Abstract
Background Prostate cancer (PCa) is the most common malignancy and the leading cause of cancer death in men. Recent studies suggest the molecular signature was more effective than the clinical indicators for the prognostic prediction, but all of the known studies focused on a single RNA type. The present study was to develop a new prognostic signature by integrating long noncoding RNAs (lncRNAs) and messenger RNAs (mRNAs) and evaluate its prognostic performance. Methods The RNA expression data of PCa patients were downloaded from The Cancer Genome Atlas (TCGA) or Gene Expression Omnibus database (GSE17951, GSE7076, and GSE16560). The PCa-driven modules were identified by constructing a weighted gene coexpression network, the corresponding genes of which were overlapped with differentially expressed RNAs (DERs) screened by the MetaDE package. The optimal prognostic signature was screened using the least absolute shrinkage and selection operator analysis. The prognostic performance and functions of the combined prognostic signature was then assessed. Results Twelve PCa-driven modules were identified using TCGA dataset and validated in the GSE17951 and GSE7076 datasets, and six of them were considered to be preserved. A total of 217 genes in these 6 modules were overlapped with 699 DERs, from which a nine-gene prognostic signature was identified (including 3 lncRNAs and 6 mRNAs), and the risk score of each patient was calculated. The overall survival was significantly shortened in patients having the risk score higher than the cut-off, which was demonstrated in TCGA (p = 5.063E − 03) dataset and validated in the GSE16560 (p = 3.268E − 02) dataset. The prediction accuracy of this risk score was higher than that of clinical indicators (the Gleason score and prostate-specific antigen) or the single RNA type, with the area under the receiver operator characteristic curve of 0.945. Besides, some new therapeutic targets and mechanisms (MAGI2-AS3-SPARC/GJA1/CYSLTR1, DLG5-AS1-DEFB1, and RHPN1-AS1-CDC45/ORC) were also revealed. Conclusion The risk score system established in this study may provide a novel reliable method to identify PCa patients at a high risk of death.
Highlights
Prostate cancer (PCa) is the most frequently diagnosed malignancy in men, with an estimated 174,650 new cases and 31,620 deaths in 2019 in the United States [1]
The objectives of the current study were (1) to establish a signature for overall survival (OS) prediction based on the PCadriven long noncoding RNAs (lncRNAs) and messenger RNAs (mRNAs) which were screened by weighted gene coexpression network analysis (WGCNA) [15], a systematic biological method to cluster highly correlated genes; (2) to confirm the superiority of the integrated molecular signature to clinical indicators or single molecular type; and (3) to reveal the underlying functions of the gene signature
Three gene microarray datasets were downloaded from Gene Expression Omnibus (GEO, http://www. http://ncbi.nlm.nih.gov/geo/) because they analyzed the gene expression profiles in tumor and control tissues of PCa patients (GSE17951: control, n = 14; tumor, n = 109 [16, 17] in which some samples without clear type description were deleted; GSE70768: control, n = 73; tumor, n = 126 [18]; these two datasets were used to estimate module preservation) or recorded the prognostic outcomes in all PCa patients (GSE16560: tumor, n = 281 [19]; this dataset was used for signature validation)
Summary
Prostate cancer (PCa) is the most frequently diagnosed malignancy in men, with an estimated 174,650 new cases and 31,620 deaths in 2019 in the United States [1]. A total of 217 genes in these 6 modules were overlapped with 699 DERs, from which a nine-gene prognostic signature was identified (including 3 lncRNAs and 6 mRNAs), and the risk score of each patient was calculated. The overall survival was significantly shortened in patients having the risk score higher than the cut-off, which was demonstrated in TCGA (p = 5:063E − 03) dataset and validated in the GSE16560 (p = 3:268E − 02) dataset. The prediction accuracy of this risk score was higher than that of clinical indicators (the Gleason score and prostate-specific antigen) or the single RNA type, with the area under the receiver operator characteristic curve of 0.945. The risk score system established in this study may provide a novel reliable method to identify PCa patients at a high risk of death
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
