The present study aimed to explore specific molecular targets for the diagnosis and treatment of non‑small cell lung cancer (NSCLC). The expression profiles of microRNAs (miRNAs) and mRNAs were downloaded from the GEO (GSE102286 and GSE101929) and TCGA databases. After data preprocessing, differentially expressed genes (DEGs) and differentially expressed miRNAs (DEMs) in cancer and normal tissues were selected and used to construct a DEM‑DEG regulatory network and a protein‑protein interaction (PPI) network. The genes and miRNAs in these networks were subjected to functional enrichment and survival analyses. Several key DEMs and DEGs were verified using RT‑qPCR, and the results were statistically interpreted using a multivariate logistic regression analysis. In this study, 25 DEMs and 789 DEGs common to all datasets were identified, which were then used for the construction of a DEM‑DEG regulatory network and a PPI network module. Survival analyses of 19 DEMs in the DEM‑DEG regulatory network and 36 DEGs in the PPI network module revealed that 34 DEGs (including TOP2A, CCNB1, BIRC5, and TTK) and two miRNAs (miR‑21‑5p and miR‑31‑5p) were significantly associated with NSCLC prognosis. Moreover, RT‑qPCR analysis identified three DEGs and five DEMs that had changes in expression consistent with those observed in the bioinformatic analysis. Finally, a multivariate logistic regression analysis of the data showed that TOP2A, CCNB1, BIRC5, miR‑21‑5p, miR‑193b‑3p, miR‑210‑3p and miR‑31‑5p could be combined for the diagnosis of NSCLC. In conclusion, TOP2A, CCNB1, BIRC5, miR‑21‑5p, miR‑193b‑3p, miR‑210‑3p and miR‑31‑5p may therefore serve as important biomarkers and diagnostic targets for NSCLC.
Keywords: non‑small cell lung cancer; mRNA; microRNA; network; multivariate logistic regression analysis.