The endoscopists provided the appropriate surveillance interval recommendations in 518 patients, 92.8%, 95% CI: 90.4–94.8% of the cohort (93.5% for close view; 92.2% for standard view; p = NS), as compared with pathology based recommendations. Incorrect recommendations were too early by 2.2 ± 0.5 years in the close-view; and either early or late in the standard view. Conclusion: We observed evidence that real-time OD of all colorectal polyps can be applied in patient care. Use of colonoscopes capable of close-up view can lead to increased JQ1 in vitro number of accurate OD. Endoscopists were twice as likely to make an OD of colorectal polyps with
high confidence using colonoscopy with close view, as compared to the conventional standard view. The diagnoses were highly accurate and led to similar surveillance
intervals as compared to those made based on pathology. NCT01288833. Key Word(s): 1. colon polyps; 2. optical diagnosis; 3. resect and discard; 4. narrow band imaging; Table 2. Diagnostic Operating Characteristics of Optical Diagnosis, stratified HIF-1 cancer by Confidence Levels All Polyps, n = 1309 Close View Optical Diagnosisa n = 710 Standard View Optical Diagnosis n = 599 P-value Odds Ratio (95% Cl) a a Univariate logistic regression was performed using conditional random effects assuming a model at a single level of clustering by patient Presenting Author: NING-LI CHAI Additional Authors: EN-QIANG LING-HU Corresponding Author: NING-LI CHAI Affiliations: 301 Hospital Objective: To study the expression of tumorigenesis related stem cell markers Lgr5 and CD44 in different pathological types of intestinal polyps and their clinical predictive significance. Methods: 145 cases of colorectal polyps, adenomas and cancer tissues were obtained by colonoscopy biopsy. Immunohistochemistry was employed to detect the expression of Lgr5 and CD44 to find out their relationship with the colon/rectum cancer occurrence and prognosis. Results: The expression of CD44 DNA ligase in
colon cancer tissue was 95.65%, significantly higher than that in normal mucosa (5%), inflammatory hyperplastic polyps (22.58%), tubular adenomatous polyps (55.26%) and villous polyps (75.76%) (P < 0.05). The expression of Lgr5 in colorectal cancer was up to 95.65%, while it was also negative in normal colorectal tissue and was 16.12% in the inflammatory hyperplastic (P < 0.05). However, the expression rate of Lgr5 in both tubular adenoma (94.73%) and villous polyps (93.94%) were not distinguished different with that of colon cancer (P > 0.05). Conclusion: 1. Lgr5 and CD44 were highly expressed in colorectal cancer tissues which was consistency with the clinical and pathological features; 2. The expression of Lgr5 and CD44 were the notable features to distinguish colorectal cancer tissue cancer with normal intestinal mucosa. 3. The correlation between the expression of Lgr5 and the tumor progression of polyps was closer comparing to CD44.