Systematic Profiling of Alternative mRNA Splicing Signature for Predicting Glioblastoma Prognosis

Abstract

Emerging evidence suggests that alternative splicing (AS) is modified in cancer and is associated with cancer progression. Systematic analysis of AS signature in glioblastoma (GBM) is lacking and is greatly needed. We profiled genome-wide AS events in 498 GBM patients in TCGA using RNA-seq data, and splicing network and prognostic predictor were built by integrated bioinformatics analysis. Among 45,610 AS events in 10,434 genes, we detected 1,829 AS events in 1,311 genes, and 1,667 AS events in 1,146 genes that were significantly associated with overall survival and disease-free survival of GBM patients, respectively. Five potential feature genes, S100A4, ECE2, CAST, ASPH, and LY6K, were discovered after network mining as well as correlation analysis between AS and gene expression, most of which were related to carcinogenesis and development. Multivariate survival model analysis indicated that these five feature genes could classify the prognosis at AS event and gene expression level. This report opens up a new avenue for exploration of the pathogenesis of GBM through AS, thus more precisely guiding clinical treatment and prognosis judgment.

Keywords: alternative splicing (AS) events; disease-free survival; glioblastoma (GBM); overall survival; prognostic predictor.

Figure 1

Overview of seven types of AS in this study. (A) Illustrations for 7 different AS events. (B) Number of AS events from the 498 GBM patients.

Figure 2

Overview of prognosis-related alternative splicing events in GBM. (A) Venn diagram of AS events of genes significantly related to overall survival and those related to recurrence after radio- and chemo-therapy. (B) Venn diagram of gene intersections in AS events of genes significantly related to overall survival and those related to recurrence after radio- and chemo-therapy. (C) Histogram of the 7 types of AS events that were markedly correlated with overall survival in the gene AS events. (D) Histogram of the 7 types of AS events that were remarkably correlated with recurrence after radio- and chemo-therapy for gene AS events.

Figure 3

Forest plots of survival-associated AS events in GBM. (A–G) Hazard ratios of top 10 survival-associated AA, AD, AP, AT, ES, ME, and RI events.

Figure 4

Kaplan-Meier plots and ROC curves of prognostic predictor for GBM patients. (A–G) Kaplan-Meier curves for prognostic prediction model built with one type of survival-associated AS event for GBM patients, respectively. The red line indicates a high-risk group, while the blue line indicates a low risk group. (H) ROC curves with AUC of prognostic predictor built by one type of all seven types of survival-associated AS events in GBM.

Figure 5

Representative dot plots of correlations between expression of 5 feature genes [S100A4 (A), ECE2 (B), CAST (C), ASPH (D), and LY6K (E)] and PSI values of AS events (p < 0.05).

Figure 6

ROC curve and Kaplan-Meier plot of prognostic predictor for GBM patients. (A,C) ROC curve with AUC of prognostic predictor related to overall survival (A) and disease-free survival (C) built by alternative events of 5 feature genes in GBM. (B,D) Kaplan-Meier curves of prognostic predictor related to overall survival (B) and disease-free survival (D) built by alternative events of 5 feature genes in GBM. (E,G) ROC curve with AUC of prognostic predictor related to overall survival (E) and disease-free survival (G) built by alternative events and transcriptome levels of 5 feature genes in GBM. (F,H) Kaplan-Meier curves of prognostic predictor related to overall survival (F) and disease-free survival (H) built by alternative events and transcriptome levels of 5 feature genes in GBM.