However, the role of chronic inflammation has not yet been fully identified. Our aim was to determine the effect disease activity on the risk of lymphoma among UC patients unexposed to immunomodulators. Methods: Nationwide data was obtained from the Veterans Affairs healthcare system 2001–2011. We performed a retrospective Fulvestrant supplier cohort study following UC patients unexposed to immunomodulators from the date of UC diagnosis to the date of lymphoma

development. UC and lymphoma patients were identified by ICD9 codes using validated algorithms supplemented by chart review. Disease activity was assessed using the rate of steroid utilization. Steroids users were classified into three equal groups according to the annual cumulative dose of steroids tertiles and were compared to steroids non-users. Multivariate cox regression analysis was performed to account for other confounding factors. Results: we included 10,780 patients with median follow-up time of 8 years, 3,441 (32%)

used steroids. We identified 34 cases of lymphoma. The incidence rate of lymphoma was 0.4, 0.3, 0.6, and 0.7 per 1000 person-years for non-users, low, intermediate, and high AZD9668 in vitro steroid users respectively. Using Cox regression analysis the age-, sex- and race-adjusted hazard ratio for lymphomas was 0.85, 1.66, and 2.23 (non-significant, Table 1) respectively as compared to non-steroid users. Conclusion: There is a non-significant trend towards increased risk of lymphoma with increased disease activity as measured by the amount of steroid use in the absence of immunomodulator therapy. Key Word(s): 1. Lymphoma; 2. Ulcerative Colitis; 3. Inflammation; 4. Steroids; Table 1 Results of the multivariate Cox regression analysis     no p/y events IR HR LCI UCI p Age 1 year increment         1.03 MCE 1.00 1.05 0.08 Notes: p/y: person year of follow up, IR: Incident rate per 1000 person year, HR: Hazard Ratio, UCI and LCI: upper and the lower limits of the 95% confidence interval respectively. Presenting Author: LICHUAN FENG Corresponding Author: LICHUAN FENG Affiliations: Third Hospital

of Peking University Objective: to observate endoscopic feature of dysplasia and canceration related to ulcerative colitis (UC) Methods: conclude the endoscopic manifestation of UC patients with dysplasia and canceration in Third Hospital of Peking University from 2005 to 2013. Results: 1. epidemiology: there were 869 UC patients who had colonoscopy in the same period and 68 patients (7.8%)had dysplasia and canceration which concluded 44 men and 24 women, average age was 39 ± 2.34 years old. 2. degree of dysplasia: there were 52 mild dysplasia, 12 moderate dysplasia, 2 severe dysplasia and 2 early adenocarcinoma. the percentage of mild dysplasia was higher than other groups (P < 0.05)3. location: 52 patients’dysplasia happened on ulcer and erosion and 16 patients on polyp, uneveness and uplift. there was signifcant diferrence between two groups (P < 0.05). 4.