Comparison of COX-2, Ki-67, and BCL-2 expression in normal esophageal mucosa, Barrett's esophagus, dysplasia, and adenocarcinoma with postablation mucosa and implications for ablative therapies.

backgroundThe neosquamous mucosa that replaces ablated esophageal endothelium after endoscopic mucosal ablation for Barrett's metaplasia or high-grade dysplasia (HGD) may retain buried glandular tissue. This study aimed to assess the neoplastic potential, cellular proliferation, and resistance to apoptosis of this buried glandular tissue by measuring COX-2, Ki-67, and BCL-2 expression in these tissues.

methodsA prospectively collected database was sourced for esophageal biopsy specimens with normal histologic appearance, Barrett's metaplasia, HGD, adenocarcinoma, and postablation mucosa comprising ablated Barrett's and ablated HGD. Quantitative analysis of cellular markers was achieved immunohistochemically using monoclonal antibodies for the COX-2 enzyme (suggesting increased neoplastic potential), Ki-67 antigen (suggesting cellular proliferation), and BCL-2 oncoprotein (suggesting oncogenic resistance to apoptosis). Grading was performed by independent, blinded observers, and the pre- and postablation cellular disparities were subsequently noted.

resultsThe buried glandular elements of postablation mucosa demonstrated universally greater COX-2, Ki-67, and BCL-2 expression than normal esophagus. Barrett's esophagus and adenocarcinoma expressed significantly greater COX-2 and Ki-67 at the deep glandular level than postablation mucosa. HGD demonstrated greater Ki-67 expression than the postablation tissue but only within the superficial glands. Overall, the expression of COX-2 correlated significantly with Ki-67 expression in deep glandular tissue.

conclusionsAblation of pathologic mucosa in Barrett's esophagus and HGD reduces the expression of some markers of neoplastic behavior. However, the buried glandular tissue of the postablation mucosa still exhibits a higher expression than normal esophageal epithelium. This has potential implications for the follow-up treatment of these patients because it is unclear whether the true risk of neoplastic progression is adequately reduced. A more comprehensive study is required to address this issue.