Clinical Validation And Construction Of Prognostic Model Of Alternative Splicing Genes In Non-Small Cell Lung Cancer

Objective:At least 95% of human genes are prone to the process known as alternative splicing(AS),which is common in eukaryotes.There aren’t many studies that are relevant,however previous studies have shown that AS and the development of non-small cell lung cancer(NSCLC)are closely related.The purpose of this work was to identify differential AS genes in NSCLC for clinical validation and to build a prognostic model with clinical applicability based on the identified AS genes.Methods:1.In order to compare the differential expression of AS genes in cancerous and para-cancerous normal tissues and to investigate the AS genes that co-occur in lung adenocarcinoma(LUAD)and lung squamous cell carcinoma(LUSC),the AS data of NSCLC were downloaded from the TCGA Splice Seq database.2.To clinically validate the AS genes that were differently expressed by LUAD and LUSC together in the database,40 cancer tissues and paired para-cancerous normal tissues from NSCLC patients who underwent surgical resection in Chifeng from April 2021 to April2022 were collected.Quantitative real-time polymerase chain reaction(q RT-PCR)was applied to detect AS gene expression in the tissues.3.Download the corresponding clinical data from the TCGA database in accordance with the patient TCGA number found in the alternative splicing data from step 1;combine the two sets of acquired data;screen prognosis-related AS events;perform survival analysis;identify potential prognostic factors for NSCLC and construct a model for predicting survival.4.Statistical software was applied to R software with Graph Pad software,the Mann-Whitney U test was used for non-normally distributed data of paired samples,LASSO regression analysis was used for screening variables,COX regression analysis was used for screening prognostic variables,and the Kaplan-Meier method was used for survival analysis.P<0.05 was considered statistically significant.Results:1.58 pairs of LUAD cancer tissues and 49 pairs of LUSC cancer tissues were chosen using the TCGA Splice Seq database,along with para-cancerous normal tissues.Using the percent splice in(PSI)ratio of cancer tissue to para-cancerous normal tissue,more than 50 pairs of genes with a ratio larger than 2 or less than 0.5 were screened in 58 LUAD samples and more than 45 pairs of genes with a ratio more than 2 or less than 0.5 in 49 LUSC samples.In both LUAD and LUSC cancer tissues,the results showed that the PSI value of the FN1gene(ENST00000323926.10 exon 25 skip)was higher.q RT-PCR was utilized to measure the expression of the FN1 gene(ENST00000323926.10 exon 25 skip)in NSCLC cancer tissues and associated para-cancerous normal tissues.The results demonstrated that the expression of the FN1 gene was significantly higher in para-cancerous normal tissues(ENST00000323926.10 exon 25 skip)than in cancerous tissues(P<0.0001),which is consistent with the database results.2.From the previous two databases,the AS data and clinical data of 58 LUAD patients were downloaded,respectively.By using LASSO regression analysis,3 survival-related AS genes were identified.COX regression analysis found 4 prognostic associations,including age,gender,pathological stage,and risk score,of which pathological stage and risk score were independent prognostic variables for LUAD.The prognostic model for LUAD was constructed using the previous prognostic associations,and its area under the curve(AUC)value of 0.716 and C-index value of 0.70 show an acceptable predictive performance and clinical applicability.Conclusion:1.The PSI value of the FN1 gene(ENST00000323926.10 exon 25 skip)in NSCLC tissues has been shown to be significantly higher than that in para-cancerous normal tissues by screening pairs of samples in the TCGA Splice Seq database.As a result,it suggests that the FN1 gene(ENST00000323926.10 exon 25 skip)may be one of the probable factors influencing the development and progression of NSCLC.2.In this study,we studied two public databases to construct a prognostic model based on the AS genes of LUAD,and we confirmed that its predictive ability and applicability are good for predicting 1-year,3-year,and 5-year survival rates of LUAD patients.This study offers a theoretical foundation for the precise treatment of NSCLC.