A major challenge in osteosarcoma (OS) is the selection of the most effective chemotherapeutic agents for individual patients, while the administration of ineffective chemotherapy increases mortality and decreases quality of life in patients. This emphasizes the need to evaluate every patient's probability of responding to each chemotherapeutic agent. We developed a profiling strategy for serum exosomal microRNAs and mRNAs in OS patients with differential chemotherapeutic responses. Twelve miRNAs were up regulated and 18 miRNAs were under regulated significantly in OS patient with poor chemotherapeutic response when compared with those in good chemotherapeutic response (p<0.05). In addition, miR-124, miR133a, miR-199a-3p, and miR-385 were validated and significantly reduced in poorly responded patients with an independent OS cohort. While miR-135b, miR-148a, miR-27a, and miR-9 were significantly over expressed in serum exosomes. Bioinformatic analysis by DIANA-mirPath demonstrated that Proteoglycans in cancer, Hippo signaling pathway, Pathways in cancer, Transcriptional misregulation in cancer, PI3K-Akt signaling pathway, Ras signaling pathway, Ubiquitin mediated proteolysis, Choline metabolism in cancer were the most prominent pathways enriched in quantiles with the miRNA patterns related to poor chemotherapeutic response. Messenger RNAs(mRNAs) includingAnnexin2, Smad2, Methylthioadenosine phosphorylase (MTAP), Cdc42-interacting protein 4 (CIP4), Pigment Epithelium-Derived Factor (PEDF), WW domain-containing oxidoreductase (WWOX), Cell division cycle 5-like (Cdc5L), P27 were differentially expressed in exosomes in OS patients with different chemotherapeutic response. These data demonstrated that exosomal RNA molecules are reliable biomarkers in classifying osteosarcoma with different chemotherapy sensitivity.