A series of meta-analyses have examined relationships between regular physical activity and susceptibility to various forms of cancer. Regular physical activity protects animals against cancer from a variety of sources: subcutaneous, intraperitoneal or intragastric carcinogens, intravenous infusion of tumor cells, or tumor implantation. In humans, regular exercise reduces susceptibility to all-cause cancer, colonic adenomas, colon but not rectal cancers, breast cancers, uterine tumors, prostate and testicular tumors, and possibly lung cancers. At most tumor sites, the average response of women is similar to that of men, but because of a limited number of studies, the effect in women is commonly nonsignificant. The relative effects of occupational and leisure activity are generally similar, an observation that suggests that the optimum response of cancer defense mechanisms is obtained from moderate levels of energy expenditure. In general, the data show a dose-response relationship, the risk of a sedentary lifestyle approximately doubling on passing from a moderate to a low level of habitual physical activity. To date, the findings have not demonstrated the postulated j-shaped relationship, but the cross-sectional comparison of low vs. high levels of leisure activity suggests that the adoption of an active lifestyle could reduce all-cause cancer rates by as much as 46%.