Epidemiological studies have established obesity as a risk factor for colorectal neoplasia, and this association is stronger for men than for women. In addition, physical activity has been shown to reduce the odds for colon but not rectal cancer. Because body size and physical activity are closely associated, it is difficult to elucidate which of these factors is more important in carcinogenesis or if the combined effect is most relevant. Evidence for the inverse association between physical activity and colorectal neoplasia is strong; however, the relationship between activity levels and preneoplasia is not fully understood. The aim of this study was to determine whether levels of physical activity affect metachronous colorectal neoplasia in the pooled study populations of two prospective, randomized clinical intervention trials of colorectal neoplasia: the Wheat Bran Fiber Trial and the Ursodeoxycholic Acid Trial. Logistic regression models were used in the analysis of 2,189 men and women to examine the association between quartiles of total energy expenditure per day and metachronous colorectal neoplasia. Odds ratios and 95% confidence intervals were obtained using the lowest quartile of total energy expenditure per day as the reference group. Compared to the first quartile of physical activity level, the adjusted odds ratios for metachronous colorectal neoplasia were 1.04 (95% CI: 0.80, 1.34), 1.00 (95% CI: 0.76, 1.34), and 0.88 (95% CI: 0.65, 1.20), for the second, third, and fourth quartiles, respectively. Stratification by sex yielded odds ratios of 0.88 (95% CI: 0.60, 1.30), 0.77 (95% CI: 0.53, 1.12), and 0.71 (95% CI: 0.48, 1.04) for males (P for trend = 0.048) and odds ratios of 1.15 (95% CI: 0.79, 1.66), 1.85 (95% CI: 1.08, 3.16), and 1.34 (95% CI: 0.53, 3.37) for females (P for trend = 0.061), respectively. Preliminary analyses indicate no statistically significant association between physical activity levels and metachronous colorectal neoplasia. However, among men there was a statistically significant inverse trend in the odds of metachronous colorectal neoplasia and quartile of total, daily energy expenditure. Additional analyses will assess the affects of different types of physical activity on metachronous colorectal neoplasia and whether these activity patterns vary across race and ethnicity. Further, the potential effect modification by sex must be explored in detail to ascertain whether body size, physical activity, or a combination of the two is the key exposure variable related to colorectal preneoplasia risk in this population.

Second AACR International Conference on the Science of Cancer Health Disparities— Feb 3–6, 2009; Carefree, AZ