A213

OBJECTIVE: To examine the association between cigarette smoking, second-hand smoke (SHS) exposure, and colorectal cancer risk. METHODS: Patients who attended Roswell Park Cancer Institute between 1982 and 1997 and completed an epidemiological questionnaire were eligible for this hospital-based case-control analysis. Approximately 1,203 cases of primary, incident colorectal cancer were frequency matched to 2,406 malignancy-free controls on age (5-year intervals), gender, and year of questionnaire completion. Logistic regression was used to calculate colorectal cancer risk by various tobacco smoke exposure measures while controlling for relevant covariates (body mass index, residence, race, education, income, family history, vegetable intake, meat intake, other tobacco use, alcohol use, and aspirin use). RESULTS: Overall, those who had more that 40 pack-years of exposure had a small, statistically significant increase in colorectal cancer risk (adjusted OR = 1.23; 95%CI = 1.02-1.48). The risk for those with more than 40 pack-years rose when individuals with current SHS exposure were removed from the analysis (adjusted OR = 2.12; 95%CI = 1.42-3.15), as a positive trend for increasing risk for increasing pack-years was also noted. Risk estimates were generally higher for males compared to females. When SHS exposure was the main predictor variable, never smoking males exposed to 7 or more hours a day of SHS had an increased colorectal cancer risk that was not statistically significant (adjusted OR = 1.58; 95%CI = 0.93-2.69), although the risk estimate for females was near the null value. CONCLUSIONS: Long-term, heavy smokers had a slight, statistically significant increase in colorectal cancer risk. Current SHS exposure may be an independent risk factor for colorectal cancer and failure to account for this exposure may result in exposure misclassification.

[Fifth AACR International Conference on Frontiers in Cancer Prevention Research, Nov 12-15, 2006]