[Show abstract] [Hide abstract]
ABSTRACT: The plasmacytoid variant of urothelial carcinoma is rare, of which less than 40 cases have been reported in the English language literature. Herein we report the largest series to date of 17 cases of urothelial carcinoma with plasmacytoid features and report the associated clinicopathologic findings. The architectural pattern of the tumor varied from cells arranged in cords and single cells (35%), small nests (17%), solid sheetlike growth (29%), and diffuse discohesive patternless architecture (23%). The plasmacytoid component varied from 15% to 100% of the specimen analyzed; in 12 cases the plasmacytoid component composed greater than 50% of the tumor. The individual tumor cells had striking morphologic overlap with plasma cells with an eccentrically placed nucleus and abundant amphophilic to eosinophilic cytoplasm. The nuclei were of low to intermediate nuclear grade with minimal nuclear pleomorphism. Thirteen of 17 cases (76%) were associated with conventional high-grade invasive urothelial carcinoma and 9 cases showed very focal intracytoplasmic vacuoles mimicking signet ring cells. One case also showed sarcomatoid dedifferentiation. The tumor cells were positive for cytokeratin 7 (94%) and cytokeratin 20 (31%); CD138 was positive in 94% of cases. All cases were invasive -- 7 into at least the lamina propria, 7 into at least the muscularis propria, and 3 into perivesical fat. Follow-up information was available in 16 cases (range: 2 wk to 43 mo; mean 10 mo; median 5.5 mo). Eleven patients died of disease and 5 patients were alive with disease. Plasmacytoid variant of urothelial carcinoma is an aggressive subtype associated with poor prognosis. In limited samples, it may be misdiagnosed as chronic cystitis or plasmacytoma, a pitfall further compounded by CD138 expression. Distinction from metastatic carcinoma from other primary sites such as stomach and breast is critical due to differing therapeutic implications.
The American journal of surgical pathology 12/2008; 33(3):417-24. DOI:10.1097/PAS.0b013e318186c45e · 4.59 Impact Factor