MicroRNA-mediated modulation of cancer-associated fibroblasts in HER2-positive breast tumor microenvironment: a comprehensive review

Abstract

The intricate interplay within the tumor microenvironment (TME) significantly dictates the trajectory of cancer progression and therapeutic response. In HER2-positive breast cancer, a particularly aggressive subtype, cancer-associated fibroblasts (CAFs) emerge as pivotal stromal components, actively orchestrating malignant behaviors. Concurrently, microRNAs (miRNAs), small non-coding RNAs, serve as potent post-transcriptional regulators and critical mediators of intercellular communication, often encapsulated within exosomes. This review provides a comprehensive analysis of the reciprocal miRNA-mediated modulation between HER2-positive breast cancer cells and CAFs. It elucidates how tumor cell-derived miRNAs reprogram normal fibroblasts into pro-tumorigenic CAFs, and how CAF-derived miRNAs, in turn, influence HER2-positive cancer cell proliferation, invasion, metastasis, and crucially, resistance to HER2-targeted therapies. Understanding this dynamic axis reveals a self-sustaining feedback loop that drives disease advancement and therapeutic evasion. This synthesis underscores the immense potential of targeting these complex miRNA-CAF interactions as a novel strategy for diagnostic, prognostic, and therapeutic interventions, aiming to overcome the persistent challenge of resistance in HER2-positive breast cancer.

Introduction

Background

Breast cancer remains a formidable global health challenge, with the HER2-positive subtype representing a particularly aggressive form characterized by its propensity for rapid proliferation and metastasis [1, 2]. For decades, the prevailing paradigm in cancer research centered primarily on the genetic aberrations within the malignant cells themselves [3]. However, a profound shift in understanding has increasingly highlighted the critical role of the tumor microenvironment (TME) as a dynamic and influential ecosystem, profoundly shaping tumor behavior, progression, and response to therapy [3, 4]. This evolving perspective recognizes that non-mutant stromal cells within the TME exert a strong modulatory influence on disease progression and therapeutic outcomes [3, 5].

Rationale

Among the diverse cellular populations comprising the TME, cancer-associated fibroblasts (CAFs) stand out as the most abundant mesenchymal cell type in breast cancer, playing a multifaceted and often pro-tumorigenic role [3, 6]. These activated fibroblasts are far from passive bystanders; they actively contribute to tumor energy metabolism, promote angiogenesis, impair immune cell function, and extensively remodel the extracellular matrix, thereby facilitating cancer progression and fostering therapeutic resistance [7–10]. CAFs are recognized as dynamic entities, undergoing modification throughout the initiation and advancement of tumorigenesis, rather than being a static, independent cell population [11, 12]. The persistent activation of fibroblasts within the tumor, unlike the transient nature of normal wound healing, signifies a fundamental dysregulation that drives chronic pathological activation, contributing to the sustained pro-tumorigenic state [13]. This recognition implies that comprehending and disrupting the mechanisms that maintain this chronic activation, potentially involving microRNAs (miRNAs) or their upstream regulators, could offer innovative therapeutic avenues to re-educate these cells [14].

Concurrently, microRNAs have emerged as pivotal regulators of gene expression, operating at the post-transcriptional level [15]. These small non-coding RNA molecules are capable of influencing a vast array of cellular processes, from proliferation and differentiation to apoptosis and metabolism [16–18]. Beyond their intracellular roles, miRNAs are increasingly recognized as key mediators of intercellular communication, often secreted into the extracellular milieu, frequently encapsulated within exosomes, where they act as chemical messengers to modulate the behavior of distant or neighboring cells [15, 18, 19]. Aberrant miRNA expression is a hallmark of cancer, with specific miRNAs acting as either oncogenes or tumor suppressors, depending on their context and target genes [20, 21]. Despite the established importance of both stromal cells and non-coding RNAs in cancer biology, significant knowledge gaps remain regarding their intersection in HER2-positive disease. Most current literature treats CAF activation and HER2-targeted therapy resistance as separate phenomena, failing to integrate them into a unified mechanistic model. Furthermore, while miRNA dysregulation is well-documented in triple-negative breast cancer, specific data distinguishing HER2-driven stromal reprogramming from general breast cancer fibrosis is sparse [22]. This review addresses these critical voids by synthesizing recent evidence on the reciprocal miRNA-CAF axis, specifically highlighting how these interactions create a unique, resistance-prone microenvironment that standard anti-HER2 protocols fail to dismantle.

Scope of review

This review posits that miRNAs orchestrate a complex and reciprocal interplay between HER2-positive breast cancer cells and CAFs. This intricate modulation profoundly influences tumor survival, metastasis, and the evasion of targeted therapies. A comprehensive understanding of these dynamic interactions is essential for developing more effective diagnostic, prognostic, and therapeutic strategies to overcome the significant clinical challenge posed by HER2-positive breast cancer.

Cancer-associated fibroblasts: orchestrators of the tumor microenvironment

Origins and heterogeneity

Cancer-associated fibroblasts (CAFs) represent a highly dynamic and diverse population of stromal cells within the tumor microenvironment, playing a substantial role in tumor progression [3, 23, 24]. Their origins are remarkably heterogeneous. CAFs can differentiate from resident normal fibroblasts (NFs) present within the breast tissue, a conversion influenced by various factors including growth factors, cytokines, exosomal microRNAs (miRNAs), and proteins [11, 25]. Beyond NFs, CAFs can also arise from other cell types through various mesenchymal transitions, such as tumor cells undergoing epithelial-mesenchymal transition (EMT), endothelial cells undergoing endothelial-mesenchymal transition (EndMT), pericytes, adipocytes, and bone marrow mesenchymal stem cells (MSCs) [11]. This diverse lineage contributes to the functional variability observed among CAF populations.

Mechanisms of activation

The activation of fibroblasts into CAFs is a critical step in their pro-tumorigenic transformation [26]. This process is driven by inflammatory cytokines, local hypoxia, and CAF-derived exosomes within the TME [13, 27]. For instance, in the early stages of tumor progression, secreted cytokines from cancer cells, such as IL-1β, guide normal fibroblasts to differentiate into pro-inflammatory CAFs, often depending on NF-κB activation [13, 28]. Unlike the transient activation of fibroblasts during normal wound healing, CAFs in the tumor microenvironment remain in constitutively activated state (Fig. 1).

Fig. 1.

Schematic representation of cancer-associated fibroblasts (CAFs) within the tumor microenvironment

Diverse roles in breast cancer progression

ECM remodeling and physical barriers

One of the primary functions of CAFs is the excessive production and remodeling of extracellular matrix (ECM) components, particularly collagen [13, 29]. This desmoplastic reaction creates a dense physical barrier that elevates interstitial fluid pressure, thereby limiting the diffusion of monoclonal antibodies like trastuzumab and impeding the physical infiltration of cytotoxic T-cells [7, 13].

Immune modulation

CAFs actively suppress anti-tumor immunity. They secrete cytokines such as IL-6, which downregulates the tumor suppressor HIC1, and IL-1β, which modulates the immune landscape to favor lung metastasis [11, 30]. Furthermore, CAFs promote the differentiation of T-cells into immunosuppressive regulatory T-cells (Tregs) via antigen presentation lacking costimulatory molecules, effectively shielding HER2-positive cells from immune surveillance [13, 31].

Angiogenesis and metastasis

To support rapid tumor growth, CAFs induce angiogenesis through the secretion of VEGF and PDGF, as well as through VEGF-independent STAT3 signaling pathways [13]. Beyond local growth, CAF-derived chemokines like CXCL12, CXCL14, and CCL5 act as potent pro-metastatic factors, recruiting bone marrow-derived cells to form pre-metastatic niches in distant organs [32, 33].

CAFs are central orchestrators of the tumor microenvironment. One of their primary functions is the excessive production and remodeling of extracellular matrix (ECM) components, such as collagen [13, 29]. This creates a dense physical barrier that not only mechanically supports cancer tissues but also acts as a reservoir for growth factors.

Beyond structural support, CAFs secrete a wide array of paracrine factors that promote malignant behavior. These include cytokines like interleukin-1β (IL-1β), which promotes breast cancer metastasis to the lungs by modulating the immune cell environment and upregulating adhesion molecules [11]. IL-8/CXCR1/2 signaling from CAFs activates and induces migration and invasion of breast cancer cells, while IL-6 downregulates the tumor suppressor HIC1, facilitating breast cancer development [11, 34]. Growth factors such as vascular endothelial growth factor (VEGF), which induces angiogenesis, are frequently induced in the TME, with CAFs being a primary source of VEGF [13]. CAFs also promote angiogenesis via VEGF-independent STAT3 signaling pathways and through the secretion of platelet-derived growth factor (PDGF), which recruits stromal fibroblasts to secrete VEGF [13]. Chemokines like CXCL12, CXCL14, and CCL5 secreted by CAFs are identified as pro-metastatic factors, recruiting bone marrow-derived cells and immune cells [13, 35–37]. The collective actions of these secreted factors and ECM remodeling enable tumor cells to invade the TME, establish interactions with other cells, and promote proliferation, invasion, metastasis, angiogenesis, immunosuppression, and drug resistance [7, 38, 39].

Left section (Origins)

CAFs can originate from multiple sources, including resident normal fibroblasts, tumor cells via epithelial-mesenchymal transition (EMT), endothelial cells via endothelial-mesenchymal transition (EndMT), pericytes, adipocytes, and bone marrow–derived mesenchymal stem cells (MSCs).

Middle section (Activation)

Normal fibroblasts are converted into CAFs through tumor-derived cytokines (e.g., IL-1β), hypoxia, and CAF-derived exosomes, often involving NF-κB signaling. Unlike transient wound-healing fibroblasts, CAFs remain persistently activated, producing excess extracellular matrix components such as collagen.

Right section (Heterogeneity)

CAF populations are diverse, with distinct phenotypes marked by α-smooth muscle actin (α-SMA) and fibroblast activation protein (FAP). Functional diversity among CAF subtypes can promote or, in some cases, restrain tumor progression, underscoring the need for selective therapeutic targeting or normalization strategies.

HER2-positive breast cancer: a distinct subtype and therapeutic challenges

Molecular characteristics and clinical implications

HER2-positive breast cancer is defined by the overexpression or amplification of the ERBB2 oncogene, which encodes the human epidermal growth factor receptor 2 (HER2) tyrosine-protein kinase [1, 40]. This subtype constitutes approximately 15–20% of all breast cancers [41]. Historically, HER2-positive breast cancer was associated with an aggressive phenotype, characterized by high proliferation rates, increased invasiveness, and a poor prognosis when left untreated [1, 30].

HER2 functions as a proto-oncogene, playing a crucial role in cell growth and differentiation [1, 32]. Uniquely, HER2 lacks a physiological ligand and its extracellular domain is constitutively poised for dimerization [1]. In cancer cells, the massive overexpression of HER2 overwhelms this natural restraint, leading to uncoupling of kinase domain dimerization and signaling from the extracellular domains [1, 33]. Dimerization, particularly with HER3, results in the phosphorylation of C-terminal tails and the subsequent activation of critical downstream signaling pathways, including the phosphatidylinositol 3-kinase (PI3K)/AKT and mitogen-activated protein kinase (MAPK)/ERK pathways, which are central to cell growth, survival, and migration [1, 31].

Despite its defining molecular features, HER2-positive breast cancer itself exhibits significant heterogeneity. Molecular classification studies have revealed distinct subtypes within HER2-positive breast cancer, such as HER2-enriched, estrogen receptor (ER)-activated, immunomodulatory, and highly heterogeneous subtypes [42, 43]. Each of these molecular classifications presents unique characteristics and implies different therapeutic strategies, underscoring the complexity even within this defined breast cancer subtype [42].

Intrinsic heterogeneity: luminal vs. HER2-enriched

Critical evaluation of miRNA dysregulation requires distinguishing between Luminal-HER2 (ER -positive/ HER2-positive) and HER2-Enriched (ER-negative/HER2-positive) subtypes, as their stromal interactions differ fundamental. In Luminal-HER2 tumors, the interplay between Estrogen Receptor (ER) and HER2 signaling creates a unique miRNA landscape. For instance, miR-205 and let-7 are frequently downregulated in HER2-Enriched tumors, contributing to their high proliferative index and stemness features [42, 43]. Conversely, in Luminal-HER2 subtypes, miR-221/222 are critical mediators that can target ERα, promoting a switch from hormone-dependence to growth factor-dependence, a key driver of resistance to endocrine therapies like tamoxifen [44]. Understanding these subtype-specific profiles is essential, as the CAF secretome in a Luminal-HER2 tumor may prioritize immune exclusion, whereas in HER2-Enriched tumors, it may drive rapid angiogenesis.

Mechanisms of HER2-targeted therapy resistance

The advent of HER2-targeted therapies, including monoclonal antibodies like trastuzumab and pertuzumab, and tyrosine kinase inhibitors (TKIs) such as lapatinib, neratinib, and tucatinib, has dramatically improved the prognosis for patients with HER2-positive breast cancer [1, 45]. However, both primary (pre-existing) and acquired (developing during treatment) drug resistance remain significant clinical obstacles, limiting the long-term efficacy of these otherwise successful treatments [41, 46].

The phenomenon of HER2 addiction, where the oncogene is continuously required throughout the entire disease process, from initiation to advanced metastatic disease, is well-established [1]. Yet, despite this fundamental dependence, resistance to therapy persists as a major challenge [41, 47]. This seeming paradox arises because tumors develop adaptive mechanisms to circumvent HER2 inhibition, often by relying on alternative signaling pathways or exploiting external cues from the TME [7, 48]. For instance, microenvironmental signals can profoundly influence resistance. In HER2-enriched (HER2E) cells, hepatocyte growth factor (HGF), a ligand for MET, can induce resistance that can be reversed by a MET inhibitor [44]. Similarly, in luminal-like HER2-positive (L- HER2-positive) cells, neuregulin1-β1 (NRG1β), a ligand for HER3, induces resistance that can be reversed by inhibiting HER2-HER3 heterodimerization [44]. These observations highlight that targeting the oncogene alone is often insufficient, and overcoming resistance necessitates disrupting the adaptive mechanisms that allow the tumor to escape its HER2 dependence [49].

Beyond direct signaling pathways, the tumor immune microenvironment also plays a crucial role in modulating the response to treatment in HER2-positive breast cancer [50]. Studies have shown that the infiltration of specific immune cell populations, such as B-cells and regulatory T-cells (Tregs), in the primary tumor is associated with unfavorable overall survival in patients with HER2-positive metastatic breast cancer [50, 51]. This indicates that the immune contexture within the TME is a significant determinant of therapeutic response and patient outcomes. Therefore, future strategies for HER2-positive breast cancer should integrate immunomodulatory approaches, potentially by targeting miRNAs that influence immune cell infiltration or function within the TME, to improve patient survival. This represents a critical, yet still underexplored, area for therapeutic intervention (Fig. 2).

Fig. 2.

Mechanisms of Resistance to HER2-Targeted Therapies. A Therapeutic Action: Monoclonal antibodies (Trastuzumab, Pertuzumab) bind extracellular domains to prevent dimerization, while Tyrosine Kinase Inhibitors (Lapatinib, Tucatinib) block intracellular phosphorylation. B Bypass Signaling (Resistance): Tumor cells evade inhibition via alternative receptor activation. The binding of HGF to the MET receptor, or Neuregulin-1β (NRG1β) to HER3, sustains downstream PI3K/AKT and MAPK/ERK signaling despite HER2 blockade. C Immune Microenvironment: Infiltration of regulatory T-cells (Tregs) and immunosuppressive B-cells hampers the antibody-dependent cellular cytotoxicity (ADCC) effect of trastuzumab

MicroRNAs: master regulators of gene expression and intercellular communication

Biogenesis and processing

MicroRNAs (miRNAs) are a class of small, non-coding RNAs, typically around 22 nucleotides in length. Their biogenesis follows a highly conserved, sequential pathway. The process begins in the nucleus, where miRNAs are initially transcribed from DNA sequences by RNA polymerase II into long primary miRNA transcripts (pri-miRNAs) [15]. These pri-miRNAs are then processed by the microprocessor complex (Drosha and DGCR8) into precursor miRNAs (pre-miRNAs) [15, 52, 53] (Fig. 3).

Fig. 3.

Canonical MicroRNA Biogenesis and Mechanism of Action. A Nuclear Processing: miRNA genes are transcribed by RNA Polymerase II into primary miRNAs (pri-miRNAs). The Microprocessor complex, comprising the RNase III enzyme Drosha and its cofactor DGCR8, cleaves the pri-miRNA into a hairpin-shaped precursor miRNA (pre-miRNA). B Nuclear Export: The pre-miRNA is recognized by Exportin-5 and actively transported from the nucleus to the cytoplasm in a RanGTP-dependent manner. C Cytoplasmic Processing: In the cytoplasm, the RNase III enzyme Dicer removes the terminal loop of the pre-miRNA, generating a mature miRNA duplex (~ 22 nucleotides). D Target Recognition and Silencing: The functional guide strand is selectively loaded into an Argonaute (AGO) protein to form the miRNA-induced silencing complex (miRISC). The miRISC binds to complementary sequences (typically in the 3’ UTR) of target mRNAs, resulting in either translational repression or mRNA degradation

Mechanisms of silencing

Following nuclear processing, pre-miRNAs are actively transported to the cytoplasm by Exportin-5. There, the enzyme Dicer processes the pre-miRNA into a mature duplex. One strand is loaded into an Argonaute (AGO) family protein to form the miRNA-induced silencing complex (miRISC) [15]. The miRISC binds to complementary sequences on target mRNAs, leading to gene silencing primarily through translational inhibition or mRNA degradation [15].

Mechanisms of miRNA dysregulation in malignancy

Aberrant expression of miRNAs is a widely recognized hallmark of cancer, including breast cancer, where it can either promote or inhibit disease progression [20]. The dysregulation of miRNAs in cancer can arise from various mechanisms, including genomic amplification or deletion of miRNA genes, epigenetic modifications (such as DNA methylation or histone modifications), transcriptional dysregulation, and defects in the miRNA biogenesis machinery itself [20, 54]. For instance, elements of the miRNA machinery, including Dicer and Drosha, are known to be involved in the progression of several cancer subtypes and can serve as predictive markers [55, 56]. The observation that dysregulation can occur at the level of core biogenesis machinery suggests that targeting these fundamental processes, rather than individual miRNAs, could offer a broad-spectrum approach to globally restore tumor-suppressive miRNA profiles or inhibit oncogenic ones. However, this strategy also carries the inherent risk of significant off-target effects, given the widespread physiological roles of miRNAs [57].

Depending on their specific target genes and the cellular context, dysregulated miRNAs can function as either oncogenes (oncomiRs) or tumor suppressors (miRsupps) [58, 59]. For example, some miRNAs may be highly expressed and contribute to tumor progression, while others are frequently downregulated and inhibit tumorigenesis by regulating processes like cell growth and apoptosis [60, 61]. This context-dependent nature means that a single miRNA can exhibit dual functions, acting as an oncogene in one cancer type and a tumor suppressor in another, or even within different subtypes of the same cancer [60, 62]. This complexity underscores that therapeutic strategies involving miRNAs must be highly specific to the cancer type and even subtype (e.g., HER2-positive vs. triple-negative breast cancer). While this adds to the complexity of therapeutic development, it also opens avenues for highly precise interventions that exploit these context-specific roles, potentially minimizing off-target effects in healthy tissues and maximizing therapeutic precision [63].

Dysregulated miRNAs contribute to virtually all hallmarks of cancer, including uncontrolled cell proliferation, enhanced cell growth, evasion of apoptosis, increased invasion and metastasis, immune escape, and altered cellular metabolism [16]. Their profound impact on these fundamental biological processes, coupled with their remarkable stability in biological fluids, makes them highly promising non-invasive biomarkers. Circulating miRNAs can be detected in blood and other body fluids, offering potential for early diagnosis, prognostic assessment, and real-time monitoring of treatment response [16, 64].

Extracellular signaling and exosomes

Beyond intracellular functions, miRNAs are actively secreted into the extracellular milieu. This is not a random process; specific miRNAs are selectively sorted into exosomes based on specific motifs. RNA-binding proteins, such as hnRNPA2B1 and synaptotagmin-binding cytoplasmic RNA-interacting protein (SYNCRIP), recognize specific ‘EXO-motifs’ (e.g., GGAG sequences) on miRNAs, guiding their loading into the intraluminal vesicles of multivesicular bodies (MVBs) [15]. This sorting mechanism explains why the exosomal miRNA profile often differs significantly from the cellular profile of the donor cell. Once released, these exosomes exhibit distinct uptake kinetics; they are often internalized via endocytosis or membrane fusion, allowing the transferred miRNAs to repress target mRNAs in recipient cells, thereby altering the TME landscape [65, 66].

Reciprocal microRNA-mediated modulation of CAFs

The influence within the TME is profoundly bidirectional. This section dissects the “two-way street” of communication: first, how tumor cells corrupt normal fibroblasts, and second, how these activated CAFs subsequently fuel tumor malignancy.

Tumor cell-derived miRNAs: reprogramming the stroma

HER2-positive breast cancer cells actively reprogram their surrounding microenvironment, although the specificity of these interactions often overlaps with broader breast cancer mechanisms. A pivotal HER2-specific interaction involves miR-146a-5p. Cabello et al. demonstrated that miR-146a-5p is the major dysregulated miRNA in trastuzumab-resistant HER2-positive cells and is highly enriched in their exosomes. These exosomes can horizontally transfer resistance traits and have been implicated in activating endothelial cells and fibroblasts within the HER2-positive niche [41, 67].

In contrast, other significant miRNA-CAF interactions, while highly relevant to HER2-positive disease, have primarily been characterized in general breast cancer or triple-negative models. For instance, miR-425-5p facilitates the transition of normal fibroblasts to CAFs via the TGFβ1/ROS signaling pathway [11], while miR-105 reprograms CAF metabolism to provide nutritional support for tumor growth [68].

CAF-derived miRNAs: feedback to the tumor

Once activated, CAFs release their own repertoire of miRNAs, often via exosomes, which act as a feedback loop to enhance HER2-positive cell survival. While specific HER2-exclusive data is emerging, general breast cancer studies illustrate the potency of this feedback. For example, CAF-released miR-29a interferes with p38-STAT1 signaling to promote drug resistance, and miR-3613-3p acts as an oncogene to promote ROS production and metastasis [68–70]. Additionally, CAFs secrete miR-221/222, which can suppress estrogen receptor expression, fostering the aggressive, drug-resistant phenotype characteristic of HER2-enriched tumors.

Site-specific metastasis: the “Educated” niche

Emerging evidence highlights that miRNA-mediated reprogramming is not uniform; it is dictated by the specific metastatic niche. In brain metastases, which affect up to 50% of metastatic HER2-positive patients, the ‘CAF’ function is often assumed by reactive astrocytes. Ahmad et al. demonstrated that miR-20b is significantly upregulated in HER2-positive brain metastases compared to primary tumors, facilitating the colonization of the brain parenchyma [71]. Conversely, the downregulation of miR-1258 targets heparanase to promote brain invasion [72]. In the liver, tumor-derived exosomes carrying miR-122 have been shown to reprogram hepatic stellate cells—the liver’s resident fibroblasts—into a pro-metastatic phenotype [73, 74]. This suggests that therapeutic strategies must be organ-specific: a miRNA inhibitor effective against breast CAFs might fail to normalize the distinct stromal phenotype of a brain or liver metastasis.

The immune-CAF-miRNA axis: orchestrating immunosuppression

A critical gap in earlier literature is the role of miRNA-CAF crosstalk in mediating immune evasion. CAFs do not merely block T-cells physically; they actively remodel the immune landscape via miRNA signaling.

• Treg recruitment The specific exclusion of cytotoxic T-cells is often driven by the recruitment of FOXP3 + Regulatory T-cells (Tregs). High levels of stromal miR-21 have been shown to correlate with increased Treg infiltration in the breast tumor microenvironment, creating an immunosuppressive sanctuary that protects HER2-positive cells from antibody-dependent cellular cytotoxicity (ADCC) [7–10, 50, 51].

• Macrophage plarization CAFs actively reprogram infiltrating innate immune cells. Recent comprehensive reviews highlight that CAF-derived exosomes transfer specific miRNA cargos to monocytes, polarizing them from an anti-tumor M1 state to a pro-tumor M2 macrophage phenotype [75]. These M2 macrophages subsequently secrete immunosuppressive cytokines (e.g., IL-10, TGF-β) that further inhibit T-cell function and promote therapeutic resistance.

MicroRNA-mediated CAF influence on HER2-targeted therapy resistance

Cancer-associated fibroblasts (CAFs) are critical contributors to the development of therapeutic resistance in breast cancer, including the HER2-positive subtype [7]. Their mechanisms of resistance are diverse, encompassing the formation of physical barriers, metabolic reprogramming, secretion of exosomes, enhancement of DNA repair, activation of bypass signaling pathways, upregulation of multidrug resistance proteins, and inhibition of immune checkpoints [7]. MicroRNAs (miRNAs) play a central role in regulating this acquired drug resistance [76]. Table 1 summarizes key miRNAs involved in modulating CAF function and HER2-positive breast cancer progression and resistance, highlighting their source, targets, and effects [77].

Table 1.

Key MicroRNAs modulating CAF function and HER2-Positive breast cancer Progression/Resistance

miRNA name	Model context	Source	Target genes/pathways	Effect on HER2-positive biology	References
miR-146a-5p	HER2-positive Specific (Trastuzumab-resistant lines & patients)	Tumor Exosomes	CDKN1A, NF-κB	Directly confirmed: Promotes trastuzumab resistance, angiogenesis, and EMT.	[41]
miR-1246	HER2-positive Specific (Patient serum)	Tumor Exosomes	Cyclin G2	Directly confirmed: Predictive biomarker for trastuzumab resistance; correlates with poor survival.	[79]
miR-155	HER2-positive Specific (Patient serum)	Tumor Exosomes	FOXO3a	Directly confirmed: Linked to therapeutic failure and immune evasion in HER2-positive patients.	[79]
miR-21	HER2-positive Specific	Tumor Cells	PTEN	Directly confirmed: Upregulated by HER2 signaling; confers resistance to HER2-targeted therapy.	[80]
miR-425-5p	General Breast Cancer (General BC models)	Tumor Exosomes	TGFβ1/ROS	Extrapolated: likely drives pro-fibrotic CAF activation in HER2-positive stroma.	[11]
miR-105	General Breast Cancer (MYC-driven models)	Tumor Cells	MYC signaling	Extrapolated: metabolic reprogramming (glucose/glutamine) relevant to MYC-amplified HER2-positive tumors.	[68]
miR-29a	General Breast Cancer (TNBC/Metastatic models)	CAF-derived	p38-STAT1	Extrapolated: promotes drug resistance and metastasis; relevant to aggressive HER2-positive phenotypes.	[69]
miR-3613-3p	General Breast Cancer (CAF exosomes)	CAF-derived	SOCS2	Extrapolated: promotes ROS and metastasis; potential mechanism for HER2-positive dissemination.	[70]

Clinical utility relies on distinguishing between resistance to monoclonal antibodies (trastuzumab/pertuzumab) versus Tyrosine Kinase Inhibitors (TKIs like lapatinib, neratinib).

• Trastuzumab resistance Is predominantly associated with miR-146a-5p and miR-21, which activate bypass signaling (PI3K/AKT) and induce EMT to evade antibody-dependent cellular cytotoxicity (ADCC) [41].

• TKI (Lapatinib) resistance Displays a distinct signature driven by miR-221. Hynh et al. revealed that miR-221 confers resistance to lapatinib by specifically targeting p27Kip1, a cell cycle regulator, downstream of the Src/NF-κB pathway [78].

This distinction is clinically vital: a rise in circulating miR-221 in a patient progressing on trastuzumab might indicate a specific vulnerability to TKI-based rescue therapies, whereas elevated miR-146a might suggest a need for broader PI3K pathway blockade.

The dual-function synthesis: miRNAs bridging CAF activation and resistance

While many studies view CAF activation and therapeutic resistance as separate events, emerging evidence suggests they are driven simultaneously by a subset of ‘dual-function’ miRNAs. This synthesis is critical for understanding why stroma-rich tumors are difficult to treat.

The most prominent example in HER2-positive models is miR-146a-5p. It does not function linearly; rather, it creates a synchronized response. By targeting CDKN1A and NF-κB regulators, it concurrently promotes the expansion of the CAF population (stromal activation) and induces epithelial-to-mesenchymal transition (EMT) in the tumor cells (resistance execution) [41]. Similarly, miR-21 acts as a dual node: its transfer from CAFs suppresses PTEN in cancer cells to bypass HER2 blockade (resistance), while its intracellular upregulation in fibroblasts maintains their myofibroblastic state (activation) [80]. Identifying these dual-function miRNAs offers a strategic advantage: targeting a single molecule could simultaneously normalize the stroma and re-sensitize the tumor to anti-HER2 therapies.

miR-146a-5p

This miRNA is a prominent player in trastuzumab resistance in HER2-positive breast cancer. It is significantly upregulated in trastuzumab-resistant HER2-positive breast cancer cells [41]. High expression of miR-146a-5p in primary tumor tissue of HER2-positive patients correlates with shorter disease-free survival, indicating its clinical relevance [41]. Mechanistically, overexpression of miR-146a-5p in sensitive HER2-positive breast cancer cells reduces their response to trastuzumab, while its inhibition in resistant cells increases their sensitivity [41]. This miRNA promotes increased migration and induces epithelial-to-mesenchymal transition (EMT), a process associated with trastuzumab resistance, by altering the expression of mesenchymal and epithelial markers [41, 81]. Furthermore, miR-146a-5p promotes angiogenesis and increases cell proliferation by targeting.

CDKN1A (encoding p21), a cell cycle regulator, leading to increased cell cycle progression [41].

Crucially, miR-146a-5p is highly enriched in exosomes derived from trastuzumab-resistant cells and can be horizontally transferred to sensitive cells, thereby partially transmitting trastuzumab resistance [41]. While direct experimental evidence in the provided literature primarily focuses on the effects of miR-146a-5p on breast cancer cells and endothelial cells, external research cited suggests that miR-146a-5p can also promote invasion and metastasis by activating cancer-associated fibroblasts (CAFs) in the tumor microenvironment via exosomes generated from breast cancer cells [41, 82]. This establishes a clear link between a specific miRNA, exosomal transfer, and HER2-targeted therapy resistance, with a suggested role for CAF activation, forming a self-sustaining loop that drives acquired resistance and facilitates metastatic spread. This axis, where HER2-positive cancer cells, upon encountering therapy, adapt by releasing specific miRNAs via exosomes that can reprogram surrounding normal fibroblasts into CAFs or directly transfer resistance-promoting signals to other cancer cells, represents a critical vulnerability. Targeting the production, cargo, or uptake of these resistance-driving exosomal miRNAs could be a potent strategy to prevent or reverse acquired resistance, potentially re-sensitizing tumors to existing HER2-targeted therapies.

Exosomal miR-1246 and miR-155

Beyond miR-146a-5p, other exosomal miRNAs have been identified as significant players in HER2-positive breast cancer resistance. Exosomal miR-1246 and miR-155 are significantly upregulated in trastuzumab-resistant HER2-positive breast cancer patients [79]. These miRNAs serve as valuable predictive and prognostic biomarkers, strongly correlating with poor event-free survival (EFS) for early-stage patients and poor progression-free survival (PFS) for metastatic patients [79]. Their elevated levels indicate their direct involvement in resistance mechanisms or their presence as indicators of a resistance-prone microenvironment. The ability to detect these specific resistance-associated exosomal miRNAs in circulation means they can serve as non-invasive, real-time indicators of treatment response and emerging resistance, potentially even before clinical progression. This opens a new paradigm for personalized medicine in HER2-positive breast cancer, where liquid biopsies based on exosomal miRNA profiling could allow for earlier intervention and adaptive treatment strategies.

Other miRNAs influencing the HER2 pathway

While not directly linked to CAFs in the provided information, several miRNAs directly modulate HER2 signaling, and their dysregulation could indirectly affect CAF interactions or resistance. For example, miR-205 directly targets HER3, inhibiting AKT phosphorylation and suppressing clonogenic potential in HER2-overexpressing cells; its downregulation by HER2 is essential for tumorigenesis [80, 83]. Conversely, miR-342-5p inhibits HER2 signaling and cell growth and is associated with better survival in HER2-positive patients [55]. miR-21 is upregulated by HER2 signaling, contributes to cell invasion, and is linked to trastuzumab resistance [80]. These miRNAs, by modulating the core HER2 signaling pathway, can alter the tumor cell’s signaling output to the TME, thereby indirectly influencing CAF activation or response to therapy (Fig. 4).

Fig. 4.

Schematic representation of the miRNA–CAF axis in HER2-positive breast cancer resistance to HER2-targeted therapies. Left panel (miR-146a-5p): Trastuzumab-resistant HER2-positive breast cancer cells overexpress miR-146a-5p, promoting epithelial-to-mesenchymal transition (EMT), angiogenesis, and proliferation by targeting CDKN1A (p21). miR-146a-5p is enriched in exosomes and can be transferred to sensitive cells or fibroblasts, activating cancer-associated fibroblasts (CAFs). Middle panel (CAF communication): CAFs, once activated, contribute to a feedback loop that supports resistance, tumor progression, and metastasis through paracrine signaling and exosomal miRNA transfer. Right panel (Resistance-associated miRNAs): Additional exosomal miRNAs (miR-1246, miR-155) are elevated in resistant patients and serve as biomarkers of poor prognosis. Other miRNAs such as miR-205, miR-342-5p, and miR-21 directly modulate HER2/HER3 signaling, AKT phosphorylation, and clonogenic potential, indirectly influencing CAF activity and resistance phenotypes

Therapeutic implications and future directions

The elucidation of the reciprocal miRNA-CAF axis described above is not merely of academic interest; it exposes critical vulnerabilities that can be translated into clinical practice. Current HER2-targeted therapies, such as trastuzumab and lapatinib, focus almost exclusively on extinguishing oncogenic signaling within the epithelial tumor cell. However, as detailed in Sect. 6, the stromal compartment frequently provides a reservoir of resistance via exosomal miRNAs. Therefore, a paradigm shift is required: moving from a ‘cancer-centric’ treatment model to a ‘TME-inclusive’ strategy. This section outlines how targeting the miRNA-mediated communication network offers a novel frontier to overcome the persistence of HER2-positive disease.

Targeting miRNA-CAF interactions for HER2-positive breast cancer treatment

Strategies for targeting the miRNA-CAF axis can be broadly categorized into direct miRNA modulation, CAF-specific therapies, and exosome-based interventions.

Direct miRNA modulation

This approach involves directly altering the expression levels of specific miRNAs to restore tumor-suppressive functions or inhibit oncogenic ones. For instance, delivering synthetic miRNA mimics, such as miR-342-5p, miR-205, or miR-125, could inhibit HER2 signaling or sensitize HER2-positive cells to existing therapies [84, 85]. Conversely, anti-miRNAs (antagomiRs) can be used to inhibit oncogenic miRNAs like miR-146a-5p, miR-1246, or miR-155, which are associated with trastuzumab resistance [41].

CAF-targeting therapies

Recognizing the multifaceted contributions of CAFs, therapeutic strategies can aim to neutralize their pro-tumorigenic effects. Approaches include CAF depletion, inhibition of their signal transduction pathways, or, more subtly, CAF normalization [13]. The goal of CAF normalization is to revert their pro-tumorigenic phenotype to a more quiescent or even anti-tumorigenic state, thereby disrupting their ability to produce excessive ECM, secrete pro-tumorigenic factors, or modulate the immune system [13]. Combining CAF-targeting therapies with conventional treatments has the potential to improve drug penetration into the tumor and enhance T-cell infiltration, overcoming physical and immunosuppressive barriers to therapy [13].

Exosome-based interventions

Given the crucial role of exosomes in mediating intercellular communication and transferring resistance-driving miRNAs, targeting exosomes presents a sophisticated therapeutic strategy. Inhibiting the secretion or uptake of specific resistance-associated exosomal miRNAs, such as miR-146a-5p, miR-1246, or miR-155, could prevent the spread of resistance within the tumor microenvironment [41]. Conversely, engineered exosomes could be utilized as highly sophisticated delivery vehicles for therapeutic miRNAs or anti-miRNAs, offering targeted delivery to specific cells within the TME, thereby overcoming current delivery challenges associated with synthetic miRNA constructs [58, 86]. This approach leverages the natural ability of exosomes to carry and deliver biological messages, transforming them from carriers of pathological signals into precise therapeutic agents.

Challenges and opportunities in miRNA-based therapies

Challenges to clinical translation

Despite their promise, four major barriers currently stall the translation of exosomal miRNA therapeutics into the clinic for HER2-positive patients.

1. Delivery and stability Naked miRNAs are rapidly degraded by serum nucleases. While exosomes offer protection, they are cleared rapidly by the reticuloendothelial system (RES) (liver/spleen), preventing them from reaching the tumor site [87].

2. Manufacturing and scalability Producing clinical-grade exosomes with consistent miRNA cargo is technically challenging. Current isolation methods (e.g., ultracentrifugation) are difficult to scale for mass production [58].

3. The ‘Blood-Brain Barrier’ (BBB) problem For HER2-positive patients prone to brain metastasis, standard miRNA mimics cannot cross the BBB. Engineered exosomes expressing brain-targeting peptides (e.g., RVG peptide) are currently under investigation to overcome this physical barrier [88].

4. Off-target effects A single miRNA (e.g., miR-21) targets hundreds of transcripts. High-dose mimics may inadvertently silence tumor suppressors in healthy tissues, necessitating highly specific delivery vehicles to minimize systemic toxicity.

However, significant opportunities exist. The multi-targeting capability of miRNAs, where a single miRNA can modulate multiple genes often within the same signaling pathway, presents an attractive advantage for simultaneously influencing several pathways involved in HER2-positive breast cancer progression and resistance [20]. This inherent ability to broadly impact cellular networks could lead to more robust and durable therapeutic responses compared to targeting single genes. Furthermore, the field requires more standardized miRNA profiling approaches and comprehensive meta-analyses in large, well-balanced patient cohorts to establish a clearer consensus on miRNA signatures associated with HER2-positive breast cancer and its resistance mechanisms [80, 89].

The complexity of the tumor microenvironment and the reciprocal nature of miRNA-CAF interactions strongly suggest that monotherapy, even with highly effective agents like trastuzumab, is often insufficient to achieve durable responses. Therefore, future therapeutic strategies for HER2-positive breast cancer must adopt combination approaches [90]. This could involve combining existing HER2-targeted agents with therapies that normalize CAF function, disrupt exosomal miRNA communication, or directly modulate resistance-associated miRNAs. This multi-modal approach is essential to overcome the adaptive capabilities of the tumor and its microenvironment, moving beyond single-target interventions towards more comprehensive strategies (Table 2).

Table 2.

Therapeutic strategies targeting the miRNA-CAF axis in HER2-Positive breast cancer

Strategy type	Specific miRNA/target/pathway	Mechanism of action	Potential benefits	Challenges/considerations	References
miRNA Mimic Delivery	miR-342-5p, miR-205, miR-125	Restore tumor suppressor function, Inhibit HER2 signaling, Sensitize cells	Sensitize to HER2 therapy, Reduce proliferation, Suppress clonogenic potential	Delivery efficiency, Off-target effects, Stability, Immunogenicity	[84]
Anti-miRNA Therapy	Anti-miR-146a-5p, Anti-miR-1246, Anti-miR-155	Inhibit oncogenic miRNA function, Block resistance signaling	Overcome acquired resistance, Reduce metastasis, Re-sensitize to trastuzumab	Delivery efficiency, Off-target effects, Stability, Immunogenicity	[41]
CAF Normalization	CAF activation pathways	Revert CAF pro-tumorigenic phenotype	Improve drug penetration, Enhance T-cell infiltration, Reduce ECM barrier	Identifying specific CAF subtypes, Sustaining normalization, Potential for unintended effects	[11]
Exosome Inhibition	Exosome biogenesis/secretion/uptake of resistance-driving miRNAs	Block resistance transfer, Disrupt TME communication	Prevent acquired resistance, Reduce metastatic dissemination	Specificity of inhibition, Systemic effects, Delivery of inhibitors	[41]
Engineered Exosome Delivery	Therapeutic miRNA cargo (e.g., miR-342-5p mimic)	Deliver therapeutic cargo specifically to target cells (e.g., CAFs, cancer cells)	Targeted therapy, Reduced systemic toxicity, Overcome delivery challenges	Engineering complexity, Scalability, Immunogenicity of engineered exosomes	[58]

Conclusion

This comprehensive review underscores the pivotal and multifaceted role of the miRNA-CAF axis in shaping the HER2-positive breast tumor microenvironment, driving disease progression, fostering metastasis, and, critically, mediating resistance to HER2-targeted therapies. Cancer-associated fibroblasts are not merely passive bystanders but dynamic, heterogeneous entities actively reprogrammed by tumor cells, often through the sophisticated exchange of microRNAs encapsulated within exosomes. These miRNAs serve as key mediators of a complex, reciprocal communication network, establishing a self-sustaining feedback loop that promotes tumor survival and therapeutic evasion.

The detailed examination of specific miRNA-CAF interactions, such as the role of exosomal miR-146a-5p in transmitting trastuzumab resistance and activating CAFs, or the prognostic value of circulating exosomal miR-1246 and miR-155, highlights critical vulnerabilities within this axis. The pervasive influence of the TME, including the immune microenvironment, on HER2-positive breast cancer outcomes further emphasizes that a cancer cell-centric view is insufficient to fully address the challenges of this aggressive disease.

The immense potential of targeting these intricate miRNA-CAF interactions is clear. Strategies ranging from direct miRNA modulation and CAF normalization to advanced exosome-based interventions offer promising avenues for novel, more effective, and personalized therapeutic approaches. Furthermore, the stability and detectability of exosomal miRNAs in circulation present an exciting opportunity for developing non-invasive, real-time “liquid biopsy” biomarkers for dynamic monitoring of treatment response and early detection of emerging resistance. By disrupting this vicious cycle of mutual reprogramming and leveraging the communication pathways within the TME, the scientific community can move towards overcoming the persistent challenge of HER2-positive breast cancer resistance and ultimately improve patient outcomes.

Author contributions

Mustafa T. Ardah, Waleed K. Abdulsahib, Hasanain Amer Naji, S. Renuka Jyothi, Samir Sahoo, J. Bethanney Janney, Vipasha Sharma, Aashna Sinha, and Mohigul Kholiyeva contributed to the conception, design, and drafting of the manuscript. All authors read and approved the final version of the manuscript.

Funding

Not applicable.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.