FCGR3A: A new biomarker with potential prognostic value for prostate cancer

Abstract

To screen target gene cluster by bioinformatics analysis and verify them by in vitro experiment and clinicopathological correlation analysis. We try to find a new biomarker with prognostic value for prostate cancer (PCa). 42 candidate marker genes were constructed by protein protein interaction (PPI) network and enriched by KEGG pathway to find out the gene cluster we are interested in. Prognostic model was established to preliminarily analyze the prognostic value of this gene cluster in PCa, and Cox risk regression was used for comparative analysis. Immunohistochemistry was used to detect the expression of each gene in clinical tissue microarray. Finally, we analyzed the correlation between each gene and their clinicopathological features of PCa combined with TCGA clinical data. Based on the analysis of PPI and KEGG, we found the target gene cluster (FCGR3A, HAVCR2, CCR7 and CD28). Prognostic model analysis showed that this gene cluster had the ability to predict biochemical recurrence, and the survival rate and ROC analysis showed favorable prediction effect. Univariate Cox regression analysis showed that the risk scores of Gleason score (GS), T stage, N stage and PSA were significantly different (P<0.05), and the risk ratio of high expression was 2.30 times that of low expression (P=0.004). However, it was not statistically significant in multivariate Cox regression analysis (P>0.05). The results of tissue microarray showed that FCGR3A and HAVCR2 were highly expressed in PCa (P<0.01), while the expression of CCR7 and CD28 had no significant difference (P>0.05). Kaplan-Meier analysis showed that there was significant difference in BCR free survival of FCGR3A and HAVCR2 (FCGR3A, P=0.010; HAVCR2, P=0.018), while the expression of CCR7 and CD28 had no significant difference on the survival and prognosis of PCa patients (P>0.05). TCGA clinical data analysis found that the expression of FCGR3A had a unique correlation with the clinicopathological features of PCa, which was closely related to the tumor stage. The expression of FCGR3A is related to BCR free survival of PCa patients. Therefore, FCGR3A is a new biomarker with potential prognostic value of PCa.

Keywords: FCGR3A; bioinformatics analysis; prognosis; prostate cancer; tumor microenvironment.

Figure 1

PPI network construction and KEGG pathway enrichment analysis. (A) Protein-protein interaction network for 42 candidate marker genes in String database. (B) The top 10 pathways of KEGG pathway enrichment analysis.

Figure 2

Prognostic model analysis of entire set. (A) Distribution of 4 target genes in low-risk and high-risk groups in TCGA-PRAD dataset. (B) BCR free survival curve of 4 target genes in TCGA-PRAD dataset (P=0.0012). (C) 5-year ROC curve of 4 target genes in TCGA-PRAD dataset (AUC>0.6).

Figure 3

BCR free survival analysis of 4 target genes. (A) BCR free survival curve of FCGR3A in TCGA-PRAD dataset (P=0.0104). (B) BCR free survival curve of HAVCR2 in TCGA-PRAD dataset (P=0.0180). (C) BCR free survival curve of CCR7 in TCGA-PRAD dataset (P=0.2864) (D). BCR free survival curve of CCR7 in TCGA-PRAD dataset (P=0.6035).

Figure 4

Immunohistochemical staining for 4 target genes expression in PCa and normal prostate tissue samples. (A) Full view of the immunohistochemistry staining for 4 target genes expression. (B) (no significance, ns P>0.05) Immune risk score (IRS) of 4 target genes in PCa and normal prostate tissue samples (Cancer vs. Non-Cancer, ^** P<0.01). (C) 4 target genes expression in PCa tissue samples with different tumor stages and Gleason score (GS).