Diagnostic cytopathology 2016 April 28 [Epub ahead of print] [Link]
Walts AE, Hiroshima K, McGregor SM, Wu D, Husain AN, Marchevsky AM.
BACKGROUND: Loss of BAP1 by immunohistochemistry (IHC) and CDKN2A(p16) deletion by fluorescence in situ hybridization (FISH) have been proposed to distinguish malignant mesothelioma (MM) from atypical reactive mesothelial proliferations (ARMP) in effusions but it is uncertain whether both tests are needed routinely.
METHODS: Paraffin embedded blocks from 67 effusions (32 MM, 35 ARMP) were evaluated with BAP1 IHC. 38 of them (17 MM, 21 ARMP) were also analyzed with CDKN2A (p16) FISH. Criteria for MM were absence of BAP1 nuclear staining in >50% of atypical mesothelial cells in the presence of a positive internal control, and/or CDKN2A(p16) homozygous or hemizygous deletion pattern in >15% or >41.5% of atypical mesothelial cells, respectively. Sensitivity (SS), specificity (SP), positive, and negative predictive values (PPV, NPV) for MM were calculated.
RESULTS: 17 of 32 MM were correctly diagnosed by IHC (SS 53.1%, SP 85.7%, PPV 77.3%, NPV 64.3% for MM). In the subset of 38 cases studied by IHC and FISH, the SS, SP, PPV, and NPV of IHC for MM were 41.2, 81.0, 63.6, and 63.0%, respectively. 7 of 17 MM were diagnosed by FISH (SS 41.2%, SP 100%, PPV 100%, NPV 67.7% for MM).
CONCLUSION: BAP1 IHC and CDKN2A(p16) FISH yield similar sensitivities for MM in paired samples but FISH has a higher specificity. To diagnose effusions with atypical mesothelial cells, use of both tests is optimal. However, BAP1 IHC alone may be sufficient to diagnose MM when clinical, imaging, and cytological findings strongly suggest MM.