Chemo-resistance is a huge obstacle encountered in the osteosarcoma (OS) treatment. Protein-coding mRNAs, as well as non-coding RNAs (ncRNAs), including long ncRNA (lncRNA), circular RNA (circRNA), and microRNA (miRNA), have been demonstrated to play an essential role in the regulation of cancer biology. However, the comprehensive expression profile and competing endogenous RNA (ceRNA) regulatory network between mRNAs and ncRNAs in the OS chemo-resistance still remain unclear. In the current study, we developed whole-transcriptome sequencing (RNA sequencing RNA-seq) in the three paired multi-drug chemo-resistant and chemo-sensitive OS cell lines to comprehensively identify differentially expressed lncRNAs, circRNAs, miRNAs, and mRNAs. Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were performed for mRNAs with significantly different expression. Then the ceRNA networks combining lncRNAs, circRNAs, miRNAs, and mRNAs were predicted and constructed on the basis of the authoritative miRanda and TargetScan databases combined with the widely accepted vital drug resistance-related genes and signal transduction pathways. In addition, two constructed ceRNA regulatory pathways, lncRNAMEG3/hsa-miR-200b-3p/AKT2 and hsa_circ_0001258/hsa-miR-744-3p/GSTM2, were randomly selected and validated by real-time qPCR, RNA immunoprecipitation (RIP), RNA pull-down assay, and dual luciferase reporter gene system. Taken together, our findings may provide new evidence for the underlying mechanism of OS chemo-resistance and uncover some novel targets for reversing it. Zhang et al. identify the comprehensive expression profile and competing endogenous RNA (ceRNA) regulatory network between mRNAs and ncRNAs (long ncRNA lncRNA, circular RNA (circRNA), and microRNA miRNA) in the OS chemo-resistance by whole-transcriptome RNA sequencing, bioinformatics analysis, and assay validation in the three paired multi-drug chemo-resistant and chemo-sensitive OS cell lines.