Comprehensive analysis of the FOXA1-related ceRNA network and identification of the MAGI2-AS3/DUSP2 axis as a prognostic biomarker in prostate cancer.

BACKGROUND: Prostate cancer (PCa) is the second most common cause of cancer-related deaths in American men. Even though increasing evidence has disclosed the competitive endogenous RNA (ceRNA) regulatory networks among cancers, the complexity and behavior characteristics of the ceRNA network in PCa remain unclear. Our study aimed to investigate the forkhead box A1 (FOXA1)-related ceRNA regulatory network and ascertain potential prognostic markers associated with PCa. METHODS: RNA sequence profiles downloaded from The Cancer Genome Atlas (TCGA) were analyzed to recognize differentially expressed genes (DEGs) derived from tumor and non-tumor adjacent samples as well as FOXA1(low) and FOXA1(high) tumor samples. The enrichment analysis was conducted for the dysregulated mRNAs. The network for the differentially expressed long non-coding RNA (lncRNA)-associated ceRNAs was then established. Survival analysis and univariate Cox regression analysis were executed to determine independent prognostic RNAs associated with PCa. The correlation between DUSP2 and immune cell infiltration level was analyzed. Tissue and blood samples were collected to verify our network. Molecular experiments were performed to explore whether DUSP2 is involved in the development of PCa. RESULTS: A ceRNA network related to FOXA1 was constructed and comprised 18 lncRNAs, 5 miRNAs, and 44 mRNAs. The MAGI2-AS3~has-mir-106a/has-mir-204~DUSP2 ceRNA regulatory network relevant to the prognosis of PCa was obtained by analysis. We markedly distinguished the MAGI2-AS3/DUSP2 axis in the ceRNA. It will most likely become a clinical prognostic model and impact the changes in the tumor immune microenvironment of PCa. The abnormal MAGI2-AS3 expression level from the patients' blood manifested that it would be a novel potential diagnostic biomarker for PCa. Moreover, down-expressed DUSP2 suppressed the proliferation and migration of PCa cells. CONCLUSIONS: Our findings provide pivotal clues to understanding the role of the FOXA1-concerned ceRNA network in PCa. Simultaneously, this MAGI2-AS3/DUSP2 axis might be a new significant prognostic factor associated with the diagnosis and prognosis of PCa.