When colorectal cancer (CRC) is detected at an early stage, the 5-year survival exceeds 95%. Although colonoscopy is an excellent screening tool and considered the current gold standard, there is a miss rate for polyps as high as 22%. In particular, """"""""flat lesions"""""""" in the colon are more commonly missed and may be more likely to contain areas of dysplasia. The problem is further exacerbated in ulcerative colitis (UC), in which dysplasia can develop in macroscopically normal-appearing mucosa. Thus, there is a need to develop novel technologies that would permit the early detection and in situ characterization of early neoplastic colonic lesions with high sensitivity and specificity. The overall goal of this proposal is to clinically translate novel imaging agents and devices we have developed to address this unmet need. Specifically, we will utilize a class of """"""""smart"""""""" agents that increase their near infrared (NIR) fluorescence after selective interaction with a target protease (cathepsin) that is overexpressed in colonic adenomas and adenocarcinomas. Utilizing this technology, our preliminary studies have demonstrated superior endoscopic detection of preneoplastic lesions in mouse models of colon cancer when compared to conventional white light examinations. Moreover, we have observed comparable sensitivity and specificity of this technique for neoplasia that arises in the background of chronic UC. We seek to optimize and characterize this agent in new mouse models that spontaneously develop focal colonic adenomas and adenocarcinomas of known age and location. Cathepsin protease expression will be correlated with lesion progression in these mouse models as well as in a broad spectrum of ex vivo human neoplastic lesions.. The culmination of this effort will be a pilot clinical trial in which the feasibility and diagnostic performance of this novel technology will be evaluated in patients with sporadic invasive CRC, patients with polyposis syndromes, and patients with dysplasia in the setting of UC.
Showing the most recent 10 out of 590 publications
