Development of a novel lipid metabolism-based signature to predict survival and immune response in triple negative breast cancer

Development of a novel lipid metabolism-based signature to predict survival and immune response in triple negative breast cancer | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Help Center Sign In Submit a Preprint Cite Share Download PDF Article Development of a novel lipid metabolism-based signature to predict survival and immune response in triple negative breast cancer Xia Yang, Wen Tang, Yongtao He, Huimin An, Jin Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-1941091/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background The high heterogeneity of triple negative breast cancer (TNBC) is the main clinical challenge for individualized therapy. Considering that fatty acid metabolism (FAM) plays an indispensable role in tumorigenesis and development of TNBC, we proposed a novel FAM-based classification to characterize the tumor microenvironment immune profiles and heterogeneous for TNBC. Methods Weighted gene correlation network analysis (WGCNA) was performed to identify FAM-related genes from 221 TNBC samples in Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) dataset. Then, non-negative matrix factorization (NMF) clustering analysis was applied to determine FAM clusters based on the prognostic FAM-related genes, which chosen from the univariate/multivariate cox regression model and the least absolute shrinkage and selection operator (LASSO) regression algorithm. Then, a FAM scoring scheme was constructed to further quantify FAM features of individual TNBC patient based on the prognostic differentially expressed genes (DEGs) between different FAM clusters. Systematically analyses were performed to evaluate the correlation between the FAM scoring system (FS) with survival outcomes, genomic characteristics, tumor microenvironment (TME) features and immunotherapeutic response for TNBC, which were further validated in the Cancer Genome Atlas (TCGA) and GSE58812 datasets. Moreover, the expression level and clinical significancy of the selected FS gene signatures were further validated in our cohort. Results 1860 FAM-genes were screened out using WGCNA. Three distinct FAM clusters were determined by NMF clustering analysis, which allowed to distinguish different groups of patients with distinct clinical outcomes and tumor microenvironment (TME) features. Then, prognostic gene signatures based on the DEGs between different FAM clusters were identified using univariate cox regression analysis and Lasso regression algorithm. A FAM scoring scheme was constructed, which could divide TNBC patients into high and low-FS subgroups. Low FS subgroup, characterized by better prognosis and abundance with effective immune infiltration. While patients with higher FS were featured with poorer survival and lack of effective immune infiltration. In addition, two independent immunotherapy cohorts (Imvigor210 and GSE78220) confirmed that patients with lower FS demonstrated significant therapeutic advantages from anti-PD-1/PD-L1 immunotherapy and durable clinical benefits. Further analyses in our cohort found that the differential expression of CXCL13, FBP1 and PLCL2 were significantly associated with clinical outcomes of TNBC samples. Conclusions This study revealed FAM plays an indispensable role in formation of TNBC heterogeneity and TME diversity. The novel FAM-based classification could provide a promising prognostic predictor and guide more effective immunotherapy strategies for TNBC. Fatty acid metabolism Tumor microenvironment Immunotherapy Triple negative breast cancer Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Introduction Triple-negative breast cancer (TNBC), as defined by the absence of estrogen receptor (ER) and progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2), is a heterogeneous breast cancer subtype that carries the worst prognosis due to its aggressive characteristics and limited therapeutic options [ 1 , 2 ] . Much effort has been devoted over the past decade in classifying TNBCs into several molecular subtypes with distinct mutational profiles, genomic alterations, and biological processes that could guide treatment decisions [ 3 – 7 ] . Lipid metabolism, especially the synthesis of fatty acids (FAs), is an important cellular process that converts nutrients into metabolic intermediates for membrane biosynthesis, energy storage, and signal molecule production [ 8 ] . Abnormal lipid metabolism is one of the hallmarks of cancer. Increasing evidence confirmed that the significance role of FAM in carcinogenesis, including cell–matrix interaction, cell signaling and communication, tumor angiogenesis and metastasis, and immune modulation [ 9 , 10 ] . Targeting FAM process has become a promising therapeutic strategy for tumors [ 11 , 12 ] . However, research on FAM-relevant molecular classification of TNBC has not yet been established. Recently, immunotherapy represented by specific immune checkpoint inhibitors (ICIs), such as anti-CTLA-4 and anti-PD-1/L1, have achieved a marked durable response in some types of cancer, but the overall response rate is still unsatisfied in TNBC [ 13 – 15 ] . The functional status of T cells is a key determinant of effective antitumor immunity and immunotherapy. Therefore, it is important to further elucidate the molecular mechanism of T cell dysfunction in tumor microenvironment. The mechanism of Immunometabolism in regulating the function and fate of immune cells has been widely concerned. Previous studies found that tumor cells and Treg cells drove elevated expression of group IVA phospholipase A2, then altered lipid metabolism and senescence of T cells. Inhibition of group IVA phospholipase A2 reprogrammed effector T cell lipid metabolism, effectively prevented T cell senescence, and enhanced anti-tumor immunity and immunotherapy efficacy [ 16 ] . Meanwhile, S-palmitoylation, a lipid process that covalently binds palmitic acid to protein residues, has been found to play an indispensable role in maintaining PD-L1 stability and inhibiting T cell cytotoxicity [ 17 ] . However, the specific effects of FAM on the tumor microenvironment immune profiles in TNBC are not completed recognized. In this study, we comprehensively evaluated the association between FAM and TME cell-infiltrating characteristics and heterogeneous by integrating the transcriptomic and genomic data of 470 TNBC samples from METABRIC, TCGA and GEO databases. WGCNA was performed to identify FAM-related genes in TNBC patients. Then, 3 distinct FAM clusters with nonnegative matrix factorization (NMF) clustering were identified. Moreover, we constructed a scoring scheme to quantify the FAM features in individual TNBC patient. The prognosis traits, genomic variations, transcriptome features, as well as immune infiltration among the different FAM subtypes were further analyzed and verified. These findings suggested that FAM plays an indispensable role in shaping the tumor immune microenvironment profiles and heterogeneous for TNBC. Methods Data acquisition and preparation The workflow was shown in Figure S1A. TNBC patients with full clinical annotations and RNA-seq data were searched from Molecular Taxonomy of Breast Cancer International Consortium (METABRIC, http://www.cbioportal.org/datasets), the Cancer Genome Atlas (TCGA, https://portal. gdc.cancer.gov/repository), and Gene-Expression Omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo/) datasets. In total, 470 TNBC samples were included in our study for further analysis. The microarray data of the METABRIC (N=221) was served as training dataset. The normalized RNA-seq data which acquired from TCGA database (N=142) and the gene expression profiles from GSE58812 (N=107) were used as independent validation datasets. R package ‘limma’ was applied for gene expression normalization [18] . Identification of FAM-related genes The hallmark gene sets of fatty acid metabolism (FAM) which including 158 genes were obtained from the Molecular Signatures Data base (MSigDB) (https://www.gseamsigdb.org/gsea/msigdb/). First, we evaluated the FAM ssGSEA score in METABRIC-TNBC samples by the ssGSEA algorithm (R package “gsva”) [19] . Then, we screened out the FAM module and FAM genes by the R package “wgcna” [20] . Totally, we employed Pearson’s correlation matrices, co-expression similarity matrix, and average linkage method to evaluate the correlation coefficient between any two genes. A weighted adjacency matrix with a scale free co-expression network and topological overlap matrix (TOM) were constructed to evaluate the connectivity and dissimilarity of the co-expression network. Based on the TOM dissimilarity, a hierarchical clustering tree was established, which could identify the key modules and genes. Then, we set the module membership (MM) >0.8 and gene significance (GS) > 0.5 to identify the correlation between genes and FAM ssGSEA score. Totally, 1860 candidate FAM-related genes were identified from the FAM module. Prognostic value analysis The prognostic significance of the FAM-related genes which obtained from the WGCNA was analyzed by univariate Cox regression model using the R package “survival”. Then, LASSO cox regression algorithm was applied to further select prognostic FAM genes using the R package “glmet” [21] . Subsequently, multivariate Cox regression model was performed to identify the most robust prognostic gene signatures for OS in METABRIC-TNBC patients. Non-negative matrix factorization (NMF) clustering analysis Non-negative matrix factorization (NMF) clustering analysis was applied to identify distinct FAM clusters based on the expression of 8 prognostic FAM-related genes. The optimal number of clusters and their stability were determined by the consensus clustering algorithm. The R package “NMF” was used to perform the consensus clustering [22] . Identification of differentially expressed genes (DEGs) between distinct FAM clusters The differentially expressed genes (DEGs) between different FAM clusters were identify using the R package “limma” [18] . The significance criteria for determining DEGs was set as adjusted P value < 0.01. Gene set variation analysis (GSVA) and functional annotation GSVA enrichment analysis using “GSVA” R packages were performed to investigate the variation in biological process between different FAM clusters [19] . The gene sets of “c2.cp.kegg.v7.1.symbols” were downloaded from MSigDB database for running GSVA analysis. Adjusted P value <0.05 was considered as statistically significance. The “clusterProfiler” R package was used to perform functional annotation for DEGs between different FAM clusters, with the cutoff value of FDR < 0.05 and P < 0.05 [23] . Estimation of TME cell infiltration ssGSEA algorithm was used to quantify the relative abundance of 28 immune cell types in the TME [24, 25] . The relative abundance of each immune cell type in each sample were represented by the enrichment scores which calculated by ssGSEA analysis. CIBERSORT algorithm was applied to analyze the compositions of 22 types of tumor-infiltrating immune cells among different FAM subgroups [26] . MCP-counter scores regarding immune-related activity and fibroblasts were evaluated using the ‘MCPcounter’ package [27] . The immune and stromal scores were further investigated to predict the level of infiltrating immune and stromal components in the TME using the ‘estimate’ package [28] . Moreover, the expression of key immune profiles were compared between different FAM clusters. Generation of a novel FAM-based classification To quantify the FAM features of individual TNBC patient, we explored a novel FAM-based classification—the FAM scoring system (FS) to investigate the FAM features of individual TNBC patient. Specifically, 782 overlap DEGs were identified from different FAM clusters, we then extracted the prognostic gene signatures using univariate Cox regression model and Lasso regression algorithm. Finally, 8 genes were chosen to construct the FAM scoring system. The FAM Score (FS) was calculated by the corresponding coefficients of selected gene signatures: FAM Score = Σ i Coefficient（mRNA）* Expression（mRNA） Where i represent the selected gene signatures. Gene set variation analysis (GSVA) We then performed GSVA to further reveal the most significantly enriched molecular pathways between different FS subgroups using the R package “GSVA” [19] . The gene sets of “c2.cp.kegg.v7.1.symbols” were downloaded from MSigDB database. Adjusted P value < 0.05 was considered as statistically significance. Significantly mutated genes and tumor mutation burden in different FS groups Principal component analysis for the expression profiles of 8 FS gene signatures were analyzed and presented between tumor and normal samples in TCGA cohort. Moreover, the CNV variation frequency of 8 FS gene signatures were further evaluated in TCGA-TNBC cohort. The R package of RCircos was adopted to plot the copy number variation landscape of these selected gene signatures in 23 pairs of chromosomes [29] . Using the R package maftools [30] , the overall mutation landscape was summarized and present in patients with high and low FS subgroups in TCGA cohort. Then, TMB scores based on the TGCA somatic mutation data were calculated to evaluate the mutation status between different FS subgroups. Genomic and clinical data sets with immune-checkpoint blockade The immunotherapeutic cohorts: IMvigor210 cohort (advanced urothelial cancer treated with atezolizumab, anti PD-L1 antibody) [31] and GSE78220 cohort (metastatic melanoma with intervention of pembrolizumab, an anti PD-1 antibody) [32] were chosen to analyze the predictive efficiency of FS scheme for immunotherapy. Exploration of potential compounds targeting the selected FAM-related gene signatures To explore potential compounds targeting the selected FS-related gene signatures for treatment of TNBC, we calculated the therapeutic response based on the half-maximal inhibitory concentration (IC50) of various molecular obtained from the CellMiner database for each sample [33] . Validation of the bioinformatics results using RT-qPCR assay 63 paired TNBC tissues and adjacent normal tissues were acquired from Sir Run Run Shaw hospital of Zhejiang University School of Medicine. Total RNA was extracted using TRIzol reagent (Invitrogen, USA) according to the manufacturer’s protocol. Reverse transcription was conducted using PrimeScript RT MasterMix (Takara, China). RT qPCR was performed using SYBR Green PCR MasterMix (Takara, China) according to the manufacturer’s protocol. The primers were listed in (Table S3). The average Ct value were used for each gene which repeated three times, β-actin was applied to normalize the target genes mRNA expression. Statistical analysis Student’s t-tests were applied to analyze normally distributed variables, Wilcoxon rank-sum test was performed to evaluate non-normally distributed variables. One-way ANOVA and Kruskal-Wallis tests were used to conduct difference comparisons of more than two groups. Kaplan-Meier survival analysis and Cox proportional hazards model were chosen to investigate the prognostic significance of FAM-related genes and FAM subtypes. All statistical P value were two side, with p < 0.05 as statistically significance. All data processing was done in R 4.0.1 software. Results Identification of FAM-related module and genes Based on the FAM ssGSEA score we calculated by the ssGSEA algorithm, a gene co-expression network was constructed with the WGCNA algorithm to identify FAM-related module. The most critical parameter of the soft threshold power was set at 4 (Figure 1a). Then, a hierarchical clustering tree was established to identify the key modules and genes (Figure 1b). Figure 1c showed that the blue module was positively correlated with FAM, which termed as FAM-related module, and the genes in the blue module were regarded as FAM-related genes (n=1860) (Table S1). Different FAM clusters mediated by FAM-related genes Prognostic analysis (including univariate/multivariate Cox regression model and Lasso regression algorithm) were performed to identify the prognostic values of these FAM-related genes (Figure 2a-c and Table S1). Then, Consensus Clustering analysis of the NMF algorithm were applied to classify patients with qualitatively different FAM clusters based on the expression of prognostic FAM-related genes (Figure 2d). Three distinct FAM clusters were eventually identified, including 90 cases in FAM cluster C1, 76 cases in FAM cluster C2 and 55 cases in FAM cluster C3 (Table S2). Prognostic analysis revealed that particularly prominent survival advantage in FAM cluster-C2, whereas the worst prognosis found in FAM cluster-C3 (Figure 2e-f). GSVA algorithm showed significant differences in KEGG pathways among these distinct FAM clusters (Figure 2 g-h). TME cell infiltration characteristics in distinct FAM clusters ssGSEA was performed to explore the enrichment levels of 28 immune signatures in the TME of TNBC, which classified METABRIC-TNBC samples into three distinct immune subtypes (Figure 3a). 67 patients were set at the high-immunity group, which characterized by abundance of immune cell infiltration; The low-immunity group contained 57 patients, which characterized by the suppression of immunity, and 97 patients were present in the modulate-immunity group, which characterized by incomplete immune cell infiltration and immune activation. Then, we analyzed the distribution of immune signatures among these distinct FAM clusters. As shown in Figure 3b, FAM cluster-C2 was markedly associated with high-immunity group, whereas FAM cluster-C1 presented high proportion of low-immunity group. Then, CIBERSORT algorithm and MCPcounter method were used to show the differences on the compositions of TME cell types between the three FAM patterns. As shown in Figure 3c-d, FAM cluster-C2 was remarkably enriched in innate immune cell infiltration including natural killer cell, plasma cells, Myeloid dendritic cell, and cytotoxic lymphocytes compared to other FAM clusters. Moreover, ESTIMATE analysis found that the diversity distribution of the immune and stromal scores between different FAM clusters, which suggested that FAM plays an inevitable role in tumor microenvironment immune profiles (Figure 3e-f). Prognostic DEGs between different FAM clusters Considering the prominently prognostic difference among the FAM-clusters, we further examined the potential FAM-related transcriptional expression change across three FAM clusters in METABRIC-TNBC samples. A total of 782 DEGs were identified, which depicted in Figure 4a and Table S2. Univariate Cox regression model based on the 782 DEGs was performed to find prognostic FAM-related DEGs. Functional enrichment analysis of these prognostic FAM-related DEGs revealed that 49 biological processes (BP) related to immune response and T cell activation, 23 cellular components (CC), including T cell complex and plasma membrane, and 14 molecular functions (MF) refer to SH3/SH2 adaptor activity and transmembrane signaling receptor activity were significant enriched; 13 KEGG pathways related to primary immunodeficiency and T cell receptor signaling pathway were significant over-represented (Figure 4b-e). Construction of FAM gene signature and FAM scoring system Lasso regression analysis based on the obtained 140 prognostic FAM-related DEGs were performed to classify patients into different genomic subtypes (Figure 5a-b). 8 selected FAM-related prognostic genes were identified from the Lasso regression algorithm, which were defined as FAM-related gene signatures. Then, a set of FAM scoring system was constructed to quantify the FAM features of individual patients with TNBC, we termed as FAM Score (FS). Consistent with the clustering grouping of FAM clusters, two distinct FAM score subgroups were found and we named these 2 subgroups as FS-low and -high. Further survival analysis indicated that significant prognostic differences between the low- and high- FS subgroups (Figure 5 c-d). The low-FS subgroup was proven to be associated with better prognosis, while patients with higher-FS showed worse survival outcomes. Furthermore, multivariate Cox regression model analysis confirmed that the FS scheme could serve as an independent prognostic biomarker for METABRIC-TNBC patients (Figure 5e). To better illustrate the association between FAM features with prognosis, the alluvial diagram was used to visualize the attribute changes of individual patients (Figure 5f). Moreover, obvious differences in KEGG pathways were found between the low- and high FS groups (Figure 5g). TME cell infiltration characteristics in distinct FS subgroups Regarding the immune classification, low-FS subtype consisted of more proportions of high- and medium-immunity tumors, whereas high-FS subtype contained mainly low-immunity tumors. (Figure 6a). The results of Cibersort and MCPcounter methods revealed the different infiltrating abundances of TME immune infiltrating cell types between the low- and high-FS groups. As shown in Figure 6b-c, low-FS subtype was remarkably enriched in activate immune cell infiltration including CD8+ T cells, cytotoxic lymphocytes, plasma cells, NK cells, Myeloid dendritic cell, and mast cells compared to the high FS subtype. ESTIMATE analysis exhibited the diversity of the immune and stromal score between the low- and high-FS groups (Figure 6d-e). Moreover, significant reverse correlation between the expression of immune profiles and FS were found in METABRIC-TNBC cohort (Figure 6f). The above findings demonstrated that low-FS subtype tumors had relatively higher immune infiltration levels compared with that in high-FS subtype. Clinical application of the FAM scoring system in two independent cohorts To further explore the clinical application value of the novel FAM-based classification, we drew attention to the TCGA and GSE58812 cohorts, which comprised 142 and 107 TNBC samples, respectively. First, we calculated the FAM score for each patient and then categorized patients into high- and low-FS subgroups based on the cutoff value of their individual FS. Familiar to the results of METABRIC-TNBC dataset, prognostic analysis also revealed FS-low group was remarkable related to prolonged survival, while FS-high group was characterized by poorer survival (Figure S2a-b and Figure S3a-b). In addition, multivariate Cox regression analysis for TCGA-TNBC cohort also confirmed that FAM score could act as an independent prognostic biomarker in TNBC (Figure S2c). As for TME immune features, ssGSEA algorithm also identified three distinct immunity phenotypes for TCGA and GSE58812 cohorts (Figure S2d and Figure S3c). Regarding the immune classification, low-FS subtype also showed greater proportion of high- and medium-immunity (Figure S2e and Figure S3d). Cibersort and MCPcounter algorithm also revealed that effective immune cells were markedly abundant in low-FS groups (Figure S2f-g and Figure S3e-f). ESTIMATE analysis also exhibited higher immune and stromal score in the low-FS group (Figure S2h-i and Figure S3g-h). Significantly different expression of immune profiles between the low- and high-FS groups were identified in TCGA and GSE58812 cohorts (Figure S2j and Figure S3i). All the above findings demonstrated that the FAM-based classification could be a reliable clinically application for predicting survival and immunotherapy response in TNBC. Landscape of genomic variation and expression of different FS subgroups in TNBC First, we summarized the incidence of copy number variations (CNV) of 8 selected FS gene signatures in TCGA-TNBC samples (Fiure 7a). The locations of CNV alterations of these mutated FAM genes on chromosomes are shown in Figure 7b. Then, we evaluated whether the differential expression of 8 selected FS gene signatures could distinguish TNBC samples from normal samples in TCGA cohort by performing principal component analysis (PCA) (Figure 7c). Furthermore, the mRNA expression levels of these selected FAM gene signature were depicted in Figure 7d, which shown wide diversity between normal and TNBC samples. Next, we further analyzed the distribution differences of somatic mutation and TMB between low- and high-FS subgroups in TCGA-TNBC cohort using the R package “maftools”. As shown in Figure 7e-f, the top 20 genes of mutation frequency were significant different between the low- and high-FS subtypes. The TMB score between the low- and high-FS subgroups in TCGA-TNBC cohort were further summarized in Figure 7g. The above results indicated that the potentially complex interaction between genomic variation and FS classification in TNBC. The role of FS scheme in anti-PD-1/L1 immunotherapy Immunotherapies represented by PD-L1 and PD-1 inhibitors were strongly recommended in antitumor therapy of advanced TNBC patients. Next, we investigated whether the FS system could predict patients’ response to immune checkpoint blockade therapy based on two immunotherapy cohorts, anti-PD-L1 cohort (IMvigor210) and anti-PD-1 cohort (GSE78220). Patients with low FS exhibited significantly clinical benefits (Figure 8b-c and 8f) and a markedly prolonged survival (Figure 8a, e). In addition, we found that higher FS was remarkably associated with desert immune phenotype, whereas patients with lower FS were enriched in inflamed immune phenotype (Figure8d). In summary, the above findings implied that the established FAM based classification was significantly correlated with tumor immune phenotypes and response to anti-PD-1/L1 immunotherapy, which could be a potential and robust biomarker for predicting the clinical response to immunotherapy and survival outcomes in TNBC. Identification of novel candidate compounds targeting the selected FAM-related gene signatures As shown in Figure S4, robust positive correlation was found between the expression level of SH2D1A with IC50 of Nelarabine, Dexamethasone Decadron, Fluphenazine and Asparaginase. The IC50 of Fulvestrant and Raloxifene appeared to be positively associated with the expression level of FBP1, similar results were found in the IC50 of Nelarabine and Hydroxyurea with the expression level of PLCL2. A significantly positive correlation was observed between the expression level of IL18RAP and IC50 of Imatinib (all p < 0.001). The above finding may contribute to exploiting the development of novel treatment strategies for targeting the FAM-related gene signatures in TNBC patients. The mRNA levels and prognostic value of selected FAM-related genes in our cohort The RT qPCR assay showed that the mRNA expression level of selected FAM-related genes (PLCL2, DLL3, DCAF4, CXCL13, RASGEF1A, FBP1, SH2D1A and IL18RAP) in adjuvant tumor tissue and TNBC tissues. In detail, CXCL13, IL18RAP and PLCL2 mRNA were downregulated, while DLL3, DCAF4 and FBP1 were significantly upregulated in TNBC samples compared with that in the paired ANTs (Figure 9A). Furthermore, Kaplan-Meier survival analysis indicated that low expression of CXCL13 and PLCL2 were significantly associated with worse DFS (Figure 9D,I), whereas high expression of FBP1 was correlated with worse DFS in TNBC (Figure 9F). Discussion Studies on classification of tumors based on their FAM relevant profiles are beginning to emerge [ 34 , 35 ] . Increasing evidence demonstrated that FAM took on an indispensable role in the tumorigenesis, immunity regulation as well as chemoresistance in TNBC [ 36 – 38 ] .Thus, understanding the FAM features in shaping immune contexture and heterogeneous of TNBC will providing insights into the interaction of FAM and TME, and guiding more effective immunotherapy strategies for TNBC. To date, the association between FAM and the overall TME infiltration characterizations and heterogeneity of TNBC has not been comprehensively recognized. In this study, we successfully classified TNBCs into two heterogeneous subtypes defined by their intact FAM features, with distinct survival outcomes, genomic alternations, immune profiles. Further analyses highlighted the FS scheme could act as an independent prognostic biomarker for predicting survival in TNBC. In accordance with previous studies, we demonstrated that TNBC displayed distinct immune phenotypes among different FAM features [ 38 ] . More importantly, the FAM-based classification we constructed could predict the response to anti-PD-1/PD-L1 immunotherapy in two cohorts. With the aim of identifying the molecular drivers of distinct FS subtypes in TNBC, we observed that significantly mutated genes in different FS groups. The top 20 genes of mutation frequency were significant different between the low- and high-FS subtypes. Moreover, the 8 FS gene signatures could markedly identify tumor tissues from normal breast tissues. We next confirmed the mRNA levels of these FS gene signatures in our cohort by RT qPCR assay. In accord with the bioinformatics results, we found most of the selected FS genes were differentially expressed in TNBC samples. Kaplan–Meier curve further indicated that differential expression of CXCL13, FBP1 and PLCL2 were significantly associated with clinical outcomes in TNBC patients. Among the selected gene signatures, the expression level of CXCL13 has been found linked to the proinflammatory features of macrophages, could predict the response to the combination of chemotherapy with checkpoint inhibitors for TNBCs [ 39 ] . FBP1, a gluconeogenesis regulatory enzyme, has been found modulate cell proliferation and chemosensitivity by targeting c-myc in breast cancer [ 40 ] . Further mechanism analysis showed that MYC-overexpressing TNBC relied on fatty acid oxidation (FAO), inhibition of FAO may as a potential treatment strategy for MYC-overexpressing TNBC. Similarly, DLL3 has been found high expression in breast cancer and was an independent prognostic factor for OS [ 41 ] . The interesting findings yielded several novel insights for the molecular drivers in creating subtype-specific FAM reprograming in TNBC. Our study also has important implications for clinical translations. First, novel therapeutic strategies targeting FAM vulnerabilities are warranted according to subtype-specific features for individual TNBC patients. Second, although immune checkpoint inhibitors have been shown to be successful across multiple tumor types, including TNBC. However, the response to immunotherapy is still low in this tumor type [ 42 , 43 ] , highlighting other underlying mechanisms may influence the immune responsiveness. Mounting evidence have shown metabolites in the tumor microenvironment affecting the fate of immune cells, and therefore, modulate immune responses [ 44 , 45 ] . On the basis of the FAM features in TNBC, we hypothesized that targeting FAM therapeutic strategies and anti-PD-1/PD-L1 immunotherapy could have a synergistic effect for TNBCs. Although we set a novel FAM-based classification for predicting immunotherapy efficiency and survival on TNBC, some limitations are needed to declare. First, the FS scheme was identified by bioinformatic analysis using retrospective datasets; thus, a prospective cohort of TNBC patients is needed to validate our findings. Besides, in the absence of an appropriate ICI-based TNBC dataset, we hope that the effects of FS scheme on immunotherapy could further verified to strengthen our conclusion. In conclusion, this work demonstrated the indispensable role of FAM in shaping tumor microenvironment and heterogeneous for TNBC. Evaluating the FAM features of individual TNBC patient will contribute to enhance our cognition of tumor heterogeneity and TME infiltration features in TNBC, then providing new potential therapeutic targets and guiding more effective immunotherapy strategies. Declarations Acknowledgements Not applicable. Authors’ contributions XY and YTH designed this work. HMA and WT integrated and analyzed the data. XY wrote this manuscript. XY and JW edited and revised the manuscript. All authors approved this manuscript. Funding No funding. Availability of data and materials All data used in this work can be acquired from the Gene-Expression Omnibus (GEO; https://www.ncbi.nlm.nih.gov/geo/) under the accession number GSE78220 and GSE58812, METABRIC and the TCGA portal (https://portal.gdc.cancer.gov/). Ethics approval and consent to participate The patient data in this work were approved by Sir Run Run Shaw hospital of Zhejiang University School of Medicine. Consent for publication Not applicable. Competing interests The authors declare that they have no competing interests. References Bianchini G, Balko J M, Mayer I A, Sanders M E, Gianni L.Triple-negative breast cancer: challenges and opportunities of a heterogeneous disease. Nat Rev Clin Oncol 2016;13,674-90. Denkert C, Liedtke C, Tutt A, von Minckwitz G.Molecular alterations in triple-negative breast cancer-the road to new treatment strategies. Lancet 2017;389,2430-42. Lehmann B D, Bauer J A, Chen X, Sanders M E, Chakravarthy A B, Shyr Y, Pietenpol J A.Identification of human triple-negative breast cancer subtypes and preclinical models for selection of targeted therapies. J Clin Invest 2011;121,2750-67. Burstein M D, Tsimelzon A, Poage G M, Covington K R, Contreras A, Fuqua S A, Savage M I, Osborne C K, Hilsenbeck S G, Chang J C, Mills G B, Lau C C, Brown P H.Comprehensive genomic analysis identifies novel subtypes and targets of triple-negative breast cancer. Clin Cancer Res 2015;21,1688-98. Bareche Y, Venet D, Ignatiadis M, Aftimos P, Piccart M, Rothe F, Sotiriou C.Unravelling triple-negative breast cancer molecular heterogeneity using an integrative multiomic analysis. Ann Oncol 2018;29,895-902. Garrido-Castro A C, Lin N U, Polyak K.Insights into Molecular Classifications of Triple-Negative Breast Cancer: Improving Patient Selection for Treatment. Cancer Discov 2019;9,176-98. Jiang Y Z, Ma D, Suo C, Shi J, Xue M, Hu X, Xiao Y, Yu K D, Liu Y R, Yu Y, Zheng Y, Li X, Zhang C, Hu P, Zhang J, Hua Q, Zhang J, Hou W, Ren L, Bao D, Li B, Yang J, Yao L, Zuo W J, Zhao S, Gong Y, Ren Y X, Zhao Y X, Yang Y S, Niu Z, et al.Genomic and Transcriptomic Landscape of Triple-Negative Breast Cancers: Subtypes and Treatment Strategies. Cancer Cell 2019;35,428-40 e5. Rohrig F, Schulze A.The multifaceted roles of fatty acid synthesis in cancer. Nat Rev Cancer 2016;16,732-49. Bian X, Liu R, Meng Y, Xing D, Xu D, Lu Z.Lipid metabolism and cancer. J Exp Med 2021;218, Vriens K, Christen S, Parik S, Broekaert D, Yoshinaga K, Talebi A, Dehairs J, Escalona-Noguero C, Schmieder R, Cornfield T, Charlton C, Romero-Perez L, Rossi M, Rinaldi G, Orth M F, Boon R, Kerstens A, Kwan S Y, Faubert B, Mendez-Lucas A, Kopitz C C, Chen T, Fernandez-Garcia J, Duarte J A G, Schmitz A A, Steigemann P, Najimi M, Hagebarth A, Van Ginderachter J A, Sokal E, et al.Evidence for an alternative fatty acid desaturation pathway increasing cancer plasticity. Nature 2019;566,403-6. Munir R, Lisec J, Swinnen J V, Zaidi N.Too complex to fail? Targeting fatty acid metabolism for cancer therapy. Prog Lipid Res 2022;85,101143. Wang Y, Wang Y, Ren Y, Zhang Q, Yi P, Cheng C.Metabolic modulation of immune checkpoints and novel therapeutic strategies in cancer. Semin Cancer Biol 2022; Jia H, Truica C I, Wang B, Wang Y, Ren X, Harvey H A, Song J, Yang J M.Immunotherapy for triple-negative breast cancer: Existing challenges and exciting prospects. Drug Resist Updat 2017;32,1-15. Liu L, Wang Y, Miao L, Liu Q, Musetti S, Li J, Huang L.Combination Immunotherapy of MUC1 mRNA Nano-vaccine and CTLA-4 Blockade Effectively Inhibits Growth of Triple Negative Breast Cancer. Mol Ther 2018;26,45-55. Liu J, Liu Q, Li Y, Li Q, Su F, Yao H, Su S, Wang Q, Jin L, Wang Y, Lau W Y, Jiang Z, Song E.Efficacy and safety of camrelizumab combined with apatinib in advanced triple-negative breast cancer: an open-label phase II trial. J Immunother Cancer 2020;8, Liu X, Hartman C L, Li L, Albert C J, Si F, Gao A, Huang L, Zhao Y, Lin W, Hsueh E C, Shen L, Shao Q, Hoft D F, Ford D A, Peng G.Reprogramming lipid metabolism prevents effector T cell senescence and enhances tumor immunotherapy. Sci Transl Med 2021;13, Zheng S, Song Q, Zhang P.Metabolic Modifications, Inflammation, and Cancer Immunotherapy. Front Oncol 2021;11,703681. Ritchie M E, Phipson B, Wu D, Hu Y, Law C W, Shi W, Smyth G K.limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 2015;43,e47. Hanzelmann S, Castelo R, Guinney J.GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 2013;14,7. Langfelder P, Horvath S.WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 2008;9,559. Tibshirani R.The lasso method for variable selection in the Cox model. Stat Med 1997;16,385-95. Gaujoux R, Seoighe C.A flexible R package for nonnegative matrix factorization. BMC Bioinformatics 2010;11,367. Yu G, Wang L G, Han Y, He Q Y.clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS 2012;16,284-7. Charoentong P, Finotello F, Angelova M, Mayer C, Efremova M, Rieder D, Hackl H, Trajanoski Z.Pan-cancer Immunogenomic Analyses Reveal Genotype-Immunophenotype Relationships and Predictors of Response to Checkpoint Blockade. Cell Rep 2017;18,248-62. Barbie D A, Tamayo P, Boehm J S, Kim S Y, Moody S E, Dunn I F, Schinzel A C, Sandy P, Meylan E, Scholl C, Frohling S, Chan E M, Sos M L, Michel K, Mermel C, Silver S J, Weir B A, Reiling J H, Sheng Q, Gupta P B, Wadlow R C, Le H, Hoersch S, Wittner B S, Ramaswamy S, Livingston D M, Sabatini D M, Meyerson M, Thomas R K, Lander E S, et al.Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature 2009;462,108-12. Newman A M, Liu C L, Green M R, Gentles A J, Feng W, Xu Y, Hoang C D, Diehn M, Alizadeh A A.Robust enumeration of cell subsets from tissue expression profiles. Nat Methods 2015;12,453-7. Becht E, Giraldo N A, Lacroix L, Buttard B, Elarouci N, Petitprez F, Selves J, Laurent-Puig P, Sautes-Fridman C, Fridman W H, de Reynies A.Estimating the population abundance of tissue-infiltrating immune and stromal cell populations using gene expression. Genome Biol 2016;17,218. Yoshihara K, Shahmoradgoli M, Martinez E, Vegesna R, Kim H, Torres-Garcia W, Trevino V, Shen H, Laird P W, Levine D A, Carter S L, Getz G, Stemke-Hale K, Mills G B, Verhaak R G.Inferring tumour purity and stromal and immune cell admixture from expression data. Nat Commun 2013;4,2612. Zhang H, Meltzer P, Davis S.RCircos: an R package for Circos 2D track plots. BMC Bioinformatics 2013;14,244. Mayakonda A, Lin D C, Assenov Y, Plass C, Koeffler H P.Maftools: efficient and comprehensive analysis of somatic variants in cancer. Genome Res 2018;28,1747-56. Mariathasan S, Turley S J, Nickles D, Castiglioni A, Yuen K, Wang Y, Kadel E E, III, Koeppen H, Astarita J L, Cubas R, Jhunjhunwala S, Banchereau R, Yang Y, Guan Y, Chalouni C, Ziai J, Senbabaoglu Y, Santoro S, Sheinson D, Hung J, Giltnane J M, Pierce A A, Mesh K, Lianoglou S, Riegler J, Carano R A D, Eriksson P, Hoglund M, Somarriba L, Halligan D L, et al.TGFbeta attenuates tumour response to PD-L1 blockade by contributing to exclusion of T cells. Nature 2018;554,544-8. Hugo W, Zaretsky J M, Sun L, Song C, Moreno B H, Hu-Lieskovan S, Berent-Maoz B, Pang J, Chmielowski B, Cherry G, Seja E, Lomeli S, Kong X, Kelley M C, Sosman J A, Johnson D B, Ribas A, Lo R S.Genomic and Transcriptomic Features of Response to Anti-PD-1 Therapy in Metastatic Melanoma. Cell 2017;168,542. Reinhold W C, Sunshine M, Liu H, Varma S, Kohn K W, Morris J, Doroshow J, Pommier Y.CellMiner: a web-based suite of genomic and pharmacologic tools to explore transcript and drug patterns in the NCI-60 cell line set. Cancer Res 2012;72,3499-511. Hu B, Yang X B, Sang X T.Construction of a lipid metabolism-related and immune-associated prognostic signature for hepatocellular carcinoma. Cancer Med 2020;9,7646-62. Bao M, Shi R, Zhang K, Zhao Y, Wang Y, Bao X.Development of a membrane lipid metabolism-based signature to predict overall survival for personalized medicine in ccRCC patients. EPMA J 2019;10,383-93. Camarda R, Zhou A Y, Kohnz R A, Balakrishnan S, Mahieu C, Anderton B, Eyob H, Kajimura S, Tward A, Krings G, Nomura D K, Goga A.Inhibition of fatty acid oxidation as a therapy for MYC-overexpressing triple-negative breast cancer. Nat Med 2016;22,427-32. Al-Bahlani S, Al-Lawati H, Al-Adawi M, Al-Abri N, Al-Dhahli B, Al-Adawi K.Fatty acid synthase regulates the chemosensitivity of breast cancer cells to cisplatin-induced apoptosis. Apoptosis 2017;22,865-76. Gong Y, Ji P, Yang Y S, Xie S, Yu T J, Xiao Y, Jin M L, Ma D, Guo L W, Pei Y C, Chai W J, Li D Q, Bai F, Bertucci F, Hu X, Jiang Y Z, Shao Z M.Metabolic-Pathway-Based Subtyping of Triple-Negative Breast Cancer Reveals Potential Therapeutic Targets. Cell Metab 2021;33,51-64 e9. Zhang Y, Chen H, Mo H, Hu X, Gao R, Zhao Y, Liu B, Niu L, Sun X, Yu X, Wang Y, Chang Q, Gong T, Guan X, Hu T, Qian T, Xu B, Ma F, Zhang Z, Liu Z.Single-cell analyses reveal key immune cell subsets associated with response to PD-L1 blockade in triple-negative breast cancer. Cancer Cell 2021;39,1578-93 e8. Liu W, Xiong X, Chen W, Li X, Hua X, Liu Z, Zhang Z.High expression of FUSE binding protein 1 in breast cancer stimulates cell proliferation and diminishes drug sensitivity. Int J Oncol 2020;57,488-99. Yuan C, Chang K, Xu C, Li Q, Du Z.High expression of DLL3 is associated with a poor prognosis and immune infiltration in invasive breast cancer patients. Transl Oncol 2021;14,101080. Schmid P, Rugo H S, Adams S, Schneeweiss A, Barrios C H, Iwata H, Dieras V, Henschel V, Molinero L, Chui S Y, Maiya V, Husain A, Winer E P, Loi S, Emens L A, Investigators I M.Atezolizumab plus nab-paclitaxel as first-line treatment for unresectable, locally advanced or metastatic triple-negative breast cancer (IMpassion130): updated efficacy results from a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol 2020;21,44-59. Schmid P, Adams S, Rugo H S, Schneeweiss A, Barrios C H, Iwata H, Dieras V, Hegg R, Im S A, Shaw Wright G, Henschel V, Molinero L, Chui S Y, Funke R, Husain A, Winer E P, Loi S, Emens L A, Investigators I M T.Atezolizumab and Nab-Paclitaxel in Advanced Triple-Negative Breast Cancer. N Engl J Med 2018;379,2108-21. Patel C H, Leone R D, Horton M R, Powell J D.Targeting metabolism to regulate immune responses in autoimmunity and cancer. Nat Rev Drug Discov 2019;18,669-88. Pearce E L, Walsh M C, Cejas P J, Harms G M, Shen H, Wang L S, Jones R G, Choi Y.Enhancing CD8 T-cell memory by modulating fatty acid metabolism. Nature 2009;460,103-7. Additional Declarations No competing interests reported. Supplementary Files TableS1.xlsx TableS2.xlsx Tables3.xlsx FigureS14.pdf Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-1941091","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":127920176,"identity":"091c07ad-d1a5-4f20-a7e5-09cadd94e7bd","order_by":0,"name":"Xia Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3UlEQVRIiWNgGAWjYBACef7mAwcSeNjkGBgOALlsRGgxnHEs8cADGT5j4rUwHMgxPvjARi6xAcwjRgtjwwGDAwk5ZunzG88YMHwoO8zAP7sBvxZ25oaEAwln0nIbG84YMM44d5hB4s4BgrYcOJDYcyy3meGMATNv22EGA4kEQn5JbDiQ+O9/OhtIy1/itCQzgAI5gQekhZEYLcBABmsxnMFwrOBgz7l0HokbBLTI8/d//viDh01efsbhjQ9+lFnL8c8g5DA4kDgAjkweYtUDAX8DCYpHwSgYBaNgRAEA88lKqdDVorgAAAAASUVORK5CYII=","orcid":"","institution":"Sir Run Run Shaw hospital of Zhejiang University School of Medicine","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Xia","middleName":"","lastName":"Yang","suffix":""},{"id":127920177,"identity":"3bf8918c-4a88-40f7-89cb-716ea5861b65","order_by":1,"name":"Wen Tang","email":"","orcid":"","institution":"Sir Run Run Shaw hospital of Zhejiang University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Wen","middleName":"","lastName":"Tang","suffix":""},{"id":127920178,"identity":"b850ba40-f862-45a1-a800-fd587de07998","order_by":2,"name":"Yongtao He","email":"","orcid":"","institution":"Sir Run Run Shaw hospital of Zhejiang University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yongtao","middleName":"","lastName":"He","suffix":""},{"id":127920179,"identity":"25a61812-a1fc-4830-8758-47533f1721af","order_by":3,"name":"Huimin An","email":"","orcid":"","institution":"Sir Run Run Shaw hospital of Zhejiang University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Huimin","middleName":"","lastName":"An","suffix":""},{"id":127920180,"identity":"a8f6f779-e609-42d0-b0cc-b2975b771d0a","order_by":4,"name":"Jin Wang","email":"","orcid":"","institution":"Sir Run Run Shaw hospital of Zhejiang University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Jin","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2022-08-08 11:44:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-1941091/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-1941091/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":25284467,"identity":"bd57c180-5005-4ab5-883f-40dac79a428b","added_by":"auto","created_at":"2022-08-16 19:43:01","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":431169,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification of FAM-related module and genes. a The optimal soft threshold power was chosen as β = 4. b A hierarchical clustering tree was established to identify the FAM-related module. c The blue module was identified positively correlated with FAM, which we termed as FAM moldule.\u0026nbsp;\u003c/p\u003e","description":"","filename":"fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-1941091/v1/2b0256d6aa92605c59db83dd.png"},{"id":25284322,"identity":"c8d776b2-6840-4407-b4b5-da125aa19009","added_by":"auto","created_at":"2022-08-16 19:38:01","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":553884,"visible":true,"origin":"","legend":"\u003cp\u003eDifferent FAM clusters mediated by FAM-related genes. a 78 prognostic FAM genes were further selected by Lasso regression algorithm. b 8 prognostic FAM-related genes were finally identified applying multivariate Cox regression model. c Three distinct FAM clusters were established based on the expression of prognostic FAM genes using Consensus Clustering analysis of the NMF algorithm. d-e Survival analyses for RFS (d) and OS (e) among different FAM clusters in METABRIC-TNBC cohort.\u003c/p\u003e","description":"","filename":"fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-1941091/v1/8fa312ba119c725b71366b60.png"},{"id":25284325,"identity":"9bb620d6-a956-47d6-9ba9-3357ec26cd0c","added_by":"auto","created_at":"2022-08-16 19:38:01","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":603287,"visible":true,"origin":"","legend":"\u003cp\u003eTME cell infiltration characteristics in distinct FAM clusters. a ssGSEA showed three distinct immunity phenotypes were identified in METABRIC-TNBC cohort. b The rate of different immunity phenotypes among different FAM clusters. c Cibersort revealed the abundance of each TME infiltrating cells among different FAM clusters. d MCPcounter revealed the abundance of each immune infiltrating cell types among different FAM clusters. e-f ESTIMATE analysis exhibited the diversity of the immune (e) and stromal score (f) among different FAM clusters\u003c/p\u003e","description":"","filename":"fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-1941091/v1/7065756e5f525fe3b7367c7d.png"},{"id":25284329,"identity":"97e0ef26-e3e2-48b7-86ae-ee01ca97e5db","added_by":"auto","created_at":"2022-08-16 19:38:01","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":432300,"visible":true,"origin":"","legend":"\u003cp\u003ePrognostic DEGs between different FAM clusters.\u003cstrong\u003e \u003c/strong\u003ea\u003cstrong\u003e \u003c/strong\u003e782 DEGs were identified across three FAM clusters using the R package “limma”. b-e Functional annotation for these DEGs using GO enrichment analysis, with the enrichment of GO-BP (b), GO-CC (c), GO-MF (d) and KEGG pathways of the 140 prognostic DEGs using univariate Cox regression analysis.\u0026nbsp;\u003c/p\u003e","description":"","filename":"fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-1941091/v1/cfbdaf7da49e86bfd718e4ae.png"},{"id":25284643,"identity":"79bd190d-38f8-463c-94fe-65192a606d9c","added_by":"auto","created_at":"2022-08-16 19:48:01","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":386240,"visible":true,"origin":"","legend":"\u003cp\u003eConstruction of FAM gene signature and FAM scoring system. a-b Lasso cox regression algorithm found the prognostic genes from the 140 prognostic DEGs. c-d Survival analyses for RFS (c) and OS (d) between low- and high- FS groups in METABRIC-TNBC cohort. e Multivariate Cox regression analysis confirmed that FS could serve as an independent prognostic biomarker for METABRIC-TNBC samples. f Alluvial diagram showing the changes of FAM clusters, FS subtypes and overall survival. g Differences in KEGG pathways between the low- and high FS groups.\u0026nbsp;\u003c/p\u003e","description":"","filename":"fig5.png","url":"https://assets-eu.researchsquare.com/files/rs-1941091/v1/f1f76bfc82b4677c7b45b1f0.png"},{"id":25284332,"identity":"ad881647-cf60-4631-910f-57c0275634ae","added_by":"auto","created_at":"2022-08-16 19:38:01","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":346696,"visible":true,"origin":"","legend":"\u003cp\u003eTME cell infiltration characteristics in distinct FS subgroups in METABRIC-TNBC cohort. a The rate of different immunity phenotypes between the low- and high FS groups. b Cibersort revealed the abundance of each TME infiltrating cells between the low- and high-FS groups. c MCPcounter revealed the abundance of each immune infiltrating cell types between the low- and high-FS groups. d-e ESTIMATE analysis exhibited the diversity of the immune (d) and stromal score (e) between the low- and high-FS groups. f The expression of immune profiles between the low- and high-FS groups in METABRIC-TNBC cohort.\u003c/p\u003e","description":"","filename":"fig6.png","url":"https://assets-eu.researchsquare.com/files/rs-1941091/v1/7df3039b852f0168c8107533.png"},{"id":25284334,"identity":"31db8e0b-90f2-4f10-971a-66c3a357ad12","added_by":"auto","created_at":"2022-08-16 19:38:01","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":399037,"visible":true,"origin":"","legend":"\u003cp\u003eLandscape of genetic and expression variation of FS in TNBC. a The CNV variation frequency of FS gene signatures in TNBC-TNBC cohort. b The location of CNV alteration of FS gene signatures on 23 chromosomes in TCGA-TNBC cohort. c Principal component analysis for the expression profiles of 8 FS gene signatures to distinguish tumors from normal samples in TCGA-TNBC cohort. d The expression of 8 FS gene signatures between normal and tumor tissues. e-f The waterfall plot of tumor somatic mutation of the low- (e) and high-FS subgroups (f) in TCGA-TNBC cohort. g The TMB status between the low- and high-FS subgroups in TCGA-TNBC cohort.\u003c/p\u003e","description":"","filename":"Fig7.png","url":"https://assets-eu.researchsquare.com/files/rs-1941091/v1/f97281463f2f3ce03fc2f5fc.png"},{"id":25284471,"identity":"af883510-f620-4e50-9c16-477861a516ba","added_by":"auto","created_at":"2022-08-16 19:43:01","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":131335,"visible":true,"origin":"","legend":"\u003cp\u003eThe role of FS scheme in anti-PD-1/L1 immunotherapy. a Survival analyses for low- and high-FS subgroups in the anti-PD-L1 immunotherapy cohort (IMvigor 210). b The proportion of patients with response/nonresponse to PD-L1 blockade immunotherapy in low and high FS groups (IMvigor 210). c The proportion of patients with different therapeutic responses to PD-L1 blockade in low and high FS groups (IMvigor 210). d The proportion of patients with different immunity phenotypes in low and high FS groups (IMvigor 210). e The proportion of patients with different survival status in low and high FS groups (GSE78220). f The proportion of patients with different therapeutic response to PD-1 blockade immunotherapy in low and high FS groups (GSE78220). \u003c/p\u003e\u003cp\u003eSD, stable disease; PD, progressive disease; CR, complete response; PR, partial response.\u003c/p\u003e","description":"","filename":"fig8.png","url":"https://assets-eu.researchsquare.com/files/rs-1941091/v1/861ce8ad9fe7ac9b8a8a2349.png"},{"id":25284472,"identity":"fbf170ed-3ecc-4887-91b8-8d1934f85d36","added_by":"auto","created_at":"2022-08-16 19:43:01","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":222267,"visible":true,"origin":"","legend":"\u003cp\u003eThe mRNA levels and prognostic value of selected FS gene signatures in our cohort. (A) Comparison of mRNA expression levels of selected FS gene signatures in adjacent normal tissues and Tumor tissues by RT-qPCR assay. Kaplan–Meier curve shows the survival diversity between differential expression of DLL3 (B), DCAF4 (C), CXCL13 (D), IL18RAP (E), FBP1 (F), SH2D1A (G), PLCL2 (H) and RASGEF1A (I) in our cohort. Non-significant (ns) P \u0026gt; 0.05, * P \u0026lt; 0.05, ** P \u0026lt; 0.01.\u003c/p\u003e","description":"","filename":"Fig9.png","url":"https://assets-eu.researchsquare.com/files/rs-1941091/v1/55e8bcacb2b99616ddfdd52c.png"},{"id":30654639,"identity":"568b961b-5a07-4aae-ba95-22dbf0b2159d","added_by":"auto","created_at":"2022-12-22 03:59:29","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2610546,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-1941091/v1/e730182d-32d3-4da6-80a6-b1ec227cdbfd.pdf"},{"id":25284469,"identity":"397bcb48-4c5e-45a2-a4b7-d139f492c1ce","added_by":"auto","created_at":"2022-08-16 19:43:01","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":35542,"visible":true,"origin":"","legend":"","description":"","filename":"TableS1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-1941091/v1/9b2064ffb5fb5e8880665068.xlsx"},{"id":25284324,"identity":"d741b212-7044-4d59-957d-7608d3d5313e","added_by":"auto","created_at":"2022-08-16 19:38:01","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":25320,"visible":true,"origin":"","legend":"","description":"","filename":"TableS2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-1941091/v1/8feeab28fa7008308bf3ee7f.xlsx"},{"id":25284852,"identity":"ef906d50-87d4-40fb-ae58-194456bdc2a1","added_by":"auto","created_at":"2022-08-16 19:53:01","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":10789,"visible":true,"origin":"","legend":"","description":"","filename":"Tables3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-1941091/v1/07dabfb48c0461a90bffa156.xlsx"},{"id":25284645,"identity":"28c4c495-b18a-42b6-a063-eb5e383b662f","added_by":"auto","created_at":"2022-08-16 19:48:01","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":566196,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS14.pdf","url":"https://assets-eu.researchsquare.com/files/rs-1941091/v1/ec1dcc01e291edc9224ece72.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Development of a novel lipid metabolism-based signature to predict survival and immune response in triple negative breast cancer","fulltext":[{"header":"Introduction","content":"\u003cp\u003eTriple-negative breast cancer (TNBC), as defined by the absence of estrogen receptor (ER) and progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2), is a heterogeneous breast cancer subtype that carries the worst prognosis due to its aggressive characteristics and limited therapeutic options\u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]\u003c/sup\u003e. Much effort has been devoted over the past decade in classifying TNBCs into several molecular subtypes with distinct mutational profiles, genomic alterations, and biological processes that could guide treatment decisions\u003csup\u003e[\u003cspan additionalcitationids=\"CR4 CR5 CR6\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eLipid metabolism, especially the synthesis of fatty acids (FAs), is an important cellular process that converts nutrients into metabolic intermediates for membrane biosynthesis, energy storage, and signal molecule production\u003csup\u003e[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/sup\u003e. Abnormal lipid metabolism is one of the hallmarks of cancer. Increasing evidence confirmed that the significance role of FAM in carcinogenesis, including cell\u0026ndash;matrix interaction, cell signaling and communication, tumor angiogenesis and metastasis, and immune modulation \u003csup\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e. Targeting FAM process has become a promising therapeutic strategy for tumors\u003csup\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e. However, research on FAM-relevant molecular classification of TNBC has not yet been established.\u003c/p\u003e \u003cp\u003eRecently, immunotherapy represented by specific immune checkpoint inhibitors (ICIs), such as anti-CTLA-4 and anti-PD-1/L1, have achieved a marked durable response in some types of cancer, but the overall response rate is still unsatisfied in TNBC\u003csup\u003e[\u003cspan additionalcitationids=\"CR14\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/sup\u003e. The functional status of T cells is a key determinant of effective antitumor immunity and immunotherapy. Therefore, it is important to further elucidate the molecular mechanism of T cell dysfunction in tumor microenvironment.\u003c/p\u003e \u003cp\u003eThe mechanism of Immunometabolism in regulating the function and fate of immune cells has been widely concerned. Previous studies found that tumor cells and Treg cells drove elevated expression of group IVA phospholipase A2, then altered lipid metabolism and senescence of T cells. Inhibition of group IVA phospholipase A2 reprogrammed effector T cell lipid metabolism, effectively prevented T cell senescence, and enhanced anti-tumor immunity and immunotherapy efficacy\u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/sup\u003e. Meanwhile, S-palmitoylation, a lipid process that covalently binds palmitic acid to protein residues, has been found to play an indispensable role in maintaining PD-L1 stability and inhibiting T cell cytotoxicity\u003csup\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/sup\u003e. However, the specific effects of FAM on the tumor microenvironment immune profiles in TNBC are not completed recognized.\u003c/p\u003e \u003cp\u003eIn this study, we comprehensively evaluated the association between FAM and TME cell-infiltrating characteristics and heterogeneous by integrating the transcriptomic and genomic data of 470 TNBC samples from METABRIC, TCGA and GEO databases. WGCNA was performed to identify FAM-related genes in TNBC patients. Then, 3 distinct FAM clusters with nonnegative matrix factorization (NMF) clustering were identified. Moreover, we constructed a scoring scheme to quantify the FAM features in individual TNBC patient. The prognosis traits, genomic variations, transcriptome features, as well as immune infiltration among the different FAM subtypes were further analyzed and verified. These findings suggested that FAM plays an indispensable role in shaping the tumor immune microenvironment profiles and heterogeneous for TNBC.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eData acquisition and preparation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe workflow was shown in Figure S1A. TNBC patients with full clinical annotations and RNA-seq data were searched from Molecular Taxonomy of Breast Cancer International Consortium (METABRIC, http://www.cbioportal.org/datasets), the Cancer Genome Atlas (TCGA, https://portal. gdc.cancer.gov/repository), and Gene-Expression Omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo/) datasets. In total, 470 TNBC samples were included in our study for further analysis. The microarray data of the METABRIC (N=221) was served as training dataset. The normalized RNA-seq data which acquired from TCGA database (N=142) and the gene expression profiles from GSE58812 (N=107) were used as independent validation datasets. R package \u0026lsquo;limma\u0026rsquo; was applied for gene expression normalization\u003csup\u003e[18]\u003c/sup\u003e. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIdentification of FAM-related genes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe hallmark gene sets of fatty acid metabolism (FAM) which including 158 genes were obtained from the Molecular Signatures Data base (MSigDB) (https://www.gseamsigdb.org/gsea/msigdb/). First, we evaluated the FAM ssGSEA score in METABRIC-TNBC samples by the ssGSEA algorithm (R package \u0026ldquo;gsva\u0026rdquo;) \u003csup\u003e[19]\u003c/sup\u003e. Then, we screened out the FAM module and FAM genes by the R package \u0026ldquo;wgcna\u0026rdquo;\u003csup\u003e[20]\u003c/sup\u003e. Totally, we employed Pearson\u0026rsquo;s correlation matrices, co-expression similarity matrix, and average linkage method to evaluate the correlation coefficient between any two genes. A weighted adjacency matrix with a scale free co-expression network and topological overlap matrix (TOM) were constructed to evaluate the connectivity and dissimilarity of the co-expression network. Based on the TOM dissimilarity, a hierarchical clustering tree was established, which could identify the key modules and genes. Then, we set the module membership (MM) \u0026gt;0.8 and gene significance (GS) \u0026gt; 0.5 to identify the correlation between genes and FAM ssGSEA score. Totally, 1860 candidate FAM-related genes were identified from the FAM module.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePrognostic value analysis \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe prognostic significance of the FAM-related genes which obtained from the WGCNA was analyzed by univariate Cox regression model using the R package \u0026ldquo;survival\u0026rdquo;. Then, LASSO cox regression algorithm was applied to further select prognostic FAM genes using the R package \u0026ldquo;glmet\u0026rdquo;\u003csup\u003e[21]\u003c/sup\u003e. Subsequently, multivariate Cox regression model was performed to identify the most robust prognostic gene signatures for OS in METABRIC-TNBC patients.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNon-negative matrix factorization (NMF) clustering analysis \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNon-negative matrix factorization (NMF) clustering analysis was applied to identify distinct FAM clusters based on the expression of 8 prognostic FAM-related genes. The optimal number of clusters and their stability were determined by the consensus clustering algorithm. The R package \u0026ldquo;NMF\u0026rdquo; was used to perform the consensus clustering\u003csup\u003e[22]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIdentification of differentially expressed genes (DEGs) between distinct FAM clusters\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe differentially expressed genes (DEGs) between different FAM clusters were identify using the R package \u0026ldquo;limma\u0026rdquo;\u003csup\u003e[18]\u003c/sup\u003e. The significance criteria for determining DEGs was set as adjusted P value \u0026lt; 0.01.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGene set variation analysis (GSVA) and functional annotation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGSVA enrichment analysis using \u0026ldquo;GSVA\u0026rdquo; R packages were performed to investigate the variation in biological process between different FAM clusters\u003csup\u003e[19]\u003c/sup\u003e. The gene sets of \u0026ldquo;c2.cp.kegg.v7.1.symbols\u0026rdquo; were downloaded from MSigDB database for running GSVA analysis. Adjusted P value \u0026lt;0.05 was considered as statistically significance. The \u0026ldquo;clusterProfiler\u0026rdquo; R package was used to perform functional annotation for DEGs between different FAM clusters, with the cutoff value of FDR \u0026lt; 0.05 and \u003cem\u003eP \u003c/em\u003e\u0026lt; 0.05\u003csup\u003e[23]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEstimation of TME cell infiltration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003essGSEA algorithm was used to quantify the relative abundance of 28 immune cell types in the TME\u003csup\u003e[24, 25]\u003c/sup\u003e. The relative abundance of each immune cell type in each sample were represented by the enrichment scores which calculated by ssGSEA analysis. CIBERSORT algorithm was applied to analyze the compositions of 22 types of tumor-infiltrating immune cells among different FAM subgroups \u003csup\u003e[26]\u003c/sup\u003e. MCP-counter scores regarding immune-related activity and fibroblasts were evaluated using the \u0026lsquo;MCPcounter\u0026rsquo; package\u003csup\u003e[27]\u003c/sup\u003e. The immune and stromal scores were further investigated to predict the level of infiltrating immune and stromal components in the TME using the \u0026lsquo;estimate\u0026rsquo; package\u003csup\u003e[28]\u003c/sup\u003e. Moreover, the expression of key immune profiles were compared between different FAM clusters.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGeneration of a novel FAM-based classification \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo quantify the FAM features of individual TNBC patient, we explored a novel FAM-based classification\u0026mdash;the FAM scoring system (FS) to investigate the FAM features of individual TNBC patient. Specifically, 782 overlap DEGs were identified from different FAM clusters, we then extracted the prognostic gene signatures using univariate Cox regression model and Lasso regression algorithm. Finally, 8 genes were chosen to construct the FAM scoring system. The FAM Score (FS) was calculated by the corresponding coefficients of selected gene signatures:\u003c/p\u003e\n\u003cp\u003eFAM Score = \u0026Sigma;\u003cem\u003ei \u003c/em\u003eCoefficient（mRNA）* Expression（mRNA）\u003c/p\u003e\n\u003cp\u003eWhere\u003cem\u003e i\u003c/em\u003e represent the selected gene signatures.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGene set variation analysis (GSVA) \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe then performed GSVA to further reveal the most significantly enriched molecular pathways between different FS subgroups using the R package \u0026ldquo;GSVA\u0026rdquo;\u003csup\u003e[19]\u003c/sup\u003e. The gene sets of \u0026ldquo;c2.cp.kegg.v7.1.symbols\u0026rdquo; were downloaded from MSigDB database. Adjusted P value \u0026lt; 0.05 was considered as statistically significance.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSignificantly mutated genes and tumor mutation burden in different FS groups\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePrincipal component analysis for the expression profiles of 8 FS gene signatures were analyzed and presented between tumor and normal samples in TCGA cohort. Moreover, the CNV variation frequency of 8 FS gene signatures were further evaluated in TCGA-TNBC cohort. The R package of RCircos was adopted to plot the copy number variation landscape of these selected gene signatures in 23 pairs of chromosomes\u003csup\u003e[29]\u003c/sup\u003e. Using the R package maftools\u003csup\u003e[30]\u003c/sup\u003e, the overall mutation landscape was summarized and present in patients with high and low FS subgroups in TCGA cohort. Then, TMB scores based on the TGCA somatic mutation data were calculated to evaluate the mutation status between different FS subgroups. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGenomic and clinical data sets with immune-checkpoint blockade \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe immunotherapeutic cohorts: IMvigor210 cohort (advanced urothelial cancer treated with atezolizumab, anti PD-L1 antibody)\u003csup\u003e[31]\u003c/sup\u003e and GSE78220 cohort (metastatic melanoma with intervention of pembrolizumab, an anti PD-1 antibody)\u003csup\u003e[32]\u003c/sup\u003e were chosen to analyze the predictive efficiency of FS scheme for immunotherapy.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eExploration of potential compounds targeting the selected FAM-related gene signatures \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo explore potential compounds targeting the selected FS-related gene signatures for treatment of TNBC, we calculated the therapeutic response based on the half-maximal inhibitory concentration (IC50) of various molecular obtained from the CellMiner database for each sample\u003csup\u003e[33]\u003c/sup\u003e. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eValidation of the bioinformatics results using RT-qPCR assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e63 paired TNBC tissues and adjacent normal tissues were acquired from Sir Run Run Shaw hospital of Zhejiang University School of Medicine. Total RNA was extracted using TRIzol reagent (Invitrogen, USA) according to the manufacturer\u0026rsquo;s protocol. Reverse transcription was conducted using PrimeScript RT MasterMix (Takara, China). RT qPCR was performed using SYBR Green PCR MasterMix (Takara, China) according to the manufacturer\u0026rsquo;s protocol. The primers were listed in (Table S3). The average Ct value were used for each gene which repeated three times, \u0026beta;-actin was applied to normalize the target genes mRNA expression. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStudent\u0026rsquo;s t-tests were applied to analyze normally distributed variables, Wilcoxon rank-sum test was performed to evaluate non-normally distributed variables. One-way ANOVA and Kruskal-Wallis tests were used to conduct difference comparisons of more than two groups. Kaplan-Meier survival analysis and Cox proportional hazards model were chosen to investigate the prognostic significance of FAM-related genes and FAM subtypes. All statistical P value were two side, with p \u0026lt; 0.05 as statistically significance. All data processing was done in R 4.0.1 software.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eIdentification of FAM-related module and genes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBased on the FAM ssGSEA score we calculated by the ssGSEA algorithm, a gene co-expression network was constructed with the WGCNA algorithm to identify FAM-related module. The most critical parameter of the soft threshold power was set at 4 (Figure 1a). Then, a hierarchical clustering tree was established to identify the key modules and genes (Figure 1b). Figure 1c showed that the blue module was positively correlated with FAM, which termed as FAM-related module, and the genes in the blue module were regarded as FAM-related genes (n=1860) (Table S1).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDifferent FAM clusters mediated by FAM-related genes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePrognostic analysis (including univariate/multivariate Cox regression model and Lasso regression algorithm) were performed to identify the prognostic values of these FAM-related genes (Figure 2a-c and Table S1). Then, Consensus Clustering analysis of the NMF algorithm were applied to classify patients with qualitatively different FAM clusters based on the expression of prognostic FAM-related genes (Figure 2d). Three distinct FAM clusters were eventually identified, including 90 cases in FAM cluster C1, 76 cases in FAM cluster C2 and 55 cases in FAM cluster C3 (Table S2). Prognostic analysis revealed that particularly prominent survival advantage in FAM cluster-C2, whereas the worst prognosis found in FAM cluster-C3 (Figure 2e-f). GSVA algorithm showed significant differences in KEGG pathways among these distinct FAM clusters (Figure 2 g-h). \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTME cell infiltration characteristics in distinct FAM clusters \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003essGSEA was performed to explore the enrichment levels of 28 immune signatures in the TME of TNBC, which classified METABRIC-TNBC samples into three distinct immune subtypes (Figure 3a). 67 patients were set at the high-immunity group, which characterized by abundance of immune cell infiltration; The low-immunity group contained 57 patients, which characterized by the suppression of immunity, and 97 patients were present in the modulate-immunity group, which characterized by incomplete immune cell infiltration and immune activation. Then, we analyzed the distribution of immune signatures among these distinct FAM clusters. As shown in Figure 3b, FAM cluster-C2 was markedly associated with high-immunity group, whereas FAM cluster-C1 presented high proportion of low-immunity group. Then, CIBERSORT algorithm and MCPcounter method were used to show the differences on the compositions of TME cell types between the three FAM patterns. As shown in Figure 3c-d, FAM cluster-C2 was remarkably enriched in innate immune cell infiltration including natural killer cell, plasma cells, Myeloid dendritic cell, and cytotoxic lymphocytes compared to other FAM clusters. Moreover, ESTIMATE analysis found that the diversity distribution of the immune and stromal scores between different FAM clusters, which suggested that FAM plays an inevitable role in tumor microenvironment immune profiles (Figure 3e-f).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePrognostic DEGs between different FAM clusters\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConsidering the prominently prognostic difference among the FAM-clusters, we further examined the potential FAM-related transcriptional expression change across three FAM clusters in METABRIC-TNBC samples. A total of 782 DEGs were identified, which depicted in Figure 4a and Table S2. Univariate Cox regression model based on the 782 DEGs was performed to find prognostic FAM-related DEGs. Functional enrichment analysis of these prognostic FAM-related DEGs revealed that 49 biological processes (BP) related to immune response and T cell activation, 23 cellular components (CC), including T cell complex and plasma membrane, and 14 molecular functions (MF) refer to SH3/SH2 adaptor activity and transmembrane signaling receptor activity were significant enriched; 13 KEGG pathways related to primary immunodeficiency and T cell receptor signaling pathway were significant over-represented (Figure 4b-e). \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConstruction of FAM gene signature and FAM scoring system\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLasso regression analysis based on the obtained 140 prognostic FAM-related DEGs were performed to classify patients into different genomic subtypes (Figure 5a-b). 8 selected FAM-related prognostic genes were identified from the Lasso regression algorithm, which were defined as FAM-related gene signatures. Then, a set of FAM scoring system was constructed to quantify the FAM features of individual patients with TNBC, we termed as FAM Score (FS). Consistent with the clustering grouping of FAM clusters, two distinct FAM score subgroups were found and we named these 2 subgroups as FS-low and -high. Further survival analysis indicated that significant prognostic differences between the low- and high- FS subgroups (Figure 5 c-d). The low-FS subgroup was proven to be associated with better prognosis, while patients with higher-FS showed worse survival outcomes. Furthermore, multivariate Cox regression model analysis confirmed that the FS scheme could serve as an independent prognostic biomarker for METABRIC-TNBC patients (Figure 5e). To better illustrate the association between FAM features with prognosis, the alluvial diagram was used to visualize the attribute changes of individual patients (Figure 5f). Moreover, obvious differences in KEGG pathways were found between the low- and high FS groups (Figure 5g). \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTME cell infiltration characteristics in distinct FS subgroups \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRegarding the immune classification, low-FS subtype consisted of more proportions of high- and medium-immunity tumors, whereas high-FS subtype contained mainly low-immunity tumors. (Figure 6a). The results of Cibersort and MCPcounter methods revealed the different infiltrating abundances of TME immune infiltrating cell types between the low- and high-FS groups. As shown in Figure 6b-c, low-FS subtype was remarkably enriched in activate immune cell infiltration including CD8+ T cells, cytotoxic lymphocytes, plasma cells, NK cells, Myeloid dendritic cell, and mast cells compared to the high FS subtype. ESTIMATE analysis exhibited the diversity of the immune and stromal score between the low- and high-FS groups (Figure 6d-e). Moreover, significant reverse correlation between the expression of immune profiles and FS were found in METABRIC-TNBC cohort (Figure 6f). The above findings demonstrated that low-FS subtype tumors had relatively higher immune infiltration levels compared with that in high-FS subtype.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical application of the FAM scoring system in two independent cohorts\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo further explore the clinical application value of the novel FAM-based classification, we drew attention to the TCGA and GSE58812 cohorts, which comprised 142 and 107 TNBC samples, respectively. First, we calculated the FAM score for each patient and then categorized patients into high- and low-FS subgroups based on the cutoff value of their individual FS. Familiar to the results of METABRIC-TNBC dataset, prognostic analysis also revealed FS-low group was remarkable related to prolonged survival, while FS-high group was characterized by poorer survival (Figure S2a-b and Figure S3a-b). In addition, multivariate Cox regression analysis for TCGA-TNBC cohort also confirmed that FAM score could act as an independent prognostic biomarker in TNBC (Figure S2c). As for TME immune features, ssGSEA algorithm also identified three distinct immunity phenotypes for TCGA and GSE58812 cohorts (Figure S2d and Figure S3c). Regarding the immune classification, low-FS subtype also showed greater proportion of high- and medium-immunity (Figure S2e and Figure S3d). Cibersort and MCPcounter algorithm also revealed that effective immune cells were markedly abundant in low-FS groups (Figure S2f-g and Figure S3e-f). ESTIMATE analysis also exhibited higher immune and stromal score in the low-FS group (Figure S2h-i and Figure S3g-h). Significantly different expression of immune profiles between the low- and high-FS groups were identified in TCGA and GSE58812 cohorts (Figure S2j and Figure S3i). All the above findings demonstrated that the FAM-based classification could be a reliable clinically application for predicting survival and immunotherapy response in TNBC.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLandscape of genomic variation and expression of different FS subgroups in TNBC \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFirst, we summarized the incidence of copy number variations (CNV) of 8 selected FS gene signatures in TCGA-TNBC samples (Fiure 7a). The locations of CNV alterations of these mutated FAM genes on chromosomes are shown in Figure 7b. Then, we evaluated whether the differential expression of 8 selected FS gene signatures could distinguish TNBC samples from normal samples in TCGA cohort by performing principal component analysis (PCA) (Figure 7c). Furthermore, the mRNA expression levels of these selected FAM gene signature were depicted in Figure 7d, which shown wide diversity between normal and TNBC samples. Next, we further analyzed the distribution differences of somatic mutation and TMB between low- and high-FS subgroups in TCGA-TNBC cohort using the R package \u0026ldquo;maftools\u0026rdquo;. As shown in Figure 7e-f, the top 20 genes of mutation frequency were significant different between the low- and high-FS subtypes. The TMB score between the low- and high-FS subgroups in TCGA-TNBC cohort were further summarized in Figure 7g. The above results indicated that the potentially complex interaction between genomic variation and FS classification in TNBC.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe role of FS scheme in anti-PD-1/L1 immunotherapy \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eImmunotherapies represented by PD-L1 and PD-1 inhibitors were strongly recommended in antitumor therapy of advanced TNBC patients. Next, we investigated whether the FS system could predict patients\u0026rsquo; response to immune checkpoint blockade therapy based on two immunotherapy cohorts, anti-PD-L1 cohort (IMvigor210) and anti-PD-1 cohort (GSE78220). Patients with low FS exhibited significantly clinical benefits (Figure 8b-c and 8f) and a markedly prolonged survival (Figure 8a, e). In addition, we found that higher FS was remarkably associated with desert immune phenotype, whereas patients with lower FS were enriched in inflamed immune phenotype (Figure8d). In summary, the above findings implied that the established FAM based classification was significantly correlated with tumor immune phenotypes and response to anti-PD-1/L1 immunotherapy, which could be a potential and robust biomarker for predicting the clinical response to immunotherapy and survival outcomes in TNBC.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIdentification of novel candidate compounds targeting the selected\u003c/strong\u003e \u003cstrong\u003eFAM-related gene signatures \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs shown in Figure S4, robust positive correlation was found between the expression level of SH2D1A with IC50 of Nelarabine, Dexamethasone Decadron, Fluphenazine and Asparaginase. The IC50 of Fulvestrant and Raloxifene appeared to be positively associated with the expression level of FBP1, similar results were found in the IC50 of Nelarabine and Hydroxyurea with the expression level of PLCL2. A significantly positive correlation was observed between the expression level of IL18RAP and IC50 of Imatinib (all p \u0026lt; 0.001). The above finding may contribute to exploiting the development of novel treatment strategies for targeting the FAM-related gene signatures in TNBC patients. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe mRNA levels and prognostic value of selected FAM-related genes in our cohort\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe RT qPCR assay showed that the mRNA expression level of selected FAM-related genes (PLCL2, DLL3, DCAF4, CXCL13, RASGEF1A, FBP1, SH2D1A and IL18RAP) in adjuvant tumor tissue and TNBC tissues. In detail, CXCL13, IL18RAP and PLCL2 mRNA were downregulated, while DLL3, DCAF4 and FBP1 were significantly upregulated in TNBC samples compared with that in the paired ANTs (Figure 9A). Furthermore, Kaplan-Meier survival analysis indicated that low expression of CXCL13 and PLCL2 were significantly associated with worse DFS (Figure 9D,I), whereas high expression of FBP1 was correlated with worse DFS in TNBC (Figure 9F).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eStudies on classification of tumors based on their FAM relevant profiles are beginning to emerge\u003csup\u003e[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]\u003c/sup\u003e. Increasing evidence demonstrated that FAM took on an indispensable role in the tumorigenesis, immunity regulation as well as chemoresistance in TNBC\u003csup\u003e[\u003cspan additionalcitationids=\"CR37\" citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]\u003c/sup\u003e.Thus, understanding the FAM features in shaping immune contexture and heterogeneous of TNBC will providing insights into the interaction of FAM and TME, and guiding more effective immunotherapy strategies for TNBC. To date, the association between FAM and the overall TME infiltration characterizations and heterogeneity of TNBC has not been comprehensively recognized.\u003c/p\u003e \u003cp\u003eIn this study, we successfully classified TNBCs into two heterogeneous subtypes defined by their intact FAM features, with distinct survival outcomes, genomic alternations, immune profiles. Further analyses highlighted the FS scheme could act as an independent prognostic biomarker for predicting survival in TNBC. In accordance with previous studies, we demonstrated that TNBC displayed distinct immune phenotypes among different FAM features\u003csup\u003e[\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]\u003c/sup\u003e. More importantly, the FAM-based classification we constructed could predict the response to anti-PD-1/PD-L1 immunotherapy in two cohorts.\u003c/p\u003e \u003cp\u003eWith the aim of identifying the molecular drivers of distinct FS subtypes in TNBC, we observed that significantly mutated genes in different FS groups. The top 20 genes of mutation frequency were significant different between the low- and high-FS subtypes. Moreover, the 8 FS gene signatures could markedly identify tumor tissues from normal breast tissues. We next confirmed the mRNA levels of these FS gene signatures in our cohort by RT qPCR assay. In accord with the bioinformatics results, we found most of the selected FS genes were differentially expressed in TNBC samples. Kaplan\u0026ndash;Meier curve further indicated that differential expression of CXCL13, FBP1 and PLCL2 were significantly associated with clinical outcomes in TNBC patients.\u003c/p\u003e \u003cp\u003eAmong the selected gene signatures, the expression level of CXCL13 has been found linked to the proinflammatory features of macrophages, could predict the response to the combination of chemotherapy with checkpoint inhibitors for TNBCs\u003csup\u003e[\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]\u003c/sup\u003e. FBP1, a gluconeogenesis regulatory enzyme, has been found modulate cell proliferation and chemosensitivity by targeting c-myc in breast cancer\u003csup\u003e[\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]\u003c/sup\u003e. Further mechanism analysis showed that MYC-overexpressing TNBC relied on fatty acid oxidation (FAO), inhibition of FAO may as a potential treatment strategy for MYC-overexpressing TNBC. Similarly, DLL3 has been found high expression in breast cancer and was an independent prognostic factor for OS\u003csup\u003e[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]\u003c/sup\u003e. The interesting findings yielded several novel insights for the molecular drivers in creating subtype-specific FAM reprograming in TNBC.\u003c/p\u003e \u003cp\u003eOur study also has important implications for clinical translations. First, novel therapeutic strategies targeting FAM vulnerabilities are warranted according to subtype-specific features for individual TNBC patients. Second, although immune checkpoint inhibitors have been shown to be successful across multiple tumor types, including TNBC. However, the response to immunotherapy is still low in this tumor type\u003csup\u003e[\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]\u003c/sup\u003e, highlighting other underlying mechanisms may influence the immune responsiveness. Mounting evidence have shown metabolites in the tumor microenvironment affecting the fate of immune cells, and therefore, modulate immune responses\u003csup\u003e[\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]\u003c/sup\u003e. On the basis of the FAM features in TNBC, we hypothesized that targeting FAM therapeutic strategies and anti-PD-1/PD-L1 immunotherapy could have a synergistic effect for TNBCs.\u003c/p\u003e \u003cp\u003eAlthough we set a novel FAM-based classification for predicting immunotherapy efficiency and survival on TNBC, some limitations are needed to declare. First, the FS scheme was identified by bioinformatic analysis using retrospective datasets; thus, a prospective cohort of TNBC patients is needed to validate our findings. Besides, in the absence of an appropriate ICI-based TNBC dataset, we hope that the effects of FS scheme on immunotherapy could further verified to strengthen our conclusion.\u003c/p\u003e \u003cp\u003eIn conclusion, this work demonstrated the indispensable role of FAM in shaping tumor microenvironment and heterogeneous for TNBC. Evaluating the FAM features of individual TNBC patient will contribute to enhance our cognition of tumor heterogeneity and TME infiltration features in TNBC, then providing new potential therapeutic targets and guiding more effective immunotherapy strategies.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eXY and YTH designed this work. HMA and WT integrated and analyzed the data. XY wrote this manuscript. XY and JW edited and revised the manuscript. All authors approved this manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNo funding.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data used in this work can be acquired from the Gene-Expression Omnibus (GEO; https://www.ncbi.nlm.nih.gov/geo/) under the accession number GSE78220 and GSE58812, METABRIC and the TCGA portal (https://portal.gdc.cancer.gov/).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe patient data in this work were approved by Sir Run Run Shaw hospital of Zhejiang University School of Medicine.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBianchini G, Balko J M, Mayer I A, Sanders M E, Gianni L.Triple-negative breast cancer: challenges and opportunities of a heterogeneous disease. Nat Rev Clin Oncol 2016;13,674-90.\u003c/li\u003e\n\u003cli\u003eDenkert C, Liedtke C, Tutt A, von Minckwitz G.Molecular alterations in triple-negative breast cancer-the road to new treatment strategies. Lancet 2017;389,2430-42.\u003c/li\u003e\n\u003cli\u003eLehmann B D, Bauer J A, Chen X, Sanders M E, Chakravarthy A B, Shyr Y, Pietenpol J A.Identification of human triple-negative breast cancer subtypes and preclinical models for selection of targeted therapies. J Clin Invest 2011;121,2750-67.\u003c/li\u003e\n\u003cli\u003eBurstein M D, Tsimelzon A, Poage G M, Covington K R, Contreras A, Fuqua S A, Savage M I, Osborne C K, Hilsenbeck S G, Chang J C, Mills G B, Lau C C, Brown P H.Comprehensive genomic analysis identifies novel subtypes and targets of triple-negative breast cancer. Clin Cancer Res 2015;21,1688-98.\u003c/li\u003e\n\u003cli\u003eBareche Y, Venet D, Ignatiadis M, Aftimos P, Piccart M, Rothe F, Sotiriou C.Unravelling triple-negative breast cancer molecular heterogeneity using an integrative multiomic analysis. Ann Oncol 2018;29,895-902.\u003c/li\u003e\n\u003cli\u003eGarrido-Castro A C, Lin N U, Polyak K.Insights into Molecular Classifications of Triple-Negative Breast Cancer: Improving Patient Selection for Treatment. Cancer Discov 2019;9,176-98.\u003c/li\u003e\n\u003cli\u003eJiang Y Z, Ma D, Suo C, Shi J, Xue M, Hu X, Xiao Y, Yu K D, Liu Y R, Yu Y, Zheng Y, Li X, Zhang C, Hu P, Zhang J, Hua Q, Zhang J, Hou W, Ren L, Bao D, Li B, Yang J, Yao L, Zuo W J, Zhao S, Gong Y, Ren Y X, Zhao Y X, Yang Y S, Niu Z, et al.Genomic and Transcriptomic Landscape of Triple-Negative Breast Cancers: Subtypes and Treatment Strategies. Cancer Cell 2019;35,428-40 e5.\u003c/li\u003e\n\u003cli\u003eRohrig F, Schulze A.The multifaceted roles of fatty acid synthesis in cancer. Nat Rev Cancer 2016;16,732-49.\u003c/li\u003e\n\u003cli\u003eBian X, Liu R, Meng Y, Xing D, Xu D, Lu Z.Lipid metabolism and cancer. J Exp Med 2021;218,\u003c/li\u003e\n\u003cli\u003eVriens K, Christen S, Parik S, Broekaert D, Yoshinaga K, Talebi A, Dehairs J, Escalona-Noguero C, Schmieder R, Cornfield T, Charlton C, Romero-Perez L, Rossi M, Rinaldi G, Orth M F, Boon R, Kerstens A, Kwan S Y, Faubert B, Mendez-Lucas A, Kopitz C C, Chen T, Fernandez-Garcia J, Duarte J A G, Schmitz A A, Steigemann P, Najimi M, Hagebarth A, Van Ginderachter J A, Sokal E, et al.Evidence for an alternative fatty acid desaturation pathway increasing cancer plasticity. Nature 2019;566,403-6.\u003c/li\u003e\n\u003cli\u003eMunir R, Lisec J, Swinnen J V, Zaidi N.Too complex to fail? Targeting fatty acid metabolism for cancer therapy. Prog Lipid Res 2022;85,101143.\u003c/li\u003e\n\u003cli\u003eWang Y, Wang Y, Ren Y, Zhang Q, Yi P, Cheng C.Metabolic modulation of immune checkpoints and novel therapeutic strategies in cancer. Semin Cancer Biol 2022;\u003c/li\u003e\n\u003cli\u003eJia H, Truica C I, Wang B, Wang Y, Ren X, Harvey H A, Song J, Yang J M.Immunotherapy for triple-negative breast cancer: Existing challenges and exciting prospects. Drug Resist Updat 2017;32,1-15.\u003c/li\u003e\n\u003cli\u003eLiu L, Wang Y, Miao L, Liu Q, Musetti S, Li J, Huang L.Combination Immunotherapy of MUC1 mRNA Nano-vaccine and CTLA-4 Blockade Effectively Inhibits Growth of Triple Negative Breast Cancer. Mol Ther 2018;26,45-55.\u003c/li\u003e\n\u003cli\u003eLiu J, Liu Q, Li Y, Li Q, Su F, Yao H, Su S, Wang Q, Jin L, Wang Y, Lau W Y, Jiang Z, Song E.Efficacy and safety of camrelizumab combined with apatinib in advanced triple-negative breast cancer: an open-label phase II trial. J Immunother Cancer 2020;8,\u003c/li\u003e\n\u003cli\u003eLiu X, Hartman C L, Li L, Albert C J, Si F, Gao A, Huang L, Zhao Y, Lin W, Hsueh E C, Shen L, Shao Q, Hoft D F, Ford D A, Peng G.Reprogramming lipid metabolism prevents effector T cell senescence and enhances tumor immunotherapy. Sci Transl Med 2021;13,\u003c/li\u003e\n\u003cli\u003eZheng S, Song Q, Zhang P.Metabolic Modifications, Inflammation, and Cancer Immunotherapy. Front Oncol 2021;11,703681.\u003c/li\u003e\n\u003cli\u003eRitchie M E, Phipson B, Wu D, Hu Y, Law C W, Shi W, Smyth G K.limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 2015;43,e47.\u003c/li\u003e\n\u003cli\u003eHanzelmann S, Castelo R, Guinney J.GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 2013;14,7.\u003c/li\u003e\n\u003cli\u003eLangfelder P, Horvath S.WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 2008;9,559.\u003c/li\u003e\n\u003cli\u003eTibshirani R.The lasso method for variable selection in the Cox model. Stat Med 1997;16,385-95.\u003c/li\u003e\n\u003cli\u003eGaujoux R, Seoighe C.A flexible R package for nonnegative matrix factorization. BMC Bioinformatics 2010;11,367.\u003c/li\u003e\n\u003cli\u003eYu G, Wang L G, Han Y, He Q Y.clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS 2012;16,284-7.\u003c/li\u003e\n\u003cli\u003eCharoentong P, Finotello F, Angelova M, Mayer C, Efremova M, Rieder D, Hackl H, Trajanoski Z.Pan-cancer Immunogenomic Analyses Reveal Genotype-Immunophenotype Relationships and Predictors of Response to Checkpoint Blockade. Cell Rep 2017;18,248-62.\u003c/li\u003e\n\u003cli\u003eBarbie D A, Tamayo P, Boehm J S, Kim S Y, Moody S E, Dunn I F, Schinzel A C, Sandy P, Meylan E, Scholl C, Frohling S, Chan E M, Sos M L, Michel K, Mermel C, Silver S J, Weir B A, Reiling J H, Sheng Q, Gupta P B, Wadlow R C, Le H, Hoersch S, Wittner B S, Ramaswamy S, Livingston D M, Sabatini D M, Meyerson M, Thomas R K, Lander E S, et al.Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature 2009;462,108-12.\u003c/li\u003e\n\u003cli\u003eNewman A M, Liu C L, Green M R, Gentles A J, Feng W, Xu Y, Hoang C D, Diehn M, Alizadeh A A.Robust enumeration of cell subsets from tissue expression profiles. Nat Methods 2015;12,453-7.\u003c/li\u003e\n\u003cli\u003eBecht E, Giraldo N A, Lacroix L, Buttard B, Elarouci N, Petitprez F, Selves J, Laurent-Puig P, Sautes-Fridman C, Fridman W H, de Reynies A.Estimating the population abundance of tissue-infiltrating immune and stromal cell populations using gene expression. Genome Biol 2016;17,218.\u003c/li\u003e\n\u003cli\u003eYoshihara K, Shahmoradgoli M, Martinez E, Vegesna R, Kim H, Torres-Garcia W, Trevino V, Shen H, Laird P W, Levine D A, Carter S L, Getz G, Stemke-Hale K, Mills G B, Verhaak R G.Inferring tumour purity and stromal and immune cell admixture from expression data. Nat Commun 2013;4,2612.\u003c/li\u003e\n\u003cli\u003eZhang H, Meltzer P, Davis S.RCircos: an R package for Circos 2D track plots. BMC Bioinformatics 2013;14,244.\u003c/li\u003e\n\u003cli\u003eMayakonda A, Lin D C, Assenov Y, Plass C, Koeffler H P.Maftools: efficient and comprehensive analysis of somatic variants in cancer. Genome Res 2018;28,1747-56.\u003c/li\u003e\n\u003cli\u003eMariathasan S, Turley S J, Nickles D, Castiglioni A, Yuen K, Wang Y, Kadel E E, III, Koeppen H, Astarita J L, Cubas R, Jhunjhunwala S, Banchereau R, Yang Y, Guan Y, Chalouni C, Ziai J, Senbabaoglu Y, Santoro S, Sheinson D, Hung J, Giltnane J M, Pierce A A, Mesh K, Lianoglou S, Riegler J, Carano R A D, Eriksson P, Hoglund M, Somarriba L, Halligan D L, et al.TGFbeta attenuates tumour response to PD-L1 blockade by contributing to exclusion of T cells. Nature 2018;554,544-8.\u003c/li\u003e\n\u003cli\u003eHugo W, Zaretsky J M, Sun L, Song C, Moreno B H, Hu-Lieskovan S, Berent-Maoz B, Pang J, Chmielowski B, Cherry G, Seja E, Lomeli S, Kong X, Kelley M C, Sosman J A, Johnson D B, Ribas A, Lo R S.Genomic and Transcriptomic Features of Response to Anti-PD-1 Therapy in Metastatic Melanoma. Cell 2017;168,542.\u003c/li\u003e\n\u003cli\u003eReinhold W C, Sunshine M, Liu H, Varma S, Kohn K W, Morris J, Doroshow J, Pommier Y.CellMiner: a web-based suite of genomic and pharmacologic tools to explore transcript and drug patterns in the NCI-60 cell line set. Cancer Res 2012;72,3499-511.\u003c/li\u003e\n\u003cli\u003eHu B, Yang X B, Sang X T.Construction of a lipid metabolism-related and immune-associated prognostic signature for hepatocellular carcinoma. Cancer Med 2020;9,7646-62.\u003c/li\u003e\n\u003cli\u003eBao M, Shi R, Zhang K, Zhao Y, Wang Y, Bao X.Development of a membrane lipid metabolism-based signature to predict overall survival for personalized medicine in ccRCC patients. EPMA J 2019;10,383-93.\u003c/li\u003e\n\u003cli\u003eCamarda R, Zhou A Y, Kohnz R A, Balakrishnan S, Mahieu C, Anderton B, Eyob H, Kajimura S, Tward A, Krings G, Nomura D K, Goga A.Inhibition of fatty acid oxidation as a therapy for MYC-overexpressing triple-negative breast cancer. Nat Med 2016;22,427-32.\u003c/li\u003e\n\u003cli\u003eAl-Bahlani S, Al-Lawati H, Al-Adawi M, Al-Abri N, Al-Dhahli B, Al-Adawi K.Fatty acid synthase regulates the chemosensitivity of breast cancer cells to cisplatin-induced apoptosis. Apoptosis 2017;22,865-76.\u003c/li\u003e\n\u003cli\u003eGong Y, Ji P, Yang Y S, Xie S, Yu T J, Xiao Y, Jin M L, Ma D, Guo L W, Pei Y C, Chai W J, Li D Q, Bai F, Bertucci F, Hu X, Jiang Y Z, Shao Z M.Metabolic-Pathway-Based Subtyping of Triple-Negative Breast Cancer Reveals Potential Therapeutic Targets. Cell Metab 2021;33,51-64 e9.\u003c/li\u003e\n\u003cli\u003eZhang Y, Chen H, Mo H, Hu X, Gao R, Zhao Y, Liu B, Niu L, Sun X, Yu X, Wang Y, Chang Q, Gong T, Guan X, Hu T, Qian T, Xu B, Ma F, Zhang Z, Liu Z.Single-cell analyses reveal key immune cell subsets associated with response to PD-L1 blockade in triple-negative breast cancer. Cancer Cell 2021;39,1578-93 e8.\u003c/li\u003e\n\u003cli\u003eLiu W, Xiong X, Chen W, Li X, Hua X, Liu Z, Zhang Z.High expression of FUSE binding protein 1 in breast cancer stimulates cell proliferation and diminishes drug sensitivity. Int J Oncol 2020;57,488-99.\u003c/li\u003e\n\u003cli\u003eYuan C, Chang K, Xu C, Li Q, Du Z.High expression of DLL3 is associated with a poor prognosis and immune infiltration in invasive breast cancer patients. Transl Oncol 2021;14,101080.\u003c/li\u003e\n\u003cli\u003eSchmid P, Rugo H S, Adams S, Schneeweiss A, Barrios C H, Iwata H, Dieras V, Henschel V, Molinero L, Chui S Y, Maiya V, Husain A, Winer E P, Loi S, Emens L A, Investigators I M.Atezolizumab plus nab-paclitaxel as first-line treatment for unresectable, locally advanced or metastatic triple-negative breast cancer (IMpassion130): updated efficacy results from a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol 2020;21,44-59.\u003c/li\u003e\n\u003cli\u003eSchmid P, Adams S, Rugo H S, Schneeweiss A, Barrios C H, Iwata H, Dieras V, Hegg R, Im S A, Shaw Wright G, Henschel V, Molinero L, Chui S Y, Funke R, Husain A, Winer E P, Loi S, Emens L A, Investigators I M T.Atezolizumab and Nab-Paclitaxel in Advanced Triple-Negative Breast Cancer. N Engl J Med 2018;379,2108-21.\u003c/li\u003e\n\u003cli\u003ePatel C H, Leone R D, Horton M R, Powell J D.Targeting metabolism to regulate immune responses in autoimmunity and cancer. Nat Rev Drug Discov 2019;18,669-88.\u003c/li\u003e\n\u003cli\u003ePearce E L, Walsh M C, Cejas P J, Harms G M, Shen H, Wang L S, Jones R G, Choi Y.Enhancing CD8 T-cell memory by modulating fatty acid metabolism. Nature 2009;460,103-7.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Fatty acid metabolism, Tumor microenvironment, Immunotherapy, Triple negative breast cancer","lastPublishedDoi":"10.21203/rs.3.rs-1941091/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-1941091/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eThe high heterogeneity of triple negative breast cancer (TNBC) is the main clinical challenge for individualized therapy. Considering that fatty acid metabolism (FAM) plays an indispensable role in tumorigenesis and development of TNBC, we proposed a novel FAM-based classification to characterize the tumor microenvironment immune profiles and heterogeneous for TNBC.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWeighted gene correlation network analysis (WGCNA) was performed to identify FAM-related genes from 221 TNBC samples in Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) dataset. Then, non-negative matrix factorization (NMF) clustering analysis was applied to determine FAM clusters based on the prognostic FAM-related genes, which chosen from the univariate/multivariate cox regression model and the least absolute shrinkage and selection operator (LASSO) regression algorithm. Then, a FAM scoring scheme was constructed to further quantify FAM features of individual TNBC patient based on the prognostic differentially expressed genes (DEGs) between different FAM clusters. Systematically analyses were performed to evaluate the correlation between the FAM scoring system (FS) with survival outcomes, genomic characteristics, tumor microenvironment (TME) features and immunotherapeutic response for TNBC, which were further validated in the Cancer Genome Atlas (TCGA) and GSE58812 datasets. Moreover, the expression level and clinical significancy of the selected FS gene signatures were further validated in our cohort.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003e1860 FAM-genes were screened out using WGCNA. Three distinct FAM clusters were determined by NMF clustering analysis, which allowed to distinguish different groups of patients with distinct clinical outcomes and tumor microenvironment (TME) features. Then, prognostic gene signatures based on the DEGs between different FAM clusters were identified using univariate cox regression analysis and Lasso regression algorithm. A FAM scoring scheme was constructed, which could divide TNBC patients into high and low-FS subgroups. Low FS subgroup, characterized by better prognosis and abundance with effective immune infiltration. While patients with higher FS were featured with poorer survival and lack of effective immune infiltration. In addition, two independent immunotherapy cohorts (Imvigor210 and GSE78220) confirmed that patients with lower FS demonstrated significant therapeutic advantages from anti-PD-1/PD-L1 immunotherapy and durable clinical benefits. Further analyses in our cohort found that the differential expression of CXCL13, FBP1 and PLCL2 were significantly associated with clinical outcomes of TNBC samples.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThis study revealed FAM plays an indispensable role in formation of TNBC heterogeneity and TME diversity. The novel FAM-based classification could provide a promising prognostic predictor and guide more effective immunotherapy strategies for TNBC.\u003c/p\u003e","manuscriptTitle":"Development of a novel lipid metabolism-based signature to predict survival and immune response in triple negative breast cancer","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2022-08-16 19:37:59","doi":"10.21203/rs.3.rs-1941091/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4ca7c44f-5b99-4ba0-9a14-e39cc3c84bcc","owner":[],"postedDate":"August 16th, 2022","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2022-12-22T03:59:14+00:00","versionOfRecord":[],"versionCreatedAt":"2022-08-16 19:37:59","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-1941091","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-1941091","identity":"rs-1941091","version":["v1"]},"buildId":"GqpaHPwrfC8PjnIFayRh5","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}