| Objective: Ovarian cancer(OC)is a gynecological cancer with a high mortality rate,seriously endangering women’s health.Finding biomarkers to predict the prognosis of ovarian cancer is of great value.Panoptosis is a newly discovered pro-inflammatory programmed cell death pathway,which mainly emphasizes the crosstalk between cell pyrosis,apoptosis and necrotic apoptosis.This paper uses bioinformatics to conduct data mining on databases such as TCGA and GTEx,aiming to screen out lncRNAs that are related to panoptosis and have a significant impact on the prognosis of ovarian cancer patients,and use these lncRNAs to build a prognosis prediction model,and verify its effectiveness.The prediction model built in this paper can effectively predict the prognosis and survival of patients and promote the development of precision medicine.Methods: 1.35 panoptosis-related genes were obtained by searching for the keyword "panoptosis" in various database literature.Download the transcriptome information of ovarian cancer,normal ovarian tissue,and clinical-related information of ovarian cancer patients based on the Cancer Genome Atlas database(TCGA)and Genotype-Tissue Expression(GTEx)database.Using the absolute value of differential expression multiple>1(FDR<0.05,|log2Fold Change |>1)as the threshold,we screened differentially expressed genes related to panoptosis.2.Normalize the transcriptome information of ovarian cancer and normal ovarian tissue,and find the lncRNAs co-expressed with differential genes related to panoptosis.The identification criteria are cor Filter>0.4,P<0.001.These lncRNAs are called panoptosis-related lncRNAs.A single factor,LASSO-COX risk regression analysis was performed on panoptosis-related lncRNAs,with the screening criteria being cox Pfilter=0.05,P<0.05.Finally,14 panoptosis-related lncRNAs were selected and analyzed for correlation with 35 panoptosis-related genes.3.Use these 14 panoptosis-related lncRNAs to construct a risk assessment model,obtain the risk score of each patient,and divide the patients into high and low-risk groups based on the median risk score.Use the Kaplan-Meier method to analyze the survival status of patients in the high and low-risk groups.Compare the independent prognostic value of risk scores with clinical pathological features such as patient age,tumor grade,and stage in ovarian cancer patients using univariate and multivariate independent prognostic analysis.4.Use receiver operating characteristic(ROC),consistency index(C-index),and comparison with published prognostic models to verify the predictive effectiveness of the risk assessment model.At the same time,integrate patient clinical information to construct a column chart,and use calibration curves to verify the accuracy of column chart prediction.5.Use R packages such as "ggpubr" and "Limma" to calculate the Tumor Mutational Burden(TMB)score of ovarian cancer patients.Use the Tumor Immune Dysfunction and Exclusion(TIDE)website to obtain the TIDE score of ovarian cancer patients,then explore the correlation between TMB,TIDE,and risk score.Evaluate the correlation between the risk model and tumor-infiltrating immune cells using immune scoring software such as CIBERSORT and TIMER.Analyze and visualize immune checkpoints with significant differential expression between high and low-risk groups through R packets such as "reshape2".Results:1.Differential analysis of panoptosis-related genes expression between normal ovarian tissue and ovarian cancer tissue.35 panoptosis-related genes were included in the study through literature search.Through gene enrichment analysis,the results showed that the enrichment pathway was mainly related to immune response and programmed cell death.At the same time,panoptosis-related genes with differential expression in ovarian cancer and normal ovarian tissues were screened(P<0.05): CASP3,CASP6,CASP8,DDX58,FADD,RIPK1,GSDMD,NFS1,NAIP,NLRP1.2.Screening for lncRNAs related to panoptosis and constructing a risk assessment model.1196 lncRNAs were obtained through co-expression analysis of differential genes in panoptosis.41 prognosis-related lncRNAs with differential expression were selected using univariate COX regression analysis(P<0.05).14 lncRNAs(CAPN10-DT,AC007848.1,MACORIS,AL109615.3,AC020661.1,AC100860.1,LINC00861,AC069120.1,AC006001.2,AC067930.2,AC079866.2,AC008280.2,AL590652.1,LINC02408)were further screened using LASSO-COX analysis to construct a risk model.All 14 panoptosis-related lncRNAs were risk factors with HR>1(P<0.05).Construct a risk assessment model using the regression coefficients and expression levels of panoptosis-related lncRNAs.3.Using survival analysis and other methods to validate the risk assessment model.The median risk score can effectively distinguish between high-risk and low-risk populations(P<0.01).The K-M survival curve shows that compared to the low-risk group,the high-risk group has poorer overall survival and progression-free survival(P<0.001).Using ROC curve analysis to compare the Area Under Curve(AUC)of risk score,patient age,tumor grade,and stage,the results in the entire set were 0.749,0.578,0.525,and 0.523,respectively,indicating that the risk score has better prognostic value compared to existing clinical pathological features.And in the training set,test set,and complete set,the AUC values of the 1-year,3-year,and5-year survival rates of the risk score were all greater than 0.6,indicating that the risk assessment model has great predictive ability for the prognosis of patients.4.Analysis of immune characteristics of patients in high and low-risk groups.The correlation analysis between immune-related cell infiltration and risk score found that follicular helper T cells,resting CD8+T cells were negatively correlated with the risk score.In contrast,M2 macrophages,cancer-related fibroblasts,mast cell were positively correlated with the risk score(P<0.001).Through the analysis of immune checkpoints,it was found that there was a significant difference in the expression of the immune checkpoint gene NRP1 between high and low-risk groups(P<0.001),indicating that it is expected to become a breakthrough point for immunotherapy of ovarian cancer.Conclusion: We constructed a risk assessment model using 14 panoptosis-related lncRNAs and validated its effectiveness and accuracy in predicting the prognosis of ovarian cancer.We found that the model helps evaluate the prognosis of ovarian cancer patients and providing recommendations for their treatment. |
PDF Full Text Request