Salivary gland tumors (SGTs) are rare tumors of the head and neck with different clinical behavior. Preoperative diagnosis, based on instrumental and cytologic examinations, is crucial for their correct management. The identification of molecular markers might improve the accuracy of pre-surgical diagnosis helping to plan the proper treatment especially when a definitive diagnosis based only on cytomorphology cannot be achieved. miRNAs appear to be new promising biomarkers in the diagnosis and prognosis of cancer. Studies concerning the useful of miRNA expression in clinical decision-making regarding SGTs remain limited and controversial.The expression of a panel of 798 miRNAs was investigated using Nanostring technology in 14 patients with malignant SGTs (6 mucoepidermoid carcinomas, 4 adenoid cystic carcinomas, 1 acinic cell carcinoma, 1 ductal carcinoma, 1 cystadenocarcinoma and 1 adenocarcinoma) and in 10 patients with benign SGTs (pleomorphic adenomas). The DNA Intelligent Analysis (DIANA)-miRPath v3.0 software was used to determinate the miRNA regulatory roles and to identify the controlled significant Kyoto Encyclopedia of Genes and Genomes (KEGG) molecular pathways. Forty six miRNAs were differentially expressed (False Discovery Rate-FDR<0.05) between malignant and benign SGTs. DIANA miRPath software revealed enriched pathways involved in cancer processes as well as tumorigenesis, cell proliferation, cell growth and survival, tumor suppressor expression, angiogenesis and tumor progression. Interestingly, clustering analysis showed that this signature of 46 miRNAs is able to differentiate the two analyzed groups. We found a correlation between histological diagnosis (benign or malignant) and miRNA expression profile.The molecular signature identified in this study might become an important preoperative diagnostic tool.