Background: Alternative splicing (AS) is believed to play a vital role in tumor development. Therefore, comprehensive investigation of AS and its biological function in prostate cancer (PCa) is crucial.
Methods: The AS profiling of 489 patients with PCa was obtained from The Cancer Genome Atlas (TCGA) SpliceSeq database. Bioinformatics tools were used to describe splicing associations and build prognostic models. Unsupervised clustering of the determined prognostic AS events and the relationship with immune characteristics were also explored.
Results: In total, 20,723 AS events were detected and 2,805 were identified in PCa. In the regulatory networks, the data suggested a significant correlation between splicing factor (SF) expression and AS events. To stratify the progression risk of PCa patients, prognostic models were constructed using splicing patterns. Six AS events were screened out as independent prognostic factors for progression-free survival. Based on the gene features, we constructed the combined prognostic predictors model, and the receiver operating characteristic (ROC) curve for this model reached a high area under the ROC curve (AUC) of 0.729793, indicating a favorable ability to predict patient outcomes. Through unsupervised clustering analysis, the correlations between AS-based clusters and prognosis as well as immune characteristics were revealed. The correlation analysis on TIMER revealed the relationship between gene expression and immune cell infiltration.
Conclusions: This in-depth genome-wide analysis of the AS profiling in PCa revealed unique AS events associated with cancer progression and the infiltration of immune cells, with potential for predicting outcomes and therapeutic responses.
Keywords: Alternative splicing (AS); immune cells infiltration; prognosis; prostate cancer (PCa).
2021 Translational Andrology and Urology. All rights reserved.