Purpose: The association between physical activity and overall and site-specific cancer risk is elaborated in relation to whether any observed dose-response association between physical activity and cancer can be interpreted in terms of how much physical activity (type, intensity, duration, frequency) is needed to influence site- and gender-specific cancer risk.

Methods: Observational studies were reviewed that have examined the independent effect of the volume of occupational physical activity (OPA) and/or leisure time physical activity (LPA) on overall and site-specific cancer risk.

Results: The evidence of cohort and case-control studies suggests that both leisure time and occupational physical activity protect against overall cancer risk, with a graded dose-response association suggested in both sexes. Confounding effects such as diet, body weight, and parity are often included as a covariate in the analyses, with little influence on the observed associations. A crude graded inverse dose-response association was observed between physical activity and colon cancer in 48 studies including 40,674 colon/colorectal cancer cases for both sexes. A dose-response effect of physical activity on colon cancer risk was especially observed, when participation in activities of at least moderate activity (>4.5 MET) and demonstrated by activities expressed as MET-hours per week. An observed inverse association with a dose-response relationship between physical activity and breast cancer was also identified in the majority of the 41 studies including 108,031 breast cancer cases. The dose-response relationship was in particular observed in case-control studies and supported by observations in cohort studies when participation in activities of at least moderate activity (>4.5 MET) and demonstrated by activities expressed by MET-hours per week. This association between physical activity and breast cancer risk is possibly dependent on age at exposure, age at diagnosis, menopausal status and other effect modifiers, e.g., body mass index. Furthermore, data concerning carcinoma of other cancers (prostate, lung, endometrium, ovary, and testicular cancers) are required.

Conclusion: A protective effect of physical activity on site-specific cancer risk with a dose-response association between physical activity and colon and pre- and postmenopausal breast cancer supported by identified biological mechanisms has been observed. The optimal permutation of type, intensity, duration, and frequency of physical activity across the lifespan is unclear, but it is gender, age, and site specific and supports moderate activity (>4.5 MET) more than light activities (<4.5 MET). The complicated nature of the physical activity variable, combined with lack of knowledge regarding possible biological mechanisms operating between physical activity and cancer, warrants further studies including controlled clinical randomized trials.