Background Malignant pleural mesothelioma (MPM) is characterized by mutations in several genes, including cyclin-dependent kinase-inhibitor 2A/p16 in the 9p21 locus, BRCA1-associated protein 1 (BAP1), and neurofibromatosis type 2 (NF2) in the 22q12 locus. Recent studies indicate that fluorescence in situ hybridization (FISH) detects hemizygous loss of NF2 in tissue specimens of MPM. The authors investigated whether NF2 FISH, either alone or in combination with other diagnostic assays (9p21 FISH, methylthioadenosine phosphorylase [MTAP] immunohistochemistry [IHC], and BAP1 IHC), effectively distinguishes MPM cells from reactive mesothelial cells (RMCs) in cell blocks prepared from pleural effusions. Methods FISH assays were used to examine the deletion status of NF2 and 9p21, and IHC was used to determine the expression of MTAP and BAP1 in cell blocks from 54 cases with MPM and 18 cases with RMCs. Results Hemizygous NF2 loss (chromosome 22 monosomy or hemizygous deletion) showed 51.9% sensitivity (48.1% for chromosome 22 monosomy and 3.7% for hemizygous deletion) and 100% specificity in differentiating MPM cells from RMCs. Combinations of NF2 FISH, 9p21 FISH, and BAP1 IHC assays yielded greater sensitivity (98.1%) than any assay alone (9p21 FISH, 61.1%; MTAP IHC, 52.8%; or BAP1 IHC, 60.4%). The level of hemizygous NF2 loss in cell blocks positively correlated with that in corresponding tissues. Furthermore, to overcome cytologic specimen-specific challenges, FISH combined with cytokeratin AE1/AE3 immunofluorescence was necessary in 25.9% of MPM cases for FISH assessment of predominantly scattered MPM cells. Conclusions NF2 FISH alone or in combination with other diagnostic assays effectively differentiates MPM cells from RMCs in cell blocks prepared from pleural effusions.
