A Novel Prognostic Index Based on Alternative Splicing in Papillary Renal Cell Carcinoma.

Zhipeng Wu,Xianlin Xu,Kai Wang,Jinhui Liu,Rui Sun,Dongming Chen,Changchun Cao

doi:10.3389/fgene.2019.01333

Abstract

BackgroundPapillary renal cell carcinoma (pRCC) is a heterogeneous multifocal or isolated tumor with an invasive phenotype. Previous studies presented that alternative splicing, as a crucial posttranscriptional regulator in gene expression, is associated with tumorigenesis. However, the association between alternative splicing and pRCC has not been clarifiedMethodsThe RNA sequencing data and clinical information were downloaded from The Cancer Genome Atlas database and mRNA splicing profiles from TCGASpliceSeq. The percent spliced in data of alternative splicing merged with survival information was firstly calculated by univariate Cox regression analysis to screen for survival‐associated alternative splicing events, and survival‐associated alternative splicing events were then analyzed by Gene Ontology categories using Kyoto Encyclopedia of Genes and Genomes. Meanwhile, the least absolute shrinkage and selection operator Cox analysis and multivariate Cox analysis were performed to calculate the prognostic index for each alternative splicing type. In addition, clinical factors were introduced to assess the performance of prognostic index.ResultsA total of 4,084 candidate survival-associated alternative splicing events in 2,558 genes were screened out. Patients were divided into the low-risk group and the high-risk group based on the median prognostic index value. The Kaplan-Meier survival analysis (p < 0.05) and receiver operating characteristics curves (AUC>0.9) indicated that prognostic index was effective and stable for predicting the prognosis of pRCC patients. Furthermore, a regulatory network was constructed incorporating alternative splicing events and survival-associated splicing factors.ConclusionOur study provides new insights into the mechanism of alternative splicing events in tumorigenesis and their clinical potential for pRCC.

Highlights

Papillary renal cell carcinoma, which accounts for up to 15% of renal cell carcinoma, is the second most common histological subtype of kidney cancer (Jonasch et al, 2014; Malouf et al, 2016)
The percent spliced in data of alternative splicing merged with survival information was firstly calculated by univariate Cox regression analysis to screen for survival‐associated alternative splicing events, and survival‐associated alternative splicing events were analyzed by Gene Ontology categories using Kyoto Encyclopedia of Genes and Genomes
Patients were divided into the low-risk group and the high-risk group based on the median prognostic index value

Summary

Introduction

Papillary renal cell carcinoma (pRCC), which accounts for up to 15% of renal cell carcinoma, is the second most common histological subtype of kidney cancer (Jonasch et al, 2014; Malouf et al, 2016). Vascular endothelial growth factor (VEGF) pathway has been proven to be involved in metastatic pRCC (Armstrong et al, 2016), but we still speculate that multiple mechanisms lie behind these pRCC with diverse presentations. We designed this systematic and comprehensive analysis to drill into the oncogenic mechanism of pRCC. The GENCODE (Frankish et al, 2019) genome maintained by EBI currently counts 19,965 protein-coding genes, 17,910 long noncoding RNA genes, and 7,576 small noncoding genes in human Previous studies presented that alternative splicing, as a crucial posttranscriptional regulator in gene expression, is associated with tumorigenesis. The association between alternative splicing and pRCC has not been clarified

Methods

Results

Conclusion