Prognosis-related genes participate in immunotherapy of renal clear cell carcinoma possibly by targeting dendritic cells

Abstract

Tumor immunotherapy has become one of the most promising approaches to tumor treatment. This study aimed to screen genes involved in the response of clear cell renal cell carcinoma (ccRCC) to immunotherapy and analyze their function. Based on the Gene Expression Omnibus and The Cancer Genome Atlas datasets, we screened out nine differentially expressed genes (TYROBP, APOC1, CSTA, LY96, LAPTM5, CD300A, ALOX5, C1QA, and C1QB) associated with clinical traits and prognosis. A risk signature constructed by these nine genes could predict the survival probability for patients at 1 year, 3 years, and 5 years. The immune checkpoint blockade response rate in the high-risk group was significantly higher than in the low-risk group (49.25% vs. 24.72%, p ≤ 0.001). The nine prognosis-related genes were negatively correlated with activated dendritic cells in the low-risk group but not in the high-risk group. qRT-PCR, immunohistochemistry, and immunofluorescence showed that the nine prognosis-related genes were associated with dendritic cell activity and the PD-1 positive staining rate. In conclusion, the nine prognosis-related genes have a high prognostic value. The patients in the high-risk group were more likely to benefit from immunotherapy, and the mechanism might be related to the release of dendritic cell-mediated immunosuppression.

Keywords: clear cell renal cell carcinoma; dendritic cell; immunotherapy; prognosis; risk score.

FIGURE 1

Screening for DEGs. (A) Screening for DEGs of GSE14994. (B) Screening for DEGs of GSE15641. (C) Screening for DEGs of GSE53757. (D) Venn diagram presents common upregulated DEGs for the three gene microarrays. (E) Venn diagram presents common downregulated DEGs for the three gene microarrays. (F) Venn diagram presents the downregulated and upregulated DEGs common to the three gene microarrays and the TCGA database.

FIGURE 2

Identification of the clinical trait-related genes. (A) Clustering dendrogram of all samples. (B) Analyses of the scale-free index and mean connectivity for multiple soft-threshold powers. (C) Cluster dendrogram of all genes. (D) Correlation heatmap of the module eigengenes and clinical traits. (E) Expression of the 20 clinical trait-related genes based on the TCGA database.

FIGURE 3

Construction of the risk signature with nine prognosis-related genes in the training set. (A) Univariate Cox regression analysis. (B) Multivariate Cox regression analysis. (C) Distribution of the risk score and survival of patients and the expression pattern of the nine prognosis-related genes in the high- and low-risk groups. (D) Receiver operating characteristic (ROC) curves of overall survival (OS) at 1 year, 3 years, and 5 years (E) K–M survival curve of the OS in the high- and low-risk groups.

FIGURE 4

Construction of the risk signature with nine prognosis-related genes in the verification set. (A) Univariate Cox regression analysis. (B) Multivariate Cox regression analysis. (C) Distribution of the risk score and survival of patients and the expression pattern of the nine prognosis-related genes in the high- and low-risk groups. (D) ROC curves of OS at 1 year, 3 years, and 5 years. (E) K–M survival curve of the OS in the high- and low-risk groups.

FIGURE 5

Correlation between the risk score and clinical traits and their predictive value. (A) Multivariate Cox regression analysis for predicting independent prognostic factors. (B) Nomogram based on the Cox regression analysis. (C) Nomogram based on the logistic regression analysis. (D) The calibration curve for 1 year. (E) Calibration curve for 3 years. (F) Calibration curve for 5 years.

FIGURE 6

Functional enrichment analysis of the nine prognosis-related genes.

FIGURE 7

Relationship between the risk score and immunotherapy. (A) TMB score. (B) IFNG. (C) CD8. (D) TAM_M2. (E) CAF. (F) TIDE score. (G) PD-L1. (H) MSI_Exp_Sig. (I) MDSC. (J) IPS score with CTLA4_neg_PD1_pos. (K) IPS score with CTLA4_pos_PD1_pos. (L) Total IPS score. (M) IPS score with CTLA4_neg_PD1_neg. (N) IPS score with CTLA4_pos_PD1_neg.

FIGURE 8

Relationship between the nine prognosis-related genes and immunity. (A) Immune score, stromal score, and ESTIMATE score. (B) Fractions of 22 immune cells in the high- and low-risk groups. (C) Correlation heatmap between the nine prognosis-related genes and significant immune cells in the high-risk group. (D) Correlation heatmap between the nine prognosis-related genes and significant immune cells in the low-risk group. (E) Correlation network between the nine prognosis-related genes and markers of activated dendritic cells in the high-risk group. (F) Correlation network between the nine prognosis-related genes and markers of activated dendritic cells in the low-risk group. × means not significant (C–D). Red represents the nine prognosis-related genes, gray represents the markers with activated dendritic cells, and the thickness of the lines represents the strength of the correlation (E–F).

FIGURE 9

Validation of the relationship between the prognosis-related genes and immunotherapy. (A) Correlation of the prognosis-related genes and the activated dendritic cell markers (red box). (B) Expression levels of the prognosis-related genes in patients grouped with different clinical traits. T1–T4 represents T stage, which refers to the size and extent of the primary tumor, and is represented by T1–T4 in turn. G1–G4 represents clinicopathological grades, and the larger the number, the lower the degree of tumor differentiation, and the higher the degree of malignancy. Stage 1–Stage 4 represents clinical staging. (C) HE staining and immunohistochemistry (PD-1/PD-L1) of low-risk and high-risk patients, 400×. (D) Percentage of positive cells, H-Score, and IRS for immunohistochemical analysis of PD-1/PD-L1. (E) Immunofluorescence of PD-1 (green) and CD11c (red) in low-risk and high-risk patients, 1,000×, white arrows indicate red and green overlap.