Abstract
Background: Sarcomas are heterogeneous rare malignancies constituting approximately 1% of all solid cancers in adults and including more than 70 histological and molecular subtypes with different pathological and clinical development characteristics. Method: We identified prognostic biomarkers of sarcomas by integrating clinical information and RNA-seq data from TCGA and GEO databases. In addition, results obtained from cell cycle, cell migration, and invasion assays were used to assess the capacity for Tanespimycin to inhibit the proliferation and metastasis of sarcoma. Results: Sarcoma samples (N = 536) were divided into four pathological subtypes including DL (dedifferentiated liposarcoma), LMS (leiomyosarcoma), UPS (undifferentiated pleomorphic sarcomas), and MFS (myxofibrosarcoma). RNA-seq expression profile data from the TCGA dataset were used to analyze differentially expressed genes (DEGs) within metastatic and non-metastatic samples of these four sarcoma pathological subtypes with DEGs defined as metastatic-related signatures (MRS). Prognostic analysis of MRS identified a group of genes significantly associated with prognosis in three pathological subtypes: DL, LMS, and UPS. ISG15, NUP50, PTTG1, SERPINE1, and TSR1 were found to be more likely associated with adverse prognosis. We also identified Tanespimycin as a drug exerting inhibitory effects on metastatic LMS subtype and therefore can serve a potential treatment for this type of sarcoma. Conclusions: These results provide new insights into the pathogenesis, diagnosis, treatment, and prognosis of sarcomas and provide new directions for further study of sarcoma.