Identification of Spindle and Kinetochore-Associated Family Genes as Therapeutic Targets and Prognostic Biomarkers in Pancreas Ductal Adenocarcinoma Microenvironment

Abstract

Aim: The role of spindle and kinetochore-associated (SKA) genes in tumorigenesis and cancer progression has been widely studied. However, so far, the oncogenic involvement of SKA family genes in pancreatic cancer and their prognostic potential remain unknown.

Methods: Here, we carried out a meta-analysis of the differential expression of SKA genes in normal and tumor tissue. Univariate and multivariate survival analyses were done to evaluate the correlation between SKA family gene expression and pancreas ductal adenocarcinoma (PDAC) prognosis. Joint-effect and stratified survival analysis as well as nomogram analysis were used to estimate the prognostic value of genes. The underlying regulatory and biological mechanisms were identified by Gene set enrichment analysis. Interaction between SKA prognosis-related genes and immune cell infiltration was assessed using the Tumor Immune Estimation Resource tool.

Results: We find that SKA1-3 are highly expressed in PDAC tissues relative to non-cancer tissues. Survival analysis revealed that high expression of SKA1 and SKA3 independently indicate poor prognosis but they are not associated with relapse-free survival. The prognostic value of SKA1 and SKA3 was further confirmed by the nomogram, joint-effect, and stratified survival analysis. Analysis of underlying mechanisms reveals that these genes influence cancer-related signaling pathways, kinases, miRNA, and E2F family genes. Notably, prognosis-related genes are inversely correlated with several immune cells infiltrating levels.

Conclusion: We find that SKA1 and SKA3 expression correlates with prognosis and immune cell infiltration in PDAC, highlighting their potential as pancreatic cancer prognostic biomarkers.

Keywords: bioinformatics analysis; biomarker; immune infiltration; pancreatic cancer; prognosis; spindle and kinetochore associated.

FIGURE 1

Flowchart presenting the general work flow of this study.

FIGURE 2

Meta-analysis of 12 datasets of pancreatic cancer. (A) Forest plot showing SKA1 expression difference. (B) sROC curve of SKA1. (C) Forest plot showing SKA2 expression difference. (D) sROC curve of SKA2. (E) Forest plot showing SKA3 expression difference. (F) sROC curve of SKA3.

FIGURE 3

Interaction networks and co-expression matrix of spindle and kinetochore associated genes. (A) Gene–gene interaction network created using the Gene Multiple Association Network Integration Algorithm (GeneMANIA). (B) Protein–protein interaction network created using the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING). (C–E) Co-expression matrix of SKA genes in TCGA, GTEx, and GSE62452 datasets.

FIGURE 4

Kaplan-Meier survival curves showing the association of spindle and kinetochore related genes with the overall survival and relapse-free survival of pancreatic ductal adenocarcinoma patients from TCGA and GEO datasets. Overall survival in TCGA: (A) SKA1; (B) SKA2; (C) SKA3; Overall survival in GEO62452: (D) SKA1; (E) SKA2; (F) SKA3; Overall survival in GEO28735: (G) SKA1; (H) SKA2; (I) SKA3; Relapse-free survival in TCGA stratified by (J) SKA1; (K) SKA2; (L) SKA3.

FIGURE 5

Joint-effect survival analysis for overall survival of patients with pancreatic ductal adenocarcinoma. (A) Combination of SKA1 and SKA3. Combination of SKA1 and prognosis-related clinical factors: (B) histologic grade; (C) radiation therapy; (D) targeted molecular therapy; (E) radical resection. Combination of SKA3 and prognosis-related clinical factors: (F) histologic grade; (G) radiation therapy; (H) targeted molecular therapy; (I) radical resection.

FIGURE 6

The relationship of spindle and kinetochore-associated genes with the clinical information. (A,B) Stratified survival analysis of SKA1 and SKA3 in various clinicopathological parameters. (C) A prognostic nomogram based on SKA1 and SKA3 for predicting the 1-, 2-, and 3-year overall survival rate of patients with pancreatic ductal adenocarcinoma. HR, hazard ratio; CI, confidence interval; L/L: SKA1^low + SKA3^low; L/H: SKA1^low + SKA3^high; H/L: SKA1^high + SKA3^low; H/H: SKA1^high + SKA3^high.

FIGURE 7

Analysis of the prognostic value and clinical relevance of SKA1 and SKA3 in pancreatic ductal adenocarcinoma patients. (A–D) Time-dependent receiver operating characteristic (ROC) curves of SKA1 and SKA3 showing the 1-, 2-, and 3-year overall survival rate of patients with PDAC from TCGA and GSE62452 datasets. (E) From top to bottom are the expression values of SKA1, patients’ survival status distribution, and the expression heat map of SKA1 in the low- and high-expression groups. The expression distribution of SKA1 and SKA2 genes in different (F,G) AJCC stages and (J,K) grades in TCGA and GSE62452 datasets. (H,I) The expression distribution of SKA1 and SKA2 genes in different pathological T grade and cancer status in TCGA cohort. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 8

Gene set enrichment analysis of SKA1 and SKA3 in TCGA dataset. (A–C, J, K) GSEA results of C5 gene sets for high SKA1 and SKA3 expression groups; (D–I, L–O) GSEA results of C2 gene sets for high SKA1 and SKA3 expression groups; NES, normalized enrichment score; FDR, false discovery rate.

FIGURE 9

Genomic alterations and DNA methylation level of spindle and kinetochore-associated genes. (A) OncoPrint of SKA1–3 alterations in TCGA cohort. Different types of genetic alterations shown in different colors. (B,C) SKA1 and SKA3 expression in different CNV groups. Data were obtained from the cBioportal (https://www.cbioportal.org/). (D,E) Two heatmaps showing the methylation profile of 11 CpG sites in SKA1 DNA locus and 15 CpG sites in SKA3 DNA locus. Data were obtained from UCSC Xena (http://xena.ucsc.edu/) using Infinium HumanMethylation450 BeadChip.

FIGURE 10

The impact of SKA1 and SKA3 gene expression and mutation on tumor immunity. (A) SKA1 expression showed significant negative correlation with infiltration levels of CD8+ T cells and macrophages. (B) SKA1 CNV influenced the infiltration levels of all the immune cells. (C) SKA3 expression showed significant negative correlation with infiltration levels of CD8+ T cells, CD4+ T cells, macrophages, neutrophils, and dendritic cells. (D) Changes in SKA3 CNV altered infiltration levels of B cells and CD4+ T cells. (E–G) Infiltration level of CD4+ T and CD8+ T cell and distribution of immune scores in high expression and low expression groups of SKA1 and SK A3 in TCGA datasets. Immune scores were calculated using the ESTIMATE algorithm. (H,I) The expression distribution of SKA1 and SKA3 genes in different mutation status of TP53 and KRAS in TCGA dataset. *p < 0.05, **p < 0.01, ***p < 0.001.