BAP1 immunohistochemistry has limited prognostic utility as a complement of CDKN2A (p16) fluorescence in situ hybridization in malignant pleural mesothelioma.

BRCA-associated protein 1 (BAP1) immunohistochemistry (IHC) and CDKN2A (p16) fluorescence in situ hybridization (FISH) have shown clinical utility in confirming the diagnosis of malignant pleural mesothelioma (MPM), but the role for using these 2 markers to guide clinical management is not yet clear. Although p16 loss is predictive of poor prognosis, there is controversy as to whether BAP1 loss is predictive of a more favorable prognosis; how these results interact with one another has not been explored. We performed CDKN2A FISH on a previously published tissue microarray on which we had performed BAP1 IHC, revealing combined BAP1/p16 status for 93 MPM cases. As expected, BAP1 IHC in combination with CDKN2A FISH resulted in high sensitivity (84%) and specificity (100%) for MPM, and p16 loss was an independent predictor of poor survival (hazard ratio, 2.2553; P = .0135). There was no association between BAP1 loss and p16 loss, as 26%, 28%, 30%, and 16% of overall cases demonstrated loss of BAP1 alone, loss of p16 alone, loss of both BAP1 and p16, or neither abnormality, respectively. Although multivariate analysis demonstrated that BAP1 IHC is not an independent predictor of prognosis, when viewed in combination with homozygous CDKN2A deletion, risk stratification was evident. More specifically, patients with CDKN2A disomy and loss of BAP1 expression had improved outcomes compared with those with CDKN2A disomy and retained BAP1 expression (hazard ratio, 0.2286; P = .0017), and this finding was notably evident among epithelioid cases. We conclude that BAP1 IHC provides prognostic information within the context of CDKN2A FISH that may have clinical utility beyond diagnosis.

BAP1 facilitates diagnostic objectivity, classification, and prognostication in malignant pleural mesothelioma.

BRCA-associated protein 1 (BAP1) has emerged as a promising biomarker for malignant pleural mesothelioma (MPM). Loss of BAP1 expression can occur by a variety of mechanisms, but reports on incidence are variable and the clinical significance is unclear. In order to investigate the diagnostic and prognostic significance of BAP1, we constructed a tissue microarray consisting of 111 MPM cases and performed BAP1 immunohistochemistry. BAP1 was lost in 77% of epithelioid cases (n=58) but was retained in all sarcomatoid cases (n=10); 49% of biphasic cases showed loss (n=43), and BAP1-negative cases demonstrated loss of staining in both the epithelioid and sarcomatoid components. All non-neoplastic mesothelial tissues (n=20) retained BAP1, resulting in a sensitivity, specificity, positive predictive value, and negative predictive value of 61%, 100%, 100%, and 32%, respectively. Moreover, BAP1 expression in spindled mesothelium enabled discrimination of reactive and malignant cells, thus providing a more objective means of distinguishing epithelioid from biphasic morphology compared to histology alone. Nonetheless, BAP1 staining was patchy in some benign mesothelial neoplasms, which raises concern for using BAP1 in small biopsies. Kaplan-Meier analysis demonstrated a significant improvement in overall survival with BAP1 loss, but this did not reach significance in multivariate analysis accounting for histologic subtype. When only epithelioid cases were analyzed there was a trend toward increased survival, but it did not reach significance. We conclude that BAP1 loss is frequent in epithelioid MPM, which is in turn associated with improved survival, and that it can have additional clinical significance by facilitating histologic classification.