PTH-047 Outcomes of colonoscopic surveillance and molecular phenotyping in patients with family history of colorectal cancer. (June 2019)

• Record Type: Journal Article
• Authors: Norton, Benjamin; Coultas, James; Nicholas, Victoria; Wright, Thomas; Monahan, Kevin
• Is Part Of: Gut. Volume 68(2019)Supplement 2
• Page Start: A35
• Page End: A36
• Publication Date: 2019-06
• Subjects: Gastroenterology -- Periodicals; 616.33
• DOI: 10.1136/gutjnl-2019-BSGAbstracts.72
• Languages: English
• ISSNs: 0017-5749

Abstract

Abstract : Introduction: An estimated 35% of cases of colorectal cancer (CRC) are due to heritable factors, and approximately 30% of the UK population has a family history of CRC. Those at hereditary risk should be effectively managed through registration, phenotypic and genotypic characterisation, and risk-stratified colonoscopic surveillance. We assessed the impact of surveillance in patients at hereditary risk of CRC managed through the Family History of Bowel Cancer Registry at West Middlesex University Hospital (WMUH). Through analysis of this registry data, we assessed the diagnostic yield of colonoscopic surveillance and assessed the role of molecular testing. Methods: We analysed prospectively collected colonoscopic surveillance data in 361 patients undergoing surveillance at WMUH between 2010–2018. Patients were divided into five risk groups based on current BSG guidelines. Patient demographics including age, gender and family history were collated alongside colonoscopy findings and molecular data including mismatch repair (MMR) status. Impact of these variables on the prevalence of non-advanced adenomas (NAAs) and advanced adenomas (AAs) were assessed by logistic regression using SPSS software. Time to adenoma or AA was determined by survival analysis and findings were compared between index and surveillance colonoscopy. Results: In total, 640 colonoscopies were performed with 1000.1 years of follow-up. Five CRCs, 49 AAs and 170 NAAs were identified. The prevalence of CRC, AA and NAAs in patients without Lynch syndrome (LS) was 0.58%, 8.19% and 26.02%, respectively. Only age was significantly associated with both NAA and AA detection on multivariable analysis (P<0.05). Time to AA was earlier in LS patients (figure 1). A normal index colonoscopy was strongly associated with a normal finding during surveillance (p<0.001). In 38.6% of patients, molecular testing significantly altered surveillance strategies by recategorising familial risk. Conclusion: This data emphasises the strong association of colorectal neoplasia with MMR status and the need to exclude LS in patients at familial risk. Age is independently associated with colorectal neoplasia risk in this analysis, however patients often undergo colonoscopic surveillance inappropriately early. Finally, a normal index colonoscopy is associated with a low diagnostic yield in subsequent colonoscopies in this population. This low yield may influence future guideline strategies.