Initially the ETS family gene fusions in prostate cancer was identified by cancer outlier profile analysis (COPA)43 using the gene expression microarray data. Given the unprecedented high prevalence of gene rearrangements and associated gene fusions in prostate cancer, we hypothesized that additional non-ETS gene fusions may occur in the ETS negative prostate cancer. This view is supported by the recent identification of RAF family gene rearrangements44 in a small subset of ETS negative prostate cancer. We have also reported RAF rearrangement in a small subset of gastric cancer (AGTRAP-BRAF) and melanoma (BRAF and RAF1). AKAP9-BRAF in thyroid cancer45 and KIAA1549-BRAF in pilocytic astrocytoma46 have also been reported by others and recently pancreatic acinar cell carcinoma was known to have RAF kinase rearrangements similar to the fusion genes that we identified in prostate cancer but with different tissue specific 5’partner gene27. Similarly “druggable” kinase gene fusion, EML4-ALK47 was identified in 1-5% of non-small cell lung cancer. Based on our experience in using the next generation sequencing technology we believe that transcriptome sequencing is the rational and an unbiased approach for the identification of gene fusions in the ETS negative AA men with prostate cancer44; …show more content…
We found rearrangements only in ERG, ETV1, ETV4 and ETV5 genes with different 5’ partner genes. Many other studies have also reported similar results, suggesting non-ETS gene fusions may occur in ETS negative prostate cancer. Five to ten percent of the cases show overexpression of SPINK1 in ETS negative EA PCa6. To test our hypothesis on the presence of non-ETS gene fusions in PCa we undertook an alternate and unbiased high throughput transcriptome sequencing approach using NGS technology (Illumina ) in a small set of ETS negative samples and identified gene fusions involving BRAF and RAF1 genes in 1-2% of ETS negative prostate cancer44. Although RAF genes are not an outlier gene in PCa, based on our scoring system for nominating driver gene fusions we identified BRAF and RAF1 as the outlier gene fusions in the respective samples (Figure 9). Moreover, in the RAF1 fusion positive samples we identified the balanced reciprocal transcripts (ESRP1-RAF1; RAF1-ESRP1) which is a rare phenomenon not readily detected by conventional approaches. We studied the oncogenic properties of the SLC45A3-BRAF and ESRP1-RAF1 using both in vivo and