LB-224

Pancreatic cancer risk: associations with meat-derived carcinogen intake

Tuesday, Apr 21, 2009, 1:00 PM - 5:00 PM

Hall B-F, Poster Section 14

14

4

Kristin E. Anderson, Steven J. Mongin, Rashmi Sinha, Myron D. Gross, Timothy R. Church. University of Minnesota - SPH, Minneapolis, MN, University of Minnesota, Minneapolis, MN, National Cancer Institute, Bethesda, MD

Previous epidemiological studies found associations between high-temperature cooked meat (HTCM) and pancreatic cancer (PC). Heterocyclic amines (HCAs) and polycyclic aromatic hydrocarbons, formed in temperature- and time-dependent manners during the cooking of meat, are mutagens and carcinogens. We assessed the association between dietary intake of HCAs, benzo(a)pyrene [B(a)P] and exocrine PC in a large, prospective cohort study. Methods: Subjects randomized to screening in the PLCO (Prostate, Lung, Colorectal and Ovarian) multi-center screening trial completed a dietary survey soon after randomization (Nov 1993 to Sept 2001). Of about 77,500 subjects, 62,581 provided details on meat intake, preferred cooking methods, and doneness preferences. These subjects defined the prospective cohort, and were surveyed annually for all cancer incidence through Dec 2006. The cohort was 50% male, and 91% white, with mean age 63 yrs at entry. Current or former smokers comprised 53%, and 7% had diabetes. The CHARRED meat software was used to estimate HCA and B(a)P intake, and a mutagenic activity index (a summary measure, based on the Ames assay performed on cooked meat samples), from the survey responses on meat intake and cooking preferences. The effects of these preferences and mutagens on risk of PC were estimated as proportional hazard ratios (HRs), and adjusted for age, sex, smoking, education, race, and diabetes. Subject-level non-response was adjusted for by re-weighting in the estimation of mutagen effects. Results: Preference for HTCM was reported for bacon and sausage by 79% of respondents; hamburger, 46%; and steak, 15%. After a median 9 yrs of follow-up, 208 cases of PC were confirmed. Preferences for HTCM were generally associated with increased risk. While none were statistically significant, very well done steak yielded an HR (95% confidence interval) of 1.95 (0.99, 3.84). For specific mutagens, upward trends in risk of PC were observed with increasing quintiles of DiMeIQx and PhIP (P for linear trend: 0.02 and 0.19, respectively). Significantly high HRs were found with the mutagenic activity index, where the 4th and 5th quintiles of exposure (vs. the first) corresponded to HRs (95% CI) of 2.1 (1.3, 3.3) and 1.7 (1.1, 2.9), respectively. Conclusion: Estimation of the effects of meat intake on PC risk may benefit from careful assessment of cooking practices. HTCM, and the resulting production of mutagenic compounds, appear to confer a higher risk of PC.