Dataset Information

Identification of an independent immune-genes prognostic index for renal cell carcinoma.

ABSTRACT:

Background

Considerable evidence has indicated an association between the immune microenvironment and clinical outcome in ccRCC. The purpose of this study is to extensively figure out the influence of immune-related genes of tumors on the prognosis of patients with ccRCC.

Methods

Files containing 2498 immune-related genes were obtained from the Immunology Database and Analysis Portal (ImmPort), and the transcriptome data and clinical information relevant to patients with ccRCC were identified and downloaded from the TCGA data-base. Univariate and multivariate Cox regression analyses were used to screen out prognostic immune genes. The immune risk score model was established in light of the regression coefficient between survival and hub immune-related genes. We eventually set up a nomogram for the prediction of the overall survival for ccRCC. Kaplan-Meier (K-M) and ROC curve was used in evaluating the value of the predictive risk model. A P value of < 0.05 indicated statistically significant differences throughout data analysis.

Results

Via differential analysis, we found that 556 immune-related genes were expressed differentially between tumor and normal tissues (p < 0. 05). The analysis of univariate Cox regression exhibited that there was a statistical correlation between 43 immune genes and survival risk in patients with ccRCC (p < 0.05). Through Lasso-Cox regression analysis, we established an immune genetic risk scoring model based on 18 immune-related genes. The high-risk group showed a bad prognosis in K-M analysis. (p < 0.001). ROC curve showed that it was reliable of the immune risk score model to predict survival risk (5 year over survival, AUC = 0.802). The model indicated satisfactory AUC and survival correlation in the validation data set (5 year OS, Area Under Curve = 0.705, p < 0.05). From Multivariate regression analysis, the immune-risk score model plays an isolated role in the prediction of the prognosis of ccRCC. Under multivariate-Cox regression analysis, we set up a nomogram for comprehensive prediction of ccRCC patients' survival rate. At last, it was identified that 18 immune-related genes and risk scores were not only tremendously related to clinical prognosis but also contained in a variety of carcinogenic pathways.

Conclusion

In general, tumor immune-related genes play essential roles in ccRCC development and progression. Our research established an unequal 18-immune gene risk index to predict the prognosis of ccRCC visually. This index was found to be an independent predictive factor for ccRCC.

SUBMITTER: Li G

PROVIDER: S-EPMC8240194 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Background: Immune checkpoint inhibitor (ICI) therapy has been proved to be a promising therapy to many types of solid tumors. However, effective biomarker for estimating the response to ICI therapy and prognosis of hepatocellular carcinoma (HCC) patients remains underexplored. The aim of this study is to build a novel immune-related prognostic index based on transcriptomic profiles. Methods: Weighted gene co-expression network analysis (WGCNA) was conducted to identify immune-related hub genes that are differentially expressed in HCC cohorts. Next, univariate Cox regression analysis and least absolute shrinkage and selection operator (LASSO) analysis were used to detect hub genes associated to overall survival (OS). To validate the immune-related prognostic index, univariate and multivariate Cox regression analysis were performed. CIBERSORT and ESTIMATE were used to explore the tumor microenvironment and immune infiltration level. Results: The differential expression analysis detected a total of 148 immune-related genes, among which 25 genes were identified to be markedly related to overall survival in HCC patients. LASSO analysis yielded 10 genes used to construct the immune-related gene prognostic index (IRGPI), by which a risk score is computed to estimate low vs. high risk indicating the response to ICI therapy and prognosis. Further analysis confirmed that this immune-related prognostic index is an effective indicator to immune infiltration level, response to ICI treatment and OS. The IRGPI low-risk patients had better overall survival (OS) than IRGPI high-risk patients on two independent cohorts. Moreover, we found that IRGPI high-risk group was correlated with high TP53 mutation rate, immune-suppressing tumor microenvironment, and these patients acquired less benefit from ICI therapy. In contrast, IRGPI-low risk group was associated with low TP53 and PIK3CA mutation rate, high infiltration of naive B cells and T cells, and these patients gained relatively more benefit from ICI therapy.

Project description:BackgroundRenal cell carcinoma (RCC) is one of the ten most prevalent cancers in the world and its incidence has been rising over the past decade. However, effective biomarkers to predict the prognosis of patients remains absent, and the exact molecular mechanism of the disease remains unclear. Therefore, the identification of key genes and their biological pathways are of great significance to identify the differential expressed genes associated with the prognosis for patients with RCC, and to further explore their potential protein-protein interactions (PPIs) in tumorigenesis.MethodsThe gene expression microarray data for GSE15641 and GSE40435 were extracted from the Gene Expression Omnibus (GEO) database, including 150 primary tumors and their matched adjacent non-tumor tissues. Afterwards, gene expression for fold changes (FCs) and P value for tumor and non-tumor tissues were analyzed using online tool GEO2R. Gene expression with logFCs of greater than two combined with P value of lower than 0.01 were considered as candidate targets for treatment of RCC. The survival analysis of candidate genes was performed by online software OncoLnc. The PPI network was implemented with Search Tool for the Retrieval of Interacting Genes (STRING).ResultsIn total, there were 625 differentially expressed genes (DEGs) in GSE15641, including 415 increased and 210 decreased genes. A total of 343 DEGs were identified in the GSE40435 with 101 upregulated and 242 downregulated genes, the 20 genes with highest FC in high or low expression in each database were summarized. Five candidate genes were overlapped genes in the two GEO datasets. However, aldolase, fructose-bisphosphate B (ALDOB) was found to be the only gene affecting the prognosis. A number of critical genes were identified behind the mechanism, of which they interacted with ALDOB. Among them, phosphofructokinase, platelet (PFKP), phosphofructokinase, muscle (PFKM), pyruvate kinase L/R (PKLR), and fructose-bisphosphatase 1 (FBP1) showed a better prognosis, whereas only glyceraldehyde-3-phosphate dehydrogenase (GAPDH) rendered a bleak outcome.ConclusionsFive genes were found to be overlappingly expressed in the top 20 greatest FC in two human GEO datasets. This is of great value in the treatment and prognosis of RCC.