TY - JOUR
T1 - Re-Evaluate Fusion Genes in Prostate Cancer
AU - Wei, Ting
AU - Lu, Ji
AU - Ma, Tao
AU - Huang, Haojie
AU - Kocher, Jean Pierre
AU - Wang, Liguo
N1 - Funding Information:
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was support in part by the Center for Individualized Medicine, Mayo Clinic; National Natural Science Foundation of China (31501052); and Natural Science Foundation of Jilin Province of China (20200201315JC).
Publisher Copyright:
© The Author(s) 2021.
PY - 2021
Y1 - 2021
N2 - Background: Thousands of gene fusions have been reported in prostate cancer, but their authenticity, incidence, and tumor specificity have not been thoroughly evaluated, nor have their genomic characteristics been carefully explored. Methods: We developed FusionVet to dedicatedly validate known fusion genes using RNA-seq alignments. Using FusionVet, we re-assessed 2727 gene fusions reported from 36 studies using the RNA-seq data generated by The Cancer Genome Atlas (TCGA). We also explored their genomic characteristics and interrogated the transcriptomic and DNA methylomic consequences of the E26 transformation-specific (ETS) fusions. Results: We found that nearly two-thirds of reported fusions are intra-chromosomal, and 80% of them were formed between 2 protein-coding genes. Although most (76%) genes were fused to only 1 partner, we observed many fusion hub genes that have multiple fusion partners, including ETS family genes, androgen receptor signaling pathway genes, tumor suppressor genes, and proto-oncogenes. More than 90% of the reported fusions cannot be validated by TCGA RNA-seq data. For those fusions that can be validated, 5% were detected from tumor and normal samples with similar frequencies, and only 4% (120 fusions) were tumor-specific. The occurrences of ERG, ETV1, and ETV4 fusions were mutually exclusive, and their fusion statuses were tightly associated with overexpressions. Besides, we found ERG fusions were significantly co-occurred with PTEN deletion but mutually exclusive with common genomic alterations such as SPOP mutation and FOXA1 mutation. Conclusions: Most of the reported fusion genes cannot be validated by TCGA samples. The ETS family and androgen response genes were significantly enriched in prostate cancer–specific fusion genes. Transcription activity was significantly repressed, and the DNA methylation was significantly increased in samples carrying ERG fusion.
AB - Background: Thousands of gene fusions have been reported in prostate cancer, but their authenticity, incidence, and tumor specificity have not been thoroughly evaluated, nor have their genomic characteristics been carefully explored. Methods: We developed FusionVet to dedicatedly validate known fusion genes using RNA-seq alignments. Using FusionVet, we re-assessed 2727 gene fusions reported from 36 studies using the RNA-seq data generated by The Cancer Genome Atlas (TCGA). We also explored their genomic characteristics and interrogated the transcriptomic and DNA methylomic consequences of the E26 transformation-specific (ETS) fusions. Results: We found that nearly two-thirds of reported fusions are intra-chromosomal, and 80% of them were formed between 2 protein-coding genes. Although most (76%) genes were fused to only 1 partner, we observed many fusion hub genes that have multiple fusion partners, including ETS family genes, androgen receptor signaling pathway genes, tumor suppressor genes, and proto-oncogenes. More than 90% of the reported fusions cannot be validated by TCGA RNA-seq data. For those fusions that can be validated, 5% were detected from tumor and normal samples with similar frequencies, and only 4% (120 fusions) were tumor-specific. The occurrences of ERG, ETV1, and ETV4 fusions were mutually exclusive, and their fusion statuses were tightly associated with overexpressions. Besides, we found ERG fusions were significantly co-occurred with PTEN deletion but mutually exclusive with common genomic alterations such as SPOP mutation and FOXA1 mutation. Conclusions: Most of the reported fusion genes cannot be validated by TCGA samples. The ETS family and androgen response genes were significantly enriched in prostate cancer–specific fusion genes. Transcription activity was significantly repressed, and the DNA methylation was significantly increased in samples carrying ERG fusion.
KW - Prostate cancer
KW - TMPRSS2-ERG
KW - fusion gene
KW - gene fusion
KW - meta-analysis
UR - http://www.scopus.com/inward/record.url?scp=85108585689&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85108585689&partnerID=8YFLogxK
U2 - 10.1177/11769351211027592
DO - 10.1177/11769351211027592
M3 - Article
AN - SCOPUS:85108585689
SN - 1176-9351
VL - 20
JO - Cancer Informatics
JF - Cancer Informatics
ER -