High serum selenium and reduced risk of advanced colorectal adenoma in a colorectal cancer early detection program.

BACKGROUND: Epidemiologic and animal studies suggest that selenium may reduce risk of colorectal cancer. However, the epidemiologic data is mainly from relatively small investigations, limiting their interpretation. Although substantial evidence suggests that smoking is a strong effect modifier for other antioxidative nutrients, little is known about smoking-selenium interactions in colorectal tumors. METHODS: We studied the association of serum selenium and advanced colorectal adenoma, a cancer precursor, in 758 cases and 767 sex- and race-matched controls, randomly selected from the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial. Cases had at least one verified advanced adenoma (> or = 1 cm or villous elements, or high-grade dysplasia) of the distal colon, and controls had a negative sigmoidoscopy. RESULTS: The multivariable odds ratio (OR) comparing participants in the highest quintile of serum selenium with those in the lowest quintile was 0.76 [95% confidence interval (95% CI), 0.53-1.10; P(trend) = 0.01]. The inverse association between serum selenium and advanced colorectal adenoma was significant among recent smokers (OR, 0.53; 95% CI, 0.27-1.01 for highest versus lowest tertile; P(trend) = 0.008). Serum selenium was unrelated to adenoma risk in nonsmokers and former smokers who quit smoking > or = 10 years ago. CONCLUSION: Selenium may reduce the risk of developing advanced colorectal adenoma, particularly among the high-risk group of recent smokers.

Comorbidity independently predicted death in older prostate cancer patients, more of whom died with than from their disease.

OBJECTIVE: The purpose of this study was to examine the proportion of men who died from prostate cancer (PrCa) vs. with PrCa and assess the comorbid conditions associated with other-cause deaths. STUDY DESIGN AND SETTING: We identified all male decedents aged >/=65 years in Jefferson County, AL, in 1993-1995. By crosslinking three databases (death certificate, Medicare, and Veteran's Administration), we identified men whose deaths might have been caused by PrCa. We abstracted and reviewed medical records to rate comorbid conditions and determine whether or not death was due to PrCa. RESULTS: Of 561 men with a premortem diagnosis of PrCa, 42% died from PrCa and 53% died with PrCA; 50.2% of blacks died from PrCa vs. 36.9% of Whites. Other factors related to dying with PrCa included older age at death and a serious, or very serious, comorbid condition. Treatment did not have an independent effect on cause of death (i.e., death with vs. from PrCa). CONCLUSIONS: Comorbidity was an independent predictor of dying with PrCa, even after adjustment for ethnicity, age, and treatment. Given the as yet unproven benefit of PrCa screening, our results extend the body of information relevant to the screening decision; among men dying with a diagnosis of PrCa, only about 1/3 to 1/2 died from the disease.