Bioinformatics-Based Learning To Identify Prognostic Markers Associated With The Tumor Microenvironment In Metastatic Cutaneous Melanoma

Objective:Skin melanoma is a highly malignant skin tumor in which the morbidity and mortality of metastatic cutaneous melanoma is increasing.The tumor microenvironment(TME)is significantly associated with tumorigenesis and prognosis.Our study aimed to identify prognostic immune-related genes in the tumor microenvironment of metastatic cutaneous melanoma.The relationship between the metastatic SKCM tumor microenvironment and immunotherapy is becoming increasingly important in clinical practice.We assessed prognosis and immune microenvironment by building a model based on tumor microenvironment-associated genes.Methods:The dataset containing RNA sequencing and corresponding clinical data of patients with metastatic cutaneous melanoma was obtained from the GTEx database by searching the TCGA database,GEO database for transcriptomic data of normal tissues.Using TCGA data as the training set and GEO data as the validation set,weighted gene co-expression network analysis(WGCNA)was used to construct a gene co-expression network to screen the gene modules with the highest correlation to a subgroup.Difference analysis was performed to identify the difference genes,and the highest correlation module genes were crossed with the difference genes to obtain the significant difference genes.The non-negative matrix decomposition(NMF)algorithm was used to classify the molecular subtypes and study the distribution characteristics among different subtypes.A risk score model was constructed using univariate Cox regression,least absolute shrinkage and selection operator(LASSO)and multivariate Cox regression to classify patients into high-risk and low-risk groups,and the prediction model was validated by column line plots incorporating clinical characteristics.set was used for an extended validation.Then,the immune microenvironment was evaluated by the CIBERSORT algorithm,while enrichment analysis was performed to explore its biological significance.t IDE was used to predict immune checkpoint blockade responses,mimicking tumor immune evasion mechanisms,and immunophenoscore(IPS)-PD1/PD-L1 blockers and IPS-CTLA4 blocker data to predict the response to immune checkpoint inhibitors(ICI)in metastatic SKCM.Expression levels of immune checkpoint molecules and disease markers were analyzed.Drug sensitivity analysis of genes to study potential therapeutic agents.Results:Two molecular subgroups(group 1 and group 2)with different tumor microenvironment-associated gene expression patterns were identified.A tumor microenvironment-related model based on 5 genes(PLLP,GPRIN1,GPR143,TTYH3 and HSD11B2)was developed through the TCGA database to identify poor prognosis cases.The overall survival rate of patients in group 1 was significantly lower compared to group 2(P< 0.05).Both LASSO regression and Cox regression analyses were used to establish the independence of prognostic models.and was further validated in the GEO cohort.Patients in the low-risk group had better overall survival(OS)than those in the high-risk group(P<0.05).In addition,multivariate Cox regression showed that our constructed hypoxia-related and immune-related prognostic features could be used as independent factors for prognostic prediction(P<0.05).The validity of column line plot prognostic prediction has been well demonstrated in internal and external data validation cohorts,and receiver operating characteristic(ROC)curve analysis demonstrates the accurate predictive performance exhibited by the 5-gene signature for 1-year,3-year,and 5-year survival.The low-risk subgroup showed longer overall survival times.Gene function enrichment analysis revealed major enrichment in various tumor-associated signaling pathways.Using the CIBERSORT algorithm,we found that patients in the highrisk group had lower expression of immune scores,stromal scores,and different immune cell infiltration status compared to patients in the low-risk group.5tumor microenvironment-related signature genes had good predictive value for clinical prognosis and ICI outcome in metastatic SKCM.Low-risk patients may benefit from ICI.In conclusion,tumor microenvironment and immune-related prognostic features could be used as a method to stratify the risk of metastatic SKCM.Finally,sensitivity drugs for each subgroup were identified by pharmacovigilance analysis.Conclusion:The tumor microenvironment is associated with the prognosis and occurrence of immune cell infiltration in patients with metastatic SKCM,and our establishment of a 5-gene signature based on the tumor microenvironment(including PLLP,GPRIN1,GPR143,TTYH3,and HSD11B2)to predict survival outcome and immunotherapy response could provide an immunological perspective for the development of personalized therapy.