Mesotheliomas with deciduoid morphology - A morphologic spectrum and a variant not confined to young females

Deciduoid mesotheliomas are rare with only four previously reported cases, all affecting the peritoneum of young females. We describe another six cases (three men and three women; age range 52-65 yrs, median 55 yrs; five peritoneal and one pleural). Three patients had an occupational history of asbestos exposure. The deciduoid appearance predominated in four cases, whereas in two it represented a minor component within conventional tubulopapillary epithelioid mesothelioma. All tumors were strongly cytokeratin-positive (including CK5/6) and all showed at least focal staining for thrombomodulin, HBME-1, and calretinin. All were negative for epithelial mucin (D/PAS), CEA, BerEP4, LeuM1 (CD15), CD21, CD35, and S100 protein, Five of six cases (83%) were vimentin-positive and two (33%) were focally positive for alpha-smooth muscle actin. A differential diagnosis of gastrointestinal autonomic nerve tumor (GANT) had been initially considered from the morphology of one case, and we found positivity for some of the "neuronal" markers described in GANTs. This prompted us to apply such a panel to the other five tumors, accepting that the cytokeratin positivity encountered in all of our cases would exclude GANT. All cases of deciduoid mesothelioma (100%) were positive for PGP 9.5 and NSE and four of six (66%) were positive for NKI/C3. Weak focal staining (<5% cells) for syn aptophysin was seen in two of six tumors. All cases were chromogranin-negative. All cases examined by electron microscopy showed desmosomes and smooth microvilli without rootlets but no neurcendocrine granules. In conclusion, a de ciduoid morphology appears to be part of the histopathologic spectrum encountered in epithelioid mesothelioma. This variant is not confined to female patients and occurs over a wider age range than previously recognized. The overlapping immunophenotype with GANTs illustrates that caution should be exercised when interpreting positivity for "neuronal" markers in this context. An immunohistochemical panel that includes cytokeratins should always be used.