Is Ki-67 a better proliferative marker in the colon than proliferating cell nuclear antigen?

Endogenous markers of proliferating cells have increasingly supplanted the use of incubation of biopsy tissues in vitro with tritiated thymidine or with bromodeoxyuridine, thus avoiding the potential variation resulting from the incubation procedure. Antibodies to proliferating cell nuclear antigen (PCNA) such as PC10 have been promoted as optimal for this purpose, although considerable variation in colonic proliferating cells with this antibody has been reported. We have compared the detection of colonic proliferating cells in normal mucosa and adenomata using the PC10 monoclonal antibody (mAb) to PCNA and the Mib-1 mAb to Ki-67 in formalin-fixed tissues using antigen retrieval solutions with microwaving. The PC10 antibody showed variable immunostaining of proliferating and nonproliferating cells with minor changes in primary antibody concentration or microwave conditions and between normal and adenomatous tissue. In contrast, Mib-1 immunostaining was quite constant with differing antigen retrieval and antibody conditions and similar staining of proliferating cells in colonic adenomas. Some loss of immunoreactivity occurred if the cut sections were not immunostained within approximately 1 week. These data suggest that whereas PCNA immunohistochemistry is satisfactory when carefully controlled in large chemopreventive studies, the Mib-1 mAb to Ki-67 is superior to PCNA antibodies in immunostaining proliferating cells in the formalin-fixed human colon.