Patterns of sialyl-Lewis X expression predict gastric histopathology.

Abstract

Introduction: Gastric cancer develops through a series of pre-cancerous changes over decades of chronic inflammation. Chronic atrophic gastritis (CAG) represents a critical transition in the progression to gastric cancer, though validated histologic markers are needed to more accurately detect and assess the extent of CAG. We previously identified sialyl-Lewis X (sLe^x) as a marker of atrophic gastric epithelium in mice. Here, we establish patterns of sLe^x expression that can be used to detect and distinguish human gastric pre-cancerous lesions.

Methods: We obtained gastric corpus and/or antrum biopsies from 149 adult patients with dyspepsia. Biopsies were stained with hematoxylin/eosin and a commercially available antibody to sLe^x. Histologic diagnoses included normal, chronic non-atrophic gastritis (CNG), or CAG with or without intestinal metaplasia (IM) and were determined by a single pathologist. A second pathologist graded each biopsy according to consensus criteria, based on the presence, intensity, and glandular distribution of sLe^x staining. Log-linear models were used to determine the association between patterns of sLe^x expression and gastric pathology.

Results: The majority of patients (70%) had gastric pathology (CNG or CAG ± IM). The presence of sLe^x could be used to detect gastric pathology (97% sensitivity), and the absence of sLe^x staining could reliably predict normal histology (76% specificity). The intensity of sLe^x staining significantly correlated with gastric pathology. Moreover, a deeper (≥ 50%) glandular sLe^x distribution in the antrum was significantly associated with CAG, while a more superficial (< 50%) distribution significantly correlated with CNG.

Conclusion: Patterns of sLe^x expression can be used to detect and refine the histologic assessment of gastric pre-neoplastic lesion severity.