Prognostic analysis of cuproptosis-related gene in triple-negative breast cancer

Abstract

Background: Cuproptosis is a copper-dependent cell death mechanism that is associated with tumor progression, prognosis, and immune response. However, the potential role of cuproptosis-related genes (CRGs) in the tumor microenvironment (TME) of triple-negative breast cancer (TNBC) remains unclear.

Patients and methods: In total, 346 TNBC samples were collected from The Cancer Genome Atlas database and three Gene Expression Omnibus datasets, and were classified using R software packages. The relationships between the different subgroups and clinical pathological characteristics, immune infiltration characteristics, and mutation status of the TME were examined. Finally, a nomogram and calibration curve were constructed to predict patient survival probability to improve the clinical applicability of the CRG_score.

Results: We identified two CRG clusters with immune cell infiltration characteristics highly consistent with those of the immune-inflamed and immune-desert clusters. Furthermore, we demonstrated that the gene signature can be used to evaluate tumor immune cell infiltration, clinical features, and prognostic status. Low CRG_scores were characterized by high tumor mutation burden and immune activation, good survival probability, and more immunoreactivity to CTLA4, while high CRG_scores were characterized by the activation of stromal pathways and immunosuppression.

Conclusion: This study revealed the potential effects of CRGs on the TME, clinicopathological features, and prognosis of TNBC. The CRGs were closely associated with the tumor immunity of TNBC and are a potential tool for predicting patient prognosis. Our data provide new directions for the development of novel drugs in the future.

Keywords: cuproptosis; immunotherapy; triple-negative breast cancer; tumor microenvironment; tumor mutation burden.

Figure 1

Landscape of genetic variation of Cuprotosis related genes in BRCA (A) Mutation frequency of 16 CRGs in 986 BRCA from TCGA and GEO combined samples. The number on the right indicated the mutation frequency of each gene. The bar chart on the right showed the proportion of mutations. The stacked bar chart below shows the fraction of conversions. (B) The position of CRGs CNV changes on 23 chromosomes. (C) Interaction between CRGs in BRCA. The line connecting CRGs indicated their interaction, and the thickness of the line indicated the correlation strength between CRGs. Purple and green represent negative and positive correlation respectively. (D) The figure of the PCA of the TCGA and GEO (GSE13356, GSE65194, GSE58812) datasets. (E) CNV mutation frequency of CRGs. The deletion frequency, blue dot; The amplification frequency, red dot. BRCA, breast cancer; CRGs, cuprotosis related genes; CNV, copy number variant.

Figure 2

Enrichment of five prognosis-related CRGs (A) Gene ontology enrichment (B) KEGG metabolic pathway Enrichment (C) Disease enrichment.

Figure 3

CRG subtypes and clinicopathological and biological characteristics of two distinct subtypes of samples divided by consistent clustering. (A) Consensus matrix heatmap defining two clusters (k = 2) and their correlation area. (B) relative change area under cumulative distribution function curve. (C) PCA analysis revealed significant differences in transcriptome between the two subtypes. (D) Kaplan-Meier curve showed that there were significant survival differences between cluster A and B (P=0.041). (E) The abundance of each TME infiltrating cell in two clusters. The upper and lower ends of the boxes represented interquartile range of values. The lines in the boxes represented median value, and black dots showed outliers. The asterisks represented the statistical p value (*P < 0.05; **P < 0.01; ***P < 0.001). (F) Differences in clinicopathologic features and expression levels of CRGs between the two distinct subtypes. (G) GSVA of biological pathways between two distinct subtypes, in which red and blue represent activated and inhibited pathways, respectively.

Figure 4

Generation and functional annotation of CRGs (A, B) Two subgroups was identified the optimal value for consensus clustering,and was designated genecluster 1 (C1) and genecluster 2 (C2). (C) Survival curve of the patients between C1 and C2(P = 0.034). (D) Differences in clinicopathologic features and expression levels of CRGs between the C1 and C2. (E) The expression of 3 CRGs between C1 and C2. The asterisks represented the statistical p value (**P < 0.01; ***P < 0.001).

Figure 5

Construction of the prognosis risk prediction model (A, B) The least absolute shrinkage and selection operator (LASSO) regression was performed with the minimum criteria. (C, D) Differences in CRG_score between gene subtypes and cuprotosis subtypes (E) Differences in clinicopathologic features and expression levels of CRGs between the high-and low-risk group. (F) Alluvial diagram of subtype distributions in groups with different CRG_scores and survival outcomes. (G) Six differential expressions of CRGs in high and low risk groups. The asterisks represented the statistical p value (*P < 0.05; **P < 0.01; ***P < 0.001).

Figure 6

Prognosis value of the risk model in the train, test, and entire sets (A–C) Kaplan–Meier survival curves of survival probability of patients between low-and high-risk groups in the train,test, and entire sets, respectively. (D–F) Exhibition of CRGs model based on risk score ofthe train, test, and entire sets, respectively. (G–I) Survival time and survival status between low-and high-risk groups in the train, test, and entire sets, respectively. (J–L) ROC curves to predict the sensitivity and specificity of 1-, 3-, and 5-year survival according to the CRG_score in the train,test, and entire sets, respectively. (M, N) Survival probability in the high-risk and low-risk subgroups of the GEO cohort and TCGA cohort, respectively.

Figure 7

Independence detection of the constructed risk prediction model (A) Relationships between CRG_score and Age. (B) Relationships between CRG_score and Position. (C–E) Kaplan–Meier survival curves of survival probability prognostic value stratified by age and stage. (F, G) Uni- and multi-Cox analyses of clinical factors and risk score with OS.

Figure 8

Comparison of immune activity between subgroups (A) The immune cell bubble of risk groups (B) Correlations between the abundance of immune cells and 2 genes in the proposed model. (C) Correlations between CRG_score and both immune and stromal scores. (D) Correlations between CRG_score and TIDE, MSI, Dysfunction and Exclusion. (E) Differences in immune scores, stromal scores and expression levels of immune cells between the high-and low-risk group. (F) Correlation of 12 immune checkpoint gene risk scores. (G–I) The association between IPS and the CRGs based on TCIA database, (G) CTLA4– PD1–(H) CTLA4– PD1+ (I) CTLA4+ PD1– (J) CTLA4+ PD1+. The asterisks represented the statistical p value (*P < 0.05; **P < 0.01; ***P < 0.001).

Figure 9

Genetic characteristics of CRG_score and tumor somatic mutation of TNBC (A, B) The waterfall plot of tumor somatic mutation established by those with high-and low-risk group. (C) The overall survival of H-TMB and L-TMB using Kaplan–Meier in Log-rank test. (D) The overall survival of the patients stratified by both the CRG-score signature and TMB using Kaplan–Meier curves. (E, F) Correlations between CRGs and Pathway Through KEGG and HALLMARK Enrichment Analysis. The asterisks represented the statistical p value (*P < 0.05; **P < 0.01; ***P < 0.001).

Figure 10

Construction and Evaluation of Nomogram Based on CRG_score (A) Nomogram for predicting the 1-, 3-, and 5-year OS of TNBC patients. (B) Calibration curves of the nomogram. (C–E) ROC curves for predicting the 1-, 3-, and 5-year ROC curves in the train, test, and entire sets. The asterisks represented the statistical p value (*P < 0.05; **P < 0.01; ***P < 0.001).

Figure 11

Drug sensitivity (A–H) Estimated drug sensitivity in patients with high and low FRLM risk.

Figure 12

Inhibition of ATP7A restrained proliferation and migration capacities of breast cancer cells (A) Western blot analysis revealed the efficiency of ATP7A knocking-down in BT549 and MDA-MB-231 cell lines. (B) CCK-8 cell proliferation assay after ATP7A knockdown in BT549 and MDA-MB-231 cell lines. (C) The Transwell invasion assay showed that knocking-down of ATP7A inhibited the cellular invasion of the BT549 and the MDA-MB-231.Graphical representation of the number of invasive BT549 and MDA-MB-231 cells per microscopic field. Data were shown as the mean ± SD from three independent experiments. **P < 0.01, vs. control group.