P16 loss and mitotic activity predict poor survival in patients with peritoneal malignant mesothelioma.

PURPOSE: Peritoneal malignant mesothelioma is an aggressive neoplasm for which intensive therapy improves survival in a subset of patients. We hypothesized that pathologic variables would stratify patients into favorable and unfavorable survival subgroups. EXPERIMENTAL DESIGN: Fifty-four patients with peritoneal malignant mesothelioma were evaluated for trimodal therapy from 1995 to 2003. Two pathologists evaluated pathologic variables independently, and p16 status was analyzed by immunohistochemistry. RESULTS: Patients not receiving trimodal therapy had a significantly increased risk of death [hazard ratio (HR), 9.6; 4.3-21.6; P < 0.0001]. Biphasic histology was also associated with increased risk of death (HR, 8.5; 3.4-21.8; P < 0.0001). In multivariate analysis adjusting for treatment modality and histologic type, high mitotic rate and p16 loss were associated with increased risk of death (HR, 3.074; 1.05-9.0; P < 0.04 and HR, 3.65; 1.3-10.2; P < 0.014, respectively). CONCLUSIONS: Biphasic histology, increased mitotic rate, and p16 loss were independently associated with poorer survival in peritoneal malignant mesothelioma. Among the trimodal treated patients, increased mitotic rate was associated with increased risk of death.

Usefulness of CDX2 and TTF-1 in differentiating gastrointestinal from pulmonary carcinoids.

Carcinoids of different organs appear morphologically indistinguishable. We studied the usefulness of differential expression of CDX2 and thyroid transcription factor-1 (TTF-1) in 78 gastrointestinal and pulmonary carcinoids and their metastases (n = 10). CDX2 staining of gastric biopsy specimens with neuroendocrine hyperplasia (n = 11) and various gastritides (n = 10) was also performed. All ileal (6/6 [100%]), 6 (86%) of 7 appendiceal, 3 (75%) of 4 duodenal, 1 (50%) of 2 ampullary, 12 (33%) of 18 rectal, 6 (30%) of 20 pancreatic, and 1 (17%) of 6 gastric carcinoids expressed CDX2 with variable intensity; none of the pulmonary carcinoids stained. Of 15 pulmonary carcinoids, 8 (53%) stained with TTF-1, but none of the gastrointestinal carcinoids did. CDX2 and TTF-1 staining profiles of primary and metastatic carcinoids were similar. CDX2+ gastric endocrine cells had a distribution similar to that of gastrin and enterochromaffin cells but not enterochromaffin-like cells. Our results suggest that CDX2 and TTF-1 have high specificity for gastrointestinal and pulmonary carcinoids, respectively.

Utility of CD138 (syndecan-1) in distinguishing carcinomas from mesotheliomas.

CD138 (Syndecan-1) is a transmembrane heparan sulfate proteoglycan present on the surface of plasma cells and epithelial cells. CD138 is also expressed in some hematopoietic neoplasms and has recently been observed in carcinomas. We characterized CD138 expression in cell-block preparations of fluids/effusions, focusing on the distinction between carcinoma and mesothelioma. One hundred formalin-fixed, paraffin-embedded cell-block sections of fluids/effusions consisting of 58 metastatic carcinomas, 24 mesotheliomas, 11 reactive mesothelial cell proliferations, 3 lymphomas, 3 metastatic sarcomas, and 1 metastatic melanoma were stained with a monoclonal antibody against CD138. CD138 staining was observed in 32/58 (55%) metastatic carcinomas and 2/24 (8%) mesotheliomas; all reactive mesothelial cells, lymphomas, sarcomas, and melanoma were negative. CD138 is a highly specific marker in the differential diagnosis of carcinoma vs. mesothelioma. Distinct membranous staining without background staining of most inflammatory cells makes CD138 an ideal marker for cell-block preparations of fluids/effusions. It should be an integral component of the epithelial-mesothelial antibody panel.