Prognostic Value of Vascular-Expressed PSMA and CD248 in Urothelial Carcinoma of the Bladder

Abstract

Background: Urothelial carcinoma of the bladder (UCB) is a common cancer of the urinary system. Despite substantial improvements in available treatment options, the survival outcome of patients with advanced UCB is unsatisfactory. Therefore, it is necessary to identify new prognostic biomarkers for monitoring and therapy guidance of UCB. In recent years, prostate-specific membrane antigen (PSMA) and CD248 have been identified promising candidate bio7markers.

Methods: In this study, we first examined PSMA and CD248 expression in tissues from 124 patients with UCB using immunohistochemical and immunofluorescent staining. We then analyzed the association between the expression of the two biomarkers and other clinicopathological features and prognosis. Finally, we performed bioinformatic analysis of CD248 and FOLH 1 (PSMA) using the TCGA-BLCA dataset to explore the underlying mechanism of PSMA and CD248 in the progression of UCB.

Results: Among the 124 cases, PSMA and CD248 were confirmed to be expressed in tumor-associated vessels. Vascular PSMA and CD248 expression levels were associated significantly with several deteriorated clinicopathological features. Furthermore, using univariate and multivariate Cox analyses, high vascular PSMA and CD248 expression levels were observed to be associated significantly with poor prognosis in patients with UCB. As risk factors, both PSMA and CD248 expression showed good performance to predict prognosis. Furthermore, combining these vascular molecules with other clinical risk factors generated a risk score that could promote predictive performance. Bioinformatic analysis showed that both PSMA and CD248 might contribute to angiogenesis and promote further progression of UCB.

Conclusion: Both PSMA and CD248 are specifically expressed in the tumor-associated vasculature of UCB. These two molecules might be used as novel prognostic biomarkers and vascular therapeutic targets for UCB.

Keywords: CD248; PSMA; angiogenesis; prognosis; urothelial carcinoma of the bladder.

Figure 1

PSMA expression in the tumor-associated vasculature of UCB and the comparison with clinicopathological parameters. (A) Representative IHC staining to show different PSMA expression levels in HCC, using CD31 as the positive control. Case 1: No PSMA expression (score = 0), Case 2: Positive PSMA expression in ≤50% of the tumor-associated vasculature (score = 1), and Case 3: Positive PSMA expression in >50% of the tumor-associated vasculature (score = 2). (B) Comparison between PSMA expression and the Ki-67 index. (C) Comparison between PSMA expression and invasive stage. (D) Comparison between PSMA expression and metastasis. (E) Comparison between PSMA expression and clinical stage. Scale bar = 200 μm. Representative images are shown. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 2

CD248 expression in the tumor-associated vasculature of UCB and the comparison with clinicopathological parameters. (A) Representative IHC staining to show different CD248 expression levels in HCC, using CD31 as the positive control. Case 1: No CD248 expression (score = 0), Case 2: Positive CD248 expression in ≤50% of the tumor-associated vasculature (score = 1), and Case 3: Positive CD248 expression in >50% of the tumor-associated vasculature (score = 2). (B) Comparison between CD248 expression and the Ki-67 index. (C) Comparison between CD248 expression and invasive stage. (D) Comparison between CD248 expression and metastasis. (E) Comparison between CD248 expression and clinical stage. Scale bar = 200 μm. Representative images are shown. *P < 0.05; ns, no statistically significant.

Figure 3

The correlation between the expression levels of CD248 and PSMA in UCB vessels. (A) Representative IHC staining in serial paraffin sections of three cases with UCB, using CD31 as the positive control. Scale bar = 100 μm. (B) The correlation between the expression levels of CD31, PSMA, and CD248 in UCB, respectively, using data from the GEPIA database. (C) Representative triple-immunofluorescence histochemistry for CD31 (red), PSMA (pink), and CD248 (green) in serial paraffin sections of three cases with UCB. Scale bar = 250 and 50 μm, respectively. Representative images are shown.

Figure 4

Prognostic value of vascular CD248 and PSMA expression in UCB. (A, B) Kaplan–Meier survival curve showing the association between vascular PSMA (A)/CD248 (B) expression and overall survival of patients with UCB. (C) ROC curves for PSMA and CD248 expression. (D, E) Kaplan–Meier survival curve showing the association between (D) PSMA-/(E) CD248-related risk score and overall survival of patients with UCB. (F, G) The distribution of the (F) PSMA-/(G) CD248-related risk score and the survival status of each patient. (H) ROC curves of PSMA expression and the PSMA-related risk score. (I) ROC curves of CD248 expression and the CD248-related risk score. (J) The flow direction of the final outcome of patients with different expression levels of PSMA and CD248 in a Sankey plot. Nomogram of the vascular-PSMA-based signature for survival probability prediction (K) and risk of death prediction (L). Nomogram of the vascular-CD248-based signature for survival probability prediction (M) and risk of death prediction (N).

Figure 5

Bioinformatic analysis of CD248 and PSMA using TCGA-BLCA dataset. Kaplan-Meier curve of CD248 (A), PSMA (B), and the endothelial cell score (C). Correlation between the endothelial cell score and CD248 expression (D). Correlation between the endothelial cell score and PSMA expression (E). Heat map of DEGs (F). Co-expressed heatmap of top 20 Cor-DEGs (G). Green to red spectrum indicates low to high gene expression. GO enrichment analysis of Cor-DEGs (H). Heat map of DETFs (I). Forest graph of PCor-DEGs (J). The red and green dots represent PCor-DEGs with a hazard ratio >1 and ≤1, respectively. TFs-based regulatory network for PCor-DEGs (K). P < 0.05 or FDR < 0.05 was considered statistically significant.