AGA Abstracts T1981 Plasma Levels of Insulinotropic and Digestive Hormones Predict the Risk of Colonic Adenomas Martin J. Wolff, Neal Joseph, Maryann L. Huie, Sergio Quijano, Nora Muakkassa, Frank Martiniuk, Guillermo I. Perez-Perez, Fritz Francois Background: Obesity is associated with an increased risk of colorectal cancer. Derangements in gut hormones that regulate energy balance and metabolism may influence epithelial growth. The aim was to evaluate insulinotropic and digestive hormones in an average risk colorectal cancer (CRC) screening population. Methods: Consecutive patients referred for CRC screening at a municipal hospital were enrolled, and a detailed medical evaluation including height and weight was obtained prior to colonoscopy. A total of 15cc of blood was collected and plasma concentrations of total amylin, acylated ghrelin, glucose-dependent insulinotropic peptide (GIP), active glucagon-like peptide-1 (GLP-1), insulin, leptin, pancre- atic polypeptide (PP), and peptide YY (PYY) were measured using a multiplex immunoassay. Based on histologic confirmation of colonoscopic findings, cases (CAs) were defined as individuals with any adenomas regardless of size, including those with villous histology, high-grade dysplasia, or adenocarcinoma. Controls (COs) were selected from individuals without any colonic lesions and matched to the cases based on age, gender, ethnicity, and body mass index. Results: A total of 108 CAs, 56 men and 52 women, (mean age 61 ± 7 years, BMI 26.7 ± 4) were enrolled and matched to 108 COs. The study cohort was composed of Hispanics (58%), Asians (25%), African Americans (14%), and Caucasians (3%). Reflecting fat stores, plasma leptin was significantly correlated to BMI in CAs (r=0.92, p<0.001) and COs (r=0.77, p<0.001). Plasma level of active GLP-1 was markedly elevated in CAs (1130 pg/ml, IQR 548-1610) compared to COs (315 pg/ml, IQR 171-789); p<0.001. Similarly total amylin was significantly elevated in CAs (median 496 pg/ml, IQR 135-754) compared to COs (median 214 pg/ml, IQR 75-471); p<0.001. Although the prevalence of diabetes did not differ between the two groups (25% vs 21%), plasma insulin was lower in CAs (380 pg/ml, IQR 180-566) compared to COs (455 pg/ml, IQR 331-621); p<0.001. Addition- ally, among women active ghrelin was significantly elevated in CAs (31 pg/ml, IQR 19-49) compared to COs (16 pg/ml, IQR 8-39); p=0.005. Among men, leptin was also significantly elevated in CAs (4235 pg/ml, IQR 1992-6617) compared to COs (2710 pg/ml, IQR 1225- 5245); p=0.047. Conclusions: In this ethnically diverse average risk CRC screening popula- tion insulinotropic peptides differentiated individuals with and without colon pathology independent of diabetes. The association of leptin and ghrelin with colon adenomas differs according to gender. Further studies are warranted on the role of gut hormones involved in energy homeostasis and colon cancer development. T1982 Alcohol Stimulates Activation of Snail, Epidermal Growth Factor Receptor Signaling, and Markers of Epithelial-Mesenchymal Transition in Colon and Breast Cancer and Normal Intestinal Epithelial Cells Christopher B. Forsyth, Yueming Tang, Maliha Shaikh, Lijuan Zhang Even moderate alcohol use can increase the risk of several cancers including colon and breast cancer by mechanisms that have not been fully identified. Epithelial-mesenchymal transition (EMT) is a cellular program established to play a role in developmental morphogen- esis. EMT results in cells losing their stationary epithelial characteristics and becoming more migratory/invasive with a phenotype expressing mesenchymal proteins such as vimentin. Recent data supports a role for EMT in cancer initiation, progression, and metastases. Snail is one transcription factor sometimes called the master regulator of EMT. We hypothesized that alcohol might promote cancer through stimulation of the EMT gene program through stimulation of Snail activity and transcription. To test this hypothesis we determined the effect of alcohol stimulation on cellular phenotypic markers of EMT as well as activation and nuclear localization of Snail and Snail transcription. We also attempted to identify key signaling pathways stimulated by alcohol resulting in activation of Snail. Human colon and breast cancer cell lines and a normal intestinal epithelial cell line were tested. Cells were treated with alcohol and assessed for EMT-related changes using immunofluorescent micro- scopy, western blotting, reporter assays, RT-PCR, and knockdown of Snail with siRNA. Our data show physiological alcohol (0.2%, 24h) upregulated the signature EMT phenotypic marker vimentin in as well as MMP-2, MMP-7, and MMP-9 also characteristic of the EMT proteome and important in cancer metastases. In addition alcohol stimulated cancer cell migration, also a characteristic of Snail expression and EMT. Timecourse experiments showed alcohol stimulated nuclear localization of activated Snail phosphorylated at Ser246 within 30 min in all four cell lines as well as transcription from a Snail reporter plasmid (6h), and Snail gene expression by RT-PCR (24h). Knockdown of Snail expression with siRNA pre- vented alcohol-stimulated vimentin expression and thus EMT. Supporting these data In Vivo Snail expression was significantly elevated in colonic mucosal biopsies from active alcoholics (IRB approved). Finally, in terms of pathways stimulating Snail, we found alcohol stimulated activation of epidermal growth factor receptor (EGFR) signaling (a known inducer of EMT) and EGFR inhibition blocked the alcohol-stimulated migration of cancer cells as well as Snail transcription. These data support a novel mechanism for alcohol promoted carcinogen- esis through the EMT program via a Snail-mediated mechanism that could be targeted for therapy or prevention of alcohol-related cancers. T1983 Cigarette Smoking and Risk of Colorectal Cancer in the European Prospective Investigation Into Cancer and Nutrition (EPIC) Anke M. Leufkens, Fränzel J. van Duijnhoven, Peter D. Siersema, Alina Vrieling, Antonio Agudo, Inger T. Gram, Elisabete Weiderpass, Bas B. Bueno-de-Mesquita Background: Active smoking has consistently been shown to be related to an increased risk of colorectal adenomas. However, the evidence for a relationship with colorectal carcinoma (CRC) is inconsistent. Furthermore, findings from previous studies were inconclusive on whether the colon or the rectum is more sensitive to the effect of smoking. Aim: To investigate the relationship between cigarette smoking and the risk of CRC overall and by tumor A-614 AGA Abstracts localization (proximal or distal colon, or rectum) in particular. Methods: We used data from the European Prospective Investigation into Cancer and Nutrition (EPIC), a multicenter prospective cohort study to investigate the relationship between diet, nutritional and meta- bolic characteristics, various lifestyle factors and the risk of cancer. Individuals were enrolled between 1992 and 2000 in 23 collaborating centers in 10 European countries. We studied data from 465,876 participants of whom 2,855 developed CRC during follow-up (mean 8.7 years). Cox proportional hazard regression analysis, stratified by age, gender and center, was used to estimate hazard ratios (HR) and 95% confidence intervals (95%CI) for the effect of smoking status (never, former and current) on the occurrence of CRC. All analyses were adjusted for the consumption of fruit, vegetables, red meat, processed meat, (shell) fish, intake of total fiber and energy from fat and non-fat, body mass index, physical activity, educational level, weight and height. Results: CRC was located in the colon in 1,802 cases, of which 774 were proximal, 772 distal and 256 not specified, whereas in 966 cases, it was located in the rectum. Former cigarette smokers had stopped smoking for a mean of 15.2 years (at least 0.5 years) and were shown to be at an increased risk of CRC compared to never smokers (HR 1.18, 95%CI 1.07-1.29). Risk for current cigarette smoking was not significantly increased (HR 1.10, 95%CI 0.98-1.22). When we stratified for tumor location, both former (HR 1.24, 95%CI 1.04-1.49) and current smokers (HR 1.34, 95%CI 1.08-1.66) were at increased risk of cancer in the proximal colon, whereas no relationship was found with the distal colon or rectum. Conclusions: Smokers are at increased risk of colon cancer, particularly in the proximal colon, whereas this is not the case for cancer in the distal colon or rectum. The effect of smoking intensity, duration and pack-years of smoking, and time since quitting smoking needs further elucidation. T1984 The Nutritional Role of the Microbiota in Populations with a High and Low Risk of Colon Cancer Junhai Ou, Tsutomu Fukuwatari, Deborah O'Connor, Stephen J. O'Keefe Recent advances in our quantitation of the mass and diversity of the human microbiota have revealed the importance of mutualism between humans and microbes in preventing disease. Microbes interact with diet and host factors to produce vitamins and short chain fatty acids, essential nutrients for mucosal health. In order to determine whether differences in the rate of production of these nutrients could explain differences in colon cancer risk, we compared their content in colonic evacuants from samples of the high risk (65:100,000 population) African American (AAs) and low-risk (<1:100,000) native African (NAs) popula- tions. Methods: Following a 12h fast, groups of 16 healthy 50-65 year old AAs and NAs were given 2L Golytely solution to drink over 30 minutes at 8am. For 3hours, all stools were collected, weighed, homogenized, sampled and frozen at -200C to await analysis. Fecal fluid was extracted and analyzed for SCFAs by GC and folate and biotin were measured by microbiological assay using Lactbacillus rhamnoses for folate, and L. plantarum ATCC 8014 for biotin. Pooled samples (n=12 per group) were also assayed for total thiamin and vitamin C with HPLC, and vitamin B12 was assayed by the competitive solid phase immune method. Results(Table)showed that the colonic content of major SCFAs (acetate, butyrate, and pro- prionate) were significantly higher (p<0.05) in NAs. Colonic content of folate, biotin and B12 were similar in both groups and similar to or higher than usual dietary intakes (RDA), while vitamin C and thiamine were considerably lower (0.347 and 1.36mg vs 75-90mg/day (RDA), and 0.002 and 0.002mg vs 1.5mg/d respectively). In conclusion, the low risk of colon cancer in native Africans might be partly explained by the high production of butyrate by the microbiota. As patients were fasted, and dietary vitamin absorption is predominantly in the small intestine, the detection of large quantities of folate and biotin in colonic contents likely represents de novo microbiotal synthesis. This might also play a role in the suppression of colon cancer in impoverished communities as recent studies have demonstrated active colonic vitamin absorption, and both these vitamins are involved in epigenetic regulation of epithelial proliferation. *=p<0.05, Mann Whitney NP test T1985 Cigarette Smoking and the Risk of Colorectal Cancer: A Meta-Analysis of Prospective Cohort Studies Kelvin K. Tsoi, Yee Yu Pau, William K. Wu, Francis K. L. Chan, Sian M. Griffiths, Joseph J. Sung Background. Smoking has been implicated in many malignant diseases but its association with colorectal cancer (CRC) is controversial. Aim. We quantitatively evaluated the relation between smoking and incidence of CRC in a meta-analysis of cohort studies. Methods. Full publications of prospective cohort studies were identified in MEDLINE and EMBASE from 1950 to 2008. Subjects were classified as either current smokers, former smokers or non- smokers. The quantity of smoking was assessed by number of cigarettes per day, years of smoking, and pack-years. The reported relative risks of CRC were pooled by random-effects model. Sensitivity analysis was conducted and publication bias evaluated. Results. A total of 1,463,796 subjects were recruited in 28 prospective cohorts from America, Europe and Asia with median follow-up of 13 years (4 - 30 years). Current smokers showed a modestly higher risk of CRC (RR: 1.20, 95% CI: 1.10-1.30) than non-smokers. The risk of CRC among male smokers (RR: 1.38, 95% CI: 1.22-1.56) was more significant than female smokers (RR: 1.06, 95% CI: 0.95-1.19). Rectal cancer is more closely related to smoking (RR: 1.36, 95% CI: 1.15-1.61) than colonic cancer. Former smokers still carry a higher CRC risk than non-smokers. The increase risk of CRC is related to cigarettes per day, longer years of smoking, or larger pack-years. Conclusion. Smoking significantly increased the risk