| The associations of serum micronutrients including retinol, carotenoids ($alpha$-carotene, $beta$-carotene, and lycopene), vitamin E ($alpha$-tocopherol and $gamma$-tocopherol), and selenium with the risk of colon cancer were examined in the current study. In addition, the relationships between selected serum biochemical indices including cholesterol, total bilirubin, total protein, albumin, as well as uric acid and the risk of colon cancer were also explored. Genetic factors have been proposed to play a greater role in the development of cancer of the proximal colon; while environmental factors, the cancer of the distal colon. Therefore, the anatomic-site specific associations of serum micronutrients and serum biochemical indices with the risk of colon cancer were also of major interest in the current study.;The cohort of the CLUE study, a serum-collection study conducted in Washington County, Maryland in 1974, was used as the study population for the current study. Since 1975, 121 colon cancer cases were identified through the Washington County Cancer Registry. One matched control was selected for each case. Their serum levels of micronutrients and biochemical indices were measured using the serum specimen collected in the CLUE study.;A strong inverse association between serum $alpha$-tocopherol and the risk of colon cancer was observed (odds ratios of colon cancer per 1 mg/dl change in the serum concentration of $alpha$-tocopherol was 0.50; 95 percent confidence interval, 0.27 and 0.91). A weak inverse association with the risk of colon cancer was also suggested for serum $alpha$-carotene and $gamma$-tocopherol (odds ratios of colon cancer per unit change in the serum concentration was 0.93 and 0.73 respectively; 95% confidence interval, 0.83 and 1.05, 0.09 and 5.89 respectively). As to serum retinol, $beta$-carotene, lycopene, and selenium, no association with the risk of colon cancer could be found.;A strong inverse association between serum albumin and the risk of colon cancer was observed (odds ratios of colon cancer per 1 g/dl change in the serum albumin concentration was 0.23; 95 percent confidence interval, 0.07 and 0.76). An inverse association with the risk of colon cancer was also suggested for serum total protein and total bilirubin (odds ratios of colon cancer per unit change in the serum concentration was 0.56 and 0.68 respectively; 95% confidence interval, 0.31 and 1.00, 0.39 and 1.17 respectively). As to serum cholesterol and uric acid, no association with the risk of colon cancer could be found.;All the associations mentioned above were especially prominent among cancer of the proximal colon. Thus, the hypothesis that genetic factors play a greater role in the development of proximal colon cancer was not supported by the current study.;These results support a role for $alpha$-tocopherol in the prevention of colon cancer. As to the meaning of the inverse association between serum albumin and the risk of colon cancer, further studies are needed. |
