The separation of reactive from malignant mesothelial proliferations is often a difficult morphologic problem. There is contradictory information in the literature on whether methylthioadenosine phosphorylase (MTAP) immunohistochemistry can be used for this purpose.
To determine the utility of MTAP immunohistochemistry in distinguishing reactive from malignant mesothelial proliferations.
We stained a tissue microarray containing 20 epithelioid malignant mesotheliomas and 17 reactive mesothelial proliferations. For the mesotheliomas, comparisons were made between MTAP staining and BRCA-associated nuclear protein 1 (BAP1) immunohistochemistry, cyclin-dependent kinase inhibitor 2A ( CDKN2A) fluorescence in situ hybridization, and neurofibromin 2 ( NF2) fluorescence in situ hybridization, which are established techniques for making this separation.
Loss of MTAP was seen in 0 of 17 reactive mesothelial proliferations and 13/20 (65%) malignant mesotheliomas. Almost all cases with loss showed loss in 100% of mesothelial cells. Background inflammatory and stromal cells served as a positive internal control. CDKN2A fluorescence in situ hybridization on the mesotheliomas showed concordance with MTAP staining in 14 of 17 evaluable cases. BAP1 immunohistochemistry showed loss of nuclear staining in 11 of 20 mesotheliomas (55%). No cases showed loss of NF2. A total of 18 of 20 mesotheliomas (90%) showed loss of either MTAP or BAP1.
In the context of a mesothelial proliferation, loss of MTAP staining is 100% specific for malignant mesothelioma. In this study the combination of MTAP and BAP1 immunohistochemical staining allowed separation of reactive from epithelial malignant mesothelial proliferations in 90% of cases.