The prognostic value of Dickkopf-3 (Dkk3), TGFB1 and ECM-1 in prostate cancer

Abstract

Prostate cancer (PCa) is considered one of the most common cancers worldwide. Despite advances in patient diagnosis, management, and risk stratification, 10%-20% of patients progress to castration-resistant disease. Our previous report highlighted a protective role of Dickkopf-3 (DKK3) in PCa stroma. This role was proposed to be mediated through opposing extracellular matrix protein 1 (ECM-1) and TGF-β signalling activity. However, a detailed analysis of the prognostic value of DKK3, ECM-1 and members of the TGF-β signalling pathway in PCa was not thoroughly investigated. In this study, we explored the prognostic value of DKK3, ECM-1 and TGFB1 using a bioinformatical approach through analysis of large publicly available datasets from The Cancer Genome Atlas Program (TGCA) and Pan-Cancer Atlas databases. Our results showed a significant gradual loss of DKK3 expression with PCa progression (p < 0.0001) associated with increased DNA methylation in its promoter region (p < 1.63E-12). In contrast, patients with metastatic lesions showed significantly higher levels of TGFB1 expression compared to primary tumours (p < 0.00001). Our results also showed a marginal association between more advanced tumour stage presented as positive lymph node involvement and low DKK3 mRNA expression (p = 0.082). However, while ECM1 showed no association with tumour stage (p = 0.773), high TGFB1 expression showed a significant association with more advanced stage presented as advanced T3 stage compared to patients with low TGFB1 mRNA expression (p < 0.001). Interestingly, while ECM1 showed no significant association with patient outcome, patients with high DKK3 mRNA expression showed a significant association with favourable outcomes presented as prolonged disease-specific (p = 0.0266), progression-free survival (p = 0.047) and disease-free (p = 0.05). In contrast, high TGFB1 mRNA expression showed a significant association with poor patient outcomes presented as shortened progression-free (p = 0.00032) and disease-free survival (p = 0.0433). Moreover, DKK3, TGFB1 and ECM1 have acted as immune-associated genes in the PCa tumour microenvironment. In conclusion, our findings showed a distinct prognostic value for this three-gene signature in PCa. While both DKK3 and TGFB1 showed a potential role as a clinical marker for PCa stratification, ECM1 showed no significant association with the majority of clinicopathological parameters, which reduce its clinical significance as a reliable prognostic marker.

Keywords: Dickkopf-3; ECM-1; TGF-β signalling pathway; prognosis; prostate cancer.

FIGURE 1

The mRNA expression of DKK3 in normal versus prostate cancer samples CANCERTOOL database. Violin plots show the expression of DKK3 in non-tumoral tissue (N) and primary tumours (PT) in various datasets of the CANCERTOOL database.

FIGURE 2

The differential DNA methylation levels of DKK3 promoter between normal and primary prostate tumours.

FIGURE 3

Boxplots of the mRNA expression of DKK3, ECM1, TGFB1 and PSA (KLK3) in a large patient cohort using the TNM plot database.

FIGURE 4

The association between DKK3, ECM1 and TGFB1 mRNA expression and clinicopathological features of prostate cancer. (A) The association between DKK3 and prostate cancer clinicopathological characteristics using data from prostate adenocarcinoma (TCGA) through cBioPortal tool. (B) The association between ECM1 and TGFB1 mRNA expression prostate cancer clinicopathological characteristics using data from prostate adenocarcinoma (TCGA) through cBioPortal tool.

FIGURE 5

The association between DKK3, ECM1 and TGFB1 mRNA expression and patient survival. Log-rank test p-values were used to evaluate the statistical significance between the low and the high expression level.

FIGURE 6

Correlation of (A) DKK3, (B) ECM1 and (C) TGFB1 gene expression level with immune cell infiltration in PCa. The scatterplots displayed Spearman's rho value correlation and statistical significance showing the potential interplay between DKK3, TGFBI and ECMI biomarkers and immune cells (B-Cell, CD8+ T Cell, CD4+ T Cell, Macrophage, Neutrophil and Dendritic Cell) in the prostate cancer tumor microenvironment. The log2 TPM gene expression values are presented on the y-axis, the average immune cell infiltration levels are presented on the x-axis. The blue curve and gray area in the figures show the general trend direction. TPM: transcripts per million. (Data generated from TIMER Webtool).

FIGURE 7

Gene-Gene interaction network including DKK3, TGFB1 and ECM1. It shows interaction strength (edge thickness), interaction type (colour), multiple edges between nodes, and protein score (node size) defined using a stylesheet constructed with GeneMANIA. The interconnections between studied genes were evaluated based on physical interaction, co-expression, predicted, co-localization, common pathway, genetic interaction and shared protein domains.

FIGURE 8

Gene ontologies enrichment analysis (A) Biological Process and (B) Molecular Functions of DKK3, TGFBI and ECMI using Enricher database.