To determine if differential profile of miRNAs in peripheral blood mononuclear cells (PBMCs) could be identified in muscle-specific receptor tyrosine kinase antibody positive myasthenia gravis (MuSK-MG) and linked to disease stage, a case-control method was used to compare the difference in miRNA expression profiles of PBMCs using next generation sequencing (NGS) in MuSK-MG patients and healthy controls (HCs). Six significant miRNAs from the discovery set were then validated using RT-qPCR in 11 MuSK-MG patients and 10 HCs. A unique miRNA prediction algorithm was used to predict the target genes of differentially expressed miRNAs and a network of miRNA gene pathways. Compared with HCs, 101 differentially expressed miRNAs were screened in MuSK-MG, of which 5 miRNAs were upregulated, and 96 miRNAs were downregulated. The top six differentially expressed molecules were selected for verification; four of them (miR-340-5p, miR-106b-5p, miR-27a-3p, and miR-15a-3p) were significantly different. The network analysis of miRNA gene pathways revealed that differentially expressed miRNAs were involved in a complex set of biological processes. Clinically, the four miRNAs that were validated are not correlated to MuSK antibody titers and quantitative myasthenia gravis score. Four miRNAs that were validated in this study have specificity to distinguish MuSK-MG from HCs.