Background: Lung cancer is the leading cause of cancer-related deaths in the USA and worldwide. Yet, about 95% of new drug candidates validated in preclinical phase eventually fail in clinical trials. Such a high attrition rate is attributed mostly to the inability of conventional two-dimensionally (2D) cultured cancer cells to mimic native three-dimensional (3D) growth of malignant cells in human tumors. Thus, it is expected that 3D cell culture systems would more accurately represent the phenotype of cancer cells growing in tumors. To ascertain phenotypical differences between these two distinct culture conditions, we carried out a comparative proteomic analysis of membrane fraction obtained from 3D- and 2D-cultured NSCLC model cell line NCI-H23. Methods: Global shotgun membrane (SGM) proteomics that relies on strong cation exchange (SCX)-based peptide fractionation and accurate-mass, liquid chromatography mass spectrometry (HR/AM LC-MS) was employed to analyze microsomal fractions obtained from the NCI-H23 cells grown in both 2D and 3D culture conditions. Results: Comparative proteomics revealed a map of 1,166 (24%) nonredundant protein species regulated in culture dependent manner in NCI-H23 cell line. Of these, a subset of 234 (i.e., 21 %) proteins were found significantly dysregulated (p-value ≤ 0.05) under both culture conditions whereas a total of 334 (27.8%) and 598 (51,2%) proteins were uniquely identified in 3D and 2D culture, respectively. The Ingenuity Pathway Analysis revealed extensive metabolic changes and differential regulation of a subset of CD molecules in culture-dependent manner. Using western blotting we verified exclusive 3D-culture expression of CD99, CD146 and CD239, involved in development of malignant stroma extracellular matrix, neo-angiogenesis and metastasis. Furthermore, using label-free quantitation we unambiguously confirmed upregulation of wild type and monoallelic KRas4B G12C mutant in 3D cultured NCI-H23 cells, targeting exclusively proteoform-specific tryptic peptides. Conclusions: In this study we generated a large-scale proteomic resource/atlas of a preclinical testing model NCI-H23 cell line grown in 3D- and 2D-culture, providing insight into phenotypical/proteomic changes unique to each culture type, that would not have been discovered using only conventional 2D-culture. To reduce high attrition rate of new drug candidates it is critical to profile a large number of patient-derived NSCLC cell lines.