The hypoxic nature of the TME is a critical pathological feature that drives angiogenesis and the development of drug resistance. As tumors proliferate rapidly, the local oxygen partial pressure decreases significantly, leading to the stabilization and activation of HIF-1α (26). As a master regulator of the hypoxic response, HIF-1α promotes the formation of disorganized and dysfunctional tumor vasculature by upregulating pro-angiogenic factors such as VEGF (27). This aberrant vascular network not only impedes effective drug delivery but also supports the survival and stemness of cancer stem cells (CSCs), thereby enhancing chemoresistance (28). Meanwhile, hypoxia facilitates the infiltration of immunosuppressive cells, including myeloid-derived suppressor cells (MDSCs) and regulatory T cells, while impairing the function of cytotoxic T cells, which compromises the response to immunotherapy (29). Moreover, the hypoxia-HIF-α axis reprograms cellular metabolism and induces epigenetic remodeling, further conferring tolerance to chemotherapy and radiotherapy in breast cancer cells (30). It is noteworthy that the expression and activity of HIF-1α are regulated by multiple signaling pathways, including PI3K/AKT/mTOR and MAPK/ERK (31). Therefore, targeting these upstream pathways or directly inhibiting HIF-1α, in combination with anti-angiogenic and immunomodulatory strategies, may represent a promising therapeutic approach to overcome hypoxia-associated treatment resistance (Fig. 2).

The TME in breast cancer constitutes a complex ecosystem composed of tumor cells, immune cells, cancer-associated fibroblasts (CAFs), and the extracellular matrix (ECM), among other components. Through their interactions, these elements suppress immune responses, thereby promoting tumor progression and conferring drug resistance (32). For instance, the infiltration of immunosuppressive cells, such as CAFs, tumor-associated macrophages, and MDSCs, can secrete inhibitory cytokines that impair the activity of cytotoxic T cells, leading to resistance to immunotherapy (33,34). Furthermore, metabolic competition and immunosuppression contribute to therapeutic resistance. Due to metabolic reprogramming, breast cancer cells compete with T cells for nutrients such as glucose, amino acids, and lipids. This process results in the accumulation of metabolites including lactic acid and kynurenine, which suppress T-cell activity. The combined effects of metabolic stress and metabolite-induced immunosuppression significantly enhance immunotherapy resistance in breast cancer (35). Abnormal immune checkpoint activation further promotes tumor evasion. The binding of PD-L1 on tumor cells to PD-1 on T cells inhibits T-cell activation and facilitates immune escape. Clinical studies show that blocking the PD-1/PD-L1 interaction can restore T-cell function (36). ECM-mediated physical barrier also contributes to resistance. Excessive ECM deposition not only impedes drug penetration but also promotes tumor cell survival by activating the PI3K/AKT pathway via integrin signaling. Targeting ECM remodeling can improve drug delivery and restore immune cell infiltration (37). Therefore, developing combination strategies targeting the immunosuppressive microenvironment, such as ICIs combined with metabolic modulators or ECM targeting agents, may provide a novel approach to overcome drug resistance in breast cancer. A comprehensive analysis of the dynamic interaction network among various components within the tumor microenvironment will establish a theoretical foundation for developing more effective therapeutic strategies (Fig. 2).

Breast CSCs (BCSCs), a subpopulation within tumors endowed with self-renewal and multilineage differentiation potential, are a major driver of chemotherapy and targeted therapy resistance due to their enrichment and expansion (38). The maintenance of BCSC stemness is mediated by aberrant activation of multiple signaling pathways. For example, activation of the PI3K/AKT/mTOR axis enhances BCSC stemness, thereby conferring therapy resistance (39). Activation of the Wnt/β-catenin signaling pathway induces epithelial-mesenchymal transition (EMT) to promote metastasis and confers resistance to endocrine therapy. Notably, studies have shown that inhibiting this pathway, in combination with CDK4/6 inhibitors, can partially reverse this resistance (40,41). The stemness of BCSCs is also regulated by epigenetic mechanisms. For instance, aberrant expression of long non-coding RNAs (lncRNAs) and alterations in DNA methylation patterns promote the stem cell-like phenotype, thereby reducing treatment efficacy (42). Moreover, inflammatory factors in the TME contribute to stemness maintenance by inhibiting apoptosis, promoting drug efflux protein expression, and enhancing inherent stem-like properties, thereby reinforcing therapy resistance (43). Several therapeutic strategies targeting BCSCs have been developed, including chimeric antigen receptor T-cell (CAR-T) therapy, ICIs, and approaches using BCSCs as drug delivery vehicles. However, these treatments still face challenges such as insufficient targeting specificity, acquired resistance, and safety concerns. Future efforts should focus on in-depth characterization of BCSCs, optimization of treatment strategies, and enhanced clinical validation to improve therapeutic outcomes (44) (Fig. 2).

DNA methylation plays a pivotal role in breast cancer drug resistance. Aberrant DNA methylation patterns can lead to the silencing of tumor suppressor genes or activation of drug resistance-associated genes, thereby altering cellular response to therapy (45). Research has demonstrated a significant correlation between the mRNA expression of drug resistance genes and their DNA methylation status, with variations observed across different molecular subtypes (46). In ER-positive breast cancer, aberrant DNA methylation was shown to be closely associated with resistance to endocrine therapy (47). The DNA methyltransferase inhibitor decitabine has been shown to reverse epigenetic alterations in drug-resistant cells, restoring their sensitivity to chemotherapeutic agents (48). Protein arginine methyltransferases (PRMTs) are fundamental epigenetic enzymes; pharmacological inhibition of PRMTs has been shown to markedly sensitize tumors to diverse anticancer therapies, supporting the potential of combination strategies with conventional agents to circumvent treatment resistance (49). In addition, dysregulation of m6A RNA methylation regulators may contribute to tumor progression. Although their precise role in DNA repair remains incompletely elucidated, preliminary evidence suggests a potential influence on cancer drug resistance (50). DNA methylation-based biomarkers have the potential to identify patients who respond to platinum-based chemotherapy and may offer novel therapeutic targets to combat tumor resistance. However, research in this field continues to face several challenges. The heterogeneity of DNA methylation patterns across distinct breast cancer subtypes, the complexity of the DNA methylation regulatory network, and the limited selectivity of current epigenetic drugs pose significant challenges. The development of liquid biopsy technology has provided new opportunities for the clinical translation of DNA methylation biomarkers by enabling the simultaneous assessment of tumor gene mutations, resistance genes, and DNA methylation status through detection of circulating tumor DNA in blood, thereby offering multidimensional evidence for precision breast cancer therapy (51). Future studies should focus on elucidating the crosstalk between DNA methylation and other epigenetic mechanisms, such as histone modifications, to facilitate the development of more effective combination treatment regimens (Fig. 3).

Histone modifications refer to various chemical changes, such as methylation, acetylation, phosphorylation, and ubiquitination, that occur on histone proteins. These modifications regulate gene transcription by altering chromatin structure or recruiting modifier proteins, and they play a critical role in epigenetic regulation. Dysregulation of histone modifications is a core epigenetic mechanism underlying acquired drug resistance in breast cancer. In ER-positive breast cancer, aberrant histone H3 modifications, such as H3K27 acetylation, are associated with resistance to endocrine therapy (52). The combination of histone deacetylase inhibitors (HDACi; such as vorinostat) with conventional chemotherapeutic drugs significantly enhances treatment sensitivity in resistant cells (53). Furthermore, following chemotherapy for breast cancer, OTULIN (a deubiquitinating enzyme) stabilizes β-catenin via deubiquitination, activating the Wnt/β-catenin pathway. This activation promotes DDR and metastasis, suggesting that inhibition of this pathway may enhance chemosensitivity (54). The FDA has approved several HDACi, including romidepsin, belinostat, panobinostat, and vorinostat, for clinical use. In addition, small-molecule inhibitors targeting histone acetyltransferases are under clinical investigation (45). In summary, histone modifications drive drug resistance by regulating gene expression, chromatin architecture, and signaling pathway activity. Developing small-molecule inhibitors against specific histone-modifying enzymes, either alone or in combination with other epigenetic drugs, may provide novel therapeutic strategies to overcome drug resistance in breast cancer (Fig. 3).

ncRNAs, as crucial players in epigenetic regulation, play key roles in mediating treatment resistance in breast cancer. Diverse ncRNAs, including microRNAs, lncRNAs, and circular RNAs (circRNAs) contribute to resistance by regulating gene expression, modulating the activation of signaling pathways, and remodeling the TME (55,56). For example, in ER-positive breast cancer, ~40% of patients relapse due to acquired tamoxifen resistance, a process involving ncRN-mediated regulation of resistance-related genes (57). HOX transcript antisense intergenic RNA, a well-characterized oncogenic lncRNA, promotes breast cancer metastasis and chemoresistance through epigenetic modifications, regulation of target genes, and activation of signaling pathways (58). Certain lncRNAs enhance chemoresistance by upregulating ABC transporter expression, which reduces intracellular drug accumulation and impairs the efficacy of agents such as gemcitabine (59). Moreover, aberrant expression of circRNAs has been shown to promote tumor cell invasiveness and drug resistance by regulating EMT-related genes (60). Thus, ncRNA expression patterns show promise as biomarkers for predicting the risk of resistance and as potential therapeutic targets. Although antisense oligonucleotide-based interventions have demonstrated efficacy in restoring chemosensitivity, challenges such as a lack of assay standardization and off-target effects have impeded their translation into routine clinical use (61). Future research should focus on elucidating ncRNA-mediated crosstalk within the immune microenvironment, developing highly specific ncRNA antagonists to overcome endocrine resistance through personalized therapeutic approaches, and integrating multi-omics data to accelerate translational applications (62) (Fig. 3).

HRR deficiency is closely associated with drug resistance mechanisms in breast cancer, primarily involving restoration of homologous recombination function, alterations in DNA replication fork stability, aberrant epigenetic regulation, and compensatory activation of alternative DNA repair pathways. As core components of the HRR pathway, loss-of-function mutations in BRCA1/2 sensitize tumor cells to PARPi and platinum-based agents. However, clinical observations indicate that restoration of HRR function may lead to acquired resistance (63). Specific mechanisms of resistance include epigenetic regulation-mediated restoration of HRR, exemplified by demethylation of BRCA gene promoters, which restores BRCA protein expression and reactivates the HRR pathway (64) and compensatory upregulation of key DNA repair proteins. For instance, loss of BRCA1/2 induces compensatory activation of RAD51 recombinase (RAD51), which bypasses HRR defects to restore double-strand break (DSB) repair capacity, thereby conferring chemotherapy resistance (65). Loss of 53BP1 partially restores HRR in BRCA-deficient cells by modulating DNA end resection, thereby enabling escape from PARPi-induced synthetic lethality (66,67). Dysfunction of HRR-associated proteins such as partner and localizer of BRCA2 (PALB2), which works together with BRCA1/2 and RAD51 to mediate HRR, may also lead to HRR deficiency and sensitize tumors to PARPi. For example, loss of PALB2 similarly disrupts HRR and increases sensitivity to PARPi. However, certain PALB2 mutations may impair its binding to BRCA1/2, resulting in partially retained HRR function and subsequent drug resistance. Targeting the PALB2 recruitment machinery may thus represent a promising therapeutic strategy for BRCA1-mutated tumors (68–70). Clinical data have demonstrated that combining PARPi with chemotherapeutic agents, such as paclitaxel and gemcitabine, or immunotherapeutic agents, such as pembrolizumab, shows potential in overcoming drug resistance (71). Future precision therapies targeting HRR deficiency are expected to expand the eligible patient population in breast cancer. Further investigation of pathways that crosstalk with the HRR network may reveal novel targets for overcoming PARPi resistance. In clinical practice, molecular profiling of HRR gene alterations could enable early identification of patients at high risk of treatment resistance and guide personalized therapeutic strategies (Fig. 4).

NHEJ is a primary pathway for repairing DSBs in mammalian cells. Upon the occurrence of DSBs, the Ku70/80 heterodimer is rapidly recruited to the broken ends and binds to them in a sequence-nonspecific manner. This recruitment is followed by the assembly and activation of the DNA-dependent protein kinase catalytic subunit, facilitating synapsis of the broken DNA ends. Subsequently, end-processing factors, including the Artemis nuclease, are engaged to prepare the termini for ligation. Ultimately, the X-ray repair cross-complementing (XRCC) 4-like factor complex, in conjunction with DNA ligase IV, catalyzes the final ligation step to complete the repair process (72–74). The repair process frequently introduces base deletion or insertion mutations, leading to reduced sequence fidelity. In breast cancer, core NHEJ components are frequently upregulated. This elevation is regarded as a compensatory adaptation to genomic instability or impairments in other DSB repair pathways, thereby conferring resistance to radiotherapy and numerous chemotherapeutic agents (75,76). Additionally, the alternative non-homologous end joining (Alt-NHEJ) pathway is mediated by key factors such as DNA ligase III, DNA polymerase θ, and PARP1. Their upregulation is closely associated with tumor progression and drug resistance. When canonical NHEJ is impaired, Alt-NHEJ is activated as a backup pathway to maintain genome integrity, paradoxically promoting drug resistance (77). The aforementioned evidence indicates that NHEJ and its alternative pathways contribute to drug resistance in breast cancer via a multi-molecular interaction network. Targeting key factors in the NHEJ pathway may represent a promising strategy to overcome NHEJ-mediated resistance (Fig. 4).

The BER pathway serves as a primary cellular defense mechanism against DNA single-base lesions and single-strand breaks, primarily repairing damage caused by both endogenous and exogenous factors such as reactive oxygen species (ROS), alkylating agents, and ionizing radiation (78–80). First, chemotherapy-induced DDR is considered. The BER pathway effectively reverses DNA lesions induced by chemotherapeutic agents such as cisplatin, significantly reducing the cytotoxic efficacy of these drugs against cancer cells (81). Flap endonuclease 1 (FEN1), a key enzyme in BER, is often overexpressed in TNBC and confers resistance to chemotherapeutic agents. Inhibition of FEN1 has been shown to sensitize tumor cells to treatment (82). Second, the influence of genetic variation, such as single-nucleotide polymorphisms in BER-related genes, including XRCC1, may alter BER efficiency and consequently affect treatment response and prognosis in patients with breast cancer (83). Currently, no specific therapeutic agents targeting the BER pathway are available beyond PARPi. Therefore, systematic identification of key biomarkers in the BER pathway, such as the expression level of DNA polymerase β, XRCC1 mutation status, and FEN1 activity, will contribute to the precise screening of patients likely to benefit, and provide a theoretical basis for developing novel targeted strategies (Fig. 4).

NER is an essential DNA repair mechanism in mammalian cells, dedicated to removing bulky DNA lesions that distort the helical structure, such as UV-induced cyclobutane pyrimidine dimers and DNA damage generated by chemotherapeutic agents. In breast cancer, aberrantly elevated NER activity efficiently removes cisplatin-induced DNA adducts and impedes damage accumulation, thereby mediating both intrinsic and acquired resistance to this chemotherapeutic agent (84,85). Excision repair cross-complementation (ERCC) group 1, a rate-limiting enzyme in the NER pathway, confers cellular resistance to platinum-based drugs when highly expressed due to enhanced DNA repair capacity (86). Further studies revealed that methylation of the ERCC4 gene promoter results in loss of its encoded product, xeroderma pigmentosum group F-complementing protein (XPF), thereby suppressing NER function and increasing cisplatin sensitivity. Conversely, restoring XPF expression reactivates NER and reduces cisplatin efficacy, suggesting that epigenetic modulation of ERCC4 to target NER activity holds therapeutic potential (87,88). Moreover, under specific contexts, BRCA1 facilitates the clearance of chemotherapy-induced DNA lesions and promotes genomic stability through co-activation of NER (89). However, the functional landscape of NER activity in breast cancer remains incompletely defined. Precise identification of NER deficiencies or hyperactivation is expected to guide the development of NER-targeting agents, thereby increasing chemosensitivity and overcoming drug resistance (Fig. 4).

The MMR system maintains genomic stability by recognizing and correcting base-base mismatches during DNA replication. In breast cancer chemoresistance, it mediates context-dependent resistance through bidirectional epigenetic reprogramming. On one hand, loss of MMR function, such as that induced by MutL Homolog 1 (MLH1) or MutS Homolog 2 (MSH2) gene silencing, causes microsatellite instability-high and may initially increase tumor sensitivity to DNA-damaging agents. However, sustained MMR deficiency ultimately promotes genomic instability. For example, in doxorubicin resistance, hypermethylation of the MSH2 promoter compromises MMR function, facilitates the accumulation of acquired mutations, and leads to secondary resistance (90–92). On the other hand, anthracycline-induced hypomethylation of MLH1 or MSH2 promoters can enhance MMR gene expression and repair activity. This enables cancer cells to excessively clear DNA damage, resulting in pan-chemoresistance (93). The MutL complex genes (MLH1, PMS1, PMS2, and MLH3) are core components of the MMR pathway. Defects in these genes disrupt MMR function, preventing ER-positive breast cancers from effectively suppressing CDK4 activity during endocrine therapy, thereby conferring treatment resistance. Consequently, MMR deficiency may serve as a predictive biomarker for response to CDK4/6 inhibitors (94). Furthermore, it has been shown that the HR and MMR pathways are functionally interconnected in DNA repair-deficient tumors and share key protein components. Although the precise mechanisms remain incompletely elucidated, this crosstalk offers novel insights and potential therapeutic targets for precision oncology (95). In summary, co-targeting complementary DNA damage response pathways, such as combining PARPi with MMR targeting strategies, represents a promising therapeutic approach to overcome drug resistance in breast cancer (Fig. 4).

Metabolic reprogramming in breast cancer cells is characterized by markedly enhanced glycolysis (the Warburg effect), which operates synergistically with the tricarboxylic acid cycle and the hexosamine biosynthesis pathway to supply energy and biosynthetic precursors that promote tumor cell survival, invasion, and maintenance of stemness under chemotherapy-induced stress (96). For instance, Pim-2 proto-oncogene, serine/threonine kinase was shown to bind to and promote phosphorylation of 6-phosphofructo-2-kinase at Ser478, thereby enhancing glycolytic activity and conferring paclitaxel resistance in breast cancer cells (97). Moreover, pyruvate kinase M2 was demonstrated to activate autophagy, supplying tumor cells with energy and metabolic intermediates that bolster survival advantage under drug pressure. Thus, the ‘glycolysis-autophagy axis’ represents a core mechanism underlying chemoresistance (98). Hexokinase 2 (HK2) also activates the NF-κB pathway, leading to upregulation of PD-L1 expression, which drives immune escape and influences tumor immune infiltration and patient prognosis. These findings suggest that combining PD-L1 blockade with HK2-targeted therapy may offer a novel therapeutic strategy for breast cancer (99). The end product of glycolysis, lactate, contributes to tumor progression by acidifying the tumor microenvironment, suppressing immune responses, and modulating oncogene expression (100). Therefore, developing specific inhibitors targeting key glycolytic enzymes and using them in combination with existing anticancer drugs may overcome therapy resistance by inhibiting tumor glycolytic metabolism and enhancing immune responses, thereby providing a novel strategy for breast cancer treatment (96) (Fig. 5).

Disorders in lipid metabolism contribute to treatment evasion in breast cancer cells by regulating fatty acid synthesis, cholesterol metabolism, and phospholipid remodeling, thereby providing a sustained energy supply and maintaining TME homeostasis (101,102). Key lipid metabolic enzymes are tightly regulated by transcription factors that drive aberrant synthesis and accumulation of fatty acids and cholesterol, thereby enhancing resistance to chemotherapy, endocrine therapy, and targeted therapy. Sterol regulatory element-binding protein 1 promotes fatty acid synthesis via Akt/mTOR activation, thereby increasing cell membrane fluidity and reducing drug influx, leading to doxorubicin resistance (103). Dysregulated cholesterol metabolism enhances chemoresistance by upregulating ABC transporter expression and facilitating lipid raft-mediated signaling (104). The ketone body-producing rate-limiting enzyme 3-hydroxy-3-methylglutaryl-CoA synthase 2 is highly expressed in tamoxifen-resistant breast cancer, and developing inhibitors against it represents a promising strategy for reversing endocrine resistance (105). Notably, lipid metabolic reprogramming also modulates EMT and stemness, remodels the immune microenvironment, and alters chemotherapeutic metabolism, collectively exacerbating drug resistance (106–108). This multi-layered and multi-targeted metabolic network highlights targeting key nodes of lipid metabolism as a potential therapeutic strategy to overcome drug resistance in breast cancer (Fig. 5).

Amino acids serve as essential nutrients for sustaining cellular life. Metabolic reprogramming of amino acids has been demonstrated to significantly promote the proliferation, metastasis, and therapy resistance of breast cancer cells, a process closely associated with dysregulated expression of amino acid transporters and altered activity of key enzymes. Research has revealed that solute carrier family 7 member 5 (SLC7A5), an amino acid transporter, is highly expressed across different breast cancer subtypes, particularly in Luminal B breast cancer, where its elevated expression is associated with poor patient prognosis, suggesting its potential as a therapeutic target in this subtype (109). In TNBC, chemotherapy-induced reactive ROS stress prompts cancer cells to enhance glutamine metabolism by upregulating glutaminase (GLS) and glutamate transporters, leading to massive glutamate efflux. This facilitates GSH synthesis to scavenge ROS and maintain redox homeostasis, thereby driving chemoresistance. Targeting glutamate depletion or inhibiting GLS and the cystine/glutamate transporter SLC7A11 (xCT) disrupts this antioxidant system, resulting in ROS accumulation and resensitization to treatment (110,111). Amino acid metabolism also interfaces with the TME: Tumor cells compete with immune cells, such as T cells, for key amino acids such as tryptophan and arginine, supporting tumor survival and proliferation. Meanwhile, accumulation of the tryptophan metabolite kynurenine suppresses T-cell function and fosters an immunosuppressive TME, ultimately compromising immunotherapy efficacy and indirectly contributing drug resistance in breast cancer (112,113). In summary, targeting amino acid transporters, such as with SLC7A5 inhibitors, in combination with metabolic inhibitors and immunotherapy, along with metabolism-based precision therapy, represents a promising strategy for overcoming drug-resistant breast cancer (Fig. 5).

Nucleotides function as the fundamental building blocks for DNA/RNA synthesis and participate in cellular energy metabolism, signaling transduction, and proliferation regulation. In breast cancer, upregulation of rate-limiting enzymes in the de novo nucleotide synthesis pathway promotes excessive purine and pyrimidine accumulation, which activates downstream signaling cascades, enhances tumor stemness and metastatic potential, and facilitates DDR, ultimately leading to chemoresistance (114,115). For instance, elevated expression of the purine metabolic enzyme phosphoribosylaminoimidazole succinocarboxamide synthetase enhances ERα activity through the cyclic adenosine monophosphate-protein kinase A-mTOR signaling axis, resulting in tamoxifen resistance in breast cancer (116). Furthermore, nucleotide metabolic reprogramming contributes to therapy resistance by modulating the TME. Tumor cells consume nucleotide precursors and release immunosuppressive metabolites, leading to T-cell dysfunction through nutrient competition and subsequent immune escape (117,118). Therefore, targeting nucleotide metabolism in combination with conventional therapies may offer a novel multi-target strategy to overcome drug resistance in breast cancer by synergizing metabolic intervention with immune modulation (Fig. 5).