Increased expression of proliferative Ki-67 nuclear antigen is correlated with dysplastic colorectal epithelium in ulcerative colitis.

Because patients with ulcerative colitis have an increased long-term risk of colorectal cancer, colonoscopic surveillance with multiple biopsies is commonly performed for histopathological detection of dysplasia to select high-risk patients for prophylactic colectomy. Improved differentiation between neoplastic vs. nonneoplastic changes is needed because active inflammation may cause significant misinterpretation of nonneoplastic reactive/regenerative changes in the epithelium. We investigated whether the expression of proliferative antigens is correlated with various degrees of epithelial dysplasia and inflammatory changes in biopsy specimens from patients with long-standing ulcerative colitis. Colorectal biopsy specimens from patients undergoing colonoscopic surveillance were analyzed immunohistochemically using two types of monoclonal antibodies: MIB-1 against Ki-67 and NCL-PCNA against proliferating cell nuclear antigen for structural, active inflammatory, and dysplastic changes. Specimens from patients without inflammatory bowel disease or neoplasia were used as controls; these showed no increased proliferation. However, increased staining with the MIB-1 monoclonal antibody was detected in 9% of the specimens from patients with long-standing ulcerative colitis without active inflammation or dysplasia; this was significantly more common in specimens indefinite for dysplasia, probably positive (24%), and in those with definite dysplasia of low (47%) or high grade (67%; P = 0.008). For increased PCNA staining, there was a non-significant correlation (P = 0.30) with increasing degrees of dysplasia. Increased MIB-1 immunostaining was found in 50% and increased PCNA immunostaining in 75% of the specimens displaying mild inflammation. Both antibodies had a 100% increased staining in specimens with moderate or severe inflammation. Increased proliferation as expressed by MIB-1 is thus better correlated with increasing degree of dysplasia than is PCNA. Neither staining method is able to differentiate neoplastic from inflammatory epithelial changes. However, in the absence of active inflammation, immunostaining for MIB-1 may be a valuable adjunct in the confirmation of dysplastic epithelial changes in long-standing ulcerative colitis, particularly in the indefinite changes category.