The three-dimensional (3D) tumor spheroid model exhibits enhanced fidelity in replicating the tumor microenvironment and demonstrates exceptional resistance to clinical drugs compared to the two-dimensional (2D) monolayer model. In this study, we used multi-omics (transcriptome, proteomics, and metabolomics) tools to explore the molecular mechanisms and metabolic differences of the two culture models. Analysis of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment pathways revealed that the differentially expressed genes between the two culture models were mainly enriched in cellular components and biological processes associated with extracellular matrix, extracellular structural organization and mitochondrial function. A combined analysis of three omics data revealed 11 possible drug resistance targets. Among these targets, 7 genes, AKR1B1, ALDOC, GFPT2, GYS1, LAMB2, PFKFB4 and SLC2A1, exhibited significantly up-regulation. Conversely, 4 genes, COA7, DLD, IFNGR1 and QRSL1, were significantly down regulated. Clinical prognostic analysis using the TCGAsurvival database indicated that high expression groups of SLC2A1, ALDOC and PFKFB4 exhibited a significant negative correlation with patient survival. We further validated their involvement in chemotherapy drug resistance, indicating their potential significance in improving prognosis and chemotherapy outcomes. These results provide valuable insights into potential therapeutic targets that can potentially enhance treatment efficacy and patient outcomes.