| Background:The incidence of cancer has increasingly affected younger individuals over the time and the number of cancer cases and deaths continues to rise annually.According to the 2022 Cancer Statistics Report by the American Cancer Society,lung cancer is the leading cause of mortality posing a significant health risk.Lung cancer can be classified into two types:small cell lung cancer and non-small cell lung cancer,with squamous lung cancer being a major subtype of the latter.With no chance of surgical cure for advanced squamous lung cancer and a poor overall prognosis for chemotherapy,there is a need to investigate the factors of the prognosis of squamous lung cancer to improve patient survival.In recent years,endoplasmic reticulum stress and angiogenesis have become hot topics in targeted therapy due to their pivotal roles in tumor progression and metastasis.However,many reports have solely emphasized the role of angiogenesis or endoplasmic reticulum stress,with few analyses linking the two.Consequently,there is a pressing need for a combined study to investigate the potential mechanisms affecting the prognosis of lung squamous carcinoma.Objectives:Using bioinformatics methods to investigate the prognostic impact and potential regulatory mechanisms of endoplasmic reticulum stress and angiogenesis-related genes in lung squamous carcinoma.Methods:Gene expression information of lung squamous carcinoma patients and related clinical information were downloaded from the TCGA database.Genes associated with endoplasmic reticulum stress and angiogenesis were identified through the MSigDB database.This study constructed a prognostic gene risk model by using Lasso-Cox regression analysis,and plotted the AUC curve and bar chart for evaluation purposes.The sensitivity of prognostic model-related genes to chemotherapeutic drugs was calculated by CellMiner data.Results:A total of 493 samples were included in the study.Five genes were screened by univariate Cox regression analysis.Based on the screened genes,the study samples were subjected to consistency clustering analysis and classified into two subtypes,A and B.The analysis revealed significant differences between subtypes A and B in terms of survival and functional enrichment.Gene set enrichment analysis showed that the MAPK pathway,JAK pathway,angiogenesis-related pathways such as the VEGF pathway and other genes associated with tumour malignant behaviour were significantly enriched in subtype B,confirming that subtype B has a poorer prognosis than subtype A.Given that predictive genes can classify lung squamous carcinoma into two subtypes with differential A and B,a risk model with 4 prognostic genes was constructed by Lasso-Cox regression analysis,with prognosis-related AUC curve values of 0.592,0.658 and 0.643 at 1,3 and 5 years,respectively.The median risk score was used to divide the investigated population into two groups:high and low risk groups.In order to investigate the correlation between prognostic gene expression and immunity,immune infiltration analysis was performed on the study population and the results showed that there differential expression of immune cells in the two risk groups and that there was a significant correlation between prognostic gene expression and immune cells.Mutations in TP53,TTN,CSMD3 and RYR2 were found to be present in more than 40%of the sample population based on mutation information.Comparative analysis of the relationship between model genes and prognostic sensitivity based on drug and gene expression data revealed that prognostic model-related genes correlated with the sensitivity of chemotherapeutic drugs.Conclusions:The prognostic risk score models and line graphs constructed from four key genes,APOH,CEBPB,WFS1 and VAV2,can predict the prognosis of patients with squamous lung cancer more accurately and these genes are expected to be new therapeutic targets.These genes are expected to be new therapeutic targets. |
PDF Full Text Request