| Popis: |
Childhood sarcomas can be extremely difficult to accurately diagnose on the basis of morphological characteristics alone. Ancillary methods, such as RT-PCR or fluorescence in situ hybridization, to detect pathognomonic gene fusions can help to distinguish these tumors. Two major deficiencies of these assays are their inability to identify gene fusions at nucleotide resolution or to detect multiple gene fusions simultaneously. We developed a next-generation sequencing-based assay designated ChildSeq-RNA that uses the Ion Torrent platform to screen for EWSR1-FLI1 and EWSR1-ERG, PAX3-FOXO1 and PAX7-FOXO1, EWSR1-WT1, and ETV6-NTRK3 fusions of Ewing sarcoma (ES), alveolar rhabdomyosarcoma, desmoplastic small round cell tumor, and congenital fibrosarcoma, respectively. To rapidly analyze resulting data, we codeveloped a bioinformatics tool, termed ChildDecode, that operates on a scalable, cloud-computing platform. Total RNA from four ES cell lines plus 33 clinical samples representing ES, alveolar rhabdomyosarcoma, desmoplastic small round cell tumor, and congenital fibrosarcoma tumors was subjected to ChildSeq-RNA. This accurately identified corresponding gene fusions in each tumor type, with no examples of false positive fusion detection in this proof-of-concept study. Comparison with previous RT-PCR findings demonstrated high sensitivity (96.4%; 95% CI, 82.3%-99.4%) and specificity (100%; 95% CI, 56.6%-100%) of ChildSeq-RNA to detect gene fusions. Herein, we propose ChildSeq-RNA as a novel tool to detect gene fusions in childhood sarcomas at single-nucleotide resolution. |
