Application of immunohistochemistry to the diagnosis of malignant mesothelioma.

CONTEXT: The diagnosis of malignant mesothelioma (MM) is rendered with the aid of immunohistochemistry to demonstrate the presence of "mesothelial," "epithelial," or "sarcomatous" differentiation. Antibody panels that have been proposed for the distinction between MM and other neoplasms usually include 2 or more epithelial markers used to exclude the diagnosis of a carcinoma, such as monoclonal and polyclonal carcinoembryonic antigen, Ber-EP4, B72.3, CD15, MOC-31, thyroid transcription factor 1, BG8, and others, and 2 or more mesothelial markers used to confirm the diagnosis of MM, such as cytokeratin 5/6, calretinin, HBME-1, thrombomodulin, WT-1, mesothelin, D2-40, and podoplanin. In general, most antibody panels provide excellent sensitivity and specificity for the differential diagnosis between MM epithelial variant and adenocarcinoma, particularly of lung origin. However, the accuracy of these markers is lower for the diagnosis of sarcomatous MM and for the differential diagnosis between MM and squamous cell carcinoma and carcinomas of renal, ovarian, and other origin.
OBJECTIVE: To identify optimal antibody panels for the diagnosis of MM. DATA SOURCES: Literature review to determine how many and which mesothelial and epithelial markers need to be included in differential diagnosis antibody panels.
CONCLUSIONS: Various antibody panels have been recommended for the diagnosis of MM, with no overall consensus about how many and which markers should be used. A recent study with Bayesian statistics has demonstrated that the use of many markers does not provide higher diagnostic accuracy than the use of selected single antibodies or various combinations of only 2 markers. There is a need for the development of evidence-based or consensus-based guidelines for the diagnosis of MM in different differential diagnosis situations.