Non-small cell lung cancer (NSCLC) is one of the leading causes of cancer-associated mortality. Lung adenocarcinoma (LAC) is the most prevalent pathological subtype of NSCLC and accounts for ~40% of all lung cancer mortalities. There remains an urgent demand for the identification of novel biomarkers for the diagnosis and development of therapeutic strategies for LAC. In the present study, the profiles of the differentially-expressed circular RNAs (circRNAs) in LAC tissues compared with those in their corresponding non-cancerous tissues were obtained after analyzing the circRNA microarray dataset GSE101586. The expression pattern of the indicated circRNAs in the LAC tissues were subsequently verified using reverse transcription-quantitative PCR (RT-qPCR). The potential prognostic significance of these circRNAs in patients with LAC were then analyzed in a retrospective clinical study. A circRNA-microRNA (miR or miRNA)-mRNA regulatory network in LAC was established by using Cytoscape. In addition, a protein-protein interaction (PPI) network was plotted using the Search Tool for the Retrieval of Interacting Genes/Proteins and visualized through Cytoscape. The prognostic value of the hub genes found was then analyzed based on the Gene Expression Profiling Interactive Analysis database. In total, four differentially-expressed circRNAs were obtained from the GSE101586 microarray dataset, three of which (hsa_circ_0006220, hsa_circ_0072088 and hsa_circ_0001666) were confirmed by RT-qPCR to be highly expressed in LAC tissues. This retrospective clinical study revealed that higher expression levels of these three circRNAs were associated with poorer prognoses in patients with LAC. In addition, siRNA-mediated knockdown of these circRNAs was found to inhibit cell proliferation, migration and invasion in LAC cells. Following analysis of the molecular mechanism underlying these circRNAs, eight miRNAs, namely miR-520f, miR-1261, miR-1270, miR-620, miR-188-3p, miR-516b, miR-940 and miR-661, were identified with potential binding sites for these three circRNAs. Subsequently, 232 overlapped genes from the 795 upregulated genes in the LAC samples from The Cancer Genome Atlas database and 7,829 predicted target genes of the list of eight aforementioned miRNAs were obtained. A circRNA-miRNA-mRNA network was then constructed. A PPI network was established, with six hub genes, namely kinesin family member (KIF) 2C, KIF18B, maternal embryonic leucine zipper kinase, baculoviral IAP repeat-containing 5, polo-like kinase 1 and cytoskeleton-associated protein 2-like, determined from this network. Higher expression levels of each of these hub genes were found to be associated with poorer prognoses of patients with LAC. To conclude, data from the present study suggested that circRNAs hsa_circ_0006220, hsa_circ_0072088 and hsa_circ_0001666 have the potential to be viable biomarkers and therapeutic targets for LAC.