Molecular profiles of predictive biomarkers for platinum-based chemotherapy in Non-Small Cell Lung Cancer (NSCLC)

Abstract

Background

Non-small cell lung cancer (NSCLC) is the principal subtype of lung cancer. Among all therapeutic options, platinum-based chemotherapy agents, especially Cisplatin, are still commonly used treatment for NSCLC patients. However, developing chemoresistance in NSCLC cells often gives rise to chemotherapy failure. Therefore, more studies are required to shed light on gene interaction and cellular pathways involved in initiating and developing resistance to platinum-based chemotherapy in NSCLC. Hence, it is urgent to find the key genes, microRNA (miRNAs), and potential molecular mechanisms implicated in chemoresistance and present markers to predict response to platinum-based chemotherapy in NSCLC patients.

Methods

The microarray datasets GSE6410, GSE7035, GSE14814, GSE26704, GSE73302 were downloaded from the Gene Expression Omnibus (GEO) database and were analyzed using R software. Functional and pathway enrichment analyses were performed using the Enrich R site. Then, the protein-protein interaction (PPI) network and hub genes were obtained using the Cytoscape software. Further, the miRSystem database was performed to predict the miRNAs regulating the hub genes. Moreover, Cytoscape software and the CytoHubba plugin were used to construct the miRNA-target interaction network and hub modules. Finally, the Kaplan–Meier curve was used to demonstrate the survival curves and assess the association of the genes signature with clinical outcomes.

Results

A total of 142 differentially expressed genes (DEGs) were found. The gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses present the p53 signaling pathway as the most significant pathway.

Besides, from the top ten terms obtained of Biological Process, Molecular Function, and Cellular Component, the first ones, including cholesterol biosynthetic process, the extrinsic component of external side of plasma membrane, cytokine activity, were selected respectively. Based on the PPI network, the ten nodes with the highest degree were screened as hub genes. In addition, from the miRNA–target regulatory network in Cytoscape, ten hub nodes were found. Ultimately, according to Kaplan–Meier curve, BTG2 and TP53I3 with p-value <0.05 were associated with a better prognosis.

Conclusions

In the present study, DEGs, candidate miRNAs, and underlying mechanisms involved in chemoresistance were identified to suggest potential biomarkers to provide new clues for the prediction of response to platinum-based chemotherapy.