1. Introduction

Breast cancer is the most common cancer in women worldwide, and is a leading cause of cancer mortality in women [1]. Treatment of breast cancer is largely based on the expression of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor 2 (HER2/ERBB2). The hormone receptors ER/PR-positive patients account for around 70% of breast cancer patients, who are treated with hormonal therapy [2]. HER2 expression, which is observed in approximately 15–25% of breast cancers, is mainly relevant in the choice of anti-HER2 targeted therapy. Triple-negative breast cancer (TNBC), defined by the absence of all three markers, is the most aggressive subtype of breast cancer due to the lack of targeted therapeutic options. About 15% of breast cancers are TNBC [2]. Chemotherapy and immunotherapy are currently the most commonly used systematic therapies in TNBC patients. However, tumor recurrence, metastasis and treatment resistance remain the leading causes of mortality, particularly in TNBC patients.

Substantial advances have been made in understanding the heterogeneity of breast cancer. Cancer stem cells (CSCs) are a subpopulation of cancer cells that can self-renew and differentiate into different cell lineages, thereby contributing to tumor relapse and treatment resistance. Self-renewal refers to a process that occurs during cell division, whereby at least one daughter cell maintains stem cell-like phenotypes, including long-term proliferative potential [3]. The CSC self-renewal capacity can be assessed by serial passaging of cancer cells in vivo [4]. CSCs have been discovered in many types of cancer, including breast cancer, glioblastoma, colon cancer and prostate cancer [5]. One of the most profound implications of CSCs is their contribution to tumor heterogeneity. Due to their clinical significance, several biomarkers have been identified to characterize CSCs, correlating with diagnosis, therapy sensitivity and prognosis. Dormant CSCs are thought to be more resistant to current therapies [6]. Importantly, CSCs exhibit high plasticity, modifying their phenotypic and functional appearance under different conditions [7]. By understanding the specific markers and signaling pathways associated with CSCs, novel treatments could potentially inhibit their self-renewal or induce differentiation, thus eradicating the seeds of the tumor. Here, we discussed the heterogeneity and plasticity of CSCs in breast cancer and novel treatment strategies to target CSCs.

2. Breast Cancer Stem Cells (BCSCs)

BCSCs are a small proportion of breast cancer cells that are capable of self-renewal and generate differentiated cancer cells [8]. The origin of BCSCs is still unclear. One of the theories is that they derive from mammary stem cells (MaSC), as they share similar markers and a MaSC-enriched population has been observed in preneoplastic tissue of mammary tumor-prone mice [6]. MaSCs possess the ability to differentiate into various mammary cell lineages, contributing to the gland's functional architecture during puberty, pregnancy, and lactation. MaSCs lacks the expression of hormone receptors in mice and they might become BCSCs after gaining the mutation [9]. Further defining the regulation mechanisms of MaSC might help us understanding the origin of BCSCs.

BCSCs have been extensively studied due to their malignant behavior such as tumorigenesis, therapeutic resistance and high metastatic potential. The classic method to identify CSCs involves cell sorting by flow cytometry to isolate cell populations based on specific surface markers, followed by inoculation of cells at limited dilutions in serial xenotransplantation assay to assess the tumor-initiating cell frequency [3]. A number of alternative approaches include three-dimensional (3D) culture systems and single-cell RNA sequencing, etc [3]. The main characteristics of BCSCs are their high heterogeneity and plasticity, which contribute remarkably to their malignancy and pose a challenge for efficient cancer therapeutics. Heterogeneity refers to their subpopulations that express different cellular markers. Plasticity refers to their ability to switch between different states regulated by the tumor microenvironment and pro-tumor signaling pathways.

The mammary gland is a highly regenerative organ that goes through significant changes during the menstrual cycle, pregnancy, lactation, and involution, and stem cells play important roles during these process. How these changes affect BCSCs and breast oncogenesis is still controversial. A previous study showed that pregnancy in the adult mouse does not alter the proportion of mammary epithelial stem/progenitor cells [10]. Epidemiological studies have demonstrated that an early, full-term pregnancy at young age may reduce the lifetime risk of breast cancer [11]. However, pregnancy generally offers protection against the development of ER/PR⁺ tumors, while the breast cancers that emerge shortly after pregnancy typically do not express these hormone receptors [12]. During pregnancy, the breast undergoes significant morphological and functional changes, primarily driven by a surge in hormones such as estrogen and progesterone. These hormonal shifts stimulate mammary gland development and differentiation, leading to the proliferation of epithelial cells and increased stromal support. The microenvironment with growth factors and cytokines promotes the survival and expansion of BCSCs [13]. Furthermore, another study revealed that parity reduced mammary repopulating activity but did not affect mammary stem cells defined as CD24⁺ CD29/CD49f^hi in mice [14].

3. BCSC Heterogeneity and Plasticity

A number of subpopulations within breast cancer have been identified to possess high tumor-initiating potential. The most commonly used markers to identify BCSCs include CD24^– CD44⁺, ALDH, and CD49f [15,16].

The initial BCSC population to be identified was that of CD24^–/low CD44⁺ cells [4]. These cancer cells demonstrate a substantial capacity for tumor initiation. As few as 100 breast cancer cells with EpCAM⁺, high levels of CD44, and low levels of CD24 expression were able to efficiently form tumors when inoculated in immune-deficient mice [4]. The tumorigenic subpopulation could be serially passaged. Each time, these cells generated new tumors containing additional CD44⁺CD24^–/low Lineage^– tumorigenic cells as well as the phenotypically diverse mixed populations of non-tumorigenic cells present in the initial tumor [4]. CD44 is a cell-surface glycoprotein that binds hyaluronic acid (HA) and many other extracellular matrix components [17]. CD44 is involved in cell-cell interactions, cell adhesion and migration. Alternative splicing of CD44 could determine breast cancer stem cell state [18]. CD44 standard splice isoform (CD44s) is the predominant isoform expressed in BCSCs and positively associates with the CSC gene signatures. In contrast, the CD44 variant splice isoforms (CD44v) exhibit an inverse association with the CSC signatures [18]. Mounting evidence have suggested that CD44 functions not only as CSC markers but also as critical regulators of cancer stemness, including self-renewal, tumor initiation, and metastasis [17].

Another frequently used BCSCs marker is aldehyde dehydrogenase 1 (ALDH1). ALDH1 is a member of the aldehyde dehydrogenase protein family, which catalyzes the oxidation of intracellular aldehydes. The measurement of ALDH1 activity via the Aldefluor assay is a frequently employed method for the analysis of BCSCs in both cell lines and animal models. ALDH1 plays a role in the early differentiation of BCSCs through oxidizing retinol to retinoic acid [16]. The activity of ALDH1A1 in BCSCs has been demonstrated to remodel myeloid-derived suppressor cells, thereby promoting breast cancer progression [19].

CD49f, also known as integrin-α6, is a further BCSC marker, as well as a common CSC marker [20]. It forms homodimers with β4-integrin, thereby facilitating binding to laminin and enhancing epithelial cell adhesion to the extracellular matrix [21]. CSCs, which are enriched in the CD24⁺ CD29⁺/CD49f⁺ cell population from a Brca1-mutant mouse model, exhibited a markedly elevated migration capacity relative to CD24^– CD29^–/CD49f^– cells [22]. The dynamics of the CD49f⁺ tumor-initiating population is associated with taxane resistance in TNBC [23].

Notably, CD24^–/low CD44⁺ mesenchymal-like BCSCs are primarily quiescent and localized at the tumor invasive front, whereas ALDH⁺ epithelial-like BCSCs, are proliferative, and are located more centrally [24]. Although breast cancer cells with both CD24^–/low CD44⁺ and ALDH⁺ expression represent a rare population, they have been observed to exhibit the greatest tumorigenic and invasive capacity [24]. Importantly, a recent study has demonstrated that ALDH1A3 is the key regulator that determines the balance of ALDH⁺ and CD24^–CD44⁺ breast cancer stem cells [25]. While ALDH1A3 increases the number of ALDH⁺ breast cancer cells, it inversely suppresses the CD24^– CD44⁺ population by retinoic acid signaling. It can therefore be concluded that BCSCs constitute a heterogeneous population, which is regulated by a number of different factors.

The abundance of BCSCs in different breast cancer subtypes has also been explored. The tumors enriched with CD44⁺ CD24^– or ALDH⁺ subpopulation were more correlated with basal-like and HER2-positive breast cancers [26,27]. In line with these results, BCSCs were rare in ER⁺ PR⁺ luminal subtype breast cancers [28]. Moreover, within a tumor, CD44⁺ CD24⁻ cells showed lower ER and PR expression compared to CD44⁻ CD24⁺ cells [28]. These results may also explain the poor prognosis of TNBC and HER2-positive patients.

The causes of BCSC heterogeneity is complicated and not yet fully elucidated. One theory is the occurrence of series of mutations in differentiated somatic cells. A key aspect of this theory is the concept of de-differentiation of differentiated cell, allowing them to be reverted to a more stem-like state under certain conditions [29]. This plasticity may underlie the heterogeneity observed in BCSCs, as different differentiated cells may respond uniquely to mutagenic stress, leading to diverse tumor phenotypes and behaviors. In contrast, the second theory proposes the occurrence of mutations in MaSC or progenitors [30]. This theory suggests that BCSCs arise from mutations occurring in the normal mammary stem cell compartment or its immediate progenitors. These mutations can lead to the dysregulation of key pathways involved in stem cell maintenance and differentiation, resulting in the heterogeneity and plasticity of BCSCs. However, sequencing analysis revealed that the majority of somatic mutations are shared between CSCs and bulk tumors [31]. Therefore, future efforts are still needed to deeply understand the causes of BCSC heterogeneity and plasticity.

4. Extrinsic Factors Regulating BCSCs Heterogeneity and Plasticity

The heterogeneity of BCSCs is typically attributed to their inherent plasticity. This plasticity refers to their capacity to transition between stem-like and differentiated states [32]. Multiple mechanisms have been identified as regulators of BCSCs plasticity and heterogeneity. Here we discuss how the tumor microenvironment components, hypoxia, tumor-resident microbiota, and tumor cell intrinsic signaling pathways influence the heterogeneity and plasticity of BCSCs.

4.1 Tumor ecosystem

Breast cancer consists of multiple cell types, including a large variety of immune and other host cells, which create the tumor microenvironment (TME). The crosstalk between all these cellular and structural components is essential for breast cancer growth and progression by accumulating phenotypic and genetic heterogeneity. The tumor microenvironment has been identified as a driving force behind BCSC heterogeneity and plasticity, with interactions between immune cells, stromal cells, and the extracellular matrix influencing BCSC expansion. The immune cells in the TME regulate BCSC plasticity through the establishment of positive feedback loops [33]. Here, we summarize ways in which the various cell types that comprise the TME, including macrophages, myeloid-derived suppressor cells, T cells and cancer-associated fibroblasts, contribute to the BCSC stemness and plasticity. The mechanisms are summarized in Figure 1 (created with BioGDP.com [34]).

Figure 1. Mechanisms of BCSC plasticity regulated by the tumor ecosystem. The plasticity and heterogeneity of BCSCs are influenced by the immune cells, fibroblasts, hypoxia, and microbiota present in the tumor niche. This contributes to treatment resistance and metastasis. TAMs: Tumor-associated macrophages; MDSC: myeloid-derived suppressor cells; CAFs: cancer-associated fibroblasts; ETBF: enterotoxigenic bacteroides fragilis. Created with BioGDP.com.

4.1.1 Immune cells

Macrophage

Tumor-associated macrophages (TAMs) constitute the most abundant immune population in the breast tumor microenvironment. TAMs have been demonstrated to facilitate cancer cell proliferation and metastasis, while also exerting immunosuppressive effects on the adaptive immune cells that populate the TME. Macrophages that have been activated are typically classified into two categories: M1-like macrophages, which are primarily involved in pro-inflammatory responses, and M2-like macrophages, which are primarily involved in anti-inflammatory responses.

Recent studies have demonstrated that TAMs promote epithelial-mesenchymal transition (EMT) and the cancer stem cell properties in breast cancer via the CCL2/AKT/β-catenin signaling [35,36]. In addition, Wang et al. recently reported that Rab13, a small GTPase, is highly expressed in BCSCs. Rab13 was found to control the membrane translocation of C-X-C chemokine receptor type 1/2 (CXCR1/2), allowing tumor cells to interact with TAMs and cancer-associated fibroblasts, and establish a supportive BCSC niche [37]. Furthermore, macrophage-induced ERK-TGF-β1 signaling in MCF7 breast cancer cells result in cancer stem cell plasticity and EMT which are completely reversible [38]. Additionally, M2-like TAMs contribute to increase cancer cell stemness in TNBC cells via secreting VEGFA, which were evaluated with stem cell markers, including CD24, CD44, OCT-4, Nanog and SOX-2. Reciprocally, elevated VEGFA expression by TAM-educated TNBC cells acts as a regulator of macrophage polarization [39]. In light of these findings, it can be concluded that TAMs play a pivotal role in regulating BCSC stemness and plasticity.

Myeloid-derived suppressor cell(MDSC)

Myeloid-derived suppressor cells (MDSCs) are activated neutrophils and monocytes with potent immunosuppressive activity. Two distinct subtypes of MDSCs have been identified: monocytic myeloid-derived suppressor cells (M-MDSCs) and granulocytic polymorphonuclear MDSCs (PMN-MDSCs). M-MDSCs show a stronger immunosuppressive effect [33]. Nonetheless, PMN-MDSCs play an important role in regulating tumor featured immune responses [40].

MDSC contribute to the development of stem-like phenotypes in breast cancer cells, which in turn facilitate immune suppression and tumor recurrence. For example, MDSCs promoted tumor growth by enhancing sphere formation and increasing the percentage of ALDH⁺ BCSCs in breast cancer as well as by suppressing T-cell activation. These effects were mediated by the cross-talk between the IL6-STAT3 and NOTCH pathways in cancer cells with MDSC [41]. MDSC was found to be correlated with the presence of BCSCs in breast cancer samples and was also identified as an independent predictor of poor survival outcomes [41]. Additionally, it was demonstrated that PMN-MDSCs, which are modulated by CCL20 from cancer cells, can increase ALDH⁺ BCSCs and stemness genes (eg. KLF4, NANOG, SOX9, etc.) via the CXCL2-CXCR2 pathway in breast cancer [42].

T cell

Although T cells are known for their tumor-killing ability, recent studies reveal that ineffective immune responses not only fail to clear a tumor, but also promote stemness and tumor-forming capacity in cancer cells. Interferon gamma (IFNγ) produced by activated T cells following immunotherapy directly converts non-CSCs to CSCs [43]. IFNγ enhances several CSC phenotypes, including resistance to chemo- and radiotherapy, as well as metastasis formation [43]. Cognate interaction with nonlytic CD8⁺T cells increased the proportion of CD44⁺CD24^– BCSCs in a cell-to-cell contact- or proximity-dependent manner [44].

4.1.2 Cancer-Associated Fibroblast (CAF)

CAFs play a pivotal role in the reactive stroma within the TME, as they not only interact with cancer cells extensively via secreted cytokines or cell–cell adhesion but also indirectly influence cancer cells via remodeling of the extracellular matrix and immune cell infiltration [45]. For instance, primary CAFs activated by cocultured breast cancer cells produce higher levels of chemokine CCL2, which induces the CSC phenotype in breast cancer [46]. Increased CCL2 expression in activated fibroblasts required STAT3 activation by diverse breast cancer-secreted cytokines, and in turn, induced the mammosphere formation and self-renewal of expansion in breast cancer cells via NOTCH1, constituting a cancer-stroma-cancer signaling circuit [46]. In addition, Su et al. have found that CD10⁺GPR77⁺ CAFs provide a survival niche for CSCs, thereby promoting tumor formation and chemoresistance [47]. Mechanistically, CD10⁺GPR77⁺ CAFs are driven by persistent NF-κB activation via p65 phosphorylation and acetylation, which is maintained by complement signaling via GPR77. In line with this notion, targeting these CAFs with a neutralizing anti-GPR77 antibody abolishes tumor formation and restores chemosensitivity [47]. On the other hand, BCSCs reciprocally promote the formation of CAFs. Breast cancer stem cells are found to upregulate IRF6 in stromal fibroblasts to induce stromagenesis, in order to sustain the BCSC tumor niche [48].

4.1.3 Hypoxia

Hypoxia is a key feature of the breast cancer TME that promotes expression of the transcription factor hypoxia-inducible factor 1 (HIF-1) and is correlated with unfavorable prognosis and chemoresistance. HIF-1 induces the expression or activity of stem cell pluripotency factors, including NANOG, KLF4, OCT4 and SOX2 [49,50]. It is also noteworthy that HIF-1 facilitates the evasion of phagocytosis and the maintenance of BCSCs by activating the transcription of CD47 [51]. CD47 functions as a mediator of immune evasion by delivering a 'don't eat me' signal to macrophages [52]. Hypoxia-induced secretion regulates the dissemination of CSCs, mediated by JAK-STAT signaling [53]. A recent study found that nuclear prelamin A recognition factor (NARF) acts as a hypoxia-induced coactivator for OCT4-mediated BCSC specification [49]. It was also illustrated that HIF-2α, but not HIF-1α, induces breast cancer cell stemness under hypoxia [54]. Mechanistically, hypoxia inducible factor-2α (HIF-2α) activates superoxide dismutase 2 (SOD2) under hypoxic conditions, thereby decreasing mitochondrial reactive oxygen species (mtROS) levels. The reduced mtROS transported to the endoplasmic reticulum subsequently activate unfolded protein response (UPR) via GRP78, which eventually confer chemoresistance of cancer cells, increase of OCT4 expression, and both CD44⁺CD24⁻BCSC and ALDH1⁺ BCSC populations [54]. Accordingly, depletion or inhibition of HIF-2α was observed to abolish the hypoxia-induced sphere-forming ability and the percentage of CD44⁺CD24⁻ cell subpopulation [54]. Of note, glycolysis was found to reprogram BCSC under hypoxic conditions. Pyruvate dehydrogenase kinase 1 (PDK1), a key glycolysis gene, was found to be overexpressed in BCSCs and accumulated in hypoxic regions. The inhibition of PDK1 led to a remarkable reduction in the percentage of ALDH⁺ BCSC subpopulation, expression of stemness-related transcriptional factor, and sphere-forming ability and tumor growth in breast cancer [55]. Taken together, hypoxia has been identified as one of the critical factors in the TME that sustain the BCSC stemness.

4.1.4 Tumor-resident microbiota

The roles of microbiota for cancer cell stemness, metastasis and response to treatment remain poorly understood, having only recently emerged as a topic of investigation. Tumor-resident intracellular microbiota, albeit at low biomass, were shown to promote metastatic colonization in breast cancer [56]. It is noteworthy that microbiota enterotoxigenic Bacteroides fragilis (ETBF) was found to be highly enriched in tumors of breast cancer patients who did not respond to taxane-based neoadjuvant chemotherapy [57]. Mechanistically, the toxic protein BFT-1 secreted by ETBF, directly bound to the cytosolic pattern recognition receptor NOD1. NOD1 increased BCSCs by activating the NOTCH1-HEY1 signaling pathway. In line with this notion, NOD1 inhibition and ETBF clearance was shown to increase the chemosensitivity of breast cancer by impairing BCSCs [57]. However, evidence on how the tumor-resident microbiota regulate breast cancer stemness and plasticity is still rare, which warrants further investigation.

5. Impact of BCSC Heterogeneity and Plasticity

The heterogeneity and plasticity of BCSCs contribute remarkably to their high malignancy. BCSCs are able to switch between different states in response to corresponding signals and environmental stimuli. The consequences of such plasticity include increased invasive capacity via epithelial-mesenchymal transition, immune evasion, vasculogenic mimicry and resistance to chemotherapy and radiotherapy (Figure 2).

Figure 2. Impact of BCSC plasticity and heterogeneity in breast cancer malignancy. BCSCs can transit between mesenchymal and epithelial states in response to stimulation by TGF-β1. They are capable of evading immune cell-mediated killing through the activation of WNT/β-catenin or the reduction of antigen processing (TAP) genes and co-stimulatory molecule CD80. VEGF and TEM8 promotes vasculogenic mimicry in BCSCs. EMT: Epithelial-mesenchymal transition; MET: Mesenchymal-epithelial transition (MET). Created with BioGDP.com.

6. Preclinical/Clinical Therapeutic Strategies Targeting BCSCs

Given the pivotal roles that BCSCs play in tumorigenesis and treatment resistance, many therapeutic approaches have been tested in preclinical settings alone or in combination with chemotherapy. A summary of recent novel preclinical therapeutics, along with their respective targets and effects on BCSCs, is presented in Table 1.

Table 1. Preclinical therapeutics targeting BCSCs.

The first strategy involves the development of small molecule inhibitors targeting BCSC signalings or the TME. For instance, targeting the Hh signaling pathway may prove an effective strategy to overcome BCSC plasticity. In the phase I clinical trial EDALINE, three of the 12 patients with metastatic TNBC derived clinical benefit from combination therapy with the SMO inhibitor Sonidegib and docetaxel chemotherapy, with one patient experiencing a complete response [70]. In accordance with this hypothesis, novel combinations of JAK2 inhibitors (ruxolitinib and pacritinib) with SMO inhibitors (vismodegib and sonidegib) synergistically suppressed breast cancer stem cells, tumor growth and metastasis [96]. Additionally, CD73 expression is required for the maintenance of CD44⁺CD24^– BCSCs. Quercetin and luteolin, which are commonly used as over-the-counter supplements, combined with paclitaxel could effectively downregulate paclitaxel-enhanced CD73 and CSC-promoting pathways YAP and Wnt. The triple-drug combination could inhibit paclitaxel-enriched CSCs, suppress the growth of TNBC cell lines and patient-derived xenograft organotypic cultures [88]. The FDA-approved PI3 kinase p110а inhibitor, alpelisib (BYL-719), has the potential to inhibit the stemness and EMT phenotype in BCSCs in 3D cultures. A comparison of eribulin-resistant breast cancer cell lines revealed that BYL-719 effectively overcame drug resistance [87].

In addition to small molecule inhibitors, the use of currently approved antibody also showed promising effects. Notably, pharmacological inhibition of IL-6 by tocilizumab potentiates cisplatin cytotoxicity and targets cancer stem cells in TNBC [92]. New antibodies are also in development. The antibody targeting GD2 (Dinutuximab), which is expressed in around 60% of primary TNBC, is able to inhibit TNBC growth by targeting GD2⁺ breast cancer stem-like cells [93].

The implementation of immune cytotherapy also showed promising results. For instance, DLL4-targeted CAR-T therapy has been shown to eliminate cancer stem cells and reshape immune microenvironment in HER2⁺ breast cancer, thereby sensitizing neoadjuvant chemotherapy [94]. Likewise, the TEM8 CAR T cells targeted breast cancer stem-like cells, offsetting the formation of mammospheres [95].

Besides preclinical studies, multiple therapeutic agents targeting BCSC in combination with chemotherapy have been tested in clinical trials. Relevant studies have been summarized in Table 2. The metabolism of CSCs in hypoxia suggests an altered balance between mitochondrial oxidative phosphorylation and glycolysis [97]. In line with this notion, specific drugs targeting mitochondrial metabolism, leading to increased ROS levels, may inhibit CSCs. Many classes of FDA-approved antibiotics, including doxycycline, actually target mitochondria. A recent clinical trial (NCT06452394) aims to check for the clinical efficacy of doxycycline to target the BCSCs and improve the response to neoadjuvant chemotherapy in ER+/HER2– breast cancers.

Table 2. Clinical trials targeting BCSCs.

Moreover, cirmtuzumab is a receptor-tyrosine-kinase like orphan receptor 1 (ROR1) antibody [98]. A phase 1b study has evaluated the safety and efficacy of cirmtuzumab combined with paclitaxel for metastatic or locally advanced, unresectable breast cancer (NCT02776917). The stem cell makers, ALDH and CD133, from primary pre-treatment and post-treatment tumor specimens were measured to assess the effects on BCSCs. Furthermore, CXCR1 has been identified as a druggable target on BCSCs. The CXCR1 inhibitor, reparixin, in combination with paclitaxel has been tested in TNBC patients. Unfortunately, the primary endpoint of prolonged PFS was not met [99].

7. Concluding Remarks and Future Perspectives

Overall, the studies on the heterogeneity and plasticity of BCSCs underscores the complexity of these cells and their roles in tumorigenesis, tumor progression and treatment resistance. Accumulating evidence have proved that a number of factors contribute to the plasticity of BCSCs. These include pro-tumor signaling, immune cells and stromal cells in the TME, hypoxia, non-coding RNAs (e.g., miRNAs, and lncRNAs) and the recently discovered tumor-resident microbiota. A comprehensive understanding of the molecular mechanisms and environmental factors that drive BCSC plasticity is essential for the development of targeted therapies and the improvement of patient outcomes. A disruption of the tumor-TME network may prove an effective strategy for the treatment of breast cancer.

Despite fundamental achievements, several questions still remain elusive. First of all, the eradication of BCSCs may prove challenging due to their inherent heterogeneity. The manner in which distinct subtypes of BCSCs compete or collaborate at various stages of tumor progression remains to be elucidated. Recent advances in spatial transcriptomics and single-cell sequencing may facilitate the resolution of these questions. Secondly, it is crucial to choose the optimal treatment regimens of combining BCSC-targeted therapy with systemic therapy, while minimizing the incidence of adverse effects. Thirdly, the majority of BCSC-targeted therapies have been evaluated in preclinical models. Further clinical trials are required to validate the efficacy and safety of these treatments in patients with breast cancer.

Declarations

References