Sarcomas are malignant soft tissue and bone tumours affecting adults, adolescents and children. They represent a morphologically heterogeneous class of tumours and some entities lack defining histopathological features. Therefore, the diagnosis of sarcomas is burdened with a high inter-observer variability and misclassification rate. Here, we demonstrate classification of soft tissue and bone tumours using a machine learning classifier algorithm based on array-generated DNA methylation data. This sarcoma classifier is trained using a dataset of 1077 methylation profiles from comprehensively pre-characterized cases comprising 62 tumour methylation classes constituting a broad range of soft tissue and bone sarcoma subtypes across the entire age spectrum. The performance is validated in a cohort of 428 sarcomatous tumours, of which 322 cases were classified by the sarcoma classifier. Our results demonstrate the potential of the DNA methylation-based sarcoma classification for research and future diagnostic applications.

The NanoString nCounter assay is a high-throughput hybridization technique using target-specific probes that can be customized to test for numerous fusion transcripts in a single assay using RNA from formalin-fixed, paraffin-embedded material. We designed a NanoString assay targeting 174 unique fusion junctions in 25 sarcoma types. The study cohort comprised 212 cases, 96 of which showed fusion gene expression by the NanoString assay, including all 20 Ewing sarcomas, 11 synovial sarcomas, and 5 myxoid liposarcomas tested. Among these 96 cases, 15 showed fusion expression not identified by standard clinical assay, including EWSR1-FLI1, EWSR1-ERG, BCOR-CCNB3, ZC3H7B-BCOR, HEY1-NCOA2, CIC-DUX4, COL1A1-PDGFB, MYH9-USP6, YAP1-TFE3, and IRF2BP2-CDX1 fusions. There were no false-positive results; however, four cases were false negative when compared with clinically available fluorescence in situ hybridization or RT-PCR testing. When batched as six cases, the per-sample reagent cost was less than conventional techniques, such as fluorescence in situ hybridization, with technologist hands-on time of 1.2 hours per case and assay time of 36 hours. In summary, the NanoString nCounter Sarcoma Fusion CodeSet reliably and cost-effectively identifies fusion genes in sarcomas using formalin-fixed, paraffin-embedded material, including many fusions missed by standard clinical assays, and can serve as a first-line clinical diagnostic test for sarcoma fusion gene identification, replacing multiple individual clinical assays.