Immunohistochemical expression of prostatic antigens in adenocarcinoma and villous adenoma of the urinary bladder.

Adenocarcinomas of the bladder are rare, with the diagnosis dependent on exclusion of secondary involvement by direct extension or metastatic spread from other sites. The recent description of an unusual form of urothelial-type mucinous prostatic adenocarcinoma raises a novel differential diagnosis between adenocarcinomas of the prostate and bladder, and investigation into the utility of classic prostatic immunohistochemical antigens in bladder adenocarcinoma is warranted. We identified 37 primary infiltrating adenocarcinomas of the bladder, which included signet ring cell carcinomas (n=11), urachal adenocarcinomas (n=5), and enteric adenocarcinoma (n=21). Also included for comparison were 3 cases, each of bladder villous adenomas and bladder adenocarcinoma in situ. Tissue microarrays were constructed from each case, with each specimen represented by multiple 1.0-mm cores to assess for tumor protein heterogeneity. Immunohistochemistry for prostate-specific antigen (PSA), prostate specific acid phosphatase (PSAP), P501S (prostein), and prostate specific membrane antigen (PSMA) was performed, and moderate to strong immunoreactivity was considered a positive result. Of the 37 adenocarcinomas, all were negative for PSA and PSAP (0/37; 0%). In contrast, a minority of bladder adenocarcinomas was labeled with the prostate antigens P501S and PSMA. P501S showed moderate diffuse cytoplasmic staining in 4/37 cases (11%), including 3 enteric-type adenocarcinomas and 1 mucinous adenocarcinoma. Additionally, 1 case of adenocarcinoma in situ demonstrated diffuse cytoplasmic staining for P501S. The granular perinuclear staining pattern of P501S typically seen in prostatic adenocarcinoma was absent in all cases of bladder adenocarcinoma. PSMA showed diffuse cytoplasmic staining in 4/37 (11%) infiltrating adenocarcinomas (including 1 signet ring carcinoma and 3 enteric-type adenocarcinomas), and in 1 case of adenocarcinoma in situ. Membranous PSMA staining was evident in an additional 3 tumors, 1 urachal mucinous adenocarcinoma, 1 nonurachal mucinous and signet ring cell adenocarcinoma, and 1 nonurachal villous adenoma. In conclusion, although all cases of bladder adenocarcinoma examined were negative for PSA and PSAP, the surprising finding that a subset of invasive and in situ adenocarcinomas of the bladder demonstrated immunoreactivity for P501S and PSMA should warrant caution when using these markers in differentiating prostatic from bladder adenocarcinomas. The lack of granular perinuclear staining for P501S and the absence of membranous PSMA staining both favor a bladder adenocarcinoma, although rare cases of villous adenoma and adenocarcinoma did show PSMA membranous staining indistinguishable from that seen in prostate cancer. Although the novel antigens P501S and PSMA are fairly specific and more sensitive in the differential diagnosis of prostate and urothelial carcinoma, care must be taken when adenocarcinomas of the bladder are considered within this differential diagnosis.