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

Exercise

Introduction
Although recent recommendations emphasize the need for regular physical activity,¹ particularly for cardiovascular health, much of the American population remains sedentary. Over the past decade, mounting evidence has indicated that physical activity also may reduce the risk of cancer. Excluding lung cancer, which is closely related to smoking, the most prevalent malignancies in the United States are cancers of the colon and rectum, breast, and prostate. These malignancies are associated with low levels of physical activity and other factors common in the American lifestyle. Exercise influences many of the metabolic processes and the levels of various hormones, some of which are likely to be related to specific cancers, so it is quite plausible that exercise may reduce cancer risk. The precise ways hormones and other biochemical processes are related to cancer, however, are complex and much remains unknown. Thus, the best evidence regarding exercise and risk of cancer comes from studies that directly examine the association between physical activity and risk of specific cancers.

Level of physical activity has been assessed in several ways in epidemiologic studies. Some studies rely on questionnaires based on self-reported levels of specified activities. Other studies have used occupational status to estimate physical activity on the job, as the average level of energy expenditure varies by occupation. Others have studied certain groups, such as participants of high school athletics, who presumably have high activity levels. No method is perfect, but each can yield important information.

Colon cancer
The evidence for a protective influence of physical activity is most clear for colon cancer, although exercise does not appear to protect against rectal cancers. Cancers of the colon make up about three-quarters of total colorectal cancers. A decrease in colon cancer risk has been observed in individuals who work at occupations that require high degrees of physical exertion. Also, studies of self-reported exercise and recreationhave indicated a lower risk of colon cancer among more active individuals.^2,3 The evidence regarding colon cancers is quite strong and has been observed consistently in numerous studies in different countries.

There are at least two plausible ways in which exercise may help prevent colon cancer. Exercise is required for the proper functioning of the large bowel. For example, people who are bedridden frequently suffer from constipation. Exercise increases the rate in which stool transits through the intestine and, as a result, contact with the potential carcinogens in the stool may be minimized. Exercise has numerous other benefits, such as increasing the level of high-density lipoprotein (HDL) or 'good' cholesterol and reducing blood sugar and insulin levels. Some evidence indicates that high circulating levels of insulin may promote the growth of tumors.

Although a reduction in colon cancer risk through physical activity is very likely, the magnitude of the benefit is not well quantified. Much evidence suggests that even moderate levels of activity, such as brisk walking three hours per week, could substantially lower risk. More active individuals, such as those jogging or running five hours per week, probably cut their risk by more than half, but even modest increases in exercise are likely to be beneficial. Any increase in activity is good and should be encouraged, but the higher the activity level, the greater the reduction in colon cancer is likely to be.

Breast cancer
The relationship between physical activity and breast cancer is unclear. Breast cancer is influenced by estrogens, and exercise appears to affect levels of sex hormones. Investigators have used several approaches for estimating the relation of physical activity to breast cancer risk. Two studies were based on sports participation in college, other studies have used job title to infer workplace levels of activity, and the remaining studies were based on self reported levels of exposure. Frisch and colleagues⁴ provided the first epidemiologic evidence linking physical activity to lower breast cancer risk, comparing the lifetime prevalence of breast cancer in former college athletes and nonathletes. Subjects who did not report participation in college athletics were nearly twice as likely to report a history of breast cancer as women who had been on sports teams in college. Findings suggested that college athleticism defined a group of women who lead physically active lives, but the study was limited by the small number of breast cancer cases reported and its ability to control for other potential factors that could have accounted for the association with breast cancer. A number of studies have attempted to estimate workplace activity levels from job titles. Although these studies tend to support the exercise/breast cancer association, they are likely to be confounded by socioeconomic status, since women in professional and clerical occupations more so than other women tend to have other breast cancer risk factors, such as higher socioeconomic status, fewer children, and later age at first birth.

Four studies were based on direct assessment of physical activity through self-report. Two prospective studies, each based on approximately 120 cases, had inconsistent results. In contrast, three case-control studies, in which exercise earlier in life is compared in women who recently have had a diagnosis of breast cancer with that reported in women without the disease, found that breast cancer patients reported lower physical activity levels prior to their diagnosis. One of these, a study limited to pre menopausal breast cancer of women aged 40 and younger,⁵ found a linear trend toward lower risk with increasing hours per week spent in physical exercise during a woman's reproductive life. Consistent participation in exercise lasting four or more hours per week was associated with a nearly 60 percent reduction in breast cancer risk. Exercise within 10 years after menarche provided similar or slightly diminished protection. The impact of physical exertion on postmenopausal disease, which is much more common than premenopausal breast cancer, could not be addressed in this study. A recent case-control study of women younger than 75 years provides overall support for the observation of a protective effect for regular moderate or strenuous activity early in life but not for low-level activity. However, this study did not collect information on recent activity levels.

While results have not been entirely consistent across studies, overall they have suggested a protective role for physical activity in the prevention of breast cancer. Also, many crucial questions remain, particularly what level of activity is required, whether there is risk reduction for both pre- and postmenopausal cases, and whether the risk reduction occurs primarily for exercise during adolescence. The relationship between activity and breast cancer remains an important area of research.

Prostate cancer
Like breast cancer, prostate cancer is influenced by hormones. Three studies have suggested an inverse relation between physical activity and risk of prostate cancer, two did not find an appreciable association, while two actually found a slight increase in risk with greater levels of activity. The latter two studies assessed physical activity during young adulthood, long before clinical prostate cancer developed. In contrast, a recent study⁶ suggested that physical activity later in life might be related inversely to prostate cancer. However, in this latter study, the association was limited to a small proportion of men who sustained a very high level of physical activity. Overall, epidemiologic data do not strongly support an appreciable association between physical activity and risk of prostate cancer, but the data are limited and further study is needed to confirm this conclusion. Few data concerning the relationship between physical activity and other types of cancer are currently available. Some evidence suggests that physical activity may offer modest protection against lung cancer, although the impact would be very small compared with the risk caused by smoking.

Conclusion
The evidence is overwhelming that physical activity helps protect against colon cancer. Most likely, a benefit exists at moderate levels of activity, such as vigorous walking for three hours a week, but even more benefit is likely at greater levels of activity. Breast cancer is another cancer for which activity may be beneficial, but this is not proven. Prostate cancer does not appear to be related appreciably to exercise, although there is some evidence of a benefit at very high levels of activity. Little is known as to whether exercise protects against other cancers. Much remains to be learned, but even what is known currently about the protective effect of exercise on colon cancer is valuable given that over 100,000 men and women develop this malignancy annually in the US.

Suggestions

Summary Points

Suggested Further Reading

References