Abstract
Objective Diffuse malignant peritoneal mesothelioma (DMPM), represents 30% of all malignant mesothelioma, and is characterised by a difficult diagnosis and different presentations. Immunohistochemistry has improved the diagnostic sensitivity and specificity in the differential diagnosis between metastatic adenocarcinoma and malignant mesothelioma, and loss of BRCA-1-associated protein 1 (BAP1) expression is correlated with BAP1 somatic or constitutional genetic defects. Furthermore, cyclin-dependent kinase inhibitor 2A (CDKN2A) is frequently lost in DMPM. In the present study, we assessed the value of integrating BAP1 in the panel of antibodies used for the diagnosis of DMPM in cytological samples. Since p16 fluorescent in situ hybridisation (FISH) assay could constitute an additional useful adjunct, results of BAP1 immunostaining and p16 FISH assays have been compared. Methods Forty-eight DMPM patients and 71 peritoneal carcinomatosis patients were included. BAP1 immunohistochemical and CDKN2A FISH techniques were performed on tissue specimens of DMPM (n = 48) and peritoneal carcinomatosis (n = 71) then on cell-block of DMPM (n = 16), peritoneal carcinomatosis (n = 25) and peritoneal benign effusion (n = 5). Results Loss of BAP1 expression was observed in 56.3% of DMPM while none of the peritoneal carcinoma specimens showed BAP1 loss of expression. CDKN2A loss was observed in 34.9% DMPM and 2.1% peritoneal carcinoma. Although BAP1 immunostaining was successful in 100% of cytological DMPM samples, CDKN2A deletion status could be obtained for 75% of DMPM cases. Conclusion BAP1 immunostaining represents an objective and reproducible diagnostic biomarker for peritoneal mesothelioma in effusion cytology specimens and should be preferred to CDKN2A FISH analysis on these precious samples.