormonally active aromatic organochlorines appears to be inconclusive. In both cases, the levels of exposures are typically much lower than with active cigarette smoking, replacement estrogens, or occupational exposures or even passive exposures to environmental tobacco smoke and natural variations in estrogen levels. Consequently, the risks associated with environmental exposures are generally several orders of magnitude less but involve much larger populations and people who are subjected unwillingly to the additional burden of these exposures. These topics are an area of very active research in which the combination of improved techniques for exposure assessment and creative epidemiologic study designs should provide more precise information. Until more scientific information is available, it would be prudent to take precautions against these particular environmental pollutants.

Suggestions

Suggested Further Reading

References