Identification of a Potential Alternative Splicing Prognostic Signature Associated With Immune Infiltrates of Ovarian Cancer

Abstract

Background: The mortality rate of ovarian cancer (OC) ranks the first in gynecological tumors, which seriously threatens women's health and life. In recent years, alternative splicing (AS) has gradually been considered to play a key role in immune infiltrates of tumor. However, the prognostic significance of AS events related to immune infiltrates in OC remains unknow. The aim of our research was to investigate the potential prognostic value of AS events associated with immune infiltrates in OC.Methods: The RNA sequences (RNA-seq) and clinical data were downloaded from the Cancer Genome Atlas (TCGA) database. The AS event data was obtained from TCGA SpliceSeq database. Single sample gene set enrichment analysis (ssGSEA) was performed to calculate the abundance of 28 immune cell types in samples from TCGA-OV dataset. A consensus clustering algorithm was used to group the OC patients. Differential expression analysis was used to identify differentially expressed AS events (DEASs) between groups. Univariate Cox regression analysis was implemented to screen for AS events with prognostic value. A least absolute shrinkage and selection operator (LASSO) Cox regression and multivariate Cox regression analyses was used to further narrow the AS events with prognostic value and construct an alternative splicing prognostic signature for predicting the prognosis of OC patients. Results: The OC patients from TCGA database were classified into two groups (cluster.A and cluster.B) based on the 28 types of TIIC using a consensus clustering algorithm. The patients in the cluster.A group had increased immune infiltrates compared with the cluster.B group. 3616 DEASs were acquired and 171 DEASs have prognostic value (p<0.05). 28 DEASs with prognostic value (p<0.001) were fitted into LASSO Cox regression and multivariate Cox regression analyses. A prognostic signature with 18 DEASs was constructed to predict the prognosis of OC patients. Each patient obtained a riskscore and the patients were classified into high-and low-risk group using the median riskscore as a cutoff. Kaplan-Meier curve revealed that the patients in high-risk group have poor outcome. Conclusions: Collectively, our research identified an alternative splicing prognostic signature associated with immune infiltrates of OC, which may provide new directions for the immunotherapy of OC patients.