Background:Entrectinib, a ROS1 inhibitor, is effective in patients with ROS1-positive
non-small-cell lung cancer (NSCLC). However, entrectinib resistance remains a challenge worldwide.
The biomarkers of entrectinib resistance and molecular mechanisms have not been clarified
based on the Gene Expression Omnibus (GEO) database.Objects:The aim of this study is to identify key genes and signaling pathways involved in the development
of entrectinib-resistant NSCLC through bioinformatics analysis and experimental validation.Methods:Differentially expressed genes (DEGs) were screened between entrectinib resistant and
parental human NSCLC cell lines of the GSE214715 dataset, lung adenocarcinoma (LUAD) and
non-tumor adjacent tissues of the GSE75037 dataset, and NSCLC and non-tumor adjacent tissues
of the GSE18842 dataset. Functional enrichment analyses were performed, including Gene Ontology
(GO) term and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. Overlapped
DEGs among those three datasets were identified using the Venn diagram package. The transcriptional
levels of key genes were investigated using the University of ALabama at Birmingham CANcer
data analysis Portal (UALCAN). The association between transcriptional levels of key genes
and survival was analyzed using Kaplan-Meier Plotter (https://www.kmplot.com/analysis/). The
correlations between hub genes and immune cell infiltration were investigated using the Tumor Immune
Estimation Resource (TIMER) database. Specific signaling pathway enrichment analysis was
performed using Gene Set Enrichment Analysis (GSEA) of LinkedOmics. Competitive endogenous
RNA (ceRNA) networks, genome-wide association studies (GWAS), and drug sensitivity analyses
of key genes were further investigated. The expression of ZEB2 was subsequently confirmed in
both parental HCC78 cells and entrectinib-resistant HCC78 cells using real-time quantitative polymerase
chain reaction (qRT-PCR).Results:708 DEGs were identified between entrectinib-resistant CUTO28 (CUTO28-ER) and
parental CUTO28 cell lines in the GSE214715 dataset. One thousand three hundred and ninety-five
DEGs were identified between entrectinib resistant (CUTO37-ER) and parental CUTO37 cell lines
in the GSE214715 dataset. Eight hundred and forty-nine DEGs were identified between LUAD and
non-tumor adjacent tissues in the GSE75037 dataset. Seven hundred and sevety-three DEGs were
identified between NSCLC and non-tumor adjacent tissues in the GSE18842 dataset. Among these
three datasets, seven overlapped DEGs were identified, including ZBED2, CHI3L2, CELF2, SEMA5A,
ZEB2, S100A12, and PDK4. Among these seven overlapped DEGs, the expression levels
of CHI3L2, ZEB2, and S100A12 were downregulated in those three datasets. The results of analysis
using the UALCAN database showed that these three genes were significantly downregulated in
LUAD and LUSC patients compared with the normal population. However, only the lower transcriptional
level of ZEB2 was linked to worse survival in patients with lung cancer. GSEA analysis
revealed that ZEB2 was significantly negatively correlated with nucleotide excision repair (NER)
in LUAD, and homologous recombination (HR) and NER in LUSC, which were linked to drug resistance.
A ceRNA network of THRB-AS1/ has-miR-1293/ ZEB2 in LUAD was established.Conclusion:We have identified core genes associated with non-small cell resistance to entrectinib,
including CHI3L2, ZEB2, and S100A12. ZEB2 is a core gene associated with acquired resistance
to entetinib in NSCLC.