Vasculogenic Mimicry Related Long Noncoding RNA Signature Reveals New Therapy Strategy in Breast Cancer

Vasculogenic Mimicry Related Long Noncoding RNA Signature Reveals New Therapy Strategy in 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 Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Vasculogenic Mimicry Related Long Noncoding RNA Signature Reveals New Therapy Strategy in Breast Cancer Yukun Cao, Jing Cao, Peng Zou, Shouman Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4150302/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 Vasculogenic mimicry (VM) is linked closely to the tumorigenesis. However, VM-related lncRNAs (VRLs) involved in the mediation of breast cancer (BC) are still unknown. This research aimed to identify a prognostic signature of VRLs in BC and excavate its potential biological function. Methods We obtained RNA-seq and relevant clinical data from The Cancer Genome Atlas database. Then, Cox and the LASSO regression were utilized to construct a multigene signature. The Kaplan-Meier and ROC curves were plotted to evaluate the efficacy of the model. GO and KEGG pathway were performed for patients in high-risk and low-risk groups. SsGSEA and CIBERSORT algorithm were used to observe the relationship in high-risk and low-risk groups and immune cells. Furthermore, we analysed the inhibitory concentration (IC50) values of three representative anti-vasculogenesis drugs of BC in high-risk and low-risk groups to verify drug sensitivity. Results A VRL-based prognostic signature composed by SEMA3B-AS1, MAPT-AS1, AL355512.1 and AP005717.2 was constructed. According to the risk score calculated by this signature, BC patients were divided into high-risk and low-risk groups. Patients in the high-risk group inclined to have a worse prognosis. SsGSEA and CIBERSORT showed that the majority of immune cells e.g., macrophage and CD4 T cell expressed notably higher in high-risk group (p < 0.05). In addition, we analysed the IC50 values of sorafenib, axitinib and AZD4547 in high-risk and low-risk groups, and all these drugs demonstrated favorable sensitivity to high-risk group which indicated that patients in high-risk group might benefit from anti-vasculogenesis drugs. Conclusions Based on bioinformatic analysis, we established a VM-related gene signature to predict the overall survival of BC patients. Apart from this, we characterized the relationship in the signature, immune microenvironment and correlated drugs which may ignite a novel idea of BC therapy. vasculogenic mimicry breast cancer lncRNA immune microenvironment Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 INTRODUCTION Breast cancer (BC) is the most commonly diagnosed malignant cancer and the leading cause of women death in the world, with an estimated 2.26 million new cases and 685000 deaths reported by GLOBOCAN in 2020[ 1 ]. According to previous research, extensive proliferation of tumor cells excessively consumes the existing oxygen and nutrients within normal stomal environment, leading to intracellular hypoxia[ 2 ]. Hypoxia is highly linked to the invasion, metastasis, recurrence and drug resistance of BC and is viewed as a potential biomarker for predicting prognosis of BC[ 3 ]. Thus, it is of great significance to excavate the relationship between tumor-related hypoxia mechanism and BC. Although the pathogenesis of solid tumors caused by hypoxia remains to be discussed, increasing researches suggest that vasculogenic mimicry (VM) is inseparable from the initial hypoxic environment in tumors[ 2 ]. VM is a novel concept of providing blood supply for tumor growth independent of endothelial cells[ 4 ]. In recent years, accumulating evidence indicates that the hypoxic environment in BC cells can induce VM formation, which in turn can promote the development of BC. High hypoxia-inducible factor (HIF) expression maintains the stemness properties of cancer stem cells (CSCs) and induces epithelial-mesenchymal transformation (EMT) and epithelial-endothelial transformation (EEndT) to promote the VM formation[ 5 ]. In triple negative breast cancer (TNBC), tumors locating in hypoxic areas are capable of possessing high proportion of CD133 + CSCs to survive in an oxygen-deprived environment, which can be mediated by EMT factor Twist1[ 6 , 7 ]. Additionally, MDA-MB-231 cells could develop paralleling holoclone morphology and it was holoclone that displayed CD133 + phenotype and formed VM. In addition, holoclone also expressed endothelial cells (ECs) markers such as VE-Cadherin, MMP-2 and MMP-9, which demonstrates that these CD133 + CSCs may contribute to VM in TNBC by inducing transdifferentiation[ 7 , 8 ]. To summarize, BC cells capable of VM exhibit plasticity and transendothelial phenotype, contributing to unregulated growth and aggressive behavior, which may induced by hypoxia[ 9 ]. Plenty of oncogenic mechanisms of protein-coding genes have been well studied. However, long non-coding RNAs (lncRNAs), which were previously regarded as nonsense genome sequence, have been currently linked to multiple cancers[ 10 ]. Compared to other non-coding RNAs (ncRNAs), lncRNAs present various functions on tumorigenesis. Noted examples include the followings: (1) Knockdown of lncRNA TP73-AS1 releases the posttranslational suppression of miR-490-3p mediated by MDA-MB-231 cell VM formation[ 11 , 12 ]. (2) LncRNA RBM5-AS1 mediates the hypoxia-induced activation of Wnt/β-catenin signaling and subsequently exerts important role in proliferation, stemness maintenance, migration and invasion of BC development[ 13 ]. Nowadays, VM has become a novel direction to guide BC therapy. Therefore, identifying key VM-related lncRNAs in BC is worthy of studying. In this research, we constructed a four-LncRNA prognostic model to predict the overall survival (OS) of BC patients. In addition, this predictive signature also evaluated biological enrichment, immune infiltration and drug sensitivity response and it was also validated by internal verification. RESULTS Enrichment Analysis of VM-Related genes The workflow of this study is shown in Fig. 1 . We firstly obtained 19938 protein-coding genes from the BRCA project of the TCGA database and extracted 227 VRGs expression matrix from it. Then, 90 VMDEGs (upregulated genes: 44, downregulated genes: 46) were identified in breast tumor compared to normal breast tissue (Figs. 2 A, C). KEGG pathway analyses revealed that these 90 genes were mainly enriched in the proteoglycans in cancer, HIF-1 signaling pathway, AGE-RAGE signaling pathway, PI3K-Akt signaling pathway, bladder cancer, relaxin signaling pathway, focal adhesion (FA), EGFR tyrosine kinase inhibitor resistance, Rap1 signaling pathway, fluid shear stress and atherosclerosis (Figs. 2 B). GO analysis revealed the functional analysis landscape of these 90 genes in the following three aspects: (1) in the cellular components category, membrane microdomain, membrane raft and platelet alpha granule, etc. can be mainly detected (Figs. 2 D). (2) in the biological process category, wound healing, response to steroid hormone, and regulation of peptidase activity, etc. were mostly enriched (Figs. 2 E). (3) in the molecular function category, signaling receptor activator activity, growth factor activity and cytokine receptor binding, etc., accounted for a large proportion (Figs. 2 F). Construction of the Prognostic VRLs Signature We downloaded 13507 lncRNAs from the BRCA project of the TCGA database. Then, Spearmen correlation analysis was performed between the lncRNAs mentioned above and VMDEGs with a criteria of |R 2 |>0.4 and P < 0.001. A total of 1320 VRLs were ultimately defined [see Supplementary Table S3 ]. Then, we integrated survival information into the gene expression profiles to further identify potential prognostic VRLs. A total of 1320 VRLs were chosen for univariate Cox regression analysis, and 63 VRLs were identified to have strong relation with the prognosis of BC patients. Subsequently, these survival related 63 VRLs were adopted for implementing LASSO regression analysis. Increasing λ led to a decrease in the number of independent variables with coefficients close to 0 (Figs. 3 A). The optimal λ value was determined when partial likelihood deviance was the lowest (Figs. 3 B). Afterwards, 33 VRLs were acquired by univariate Cox and LASSO regression, and we applied bi-directional stepwise multivariate Cox regression to the remaining 33 VRLs with AIC criteria for the final screening of VRLs. Following the above selection, 4 VRLs (SEMA3B-AS1, MAPT-AS1, AL355512.1 and AP005717.2) were ascertained to further establish the prognostic signature [see Supplementary Table S4 ]. Ctyoscape software[ 14 ] intuitively visualized the interactions between these 4 VRLs and its related mRNAs (Figs. 3 D). SEMA3B-AS1 interacted strongly with following five genes: BPIFB1, EGFR, IGF2BP2, KDM4B and SLC9A3R1. AF005717.2 had tight relation with AURKA and FOXM1. AL355512.1 and MAPT-AS1 had only one correlated gene, SLC9A3R1 and KDM4B, respectively. According to multivariate cox regression analysis, filtered VRLs with hazard ratio (HR) > 1 were considered as risk factors. The Sankey diagram (Figs. 3 E) revealed that MAPT-AS1 and SEMA3B-AS1 belonged to protect-type genes as well as AL355512.1 and AF005717.2 belonged to risk-type genes. We plotted heatmap of these 4 VRLs and clinicopathological variables in high-risk and low-risk groups to show the relationship between the signature and the clinical information (Figs. 3 C). Utility of the VRGs Signature as an Independent Prognostic Indicator To verify the effectiveness of the VRLs signature prediction, K-M curves were constructed (Figs. 4 C). We found that patients in high-risk group had shorter overall survival time than those belonging to low-risk group. The risk score of the high-risk and low-risk groups are depicted in Figs. 4 A. The mortality increased along with the increment of risk score (Figs. 4 B). Forest plot for univariate Cox regression (Figs. 4 D) showed that age, T stage, N stage, M stage, Stage and risk score were closely associated with the OS of BC patients. Multivariate Cox regression forest plot (Figs. 4 E) revealed that only age and risk score could be regarded as independent predictors in BC. Satisfactory predictive ability of the signature were indicated by the ROC curve of risk score and the area under curve (AUC) at 1, 2 and 5-year survival time (Figs. 4 F). The predictive nomogram (Figs. 4 G) calculated an aggregate score on the basis of age, T stage, N stage, AJCC stage and risk score. Simultaneously, we performed calibration curve at 1, 2 and 5 years (Figs. 4 H-J) to demonstrate the probability of the survival outcome of each patient. The calibration curve was regarded as the visualization of the results of the Hosmer-Lemeshow goodness-of-fit test and the survival prediction curves at 1, 2 and 5 years fluctuated above and below the survival actuality curve, confirming good synchronization between predictive and actual OS rate. Internal Validation of the Predictive Signature A total of 1023 BC patients were divided into two cohorts (n = 513, n = 510). The clinical characteristics of these two cohorts were shown in Table 1 . In the internal cohort1, we performed K-M curve (Figs. 5 A) to depict the relationship between high-risk and low-risk group. Then we generated ROC curve (Figs. 5 C) to validate the predictive performance. The same validation method performed above was used in the internal cohort2 and relative K-M and ROC curve were showed in Figs. 5 B, D. As revealed in Figs. 5 A and 5 B, patients in the high-risk group had worse prognosis than patients in the low-risk group (Figs. 5 A, p = 2.8405e-05; Figs. 5 B, p = 2.4270e-03). In the cohort 1, the AUC reached 0.762 at 1 year, 0.713 at 3 years and 0.680 at 5 years. In the cohort 2, the AUC reached 0.738 at 1 year, 0.740 at 3 years and 0.680 at 5 years. The results denoted that patients divided into high-risk group had worse prognosis. Apart from this, the 5-years AUCs in the cohort1 and cohort2 were both lower than expectancy, which may reflect the lower accuracy of the signature’s 5-years prediction. Functional Analysis in Different Risk Groups To unveil the different biofunction and signal transduction pathways in high-risk and low-risk groups, the GSEA analysis was conducted. In GSEA, normalized enrichment score (NES) denoted the normalization of the dataset in terms of enrichment in different gene size, positive values of NES meant that genes could be enriched in high-risk group, and vice verse negative values of NES in the pathway indicated that genes could be enriched in low-risk group. The false discovery rate (FDR) could help us determine the rate of false positive finding of NES. The results (Table 2 ) showed four enriched pathways in high-risk group, including ECM receptor interaction, complement and coagulation cascades, FA and MAPK signaling (Figs. 6 A-D). Interestingly, DNA repair pathways including DNA replication, homologous recombination, mismatch repair and RNA degradation as well as immune related pathway containing nature killer cell mediated cytoxicity and immunodeficiency were enriched in low-risk group (Figs. 6 E). Evaluation of TIME Cells Infiltration Characterization We used ssGSEA to assess the landscape of 28 TIME cell infiltration in high-risk and low-risk groups. As shown in Figs. 7 A, there were 18 TIME cells presented significant difference between high-risk group and low-risk group. The majority of these TIME cells were enriched in high-risk subgroup, including gammadelta (γδ) T cell, activated CD4 T cell, regulatory T cell (Treg), myeloid-derived suppressor cell (MDSC), activated CD8 T cell, monocyte, activated dendritic cell (DC), effector memory CD4 T cell, immature B cell, type2 T helper cell (Th2 cell), T follicular helper cell, natural killer (NK) T cell and macrophage. Only five immune cells incorporating mast cell, plasmacytoid dendritic cell (pDC), eosinophil, nature killer cell and central memory CD8 T cell expressed higher in low-risk group. Spearman correlation analysis (Figs. 7 C) revealed that majority of TIME infiltrating cells had correlation with risk score. The CIBERSORT algorithm, which was based on the approach of linear support vector regression, calculated the infiltration percentage of 22 TIME cells per patient. Not all TIME cells were differentially expressed in high-risk and low-risk groups due to the fact that there existed calculation principle discrepancy in ssGSEA and CIBERSORT. However, mutually differentially expressed TIME cells reflected good consistency, e.g., monocyte cells and macrophage cells all expressed higher in high-risk group, whereas, mast cells expressed higher in low-risk group (Figs. 7 B). The heatmap (Figs. 7 D) illustrated that macrophage M0, macrophage M1, macrophage M2 and T cells CD4 memory resting cells accounting for greater fraction in TIME cells of BC patients. These findings indicated that immune biological process might function more active in the high-risk group. Therapeutic Response of the Prognostic Signature To explore potential effective drugs in two risk groups, we compared the estimated IC50 values of 198 drugs across groups. Among those, 3 candidates containing sorafenib, axitinib and AZD4547 which targeted upon vasculogenesis to combat tumorigenesis were chosen and the IC50 values of them were shown in Fig. 8 . All drugs represented lower IC50 in high-risk group, which denoting favorable sensitivity to anti-vasculogenesis therapy. MATERIALS AND METHODS Acquisition of Datasets The fragments per kilobase million (FPKM) of BC transcriptome was downloaded from the BRCA project of the Cancer Genome Atlas (TCGA) data portal ( https://portal.gdc.cancer.gov/ ), and FPKM was subsequently converted to transcripts per million (TPM) value for downstream analysis. LncRNAs and protein-coding genes were classified by using the “TCGAbiolinks” R package[ 15 ]. A total of 227 VM-related genes (VRGs) were extracted from previous articles[ 2 , 12 , 16 , 17 ] and GeneCards database ( https://www.genecards.org/ ) [see Supplementary Table S1 ]. Genes filtered from GeneCards website with a standard of scores > 15 were shortlisted. Clinicopathological information of 1023 patients encompassing complete survival data was obtained from the TCGA-BRCA cohort. The inclusion criteria were as followed: (1) histologically identified BRCA. (2) patients possessed RNA expression and related clinical data. (3) patients with a lack of survival status, tumor stage (T), lymph node stage (N), metastasis stage (M), clinical stage (Stage) and a survival time of less than 30 days were excluded. Functional Enrichment Analysis of Differentially Expressed VM-Related Genes The “limma” R package[ 18 ] was used to screen the VM-related differentially expressed genes (VMDEGs) between tumor tissues and normal tissues with a |log 2 (fold change)|>1 and adjusted p-value < 0.05 screening criteria. The details of VMDEGs were provided in [see Supplementary Table 2]. These VMDEGs, as input data, were included for further functional enrichment annotated on the basis of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) via the “clusterProfiler” R package[ 19 ]. Identification of Differentially Expressed VM-Related Genes The Pearson correlation coefficients were calculated by using “limma” R package to define the correlation between VMDEGs and lncRNAs. The qualified VM-related differentially expressed lncRNAs (VRLs) were identified in accordance with the standard of correlation coefficient |R 2 |>0.4, and p-value < 0.001. Construction of the VRLs Signature A total of 1320 VRLs were obtained after Pearson correlation analysis. Then, univariate Cox analysis based on “survival” R[ 20 ] package with p < 0.05 as the criteria was conducted for screening those genes significantly associated with OS. Moreover, we used “glmnet” R package[ 21 ] to perform the Least absolute and selection operator (LASSO) Cox regression to further filter prognostic candidates. Following above pipelines, we carried out multivariate Cox[ 20 ] regression analysis to confirm optimal VRLs and established VRLs-related prognostic signature. The Akaike information criterion (AIC) was employed to weigh the complexity of the estimated signature and the goodness of this signature to fit the data. After conducting several iterations, the AIC value of the signature kept decreasing and we chose the signature corresponding to the smallest AIC value (AIC = 533.65) as the optimal signature. The formula for VM-related prognostic risk score for each patient was as followed: $$Risk score= \sum _{i=1}^{n} ({Coef}_{i} \times {x}_{i})$$ \(Coef\) represents the coefficient value, and \(x\) represents the expression value of filtered VM-related lncRNAs. The risk score was calculated on the basis of the above formula: Risk score = (-0.1662*SEMA3B-AS1) – (0.1286*MAPT-AS1) + (0.2702*AL355512.1) + (0.1833*AP005717.2). According to the median value of the risk score, patients were divided into high-risk and low-risk groups. Validation of the Predictive Signature In order to identify the capability of the predictive model, the “caret” R package was adapted to randomly divide 1023 breast cancer patients into two cohorts (n = 513, n = 510). In the first internal cohort, we performed K-M curve via the “survminer” R package to depict the relationship between high-risk and low-risk group. Then we generated receiver operating characteristic (ROC) curves by using the “timeROC” R package to validate the predictive performance. The same validation method performed above was used in the second internal cohort. Functional and Gene Set Enrichment Analysis Gene set enrichment analysis (GSEA)[ 22 ] which could be carried out by “clusterProfiler” R package was applied for exploring enriched pathway. We downloaded KEGG and HALLMARK gene sets extracted from the Molecular Signatures Database ( https://www.gsea-msigdb.org/gsea/msigdb ) to excavate distinct biological characteristic between high-risk and low-risk groups. We considered pathways enriched with the criteria of absolute value of NES > 1, FDR < 0.25 and p-value < 0.05 is of significance. Estimation of Immune Cell Infiltration and Immune Microenvironment To further explore the relationship between immune cells and risk score, we used “IOBR” R package[ 23 ] to conduct the process. The single sample gene set enrichment analysis (ssGSEA)[ 24 ] was applied to quantify tumor immune microenvironment (TIME) cells infiltration of each patient between high-risk group and low-risk group. Information of TIME cells was acquired from the study of Charoentong[ 25 ]. The CIBERSORT algorithm[ 26 ] functioned via “IOBR” R package was used to compute the exact proportion of TIME cell infiltration for each patient. Drug Sensitivity Prediction The “oncoPredict” R package[ 27 ] was utilized to predict drug responses in BC patients. Details of the half-maximal inhibitory concentration (IC50) of each cancer cell lines to drugs could be found in Genomics of Drug Sensitivity in Cancer website (GDSC, https://www.cancerrxgene.org/ ). We calculated three presentative anti-vasculogenesis drugs between high-risk and low-risk groups to verify drug sensitivity. Statistical Analysis All data visualization and statistical analysis were performed in R software (version 4.2.2). Continuous variables were compared with log-rank test and Wilcoxon rank-sum test. The relationships between variables were estimated with Pearson test. Statistical significance was set at *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001. Discussion Breast cancer is a heterogeneous disease involving genetic and environmental factors and affects women health around the world[ 28 ]. In recent years, incremental evidence has been proposed that under the oxygen deficiency environment, hypoxia inducible factors can be activated to participate in the process of a large number of cellular pathways and genes transduction, subsequently cause VM formation, which can further promote proliferation, invasion and metastasis of plenty of solid tumors[ 2 , 9 , 29 ]. In hepatocellular carcinoma, overexpression of lncRNA n339260 induces the formation of CSC-like phenotype, which further promotes VM[ 30 ]. In colorectal cancer (CRC), lncRNA NORAD acted as a miR-495-3p sponge to regulate HIF-1α involves hypoxia-induced VM and chemoresistance[ 29 ]. In gastric cancer, a positive feed-back loop between lncRNA PVT1 expression and STAT3/VEGFA axis can be found in VM formation to continuously provide oncogenic effects[ 31 ]. Generally, it has been verified that VM can regulate tumor development via lncRNAs. However, studies on VRLs in breast cancer remain limited. Herein, we firstly collected 227 VRGs and further screened out 90 differentially expressed VMDEGs in breast tumors. The KEGG and GO enrichment analyses of VMDEGs revealed that correlated genes were mainly enriched in FA, oxygen homeostasis regulator and cellular proliferation and metabolism related biological process. These pathways recapitulated that VM cells tended to convert to an embryonic-like phenotype, and subsequently presented aggressive, metastatic and unlimited proliferation characteristics according to previous reports[ 32 ]. Then, a total of 1320 VRLs associated with 90 VMDEGs in BC (|R 2 |>0.4 and P < 0.001) was filtered by Cox and LASSO regression analysis. We founded that SEMA3B-AS1, MAPT-AS1, AL355512.1 and AP005717.2 were correlated to the OS in BC patients. Of these four VRLs, clear evidence shows that SEMA3B-AS1 (SEAS1) acts as a tumor suppressor in TNBC through SEAS1/miR-3940-3p axis[ 33 ]. In an in vitro cellular assay for TNBC, SEAS1 is regarded as a suppressor of tumor-initiating cells (TIC). The inhibitory effect of SEAS1 on the malignancy relies on activated SEMA3B transcription through H3K4 methylation mediated by MLL4[ 34 ]. In addition, MAPT-AS1 has also been identified to have close connection with tumorigenesis of BC. MAPT-AS1 silencing inhibits growth rate of BC cells in cell line derived xenografts (CDX) experiments and upregulates the sensitivity to paclitaxel[ 35 ]. Knockdown of MAPT-AS1 inversely increases the expression of Tau, a microtubule-associated protein, subsequently inclines to decrease the stabilization of microtubules which is modulated by paclitaxel[ 36 , 37 ]. To verify the prognostic accuracy of the 4 VRLs prognostic signature, we established a nomogram, K-M curves and ROC curves, etc. to intuitively show the predictive performance. We also combined different clinical parameters with the model and visualized it by forest plots. The results showed that risk score calculated by the signature could be regarded as independent indicator to predict the survival time of BC patients. In accordance with the magnitude value of risk score, BC patients were divided into high-risk and low-risk groups. GSEA analysis vividly demonstrated the staple pathogenic pathways of BC in these two groups and we perceived that FA, MAPK signaling, complement and coagulation cascades and ECM receptor interaction were mainly enriched in the high-risk group. FA is a liagtion structure when integrins attempt to across the membrane and establish connection with the ECM[ 38 ]. Integrin signaling can be activated when FA is anchored, which in turn promotes the proliferation, survival and stemness of tumor cells. In CRC, there has been proved that lncRNA ITGB8-AS1 can promote its growth and migration through integrin-mediated FA signaling [ 39 ]. Additionally, there exists evidence that FA complexes inhibition impairs EMT and decreases cell motility in TNBC[ 40 , 41 ]. Moreover, previous literatures reported that focal adhesion kinase (FAK) can also affect cell migration, invasion and metastasis via corsstalking with Src and growth factor receptor signaling pathways[ 42 ]. FAK-Src-mediated phosphorylation can recruit and trigger the activation of MAPK/ERK cascade and subsequently promote tumor angiogenesis and progression[ 43 , 44 ]. The ECM, as an important architectural component of the TME, has close interaction with cancer cells[ 45 ]. During the development of tumorigenesis, cancer cells, cancer-associated fibroblasts (CAFs)[ 46 ] and immune-related cells regulated by transforming growth factor β (TGF-β)[ 47 – 49 ] are prone to form a signaling loop, which facilitating the generation of ECM stiffness, and in return, the rigid ECM accelerates tumor growth rate[ 45 , 50 ]. In conclusion, these pathways enriched in high-risk group had strong relationship with VM formation. Furthermore, we applied ssGSEA and CIBERSORT algorithm on the signature to explore the immune landscape of high-risk and low-risk groups. Based on the signature stratification and boxplots visualized via ssGSEA and CIBERSORT, patients in high-risk group had higher infiltration of immunosuppressive cells including Tregs and macrophages, as well as lower infiltration of cytotoxic immune cells incorporating NK cells. In tumor microenvironment (TME) of BC, T effector proliferation could be inhibited via enhancing the Tregs capcity under the immunosuppressive environment exerted by CAFs[ 51 ]. Recently, a single-cell RNA (scRNA) analysis research further identified that a subset of CAFs (CAF-S1) cellular clusters composed of ecm-myCAF and TGFβ-myCAF are capable of forming a reciprocal cross-talk tightly linked to immunotherapy resistance, and this cross-talk must be accomplished in TME with highly Tregs infiltration[ 52 ]. Apart from this, a number of research have discovered that tumor-associated macrophages (TAMs) are able to secrete abundant chemokines, cytokines, growth factors and inflammatory mediators as immune effector cells[ 53 ]. It has been proved that TAMs have the tendency to convert into polarized M2 phenotype under hypoxic region and subsequently induce vascular abnormalities, thereby promoting tumor growth and metastasis[ 54 – 56 ]. In human TNBC cell lines, M2 macrophage secret growth factors VEGF to upregulate lncRNA PCAT6, and further stimulate endothelial tube formation to generate new vessels, facilitating TNBC cell proliferation, migration and invasion[ 55 ]. As depicted in Figs. 7 A and 7 B, M2 macrophage are highly expressed in high-risk group, which denotes worse prognosis. Whereas, for BC patients with low risk score, high infiltration of NK cells frequently implies the opposite prognosis to those in high-risk group. NK cells are currently consided as effector cells similar to cytotoxic T cells (CTLs), exerting natural cytotoxicity against primary tumor cells and metastasis by inhibiting distant metastasis[ 57 ]. To sum up, this 4 VRLs signature can help demonstrate the immune landscape of BC. Ultimately, drugs targeting on vasculogenesis in BC emerge promising results in recent years. A randomized, multicenter prospective trial of transarterial chemoembolization (TACE) plus sorafenib as compared with TACE alone in patients with hepatoelluilar carcinoma (TACTICS trial, NCT01217034, registered in 2nd October, 2010) has demonstrated that the combination of TACE plus sorafenib could markedly improve median progression-free survival (PFS) [25.2 vs 13.5 months; HR = 0.59; 95%CI, 0.41 to 0.87; p = 0.006][ 58 ]. In our research, drug sensitivity of three anti-vasculogenesis agents all showed lower IC50 in high-risk group, which suggeseted that patients with high VM risk score may benefit from anti-vasculogenesis therapy and this outcomes might lay the foundation for subsequent clinical trials in breast cancer. There are still some limitations in our research. We only used TCGA database for validation, and external public databases and real-world data validation are still to be needed to strengthen the evidence. Additionally, the mechanism of the VRLs in BC remains to be further verified by experiments. Conclusion In summary, our study established a novel VM-related signature in BC based on four lncRNAs (SEMA3B-AS1, MAPT-AS1, AL355512.1 and AP005717.2). The satisfactory predictive capability of the signature was verified by TCGA database. We also found that in high-risk group, cytotoxic cells expressed lower, accompanying with higher expression of immunomonitor escaping cells, which indicated poorer prognosis. Ultimately, three anti-vasculogenesis drugs (sorafenib, axitinib and AZD4547) all showed lower IC50, indicating that patients in high-risk group may be more sensitive to anti-vasculogenesis drug therapy. These findings could ignite innovative BC therapeutic ideas in clinical practice. Abbreviations AIC: Akaike information criterion AUC: Area under curve BC: Breast cancer CAF: Cancer-associated fibroblast CDX: Cell line derived xenografts CSC: Cancer stem cells EC: Endothelial cell ECM: Extracellular matrix EEndT: Epithelial-endothelial transformation EGFR: Epidermal growth factor receptor EMT: Epithelial-mesenchymal transformation FDR: False discovery rate FA: Focal adhesion GO: Gene ontology GSEA: Gene set enrichment analysis HIF: Hypoxia-induced factor HR: Hazard ratio IC50: Half maximal inhibitory concentration KEGG: Kyoto encyclopedia of genes and genomes LASSO: Least absolute shrinkage and selection operator LncRNA: Long non-coding RNA MAPK: Mitogen-activate protein kinase MMP-2: Matrix metalloproteinase 2 MMP-9: Matrix metalloproteinase 9 NcRNA: Non-coding RNA NES: Normalized enrichment score PI3K-Akt: Phosphatidyl-inositol 3 kinase-serine/threonine kinase ROC: Receiver operating characteristc RNA-seq: RNA sequencing SsGSEA: Single sample gene sample enrichment analysis TGF-β: Transforming growth factor β TIC: Tumor-initiating cell TIME: Tumor immune microenvironment TME: Tumor microenvironment TNBC: Triple-negative breast cancer VEGF: Vascular endothelial growth factor VE-Cadherin: Vascular endothelial cadherin VM: Vasculogenic mimicry VMDEGs: Vasculogenic mimicry-related differentially expressed genes VRGs: Vasculogenic mimicry-related genes VRLs: Vasculogenic mimicry-related LncRNAs Declarations Ethics Approval and Consent to Participate The research was designed and carried out abiding by the Ethics Committee at Xiangya Hospital of Central South University. Consent for Publication Not applicable. Availability of Data and Materials The data that support the findings of our study were derived from the following resources available in the public domain: The Cancer Genome Atlas Program at https://portal.gdc.cancer.gov/, GeneCards Program at https://www.genecards.org/, Genomics of Drug Sensitivity in Cancer Program at https://www.cancerrxgene.org/ and the Molecular Signatures Database Program at https://www.gsea-msigdb.org/gsea/msigdb. Competing Interests The authors declare no competing interests. Funding This study was funded by the following fundings: Chinese International Postdoctoral Exchange Fellowship Program (Talent-Introduction Program), Grant No. YJ20210330. The Key Program of the Natural Science Foundation of Hunan Province, China, Grant No. 2022SK2041. Author’s Contributions YKC, JC and SMW designed and supervised the study; PZ contributed to data collection. YKC analyzed the datasets and interpreted the results. YKC and JC advised on data analysis and manuscript drafting. All authors critically reviewed the manuscript and consented to final publication. Acknowledgements Not applicable. Additional Files Supplementary Table S1 Format: Supplementary Table S1.xls Supplementary Table S2 Format: Supplementary Table S2.xls Supplementary Table S3 Format: Supplementary Table S3.xls Supplementary Table S4 Format: Supplementary Table S4.xls References Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA A Cancer J Clinicians. 2021;71:209–49. Wei X, Chen Y, Jiang X, Peng M, Liu Y, Mo Y, et al. Mechanisms of vasculogenic mimicry in hypoxic tumor microenvironments. Mol Cancer. 2021;20:7. Lundgren K, Holm C, Landberg G. Common Molecular Mechanisms of Mammary Gland Development and Breast Cancer. Cell Mol Life Sci. 2007;64:3233–47. Maniotis AJ, Folberg R, Hess A, Seftor EA, Gardner LM, Pe’er J, et al. Vascular channel formation by human melanoma cells in vivo and in vitro: vasculogenic mimicry. Am J Pathol. 1999;155:739–52. Hu H, Ma T, Liu N, Hong H, Yu L, Lyu D, et al. Immunotherapy checkpoints in ovarian cancer vasculogenic mimicry: Tumor immune microenvironments, and drugs. International Immunopharmacology. 2022;111:109116. Zhang D, Sun B, Zhao X, Ma Y, Ji R, Gu Q, et al. Twist1 expression induced by sunitinib accelerates tumor cell vasculogenic mimicry by increasing the population of CD133+ cells in triple-negative breast cancer. Mol Cancer. 2014;13:207. Liu TJ, Sun BC, Zhao XL, Zhao XM, Sun T, Gu Q, et al. CD133+ cells with cancer stem cell characteristics associates with vasculogenic mimicry in triple-negative breast cancer. Oncogene. 2013;32:544–53. G M-G, R G-B, Ea M-P, J G-Q, E A, L D. Vasculogenic Mimicry in Breast Cancer: Clinical Relevance and Drivers. Cells. 2021;10. Seftor REB, Hess AR, Seftor EA, Kirschmann DA, Hardy KM, Margaryan NV, et al. Tumor Cell Vasculogenic Mimicry. Am J Pathol. 2012;181:1115–25. Winkle M, El-Daly SM, Fabbri M, Calin GA. Noncoding RNA therapeutics - challenges and potential solutions. Nat Rev Drug Discov. 2021;20:629–51. Tao W, Sun W, Zhu H, Zhang J. Knockdown of long non-coding RNA TP73-AS1 suppresses triple negative breast cancer cell vasculogenic mimicry by targeting miR-490-3p/TWIST1 axis. Biochem Biophys Res Commun. 2018;504:629–34. Luo Q, Wang J, Zhao W, Peng Z, Liu X, Li B, et al. Vasculogenic mimicry in carcinogenesis and clinical applications. J Hematol Oncol. 2020;13:19. Li X, Yang J, Ni R, Chen J, Zhou Y, Song H, et al. Hypoxia-induced lncRNA RBM5-AS1 promotes tumorigenesis via activating Wnt/β-catenin signaling in breast cancer. Cell Death Dis. 2022;13:95. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: A Software Environment for Integrated Models of Biomolecular Interaction Networks. Genome Res. 2003;13:2498–504. Colaprico A, Silva TC, Olsen C, Garofano L, Cava C, Garolini D, et al. TCGAbiolinks: an R/Bioconductor package for integrative analysis of TCGA data. Nucleic Acids Research. 2016;44:e71. Wang J, Xia W, Huang Y, Li H, Tang Y, Li Y, et al. A vasculogenic mimicry prognostic signature associated with immune signature in human gastric cancer. Front Immunol. 2022;13:1016612. Zhang C, Xiao J, Yuan T, He Y, Deng D, Xiao Z, et al. Molecular vasculogenic mimicry–Related signatures predict clinical outcomes and therapeutic responses in bladder cancer: Results from real-world cohorts. Front Pharmacol. 2023;14:1163115. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43:e47. 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. Therneau TM, Grambsch PM. Modeling Survival Data: Extending the Cox Model. New York, NY: Springer; 2000. Friedman JH, Hastie T, Tibshirani R. Regularization Paths for Generalized Linear Models via Coordinate Descent. Journal of Statistical Software. 2010;33:1–22. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005;102:15545–50. Zeng D, Ye Z, Shen R, Yu G, Wu J, Xiong Y, et al. IOBR: Multi-Omics Immuno-Oncology Biological Research to Decode Tumor Microenvironment and Signatures. Front Immunol. 2021;12:687975. Barbie DA, Tamayo P, Boehm JS, Kim SY, Moody SE, Dunn IF, et al. Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature. 2009;462:108–12. Charoentong P, Finotello F, Angelova M, Mayer C, Efremova M, Rieder D, et al. Pan-cancer Immunogenomic Analyses Reveal Genotype-Immunophenotype Relationships and Predictors of Response to Checkpoint Blockade. Cell Rep. 2017;18:248–62. Newman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015;12:453–7. Maeser D, Gruener RF, Huang RS. oncoPredict: an R package for predicting in vivo or cancer patient drug response and biomarkers from cell line screening data. Brief Bioinform. 2021;22:bbab260. K B, J K, Z Z, A H, Mh K, S M-K, et al. Breast cancer: Biology, biomarkers, and treatments. International immunopharmacology. 2020;84. Zhang L, Wu H, Zhang Y, Xiao X, Chu F, Zhang L. Induction of lncRNA NORAD accounts for hypoxia-induced chemoresistance and vasculogenic mimicry in colorectal cancer by sponging the miR-495-3p/ hypoxia-inducible factor-1α (HIF-1α). Bioengineered. 13:950–62. Zhao X, Sun B, Liu T, Shao B, Sun R, Zhu D, et al. Long noncoding RNA n339260 promotes vasculogenic mimicry and cancer stem cell development in hepatocellular carcinoma. Cancer Sci. 2018;109:3197–208. Zhao J, Wu J, Qin Y, Zhang W, Huang G, Qin L. LncRNA PVT1 induces aggressive vasculogenic mimicry formation through activating the STAT3/Slug axis and epithelial-to-mesenchymal transition in gastric cancer. Cell Oncol (Dordr). 2020;43:863–76. Delgado-Bellido D, Serrano-Saenz S, Fernández-Cortés M, Oliver FJ. Vasculogenic mimicry signaling revisited: focus on non-vascular VE-cadherin. Mol Cancer. 2017;16:65. Hu J, Huang H, Xi Z, Ma S, Ming J, Dong F, et al. LncRNA SEMA3B-AS1 inhibits breast cancer progression by targeting miR-3940/KLLN axis. Cell Death Dis. 2022;13:800. Chen D, Chen Z, Wang Z, Hong C, Wang Q, Yang P, et al. LncRNA SEMA3B‐AS1 suppresses the tumor‐initiating characteristics of triple negative breast cancer via engaging in MLL4‐mediated H3K4 trimethylation. Molecular Carcinogenesis. 2023;:mc.23658. Pan Y, Pan Y, Cheng Y, Yang F, Yao Z, Wang O. Knockdown of LncRNA MAPT-AS1 inhibites proliferation and migration and sensitizes cancer cells to paclitaxel by regulating MAPT expression in ER-negative breast cancers. Cell Biosci. 2018;8:7. Simone R, Javad F, Emmett W, Wilkins OG, Lourenço-Almeida F, Barahona-Torres N, et al. MIR-NATs repress MAPT translation and aid proteostasis in neurodegeneration. Nature. 2021;594:117–23. Safinya CR, Chung PJ, Song C, Li Y, Ewert KK, Choi MC. The effect of multivalent cations and Tau on paclitaxel-stabilized microtubule assembly, disassembly, and structure. Adv Colloid Interface Sci. 2016;232:9–16. Desgrosellier JS, Cheresh DA. Integrins in cancer: biological implications and therapeutic opportunities. Nat Rev Cancer. 2010;10:9–22. Lin X, Zhuang S, Chen X, Du J, Zhong L, Ding J, et al. lncRNA ITGB8-AS1 functions as a ceRNA to promote colorectal cancer growth and migration through integrin-mediated focal adhesion signaling. Mol Ther. 2022;30:688–702. Thapa N, Tan X, Choi S, Wise T, Anderson RA. PIPKIγ and talin couple phosphoinositide and adhesion signaling to control the epithelial to mesenchymal transition. Oncogene. 2017;36:899–911. Zhang Y, Sun L, Li H, Ai L, Ma Q, Qiao X, et al. Binding blockade between TLN1 and integrin β1 represses triple-negative breast cancer. Elife. 2022;11:e68481. Mitra SK, Hanson DA, Schlaepfer DD. Focal adhesion kinase: in command and control of cell motility. Nat Rev Mol Cell Biol. 2005;6:56–68. Wu S, Chen M, Huang J, Zhang F, Lv Z, Jia Y, et al. ORAI2 Promotes Gastric Cancer Tumorigenicity and Metastasis through PI3K/Akt Signaling and MAPK-Dependent Focal Adhesion Disassembly. Cancer Research. 2021;81:986–1000. Zhou F, Sun J, Ye L, Jiang T, Li W, Su C, et al. Fibronectin promotes tumor angiogenesis and progression of non-small-cell lung cancer by elevating WISP3 expression via FAK/MAPK/ HIF-1α axis and activating wnt signaling pathway. Exp Hematol Oncol. 2023;12:61. Huang J, Zhang L, Wan D, Zhou L, Zheng S, Lin S, et al. Extracellular matrix and its therapeutic potential for cancer treatment. Signal Transduct Target Ther. 2021;6:153. Li C, Teixeira AF, Zhu H-J, ten Dijke P. Cancer associated-fibroblast-derived exosomes in cancer progression. Mol Cancer. 2021;20:154. Terra M, Oberkampf M, Fayolle C, Rosenbaum P, Guillerey C, Dadaglio G, et al. Tumor-Derived TGFβ Alters the Ability of Plasmacytoid Dendritic Cells to Respond to Innate Immune Signaling. Cancer Res. 2018;78:3014–26. Batlle E, Massagué J. Transforming Grown Factor-β Signaling in Immunity and Cancer. Immunity. 2019;50:924–40. Donzelli S, Milano E, Pruszko M, Sacconi A, Masciarelli S, Iosue I, et al. Expression of ID4 protein in breast cancer cells induces reprogramming of tumour-associated macrophages. Breast Cancer Res. 2018;20:59. Najafi M, Farhood B, Mortezaee K. Extracellular matrix (ECM) stiffness and degradation as cancer drivers. J Cell Biochem. 2019;120:2782–90. Costa A, Kieffer Y, Scholer-Dahirel A, Pelon F, Bourachot B, Cardon M, et al. Fibroblast Heterogeneity and Immunosuppressive Environment in Human Breast Cancer. Cancer Cell. 2018;33:463-479.e10. Kieffer Y, Hocine HR, Gentric G, Pelon F, Bernard C, Bourachot B, et al. Single-Cell Analysis Reveals Fibroblast Clusters Linked to Immunotherapy Resistance in Cancer. Cancer Discovery. 2020;10:1330–51. Schwertfeger KL, Cowman MK, Telmer PG, Turley EA, McCarthy JB. Hyaluronan, Inflammation, and Breast Cancer Progression. Front Immunol. 2015;6. Han S, Wang W, Wang S, Yang T, Zhang G, Wang D, et al. Tumor microenvironment remodeling and tumor therapy based on M2-like tumor associated macrophage-targeting nano-complexes. Theranostics. 2021;11:2892–916. Dong F, Ruan S, Wang J, Xia Y, Le K, Xiao X, et al. M2 macrophage-induced lncRNA PCAT6 facilitates tumorigenesis and angiogenesis of triple-negative breast cancer through modulation of VEGFR2. Cell Death Dis. 2020;11:728. Chen D, Zhang X, Li Z, Zhu B. Metabolic regulatory crosstalk between tumor microenvironment and tumor-associated macrophages. Theranostics. 2021;11:1016–30. Wu S-Y, Fu T, Jiang Y-Z, Shao Z-M. Natural killer cells in cancer biology and therapy. Mol Cancer. 2020;19:120. Kudo M, Ueshima K, Ikeda M, Torimura T, Tanabe N, Aikata H, et al. Randomised, multicentre prospective trial of transarterial chemoembolisation (TACE) plus sorafenib as compared with TACE alone in patients with hepatocellular carcinoma: TACTICS trial. Gut. 2020;69:1492–501. Tables Table 1 | The clinical characteristics of patients in validation cohorts Variables Validation Cohort 1 (n=513) Cohort 2 (n=510) Age <60 283 (55%) 268 (53%) ≥60 230 (45%) 242 (47%) T T1 128 (25%) 143 (28%) T2 307 (60%) 285 (56%) T3 63 (12%) 66 (13%) T4 15 (3%) 16 (3%) N N0 249 (49%) 240 (47%) N1 172 (33%) 180 (35%) N2 56 (11%) 54 (11%) N3 36 (7%) 36 (7%) M M1 506 (99%) 499 (98%) M2 7 (1%) 11 (2%) Stage Stage I 85 (17%) 96 (19%) Stage II 303 (59%) 287 (56%) Stage III 118 (23%) 116 (23%) Stage IV 7 (1%) 11 (2%) Table 2 | Enriched gene sets in high-risk group Gene set ES NES P-value FDR ECM receptor interaction 0.58 1.96 1.80e-05 4.11e-04 Complement and coagulation cascades 0.59 1.89 1.01e-04 9.26e-04 Focal adhesion 0.50 1.88 1.04e-06 3.57e-05 MAPK signaling 0.38 1.48 1.61e-03 9.65e-03 ES, enrichment score; NES, normalized enrichment score; FDR, false discovery rate; ECM, extracellular matrix Additional Declarations No competing interests reported. Supplementary Files SupplementaryTableS1.xls Supplementary Table S1 | Vasculogenic mimicry-related genes. A total of 227 vasculogenic mimicry-related genes extracted from previous literatures and GeneCards data portal. SupplementaryTableS2.xls Supplementary Table S2 | Differentially analysis outcome of vasculogenic mimicry-related genes. Differentially expressed vasculogenic mimicry-related genes including 44 upregulated genes and 46 downregulated genes calculated by “limma” R package. SupplementaryTableS3.xls Supplementary Table S3 | Vasculogenic mimicry genes related lncRNA. Identification of vasculogenic mimicry-related lncRNAs via Spearmen correlation analysis with a criteria of |R 2 |>0.4 and P<0.001. SupplementaryTableS4.xls Supplementary Table S4 | Constituent lncRNAs of vasculogenic mimicry-related signature. A 4 vasculogenic mimicry-related lncRNAs signature after Cox and LASSO regression. 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 Advisory Board 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-4150302","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":283865566,"identity":"007a980f-7db8-4492-8483-1b0692216dff","order_by":0,"name":"Yukun Cao","email":"","orcid":"","institution":"Central South University","correspondingAuthor":false,"prefix":"","firstName":"Yukun","middleName":"","lastName":"Cao","suffix":""},{"id":283865568,"identity":"8dd6b79f-a055-40a1-84e8-99e7ac9dd9d2","order_by":1,"name":"Jing Cao","email":"","orcid":"","institution":"Central South University","correspondingAuthor":false,"prefix":"","firstName":"Jing","middleName":"","lastName":"Cao","suffix":""},{"id":283865570,"identity":"2c840fef-288c-4ca6-a1e7-ab627cded689","order_by":2,"name":"Peng Zou","email":"","orcid":"","institution":"First Affiliated Hospital of Heilongjiang University of Chinese Medicine, Heilongjiang University of Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Peng","middleName":"","lastName":"Zou","suffix":""},{"id":283865571,"identity":"6ad257a4-876e-43dc-ba3b-c06875aed2f5","order_by":3,"name":"Shouman Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxElEQVRIiWNgGAWjYDACCQglB6HYSNBiTLqWxAaitcjPbj4mdaPmTnp//xkDhg9lhxn4Zzfg12Jw51iadM6xZ7kzbuQYMM44d5hB4s4BAlokcsykc9gO526Q4DFg5m07DBRJIOCwGSAt/w6nG/CfMWD+S4wWhhtALblthxMMGHIMmBmJ0WJwIy3ZOrfvsOGMG2kFB3vOpfNI3CDosOSDt3O+HZbn7z+88cGPMms5/hmEHIYMDgAxDwnqR8EoGAWjYBTgAgBDhkBhJjVpFAAAAABJRU5ErkJggg==","orcid":"","institution":"Central South University","correspondingAuthor":true,"prefix":"","firstName":"Shouman","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2024-03-22 14:03:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4150302/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4150302/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":53750743,"identity":"94776be2-ee1f-4c6c-b10f-efc66c21eeba","added_by":"auto","created_at":"2024-03-29 18:41:03","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":48134,"visible":true,"origin":"","legend":"\u003cp\u003eThe workflow of our research. TCGA, The Cancer Genome Atlas; BRCA, Breast cancer; VRGs, VM-related genes; VMDEGs, VM-related differentially expressed genes; VRLs, VM-related differentially expressed long non-coding RNAs\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-4150302/v1/5b0f21d64c2f91e45045cb6c.png"},{"id":53750750,"identity":"6e949d1f-df4c-4dbb-855d-f089476ecf88","added_by":"auto","created_at":"2024-03-29 18:41:04","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1830844,"visible":true,"origin":"","legend":"\u003cp\u003eVisualization of VRGs and functional analyses of VRGs between BC and normal tissues. (A) Volcano plot of 227 VRGs in BC. (B) The heatmap of 227 VRGs in BC and adjacent normal tissues. (C) KEGG analysis of VRGs. (D-F) GO analysis of VRGs. FC, fold change; FDR, false discovery rate; KEGG, Kyoto Encyclopedia of Genes and Genomes; GO, Gene Ontology\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-4150302/v1/41550f0b41a4fe4ef649f11a.png"},{"id":53750742,"identity":"d40972eb-799c-41b0-9a1e-6b5a72452504","added_by":"auto","created_at":"2024-03-29 18:41:03","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":95156,"visible":true,"origin":"","legend":"\u003cp\u003eConstruction of a 4 VRLs signature and the lncRNA-mRNA network of in the predictive signature. (A-B) cvfit and lambda curves showing the LASSO regression was performed with the minimum criteria. (C) Distribution heatmap of 4 VRLs and clinicopathological variables in high/low-risk groups. (D) The co-expression network of VRLs and related mRNA. (E) Sankey diagram of VRLs, related mRNA and risk type. LASSO, the least shrinkage and selection operator.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-4150302/v1/333af2517397f74afdd82676.png"},{"id":53750741,"identity":"c71f384b-4004-400a-b67d-96909800dcf7","added_by":"auto","created_at":"2024-03-29 18:41:03","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":95609,"visible":true,"origin":"","legend":"\u003cp\u003eThe relationship between clinicopathologic variables and the VRLs signature. (A) The distribution of the risk score among BC patients. (B) K-M curves of the OS rate of BC patients in high-risk and low-risk groups. (C) The number of dead and alive patients with different risk scores. (D-E) Results of the univariate Cox regression analysis and the multivariate Cox regression analysis of the 4 VRLs signature. (F) AUCs at 1-year, 2-years and 5-years survival for the signature. (G) The nomogram to illustrate the relation between the risk score and clinicopathological variables. (H-J) The calibration curve for evaluating the accuracy of the nomogram model. K-M, Kaplan-Meier analysis; T, tumor; N, lymph node; M, metastasis; AUC, area under the curve; **p\u0026lt;0.01, ***p\u0026lt;0.001.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-4150302/v1/dd4c1bd25579bcf2515305de.png"},{"id":53752088,"identity":"1dafba26-e62e-48f3-8418-7f31ed081295","added_by":"auto","created_at":"2024-03-29 18:49:03","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":61003,"visible":true,"origin":"","legend":"\u003cp\u003eInternal validation of the predictive signature for OS based on the entire TCGA dataset. (A, C) K-M curves and AUCs at 1-year, 3-years and 5-years survival in the cohort 1. (B, D) K-M curves and AUCs at 1-year, 3-years and 5-years survival in the cohort 2.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-4150302/v1/7f52ae0824f397a7d89b1940.png"},{"id":53750747,"identity":"234f2905-fd10-4c3a-8e99-be91952c4623","added_by":"auto","created_at":"2024-03-29 18:41:04","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":555355,"visible":true,"origin":"","legend":"\u003cp\u003eGSEA showing significant enrichment in high-risk and low-risk groups in BC patients. (A-D) Enriched pathway in high-risk group. (E) Enriched pathway in high/low-risk groups. GSEA, gene enrichment analysis.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-4150302/v1/f052dbd38182eefd620a24b0.png"},{"id":53754211,"identity":"7f2c1b2e-7f7d-4987-91a5-2f10543ea257","added_by":"auto","created_at":"2024-03-29 18:57:03","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":147544,"visible":true,"origin":"","legend":"\u003cp\u003eAnalysis of immune cell infiltration landscape in BC patients. (A-B) The boxplots for the comparison of the immune cells between the high/low-risk groups calculated by ssGSEA and CIBERSORT. (C) The correlation between the risk score and 28 immune cells. (D) The heatmap of the proportion of 22 TIME cells in each BC patient. ssGSEA, single sample gene set enrichment analysis; TIME, tumor immune microenvironment; *p\u0026lt;0.05, **p\u0026lt;0.01, ***p\u0026lt;0.001, ****p\u0026lt;0.00001; ns, no significance.\u003c/p\u003e","description":"","filename":"Figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-4150302/v1/cfcf44b41227be259d3d9d75.png"},{"id":53750748,"identity":"ce3fc215-7dbc-4a1d-b319-4d24d9ec19f7","added_by":"auto","created_at":"2024-03-29 18:41:04","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":41595,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of drugs sensitivity between high-risk group and low-risk group. IC50, half-maximal inhibitory concentration.\u003c/p\u003e","description":"","filename":"Figure8.png","url":"https://assets-eu.researchsquare.com/files/rs-4150302/v1/d13e7a83a33f6b6742ec005e.png"},{"id":53756730,"identity":"4aa7b75e-d2f6-4cea-ab3a-cb2cca41f1e0","added_by":"auto","created_at":"2024-03-29 19:05:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2732256,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4150302/v1/5b47559a-6955-4105-8dca-58ed63946fad.pdf"},{"id":53750744,"identity":"be86b8cf-c642-4310-9896-10a068e4c9eb","added_by":"auto","created_at":"2024-03-29 18:41:03","extension":"xls","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":35840,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Table S1 | Vasculogenic mimicry-related genes. A total of 227 vasculogenic mimicry-related genes extracted from previous literatures and GeneCards data portal.\u003c/p\u003e","description":"","filename":"SupplementaryTableS1.xls","url":"https://assets-eu.researchsquare.com/files/rs-4150302/v1/ba408e4ecb5ed36d8c1d0d19.xls"},{"id":53750739,"identity":"b039c6d0-2e52-4b71-a457-5d1edc9b9261","added_by":"auto","created_at":"2024-03-29 18:41:03","extension":"xls","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":51712,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Table S2 | Differentially analysis outcome of vasculogenic mimicry-related genes. Differentially expressed vasculogenic mimicry-related genes including 44 upregulated genes and 46 downregulated genes calculated by “limma” R package.\u003c/p\u003e","description":"","filename":"SupplementaryTableS2.xls","url":"https://assets-eu.researchsquare.com/files/rs-4150302/v1/b591ca5b3292d27fe3fe72f3.xls"},{"id":53750740,"identity":"bd8d35c0-ac9b-4ce3-be31-d0cbb6788f94","added_by":"auto","created_at":"2024-03-29 18:41:03","extension":"xls","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":402944,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Table S3 | Vasculogenic mimicry genes related lncRNA. Identification of vasculogenic mimicry-related lncRNAs via Spearmen correlation analysis with a criteria of |R\u003csup\u003e2\u003c/sup\u003e|\u0026gt;0.4 and P\u0026lt;0.001.\u003c/p\u003e","description":"","filename":"SupplementaryTableS3.xls","url":"https://assets-eu.researchsquare.com/files/rs-4150302/v1/fda14ee0ad85d699eef2aa28.xls"},{"id":53750752,"identity":"30af1c6d-249f-487b-aa06-1df47161c61a","added_by":"auto","created_at":"2024-03-29 18:41:05","extension":"xls","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":26624,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Table S4 | Constituent lncRNAs of vasculogenic mimicry-related signature. A 4 vasculogenic mimicry-related lncRNAs signature after Cox and LASSO regression.\u003c/p\u003e","description":"","filename":"SupplementaryTableS4.xls","url":"https://assets-eu.researchsquare.com/files/rs-4150302/v1/eda02244fad6e1504e74633c.xls"}],"financialInterests":"No competing interests reported.","formattedTitle":"Vasculogenic Mimicry Related Long Noncoding RNA Signature Reveals New Therapy Strategy in Breast Cancer","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eBreast cancer (BC) is the most commonly diagnosed malignant cancer and the leading cause of women death in the world, with an estimated 2.26\u0026nbsp;million new cases and 685000 deaths reported by GLOBOCAN in 2020[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. According to previous research, extensive proliferation of tumor cells excessively consumes the existing oxygen and nutrients within normal stomal environment, leading to intracellular hypoxia[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Hypoxia is highly linked to the invasion, metastasis, recurrence and drug resistance of BC and is viewed as a potential biomarker for predicting prognosis of BC[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Thus, it is of great significance to excavate the relationship between tumor-related hypoxia mechanism and BC.\u003c/p\u003e \u003cp\u003eAlthough the pathogenesis of solid tumors caused by hypoxia remains to be discussed, increasing researches suggest that vasculogenic mimicry (VM) is inseparable from the initial hypoxic environment in tumors[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. VM is a novel concept of providing blood supply for tumor growth independent of endothelial cells[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. In recent years, accumulating evidence indicates that the hypoxic environment in BC cells can induce VM formation, which in turn can promote the development of BC. High hypoxia-inducible factor (HIF) expression maintains the stemness properties of cancer stem cells (CSCs) and induces epithelial-mesenchymal transformation (EMT) and epithelial-endothelial transformation (EEndT) to promote the VM formation[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. In triple negative breast cancer (TNBC), tumors locating in hypoxic areas are capable of possessing high proportion of CD133\u003csup\u003e+\u003c/sup\u003e CSCs to survive in an oxygen-deprived environment, which can be mediated by EMT factor Twist1[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Additionally, MDA-MB-231 cells could develop paralleling holoclone morphology and it was holoclone that displayed CD133\u003csup\u003e+\u003c/sup\u003e phenotype and formed VM. In addition, holoclone also expressed endothelial cells (ECs) markers such as VE-Cadherin, MMP-2 and MMP-9, which demonstrates that these CD133\u003csup\u003e+\u003c/sup\u003e CSCs may contribute to VM in TNBC by inducing transdifferentiation[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. To summarize, BC cells capable of VM exhibit plasticity and transendothelial phenotype, contributing to unregulated growth and aggressive behavior, which may induced by hypoxia[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e].\u003c/p\u003e \u003cp\u003ePlenty of oncogenic mechanisms of protein-coding genes have been well studied. However, long non-coding RNAs (lncRNAs), which were previously regarded as nonsense genome sequence, have been currently linked to multiple cancers[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Compared to other non-coding RNAs (ncRNAs), lncRNAs present various functions on tumorigenesis. Noted examples include the followings: (1) Knockdown of lncRNA TP73-AS1 releases the posttranslational suppression of miR-490-3p mediated by MDA-MB-231 cell VM formation[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. (2) LncRNA RBM5-AS1 mediates the hypoxia-induced activation of Wnt/β-catenin signaling and subsequently exerts important role in proliferation, stemness maintenance, migration and invasion of BC development[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Nowadays, VM has become a novel direction to guide BC therapy. Therefore, identifying key VM-related lncRNAs in BC is worthy of studying.\u003c/p\u003e \u003cp\u003eIn this research, we constructed a four-LncRNA prognostic model to predict the overall survival (OS) of BC patients. In addition, this predictive signature also evaluated biological enrichment, immune infiltration and drug sensitivity response and it was also validated by internal verification.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eEnrichment Analysis of VM-Related genes\u003c/h2\u003e \u003cp\u003eThe workflow of this study is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. We firstly obtained 19938 protein-coding genes from the BRCA project of the TCGA database and extracted 227 VRGs expression matrix from it. Then, 90 VMDEGs (upregulated genes: 44, downregulated genes: 46) were identified in breast tumor compared to normal breast tissue (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, C). KEGG pathway analyses revealed that these 90 genes were mainly enriched in the proteoglycans in cancer, HIF-1 signaling pathway, AGE-RAGE signaling pathway, PI3K-Akt signaling pathway, bladder cancer, relaxin signaling pathway, focal adhesion (FA), EGFR tyrosine kinase inhibitor resistance, Rap1 signaling pathway, fluid shear stress and atherosclerosis (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). GO analysis revealed the functional analysis landscape of these 90 genes in the following three aspects: (1) in the cellular components category, membrane microdomain, membrane raft and platelet alpha granule, etc. can be mainly detected (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). (2) in the biological process category, wound healing, response to steroid hormone, and regulation of peptidase activity, etc. were mostly enriched (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE). (3) in the molecular function category, signaling receptor activator activity, growth factor activity and cytokine receptor binding, etc., accounted for a large proportion (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eConstruction of the Prognostic VRLs Signature\u003c/h2\u003e \u003cp\u003eWe downloaded 13507 lncRNAs from the BRCA project of the TCGA database. Then, Spearmen correlation analysis was performed between the lncRNAs mentioned above and VMDEGs with a criteria of |R\u003csup\u003e2\u003c/sup\u003e|\u0026gt;0.4 and P\u0026thinsp;\u0026lt;\u0026thinsp;0.001. A total of 1320 VRLs were ultimately defined [see Supplementary Table \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e]. Then, we integrated survival information into the gene expression profiles to further identify potential prognostic VRLs. A total of 1320 VRLs were chosen for univariate Cox regression analysis, and 63 VRLs were identified to have strong relation with the prognosis of BC patients. Subsequently, these survival related 63 VRLs were adopted for implementing LASSO regression analysis. Increasing λ led to a decrease in the number of independent variables with coefficients close to 0 (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). The optimal λ value was determined when partial likelihood deviance was the lowest (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). Afterwards, 33 VRLs were acquired by univariate Cox and LASSO regression, and we applied bi-directional stepwise multivariate Cox regression to the remaining 33 VRLs with AIC criteria for the final screening of VRLs. Following the above selection, 4 VRLs (SEMA3B-AS1, MAPT-AS1, AL355512.1 and AP005717.2) were ascertained to further establish the prognostic signature [see Supplementary Table \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e]. Ctyoscape software[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] intuitively visualized the interactions between these 4 VRLs and its related mRNAs (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). SEMA3B-AS1 interacted strongly with following five genes: BPIFB1, EGFR, IGF2BP2, KDM4B and SLC9A3R1. AF005717.2 had tight relation with AURKA and FOXM1. AL355512.1 and MAPT-AS1 had only one correlated gene, SLC9A3R1 and KDM4B, respectively. According to multivariate cox regression analysis, filtered VRLs with hazard ratio (HR)\u0026thinsp;\u0026gt;\u0026thinsp;1 were considered as risk factors. The Sankey diagram (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE) revealed that MAPT-AS1 and SEMA3B-AS1 belonged to protect-type genes as well as AL355512.1 and AF005717.2 belonged to risk-type genes. We plotted heatmap of these 4 VRLs and clinicopathological variables in high-risk and low-risk groups to show the relationship between the signature and the clinical information (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eUtility of the VRGs Signature as an Independent Prognostic Indicator\u003c/h2\u003e \u003cp\u003eTo verify the effectiveness of the VRLs signature prediction, K-M curves were constructed (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). We found that patients in high-risk group had shorter overall survival time than those belonging to low-risk group. The risk score of the high-risk and low-risk groups are depicted in Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA. The mortality increased along with the increment of risk score (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). Forest plot for univariate Cox regression (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD) showed that age, T stage, N stage, M stage, Stage and risk score were closely associated with the OS of BC patients. Multivariate Cox regression forest plot (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE) revealed that only age and risk score could be regarded as independent predictors in BC. Satisfactory predictive ability of the signature were indicated by the ROC curve of risk score and the area under curve (AUC) at 1, 2 and 5-year survival time (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe predictive nomogram (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG) calculated an aggregate score on the basis of age, T stage, N stage, AJCC stage and risk score. Simultaneously, we performed calibration curve at 1, 2 and 5 years (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eH-J) to demonstrate the probability of the survival outcome of each patient. The calibration curve was regarded as the visualization of the results of the Hosmer-Lemeshow goodness-of-fit test and the survival prediction curves at 1, 2 and 5 years fluctuated above and below the survival actuality curve, confirming good synchronization between predictive and actual OS rate.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eInternal Validation of the Predictive Signature\u003c/h2\u003e \u003cp\u003eA total of 1023 BC patients were divided into two cohorts (n\u0026thinsp;=\u0026thinsp;513, n\u0026thinsp;=\u0026thinsp;510). The clinical characteristics of these two cohorts were shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. In the internal cohort1, we performed K-M curve (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA) to depict the relationship between high-risk and low-risk group. Then we generated ROC curve (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC) to validate the predictive performance. The same validation method performed above was used in the internal cohort2 and relative K-M and ROC curve were showed in Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB, D. As revealed in Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB, patients in the high-risk group had worse prognosis than patients in the low-risk group (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, p\u0026thinsp;=\u0026thinsp;2.8405e-05; Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB, p\u0026thinsp;=\u0026thinsp;2.4270e-03). In the cohort 1, the AUC reached 0.762 at 1 year, 0.713 at 3 years and 0.680 at 5 years. In the cohort 2, the AUC reached 0.738 at 1 year, 0.740 at 3 years and 0.680 at 5 years. The results denoted that patients divided into high-risk group had worse prognosis. Apart from this, the 5-years AUCs in the cohort1 and cohort2 were both lower than expectancy, which may reflect the lower accuracy of the signature\u0026rsquo;s 5-years prediction.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eFunctional Analysis in Different Risk Groups\u003c/h2\u003e \u003cp\u003eTo unveil the different biofunction and signal transduction pathways in high-risk and low-risk groups, the GSEA analysis was conducted. In GSEA, normalized enrichment score (NES) denoted the normalization of the dataset in terms of enrichment in different gene size, positive values of NES meant that genes could be enriched in high-risk group, and vice verse negative values of NES in the pathway indicated that genes could be enriched in low-risk group. The false discovery rate (FDR) could help us determine the rate of false positive finding of NES. The results (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) showed four enriched pathways in high-risk group, including ECM receptor interaction, complement and coagulation cascades, FA and MAPK signaling (Figs.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA-D). Interestingly, DNA repair pathways including DNA replication, homologous recombination, mismatch repair and RNA degradation as well as immune related pathway containing nature killer cell mediated cytoxicity and immunodeficiency were enriched in low-risk group (Figs.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003eEvaluation of TIME Cells Infiltration Characterization\u003c/h2\u003e \u003cp\u003eWe used ssGSEA to assess the landscape of 28 TIME cell infiltration in high-risk and low-risk groups. As shown in Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA, there were 18 TIME cells presented significant difference between high-risk group and low-risk group. The majority of these TIME cells were enriched in high-risk subgroup, including gammadelta (γδ) T cell, activated CD4 T cell, regulatory T cell (Treg), myeloid-derived suppressor cell (MDSC), activated CD8 T cell, monocyte, activated dendritic cell (DC), effector memory CD4 T cell, immature B cell, type2 T helper cell (Th2 cell), T follicular helper cell, natural killer (NK) T cell and macrophage. Only five immune cells incorporating mast cell, plasmacytoid dendritic cell (pDC), eosinophil, nature killer cell and central memory CD8 T cell expressed higher in low-risk group. Spearman correlation analysis (Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC) revealed that majority of TIME infiltrating cells had correlation with risk score.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe CIBERSORT algorithm, which was based on the approach of linear support vector regression, calculated the infiltration percentage of 22 TIME cells per patient. Not all TIME cells were differentially expressed in high-risk and low-risk groups due to the fact that there existed calculation principle discrepancy in ssGSEA and CIBERSORT. However, mutually differentially expressed TIME cells reflected good consistency, e.g., monocyte cells and macrophage cells all expressed higher in high-risk group, whereas, mast cells expressed higher in low-risk group (Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB). The heatmap (Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eD) illustrated that macrophage M0, macrophage M1, macrophage M2 and T cells CD4 memory resting cells accounting for greater fraction in TIME cells of BC patients. These findings indicated that immune biological process might function more active in the high-risk group.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003eTherapeutic Response of the Prognostic Signature\u003c/h2\u003e \u003cp\u003eTo explore potential effective drugs in two risk groups, we compared the estimated IC50 values of 198 drugs across groups. Among those, 3 candidates containing sorafenib, axitinib and AZD4547 which targeted upon vasculogenesis to combat tumorigenesis were chosen and the IC50 values of them were shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e. All drugs represented lower IC50 in high-risk group, which denoting favorable sensitivity to anti-vasculogenesis therapy.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"MATERIALS AND METHODS","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eAcquisition of Datasets\u003c/h2\u003e \u003cp\u003eThe fragments per kilobase million (FPKM) of BC transcriptome was downloaded from the BRCA project of the Cancer Genome Atlas (TCGA) data portal (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://portal.gdc.cancer.gov/\u003c/span\u003e\u003cspan address=\"https://portal.gdc.cancer.gov/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), and FPKM was subsequently converted to transcripts per million (TPM) value for downstream analysis. LncRNAs and protein-coding genes were classified by using the \u0026ldquo;TCGAbiolinks\u0026rdquo; R package[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. A total of 227 VM-related genes (VRGs) were extracted from previous articles[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] and GeneCards database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.genecards.org/\u003c/span\u003e\u003cspan address=\"https://www.genecards.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) [see Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e]. Genes filtered from GeneCards website with a standard of scores\u0026thinsp;\u0026gt;\u0026thinsp;15 were shortlisted. Clinicopathological information of 1023 patients encompassing complete survival data was obtained from the TCGA-BRCA cohort. The inclusion criteria were as followed: (1) histologically identified BRCA. (2) patients possessed RNA expression and related clinical data. (3) patients with a lack of survival status, tumor stage (T), lymph node stage (N), metastasis stage (M), clinical stage (Stage) and a survival time of less than 30 days were excluded.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eFunctional Enrichment Analysis of Differentially Expressed VM-Related Genes\u003c/h2\u003e \u003cp\u003eThe \u0026ldquo;limma\u0026rdquo; R package[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] was used to screen the VM-related differentially expressed genes (VMDEGs) between tumor tissues and normal tissues with a |log\u003csub\u003e2\u003c/sub\u003e (fold change)|\u0026gt;1 and adjusted p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 screening criteria. The details of VMDEGs were provided in [see Supplementary Table\u0026nbsp;2]. These VMDEGs, as input data, were included for further functional enrichment annotated on the basis of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) via the \u0026ldquo;clusterProfiler\u0026rdquo; R package[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eIdentification of Differentially Expressed VM-Related Genes\u003c/h2\u003e \u003cp\u003eThe Pearson correlation coefficients were calculated by using \u0026ldquo;limma\u0026rdquo; R package to define the correlation between VMDEGs and lncRNAs. The qualified VM-related differentially expressed lncRNAs (VRLs) were identified in accordance with the standard of correlation coefficient |R\u003csup\u003e2\u003c/sup\u003e|\u0026gt;0.4, and p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.001.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eConstruction of the VRLs Signature\u003c/h2\u003e \u003cp\u003eA total of 1320 VRLs were obtained after Pearson correlation analysis. Then, univariate Cox analysis based on \u0026ldquo;survival\u0026rdquo; R[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] package with p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 as the criteria was conducted for screening those genes significantly associated with OS. Moreover, we used \u0026ldquo;glmnet\u0026rdquo; R package[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] to perform the Least absolute and selection operator (LASSO) Cox regression to further filter prognostic candidates. Following above pipelines, we carried out multivariate Cox[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] regression analysis to confirm optimal VRLs and established VRLs-related prognostic signature. The Akaike information criterion (AIC) was employed to weigh the complexity of the estimated signature and the goodness of this signature to fit the data. After conducting several iterations, the AIC value of the signature kept decreasing and we chose the signature corresponding to the smallest AIC value (AIC\u0026thinsp;=\u0026thinsp;533.65) as the optimal signature. The formula for VM-related prognostic risk score for each patient was as followed:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$Risk score= \\sum _{i=1}^{n} ({Coef}_{i} \\times {x}_{i})$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(Coef\\)\u003c/span\u003e \u003c/span\u003e represents the coefficient value, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(x\\)\u003c/span\u003e\u003c/span\u003e represents the expression value of filtered VM-related lncRNAs. The risk score was calculated on the basis of the above formula: Risk score = (-0.1662*SEMA3B-AS1) \u0026ndash; (0.1286*MAPT-AS1) + (0.2702*AL355512.1) + (0.1833*AP005717.2). According to the median value of the risk score, patients were divided into high-risk and low-risk groups.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eValidation of the Predictive Signature\u003c/h2\u003e \u003cp\u003eIn order to identify the capability of the predictive model, the \u0026ldquo;caret\u0026rdquo; R package was adapted to randomly divide 1023 breast cancer patients into two cohorts (n\u0026thinsp;=\u0026thinsp;513, n\u0026thinsp;=\u0026thinsp;510). In the first internal cohort, we performed K-M curve via the \u0026ldquo;survminer\u0026rdquo; R package to depict the relationship between high-risk and low-risk group. Then we generated receiver operating characteristic (ROC) curves by using the \u0026ldquo;timeROC\u0026rdquo; R package to validate the predictive performance. The same validation method performed above was used in the second internal cohort.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eFunctional and Gene Set Enrichment Analysis\u003c/h2\u003e \u003cp\u003eGene set enrichment analysis (GSEA)[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] which could be carried out by \u0026ldquo;clusterProfiler\u0026rdquo; R package was applied for exploring enriched pathway. We downloaded KEGG and HALLMARK gene sets extracted from the Molecular Signatures Database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.gsea-msigdb.org/gsea/msigdb\u003c/span\u003e\u003cspan address=\"https://www.gsea-msigdb.org/gsea/msigdb\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) to excavate distinct biological characteristic between high-risk and low-risk groups. We considered pathways enriched with the criteria of absolute value of NES\u0026thinsp;\u0026gt;\u0026thinsp;1, FDR\u0026thinsp;\u0026lt;\u0026thinsp;0.25 and p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 is of significance.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eEstimation of Immune Cell Infiltration and Immune Microenvironment\u003c/h2\u003e \u003cp\u003eTo further explore the relationship between immune cells and risk score, we used \u0026ldquo;IOBR\u0026rdquo; R package[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] to conduct the process. The single sample gene set enrichment analysis (ssGSEA)[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e] was applied to quantify tumor immune microenvironment (TIME) cells infiltration of each patient between high-risk group and low-risk group. Information of TIME cells was acquired from the study of Charoentong[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. The CIBERSORT algorithm[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e] functioned via \u0026ldquo;IOBR\u0026rdquo; R package was used to compute the exact proportion of TIME cell infiltration for each patient.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eDrug Sensitivity Prediction\u003c/h2\u003e \u003cp\u003eThe \u0026ldquo;oncoPredict\u0026rdquo; R package[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e] was utilized to predict drug responses in BC patients. Details of the half-maximal inhibitory concentration (IC50) of each cancer cell lines to drugs could be found in Genomics of Drug Sensitivity in Cancer website (GDSC, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.cancerrxgene.org/\u003c/span\u003e\u003cspan address=\"https://www.cancerrxgene.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). We calculated three presentative anti-vasculogenesis drugs between high-risk and low-risk groups to verify drug sensitivity.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eAll data visualization and statistical analysis were performed in R software (version 4.2.2). Continuous variables were compared with log-rank test and Wilcoxon rank-sum test. The relationships between variables were estimated with Pearson test. Statistical significance was set at *p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, **p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ***p\u0026thinsp;\u0026lt;\u0026thinsp;0.001 and ****p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eBreast cancer is a heterogeneous disease involving genetic and environmental factors and affects women health around the world[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. In recent years, incremental evidence has been proposed that under the oxygen deficiency environment, hypoxia inducible factors can be activated to participate in the process of a large number of cellular pathways and genes transduction, subsequently cause VM formation, which can further promote proliferation, invasion and metastasis of plenty of solid tumors[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. In hepatocellular carcinoma, overexpression of lncRNA n339260 induces the formation of CSC-like phenotype, which further promotes VM[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. In colorectal cancer (CRC), lncRNA NORAD acted as a miR-495-3p sponge to regulate HIF-1α involves hypoxia-induced VM and chemoresistance[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. In gastric cancer, a positive feed-back loop between lncRNA PVT1 expression and STAT3/VEGFA axis can be found in VM formation to continuously provide oncogenic effects[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. Generally, it has been verified that VM can regulate tumor development via lncRNAs. However, studies on VRLs in breast cancer remain limited.\u003c/p\u003e \u003cp\u003eHerein, we firstly collected 227 VRGs and further screened out 90 differentially expressed VMDEGs in breast tumors. The KEGG and GO enrichment analyses of VMDEGs revealed that correlated genes were mainly enriched in FA, oxygen homeostasis regulator and cellular proliferation and metabolism related biological process. These pathways recapitulated that VM cells tended to convert to an embryonic-like phenotype, and subsequently presented aggressive, metastatic and unlimited proliferation characteristics according to previous reports[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. Then, a total of 1320 VRLs associated with 90 VMDEGs in BC (|R\u003csup\u003e2\u003c/sup\u003e|\u0026gt;0.4 and P\u0026thinsp;\u0026lt;\u0026thinsp;0.001) was filtered by Cox and LASSO regression analysis. We founded that SEMA3B-AS1, MAPT-AS1, AL355512.1 and AP005717.2 were correlated to the OS in BC patients. Of these four VRLs, clear evidence shows that SEMA3B-AS1 (SEAS1) acts as a tumor suppressor in TNBC through SEAS1/miR-3940-3p axis[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. In an \u003cem\u003ein vitro\u003c/em\u003e cellular assay for TNBC, SEAS1 is regarded as a suppressor of tumor-initiating cells (TIC). The inhibitory effect of SEAS1 on the malignancy relies on activated SEMA3B transcription through H3K4 methylation mediated by MLL4[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. In addition, MAPT-AS1 has also been identified to have close connection with tumorigenesis of BC. MAPT-AS1 silencing inhibits growth rate of BC cells in cell line derived xenografts (CDX) experiments and upregulates the sensitivity to paclitaxel[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Knockdown of MAPT-AS1 inversely increases the expression of Tau, a microtubule-associated protein, subsequently inclines to decrease the stabilization of microtubules which is modulated by paclitaxel[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eTo verify the prognostic accuracy of the 4 VRLs prognostic signature, we established a nomogram, K-M curves and ROC curves, etc. to intuitively show the predictive performance. We also combined different clinical parameters with the model and visualized it by forest plots. The results showed that risk score calculated by the signature could be regarded as independent indicator to predict the survival time of BC patients. In accordance with the magnitude value of risk score, BC patients were divided into high-risk and low-risk groups. GSEA analysis vividly demonstrated the staple pathogenic pathways of BC in these two groups and we perceived that FA, MAPK signaling, complement and coagulation cascades and ECM receptor interaction were mainly enriched in the high-risk group. FA is a liagtion structure when integrins attempt to across the membrane and establish connection with the ECM[\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. Integrin signaling can be activated when FA is anchored, which in turn promotes the proliferation, survival and stemness of tumor cells. In CRC, there has been proved that lncRNA ITGB8-AS1 can promote its growth and migration through integrin-mediated FA signaling [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. Additionally, there exists evidence that FA complexes inhibition impairs EMT and decreases cell motility in TNBC[\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. Moreover, previous literatures reported that focal adhesion kinase (FAK) can also affect cell migration, invasion and metastasis via corsstalking with Src and growth factor receptor signaling pathways[\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. FAK-Src-mediated phosphorylation can recruit and trigger the activation of MAPK/ERK cascade and subsequently promote tumor angiogenesis and progression[\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. The ECM, as an important architectural component of the TME, has close interaction with cancer cells[\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. During the development of tumorigenesis, cancer cells, cancer-associated fibroblasts (CAFs)[\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e] and immune-related cells regulated by transforming growth factor β (TGF-β)[\u003cspan additionalcitationids=\"CR48\" citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e] are prone to form a signaling loop, which facilitating the generation of ECM stiffness, and in return, the rigid ECM accelerates tumor growth rate[\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]. In conclusion, these pathways enriched in high-risk group had strong relationship with VM formation.\u003c/p\u003e \u003cp\u003eFurthermore, we applied ssGSEA and CIBERSORT algorithm on the signature to explore the immune landscape of high-risk and low-risk groups. Based on the signature stratification and boxplots visualized via ssGSEA and CIBERSORT, patients in high-risk group had higher infiltration of immunosuppressive cells including Tregs and macrophages, as well as lower infiltration of cytotoxic immune cells incorporating NK cells. In tumor microenvironment (TME) of BC, T effector proliferation could be inhibited via enhancing the Tregs capcity under the immunosuppressive environment exerted by CAFs[\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]. Recently, a single-cell RNA (scRNA) analysis research further identified that a subset of CAFs (CAF-S1) cellular clusters composed of ecm-myCAF and TGFβ-myCAF are capable of forming a reciprocal cross-talk tightly linked to immunotherapy resistance, and this cross-talk must be accomplished in TME with highly Tregs infiltration[\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e]. Apart from this, a number of research have discovered that tumor-associated macrophages (TAMs) are able to secrete abundant chemokines, cytokines, growth factors and inflammatory mediators as immune effector cells[\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e]. It has been proved that TAMs have the tendency to convert into polarized M2 phenotype under hypoxic region and subsequently induce vascular abnormalities, thereby promoting tumor growth and metastasis[\u003cspan additionalcitationids=\"CR55\" citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e]. In human TNBC cell lines, M2 macrophage secret growth factors VEGF to upregulate lncRNA PCAT6, and further stimulate endothelial tube formation to generate new vessels, facilitating TNBC cell proliferation, migration and invasion[\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e]. As depicted in Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA and \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB, M2 macrophage are highly expressed in high-risk group, which denotes worse prognosis. Whereas, for BC patients with low risk score, high infiltration of NK cells frequently implies the opposite prognosis to those in high-risk group. NK cells are currently consided as effector cells similar to cytotoxic T cells (CTLs), exerting natural cytotoxicity against primary tumor cells and metastasis by inhibiting distant metastasis[\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e]. To sum up, this 4 VRLs signature can help demonstrate the immune landscape of BC.\u003c/p\u003e \u003cp\u003eUltimately, drugs targeting on vasculogenesis in BC emerge promising results in recent years. A randomized, multicenter prospective trial of transarterial chemoembolization (TACE) plus sorafenib as compared with TACE alone in patients with hepatoelluilar carcinoma (TACTICS trial, NCT01217034, registered in 2nd October, 2010) has demonstrated that the combination of TACE plus sorafenib could markedly improve median progression-free survival (PFS) [25.2 vs 13.5 months; HR\u0026thinsp;=\u0026thinsp;0.59; 95%CI, 0.41 to 0.87; p\u0026thinsp;=\u0026thinsp;0.006][\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e]. In our research, drug sensitivity of three anti-vasculogenesis agents all showed lower IC50 in high-risk group, which suggeseted that patients with high VM risk score may benefit from anti-vasculogenesis therapy and this outcomes might lay the foundation for subsequent clinical trials in breast cancer.\u003c/p\u003e \u003cp\u003eThere are still some limitations in our research. We only used TCGA database for validation, and external public databases and real-world data validation are still to be needed to strengthen the evidence. Additionally, the mechanism of the VRLs in BC remains to be further verified by experiments.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, our study established a novel VM-related signature in BC based on four lncRNAs (SEMA3B-AS1, MAPT-AS1, AL355512.1 and AP005717.2). The satisfactory predictive capability of the signature was verified by TCGA database. We also found that in high-risk group, cytotoxic cells expressed lower, accompanying with higher expression of immunomonitor escaping cells, which indicated poorer prognosis. Ultimately, three anti-vasculogenesis drugs (sorafenib, axitinib and AZD4547) all showed lower IC50, indicating that patients in high-risk group may be more sensitive to anti-vasculogenesis drug therapy. These findings could ignite innovative BC therapeutic ideas in clinical practice.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eAIC: Akaike information criterion\u003c/p\u003e\n\u003cp\u003eAUC: Area under curve\u003c/p\u003e\n\u003cp\u003eBC: Breast cancer\u003c/p\u003e\n\u003cp\u003eCAF: Cancer-associated fibroblast\u003c/p\u003e\n\u003cp\u003eCDX: Cell line derived xenografts\u003c/p\u003e\n\u003cp\u003eCSC: Cancer stem cells\u003c/p\u003e\n\u003cp\u003eEC: Endothelial cell\u003c/p\u003e\n\u003cp\u003eECM: Extracellular matrix\u003c/p\u003e\n\u003cp\u003eEEndT: Epithelial-endothelial transformation\u003c/p\u003e\n\u003cp\u003eEGFR: Epidermal growth factor receptor\u003c/p\u003e\n\u003cp\u003eEMT: Epithelial-mesenchymal transformation\u003c/p\u003e\n\u003cp\u003eFDR: False discovery rate\u003c/p\u003e\n\u003cp\u003eFA: Focal adhesion\u003c/p\u003e\n\u003cp\u003eGO: Gene ontology\u003c/p\u003e\n\u003cp\u003eGSEA: Gene set enrichment analysis\u003c/p\u003e\n\u003cp\u003eHIF: Hypoxia-induced factor\u003c/p\u003e\n\u003cp\u003eHR: Hazard ratio\u003c/p\u003e\n\u003cp\u003eIC50: Half maximal inhibitory concentration\u003c/p\u003e\n\u003cp\u003eKEGG: Kyoto encyclopedia of genes and genomes\u003c/p\u003e\n\u003cp\u003eLASSO: Least absolute shrinkage and selection operator\u003c/p\u003e\n\u003cp\u003eLncRNA: Long non-coding RNA\u003c/p\u003e\n\u003cp\u003eMAPK: Mitogen-activate protein kinase\u003c/p\u003e\n\u003cp\u003eMMP-2: Matrix metalloproteinase 2\u003c/p\u003e\n\u003cp\u003eMMP-9: Matrix metalloproteinase 9\u003c/p\u003e\n\u003cp\u003eNcRNA: Non-coding RNA\u003c/p\u003e\n\u003cp\u003eNES: Normalized enrichment score\u003c/p\u003e\n\u003cp\u003ePI3K-Akt: Phosphatidyl-inositol 3 kinase-serine/threonine kinase\u003c/p\u003e\n\u003cp\u003eROC: Receiver operating characteristc\u003c/p\u003e\n\u003cp\u003eRNA-seq: RNA sequencing\u003c/p\u003e\n\u003cp\u003eSsGSEA: Single sample gene sample enrichment analysis\u003c/p\u003e\n\u003cp\u003eTGF-\u0026beta;: Transforming growth factor \u0026beta;\u003c/p\u003e\n\u003cp\u003eTIC: Tumor-initiating cell\u003c/p\u003e\n\u003cp\u003eTIME: Tumor immune microenvironment\u003c/p\u003e\n\u003cp\u003eTME: Tumor microenvironment\u003c/p\u003e\n\u003cp\u003eTNBC: Triple-negative breast cancer\u003c/p\u003e\n\u003cp\u003eVEGF: Vascular endothelial growth factor\u003c/p\u003e\n\u003cp\u003eVE-Cadherin: Vascular endothelial cadherin\u003c/p\u003e\n\u003cp\u003eVM: Vasculogenic mimicry\u003c/p\u003e\n\u003cp\u003eVMDEGs: Vasculogenic mimicry-related differentially expressed genes\u003c/p\u003e\n\u003cp\u003eVRGs: Vasculogenic mimicry-related genes\u003c/p\u003e\n\u003cp\u003eVRLs: Vasculogenic mimicry-related LncRNAs\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics Approval and Consent to Participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe research was designed and carried out abiding by the Ethics Committee at Xiangya Hospital of Central South University.\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\u003eAvailability of Data and Materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data that support the findings of our study were derived from the following resources available in the public domain: The Cancer Genome Atlas Program at https://portal.gdc.cancer.gov/, GeneCards Program at https://www.genecards.org/, Genomics of Drug Sensitivity in Cancer Program at https://www.cancerrxgene.org/ and the Molecular Signatures Database Program at https://www.gsea-msigdb.org/gsea/msigdb.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was funded by the following fundings: \u003c/p\u003e\n\u003cp\u003eChinese International Postdoctoral Exchange Fellowship Program (Talent-Introduction Program), Grant No. YJ20210330.\u003c/p\u003e\n\u003cp\u003eThe Key Program of the Natural Science Foundation of Hunan Province, China, Grant No. 2022SK2041.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor\u0026rsquo;s Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eYKC, JC and SMW designed and supervised the study; PZ contributed to data collection. YKC analyzed the datasets and interpreted the results. YKC and JC advised on data analysis and manuscript drafting. All authors critically reviewed the manuscript and consented to final publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAdditional Files\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSupplementary Table S1\u003c/p\u003e\n\u003cp\u003eFormat: Supplementary Table S1.xls\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Supplementary Table S2\u003c/p\u003e\n\u003cp\u003eFormat: Supplementary Table S2.xls\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Supplementary Table S3\u003c/p\u003e\n\u003cp\u003eFormat: Supplementary Table S3.xls\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Supplementary Table S4\u003c/p\u003e\n\u003cp\u003eFormat: Supplementary Table S4.xls\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA A Cancer J Clinicians. 2021;71:209\u0026ndash;49.\u003c/li\u003e\n\u003cli\u003eWei X, Chen Y, Jiang X, Peng M, Liu Y, Mo Y, et al. Mechanisms of vasculogenic mimicry in hypoxic tumor microenvironments. Mol Cancer. 2021;20:7.\u003c/li\u003e\n\u003cli\u003eLundgren K, Holm C, Landberg G. Common Molecular Mechanisms of Mammary Gland Development and Breast Cancer. Cell Mol Life Sci. 2007;64:3233\u0026ndash;47.\u003c/li\u003e\n\u003cli\u003eManiotis AJ, Folberg R, Hess A, Seftor EA, Gardner LM, Pe\u0026rsquo;er J, et al. Vascular channel formation by human melanoma cells in vivo and in vitro: vasculogenic mimicry. Am J Pathol. 1999;155:739\u0026ndash;52.\u003c/li\u003e\n\u003cli\u003eHu H, Ma T, Liu N, Hong H, Yu L, Lyu D, et al. Immunotherapy checkpoints in ovarian cancer vasculogenic mimicry: Tumor immune microenvironments, and drugs. International Immunopharmacology. 2022;111:109116.\u003c/li\u003e\n\u003cli\u003eZhang D, Sun B, Zhao X, Ma Y, Ji R, Gu Q, et al. Twist1 expression induced by sunitinib accelerates tumor cell vasculogenic mimicry by increasing the population of CD133+ cells in triple-negative breast cancer. Mol Cancer. 2014;13:207.\u003c/li\u003e\n\u003cli\u003eLiu TJ, Sun BC, Zhao XL, Zhao XM, Sun T, Gu Q, et al. CD133+ cells with cancer stem cell characteristics associates with vasculogenic mimicry in triple-negative breast cancer. Oncogene. 2013;32:544\u0026ndash;53.\u003c/li\u003e\n\u003cli\u003eG M-G, R G-B, Ea M-P, J G-Q, E A, L D. Vasculogenic Mimicry in Breast Cancer: Clinical Relevance and Drivers. Cells. 2021;10.\u003c/li\u003e\n\u003cli\u003eSeftor REB, Hess AR, Seftor EA, Kirschmann DA, Hardy KM, Margaryan NV, et al. Tumor Cell Vasculogenic Mimicry. Am J Pathol. 2012;181:1115\u0026ndash;25.\u003c/li\u003e\n\u003cli\u003eWinkle M, El-Daly SM, Fabbri M, Calin GA. Noncoding RNA therapeutics - challenges and potential solutions. Nat Rev Drug Discov. 2021;20:629\u0026ndash;51.\u003c/li\u003e\n\u003cli\u003eTao W, Sun W, Zhu H, Zhang J. Knockdown of long non-coding RNA TP73-AS1 suppresses triple negative breast cancer cell vasculogenic mimicry by targeting miR-490-3p/TWIST1 axis. Biochem Biophys Res Commun. 2018;504:629\u0026ndash;34.\u003c/li\u003e\n\u003cli\u003eLuo Q, Wang J, Zhao W, Peng Z, Liu X, Li B, et al. Vasculogenic mimicry in carcinogenesis and clinical applications. J Hematol Oncol. 2020;13:19.\u003c/li\u003e\n\u003cli\u003eLi X, Yang J, Ni R, Chen J, Zhou Y, Song H, et al. Hypoxia-induced lncRNA RBM5-AS1 promotes tumorigenesis via activating Wnt/\u0026beta;-catenin signaling in breast cancer. Cell Death Dis. 2022;13:95.\u003c/li\u003e\n\u003cli\u003eShannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: A Software Environment for Integrated Models of Biomolecular Interaction Networks. Genome Res. 2003;13:2498\u0026ndash;504.\u003c/li\u003e\n\u003cli\u003eColaprico A, Silva TC, Olsen C, Garofano L, Cava C, Garolini D, et al. TCGAbiolinks: an R/Bioconductor package for integrative analysis of TCGA data. Nucleic Acids Research. 2016;44:e71.\u003c/li\u003e\n\u003cli\u003eWang J, Xia W, Huang Y, Li H, Tang Y, Li Y, et al. A vasculogenic mimicry prognostic signature associated with immune signature in human gastric cancer. Front Immunol. 2022;13:1016612.\u003c/li\u003e\n\u003cli\u003eZhang C, Xiao J, Yuan T, He Y, Deng D, Xiao Z, et al. Molecular vasculogenic mimicry\u0026ndash;Related signatures predict clinical outcomes and therapeutic responses in bladder cancer: Results from real-world cohorts. Front Pharmacol. 2023;14:1163115.\u003c/li\u003e\n\u003cli\u003eRitchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43:e47.\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\u0026ndash;7.\u003c/li\u003e\n\u003cli\u003eTherneau TM, Grambsch PM. Modeling Survival Data: Extending the Cox Model. New York, NY: Springer; 2000.\u003c/li\u003e\n\u003cli\u003eFriedman JH, Hastie T, Tibshirani R. Regularization Paths for Generalized Linear Models via Coordinate Descent. Journal of Statistical Software. 2010;33:1\u0026ndash;22.\u003c/li\u003e\n\u003cli\u003eSubramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005;102:15545\u0026ndash;50.\u003c/li\u003e\n\u003cli\u003eZeng D, Ye Z, Shen R, Yu G, Wu J, Xiong Y, et al. IOBR: Multi-Omics Immuno-Oncology Biological Research to Decode Tumor Microenvironment and Signatures. Front Immunol. 2021;12:687975.\u003c/li\u003e\n\u003cli\u003eBarbie DA, Tamayo P, Boehm JS, Kim SY, Moody SE, Dunn IF, et al. Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature. 2009;462:108\u0026ndash;12.\u003c/li\u003e\n\u003cli\u003eCharoentong P, Finotello F, Angelova M, Mayer C, Efremova M, Rieder D, et al. Pan-cancer Immunogenomic Analyses Reveal Genotype-Immunophenotype Relationships and Predictors of Response to Checkpoint Blockade. Cell Rep. 2017;18:248\u0026ndash;62.\u003c/li\u003e\n\u003cli\u003eNewman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015;12:453\u0026ndash;7.\u003c/li\u003e\n\u003cli\u003eMaeser D, Gruener RF, Huang RS. oncoPredict: an R package for predicting in vivo or cancer patient drug response and biomarkers from cell line screening data. Brief Bioinform. 2021;22:bbab260.\u003c/li\u003e\n\u003cli\u003eK B, J K, Z Z, A H, Mh K, S M-K, et al. Breast cancer: Biology, biomarkers, and treatments. International immunopharmacology. 2020;84.\u003c/li\u003e\n\u003cli\u003eZhang L, Wu H, Zhang Y, Xiao X, Chu F, Zhang L. Induction of lncRNA NORAD accounts for hypoxia-induced chemoresistance and vasculogenic mimicry in colorectal cancer by sponging the miR-495-3p/ hypoxia-inducible factor-1\u0026alpha; (HIF-1\u0026alpha;). Bioengineered. 13:950\u0026ndash;62.\u003c/li\u003e\n\u003cli\u003eZhao X, Sun B, Liu T, Shao B, Sun R, Zhu D, et al. Long noncoding RNA n339260 promotes vasculogenic mimicry and cancer stem cell development in hepatocellular carcinoma. Cancer Sci. 2018;109:3197\u0026ndash;208.\u003c/li\u003e\n\u003cli\u003eZhao J, Wu J, Qin Y, Zhang W, Huang G, Qin L. LncRNA PVT1 induces aggressive vasculogenic mimicry formation through activating the STAT3/Slug axis and epithelial-to-mesenchymal transition in gastric cancer. Cell Oncol (Dordr). 2020;43:863\u0026ndash;76.\u003c/li\u003e\n\u003cli\u003eDelgado-Bellido D, Serrano-Saenz S, Fern\u0026aacute;ndez-Cort\u0026eacute;s M, Oliver FJ. Vasculogenic mimicry signaling revisited: focus on non-vascular VE-cadherin. Mol Cancer. 2017;16:65.\u003c/li\u003e\n\u003cli\u003eHu J, Huang H, Xi Z, Ma S, Ming J, Dong F, et al. LncRNA SEMA3B-AS1 inhibits breast cancer progression by targeting miR-3940/KLLN axis. Cell Death Dis. 2022;13:800.\u003c/li\u003e\n\u003cli\u003eChen D, Chen Z, Wang Z, Hong C, Wang Q, Yang P, et al. LncRNA SEMA3B‐AS1 suppresses the tumor‐initiating characteristics of triple negative breast cancer via engaging in MLL4‐mediated H3K4 trimethylation. Molecular Carcinogenesis. 2023;:mc.23658.\u003c/li\u003e\n\u003cli\u003ePan Y, Pan Y, Cheng Y, Yang F, Yao Z, Wang O. Knockdown of LncRNA MAPT-AS1 inhibites proliferation and migration and sensitizes cancer cells to paclitaxel by regulating MAPT expression in ER-negative breast cancers. Cell Biosci. 2018;8:7.\u003c/li\u003e\n\u003cli\u003eSimone R, Javad F, Emmett W, Wilkins OG, Louren\u0026ccedil;o-Almeida F, Barahona-Torres N, et al. MIR-NATs repress MAPT translation and aid proteostasis in neurodegeneration. Nature. 2021;594:117\u0026ndash;23.\u003c/li\u003e\n\u003cli\u003eSafinya CR, Chung PJ, Song C, Li Y, Ewert KK, Choi MC. The effect of multivalent cations and Tau on paclitaxel-stabilized microtubule assembly, disassembly, and structure. Adv Colloid Interface Sci. 2016;232:9\u0026ndash;16.\u003c/li\u003e\n\u003cli\u003eDesgrosellier JS, Cheresh DA. Integrins in cancer: biological implications and therapeutic opportunities. Nat Rev Cancer. 2010;10:9\u0026ndash;22.\u003c/li\u003e\n\u003cli\u003eLin X, Zhuang S, Chen X, Du J, Zhong L, Ding J, et al. lncRNA ITGB8-AS1 functions as a ceRNA to promote colorectal cancer growth and migration through integrin-mediated focal adhesion signaling. Mol Ther. 2022;30:688\u0026ndash;702.\u003c/li\u003e\n\u003cli\u003eThapa N, Tan X, Choi S, Wise T, Anderson RA. PIPKI\u0026gamma; and talin couple phosphoinositide and adhesion signaling to control the epithelial to mesenchymal transition. Oncogene. 2017;36:899\u0026ndash;911.\u003c/li\u003e\n\u003cli\u003eZhang Y, Sun L, Li H, Ai L, Ma Q, Qiao X, et al. Binding blockade between TLN1 and integrin \u0026beta;1 represses triple-negative breast cancer. Elife. 2022;11:e68481.\u003c/li\u003e\n\u003cli\u003eMitra SK, Hanson DA, Schlaepfer DD. Focal adhesion kinase: in command and control of cell motility. Nat Rev Mol Cell Biol. 2005;6:56\u0026ndash;68.\u003c/li\u003e\n\u003cli\u003eWu S, Chen M, Huang J, Zhang F, Lv Z, Jia Y, et al. ORAI2 Promotes Gastric Cancer Tumorigenicity and Metastasis through PI3K/Akt Signaling and MAPK-Dependent Focal Adhesion Disassembly. Cancer Research. 2021;81:986\u0026ndash;1000.\u003c/li\u003e\n\u003cli\u003eZhou F, Sun J, Ye L, Jiang T, Li W, Su C, et al. Fibronectin promotes tumor angiogenesis and progression of non-small-cell lung cancer by elevating WISP3 expression via FAK/MAPK/ HIF-1\u0026alpha; axis and activating wnt signaling pathway. Exp Hematol Oncol. 2023;12:61.\u003c/li\u003e\n\u003cli\u003eHuang J, Zhang L, Wan D, Zhou L, Zheng S, Lin S, et al. Extracellular matrix and its therapeutic potential for cancer treatment. Signal Transduct Target Ther. 2021;6:153.\u003c/li\u003e\n\u003cli\u003eLi C, Teixeira AF, Zhu H-J, ten Dijke P. Cancer associated-fibroblast-derived exosomes in cancer progression. Mol Cancer. 2021;20:154.\u003c/li\u003e\n\u003cli\u003eTerra M, Oberkampf M, Fayolle C, Rosenbaum P, Guillerey C, Dadaglio G, et al. Tumor-Derived TGF\u0026beta; Alters the Ability of Plasmacytoid Dendritic Cells to Respond to Innate Immune Signaling. Cancer Res. 2018;78:3014\u0026ndash;26.\u003c/li\u003e\n\u003cli\u003eBatlle E, Massagu\u0026eacute; J. Transforming Grown Factor-\u0026beta; Signaling in Immunity and Cancer. Immunity. 2019;50:924\u0026ndash;40.\u003c/li\u003e\n\u003cli\u003eDonzelli S, Milano E, Pruszko M, Sacconi A, Masciarelli S, Iosue I, et al. Expression of ID4 protein in breast cancer cells induces reprogramming of tumour-associated macrophages. Breast Cancer Res. 2018;20:59.\u003c/li\u003e\n\u003cli\u003eNajafi M, Farhood B, Mortezaee K. Extracellular matrix (ECM) stiffness and degradation as cancer drivers. J Cell Biochem. 2019;120:2782\u0026ndash;90.\u003c/li\u003e\n\u003cli\u003eCosta A, Kieffer Y, Scholer-Dahirel A, Pelon F, Bourachot B, Cardon M, et al. Fibroblast Heterogeneity and Immunosuppressive Environment in Human Breast Cancer. Cancer Cell. 2018;33:463-479.e10.\u003c/li\u003e\n\u003cli\u003eKieffer Y, Hocine HR, Gentric G, Pelon F, Bernard C, Bourachot B, et al. Single-Cell Analysis Reveals Fibroblast Clusters Linked to Immunotherapy Resistance in Cancer. Cancer Discovery. 2020;10:1330\u0026ndash;51.\u003c/li\u003e\n\u003cli\u003eSchwertfeger KL, Cowman MK, Telmer PG, Turley EA, McCarthy JB. Hyaluronan, Inflammation, and Breast Cancer Progression. Front Immunol. 2015;6.\u003c/li\u003e\n\u003cli\u003eHan S, Wang W, Wang S, Yang T, Zhang G, Wang D, et al. Tumor microenvironment remodeling and tumor therapy based on M2-like tumor associated macrophage-targeting nano-complexes. Theranostics. 2021;11:2892\u0026ndash;916.\u003c/li\u003e\n\u003cli\u003eDong F, Ruan S, Wang J, Xia Y, Le K, Xiao X, et al. M2 macrophage-induced lncRNA PCAT6 facilitates tumorigenesis and angiogenesis of triple-negative breast cancer through modulation of VEGFR2. Cell Death Dis. 2020;11:728.\u003c/li\u003e\n\u003cli\u003eChen D, Zhang X, Li Z, Zhu B. Metabolic regulatory crosstalk between tumor microenvironment and tumor-associated macrophages. Theranostics. 2021;11:1016\u0026ndash;30.\u003c/li\u003e\n\u003cli\u003eWu S-Y, Fu T, Jiang Y-Z, Shao Z-M. Natural killer cells in cancer biology and therapy. Mol Cancer. 2020;19:120.\u003c/li\u003e\n\u003cli\u003eKudo M, Ueshima K, Ikeda M, Torimura T, Tanabe N, Aikata H, et al. Randomised, multicentre prospective trial of transarterial chemoembolisation (TACE) plus sorafenib as compared with TACE alone in patients with hepatocellular carcinoma: TACTICS trial. Gut. 2020;69:1492\u0026ndash;501.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTable 1 | The clinical characteristics of patients in validation cohorts\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eVariables\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"65.21739130434783%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eValidation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.333333333333336%\" valign=\"top\"\u003e\n \u003cp\u003eCohort 1 (n=513)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"46.666666666666664%\" valign=\"top\"\u003e\n \u003cp\u003eCohort 2 (n=510)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"3\"\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026lt;60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e283 (55%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e268 (53%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026ge;60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e230 (45%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e242 (47%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"3\" valign=\"top\"\u003e\n \u003cp\u003eT\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eT1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e128 (25%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e143 (28%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eT2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e307 (60%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e285 (56%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eT3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e63 (12%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e66 (13%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eT4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e15 (3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e16 (3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"3\" valign=\"top\"\u003e\n \u003cp\u003eN\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eN0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e249 (49%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e240 (47%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eN1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e172 (33%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e180 (35%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eN2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e56 (11%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e54 (11%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eN3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e36 (7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e36 (7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"3\" valign=\"top\"\u003e\n \u003cp\u003eM\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eM1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e506 (99%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e499 (98%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eM2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e7 (1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e11 (2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"3\" valign=\"top\"\u003e\n \u003cp\u003eStage\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eStage I\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e85 (17%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e96 (19%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eStage II\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e303 (59%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e287 (56%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eStage III\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e118 (23%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e116 (23%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eStage IV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003e7 (1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.434782608695652%\" valign=\"top\"\u003e\n \u003cp\u003e11 (2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable 2 | Enriched gene sets in high-risk group\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"529\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"49.81132075471698%\" valign=\"top\"\u003e\n \u003cp\u003eGene set\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.056603773584905%\" valign=\"top\"\u003e\n \u003cp\u003eES\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.754716981132075%\" valign=\"top\"\u003e\n \u003cp\u003eNES\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.037735849056602%\" valign=\"top\"\u003e\n \u003cp\u003eP-value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.339622641509434%\" valign=\"top\"\u003e\n \u003cp\u003eFDR\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"49.81132075471698%\" valign=\"top\"\u003e\n \u003cp\u003eECM receptor interaction\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.056603773584905%\" valign=\"top\"\u003e\n \u003cp\u003e0.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.754716981132075%\" valign=\"top\"\u003e\n \u003cp\u003e1.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.037735849056602%\" valign=\"top\"\u003e\n \u003cp\u003e1.80e-05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.339622641509434%\" valign=\"top\"\u003e\n \u003cp\u003e4.11e-04\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"49.81132075471698%\" valign=\"top\"\u003e\n \u003cp\u003eComplement and coagulation cascades\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.056603773584905%\" valign=\"top\"\u003e\n \u003cp\u003e0.59\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.754716981132075%\" valign=\"top\"\u003e\n \u003cp\u003e1.89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.037735849056602%\" valign=\"top\"\u003e\n \u003cp\u003e1.01e-04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.339622641509434%\" valign=\"top\"\u003e\n \u003cp\u003e9.26e-04\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"49.81132075471698%\" valign=\"top\"\u003e\n \u003cp\u003eFocal adhesion\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.056603773584905%\" valign=\"top\"\u003e\n \u003cp\u003e0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.754716981132075%\" valign=\"top\"\u003e\n \u003cp\u003e1.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.037735849056602%\" valign=\"top\"\u003e\n \u003cp\u003e1.04e-06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.339622641509434%\" valign=\"top\"\u003e\n \u003cp\u003e3.57e-05\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"49.81132075471698%\" valign=\"top\"\u003e\n \u003cp\u003eMAPK signaling\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.056603773584905%\" valign=\"top\"\u003e\n \u003cp\u003e0.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.754716981132075%\" valign=\"top\"\u003e\n \u003cp\u003e1.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.037735849056602%\" valign=\"top\"\u003e\n \u003cp\u003e1.61e-03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.339622641509434%\" valign=\"top\"\u003e\n \u003cp\u003e9.65e-03\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cem\u003eES, enrichment score; NES, normalized enrichment score; FDR, false discovery rate; ECM, extracellular matrix\u003c/em\u003e\u003c/p\u003e\n"}],"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":"vasculogenic mimicry, breast cancer, lncRNA, immune microenvironment","lastPublishedDoi":"10.21203/rs.3.rs-4150302/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4150302/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eVasculogenic mimicry (VM) is linked closely to the tumorigenesis. However, VM-related lncRNAs (VRLs) involved in the mediation of breast cancer (BC) are still unknown. This research aimed to identify a prognostic signature of VRLs in BC and excavate its potential biological function.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe obtained RNA-seq and relevant clinical data from The Cancer Genome Atlas database. Then, Cox and the LASSO regression were utilized to construct a multigene signature. The Kaplan-Meier and ROC curves were plotted to evaluate the efficacy of the model. GO and KEGG pathway were performed for patients in high-risk and low-risk groups. SsGSEA and CIBERSORT algorithm were used to observe the relationship in high-risk and low-risk groups and immune cells. Furthermore, we analysed the inhibitory concentration (IC50) values of three representative anti-vasculogenesis drugs of BC in high-risk and low-risk groups to verify drug sensitivity.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eA VRL-based prognostic signature composed by SEMA3B-AS1, MAPT-AS1, AL355512.1 and AP005717.2 was constructed. According to the risk score calculated by this signature, BC patients were divided into high-risk and low-risk groups. Patients in the high-risk group inclined to have a worse prognosis. SsGSEA and CIBERSORT showed that the majority of immune cells e.g., macrophage and CD4 T cell expressed notably higher in high-risk group (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). In addition, we analysed the IC50 values of sorafenib, axitinib and AZD4547 in high-risk and low-risk groups, and all these drugs demonstrated favorable sensitivity to high-risk group which indicated that patients in high-risk group might benefit from anti-vasculogenesis drugs.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eBased on bioinformatic analysis, we established a VM-related gene signature to predict the overall survival of BC patients. Apart from this, we characterized the relationship in the signature, immune microenvironment and correlated drugs which may ignite a novel idea of BC therapy.\u003c/p\u003e","manuscriptTitle":"Vasculogenic Mimicry Related Long Noncoding RNA Signature Reveals New Therapy Strategy in Breast Cancer","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-03-29 18:40:58","doi":"10.21203/rs.3.rs-4150302/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":"06b018b2-b548-42d6-bf7c-bc269b023490","owner":[],"postedDate":"March 29th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-03-29T18:41:01+00:00","versionOfRecord":[],"versionCreatedAt":"2024-03-29 18:40:58","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4150302","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4150302","identity":"rs-4150302","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}