P282 Despite surveillance and increased use of biologics and immunosuppressives colorectal cancer detection in IBD patients remains unchanged over the past ten years

Abstract

Inflammatory bowel disease (IBD) is considered a risk factor for colorectal cancer (CRC). Patients with IBD are at higher risk of CRC compared with normal population and therefore undergo surveillance colonoscopy according to national guidelines which have changed over the past 10 years. Globally, the cumulative risk has generally reduced over the years and it is likely due to better management of IBD and robust surveillance practices. We aimed to study impact and outcomes of surveillance in our IBD patients with changing surveillance guidelines. We reviewed all patients with IBD undergoing surveillance during a 10-year period from 2007 to 2017 in our tertiary care centre. Patients with histological reports confirming any grade of dysplasia (low grade, high grade) or CRC were included. Lesions indefinite for neoplasia were excluded. All electronic clinical records were reviewed and relevant data compiled using the endoscopy, histology, and radiology databases. Patients were also analysed as a sub-group if found to have CRC. Eighty-three patients (M = 58) were diagnosed with dysplasia or cancer (median age 65 years range 34–87 years) between 2007–2017. Eighty (97%) patients had ulcerative colitis (UC) and 40 (48%) had pancolitis. Seventy-five patients had identifiable lesions at endoscopy (polypoid = 35; non-polypoid = 11; other findings including pseudopolyps = 29). 76.2 % of lesions were in the left colon and rectum. Fifty-two patients underwent surgery based on surveillance findings, the majority having a panproctocolectomy. Of 28 patients (M = 20) diagnosed to have CRC in this cohort, 20 adenocarcinomas were diagnosed on histology of biopsies at colonoscopy. Of 28 CRC, 26 were associated with UC, 2 with Crohn’s disease, and 5 had primary sclerosing cholangitis. 70% of the CRC were in left colon and rectum. Only 32% of patients had surveillance within 5 years of cancer detection and majority had white light endoscopy (WLE) with random biopsies; 4 CRC patients had previous surveillance within 1 year. 29% had active disease on endoscopy. Over the 10-year period there was no change in the detection rate of dysplasia or CRC. In our cohort of patients, male gender and UC were associated with higher risk of CRC. Our data show high interval cancer risk highlighting that WLE with random biopsies are likely to miss early dysplastic lesions. There was no change in detection rate of CRC in IBD patients over 10-year period. Unless surveillance techniques and protocols are rigorously optimised, the risk of interval CRC should be kept in mind when implementing guidelines regarding surveillance frequency.