Prognostic alternative mRNA splicing signature in non-small cell lung cancer.

Alternative splicing provides a major mechanism to generate protein diversity. Increasing evidence suggests a link of dysregulation of splicing associated with cancer. Genome-wide alternative splicing profiling in lung cancer remains largely unstudied. We generated alternative splicing profiles in 491 lung adenocarcinoma (LUAD) and 471 lung squamous cell carcinoma (LUSC) patients in TCGA using RNA-seq data, prognostic models and splicing networks were built by integrated bioinformatics analysis. A total of 3691 and 2403 alternative splicing events were significantly associated with patient survival in LUAD and LUSC, respectively, including EGFR, CD44, PIK3C3, RRAS2, MAPKAP1 and FGFR2. The area under the curve of the receiver-operator characteristic curve for prognostic predictor in NSCLC was 0.817 at 2000 days of overall survival which were also over 0.8 in LUAD and LUSC, separately. Interestingly, splicing correlation networks uncovered opposite roles of splicing factors in LUAD and LUSC. We created prognostic predictors based on alternative splicing events with high performances for risk stratification in NSCLC patients and uncovered interesting splicing networks in LUAD and LUSC which could be underlying mechanisms.