HBME-1 immunostaining in reactive mesothelial versus metastatic adenocarcinoma cells in serous fluid

Background: The cytological diagnoses of serous effusions are usually made by routine cytomorphology with certainty. However, overlapping cases sometimes exist between reactive mesothelial and adenocarcinoma cells. We tried to evaluate the diagnostic utility of HBME-1 in distinguishing between reactive mesothelial cells and adenocarcinoma in serous effusions. Materials and Methods: Fifty-two cytologic specimens processed by cell-block technique were retrieved from the archive and were assigned to two groups: Group I: 26 effusions containing reactive/benign mesothelial cells; Group II: 26 effusions containing carcinoma cells. Immunostaining with HBME-1 was performed using an Envision technique. The staining intensity of cells, according to proportion of immunoreactive cells, was scored as: 0 (negative), 1+ (<10%), 2+ (10-50%), and 3+ (≥50%); and the predominant staining pattern of positive cells were determined. Statistical analysis and tests of significance were performed using SPSS software. Results: The calculated mean values (in percentile) for stained cells in adenocarcinoma and reactive mesothelial cells were 25.57 and 67.88, respectively ( P = 0.001). Thin membranous, thick membranous, thick and thin membranous, cytoplasmic and combined patterns of staining in adenocarcinoma cells were respectively 4 cases (21.1%), 0 case, 0 case, 8 cases (42.1%), and 7 cases (36.8%), and in reactive mesothelial cells, these were 7 cases (26.9%), 1 case (3.8%), 18 cases (69.2%), 0 case, and 0 case, respectively. The intensity of staining in majority (88.5%) of reactive mesothelial cells was scored 3+, but the distribution varied in the other group. Conclusions: The staining pattern and intensity for HBME-1 is a useful panel for differentiation of adenocarcinoma and mesothelial cells.