Comparison of labelling by bromodeoxyuridine, MIB-1, and proliferating cell nuclear antigen in gastric mucosal biopsy specimens

AIMS--To compare proliferating cell nuclear antigen (PCNA) and MIB-1 with bromodeoxyuridine (BrdU) pulse labelling, a specific marker of cell proliferation, in endoscopic gastric biopsy specimens. METHODS--Twenty four biopsy specimens were obtained from 12 patients: 10 antral and eight body specimens were suitable. Each specimen was routinely processed and stained with haematoxylin and eosin. A modified Giemsa stain was used to detect the presence of Helicobacter pylori. Sections of the specimens were labelled with BrdU, MIB-1, and PC10. Gastric mucosa specimens were divided into three zones. The numbers of positively staining nuclei for 500 epithelial cell nuclei were counted in each zone and expressed as a percentage. RESULTS--The proportion of PCNA positive cells (range 0-90%) was much greater in all specimens (10 antrum, eight body). BrdU positive cells were virtually all confined to zone 2 (0-17% cells in this zone were positive) (zone 1 = surface and gastric pit, zone 2 = isthmus, zone 3 = gland base), while PCNA positive cells were present in all three zones (1 = 23-90%, 2 = 43-90%, 3 = 0-74%). Spearman's rank coefficient correlation of 0.57 confirmed that the percentage of positively staining cells varied in the same direction for both PCNA and BrdU (p < 0.001). PCNA, however, was overexpressed in all zones of the gastric epithelium compared with BrdU. In 38 biopsy specimens from 19 patients, of which 14 antrum and 11 body were suitable, the proportion of MIB-1 positive cells (0-59%) was greater than BrdU in most. As with BrdU labelling, the MIB-1 positive cells were confined to zone 2 (zone 1 = 1-11%); zone 2 = 21-59%; zone 3 = 0-13%) and the coefficient correlation for MIB-1 and BrdU was 0.63 (p < 0.001). CONCLUSIONS--MIB-1 accurately reflects the S-phase fraction in gastric mucosa, determined by BrdU labelling in conventionally processed gastric biopsy material. Caution is needed in the interpretation of PCNA labelling detected by PC10, which should not be accepted uncritically as a marker of cell proliferation in paraffin wax embedded material.