Prostate specific membrane antigen (PSMA) in urothelial cell carcinoma (UCC) is associated with tumor grading and staging.

INTRODUCTION: Prostate specific membrane antigen (PSMA) has become a target for radionuclide imaging and therapy. Previous studies have shown that the expression of PSMA is not specific to prostate tissue. In this study we examine the expression of PSMA in urothelial cell carcinoma (UCC). METHODS: Immunhistochemical PSMA-staining was performed in 89 UCC samples. PSMA expression in tumor tissue, adjacent healthy tissue and blood vessels was examined. We furthermore analyzed PSMA-mRNA expression in nine human UCC cell lines. We correlated our findings with clinical data regarding recurrence and progression of UCC. RESULTS: UCC tissue showed a significantly higher PSMA expression compared to healthy urothelial tissue (p < 0.001). Non muscle invasive bladder cancer revealed significantly higher PSMA expression compared to muscle invasive bladder cancer (p < 0.05). PSMA expression significantly differed between various T-stages (p < 0.05) and tumor differentiation (p < 0.001). In four human UCC cell lines PSMA-mRNA was detectable. Those patients who suffered recurrence showed a higher rate of PSMA expression but no correlation to recurrence-free survival was evident. Progression of disease correlated significantly with a higher PSMA expression (p = 0.036). CONCLUSIONS: Both UCC tissue and healthy urothelial tissue express PSMA, with significantly higher levels in UCC. We confirmed these findings in human UCC cell lines. In this small first cohort expression of PSMA correlates significant with progression of disease but not with recurrence and recurrence-free survival. These first results make PSMA a promising target for future diagnosis and therapy of UCC.

Prostate cancer tissues with positive TMPRSS2-ERG-gene-fusion status may display enhanced nerve density.

Innervation of prostate cancer (CaP) tissue favors tumor progression and metastasis but the regulation of innervation in CaP is unclear. The oncogenic transcription factor ERG is commonly induced by a typical TMPRSS2-ERG (TE) gene fusion in CaP and may affect innervation. Here, we analyzed whether nerve density of CaP tissue is related to TE status or perineural infiltration status of CaP tissue. In parallel, we measured several members of the neuropilin/plexin/semaphorin family (NRP, PLXN, and SEMA) as possible targets mediating innervation. The TE-gene-fusion status was determined at the mRNA level in CaP tissues by nested RT-PCR. Transcript levels were analyzed by quantitative RT-PCR in CaP tissue or cell line homogenate. ERG was analyzed by immunostaining, and the nerve density was evaluated by immunostaining for PGP9.5 and axonal neurofilament. Data were analyzed by correlation (Spearman), linear regression, Mann-Whitney U test, and contingency table analyses. TE-positive (TE-1) vs. TE-negative (TE-0) CaP tissues displayed significantly enhanced ERG-mRNA levels (TE-0: -4.183; TE-1: -2.994, P < 0.001) and ERG immunostaining (Erg-IH score; TE-0: 0.4211; TE-1: 1.391; P < 0.0001). Notably, the nerve density was significantly increased in CaP tissue samples with positive TE status compared to negative TE status (TE-0, ND score = 1.5; TE-1, ND score = 2.0; P <0.01). NRP1, NRP2, PLXNA2, PLXNB1, SEMA3A, and SEMA4B mRNAs were detectable in CaP tissues and CaP cell lines at quite heterogeneous levels. In CaP tissues, we observed significant positive correlations of ERG with NRP2, PLXNA2, PLXNB1, and SEMA4B. TE-positive CaP tissues displayed enhanced nerve density. ERG correlated with some NRP/PLXN/SEMA components suggesting possible regulatory relevance of ERG for CaP innervation.