Confocal Endomicroscopy in Patients With High Risk of Colorectal Cancer: Potential for Diagnosis of Early Neoplasia

The principle objective of this study is to validate confocal endomicroscopy (CEM) in a national, multicenter study, in terms of its ability to diagnose neoplastic lesions in vivo, in two groups of patients at high risk of colorectal cancer (CRC): patients with familial adenomatous polyposis (FAP) after colectomy in whom the neoplastic lesions are probably under-diagnosed, and patients with inflammatory bowel diseases (IBD) in whom endoscopic surveillance is particularly difficult. Methods: The study will be comprised of two phases (Phase I and II). Phase I will serve to validate at the multicenter level the results of the first, recently published, monocenter German study in terms of capacity of CEM to identify the colonic neoplastic lesions in vivo. Phase II is destined to prospectively evaluate the diagnostic yield of CEM in detection and prediction of neoplastic lesions by developing and adding new features to the confocal pattern of in vivo diagnosis. Two cohorts of patients will be studied in parallel: Patients with inflammatory bowel diseases (IBD), like ulcerative colitis (UC) or Crohn's disease (CD), including those before planned colectomy, and patients with FAP after colectomy. During lower endoscopy performed under general anaesthesia, each colonic segment will be examined before and after staining with indigo-carmin. After intra-venous fluorescein injection, all macroscopically abnormal lesions will be examined by CEM, then biopsied. In parallel, multiple random biopsies will be performed, each coupled with simultaneous CEM "optical biopsy" at the same point. In addition, during Phase II, in IBD patients before planned colectomy and in patients with FAP, a "mapping" of colonic mucosa, by obtaining a very high number of CEM "optical biopsies", will be performed, and will be correlated with standard histology performed either on colectomy specimens (IBD) or on standard biopsies (FAP). Principal analysis (Phase I and II) will include evaluation of inter-observer variation in terms of interpretation of in vivo histology and diagnostic yield of CEM with respect to the detection of neoplastic lesions by evaluation of sensitivity and specificity, using standard histology as reference method. Additional analysis (Phase II) will be performed to evaluate the diagnostic and predictive (CRC risk) value of "colonic mapping" by correlating optical images pattern score to results of standard histology. Expected results: This study should guarantee high quality data, standardization of procedures and of interpretation of CEM images, which are prerequisite for dissemination of CEM in clinical practice. The investigators expect to show that CME allows to reliably discriminate between neoplastic and non-neoplastic lesions, that, compared to standard histology, provides better characterization of lesions, especially in the context of extended lesions like in IBD, an finally, that CME images can be used to develop a new "optical biopsy"-based score allowing prediction of high CRC risk in patients with FAP and IBD. The investigators believe that CEM may increase, as compared to currently used techniques, the diagnostic yield in terms of probability of the detection of neoplastic lesions in patients at high risk of CCR, and may become a new standard for endoscopic surveillance in these patients.