Clinical Significance, Immune Landscape and Immunotherapy Efficacy Analyses of Cuproptosis-Tumor Immunological Phenotype-Related Gene Score Based Prognostic Model in Breast Cancer

doi:10.21203/rs.3.rs-4617378/v1

Download PDF

Article

Clinical Significance, Immune Landscape and Immunotherapy Efficacy Analyses of Cuproptosis-Tumor Immunological Phenotype-Related Gene Score Based Prognostic Model in Breast Cancer

https://doi.org/10.21203/rs.3.rs-4617378/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Breast invasive carcinoma (BRCA) is the most commonly diagnosed cancer in women and exhibits the highest mortality rate among female cancers. Furthermore, the tumor microenvironment (TME) plays a crucial role in the progression of BRCA. Copper ions are essential metal elements in vivo, and cuproptosis, a recently identified mode of cell death, significantly influences tumor progression, metastasis, and angiogenesis. Utilizing 10 cuproptosis-related genes (CRGs), we developed a novel cuproptosis-tumor immunological phenotype-related gene score (CTIPRGs) comprising 14 genes through Univariate Cox (UniCOX) regression and the least absolute shrinkage and selection operator (LASSO) algorithm to predict prognosis and response to immunotherapy in BRCA. Univariate and multivariate analyses indicated that CTIPRGs serve as an independent prognostic predictor. Subsequently, we constructed nomograms to provide quantitative tools for clinical practice. Kaplan-Meier (KM) plot analysis demonstrated that CTIPRGs could significantly distinguish overall survival (OS) in BRCA patients, with low-CTIPRGs groups exhibiting better prognosis. High and low-CTIPRGs groups exhibited distinct biological behaviors, mutational characteristics, and immune cell infiltration patterns. Notably, low-CTIPRGs groups were classified as an immune-inflamed phenotype with abundant immune cells, whereas high-CTIPRGs groups were identified as immune-desert phenotypes characterized by extensive stromal cell infiltration. Additionally, notable differences in tumor immunogenicity, tumor immune dysfunction and exclusion (TIDE), and tumor mutation burden (TMB) were observed between high and low-CTIPRGs groups, indicating that low-CTIPRGs groups exhibited higher sensitivity to immunotherapy. In conclusion, the identified CTIPRGs can facilitate the prediction of prognosis and treatment efficacy in BRCA patients.

Biological sciences/Cancer

Biological sciences/Immunology

Health sciences/Oncology

Cuproptosis

breast cancer

tumor immunological phenotype

tumor microenvironment

immunotherapy efficacy

drug sensitivity

Copper is one of the essential metal elements involved in human biological processes. It plays a crucial role in various catalytic processes and intermolecular interactions, including energy metabolism [1], oxidation resistance [2], transmitter release [3, 4] and Synaptic transmission12 [5]. In vivo, cooper ion has two oxidation states:oxidized Cu²⁺ and reduced Cu¹⁺ [6]. In biological systems, copper is predominantly present as Cu2 + due to the susceptibility of Cu1 + to multiple oxidants in vivo [7], and they circulate in a mutually stable state. Disturbance of copper homeostasis can lead to various diseases [8, 9], which includes neurological disorders such as Wilson disease, Menkes disease, Occipital horn syndrome, and distal hereditary motor neuropathy, as well as the development of tumors and metastasis[10]. Moreover, copper ions can stabilize nuclear hypoxia-inducible factor-1 (HIF-1), thereby further facilitating the expression of proangiogenic factors [11, 12]. A recent study[13] revealed a novel cell death mode, termed cuproptosis, which is independent of apoptosis, necroptosis, pyroptosis, and ferroptosis. Cells involved in mitochondrial respiration and the TCA cycle are the primary sites of copper-induced cell death. The combination of lipoylated TCA cycle enzymes with copper can lead to the aggregation of lipoylated proteins and the loss of Fe-S cluster-containing proteins; the induction of HSP70 subsequently results in cell death. However, few reports have addressed the effects and prognostic value of CRGs in BRCA.

Worldwide, breast cancer is the most prevalent cancer among women and has one of the highest mortality rates among female malignancies. In 2017, it was estimated that approximately 276,480 women were diagnosed with breast cancer in the United States [14]. Currently, various treatments are available for breast cancer, including surgery, radiotherapy, adjuvant chemotherapy, endocrine agents, and immunotherapy. However, the majority of patients are resistant to treatment [15, 16]. With intensive research on the TME, which is now widely recognized as a critical determinant of breast cancer progression and response to anti-neoplastic therapy, the composition of the TME in BRCA is complex, including fibroblasts, immune cells, endothelial cells, infiltrating inflammatory cells, adipocytes, and the extracellular matrix (ECM), and it varies with tumor progression, exhibiting considerable heterogeneity[17]. Breast tumors are not generally considered to be immunogenic. Tumor infiltrating lymphocytes (TILs) are the most common mononuclear immune infiltrates in breast cancer[18], and high TIL counts are associated with improved prognosis and better treatment response, especially in patients with triple-negative breast cancer[19, 20]. Consequently, an increasing number of researchers have begun to explore the association between TILs and immunotherapy and adjuvant chemotherapy, as well as the effectiveness of TILs as a prognostic factor. However, despite the continued progress in understanding the role of the TME in BRCA and the proposal of numerous potential therapeutic targets, only a few treatment options are clinically applicable, and targeting key microenvironmental hallmarks of cancer may be associated with widespread toxicity[21]. Therefore, precisely identifying individual subgroups and providing them with personalized treatment has become a challenging issue.

Tumor immunological phenotype (TIP) is an emerging concept. Solid tumors can be classified into ‘hot’ (‘immune-inflamed’) phenotypes, characterized by predominantly T-lymphocyte infiltration, and ‘cold’ (‘immune-desert’) phenotypes, characterized by the absence or exclusion of T cells. A thorough understanding of hot and cold tumors can enhance our ability to identify responders to immune checkpoint blockade (ICB)[22], and further converts cold tumors to hot tumors for the benefit of nonresponders. Recently, Wang et al. [23] used an information extraction (IE) method to identify 15 TIP-associated genes, including 12 hot tumor-related genes (CXCL9, CXCL10, CXCL11, CXCR3, CD3, CD4, CD8a, CD8b, CD274, PDCD1, CXCR4, and CCL5) and 3 cold tumor-related genes (CXCL1, CXCL2, and CCL20). These genes were significantly associated with the survival outcomes of cancer patients and exhibited great potential to enhance cancer immunotherapy. Therefore, the clinical application of TIP-related genes is highly promising.

Numerous immune infiltration-based signatures have been established to predict the prognosis of BRCA. However, the predictive accuracy of most signatures remains inadequate for clinical application. Moreover, few studies have explored the combined value of the novel cell death modality, cuproptosis, and immune profiles in predicting prognosis and treatment response in breast cancer, and the interrelationship between cuproptosis and the TME remains unclear. Therefore, in this study, we innovatively utilized cuproptosis and TIP-related features to co-construct CTIPRGs for predicting prognosis and treatment response in BRCA patients. Additionally, our results indicated that the low-CTIPRGs groups exhibited a better prognosis, greater immune cell infiltration, and a higher response to immunotherapy compared to the high-CTIPRGs groups. In conclusion, our proposed CTIPRGs, the first of their kind, could elucidate the combined role of cuproptosis and immune-related features in BRCA and may serve as a novel therapeutic guidance tool for BRCA patients.

2.1 Data Source

To obtain human breast cancer messenger RNA (mRNA) expression profiles and clinical information, we downloaded three Gene Expression Omnibus (GEO) datasets with available overall survival times and a minimum of 100 samples from the GEO database (Supplementary Table S1). We also obtained The Cancer Genome Atlas (TCGA) BRCA raw RNA-seq data and its corresponding clinical information from the UCSC Xena webserver (Supplementary Table S2). Patients with overall survival times of less than 30 days were excluded due to potential other causes of mortality[24]. The detailed clinical information was listed in Supplementary Table S3. Subsequently, RNA-seq gene expression data from 1039 patients in the Counts format were obtained from the UCSC Xena webserver(https://xenabrowser.net/), and then were transformed into Transcripts per Kilobase Million (TPM) as well as further subjected to log2 transformation for normalization. In addition, we obtained transcriptomic microarray dataset (GSE20685、GSE42568、GSE88770) [25–27] from the GEO database(https://www.ncbi.nlm.nih.gov/geo/) with 548 patients. And we obtained normal breast samples from the GTEX database (https://www.gtexportal.org/).The “ComBat” algorithm [28, 29] was applied to reduce the likelihood of batch effects from non-biological technical biases between different datasets.

2.2 Consensus Clustering of BRCA Patients

Ten CRGs were obtained from the previous report (FDX1, LIAS, LIPT1, DLD, DLAT, PDHA1, PDHB, MTF1, GLS, CDKN2A)[13]. Based on the expression profiles of 10 CRGs, we performed unsupervised clustering analysis to group 1039 patients in the TCGA-BRCA database using the “ConsensusClusterPlus” R package (version 2.60)[30]. We performed 1000 iterations to ensure a stable classification[31].

2.3 Gene set variation analysis (GSVA)

The GSVA enrichment analysis was performed to elucidate the pathway differences among different CRGs clusters by using the R package of “GSVA” (version 2.11) [32]. We downloaded “c2.cp.kegg.v7.5.1.symbols” and “h.all.v7.1.symbols”files from the MSigDB database to carry out GSVA analysis [33, 34]. An adj.P.Val of < 0.05 and absolute Log2 FC > 0.1 was considered to indicate statistical significance between different clusters by the “limma” package [35].

2.4 Construction of TIP Score

According to the previous report we have calculated the TIP score and made some modifications[23]. Briefly, we extracted the gene expression profiles of 12 hot tumor-associated genes (CXCR3, CXCR4, CXCL9, CXCL10, CXCL11, CCL5, CD3, CD4, CD8a, CD8b, CD274, and PDCD1) and predefined 3 cold tumor-associated genes (CXCL1, CXCL2, and CCL20). Then, the TIP score was derived from the z-score summary of the expression of tumor immunological phenotype genes. The estimate score, stromal score, immune score, and tumor purity for all patients in the TCGA-BRCA database were calculated using the ESTIMATE package in R (version 1.0.13) [36, 37].

2.5 Differentially Expressed Genes (DEGs)

Based on the results of the previous consensus clustering algorithm, we identified cuproptosis-related differentially expressed genes (CRDEGs). CRDEGs with an adjusted p-value < 0.05 and an absolute Log2 fold change (FC) > 1.0 were considered significant and used for further analysis.

2.6 Weighted Gene Co-Expression Network Analysis (WGCNA) and Development of CTIPRGs

WGCNA was performed on the expression data of CRDEGs to identify co-expression modules using the R package of “WGCNA” [38]. We removed outlier samples, and the optimal soft threshold power was identified to construct a weighted adjacency matrix, which was further transformed into a topological overlap matrix (TOM). The 1-TOM matrix was computed and used for gene clustering and module recognition with a minimum cluster size of 50. Similar dynamic modules were merged using a cut-off of 0.25. Subsequently, we calculated and visualized the Pearson’s correlation coefficients (PCC) and p-values between the modular signature genes and clinicopathological parameters using a heatmap. The most significant modules correlated with the TIP score were identified for further analysis. Subsequently, univariate Cox regression and the LASSO algorithm were applied to identify the optimal subset of prognostic genes using the “glmnet” R package [39]. Then, the CTIPRGs were constructed by utilizing the regression coefficients derived from univariate Cox regression analysis and multiplying them by the expression levels of the prognostic genes: CTIPRGs = Σ(coefI × ExpI), where CoefI and ExpI represent the risk coefficient and expression level of each gene, respectively.

2.7 Survival Analysis

The CTIPRGs were calculated for each patient in the TCGA-BRCA training set and the GSE20685, GSE42568, and GSE88770 validation sets. Next, we classified patients into high-CTIPRGs and low-CTIPRGs groups based on the optimal cutoff value determined from the training set using the R package “survival”. Kaplan–Meier analysis was performed to explore the prognostic significance of the CTIPRGs, using a log-rank test. Additionally, to validate the prognostic value of the CTIPRGs model, univariate and multivariate analyses were conducted to identify the independent prognostic indicators for BRCA.

2.8 Construction of a nomogram model

A nomogram model was constructed to quantitatively evaluate the prognostic risk based on the result of univariate analysis using the “arms” R package (version 6.3.0) [40]. Calibration curves for the 1-、3- and 5-year were drawn to examine the predictive capability of the nomogram.

2.9 Mutation Landscape

To determine the pattern of somatic mutations in BRCA patients in the high and low-CTIPRGs groups, a mutation annotation format (MAF) file for TCGA-BRCA was generated from the TCGA data portal (http://tcga-data.nci.nih.gov/tcga/) using the R package “maftools” (version 2.12.0) [41]. We also calculated the tumor mutation burden (TMB) score for each patient and performed Kaplan–Meier (KM) survival analysis for the high-CTIPRG and low-CTIPRG groups [42, 43].

2.10 Characteristics of Tumor Microenvironment(TME)

To investigate the relative infiltration of TME cells in the high and low-CTIPRGs groups, the ssGSEA algorithm was applied to analyze the infiltration levels of different immune cells using gene sets of 28 reported immune cells.[44] and two stromal components (fibroblasts and endothelial cells) [45]of TME. The relationship between CTIPRGs and each type of TME cell was assessed using Spearman correlation analysis. The cytolytic activity index (CYT), which characterizes the antitumor activity of cytotoxic T cells and NK cells within the tumor, was calculated from the geometric mean expression levels of GZMA and PRF1[46].

2.11 Prediction of Drug Sensitivity and Therapeutic Response

According to the previous study, we identified differences in the expression of potential immunotherapeutic markers including 49 ICB-related genes[47–49], IFN-gamma pathway markers[50] and m6A regulators[51, 52] in high-and low-CTIPRGs groups. Based on the Genomics of Drug Sensitivity in Cancer (GDSC) database, the “pRRophetic” R package (version 0.5) was used to explore the drug sensitivity of the two patient groups.[53]. The Tumor Immune Dysfunction and Exclusion (TIDE) algorithm [54] (http://tide.dfci.harvard.edu/) was used to infer the efficacy of immunotherapy in patients with TCGA-BRCA.

2.12 Immunohistochemical(IHC) staining

The 10 BRCA tissue samples and paired normal breast tissue samples were obtained from patients at the First Affiliated Hospital of Xi'an Jiaotong University. None of the patients had received antitumor therapy prior to specimen collection. Clinical samples were collected and analyzed in accordance with the Declaration of Helsinki, with approval from the Ethics Committee of the First Affiliated Hospital of Xi'an Jiaotong University. Immunohistochemical staining was performed using CCL8 antibody (1:150, 300-15-500UG, Thermo) and GZMB antibody (1:150, ab255598, Abcam). The percentage of positive cells and the intensity of staining in the stained sections were scored by two pathologists. Staining intensity was categorized as 0 (negative), 1 (yellowish), 2 (tan), and 3 (sepia). The percentage of positive cells was categorized as 1 (0–25%), 2 (26–50%), 3 (51–75%), and 4 (> 75%). The two scores were multiplied to obtain the final score.

2.13 Statistical Analyses

Spearman correlation analysis was used to calculate correlation coefficients. Kaplan–Meier (KM) analysis was performed to assess the significance of differences in survival curves using the “survival” R package (version 3.4-0) with a log-rank test. The Kruskal-Wallis test was used to compare more than two groups of data, and Wilcoxon tests were utilized to compare two groups. Univariate and multivariate Cox regression analyses were conducted to identify independent prognostic factors. All statistical analyses were conducted using R (version 4.2.0). A p-value < 0.05 was considered statistically significant (*P < 0.05; **P < 0.01; ***P < 0.001).

2.14 Ethics Statement

The study was approved by the Ethics Committee of the First Affiliated Hospital of Xi'an Jiaotong University. Clinical samples were collected and analyzed in accordance with the Declaration of Helsinki. Informed consent was obtained from all patients.

3.1 Identification of BRCA Subtypes Based on 10 Cuproptosis-Related Genes

Utilizing the expression of 10 cuproptosis-related genes (CRGs), we applied an unsupervised clustering algorithm to categorize 1039 tumor samples from TCGA-BRCA into k clusters (k = 2–9). The findings indicate that k = 5 (Fig. 1A, B) is the optimal selection for categorizing the entire dataset, with 235 samples in cluster A, 96 samples in cluster B, 271 samples in cluster C, 106 samples in cluster D, and 331 samples in cluster E (Supplementary Table S4). Survival analysis of the five subtypes (Fig. 1C) indicated that cluster D exhibited the poorest prognosis. Subsequently, to investigate the cancer hallmark pathways (Supplementary Table S5) linked with the five clusters, we performed GSVA enrichment analysis and observed heightened activity of cluster D in cancer progression (Fig. 1D), encompassing Myc targets, the mTOR signaling pathway, and the WNT-β-catenin signaling pathway. Furthermore, tumor proliferation-related pathways, including DNA repair, E2F targets, and the G2M checkpoint, exhibited substantial enrichment. Subsequent analysis revealed significant differences in the expression of 10 CRGs across all five BRCA clusters (Fig. 1E). Immune infiltration analysis of BRCA (Fig. 1F) demonstrated that cluster B exhibited a higher immune score compared to the other clusters. Similarly, the boxplot (Fig. 1G) illustrated the levels of immune cell infiltration among the different clusters as inferred by the ssGSEA algorithm.

3.2 TIP Was Related to Prognosis and TME of BRCA

Using KM survival analysis (Supplementary Table S6), we divided BRCA patients into high and low TIP score groups based on the optimal TIP score cutoff to determine the relationship between TIP score and prognosis. Approximately 44.3% (460/1039) of patients were identified as having high TIP scores, which correlated with better prognosis (Fig. 2A). Subsequently, we investigated the correlations between TIP score and immune score, stromal score, estimate score, and tumor purity (Supplementary Table S7). As shown in Fig. 2B, TIP score exhibited a strong positive correlation with immune score, stromal score, and estimate score, but a negative correlation with tumor purity. Additionally, we identified a strong positive correlation between TIP score and three immune checkpoints (Fig. 2C). Moreover, the BRCA molecular subtypes showed significant differences in TIP scores (Fig. 2D).

3.3 Development of CTIPRGs

To identify the most relevant gene modules for the TIP score, we utilized the WGCNA algorithm to evaluate expression data from the TCGA-BRCA database. We first identified 3,153 CRDEGs across five BRCA clusters (Supplementary Table S8), which were then extracted to construct co-expression networks and modules. One sample was removed prior to network construction (Supplementary Fig. 1A). Co-expressed gene modules were identified with an optimal soft-thresholding power value set to 5 (scale-free R^2 = 0.9) (Supplementary Fig. 1B). Subsequently, seven distinct modules were identified (Fig. 3A), with the heatmap plot of the clustering dendrogram among these modules shown in (Supplementary Fig. 1C). Next, we calculated the correlations between module eigengenes (MEs) and various variables using Pearson correlation coefficient (PCC) (Fig. 3B). We noted that the green module exhibited a highly positive correlation with the TIP score (PCC = 0.92, P < 2.2e-16), encompassing 178 genes. We also presented the strong correlation between green module genes and selected module membership using a scatterplot (Fig. 3C). We then performed univariate Cox regression analysis to identify genes associated with BRCA patient overall survival (OS). Among the 178 genes, 51 with a P-value < 0.05 were selected for further analysis (Supplementary Table S9). Through LASSO regression analysis with minimized lambda (Figs. 3D, E), 14 robust genes (KLHDC7B, MMP12, IL4I1, LIMD2, GZMB, TOGARAM2, CCL8, GPR84, GPR25, IGLV1-44, IGHG1, IGLV6-57, IGHV1-69, IGHD) were identified and used to construct CTIPRGs. Figure 3F shows the univariate Cox results of the selected 14 genes with their corresponding hazard ratios (HRs) and statistical significance.

3.4 Evaluation and Validation of Survival Efficacy by CTIPRGs

To better validate the stability of CTIPRGs, we calculated the CTIPRGs for each patient and divided the TCGA-BRCA (training dataset) into high- and low-CTIPRGs groups based on the optimal cutoff of the TCGA-BRCA database (Fig. 4A and Supplementary Table S4). As shown in Fig. 4B, compared to high-CTIPRGs groups, low-CTIPRGs groups had a better prognosis. Using the cutoff value from the training dataset, we also divided the external validation cohort (GSE20685, GSE42568, GSE88770) into high- and low-CTIPRGs groups. Consistent with the training datasets, the high-CTIPRGs groups in the external validation cohort had worse overall survival (OS) (Supplementary Fig. 2). To determine whether CTIPRGs is an independent prognostic indicator for BRCA, univariate and multivariate analyses were performed considering clinicopathological features. As shown in Figs. 4C, D, the hazard ratio (HR) of CTIPRGs was 1.368 (95% CI = 1.161 − 1.613) in the univariate analysis and 1.390 (95% CI = 1.170 − 1.660) in the multivariate analysis. These results suggest that CTIPRGs could be an independent prognostic factor in BRCA. To quantify the risk assessment of BRCA patients and provide clinicians with a quantitative approach to patient prognosis, a nomogram was constructed using three indicators (CTIPRGs, stage, and age) to predict individual 1-year (12 months), 3-year (36 months), and 5-year (60 months) OS. As shown in Fig. 4E, higher CTIPRGs are associated with worse prognosis. Meanwhile, the calibration plot (Fig. 4F) showed consistency between the predicted 1-year, 3-year, and 5-year OS and the actual outcomes.

3.5 The Immune Status between High-and Low CTIPRGs Groups

The tumor is a complex mass composed of proliferating tumor cells and various non-cancerous cells. These cells constitute the tumor microenvironment (TME) through the secretion of cytokines, chemokines, and the extracellular matrix (ECM), influencing both the prognosis of patients and the effectiveness of specific treatments. [55]. To further explore the differences in the infiltration of TME cells between high- and low-CTIPRGs groups, we conducted ssGSEA analysis (Supplementary Table S10). As shown in Fig. 5A, low-CTIPRGs groups exhibited enriched infiltration of adaptive and innate immune cells, including activated B cells, activated CD4 T cells, immature B cells, regulatory T cells, T follicular helper cells, activated dendritic cells, eosinophils, immature dendritic cells, MDSCs, and others. In contrast, high-CTIPRGs groups were abundant in epithelial cells and fibroblasts compared to low-CTIPRGs groups. Consistently, the boxplot (Supplementary Fig. 3) showed that the low-CTIPRGs groups had a higher immune score. In addition, Fig. 5B illustrated that, except for neutrophils, fibroblasts, and endothelial cells, CTIPRGs were negatively correlated with all the remaining cells (P-value < 0.05). Finally, to investigate the differences in biological functions between high- and low-CTIPRGs groups, we performed GSVA in these groups (Fig. 5C and Supplementary Table S11). High-CTIPRGs groups were mainly enriched in metabolism-related biological processes such as propanoate metabolism, butanoate metabolism, nitrogen metabolism, fatty acid metabolism, and others. However, immune-related pathways such as NOD-like and Toll-like receptor signaling pathways, T cell receptor signaling, RIG-like receptor signaling pathways, primary immunodeficiency, and others were more active in low-CTIPRGs groups. In conclusion, low-CTIPRGs groups can be categorized as an immune-inflamed phenotype with abundant immune cells, while high-CTIPRGs groups can be classified as an immune-desert phenotype characterized by stromal cell infiltration.

Next, we focused on the correlations between CTIPRGs and immune- and stromal-related characteristics. Figure 5D illustrated a significant negative association between CTIPRGs and CYT, immune score, stromal score, and estimate score, while showing a strong positive correlation with tumor purity. Three important immune checkpoint molecules, PDCD1, CD274, and CTLA4, were also investigated for associations with CTIPRGs in the TCGA cohort (Fig. 5D). Extracellular matrix protein 2 (ECM2) is a member of the secreted protein acidic and rich in cysteine (SPARC) family of proteins, playing important roles in cell proliferation and differentiation. [56–58]. There was a significant negative correlation between ECM2 and CTIPRGs (Fig. 5D) (P-value < 0.05). Colony-stimulating factor 1 receptor (CSF1R) is not only required for the proper development of cell populations but is also a key signaling node in microglia and macrophage biology [59, 60]. While Intercellular adhesion molecule 1 (ICAM1) belongs to the Ig superfamily and consists of five extracellular Ig domains, it can not only regulate the recruitment of leukocytes and the barrier function of endothelial and epithelial, but also impact the development and metastasis of tumors [61]. Moderate Spearman’s correlation between the expression level of CTIPRGs with ICAM1 and CSF1R was found in the TCGA breast cancer cohort (Fig. 5D).

In terms of molecular subtypes of BRCA (Fig. 5E), The Basal-like subtype had the lowest CTIPRGs among the four subtypes, whereas patients with the Luminal A subtype had the highest CTIPRGs values. Moreover, lower CTIPRGs were associated with increased immune cell infiltration. Consistently, Muenst et al. [62] found that the prevalence of PD-1 + tumor-infiltrating lymphocytes was the lowest in the Luminal A subtype while the highest in the Basal-like subtype. The previous study [63] found that the Basal-like subtype was considered the most inflamed subtype whereas the Luminal A subtype had a non-inflamed phenotype. This may explain why Basal-like subtypes had more immune infiltration. These results might provide some different immunotherapeutic options for the molecular subtypes of the BRCA.

3.6 T cell infiltration and Antitumor Immune Response

CD8 + T cells can engage tumor cells in an antigen-specific manner, secreting effector cytokines, releasing cytotoxic molecules (such as granzyme B and perforin), and inducing apoptosis to exert anti-tumor immunity. Additionally, CD8 + T cells can release CXC-chemokine receptor 3 (CXCR3), which is the receptor for the TH1-type chemokines CXC-chemokine ligand 9 (CXCL9) and CXCL10. [64]. These chemokines can inhibit tumor angiogenesis as well as promote the migration of effector T cells and NK cells into tumors [65–68]. As known cytokines, CCL4 and CCL5 can control the migration of cDC1 into tumors to initiate CD8 + T [69–71]. As shown in Fig. 6A, the expression levels of CCL4, CCL5, CXCR3, CXCL10, and CXCL9 in low-CTIPRGs groups were significantly higher than those in high-CTIPRGs groups. Furthermore, a strong negative association was observed between CTIPRGs and the chemokines and chemokine receptor (Supplementary Fig. 4A).

To quantify the differences in T-cell infiltration and anti-tumor immune responses between high- and low-CTIPRGs groups, we analyzed the TCGA cohort using eight indicators of T-cell infiltration and antitumor immune response (Fig. 6B and Supplementary Table S12). These indicator scores were calculated by ssGSEA from the corresponding gene signature [72]. All differences in ssGSEA scores of T cell infiltration were significant according to the Wilcoxon test (P-value < 0.05). Additionally, a significant negative correlation was observed between CTIPRGs and T-cell infiltration and anti-tumor immune response indicators (Supplementary Fig. 4B). These results suggested that decreased CTIPRGs were correlated with an increased antitumor immune response in the TME of TCGA-BRCA patients.

3.7 Mutation Landscape between High-and low-CTIPRGs groups

To further explore the mutation characteristics of TCGA-BRCA patients stratified by CTIPRGs, we downloaded the available somatic mutation profiles and demonstrated the mutation landscape of the high- and low-CTIPRGs groups. We presented the top 20 mutant genes in each group respectively. As shown in Fig. 7A, B, the gene with the most mutation frequency in the high-CTIPRGs groups was TP53 (65%) and that in the low-CTIPRGs groups was PIK3CA (35%). Koboldt et al.[73] and Shahbandi et al.[74] found that TP53 and PIK3CA were the most frequently mutated genes in breast cancer, with their somatic mutations occurring at an incidence of over 10% across all BRCA molecular subtypes. In addition, significant differentially mutated genes between high- and low-CTIPRGs groups were detected using Fisher’s exact test (Fig. 7C).Notably, 92 patients in the low-CTIPRGs groups and 187 patients in the high-CTIPRGs groups exhibited mutations in TP53 (OR = 4.898, P < 0.001). Furthermore, a lollipop plot (Fig. 7D) was generated to illustrate the different mutation sites of TP53 in the high- and low-CTIPRGs groups. As shown in Figs. 7E, F, we analyzed co-occurrence and mutually exclusive associations among the top 25 mutated genes in the high- and low-CTIPRGs groups. Finally, a boxplot (Fig. 7G) was generated to illustrate the differences in TMB values between the high- and low-CTIPRGs groups. As expected, the low-CTIPRGs groups had a higher TMB value, indicating a better response to immune checkpoint inhibitors (ICIs).

3.8 The Drug Sensitivity and Therapeutic Response of the CTIPRGs

Chemotherapy and immunotherapy are primary treatments for breast cancer. Therefore, we assessed the sensitivity of the high- and low-CTIPRGs groups to commonly used therapeutic agents for BRCA. Compared with the high-CTIPRGs groups, the low-CTIPRGs groups were more susceptible to methotrexate, cyclopamine, paclitaxel, gemcitabine, and doxorubicin, whereas the high-CTIPRGs groups were more sensitive to lapatinib (Fig. 8A). Elesclomol is a highly lipophilic Cu2+-binding molecule, previously described as a potent inducer of oxidative stress. It can significantly increase mitochondrial reactive oxygen species (ROS) in both GSCs and GdECs, ultimately leading to cuproptosis. [75]. We also tested the sensitivity of patients in TCGA-BRCA to elesclomol. The elesclomol IC50 of the high-CTIPRGs groups was significantly lower than low-CTIPRGs groups (Fig. 8A). Earlier research [76] has shown that the expression levels of immune checkpoint blockade (ICB)-related genes were correlated with the therapeutic responses to immune checkpoint inhibitors. Subsequently, we identified the expression of 49 immunomodulators in the high- and low-CTIPRGs groups. As shown in Fig. 8B, most ICB-related genes were expressed at significantly higher levels in the low-CTIPRGs groups than in the high-CTIPRGs groups. Moreover, given the significant negative correlation between CTIPRGs and CD8 + T cells, and the critical role of m6A methylation in impairing the anti-tumor ability of CD8 + T cells, we displayed the expression levels of m6A regulators (Fig. 8C, D) and INF-γ markers that activate CD8 + T cells in the high- and low-CTIPRGs groups. As expected, m6A regulators were more abundant in the high-CTIPRGs groups, while INF-γ-related markers were more abundant in the low-CTIPRGs groups. Finally, we used the TIDE score to predict the effect of immunotherapy in the TCGA cohort, finding that the high-CTIPRGs groups had higher TIDE scores (Fig. 8E). In summary, CTIPRGs could closely predict drug sensitivity, particularly for immunotherapy. Additionally, we found that the efficacy of immunotherapy was superior in patients with low-CTIPRGs compared to those with high-CTIPRGs.

3.9 Evaluation of CCL8 and GZMB expression in BRCA

The results above indicate that CTIPRGs have significant clinical implications. To further validate the differential expression of CTIPRGs in tumor and normal tissues, we selected the genes CCL8 and GZMB. In the TCGA cohort, we observed that the expression of both CCL8 and GZMB was elevated in BRCA tissues compared to normal samples (Fig. 9A). Upon examining the expression of CCL8 and GZMB proteins in 10 BRCA samples and paired normal breast tissues using IHC, we found that CCL8 expression was elevated in BRCA samples, whereas GZMB expression was lower (Fig. 9B). TCGA patients were categorized into two groups based on gene expression levels. We found that low CCL8 expression was associated with better prognosis, whereas low GZMB expression correlated with worse prognosis (Fig. 9C and D), thereby confirming the results presented in Figs. 4 and 5. Regarding immune cell infiltration, CCL8 expression exhibited a strong positive correlation with Macrophages M1 infiltration levels (R = 0.41, P < 0.001) and a strong negative correlation with resting Mast cells infiltration levels (R=-0.27, P < 0.001) (Fig. 9E). GZMB expression was positively correlated with activated CD4 memory T cells (R = 0.65, P < 0.001) and exhibited a strong negaFtive correlation with Macrophages M2 (R=-0.47, P < 0.001) (Fig. 9F).

It is well known that the tumor immune microenvironment plays a crucial role in tumor development [77, 78]. Furthermore, copper is involved in various biological processes in the body and maintains homeostasis. Disruption of copper homeostasis causes a novel form of cell death known as cuproptosis, characterized by an excess of copper ions and lipid-acylated proteins. Here, we constructed an independent novel model using cuproptosis and TIP-related characteristics based on multiple statistical methods—CTIPRGs. Using the training database (TCGA-BRCA) and validation databases (GSE20685, GSE42568, GSE88770), we classified BRCA patients into high- and low-CTIPRGs groups. As expected, the high-CTIPRGs group had a worse prognosis and response to immunotherapy. By building a nomogram, we could quantitatively assess and precisely stratify risk in BRCA patients, providing new ideas for future immunotherapy. In conclusion, our study showed that CTIPRGs could assess the prognostic characteristics of BRCA patients and evaluate their response to chemotherapy and immunotherapy, thereby helping to provide a more accurate clinical diagnosis and treatment for BRCA patients.

Nowadays, more and more gene signatures have surfaced to elucidate the prognostic stratification of BRCA. For example, Yang et al. [79] identified a fifteen-gene prognostic model based on immune infiltration of breast cancer and obtained favorable results, Sha et al. [80] constructed two gene signatures predicting BRCA OS based on cuproptosis. However, tumors are complex masses comprising multiple cellular components and confounding factors; most research has, however, focused on the role of a single factor in BRCA, which is limiting. Moreover, both cuproptosis and the tumor microenvironment (TME) play crucial roles in the development of BRCA.[17, 81, 82]. Therefore, a comprehensive understanding of the interrelationship between cuproptosis and immune-related features, as well as their role in BRCA, will enhance our understanding of immune regulation in the TME and aid in the development of corresponding therapeutic strategies. Here, we identified, for the first time, a green module containing 178 genes based on the expression patterns of cuproptosis and TIP-related genes using WGCNA, and constructed 14 gene signatures, termed "CTIPRGs," for predicting the prognosis of BRCA patients. This represents a novel genetic model that integrates various characteristics. In contrast to previous studies, we conducted a comprehensive analysis, elucidating differences in immune infiltration among subgroups and predicting potential sensitive drugs and treatment efficacy based on the model.

Additionally, further exploration of the differences in biological behavior between high- and low-CTIPRGs groups provided deeper insights into the regulatory mechanisms of CTIPRGs. We found that low-CTIPRGs groups were more active in well-established immune-related signaling pathways, including the NOD-like receptor signaling pathway and the T cell receptor signaling pathway. [83, 84]. In contrast, high-CTIPRGs groups were more closely associated with metabolic activities. Tumor initiation and progression require the metabolic reprogramming of cancer cells. Cancer cells autonomously alter their fluxes through various metabolic pathways to meet increased bioenergetic and biosynthetic demands and to mitigate oxidative stress necessary for their proliferation and survival.[85]. And both glycolysis and TCA cycle supports tumor growth and produces the War-burg effect through metabolite biosynthesis [86, 87]. Reactive oxygen species (ROS) generated by oxidative metabolism at levels that don’t cause cell death can further support tumorigenesis [88, 89]. The metabolic pathways enriched in the high-CTIPRGs group are likely major contributors to tumor development and poor prognosis. Furthermore, cuproptosis predominantly occurs in cells with high mitochondrial respiration, and this novel mode of cell death likely plays a critical role in metabolically active high-CTIPRGs groups, offering new directions and ideas for BRCA treatment. In contrast, patients in the low-CTIPRGs group had significantly longer overall survival (OS), higher levels of infiltration by adaptive and innate immune cell types, and were also enriched in pathways related to apoptosis, mismatch repair, and various immune responses. Apoptosis and mismatch repair can inhibit tumor cell proliferation, metastasis, and tumorigenesis through APA sites in genes. [90, 91]. The innate immune system, comprising Toll-like receptors (TLRs), NOD-like receptors (NLRs), and RIG-I-like receptors (RLRs), plays a pivotal role in immune response and immunosurveillance. Natural and synthetic agonists of these receptors can induce cell death in malignant cells and recruit immune cells, such as dendritic cells (DCs), CD8 + T cells, and NK cells, into the tumor microenvironment (TME), thereby exerting anti-tumor efficacy. [92]. The concentration of these characteristics in low-CTIPRGs groups were likely to be the main reason for their better prognosis.

In this study, we explored the correlation between CTIPRGs and T cell infiltration as well as anti-tumor immune responses. As expected, the validity of CTIPRGs in characterizing the immune landscape was further confirmed by significant negative correlations with immunization-related items. Mutation profiles and differential oncogenes revealed oncogenic pathways associated with CTIPRGs, suggesting a potential link between specific immune traits and oncogenes. Based on these analyses, we determined that CTIPRGs are potent biological indicators capable of classifying BRCA patients into distinct subgroups.

In anticancer therapies, chemotherapy remains a primary strategy for treating breast cancer patients. Chemotherapy can activate the immune system by promoting the release of danger signals from dying tumor cells and/or eliminating immunosuppressive cells, thereby enhancing the immune response and increasing the likelihood of tumor elimination following anticancer therapy. In BRCA, the crucial role of anti-tumor immunity in enhancing the effectiveness of anticancer therapies has been clearly established [93, 94]. Here, we evaluated the sensitivity of high- and low-CTIPRGs groups to commonly used chemotherapeutic agents for BRCA, including lapatinib, methotrexate, cyclopamine, paclitaxel, gemcitabine, and doxorubicin. We found that the low-CTIPRGs groups were more sensitive to most of the commonly used chemotherapeutic agents. This increased sensitivity may be related to the enrichment of various immune cell infiltrates in the low-CTIPRGs groups. Additionally, the immunogenic cancer cell death induced by chemotherapy can enhance the phagocytosis of dying tumor cells by abundant dendritic cells (DCs), leading to T-cell activation for anti-tumor effects [95]. And chemotherapy can also further induce genotoxic cellular stress triggering the expression of ligands that activate the NK cell receptor NKG2D, which leads to NK cell activation [96]. Given that CTIPRGs was constructed based on CRGs, we further explored the sensitivity of different subgroups to elesclomol. Consistent with the results of previous studies [75], we found that high-CTIPRGs groups were more susceptible than low-CTIPRGs groups. Consistent with the previous observation[97], elesclomol induces cell death by accumulating excess ROS. This finding indicates that cuproptosis plays a crucial role in the development of BRCA, and we believe that elesclomol holds promise as a therapeutic agent for BRCA.

Tumor tolerance is maintained through various mechanisms, including regulatory immune cells, immunosuppressive cytokines and chemokines, and immune checkpoint pathways that downregulate immune function. The PD-1 (also known as PDCD1)–PD-L1 receptor-ligand pair is a predominant immune checkpoint pathway in the tumor microenvironment (TME), capable of enabling cancer cells to evade immune attack [98, 99]. Unlike PD1-PDL1, CTLA acts mainly during T cell triggering and activation, and it can enhance the immunosuppressive activity of Treg cells and have a broad impact on the immune system [100]. Currently, anti-cancer immunotherapies targeting PD-1, PD-L1, and CTLA-4 have shown positive results in BRCA patients; however, only a small proportion of patients have responded to these treatments [101], the main reason for this might be the limitation of the immune status [102]. To assess the ability of CTIPRGs to predict the response rate of BRCA patients to anti-cancer immunotherapies, we examined the expression levels of 49 commonly tested immune checkpoints in high- and low-CTIPRGs groups. As expected, given the immune-inflamed phenotype characteristics of the low-CTIPRGs groups, more than half of the immune checkpoints were highly expressed in these groups, indicating that patients in the low-CTIPRGs groups were more likely to benefit from immunotherapy. Considering the critical role of CD8 + T cells in anti-tumor immunity, we evaluated key molecules in CD8 + T cell-related pathways in the high- and low-CTIPRGs groups, revealing significant differences between the groups. Additionally, we found that the low-CTIPRGs groups had a lower TIDE score. These results suggest that CTIPRGs could serve as a significant predictive marker for immunotherapy response in BRCA.

In this study, we screened and analyzed two CTIPRGs signatures, CCL8 and GZMB. Verified the high expression of CCL8 in BRCA tumor samples and GZMB in normal breast samples using IHC. Additionally, high CCL8 expression and low GZMB expression were associated with worse survival in breast cancer patients. GZMB is a serine protease with a perforin-dependent pro-apoptotic function, relying on cytotoxic immune cells like cytotoxic T-lymphocytes (CTLs) and natural killer (NK) cells to kill tumor and virally infected cells[103]. Recent studies have demonstrated GZMB expression in tumor cells, including those in breast and prostate cancer[104]. CCL8 is a member of the conserved chemokine cluster (CC cluster), located in the MCP region of mouse chromosome 11C and human chromosome 17q12[105]. It participates in the immune response by attracting T lymphocytes, monocytes, eosinophils, and natural killer (NK) cells[106]. Human studies have demonstrated that CCL8 is highly expressed during mammary gland degeneration, promoting breast carcinogenesis by enhancing the infiltration of M2 subtype macrophages[107]. CCL8 and IL-8 can also work together to mediate crosstalk between endothelial colony-forming cells (ECFCs) and triple-negative breast cancer (TNBC), leading to increased tumorigenicity[108].

In summary, we developed a prediction model-CTIPRGs based on the characteristics of CRGs and TIPs that can predict the prognosis and treatment response rate of BRCA patients. it has been validated in various aspects and provided a new guiding direction for accurate diagnosis and personalized treatment of BRCA patients.

Data availability statement

The analyzed data in this paper are available in the UCSC Xena database (https://xenabrowser.net/), the GEO database (https://www.ncbi.nlm.nih.gov/geo/) (GSE20685, GSE42568, GSE88770), and the GTEX database (https://www.gtexportal.org/). All data obtained in this study are publicly available in accordance with the statements and specifications of the respective official websites, and there are no ethical issues or copyright conflicts.

Author contributions

JR and FYH conceived the idea, design the study. JR, FYH, WJY, YKC, WJW, MJ ,SHB, XXZ, WYL, WYl , FW, MC, RJM ,SYW ,YLD and YG retrieved and analyzed the data, JR and FYH drafted and revised the manuscript. All authors contributed to the article and approved the submitted version.

Funding

This manuscript is supported by the Scientific and Technological Research Foundation of Shaan’xi Province, Key Research and Development Project, General project (JR, 2023-YBSF-666); supported by the Shaanxi Provincial Key Research and Development Plan Project (Key support projects Category A) (JR, 2021A011); supported by the Basic and Clinical Integration Innovation Project of Xi’ an Jiao tong University (JR,YXJLRH2022006； YG, YXJLRH 2022004).

Acknowledgments

The authors would like to thank the reviewers for their helpful comments on this article, as well as research groups for the TCGA, CEO,GTEX and Tumor Immune Dysfunction and Exclusion database which provided data for this collection.

Conﬂict of interest

The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their afﬁliated organizations, or those of the publisher, the editors, and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

McCord, J.M. and Fridovic.I, SUPEROXIDE DISMUTASE-AN ENZYMIC FUNCTION FOR ERYTHROCUPREIN. Federation Proceedings, 1969. 28(2): p. 346-&.
Halliwell, B. and J.M.C. Gutteridge, ROLE OF FREE-RADICALS AND CATALYTIC METAL-IONS IN HUMAN-DISEASE - AN OVERVIEW. Methods in Enzymology, 1990. 186: p. 1–85.
Kardos, J., et al., NERVE-ENDINGS FROM RAT-BRAIN TISSUE RELEASE COPPER UPON DEPOLARIZATION - A POSSIBLE ROLE IN REGULATING NEURONAL EXCITABILITY. Neuroscience Letters, 1989. 103(2): p. 139–144.
Schlief, M.L., et al., Role of the Menkes copper-transporting ATPase in NMDA receptor-mediated neuronal toxicity. Proceedings of the National Academy of Sciences of the United States of America, 2006. 103(40): p. 14919–14924.
El Meskini, R., et al., ATP7A (Menkes protein) functions in axonal targeting and synaptogenesis. Molecular and Cellular Neuroscience, 2007. 34(3): p. 409–421.
Bhattacharjee, A., K. Chakraborty, and A. Shukla, Cellular copper homeostasis: current concepts on its interplay with glutathione homeostasis and its implication in physiology and human diseases. Metallomics, 2017. 9(10): p. 1376–1388.
Arredondo, M. and M.T. Nunez, Iron and copper metabolism. Molecular Aspects of Medicine, 2005. 26(4–5): p. 313–327.
Kaler, S.G., Inborn errors of copper metabolism. Handbook of clinical neurology, 2013. 113: p. 1745–54.
Graper, M.L., et al., Introduction to Human Disorders of Copper Metabolism, in Human Disorders of Copper Metabolism I. 2014, Blackwell Science Publ: Oxford. p. V-VI.
Oliveri, V., Selective Targeting of Cancer Cells by Copper Ionophores: An Overview. Frontiers in Molecular Biosciences, 2022. 9: p. 14.
Shao, S.Q., J.X. Si, and Y.Q. Shen, Copper as the Target for Anticancer Nanomedicine. Advanced Therapeutics, 2019. 2(5): p. 16.
Lelievre, P., et al., The Multifaceted Roles of Copper in Cancer: A Trace Metal Element with Dysregulated Metabolism, but Also a Target or a Bullet for Therapy. Cancers, 2020. 12(12): p. 25.
Tsvetkov, P., et al., Copper induces cell death by targeting lipoylated TCA cycle proteins. Science, 2022. 375(6586): p. 1254-+.
Siegel, R.L., K.D. Miller, and A. Jemal, Cancer Statistics, 2018. Ca-a Cancer Journal for Clinicians, 2018. 68(1): p. 7–30.
Nanda, R., et al., A phase Ib study of pembrolizumab (MK-3475) in patients with advanced triple-negative breast cancer. Cancer Research, 2015. 75: p. 2.
Emens, L.A., et al., Inhibition of PD-L1 by MPDL3280A leads to clinical activity in patients with metastatic triple-negative breast cancer. Cancer Research, 2015. 75: p. 2.
Mittal, S., N.J. Brown, and I. Holen, The breast tumor microenvironment: role in cancer development, progression and response to therapy. Expert Review of Molecular Diagnostics, 2018. 18(3): p. 227–243.
Burugu, S., K. Asleh-Aburaya, and T.O. Nielsen, Immune infiltrates in the breast cancer microenvironment: detection, characterization and clinical implication. Breast Cancer, 2017. 24(1): p. 3–15.
Adams, S., et al., Prognostic Value of Tumor-Infiltrating Lymphocytes in Triple-Negative Breast Cancers From Two Phase III Randomized Adjuvant Breast Cancer Trials: ECOG 2197 and ECOG 1199. Journal of Clinical Oncology, 2014. 32(27): p. 2959-+.
Loi, S., et al., Prognostic and Predictive Value of Tumor-Infiltrating Lymphocytes in a Phase III Randomized Adjuvant Breast Cancer Trial in Node-Positive Breast Cancer Comparing the Addition of Docetaxel to Doxorubicin With Doxorubicin-Based Chemotherapy: BIG 02–98. Journal of Clinical Oncology, 2013. 31(7): p. 860–867.
Sierra, A., et al., Astrocyte-derived cytokines contribute to the metastatic brain specificity of breast cancer cells. Laboratory Investigation, 1997. 77(4): p. 357–368.
Zhang, J.H., et al., Turning cold tumors hot: from molecular mechanisms to clinical applications. Trends in Immunology, 2022. 43(7): p. 523–545.
Wang, H.Y., et al., Tumor immunological phenotype signature-based high-throughput screening for the discovery of combination immunotherapy compounds. Science Advances, 2021. 7(4): p. 14.
Cai, W.Y., et al., Identification of a Tumor Microenvironment-relevant Gene set-based Prognostic Signature and Related Therapy Targets in Gastric Cancer. Theranostics, 2020. 10(19): p. 8633–8647.
Clarke, C., et al., Correlating transcriptional networks to breast cancer survival: a large-scale coexpression analysis. Carcinogenesis, 2013. 34(10): p. 2300–2308.
Kao, K.J., et al., Correlation of microarray-based breast cancer molecular subtypes and clinical outcomes: implications for treatment optimization. Bmc Cancer, 2011. 11: p. 15.
Metzger-Filho, O., et al., Genomic grade adds prognostic value in invasive lobular carcinoma. Annals of oncology: official journal of the European Society for Medical Oncology, 2013. 24(2): p. 377–384.
Johnson, W.E., C. Li, and A. Rabinovic, Adjusting batch effects in microarray expression data using empirical Bayes methods. Biostatistics, 2007. 8(1): p. 118–127.
Leek, J.T., et al., The sva package for removing batch effects and other unwanted variation in high-throughput experiments. Bioinformatics, 2012. 28(6): p. 882–883.
Gentleman, R.C., et al., Bioconductor: open software development for computational biology and bioinformatics. Genome Biology, 2004. 5(10): p. 16.
Wilkerson, M.D. and D.N. Hayes, ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking. Bioinformatics, 2010. 26(12): p. 1572–1573.
Hanzelmann, S., R. Castelo, and J. Guinney, GSVA: gene set variation analysis for microarray and RNA-Seq data. Bmc Bioinformatics, 2013. 14: p. 15.
Subramanian, A., et al., Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proceedings of the National Academy of Sciences of the United States of America, 2005. 102(43): p. 15545–15550.
Liberzon, A., et al., The Molecular Signatures Database Hallmark Gene Set Collection. Cell Systems, 2015. 1(6): p. 417–425.
Ritchie, M.E., et al., limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Research, 2015. 43(7): p. 13.
Guo, C.B., et al., Mining TCGA Data for Key Biomarkers Related to Immune Microenvironment in Endometrial cancer by Immune Score and Weighted Correlation Network Analysis. Frontiers in Molecular Biosciences, 2021. 8: p. 15.
Yoshihara, K., et al., Inferring tumour purity and stromal and immune cell admixture from expression data. Nature Communications, 2013. 4: p. 11.
Langfelder, P. and S. Horvath, WGCNA: an R package for weighted correlation network analysis. Bmc Bioinformatics, 2008. 9: p. 13.
Friedman, J., T. Hastie, and R. Tibshirani, Regularization Paths for Generalized Linear Models via Coordinate Descent. Journal of Statistical Software, 2010. 33(1): p. 1–22.
Iasonos, A., et al., How to build and interpret a nomogram for cancer prognosis. Journal of Clinical Oncology, 2008. 26(8): p. 1364–1370.
Mayakonda, A., et al., Maftools: efficient and comprehensive analysis of somatic variants in cancer. Genome Research, 2018. 28(11): p. 1747–1756.
Chalmers, Z.R., et al., Analysis of 100,000 human cancer genomes reveals the landscape of tumor mutational burden. Genome Medicine, 2017. 9: p. 14.
Sun, J., et al., Characterization of immune landscape in papillary thyroid cancer reveals distinct tumor immunogenicity and implications for immunotherapy. Oncoimmunology, 2021. 10(1): p. 14.
Charoentong, P., et al., Pan-cancer Immunogenomic Analyses Reveal Genotype-Immunophenotype Relationships and Predictors of Response to Checkpoint Blockade. Cell Reports, 2017. 18(1): p. 248–262.
Becht, E., et al., Estimating the population abundance of tissue-infiltrating immune and stromal cell populations using gene expression. Genome Biology, 2016. 17: p. 20.
Rooney, M.S., et al., Molecular and Genetic Properties of Tumors Associated with Local Immune Cytolytic Activity. Cell, 2015. 160(1–2): p. 48–61.
Wu, J., et al., A risk model developed based on tumor microenvironment predicts overall survival and associates with tumor immunity of patients with lung adenocarcinoma. Oncogene, 2021. 40(26): p. 4413–4424.
Wu, X.N., et al., Identified lung adenocarcinoma metabolic phenotypes and their association with tumor immune microenvironment. Cancer Immunology Immunotherapy, 2021. 70(10): p. 2835–2850.
Liu, Z.Q., et al., Comprehensive Molecular Analyses of a Novel Mutational Signature Classification System with Regard to Prognosis, Genomic Alterations, and Immune Landscape in Glioma. Frontiers in Molecular Biosciences, 2021. 8: p. 15.
Xu, F., et al., Analysis of Lung Adenocarcinoma Subtypes Based on Immune Signatures Identifies Clinical Implications for Cancer Therapy. Molecular Therapy-Oncolytics, 2020. 17: p. 241–249.
Xu, Q.H., et al., Landscape of Prognostic m6A RNA Methylation Regulators in Hepatocellular Carcinoma to Aid Immunotherapy. Frontiers in Cell and Developmental Biology, 2021. 09: p. 17.
Li, Y.S., et al., Molecular characterization and clinical relevance of m(6)A regulators across 33 cancer types. Molecular Cancer, 2019. 18(1): p. 6.
Geeleher, P., N.J. Cox, and R.S. Huang, Clinical drug response can be predicted using baseline gene expression levels and in vitro drug sensitivity in cell lines. Genome Biology, 2014. 15(3): p. 12.
Jiang, P., et al., Signatures of T cell dysfunction and exclusion predict cancer immunotherapy response. Nature Medicine, 2018. 24(10): p. 1550-+.
Hanahan, D. and L.M. Coussens, Accessories to the Crime: Functions of Cells Recruited to the Tumor Microenvironment. Cancer Cell, 2012. 21(3): p. 309–322.
Bradshaw, A.D., Diverse biological functions of the SPARC family of proteins. International Journal of Biochemistry & Cell Biology, 2012. 44(3): p. 480–488.
Liu, C., et al., Effect of ECM2 expression on bovine skeletal muscle-derived satellite cell differentiation. Cell Biology International, 2018. 42(5): p. 525–532.
Nagaraju, G.P., et al., Molecular Mechanisms Underlying the Divergent Roles of SPARC in Human Carcinogenesis. Carcinogenesis, 2014. 35(5): p. 967-U69.
Hagan, N., et al., CSF1R signaling is a regulator of pathogenesis in progressive MS. Cell Death & Disease, 2020. 11(10): p. 25.
Chitu, V. and E.R. Stanley, Regulation of Embryonic and Postnatal Development by the CSF-1 Receptor, in Protein Kinases in Development and Disease, A. Jenny, Editor. 2017, Elsevier Academic Press Inc: San Diego. p. 229-+.
Bui, T.M., H.L. Wiesolek, and R. Sumagin, ICAM-1: A master regulator of cellular responses in inflammation, injury resolution, and tumorigenesis. Journal of Leukocyte Biology, 2020. 108(3): p. 787–799.
Muenst, S., et al., The presence of programmed death 1 (PD-1)-positive tumor-infiltrating lymphocytes is associated with poor prognosis in human breast cancer. Breast Cancer Research and Treatment, 2013. 139(3): p. 667–676.
Early Breast Canc, T., Adjuvant bisphosphonate treatment in early breast cancer: meta-analyses of individual patient data from randomised trials (vol 386, pg 1353, 2015). Lancet, 2016. 387(10013): p. 30–30.
Carmeliet, P. and R.K. Jain, Angiogenesis in cancer and other diseases. Nature, 2000. 407(6801): p. 249–257.
Nagarsheth, N., et al., PRC2 Epigenetically Silences Th1-Type Chemokines to Suppress Effector T-Cell Trafficking in Colon Cancer. Cancer Research, 2016. 76(2): p. 275–282.
Nagarsheth, N., M.S. Wicha, and W.P. Zou, Chemokines in the cancer microenvironment and their relevance in cancer immunotherapy. Nature Reviews Immunology, 2017. 17(9): p. 559–572.
Peng, D.J., et al., Epigenetic silencing of T(H)1-type chemokines shapes tumour immunity and immunotherapy. Nature, 2015. 527(7577): p. 249-+.
Romagnani, P., et al., Cell cycle-dependent expression of CXC chemokine receptor 3 by endothelial cells mediates angiostatic activity. Journal of Clinical Investigation, 2001. 107(1): p. 53–63.
Bottcher, J.P., et al., NK Cells Stimulate Recruitment of cDC1 into the Tumor Microenvironment Promoting Cancer Immune Control. Cell, 2018. 172(5): p. 1022-+.
Roberts, E.W., et al., Critical Role for CD103(+)/CD141(+) Dendritic Cells Bearing CCR7 for Tumor Antigen Trafficking and Priming of T Cell Immunity in Melanoma. Cancer Cell, 2016. 30(2): p. 324–336.
Spranger, S., et al., Tumor-Residing Batf3 Dendritic Cells Are Required for Effector T Cell Trafficking and Adoptive T Cell Therapy. Cancer Cell, 2017. 31(5): p. 711-+.
Cheng, W.C., et al., Uncoupling protein 2 reprograms the tumor microenvironment to support the anti-tumor immune cycle. Nature Immunology, 2019. 20(2): p. 206-+.
Koboldt, D.C., et al., Comprehensive molecular portraits of human breast tumours. Nature, 2012. 490(7418): p. 61–70.
Shahbandi, A., H.D. Nguyen, and J.G. Jackson, TP53 Mutations and Outcomes in Breast Cancer: Reading beyond the Headlines. Trends in Cancer, 2020. 6(2): p. 98–110.
Buccarelli, M., et al., Elesclomol-induced increase of mitochondrial reactive oxygen species impairs glioblastoma stem-like cell survival and tumor growth. Journal of Experimental & Clinical Cancer Research, 2021. 40(1): p. 17.
Goodman, A., S.P. Patel, and R. Kurzrock, PD-1-PD-L1 immune-checkpoint blockade in B-cell lymphomas. Nature Reviews Clinical Oncology, 2017. 14(4): p. 203–220.
Gajewski, T.F., H. Schreiber, and Y.X. Fu, Innate and adaptive immune cells in the tumor microenvironment. Nature Immunology, 2013. 14(10): p. 1014–1022.
Melaiu, O., et al., Influence of the Tumor Microenvironment on NK Cell Function in Solid Tumors. Frontiers in Immunology, 2020. 10: p. 18.
Peng, Y., et al., Construction and Validation of an Immune Infiltration-Related Gene Signature for the Prediction of Prognosis and Therapeutic Response in Breast Cancer. Frontiers in Immunology, 2021. 12: p. 18.
Sha, S.N., et al., Prognostic analysis of cuproptosis-related gene in triple-negative breast cancer. Frontiers in Immunology, 2022. 13: p. 21.
da Silva, D.A., et al., Copper in tumors and the use of copper-based compounds in cancer treatment. Journal of Inorganic Biochemistry, 2022. 226: p. 23.
Jiang, Y.C., et al., Copper-induced tumor cell death mechanisms and antitumor theragnostic applications of copper complexes. Nanomedicine, 2022. 17(5): p. 303–324.
Alhazmi, A., NOD-like receptor(s) and host immune responses with Pseudomonas aeruginosa infection. Inflammation Research, 2018. 67(6): p. 479–493.
Li, K., et al., Promising Targets for Cancer Immunotherapy: TLRs, RLRs, and STING-Mediated Innate Immune Pathways. International Journal of Molecular Sciences, 2017. 18(2): p. 19.
Martinez-Reyes, I. and N.S. Chandel, Cancer metabolism: looking forward. Nature Reviews Cancer, 2021. 21(10): p. 669–680.
Gatenby, R.A. and R.J. Gillies, Why do cancers have high aerobic glycolysis? Nature Reviews Cancer, 2004. 4(11): p. 891–899.
Engelman, J.A., et al., Effective use of PI3K and MEK inhibitors to treat mutant Kras G12D and PIK3CA H1047R murine lung cancers. Nature Medicine, 2008. 14(12): p. 1351–1356.
DeBerardinis, R.J. and N.S. Chandel, We need to talk about the Warburg effect. Nature Metabolism, 2020. 2(2): p. 127–129.
Ghergurovich, J.M., et al., Local production of lactate, ribose phosphate, and amino acids by human triple-negative breast cancer. Med, 2021. 2(6): p. 736-+.
Kim, N., et al., Alternative polyadenylation of single cells delineates cell types and serves as a prognostic marker in early stage breast cancer. Plos One, 2019. 14(5): p. 16.
Tan, S., et al., NUDT21 negatively regulates PSMB2 and CXXC5 by alternative polyadenylation and contributes to hepatocellular carcinoma suppression. Oncogene, 2018. 37(35): p. 4887–4900.
Wicherska-Pawlowska, K., T. Wrobel, and J. Rybka, Toll-Like Receptors (TLRs), NOD-Like Receptors (NLRs), and RIG-I-Like Receptors (RLRs) in Innate Immunity. TLRs, NLRs, and RLRs Ligands as Immunotherapeutic Agents for Hematopoietic Diseases. International Journal of Molecular Sciences, 2021. 22(24): p. 21.
Pages, F., et al., Effector memory T cells, early metastasis, and survival in colorectal cancer. New England Journal of Medicine, 2005. 353(25): p. 2654–2666.
Bindea, G., et al., Spatiotemporal Dynamics of Intratumoral Immune Cells Reveal the Immune Landscape in Human Cancer. Immunity, 2013. 39(4): p. 782–795.
Casares, N., et al., Caspase-dependent immunogenicity of doxorubicin-induced tumor cell death. Journal of Experimental Medicine, 2005. 202(12): p. 1691–1701.
Gasser, S., et al., The DNA damage pathway regulates innate immune system ligands of the NKG2D receptor. Nature, 2005. 436(7054): p. 1186–1190.
Zheng, P.J., et al., Elesclomol: a copper ionophore targeting mitochondrial metabolism for cancer therapy. Journal of Experimental & Clinical Cancer Research, 2022. 41(1): p. 13.
Pardoll, D.M., The blockade of immune checkpoints in cancer immunotherapy. Nature Reviews Cancer, 2012. 12(4): p. 252–264.
Pardoll, D., Cancer and the Immune System: Basic Concepts and Targets for Intervention. Seminars in Oncology, 2015. 42(4): p. 523–538.
Topalian, S.L., et al., Mechanism-driven biomarkers to guide immune checkpoint blockade in cancer therapy. Nature Reviews Cancer, 2016. 16(5): p. 275–287.
Duffy, A.G., et al., Tremelimumab in combination with ablation in patients with advanced hepatocellular carcinoma. Journal of Hepatology, 2017. 66(3): p. 545–551.
Ramos-Casals, M., et al., Immune-related adverse events of checkpoint inhibitors. Nature Reviews Disease Primers, 2020. 6(1): p. 21.
Garciasanz, J.A., et al., APPEARANCE OF GRANULE-ASSOCIATED MOLECULES DURING ACTIVATION OF CYTOLYTIC LYMPHOCYTE-T PRECURSORS BY DEFINED STIMULI. Immunology, 1988. 64(1): p. 129–134.
Velotti, F., et al., Granzyme B in Inflammatory Diseases: Apoptosis, Inflammation, Extracellular Matrix Remodeling, Epithelial-to-Mesenchymal Transition and Fibrosis. Frontiers in Immunology, 2020. 11: p. 9.
Nomiyama, H., et al., Organization of the chemokine genes in the human and mouse major clusters of CC and CXC chemokines: diversification between the two species. Genes and Immunity, 2001. 2(2): p. 110–113.
Ruffing, N., et al., CCR5 has an expanded ligand-binding repertoire and is the primary receptor used by MCP-2 on activated T cells. Cellular Immunology, 1998. 189(2): p. 160–168.
Farmaki, E., et al., CCL8 Promotes Postpartum Breast Cancer by Recruiting M2 Macrophages. Iscience, 2020. 23(6): p. 17.
Kim, E.S., et al., CCL8 mediates crosstalk between endothelial colony forming cells and triple-negative breast cancer cells through IL-8, aggravating invasion and tumorigenicity. Oncogene, 2021. 40(18): p. 3245–3259.

No competing interests reported.

Download PDF

Reviewers invited by journal
29 Jul, 2024
Editor assigned by journal
29 Jul, 2024
Editor invited by journal
27 Jun, 2024
Submission checks completed at journal
25 Jun, 2024
First submitted to journal
21 Jun, 2024

You are reading this latest preprint version

Clinical Significance, Immune Landscape and Immunotherapy Efficacy Analyses of Cuproptosis-Tumor Immunological Phenotype-Related Gene Score Based Prognostic Model in Breast Cancer

Status:

Version 1

Abstract

Figures

1 Introduction

2 MATERIALS AND METHODS

2.1 Data Source

2.2 Consensus Clustering of BRCA Patients

2.3 Gene set variation analysis (GSVA)

2.4 Construction of TIP Score

2.5 Differentially Expressed Genes (DEGs)

2.6 Weighted Gene Co-Expression Network Analysis (WGCNA) and Development of CTIPRGs

2.7 Survival Analysis

2.8 Construction of a nomogram model

2.9 Mutation Landscape

2.10 Characteristics of Tumor Microenvironment(TME)

2.11 Prediction of Drug Sensitivity and Therapeutic Response

2.12 Immunohistochemical(IHC) staining

2.13 Statistical Analyses

2.14 Ethics Statement

3 RESULTS

3.1 Identification of BRCA Subtypes Based on 10 Cuproptosis-Related Genes

3.2 TIP Was Related to Prognosis and TME of BRCA

3.3 Development of CTIPRGs

3.4 Evaluation and Validation of Survival Efficacy by CTIPRGs

3.5 The Immune Status between High-and Low CTIPRGs Groups

3.6 T cell infiltration and Antitumor Immune Response

3.7 Mutation Landscape between High-and low-CTIPRGs groups

3.8 The Drug Sensitivity and Therapeutic Response of the CTIPRGs

3.9 Evaluation of CCL8 and GZMB expression in BRCA

4 DISCUSSIONS

5 CONCLUSION

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1