Logo Medical Science Monitor

Call: +1.631.470.9640
Mon - Fri 10:00 am - 02:00 pm EST

Contact Us

Logo Medical Science Monitor Logo Medical Science Monitor Logo Medical Science Monitor

29 January 2022: Database Analysis  

as a Potential Independent Prognostic Factor of Breast Cancer Related to Lipid Metabolism and Immune Microenvironment Based on TCGA Database Analysis

Shilin Li CDE , Yi Fang A*

DOI: 10.12659/MSM.934597

Med Sci Monit 2022; 28:e934597

0 Comments

Abstract

BACKGROUND: Lipid metabolism has been proved to be related to the prognosis of breast cancer patients in previous studies, and the tumor immune microenvironment (TIME) plays an important role in tumorigenesis and development, but the dynamic regulation of these is still a challenge.

MATERIAL AND METHODS: This study used lipid metabolism-related pathways to score the gene expression of 980 breast cancer patients in the TCGA database. We used 4 pathways in HALLMARK related to lipid metabolism to score the genes in the database. The differentially expressed genes (DEGs) were further analyzed through survival analysis and Cox regression analysis, and MS4A1, which is associated with better prognosis, was finally determined to be a predictor. In-depth analysis found that MS4A1 was negatively correlated with patient age, clinical stage, tumor size, and distant metastasis. In the MS4A1 high-expression group, most genes were enriched in immune-related pathways, and CIBERSORT analysis found that MS4A1 expression was positively correlated with the abundance of 10 kinds of immune cells, such as CD8+T cells, which are related to the active immune status.

RESULTS: Our results suggest that MS4A1 expression can indicate the situation of lipid metabolism in breast cancer patients and reflect the status of the immune microenvironment.

CONCLUSIONS: MS4A1 has the potential to be an independent indicator of prognosis. Since the expression of MS4A1 is also related to the immune checkpoint mutation burden, detecting its expression level can also provide guidance for choosing treatment options.

Keywords: biomarkers, Breast Neoplasms, Lipid Metabolism, tumor microenvironment

Background

The incidence of breast cancer (BRCA) continues to increase. Breast cancer has a poor prognosis and is difficult to detect early, seriously endangering women’s physical and mental health. According to statistics [1], breast cancer has surpassed lung cancer in 2020, becoming the world’s deadliest cancer. In that year, the number of new breast cancer cases was 2.26 million worldwide. The number of female deaths because of cancer was 4.43 million, and 680 000 of them died from breast cancer. The current effective treatment methods mainly include surgery, chemotherapy, and endocrine therapy, but new biomarkers are needed to classify patients to facilitate individualized intervention and survival prediction, and provide new potential targets for future drug development.

Dysregulated lipid metabolism can lead to tumors, and it can also affect tumor cell proliferation, apoptosis, invasion, and metastasis [2–4]. Lipid metabolism is closely related to the occurrence and development of breast cancer: people with hypercholesterolemia and obesity have a significantly higher risk of estrogen receptor-alpha (ERα)-positive breast cancers [5–7]. 27-Hydroxycholesterol (27HC), an important factor in regulating the homeostasis of cholesterol and a selective estrogen receptor, has been experimentally verified to promote the occurrence and progression of breast cancer [8–11]. Clinical studies have further found that the use of statin therapy can improve the prognosis of breast cancer patients [12]. Classifying patients by assessing lipid metabolism and implementing interventions for lipid metabolism disorders can improve the prognosis. Furthermore, studies have found that disorders of lipid metabolism are also related to drug resistance during treatment [13].

The tumor microenvironment (TME) includes endothelial cells, pericytes, fibroblasts, adipocytes, resident and infiltrating immune cells, and extracellular matrix (ECM). ECM is mainly composed of collagen, elastin, fibronectin, laminin, and proteoglycan (PG). TME contains lipid metabolism components such as fatty acids and cholesterol, and these factors can promote tumor progression [14]. The tumor immune microenvironment (TIME) plays an important role in tumor treatment and tumor progression, which can reflect the immune conditions of tumor tissues, and provide potential targets for tumor treatment [15]. In TME, ECM is an important component, and ECM remodeling plays an important role in tumorigenesis and progression. The lysyl oxidase gene (LOX) acts in the cross-linking of elastin and collagen fibers as well as in the modulation of the structure and stiffness of tumor ECM. Studies have shown that in mouse models of breast cancer, drug therapy combined with LOX knockdown or suppression therapy can reduce tumor growth and metastasis and prolong survival [16–18]. LOX has been suggested as a promising option for the treatment of triple-negative breast cancer [19]. Moreover, LOX has prognostic value in breast cancer patients and plays a role in the formation of pre-metastatic niches in bone in ER-negative breast tumors, and patients with high levels of LOX often have a worse prognosis [20,21].

Similarly, in recent years it has been proposed the potential activity of Trabectedin (an anti-tumor compound already registered as second-line treatment of soft tissue sarcoma and for ovarian cancer) in breast cancer treatment, due to its ability to act as modulator of TME by reducing the number of tumor-associated macrophages [22,23]. Moreover, Trabectedin also showed the ability to promote the increase of timp1 expression (a matrix metalloproteases inhibitor), which can affect the activity and contributing to inhibition of cell invasiveness [24]. Studies have shown [25] that the active lipid metabolism can lead to local immunosuppression in TME and promote tumor growth.

Therefore, in the present study, we first used the 4 lipid metabolism-related pathways in HALLMARK (hallmark gene set) to score 980 patients in The Cancer Genome Atlas (TCGA), screened out differentially expressed genes, then comprehensively considered the survival prognosis, and screened out a biomarker, MS4A1, whose higher expression was related to lower lipid metabolism and better survival situation. The patients were grouped according to the level of MS4A1 expression, and enrichment analysis was performed. It was found that high expression of MS4A1 is closely related to the immune active microenvironment and activation of immune cell activation-related pathways. The MS4A family was initially determined based on the observations of pan-B cell markers, and MS4A1 (membrane-spanning 4-domain family, subfamily A) shared a high amino-acid sequence identity [26]. Previous studies have found [27] that MS4A1, which encodes a B cell surface marker, is also expressed in T cell subsets. The high expression of MS4A1 in colon cancer patients is associated with better prognosis and treatment response. However, in breast cancer, MS4A1 as a potential prognostic predictor has not been reported.

Material and Methods

DATA COLLECTION: Transcriptome RNA-seq data of 1222 BRCA cases (excluding the duplicate cases and para-cancerous cases, leaving 1091 remaining cases), single-nucleotide variation data of 986 BRCA cases and the corresponding clinical data of 1097 cases were downloaded from the TCGA database (https://portal.gdc.cancer.gov/). We chose all 980 cases with complete transcriptome, single-nucleotide variation and clinical data through the R package “venn” (Figure 1). The clinical data of the 980 cases are summarized in Table 1. The mean age of these patients was 58.46 years old.

LIPID METABOLISM-RELATED PATHWAYS-SCORE AND SCREENING FOR DIFFERENTIALLY EXPRESSED GENES (DEGS):

The 4 lipid metabolism-related pathways in HALLMARK were downloaded from http://www.gsea-msigdb.org/gsea/msigdb/index.jsp.

We used the R (version 3.6.2) package “GSVA” to score 980 patients for representing the activity level of each patient in the 4 lipid metabolism-related pathways of HALLMARK (HALLMARK_CHOLESTEROL_HOMEOSTASIS, HALLMARK_ADIPOGENESIS, HALLMARK_FATTY_ACID_METABOLISM, and HALLMARK_BILE_ACID_METABOLISM). The median score calculated by GSVA package and the 4 median scores were: HALLMARK_CHOLESTEROL_HOMEOSTASIS: −0.043, HALLMARK_ADIPOGENESIS: −0.028, HALLMARK_FATTY_ACID_METABOLISM: −0.019, and HALLMARK_BILE_ACID_METABOLISM: 0.0004 (lower than or equal to the median value was defined as the low-expression group).

To find DEGs, we performed Wilson rank sum testing on gene expression. We set the cutoff as: logFc>0 and the median difference between the high expression group and the low expression group>0. At the same time, the statistical significance of FDR <0.05 (P value corrected as FDR).

We found the intersection between the upregulated genes in the high-score group and the low-score group, and performed the same operation in the downregulated genes. Then, we merged the 2 intersections. We finally found 17 common DEGs.

SURVIVAL ANALYSIS:

We used the R package “survival” and “survminer” to perform survival analysis to determine the prognostic value of these intersected DEGs. The cutoff value was determined through the function surv_cutpoint. Kaplan-Meier method (log rank as the statistical significance test) and univariate and Cox regression analysis were used for survival analysis. P<0.05 was considered statistically significant. We selected genes that were statistically significant in both analysis methods, revealing 12 genes with predictive value for prognosis.

COX REGRESSION ANALYSIS:

Univariate and multivariate Cox regression analysis showed that MS4A1 had the lowest hazard ratio (HR). Therefore, we identified MS4A1 as the object of further analysis.

ENRICHMENT ANALYSIS:

According to the cutoff value of expression level (cutoff value was 0.008887257), 980 cases were divided into an MS4A1 high-expression group and an MS4A1 low-expression group

CLUSTERPROFILE:

R packages “clusterProfiler” and “ggplot2” were used to find DEGs. We performed Wilcoxon signed ranks testing on gene expression, with the cutoff value set as LogFc>0 and the statistical significance of FDR <0.05 (P value corrected as FDR). We then performed gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis.

GENE SET VARIATION ANALYSIS (GSVA):

We used FDR <0.05 to define the DEGs. Then GO and KEGG enrichment analysis were performed with R packages “GSVA” and “ggplot2”.

GENE SET ENRICHMENT ANALYSIS (GSEA):

GO, HALLMARK, and KEGG gene sets were analyzed using GSEA 4.0.3 software. The whole transcriptome of all tumor cases was used for GSEA, and only gene sets with NOM P<0.05 and FDR <0.25 were considered as significant.

TUMOR-INFILTRATING IMMUNE CELL (TIC) PROFILE: We used CIBERSORT [28] computation to estimate the TIC infiltration fraction in 980 cases. Pearson correlation was calculated between the proportions and MS4A1 expression. P<0.05 was selected for the following analysis.

SUB-NETWORK ANALYSIS:

We used Cytoscape (3.6.1) with the MCODE plug-in to perform sub-network analysis on the GOBP terms analyzed with GSEA, with the following parameter settings: degree cutoff 15, Mode Score cutoff 0.2, K-Core 15, Max Depth 100. We finally obtained 4 sub-networks.

HEATMAPS:

Heatmaps were used to present DEGs performed in the R package “pheatmap”. The 100 genes with most significant fold change are presented (50 genes with highest LogFc and 50 genes with lowest LogFc).

VOLCANO PLOTS:

Volcano plots were used to present DEGs performed in the R package “ggplot2”. The genes with |LogFc| >1 are colored and the genes with |LogFc| >2 are labeled.

Results

WE DETERMINED THE NUMBER OF CASES THAT NEEDED TO BE ANALYZED: We chose all 980 cases with complete transcriptome, single-nucleotide variation, and clinical data (Figure 1). The clinical data of the 980 cases are summarized in Table 1.

SCORES WERE ASSOCIATED WITH LIPID METABOLISM-RELATED PATHWAYS IN HALLMARK: Four lipid metabolism-related pathways in HALLMARK were used to obtain scores of 980 cases, and the median value was used to distinguish the high- and low-score groups. The DEGs between the high-score group and low-score group are presented in heatmaps and volcano plots (Figure 2 and Supplementary Figure 1). The 17 common genes are shown in Figure 3.

SCREEN CANDIDATE GENES BASED ON SURVIVAL PROGNOSIS: To further screen biomarkers which can predict the prognosis, we analyzed the impact of these 17 DEGs on survival. We obtained 12 genes that were statistically significant in Kaplan-Meier analysis and Cox regression analysis. Survival curves of 12 genes with prognostic value are shown in Figure 4. We used the Spearman rank correlation coefficient to reflect the correlation between gene expression and scoring, displayed in the form of circle graphs (Figure 5). The length of the ribbon represents the value of the correlation coefficient.

COX REGRESSION ANALYSIS WAS PERFORMED TO FILTER VARIABLES: We performed univariate Cox regression analysis of 12 DEGs again, this time incorporating clinical factors into variables: TNM staging and patient age. We found that there were 7 genes with protective effects on the prognosis: BGLAP, COL9A1, FAM129C, FCRL1, FGG, MS4A1 and PZP while AKR1B10, SPINK8, TAS2R31, POU3F2, and CRISP3 were risk factors for prognosis (Figure 6A). Further, we conducted multivariate Cox regression analysis, chose forward selection to filter variables, and drew forest maps (Figure 6B). It was obviously that MS4A1 was an independent protective factor for prognosis. Therefore, we finally determined MS4A1 as the target of further research.

ANALYSIS RESULTS OF 980 CASES GROUPED ACCORDING TO MS4A1 EXPRESSION: We analyzed SNP, mRNA, and clinical data of 980 cases and the results are showed in Figure 4. According to the DNA data based on 980 cases, TP53 had the highest frequency of mutations (Figure 7A). We found the DEGs between MS4A1 high-expression group and low-expression group and the 100 genes with the most significant fold change value are shown in Figure 7B. We also analyzed the differences of MS4A1 expression in different clinical characters (Figure 8). The cases were divided into 2 groups, one younger than or equal to 65 years old and the other older than 65 years old. The expression of MS4A1 in the younger group was higher than in the older group. The expression of MS4A1 in patients with clinical stage I was higher than in patients with clinical stage IV. For TNM staging, the expression of MS4A1 differed between different T stages and different M stages, meaning that the expression of MS4A1 was higher in the group with smaller tumor diameter and the group without distant metastasis. According to the analysis results, there was no significant difference of MS4A1 expression in patients with different N stages.

THE DIFFERENTIAL EXPRESSION OF MS4A1 MAY BE RELATED TO IMMUNE-RELATED PATHWAYS: Supplementary Figure 2A shows the differential expression of genes in 980 cases. We used the R package “clusterprofiler”. In GO analysis, the DEGs were enriched to the immune-related GO terms (eg, regulation of T cell activation and lymphocyte differentiation) (Figure 9A and Supplementary Figure 2B). The KEGG analysis also showed enriched immune-related pathways, such as Th1 and Th2 cell differentiation (Figure 9B and Supplementary Figure 2C). Based on the above analysis results, we concluded that most of the DEGs obtained after grouping according to the expression of MS4A1 are enriched in immune-related pathways, which indicated that the differential expression of MS4A1 may be related to immune status.

We used the R package “GSVA” for further enrichment analysis. We found that DEGs in GO, HALLMARK, and KEGG gene sets all displayed enrichment of immune-related pathways (Supplementary Figure 3).

We also performed GSEA, with similar results obtained as in the above 2 analysis methods: immune activity in the MS4A1 high-expression group was active (Figure 10). KEGG and HALLMARK results of GSEA enrichment analysis charts are displayed in Supplementary Figure 4. In GOBP-related pathways enrichment analysis, we particularly selected the terms related to T cell immunity, B cell immunity, and NK (natural killer) cell immunity to display in the chart (Figure 11A–11C). Due to the high level of redundancy in GOBP, we performed a sub-network analysis on the results of the pathway in GOBP. After setting the parameters, 4 sub-networks were finally obtained, 2 of which are related to immunity (one is related to T cells and the other is related to B cells), 1 is related to calcium transportation, and the other is related to the cytoskeleton (Figure 11D). These analysis results show that high expression of MS4A1 is related to active immunity.

RELATIONSHIP BETWEEN MS4A1 EXPRESSION AND THE PROPORTION OF TICS: To explore the relationship between MS4A1 and the immune microenvironment, immune subsets were analyzed using the CIBERSORT algorithm. A barplot shows the proportion of 22 kinds of TICs in BRCA cases (Figure 12A). A heatmap shows the correlation between 22 kinds of TICs and Pearson coefficient used for significance testing (Figure 12B). The correlation between MS4A1 and 22 kinds of immune cells is shown in a circle graph. The length of the ribbon represents the correlation coefficient (Figure 12C). The 22 kinds of immune cells are T cells CD8, T cells CD4 naïve, T cells CD4 memory resting, T cells CD4 memory activated, T cells follicular helper, T cells regulatory (Tregs), T cells gamma delta, B cells naïve, B cells memory, plasma cells, NK cells resting, NK cells activated, monocytes, macrophages M0, macrophages M1, macrophages M2, dendritic cells resting, dendritic cells activated, mast cells resting, mast cells activated, eosinophils, and neutrophils. We found that the expression MS4A1 was positively related to macrophages M1 and CD8+T cells, which are related to immune activation, and are negatively related to macrophages M2 and Tregs, which are related to immunosuppression.

In Supplementary Figure 5, the correlation between 22 kinds of immune cells and MS4A1 are presented. The correlation coefficients (R) were: B cells memory (R=0.16, FDR<0.001), B cells naïve (R=0.35, FDR<0.001), dendritic cells activated (R=−0.01, FDR=−0.867), dendritic cells resting(R=0.21, FDR<0.001), eosinophils (R=0.01, FDR=0.874), macrophages M0 (R=−0.40, FDR<0.001), macrophages M1 (R=0.40, FDR<0.001), macrophages M2 (R=−0.46, FDR<0.001), mast cells resting (R=−0.12, FDR=0.002), monocytes (R>−0.01, FDR=0.952), neutrophils (R=−0.14, FDR<0.001), NK cells activated (R=0.04, FDR=0.389), NK cells activated (R=−0.18, FDR<0.001), plasma cells (R=−0.11, FDR=0.002), T cells CD4 memory activated (R=0.46, FDR<0.001), T cells CD4 memory resting (R=0.23, FDR<0.001), T cells CD4 naïve (R<0.01, FDR=0.979), T cells CD8(R=0.42, FDR<0.001), T cells follicular helper (R=0.46, FDR<0.001), T cells gamma delta (R=0.36, FDR<0.001), and T cells regulatory (Tregs) (R−0.07, FDR=0.057). A total of 16 kind of immune cells were correlated with MS4A1. We found that the expression of MS4A1 was positively correlated with components related to immune activation such as CD8+ T cell and T cells CD4 memory activated and was is negatively correlated with macrophages M2, which are related to immunosuppression. These results indicated that the high expression of MS4A1 was associated with active immunity.

The boxplot (Figure 13A) presents differences in the infiltration of immune cells between MS4A1 high-expression group and low-expression group. There were 14 kinds of cells with different infiltration between the 2 groups. Among them, 10 kinds of cells are positively correlated with MS4A1 expression (eg, CD8+ T cell and T cells CD4 memory activated), while other 4 kinds of cells are negatively correlated with MS4A1 expression (eg, macrophages M0 and macrophages M2). We conclude that the high expression of MS4A1 is related to active immunity. The analysis and correlation analysis results showed a total of 14 common immune cell profiles are related to the expression of MS4A1 (Figure 13B).

DIFFERENTIAL EXPRESSION OF MS4A1 AND IMMUNE CHECKPOINT-RELATED GENES: To provide potential drug targets for clinical treatment, we further explored the correlation between MS4A1 expression and immune checkpoint-related genes (Supplementary Figure 5) (genes annotated in gray are not statistically significant). We found that the MS4A1 high-expression group had higher expression of immune checkpoint-related genes, which indicated that the patients with high expression of MS4A1 may be sensitive to immunotherapy.

Discussion

In this study, we tried to find genes from the TCGA database that are related to lipid metabolism and that have prognostic value. MS4A1 stood out from the screening. Interestingly, after a series of bioinformatics analysis, MS4A1 is not only related to lipid metabolism, but may also be an indicator of the immune microenvironment.

In our study, the expression of MS4A1 is correlated with the activity of lipid metabolism. Breasts are mainly composed of fatty tissue. At present, there are many known causes of breast cancer, including obesity and high-fat diet. According to statistics, obesity increases the risk of breast cancer [28–30]. Adipose tissue is the largest endocrine organ of the human body. It not only plays an important role in storing energy, but also in hormone regulation and the development of chronic inflammation, which are closely related to the occurrence and progression of tumors. In the studies of estrogen receptor (+)/progesterone receptor (+)/human epidermal growth factor receptor 2 (−) breast cancer, patients with higher body mass index (BMI) had a lower overall survival rate, and are more prone to recurrence and metastasis, which means worse prognosis [31,32]. In a prospective study, it was found that regardless of BMI, estrogen receptors, and the time interval between blood collection and diagnosis, blood lipids are related to the risk of breast cancer [33]. On the contrary, some studies have concluded that women with hypercholesterolemia have a lower risk of breast cancer and a longer survival period, but the authors believed that this result may also be due to patients with hypercholesterolemia receiving statins treatment, which reduces the risk of breast cancer [34]. If a patient can start to change her lifestyle, including reducing energy intake, adjusting diet structure, avoiding high-fat diets, and increasing exercise, the quality of life and prognosis of the patient can be significantly improved [35–38]. Some other studies found that lipid metabolism disorders are closely related to chemotherapy resistance. Lipid metabolism can be regulated by the JAK/STAT3 pathway. Inhibition of the JAK/STAT3 pathway can decrease the expression of multiple lipid metabolism-related genes, thereby inhibiting breast cancer stem cells and reducing chemotherapy resistance [39].

TME is particularly important in the occurrence and progression of tumors, and metabolic disorders can affect TME [40]. LOX (acting in the cross-linking of elastin and collagen fibers as well as in the modulation of the structure and stiffness of tumor ECM) was also discovered to be an indicator of prognosis and metastasis [20,21]. Use of Trabectedin, which is an anti-tumor compound that can adjust multiple components in TME, including tumor-associated macrophages, has been suggested in breast cancer treatment [22–24]. Many studies have confirmed that the immune components in TME are particularly important in the occurrence and development of tumors. Although breast cancer is not generally considered to be a highly immunogenic cancer, there is still a rich tumor immune microenvironment in some subtypes of breast cancer. The adaptive immune system, macrophages, and other innate immune cells in breast cancer also play important roles [41]. Some studies have found that in triple-negative breast cancer patients, TILs (tumor-infiltrating lymphocytes) are associated with better prognosis, especially prolonging recurrence-free survival, and can be used as tumor prognostic markers [42–44]. Thus, most patients with active immune response will obtain better treatment effect and prognosis.

MS4A1 encodes the B cell surface marker CD20, which participates in B cell receptor signaling and interacts with the immune microenvironment. In our analysis using lipid metabolism-related pathways to score and screen out MS4A1, the prognosis of the high-expression group was significantly better than the low-expression group. A study found that the expression of lipid metabolism-related genes in breast tissue, which is susceptible to breast cancer, is active and CD20 expression is decreased [45]. From this we speculate that MS4A1 improves the prognosis of breast cancer.

We grouped 980 cases according to the differential expression of MS4A1 and found DEGs to use in pathways enrichment analysis. The enrichment results showed that the high expression of MS4A1 was related to immune-related pathways.

CD20 encoded by MS4A1 is closely related to B cell proliferation, differentiation, and activation [46–48], and CD20 disorders have also been reported in patients with immunodeficiency diseases [49,50]. Studies have found that the expression of CD20 on tumor-infiltrating lymphocytes in patients with ovarian cancer is associated with a positive prognosis [51]; in colorectal cancer, the expression of MS4A1 is positively correlated with the survival rate of patients, and the MS4A1 expression can be used as a diagnostic marker [52,53]. These results are consistent with our finding that the prognosis of breast cancer patients in the MS4A1 high expression group was better. In addition, we found that the expression of MS4A1 was positively correlated with 10 types of immune cells, which may be one of the reasons for the better prognosis of patients with high MS4A1 expression. A study of more than 12 000 people showed that CD8+ T cells infiltration can significantly reduce the risk of death in patients with ER-negative, ER, and HER2-positive breast cancer [54]. Higher CD8+ T cells scores were associated with better survival of TNBC patients [55], and PARP inhibitor olaparib induces CD8+ T cells recruitment and activation to exert an anti-tumor effect [56].

For the pathways related to calcium channels and cytoskeleton in the enriched sub-network analysis, we speculate it may be due to the calcium ion transportation and cytoskeleton remodeling involved in the process of immune microenvironment regulation. Cytoskeleton changes are related to immune synapse formation. These physiological processes play important roles in the function of the immune system [57–59].

MS4A1 is related to the lipid metabolism and immune microenvironment status of breast cancer patients, and has potential to be an independent prognostic indicator. Since its expression is positively correlated with immune checkpoint mutation-related genes, MS4A1 could provide guidance in choosing appropriate treatment methods.

A limitation of the present study is that we only used data from a single database. In addition, due to the limitation of objective conditions, no relevant experimental verification was carried out. We will address these limitations in future related research and look forward to making further discoveries.

Conclusions

After a series of analyses of relevant data from the TCGA database, we found that MS4A1 not only reflects the lipid metabolism level of breast cancer patients, but also serves as a potential independent prognostic factor. Our research provides a new biomarker for clinical treatment, which helps to distinguish the different conditions of patients, so as to better carry out individualized treatment and maximize the treatment effect.

Figures

Figure 1. Looking for samples with all 3 pieces of information (mRNA, clinical features, single-nucleotide mutations [SNP]), a total of 980 cases were obtained. Figure 2. Four lipid metabolism-related pathways in HALLMARK were used as scoring standards. (A) Using the median value of HALLMARK_CHOLESTEROL_HOMEOSTASIS score as the standard, we performed differential gene expression analysis for the high group and the low group. (B) Using the median value of HALLMARK_ADIPOGENESIS score as the standard, we performed differential gene expression analysis for the high group and the low group. (C) Using the median value of HALLMARK_FATTY_ACID_METABOLISM score as the standard, we performed differential gene expression analysis for the high group and the low group. (D) Using the median value of HALLMARK_BILE_ACID_METABOLISM score as the standard, we performed differential gene expression analysis for the high group and the low group. Figure 3. (A) We took the intersection of DEGs in the high-score group and obtained 8 genes. (B) We took the intersection of DEGs in the low-score group and l obtained 9 genes. Figure 4. Differences in the expression of 12 genes leading to differences in survival. Figure 5. Correlation between 12 genes and 4 scoring pathways. The length of the ribbon indicates the size of the correlation. Figure 6. (A) Univariate Cox regression analysis incorporating age, TNM stage, and 12 DEGs into variables. (B) Multivariate Cox regression analysis. Because of the large number of variables, we used forward selection. MS4A1 was the only protective factor. Figure 7. (A) Single-nucleotide variation of MS4A1 high-expression group and low-expression group. (B) Transcriptome level results of MS4A1 high-expression group and low-expression group. Figure 8. (A) MS4A1 expression in different age groups and among different stage groups. (B) MS4A1 expression among different T (Topography) groups. (C) MS4A1 expression among different M (Metastasis) groups. (D) MS4A1 expression among different N (Lymph Node) groups. Figure 9. (A) We used R language with the clusterProfiler package to perform GO enrichment analysis and created a circle graph. (B)4We used R language with the clusterProfiler package to perform KEGG enrichment analysis and created a circle graph. Figure 10. (A) We used R language with the GSEA package to perform GO enrichment analysis. (B, C) We used R language with the GSEA package to perform HALLMARK enrichment analysis. (D) We used R language with the GSEA package to perform KEGG enrichment analysis. Figure 11. (A) T cell-related pathways in GOBP enrichment analysis. (B) B cell-related pathways in GOBP enrichment analysis. (C) NK cell-related pathways in GOBP enrichment analysis. (D) We used Cytoscape software to perform sub-network analysis of GOBP enrichment analysis results. The first and third sub-networks are related to immune function, the second sub-network is related to calcium channels, while the fourth is relate to cytoskeleton. Figure 12. (A) We used the CIBERSORT algorithm to explore the relationship between the expression of MS4A1 and the immune microenvironment, and 22 kinds of immune cell profiles in BRCA samples were constructed. (B) Correlation between immune cells. (C) Correlation between 22 kinds of immune cell profiles and MS4A1. Figure 13. (A) Differences in immune cells infiltrate between MS4A1 high-expression group and low-expression group. (B) Intersection of immune cells related to MS4A1 and differences in expression.

References

1. Sung H, Ferlay J, Siegel RL: Cancer J Clin, 2021; 71(3); 209-49

2. Röhrig F, Schulze A: Nat Rev Cancer, 2016; 16(11); 732-49

3. Yu XH, Ren XH, Liang XH, Tang YL: Mol Med Rep, 2018; 18(6); 5307-16

4. Esposito M, Mondal N, Greco TM: Nat Cell Biol, 2019; 21(5); 627-39

5. Boyd NF, McGuire V: J Natl Cancer Inst, 1990; 82(6); 460-68

6. Bianchini F, Kaaks R, Vainio H: Lancet Oncol, 2002; 3(9); 565-74

7. Kitahara CM, Berrington de González A, Freedman ND: J Clin Oncol, 2011; 29(12); 1592-98

8. Umetani M, Domoto H, Gormley AK: Nat Med, 2007; 13(10); 1185-92

9. DuSell CD, Umetani M, Shaul PW, Mangelsdorf DJ, McDonnell DP: Mol Endocrinol (Baltimore, MD), 2008; 22(1); 65-77

10. Nelson ER, Wardell SE, Jasper JS: Science (New York, NY), 2013; 342(6162); 1094-98

11. Wu Q, Ishikawa T, Sirianni R: Cell Rep, 2013; 5(3); 637-45

12. Kimbung S, Chang CY, Bendahl PO: Endocr Relat Cancer, 2017; 24(7); 339-49

13. Cao Y: J Clin Invest, 2019; 129(8); 3006-17

14. Gomaraschi M: Trends Endocrinol Metab, 2020; 31(3); 256-68

15. Lei X, Lei Y, Li JK: Cancer Lett, 2020; 470; 126-33

16. Rossow L, Veitl S, Vorlová S: Oncogene, 2018; 37; 4921-40

17. Miller BW, Morton JP, Pinese M: EMBO Mol Med, 2015; 7; 1063-76

18. Saatci O, Kaymak A, Raza U: Nat Commun, 2020; 11; 2416

19. De Vita A, Liverani C, Molinaro R: Sci Rep, 2021; 11(1); 5107

20. Setargew YFI, Wyllie K, Grant RD: Cancers (Basel), 2021; 13(3); 491

21. Saatci O, Kaymak A, Raza U: Nat Commun, 2020; 11(1); 2416

22. D’Incalci M, Zambelli A: Expert Opin Investig Drugs, 2016; 25(1); 105-15

23. De Vita A, Recine F, Miserocchi G: J Exp Clin Cancer Res, 2021; 40; 165

24. Souid S, Aissaoui D, Srairi-Abid N, Essafi-Benkhadir K: Curr Drug Targets, 2020; 21(10); 996-1007

25. Hao Y, Li D, Xu Y: BMC Bioinformatics, 2019; 20(Suppl 7); 195

26. Eon Kuek L, Leffler M, Mackay GA, Hulett MD: Immunol Cell Biol, 2016; 94(1); 11-23

27. Mudd TW, Lu C, Klement JD, Liu K: Cell Immunol, 2021; 360; 104260

28. Kelesidis I, Kelesidis T, Mantzoros CS: Br J Cancer, 2006; 94(9); 1221-25

29. Keum N, Greenwood DC, Lee DH: J Natl Cancer Inst, 2015; 107(2); djv088

30. Chen GC, Chen SJ, Zhang R: Obes Rev, 2016; 17(11); 1167-77

31. Robinson PJ, Bell RJ, Davis SR: Maturitas, 2014; 79(3); 279-86

32. Jeon YW, Kang SH, Park MH: BMC cancer, 2015; 15; 865

33. His M, Dartois L, Fagherazzi G: Cancer Causes Control, 2017; 28(1); 77-88

34. Zhong S, Zhang X, Chen L: Cancer Treat Rev, 2015; 41(6); 554-67

35. Prentice RL, Caan B, Chlebowski RT: JAMA, 2006; 295(6); 629-42

36. Chlebowski RT, Blackburn GL, Thomson CA: J Natl Cancer Inst, 2006; 98(24); 1767-76

37. Scott E, Daley AJ, Doll H: Cancer Causes Control, 2013; 24(1); 181-91

38. Swisher AK, Abraham J, Bonner D: Support Care Cancer, 2015; 23(10); 2995-3003

39. Wang T, Fahrmann JF, Lee H: Cell Metab, 2018; 27(1); 136-50e5

40. Dias AS, Almeida CR, Helguero LA, Duarte IF: Eur J Cancer (Oxford, England: 1990), 2019; 121; 154-71

41. Burugu S, Asleh-Aburaya K, Nielsen TO: Breast cancer (Tokyo, Japan), 2017; 24(1); 3-15

42. Adams S, Gray RJ, Demaria S: J Clin Oncol, 2014; 32(27); 2959-66

43. Loi S, Sirtaine N, Piette F: J Clin Oncol, 2013; 31(7); 860-67

44. Loi S, Michiels S, Salgado R: Ann Oncol, 2014; 25(8); 1544-50

45. Marino N, German R, Rao X: NPJ Breast Cancer, 2020; 6; 50

46. Tedder TF, Boyd AW, Freedman AS: J Immunol (Baltimore, MD: 1950), 1985; 135(2); 973-79

47. Léveillé C, R AL-D, Mourad W: Eur J Immunol, 1999; 29(1); 65-74

48. Petrie RJ, Deans JP: J Immunol (Baltimore, MD: 1950), 2002; 169(6); 2886-91

49. Kuijpers TW, Bende RJ, Baars PA: J Clin Invest, 2010; 120(1); 214-22

50. van de Ven AA, Compeer EB, Bloem AC: J Allergy Clin Immunol, 2012; 129(3); 755-61e7

51. Milne K, Köbel M, Kalloger SE: PLoS One, 2009; 4(7); e6412

52. Mudd TW, Lu C, Klement JD, Liu K: Cell Immunol, 2021; 360; 104260

53. Han M, Liew CT, Zhang HW: Clin Cancer Res, 2008; 14(2); 455-60

54. Ali HR, Provenzano E, Dawson SJ: Ann Oncol, 2014; 25(8); 1536-43

55. Oshi M, Asaoka M, Tokumaru Y: Int J Mol Sci, 2020; 21(18); 6968

56. Pantelidou C, Sonzogni O, De Oliveria Taveira M: Cancer Discov, 2019; 9(6); 722-37

57. Savic S, Caseley EA, McDermott MF: Nat Rev Rheumatol, 2020; 16(4); 222-37

58. Papa R, Penco F, Volpi S, Gattorno M: Front Immunol, 2020; 11; 604206

59. Berry CT, May MJ, Freedman BD: Methods Mol Biol (Clifton, NJ), 2021; 2366; 145-64

Figures

Figure 1. Looking for samples with all 3 pieces of information (mRNA, clinical features, single-nucleotide mutations [SNP]), a total of 980 cases were obtained.Figure 2. Four lipid metabolism-related pathways in HALLMARK were used as scoring standards. (A) Using the median value of HALLMARK_CHOLESTEROL_HOMEOSTASIS score as the standard, we performed differential gene expression analysis for the high group and the low group. (B) Using the median value of HALLMARK_ADIPOGENESIS score as the standard, we performed differential gene expression analysis for the high group and the low group. (C) Using the median value of HALLMARK_FATTY_ACID_METABOLISM score as the standard, we performed differential gene expression analysis for the high group and the low group. (D) Using the median value of HALLMARK_BILE_ACID_METABOLISM score as the standard, we performed differential gene expression analysis for the high group and the low group.Figure 3. (A) We took the intersection of DEGs in the high-score group and obtained 8 genes. (B) We took the intersection of DEGs in the low-score group and l obtained 9 genes.Figure 4. Differences in the expression of 12 genes leading to differences in survival.Figure 5. Correlation between 12 genes and 4 scoring pathways. The length of the ribbon indicates the size of the correlation.Figure 6. (A) Univariate Cox regression analysis incorporating age, TNM stage, and 12 DEGs into variables. (B) Multivariate Cox regression analysis. Because of the large number of variables, we used forward selection. MS4A1 was the only protective factor.Figure 7. (A) Single-nucleotide variation of MS4A1 high-expression group and low-expression group. (B) Transcriptome level results of MS4A1 high-expression group and low-expression group.Figure 8. (A) MS4A1 expression in different age groups and among different stage groups. (B) MS4A1 expression among different T (Topography) groups. (C) MS4A1 expression among different M (Metastasis) groups. (D) MS4A1 expression among different N (Lymph Node) groups.Figure 9. (A) We used R language with the clusterProfiler package to perform GO enrichment analysis and created a circle graph. (B)4We used R language with the clusterProfiler package to perform KEGG enrichment analysis and created a circle graph.Figure 10. (A) We used R language with the GSEA package to perform GO enrichment analysis. (B, C) We used R language with the GSEA package to perform HALLMARK enrichment analysis. (D) We used R language with the GSEA package to perform KEGG enrichment analysis.Figure 11. (A) T cell-related pathways in GOBP enrichment analysis. (B) B cell-related pathways in GOBP enrichment analysis. (C) NK cell-related pathways in GOBP enrichment analysis. (D) We used Cytoscape software to perform sub-network analysis of GOBP enrichment analysis results. The first and third sub-networks are related to immune function, the second sub-network is related to calcium channels, while the fourth is relate to cytoskeleton.Figure 12. (A) We used the CIBERSORT algorithm to explore the relationship between the expression of MS4A1 and the immune microenvironment, and 22 kinds of immune cell profiles in BRCA samples were constructed. (B) Correlation between immune cells. (C) Correlation between 22 kinds of immune cell profiles and MS4A1.Figure 13. (A) Differences in immune cells infiltrate between MS4A1 high-expression group and low-expression group. (B) Intersection of immune cells related to MS4A1 and differences in expression.

SARS-CoV-2/COVID-19

24 May 2022 : Clinical Research  

Screening for SARS-CoV-2 Infection in Students at the Medical University of Warsaw, Poland Between November...

Med Sci Monit In Press; DOI: 10.12659/MSM.936962  

04 May 2022 : Clinical Research  

Effects of Wearing Face Masks on Exercise Capacity and Ventilatory Anaerobic Threshold in Healthy Subjects ...

Med Sci Monit In Press; DOI: 10.12659/MSM.936069  

22 April 2022 : Clinical Research  

Factors Associated with Falls During Hospitalization for Coronavirus Disease 2019 (COVID-19)

Med Sci Monit In Press; DOI: 10.12659/MSM.936547  

In Press

25 May 2022 : Review article  

Narrative Review of the Mechanism of Hip Prosthesis Dislocation and Methods to Reduce the Risk of Dislocation

Med Sci Monit In Press; DOI: 10.12659/MSM.935665  

24 May 2022 : Clinical Research  

Remimazolam Tosilate Sedation with Adjuvant Sufentanil in Chinese Patients with Liver Cirrhosis Undergoing ...

Med Sci Monit In Press; DOI: 10.12659/MSM.936580  

24 May 2022 : Clinical Research  

Screening for SARS-CoV-2 Infection in Students at the Medical University of Warsaw, Poland Between November...

Med Sci Monit In Press; DOI: 10.12659/MSM.936962  

23 May 2022 : Clinical Research  

Fully Automatic Knee Joint Segmentation and Quantitative Analysis for Osteoarthritis from Magnetic Resonanc...

Med Sci Monit In Press; DOI: 10.12659/MSM.936733  

Most Viewed Current Articles

30 Dec 2021 : Clinical Research  

Retrospective Study of Outcomes and Hospitalization Rates of Patients in Italy with a Confirmed Diagnosis o...

DOI :10.12659/MSM.935379

Med Sci Monit 2021; 27:e935379

08 Mar 2022 : Review article  

A Review of the Potential Roles of Antioxidant and Anti-Inflammatory Pharmacological Approaches for the Man...

DOI :10.12659/MSM.936292

Med Sci Monit 2022; 28:e936292

01 Nov 2020 : Review article  

Long-Term Respiratory and Neurological Sequelae of COVID-19

DOI :10.12659/MSM.928996

Med Sci Monit 2020; 26:e928996

01 Jan 2022 : Editorial  

Editorial: Current Status of Oral Antiviral Drug Treatments for SARS-CoV-2 Infection in Non-Hospitalized Pa...

DOI :10.12659/MSM.935952

Med Sci Monit 2022; 28:e935952

Your Privacy

We use cookies to ensure the functionality of our website, to personalize content and advertising, to provide social media features, and to analyze our traffic. If you allow us to do so, we also inform our social media, advertising and analysis partners about your use of our website, You can decise for yourself which categories you you want to deny or allow. Please note that based on your settings not all functionalities of the site are available. View our privacy policy.

Medical Science Monitor eISSN: 1643-3750
Medical Science Monitor eISSN: 1643-3750