Antibody-drug conjugates in breast cancer: from mechanism to revolutionizing clinical practice Authors: Shuyu Li, Zhouming Xu, Wenhui Ruan, Xiqing Wang, Hongmei Yu, Ming Yi, Peifen Fu Published: 2026 ABSTRACT AbstractThe advent of antibody-drug conjugates (ADCs) has caused a paradigm shift in breast cancer management, transforming the treatment landscape from advanced lines to curative-intent settings. By integrating the specificity of monoclonal antibodies with the potency of cytotoxic payloads, ADCs have broken the limitations of traditional chemotherapy. This review aims to provide a comprehensive overview of the current state of ADCs in breast cancer, concentrating on targets under development, toxicity, and the challenges that remain to be overcome. Distinct from conventional classifications, we propose a framework categorizing ADC targets into three dimensions based on biological features: oncogenic driver antigens, lineage and oncofetal antigens, and tumor microenvironment antigens. Meanwhile, we discuss the structural evolution of ADCs, the management of toxicity, and the mechanism-based resistance. The future of treatment calls for precise ADC management, the use of next-generation bispecific ADCs and immune-oncology combinations, and the imperative of biomarker-guided sequencing. Keywords: Antibody-drug conjugates, Breast cancer, Molecular targets, Tumor microenvironment, Drug resistance, Toxicity management, Bystander effect, Combination therapy Introduction As the second most common cancer overall and the most frequently diagnosed malignancy in women, breast cancer (BC) stands as a significant global health challenge and the leading cause of cancer-related death among women. In 2022, there were approximately 20 million new cancer cases worldwide, among which 9.7 million were females, and breast cancer accounted for 2.31 million cases, causing 666,000 deaths globally in 2022 [1]. While substantial progress has been made in early detection and multimodal treatment, including surgery, radiotherapy, chemotherapy, endocrine therapy, and targeted agents, the management of advanced, metastatic, and treatment-resistant disease continues to cause difficulties in clinical management [2, 3]. Heterogeneity represents the primary obstacle in breast cancer treatment. Traditional treatments depend on the expression of hormone receptor (HR) and human epidermal growth factor receptor 2 (HER2). While effective, this framework fails to fully capture the disease’s complexity or meet all patient needs, particularly in triple-negative breast cancer (TNBC) [4]. The integration of genomic and proteomic analyses has not only discovered numerous novel tumor-associated antigens (e.g., TROP2, HER3, LIV-1, B7-H4) with therapeutic potential, but also refined molecular subtyping [5, 6]. This expansion of targetable molecules has directly promoted therapeutic innovation while driving the development of antibody-drug conjugates (ADCs), heralding a new era in targeted treatment. The concept of ADC was put forward a century ago by Paul Ehrlich, as a “magic bullet” capable of selectively delivering cytotoxic agents to tumor cells [7]. Modern ADCs have realized this vision, consisting of a monoclonal antibody (mAb) specific to a tumor-associated antigen, a stable chemical linker, and a potent cytotoxic payload [8]. This architecture aims to maximize anti-tumor efficacy by precisely delivering the payload to cancer cells while minimizing the off-target toxicity compared to conventional chemotherapy. Breast cancer has been at the forefront of ADC development and application. Trastuzumab emtansine (T-DM1) paved the way as the first ADC approved for a solid tumor, validating its effect in HER2-positive breast cancer [9]. The subsequent emergence of next-generation ADCs, such as trastuzumab deruxtecan (T-DXd) [10, 11] and sacituzumab govitecan (SG) [12, 13], has further revolutionized the treatment. As the efficacy of T-DXd in HER2-low breast cancer has been verified, it has prompted a redefinition of HER2 classification in breast cancer, enlarging the patient population eligible for HER2-targeted therapy [11, 14]. ADCs targeting other antigens, such as TROP2, have also provided new therapeutic options for TNBC and expanded the treatment for other breast cancer subtypes [15–17]. ADC Structure and mechanism of action Structure of ADCs A classical ADC consists of three key elements: a mAb that targets tumor-specific or tumor-associated antigens, a cytotoxic payload capable of effectively killing tumor cells, and a chemical linker (either cleavable or non-cleavable) that connects these two parts (Fig. 1) [7]. This sophisticated combination enables ADCs to precisely and effectively kill tumor cells without harming normal cells. Fundamental structure and next-generation optimization strategies of ADCs. The core ADC structure consists of a monoclonal antibody (mAb), a chemical linker, and a cytotoxic payload. Current optimizations involve four dimensions: (1) Antibody Optimization: Utilizes biparatopic/bispecific designs (e.g., JSKN003) and dual-epitope/dual-antigen targeting for enhanced specificity and internalization. (2) Linker Optimization: Employs enzyme-cleavable (e.g., T-DXd), acid-labile (e.g., SG with CL2A linker), and TME-specific (e.g., SHR-A1921, ICP-B794) mechanisms. (3) Payload Optimization: Involves Topo I inhibitors (e.g., DXd, SN-38), DNA alkylators or other novel classes, hydrophilic modifications, and dual-payload combinations. (4) Site-Specific Conjugation: Uses engineered cysteines (e.g., THIOMAB), glycoengineering, or enzymatic methods (e.g., mTG, FGE) to achieve homogeneous DAR and reduced toxicity. Abbreviations: ADC, antibody-drug conjugate; DAR, drug-to-antibody ratio; FGE, formylglycine-generating enzyme; mAb, monoclonal antibody; mTG, microbial transglutaminase; SG, sacituzumab govitecan; TME, tumor microenvironment Monoclonal antibody Antibodies, or immunoglobulins, are Y-shaped proteins generated by the immune system that play a critical role in precisely identifying and eliminating foreign invaders. An ideal therapeutic antibody should exhibit several essential characteristics: high specificity for the target antigen, low immunogenicity, strong binding affinity, and the ability to induce receptor-mediated internalization [18]. Produced from a single B-cell clone through asexual proliferation, mAbs offer exceptionally high homogeneity, purity, and specificity, making them suitable for targeted therapies [2]. Among the various antibody types, immunoglobulin G (IgG) serves as the most commonly used scaffold in ADCs. With a molecular weight of approximately 150 kDa, IgG demonstrates high antigen affinity, prolonged serum half-life, and contains multiple native binding sites that can be further engineered to introduce additional functional moieties [19]. IgG could be divided into four subclasses: IgG1, IgG2, IgG3, and IgG4. Among these, IgG1 and IgG3 exhibit stronger effector functions, including antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC), which can act synergistically with conjugated cytotoxic agents to enhance tumor cell killing. However, owing to its extremely short circulating half-life, IgG3 is rarely considered for the design as an antibody in ADCs. Given that the Fc fragment of IgG1 binds most strongly to the Fcγ receptor, IgG1 has become the predominant subclass used in clinical ADC development [20]. In ADCs, the mAb functions as a guidance system, delivering the cytotoxic payload precisely to the tumor by recognizing specific antigens that are only expressed on tumor cells. Several ADCs have been approved for treating breast cancer, primarily targeting two key antigens: HER2 and TROP2 [21]. The development of HER2-targeting ADCs is the most mature, encompassing both the pioneering T-DM1 [9] and the new-generation T-DXd [11], which are used across early and advanced disease settings. Meanwhile, TROP2-targeting ADCs, exemplified by SG [12], have provided an effective therapeutic option for managing refractory TNBC. Cytotoxic payload Cytotoxic payloads are the “warheads” that directly kill tumor cells. The limited and inefficient delivery of cytotoxic payloads to tumor cells necessitates the use of highly potent drugs (more powerful than traditional chemotherapies) in ADC design to achieve sufficient anti-tumor efficacy within a therapeutically viable window [18]. To ensure successful translation from a free drug to an effective conjugate, the payload must not only exhibit a validated mechanism of action (targeting core survival machineries such as DNA or tubulin), but also possess amenable chemical motifs for linker attachment while retaining biological activity upon intracellular release. Historically, the payloads have been dominated by microtubule inhibitors, which exert cytotoxicity by suppressing microtubule polymerization or depolymerization, thereby disrupting the cytoskeletal architecture and inducing mitotic arrest [22]. Representative agents, such as auristatins (e.g., MMAE, MMAF) and maytansinoids (e.g., DM1, DM4), have been successfully deployed in approved drugs. For instance, T-DM1 utilizes DM1 to dismantle the mitotic spindle, blocking cell division and precipitating apoptotic cell death. To overcome the limitations of antimitotic agents, particularly their dependence on cell division, recent advances have propelled the evolution of ADC technology toward DNA-damaging agents, including topoisomerase I inhibitors, DNA alkylating agents, etc. [23]. Topoisomerase I inhibitors, such as exatecan derivatives (DXd) and SN-38, have emerged as the milestone of next-generation ADCs. Distinct from microtubule inhibitors, these payloads could eliminate both proliferating and quiescent cells by inducing replication fork collision and lethal DNA double-strand breaks, usually coupled with high membrane permeability that facilitates a potent “bystander effect" [24]. DNA alkylating agents, however, are typically highly toxic and require more precise delivery to mitigate systemic toxicity [25]. Representative agents include calicheamicin, duocarmycins, and pyrrolobenzodiazepine dimers. Beyond established mechanisms, the selection of a payload is crucial for the overall efficacy and safety of the ADCs [26]. The hydrophobicity of the payload dictates the conjugate’s pharmacokinetics and aggregation tendency, while its charge profile influences clearance rates [27]. The “bystander effect”, which is the ability of the released payload to diffuse across membranes and eradicate adjacent antigen-negative tumor cells, is important in overcoming the intratumoral heterogeneity [24]. Current research focuses on novel therapeutic payloads to widen the therapeutic window, including RNA polymerase II inhibitors (e.g., α-Amanitin [28]) that induce apoptosis via transcription inhibition, proteolysis-targeting chimeras leveraging the ubiquitin-proteasome system, and immunostimulants, representing a shift from simple cytotoxicity to complex biological modulation. Linker In ADCs, the linker is the bridge connecting the cytotoxic payload and the antibody. Its core function is to maintain absolute stability during systemic circulation to minimize off-target toxicity, while enabling efficient cleavage and release of the cytotoxic payloads during the internalization process in tumor cells [27]. ADCs mainly utilize two classes of linkers: cleavable and non-cleavable. T-DM1 uses a non-cleavable succinimidyl-4‐(N‐maleimidomethyl)cyclohexane‐1‐carboxylate (SMCC) linker, which demonstrates tremendous stability in circulation, thereby significantly reducing systemic toxicity [29]. While this design offers outstanding safety, its major limitation is the charged nature of the released payload, which results in poor membrane permeability and a lack of bystander effect. Consequently, its efficacy is restricted to HER2-high tumor cells that can successfully internalize T-DM1. To overcome this problem, newer-generation ADCs have widely adopted cleavable linkers, which are designed to exploit special physiological features of the tumor cell for specific activation. According to the catalytic cutting method, cleavable linkers can be further classified into chemically cleavable linkers (such as hydrazine and disulfide bonds) and enzymatically cleavable linkers (such as glucuronic acid and peptide bonds) [30]. For instance, SG uses an acid-labile linker (CL2A) that is hydrolytically activated in the acidic environment of endosomes or lysosomes, leading to the release of its payload (SN-38) [13]. Unlike SG, T-DXd employs an enzyme-cleavable tetrapeptide linker, which is composed of four amino acids, Gly-Gly-Phe-Gly (GGFG). This linker is specifically cleaved by cathepsins, which are highly expressed in lysosomes. The advantage of this design lies in the excellent membrane permeability of the released DXd, allowing diffusion into adjacent antigen-low or antigen-negative tumor cells, thereby generating a potent bystander effect [31]. Due to its potent bystander effect, T-DXd not only shows great efficacy in HER2-positive breast cancer but also in HER2-low breast and heterogeneous breast cancer, thus disruptively redefining the clinical treatment [10, 11]. Conjugation technologies Serving as the nexus between the cytotoxic payload and the antibody, conjugation technology fundamentally decides the pharmacokinetic profile, therapeutic index, and physical stability of ADCs [27]. The evolution of conjugation technologies represents a paradigm shift in ADC development, from stochastic heterogeneity to molecular precision. First-generation ADCs (e.g., T-DM1) mainly adopt the stochastic conjugation that depends on the abundant naturally reactive groups on the antibody surface, such as the ε-amino groups of lysine residues or the sulfhydryl groups of partially reduced interchain cysteines [27, 32]. Despite the clinical validation of these non-specific strategies, stochastic conjugation inevitably produces a heterogeneous mixture of agents with different drug-to-antibody ratios (DARs), ranging from 0 to 8. This heterogeneity brings several challenges: high-DAR species tend to aggregate and be rapidly cleared due to high hydrophobicity, inducing off-target toxicity; low-DAR or unconjugated species competitively bind to tumor antigens without enough cytotoxicity, thereby lowering therapeutic efficacy [33, 34]. To avoid the above problems of stochastic conjugation, site-specific conjugation emerged as the “gold standard” for next-generation ADCs, ensuring precise payload attachment, enhanced stability, and homogeneous DARs [35]. Large amounts of the site-specific conjugation have been established. Genetic strategies, represented by THIOMAB™ technology, introduce a cysteine residue with chemical reactivity at a specific site of the antibody, enabling precise connection while preserving the structural integrity of the antibody [36]. Enzymatic platforms offer another alternative. For instance, SMARTag® technology utilizes formylglycine-generating enzyme (FGE) to install aldehyde handles on the antibody for specific payload attachment [37], while microbial transglutaminase (mTG) catalyzes the formation of stable isopeptide bonds between amine-containing payloads and specific glutamine residues on the antibody [38]. Furthermore, glycoengineering strategies, such as GlycoConnect™, target the native N-linked glycans (N297) in the Fc region based on copper-free click chemistry. The special advantage of this approach is the ability to achieve site-specific conjugation distal to the antigen-binding domain without altering the amino acid sequence [39]. Different from the above, affinity peptide-mediated chemistry exemplified by AJICAP™ technology utilizes removable guiding peptides to connect reactive groups with specific lysine residues, achieving regioselective conjugation without genetic or enzymatic modification [40, 41]. Mechanism of action The core mechanism of action of ADC drugs is that the mAb recognizes the target antigen and enters the cell through receptor-mediated endocytosis to release cytotoxins and kill cancer cells [42]. The detailed process is shown in Fig. 2. Structure and mechanism of action of antibody-drug conjugates (ADCs). The ADC complex is engineered with three core components: a monoclonal antibody (mAb), a chemical linker, and a cytotoxic payload. Upon systemic administration, the mAb selectively binds to tumor-associated antigens (e.g., HER2, TROP2), initiating receptor-mediated internalization. Within the acidic and enzyme-rich lysosomal compartment, the payload is liberated either via targeted linker hydrolysis (cleavable linkers; e.g., in T-DXd and SG) or complete antibody degradation (non-cleavable linkers; e.g., in T-DM1). The released cytotoxins subsequently disrupt core cellular machineries—such as DNA replication or microtubule assembly—to precipitate apoptosis. Crucially, membrane-permeable payloads can diffuse outward to eradicate adjacent antigen-negative or antigen-low cancer cells, overcoming intratumoral heterogeneity through the bystander effect. However, the diffusion could also cause side effects by killing the adjacent normal cells. Abbreviations: ADC, antibody-drug conjugate; HER2, human epidermal growth factor receptor 2; mAb, monoclonal antibody; SG, sacituzumab govitecan; T-DM1, trastuzumab emtansine; T-DXd, trastuzumab deruxtecan; TROP2, trophoblast cell surface antigen 2 Intracellular drug release. As macromolecular proteins, ADCs exhibit inefficient gastrointestinal absorption, necessitating predominant administration via the intravenous route. This route of administration offers superior efficiency and safety. Following intravenous delivery, ADCs distribute through the blood circulation. When the ADC protein reaches the tumor, the mAb component specifically binds to tumor-associated antigens (e.g., HER2, TROP2) overexpressed on the surface of cancer cells with high affinity, forming an ADC-antigen complex. After undergoing receptor-mediated endocytosis, this complex will be internalized into endosomes. Through endosomal maturation into lysosomes, the acidic pH and proteolytic enzymes (e.g., cathepsins) catalyze antibody degradation (e.g., T-DM1) or linker hydrolysis (as utilized in T-DXd and SG) to release cytotoxic payloads. Finally, the potent cytotoxic payloads induce the apoptosis of tumor cells by disrupting essential cellular processes, such as DNA replication or microtubule assembly, realizing anti-tumor efficacy [8]. Bystander effect. Beyond intracellular payload release, those permeable payloads (e.g., DXd, SN-38) are capable of diffusing across cell membranes into the extracellular space [43]. These agents can penetrate and kill adjacent tumor cells, thus enhancing ADC efficacy against heterogeneous tumors, where antigen expression is often spatially and temporally variable. By eliminating these antigen-negative or antigen-low tumor cell populations, the bystander effect further enhances the killing efficacy of ADCs. ADCs in breast cancer Since the approval of T-DM1 in 2013, ADCs have fundamentally revolutionized the therapeutic landscape of breast cancer. The paradigm has shifted swiftly from later-line therapy to frontline standards, and increasingly, into the curative intent settings of adjuvant and neoadjuvant therapy. Until 2025, seven ADCs have been approved by the Food and Drug Administration (FDA) or National Medical Products Administration (NMPA) for breast cancer indications (Fig. 3) [44, 45]. Timeline of FDA and NMPA approvals of antibody-drug conjugates for breast cancer treatment (2013–2026). The clinical evolution of ADCs highlights a rapid paradigm shift from late-line salvage therapies to curative-intent, early-stage settings. This chronological map details key milestones granted by the United States Food and Drug Administration (FDA) and the China National Medical Products Administration (NMPA). Top: HER2-targeted agents, pioneered by T-DM1, have advanced into the adjuvant setting for residual disease. Furthermore, the advent of T-DXd fundamentally redefined molecular subtyping by demonstrating robust efficacy in HER2-low and -ultralow cohorts. Recent domestic approvals (RC48, A166, and SHR-A1811) underscore the accelerating pace of ADC innovation globally. Bottom: TROP2-targeted agents, including sacituzumab govitecan (SG), datopotamab deruxtecan (Dato-DXd), and SKB264, have systematically expanded indications from refractory triple-negative disease to endocrine-pretreated HR+/HER2-negative populations. Abbreviations: 2L/3L, second-line/third-line; chemo, chemotherapy; eBC, early breast cancer; ET, endocrine therapy; HepM, hepatic metastasis; mBC, metastatic breast cancer; mTNBC, metastatic triple-negative breast cancer; RD, residual disease The success of an ADC depends on the precise recognition of antigens. The ideal target should exhibit several traits: high tumor specificity, accessible extracellular domains, minimal shedding into circulation, and rapid internalization [6]. Traditional classifications are often based on approval timelines or expression levels of the targets, which fail to capture the underlying biological characteristics of these targets [14]. To better understand the role of each target in tumor initiation and progression, ADC targets in breast cancer could be reclassified into three distinct categories according to their biological properties: Oncogenic Driver Antigens, Lineage and Oncofetal Antigens, and Tumor Microenvironment (TME) Antigens [46–49]. This functional categorization is illustrated in Fig. 4. This classification reflects not only the functional diversity of the antigens but also distinct design strategies and mechanisms of resistance. Tables 1, 2 and 3 summarizes key targets and recent advances in breast cancer research. Classification of ADC targets in breast cancer based on biological features. To align therapeutic strategies with underlying tumor biology, ADC targets are categorized into three functional dimensions. (1) Oncogenic Driver Antigens (e.g., HER2): These molecules actively drive malignant proliferation. ADCs targeting these receptors exert a dual anti-tumor mechanism: inhibiting downstream survival cascades (e.g., PI3K/AKT, MAPK) via antibody blockade, coupled with payload-induced cytotoxicity. (2) Lineage and Oncofetal Antigens (e.g., TROP2, Nectin-4): Functioning primarily as specific biomarkers of cellular identity, these targets possess minimal physiological distribution. Efficacy here relies strictly on efficient receptor-mediated internalization rather than signaling blockade. (3) Tumor Microenvironment (TME) Antigens (e.g., FAP, EDB-FN): Expressed on cancer-associated fibroblasts (CAFs) or extracellular matrix components. Targeting the TME permits extracellular payload release, exploiting the bystander effect to dismantle dense stromal barriers and eliminate heterogeneous tumor cells that lack traditional epithelial targets. (Note: These biological categories are not mutually exclusive; for instance, some lineage antigens may also exhibit driver-like signaling characteristics under specific conditions). Abbreviations: CAFs, cancer-associated fibroblasts; ECM, extracellular matrix; EDB-FN, extra-domain B fibronectin; FAP, fibroblast activation protein Table 1Landscape of Representative ADCs Targeting Oncogenic Driver Antigens in Breast CancerTargetADCAntibodyLinkerPayloadDARClinical Trial identifierconditionsdrugsPrimary OutcomePhaseStatusHER2 T-DM1* Trastuzumabnon-cleavableDM1 (MTI)3.5NCT07008976HER2 + ABCT-DM1, TQB2102PFSⅢRecruitingNCT06846437HER2 + ABCT-DM1, JSKN003PFSⅢRecruitingNCT06830889HER2 + BC with RD(post-NAT)T-DM1, BL-M07D1IDFSⅢRecruitingNCT06313086HER2 + ABCT-DM1, DP303cPFSⅢRecruitingNCT06125834HER2 + ABC (post-TKI/HP)T-DM1ORRⅡRecruiting T-DXd* TrastuzumabcleavableDXd (Topo I)8NCT07198724HR+ HER2-low/ultra ABC (Post-ET/CDK4/6i)T-DXd + ElacestrantORR, RP2DⅠ/ⅡNot yet recruitingNCT07152782HER2 + BC with BMT-DXd, T-DXd + NeratinibPFSⅡNot yet recruitingNCT07150208HER2-low BC with BMT-DXd,T-DXd + BevacizumabPFSⅡNot yet recruitingNCT07134153HER2-expressing BCwith LMD/BMT-DXdORR, OS, MTD, PKⅠ/ⅡRecruitingNCT06533826HER2-low/0 ABCT-DXd, Dato-DXdORRⅡRecruitingNCT06486883HR+ HER2-low/ultra ABC (Non-luminal)T-DXd, CDK4/6i + ETPFSⅡRecruitingNCT06328387ABCT-DXd, HCQ + T-DXd,SG, HCQ + SGORR, DLTs, AEsⅠ/ⅡRecruitingNCT06048718HER2- low BC with BMT-DXdORRⅡRecruitingRC48*DisitamabcleavableMMAE (MTI)4NCT07065435HER2 + ABC (post-T-DXd)RC48 + Bevacizumab,RC48 + PyrotinibORRⅡRecruitingNCT06966453HER2-expressing ABCRC48ORRⅠ/ⅡRecruitingNCT05904964HR+ HER2-low ABCRC48, ETPFSⅢRecruitingA166*TrastuzumabcleavableDuo-5 (MTI)2NCT06968585HER2 + ABC (post-Trastuzumab/Taxane)A166, T-DM1PFSⅢActive, not recruitingNCT03602079R/R HER2-expressing cancersA166MTDⅠ/ⅡCompletedSHR-A1811*TrastuzumabcleavableSHR9265 (Topo I)6NCT07528898HER2 + BCTHP,SHR-A1811 + PertuzumabpCRⅡNot yet recruitingNCT07393425HER2 + BCTHP,SHR-A1811 + PertuzumabpCRⅡRecruitingSYD-985Trastuzumabcleavableseco-DUBA (DNA-Alk)2.8NCT03262935HER2 + ABCSYD-985, TPCPFSⅢCompletedHER3HER3-DXdPatritumabcleavableDXd (Topo I)4NCT07060807HR+/HER2- ABCHER3-DXd, TPCPFS, OSⅢRecruitingNCT06298084ABC (Post-T-DXd)HER3-DXd,HER3-DXd + OlaparibORR, PFS, DOR, CBR, DLTs, AesⅠ/ⅡRecruitingYL-202Anti-HER3 IgG1cleavableYL0010014 (Topo I)8NCT06439771HER2-low/0 ABCYL202ORR, RP2DⅡRecruitingThis table presents a representative selection of ADCs targeting oncogenic driver antigens and is not intended to be an exhaustive list of all agents or clinical trials in the field. Clinical trial statuses were retrieved from ClinicalTrials.gov as of FebruarySymbols: Bold text indicates FDA-approved drugs; * indicates drugs approved exclusively by the NMPAAbbreviations ABC advanced breast cancer, ADC antibody-drug conjugate, AEs adverse events, BC breast cancer, BM brain metastases, DAR drug-to-antibody ratio, DDC durable disease control, DLT dose-limiting toxicity, DNA-Alk DNA alkylating agent, EB Cearly breast cancer, ET endocrine therapy, HR hormone receptor, mAb monoclonal antibody, mBC metastatic breast cancer, MTD maximum tolerated dose, MTI microtubule inhibitor, mTNBC metastatic triple-negative breast cancer, NMPA National Medical Products Administration, ORR objective response rate, OS overall survival, pCR pathologic complete response, PFS progression-free survival, PK pharmacokinetics, RP2D recommended phase 2 dose, TNBC triple-negative breast cancer, Topo I topoisomerase I inhibitor, TPC treatment of physician's choice Landscape of Representative ADCs Targeting Oncogenic Driver Antigens in Breast Cancer non- cleavable HER2 + BC with RD (post-NAT) T-DXd, T-DXd + Bevacizumab HER2-expressing BC with LMD/BM T-DXd, HCQ + T-DXd, SG, HCQ + SG RC48 + Bevacizumab, RC48 + Pyrotinib SHR- A1811* THP, SHR-A1811 + Pertuzumab THP, SHR-A1811 + Pertuzumab HER3- DXd HER3-DXd, HER3-DXd + Olaparib This table presents a representative selection of ADCs targeting oncogenic driver antigens and is not intended to be an exhaustive list of all agents or clinical trials in the field. Clinical trial statuses were retrieved from ClinicalTrials.gov as of February Symbols: Bold text indicates FDA-approved drugs; * indicates drugs approved exclusively by the NMPA Abbreviations ABC advanced breast cancer, ADC antibody-drug conjugate, AEs adverse events, BC breast cancer, BM brain metastases, DAR drug-to-antibody ratio, DDC durable disease control, DLT dose-limiting toxicity, DNA-Alk DNA alkylating agent, EB Cearly breast cancer, ET endocrine therapy, HR hormone receptor, mAb monoclonal antibody, mBC metastatic breast cancer, MTD maximum tolerated dose, MTI microtubule inhibitor, mTNBC metastatic triple-negative breast cancer, NMPA National Medical Products Administration, ORR objective response rate, OS overall survival, pCR pathologic complete response, PFS progression-free survival, PK pharmacokinetics, RP2D recommended phase 2 dose, TNBC triple-negative breast cancer, Topo I topoisomerase I inhibitor, TPC treatment of physician's choice Table 2Landscape of Representative ADCs Targeting Lineage and Oncofetal Antigens in Breast CancerTargetADCAntibodyLinkerPayloadDARClinical Trial identifierconditionsdrugsPrimary OutcomePhaseStatusTROP2 SG* SacituzumabcleavableSN-38(Topo I)7.6NCT07151586mTNBCSG, SG + T-DXdOSⅡNot yet recruitingNCT07271992BC with BMSG + IpCRⅡNot yet recruitingNCT07040644TNBCSG + Toripalimab,ToripalimabPFSⅡNot yet recruitingNCT06926920mTNBCSGORR, PFS, DLTs, AEsⅠ/ⅡRecruitingNCT06616987mTNBCSG + pegfilgrastimG ≥ 3 NeutropeniaⅡRecruiting Dato-DXd* DatopotamabcleavableDXd (Topo I)4NCT07205822HR+ HER2-0 ABCDato-DXdPFSⅢRecruitingNCT06954480mTNBC (PD-L1-)Dato-DXd,Dato-DXd + DurvalumabPFSⅡRecruitingNCT06508216mTNBC (Early Relapse)Dato-DXd,Dato-DXd + DurvalumabORR, RP2DⅠ/ⅡRecruitingSKB264*SacituzumabcleavableKL610023(Topo I)7.4NCT07182721HER2 + ABC (post-T-DXd)SKB264 + InetetamabORRⅡRecruitingNCT07109284ER+HER2- BCSKB264 + TagitanlimabpCRⅡNot yet recruitingNCT07071337HR+/HER2- ABC (Post-ET)SKB264, TPCPFSⅢRecruitingNCT06279364mTNBCSKB264, TPCPFS, OSⅢRecruitingNectin-4EVEnfortumabcleavableMMAE (MTI)4NCT04225117Adv. Solid Tumors(incl. TNBC, HR+/HER2- BC)EVORRⅡActiveLIV-1SGN-LIV1ALadiratuzuma-bcleavableMMAE (MTI)4NCT03310957mTNBCSGN-LIV1A+pembrolizumabORRⅠ/ⅡCompletedFRαMORAb-202FarletuzumabcleavableEribulin (MTI)4NCT04300556Adv. Solid Tumors(incl. TNBC)MORAb-202,MORAb-202 + OCS,MORAb-202 + LenvatinibORR, DLTs, AEs, RP2DⅠ/ⅡRecruitingSTRO-002SP8166cleavableSC209 (MTI)4------PTK7PF-06647020Anti-PTK7 IgG1cleavableAur0101 (MTI)4NCT02222922Adv. Solid Tumors(incl. TNBC)PF-06647020,PF-06647020 + fluconazole,PF-06647020 + AvelumabDLTs, AEsⅠCompleted(Discontinued)CD166CX-2009CX-191cleavableDM4 (MTI)3.5NCT04596150HER2- ABCCX-2009,CX-2009 + CX-072ORRⅡCompletedMesothelinBAY94-9343MF-TcleavableDM4 (MTI)3.2NCT03102320Adv. Solid Tumors(incl. TNBC)BAY94-9343,BAY94-9343 + Cisplatin,BAY94-9343 + GemcitabineORR, DLTs, DDCⅠCompleted(Discontinued)ROR1MK-2140CirmtuzumabcleavableMMAE (MTI)4NCT04504916Adv. Solid Tumors(incl. HER2- ABC)MK-2140ORRⅡTerminatedROR2BA3021pH-dependent CAB-IgG1cleavableMMAE (MTI)4NCT03504488TNBCBA3021,BA3021 + PD-1 inhibitorORR, DLTs, MTD, AEsⅠ/ⅡCompleted5T4 (TPBG)ASN004Novel single-chain scFv-Fc antibodycleavableAF-HPA (MTI)10–12NCT04410224Adv. Solid Tumors(incl. BC)ASN004MTDⅠCompletedTA-MUC1DS-3939Anti-TA-MUC1 IgG1cleavableDXd (Topo I)8NCT05875168Adv. Solid Tumors(incl. BC)DS-3939ORR, DLTs, AEsⅠ/ⅡRecruitingB7-H4AZD8205INT016cleavableAZ14170132 (Topo I)-NCT05123482Adv. Solid Tumors(incl. BC)AZD8205,AZD8205 + AZD2936,AZD8205 + AZD5305 ± AZD2936),AZD8205 + AZD9574 ± AZD2936DLTs, AEsⅠ/ⅡRecruitingSGN-B7H4VAnti-B7-H4 IgG1cleavableMMAE (MTI)4NCT05194072Adv. Solid Tumors(incl. BC)SGN-B7H4V,SGN-B7H4V +PembrolizumabDLTs, AEsⅠTerminatedHS-20,089Anti-B7-H4 IgG1cleavableDisclosed(Topo I)6NCT05263479Adv. Solid Tumors(incl. BC)HS-20,089MTDⅠRecruitingB7-H3I-DXdIfinatamabcleavableDXd (Topo I)4NCT04145622Adv. Solid Tumors(incl. BC)I-DXdORR, DLTs, AEsⅠ/ⅡRecruitingMGC018VobramitamabcleavableVc-seco-DUBA (DNA-Alk)2.7NCT03729596Adv. Solid Tumors(incl. TNBC)MGC018DLTs, AEsⅠ/ⅡTerminatedThis table presents a representative selection of ADCs targeting lineage and oncofetal antigens and is not intendedto be an exhaustive list of all agents or clinical trials in the field. Clinical trial statuses were retrieved from ClinicalTrials.gov as of February 2026Symbols: Bold text indicates FDA-approved drugs; * indicates drugs approved exclusively by the NMPAAbbreviations: ABC advanced breast cancer, ADC antibody-drug conjugate, AEs adverse events, BC breast cancer, BM brain metastases, DAR drug-to-antibody ratio, DDC durable diseasecontrol, DLT dose-limiting toxicity, DNA-Alk DNA alkylating agent, EBC early breast cancer, ET endocrine therapy, HR hormone receptor, mAb monoclonal antibody, mBC metastatic breast cancer, MTD maximum tolerated dose, MTI microtubule inhibitor, mTNBC metastatic triple-negativebreast cancer, NMPA National Medical Products Administration, ORR objective response rate, OS overall survival, pCR pathologic complete response, PFS progression-free survival, PK pharmacokinetics, RP2D recommended phase 2 dose, TNBC triple-negative breast cancer, Topo I topoisomerase Iinhibitor, TPC treatment of physician's choice Landscape of Representative ADCs Targeting Lineage and Oncofetal Antigens in Breast Cancer SN-38 (Topo I) SG + Toripalimab, Toripalimab Dato-DXd, Dato-DXd + Durvalumab Dato-DXd, Dato-DXd + Durvalumab KL610023 (Topo I) Adv. Solid Tumors (incl. TNBC, HR+/HER2- BC) Adv. Solid Tumors (incl. TNBC) MORAb-202, MORAb-202 + OCS, MORAb-202 + Lenvatinib Adv. Solid Tumors (incl. TNBC) PF-06647020, PF-06647020 + fluconazole, PF-06647020 + Avelumab Completed (Discontinued) CX-2009, CX-2009 + CX-072 Adv. Solid Tumors (incl. TNBC) BAY94-9343, BAY94-9343 + Cisplatin, BAY94-9343 + Gemcitabine Completed (Discontinued) Adv. Solid Tumors (incl. HER2- ABC) BA3021, BA3021 + PD-1 inhibitor Adv. Solid Tumors (incl. BC) Adv. Solid Tumors (incl. BC) Adv. Solid Tumors (incl. BC) AZD8205, AZD8205 + AZD2936, AZD8205 + AZD5305 ± AZD2936), AZD8205 + AZD9574 ± AZD2936 Adv. Solid Tumors (incl. BC) SGN-B7H4V, SGN-B7H4V + Pembrolizumab Disclosed (Topo I) Adv. Solid Tumors (incl. BC) Adv. Solid Tumors (incl. BC) Adv. Solid Tumors (incl. TNBC) This table presents a representative selection of ADCs targeting lineage and oncofetal antigens and is not intendedto be an exhaustive list of all agents or clinical trials in the field. Clinical trial statuses were retrieved from ClinicalTrials.gov as of February 2026 Symbols: Bold text indicates FDA-approved drugs; * indicates drugs approved exclusively by the NMPA Abbreviations: ABC advanced breast cancer, ADC antibody-drug conjugate, AEs adverse events, BC breast cancer, BM brain metastases, DAR drug-to-antibody ratio, DDC durable diseasecontrol, DLT dose-limiting toxicity, DNA-Alk DNA alkylating agent, EBC early breast cancer, ET endocrine therapy, HR hormone receptor, mAb monoclonal antibody, mBC metastatic breast cancer, MTD maximum tolerated dose, MTI microtubule inhibitor, mTNBC metastatic triple-negativebreast cancer, NMPA National Medical Products Administration, ORR objective response rate, OS overall survival, pCR pathologic complete response, PFS progression-free survival, PK pharmacokinetics, RP2D recommended phase 2 dose, TNBC triple-negative breast cancer, Topo I topoisomerase Iinhibitor, TPC treatment of physician's choice Table 3Landscape of Representative ADCs Targeting the Tumor Microenvironment (TME) in Breast CancerTargetADCAntibodyLinkerPayloadDARClinical Trial identifierconditionsdrugsPrimary OutcomePhaseStatusEDB-FNPYX-201Anti-EDB-FN IgG1cleavableAur0101 (MTI)4NCT05720117Adv. Solid Tumors(incl. BC)PYX-201ORR, DLTs, AEsⅠRecruitingFAPOMTX705OMTX005cleavableTAM470 (MTI)-NCT05547321Adv. Solid Tumors(incl. BC)OMTX705,OMTX705 + Tislelizumab,OMTX705 + pembrolizumabAEsⅠRecruitingLRRC15ABBV-085PR-1,498,487cleavableMMAE (MTI)2NCT02565758Adv. Solid Tumors(incl. BC)ABBV-085PK, AEsⅠCompleted(Discontinued)This table presents a representative selection of ADCs targeting tumor microenvironment targets and is not intendedto be an exhaustive list of all agents or clinical trials in the field. Clinical trial statuses were retrieved from ClinicalTrials.gov as of February 2026Abbreviations: ADC antibody-drug conjugate, AEs adverse events, BC breast cancer, DLT dose-limiting toxicity, MTI microtubule inhibitor, ORR objective response rate, PK pharmacokinetics Landscape of Representative ADCs Targeting the Tumor Microenvironment (TME) in Breast Cancer Adv. Solid Tumors (incl. BC) Adv. Solid Tumors (incl. BC) OMTX705, OMTX705 + Tislelizumab, OMTX705 + pembrolizumab Adv. Solid Tumors (incl. BC) Completed (Discontinued) This table presents a representative selection of ADCs targeting tumor microenvironment targets and is not intendedto be an exhaustive list of all agents or clinical trials in the field. Clinical trial statuses were retrieved from ClinicalTrials.gov as of February 2026 Abbreviations: ADC antibody-drug conjugate, AEs adverse events, BC breast cancer, DLT dose-limiting toxicity, MTI microtubule inhibitor, ORR objective response rate, PK pharmacokinetics Oncogenic driver antigens The malignant progression of tumors is frequently characterized by “oncogenic addiction,” where cancer cells become dependent on the aberrant activation of specific signaling pathways [50]. Oncogenic driver antigens are overexpressed cell-surface molecules that actively participate in and sustain intracellular oncogenic signaling cascades (e.g., PI3K/AKT, MAPK) crucial for tumor survival [51]. While the blockade of these pathways inherently possesses anti-tumor activity, ADCs further exploit these antigens as targets for the directed delivery of toxins, thus exerting a dual anti-tumor mechanism. This involves the direct inhibition of oncogenic signaling by the monoclonal antibody (signal blockade) coupled with the induction of apoptosis through the release of the cytotoxic payload [46]. By anchoring this classification on functional dependency rather than just clinical success or primary presentation, this framework includes boundary cases such as HER3 and EGFR. Although HER3 is predominantly discussed as a resistance-associated receptor, it functions fundamentally as a compensatory driver [52, 53]. Under therapeutic stress, it actively heterodimerizes to re-establish pro-survival signaling, thus driving the resistant phenotype. Similarly, wild-type EGFR is inherently an oncogenic driver receptor that promotes malignant proliferation in specific breast cancer subtypes, such as TNBC [54]. Its inclusion in this category reflects its undeniable biological identity as a theoretical driver, even though its clinical translation as an ADC target in breast cancer has failed due to a prohibitively narrow therapeutic window and severe on-target/off-tumor toxicities, rather than a lack of true oncogenic function [55]. HER2 HER2 is a transmembrane protein, encoded by the ERBB2 gene. Under physiological conditions, the HER2 protein serves as an integral component of the HER family of receptors (HER1/EGFR, HER2, HER3, and HER4), which collectively regulate normal cellular growth, division, and differentiation [56]. However, the amplification of the ERBB2 gene results in the aberrant overexpression of the protein on the surface of the cell, and this triggers the continuous activation of the phosphatidylinositol 3-kinase (PI3K), protein kinase B (PKB/AKT), and mitogen-activated protein kinase (MAPK) signaling pathways, thereby causing the proliferation of cancer cells [57]. The emergence of HER2-targeted ADC has been one of the most revolutionary milestones in the current treatment of breast cancer. To date, several HER2-targeted ADC therapies, namely T-DM1, T-DXd, Disitamab Vedotin (RC48), and Trastuzumab Botidotin (A166), have been approved by the FDA and the NMPA. Traditionally, the HER2 status was divided into HER2-positive (Immunohistochemistry [IHC] 3 + and IHC 2+/in situ hybridization [ISH]+) and HER2-negative types in the traditional setting [58]. Based on the results of the pivotal trial EMILIA, T-DM1, the first-generation HER2-targeted ADC, was approved by the FDA in the third-line therapy for HER2-positive metastatic breast cancer (mBC) in 2013 [59]. Later, the KATHERINE trial expanded its indication to the adjuvant setting for early-stage disease. With a median follow-up of 8.4 years, T-DM1 continued to demonstrate a significant improvement in invasive disease-free survival (iDFS) over trastuzumab (unstratified hazard ratio [HR] for invasive disease or death 0.54), with 7-year iDFS rates of 80.8% versus 67.1% [9, 60]. Mechanistically, T-DM1 retains the HER2 signal-blocking capability of trastuzumab while delivering the microtubule inhibitor DM1. However, its efficacy in low-HER2-expressing tumors is limited by a non-cleavable linker and the consequent lack of a “bystander effect”. The emergence of T-DXd has refined the classification of HER2 status. Different from T-DM1, T-DXd consists of a cleavable tetrapeptide linker, which allows the membrane-permeable topoisomerase I inhibitor DXd to cross the cell membrane [31]. This design not only suppresses tumor growth via signal blockade but also exerts a potent “bystander effect” to overcome intratumoral heterogeneity, thus showing efficacy in HER2-low (IHC1 + and IHC2+/ISH-) and even HER2-ultralow (IHC0 with faint membrane staining in > 10% of cells) breast cancers. DESTINY-Breast01 and the confirmatory Phase III DESTINY-Breast02 trial confirmed the efficacy of T-DXd treatment against pretreated, HER2-positive advanced breast cancer, while the head-to-head phase III study, DESTINY-Breast03, validated its superiority to T-DM1 [10, 61, 62]. Notably, the impressive results of the pivotal study, DESTINY-Breast04, challenged traditional molecular classification by demonstrating that T-DXd improves progression-free survival (PFS) and overall survival (OS) in patients with HER2-low disease, achieving a median PFS of 10.1 months (vs. 5.4 months) and OS of 23.9 months (vs. 17.5 months) with physician’s choice chemotherapy in the HR-positive cohort [11]. Later, DESTINY-Breast06 expanded the indications to include endocrine-pretreated, chemotherapy-naive patients with HER2-ultralow status [63]. In the early-stage setting, DESTINY-Breast05 and DESTINY-Breast11 have respectively indicate its powerful potency in adjuvant and neoadjuvant contexts [64, 65]. Currently, the ongoing DESTINY-Breast09 trial is designed to challenge the “gold standard” (taxane + trastuzumab + pertuzumab) for the first-line treatment of HER2-positive disease [66]. At present, the innovative ADCs developed in China have also made enormous progress in the treatment of HER2-positive advanced breast cancer. RC48 consists of a humanized anti-HER2 mAb conjugated to monomethyl auristatin E (MMAE) via a cleavable linker. Efficacy analyses of the RC48-C006 trial showed that RC48 significantly prolonged median progression-free survival (mPFS) compared to lapatinib plus capecitabine (9.9 months vs. 4.9 months) and was superior to the combination in the refractory subset of patients with liver metastasis [67]. Based on these results, in May 2025, the NMPA approved RC48 for specific patients with advanced breast cancer, meeting a pressing need in this niche population. A166 is a representation of the third-generation ADCs. It uses a protease-cleavable linker to precisely attach a new microtubule inhibitor (Duo-5) to the mAb by site-specific conjugation technology. This unique structural design not only enhances the stability of A166 in circulation, thus improving the safety and tolerability, but also maximizes therapeutic efficacy [68]. In HER2-positive advanced breast cancer, A166 showed an improved mPFS of 11.1 months compared to 4.4 months for T-DM1 in the head-to-head Phase III KL166-III-06 study, lowering the probability of disease progression or death by 61% [69]. Furthermore, the evolution of ADCs has recently achieved another milestone with the approval of trastuzumab rezetecan (SHR-A1811), which was granted by the NMPA in March 2026. SHR-A1811 is a novel HER2-targeting ADC equipped with a cleavable linker and a highly permeable topoisomerase I inhibitor payload (SHR9265), designed to deliver a potent bystander effect while maintaining favorable systemic stability. In HER2-positive advanced breast cancer, SHR-A1811 showed an improved mPFS of 30.6 months compared to 8.3 months for pyrotinib plus capecitabine in the Phase III HORIZON-Breast01 study, lowering the probability of disease progression or death by 78% [70, 71]. HER3 Human epidermal growth factor receptor 3 (HER3), encoded by the ERBB3 gene, is a unique member of the EGFR family. As a pseudokinase lacking intrinsic intracellular kinase activity, it functions primarily by forming heterodimeric complexes with other family members, notably HER2, to robustly activate downstream pro-survival signaling pathways [72]. HER3 plays a vital role in promoting tumor survival and mediating acquired resistance. While its expression in normal tissues is limited, it is broadly expressed across a wide range of breast cancer subtypes, including HR-positive and HER2-positive disease. More importantly, HER3 is frequently upregulated in a compensatory manner in tumors resistant to HER2-targeted therapies, making it an ideal target for overcoming drug resistance [73]. Mechanistically, broader preclinical investigations suggest that the efficacy of HER3-targeted ADCs does not strictly depend on the tumor’s oncogenic addiction to the HER3 pathway. Instead, even in tumors where HER3 acts as a compensatory receptor, agents like Patritumab deruxtecan (HER3-DXd) can effectively eradicate tyrosine kinase inhibitor (TKI)-resistant cancer cells by exploiting HER3 for efficient receptor-mediated internalization and subsequent payload release [74]. Translating these mechanistic insights into breast cancer-specific clinical evidence, HER3-DXd stands as the most advanced anti-HER3 ADC in this field. It is composed of a fully human anti-HER3 mAb (Patritumab), a cleavable tetrapeptide linker, and a topoisomerase I inhibitor payload (DXd). In a pretreated HR+/HER2- advanced breast cancer population enrolled in a Phase II ICARUS-BREAST01 trial, key efficacy assessments for HER3-DXd were achieved, including an objective response rate (ORR) of 53.5% and a mPFS of 9.2 months [75]. EGFR/HER1 Although overexpression of the Epidermal Growth Factor Receptor (EGFR/HER1) characterizes a subset of aggressive TNBC, its translation into a viable ADC target in breast cancer serves as a cautionary but informative negative example within the driver-antigen framework. Theoretical druggability has not equated to clinical success, primarily due to a critical biological divergence: unlike non-small cell lung cancer, which is frequently addicted to internalizing mutated EGFR, breast cancer is driven predominantly by wild-type EGFR. Because wild-type EGFR is also expressed in normal epithelial tissues (e.g., skin, gastrointestinal mucosa), ADCs targeting it suffer from an unacceptably narrow therapeutic index, triggering severe on-target/off-tumor toxicities [76]. Consequently, despite the theoretical druggability, its unavoidable toxicity has rendered EGFR a clinically failed ADC target in breast cancer. Lineage and oncofetal antigens Unlike oncogenic driver antigens, lineage and oncofetal antigens do not drive malignant proliferation directly. Instead, they serve as specific molecular biomarkers of identity, reflecting either the tissue of origin or a revertant, dedifferentiated phenotype. While the expression of these proteins in normal adult tissue is minimal, they are often elevated in tumors, which makes them ideal targets in ADC design [47]. Different from oncogenic driver antigens like HER2, the function of lineage and oncofetal antigens mainly relies on efficient receptor-mediated internalization and the potent cytotoxicity of their payloads, rather than on signal blockade. TROP2 TROP2 is a transmembrane glycoprotein widely distributed across epithelial malignancies, highly expressed in over 90% of TNBC and HR-positive cases. Because of the lack of actionable HER2 amplification, TROP2-targeted ADCs provide a necessary therapeutic alternative for refractory disease [77]. As the first approved agent in this class, SG conjugates the humanized hRS7 antibody to the topoisomerase I inhibitor SN-38 via a hydrolyzable CL2A linker (DAR of 7.6). Efficacy was established in the ASCENT trial, which has confirmed that SG prolonged overall survival in pretreated metastatic TNBC, establishing its position as the standard in the second-line therapy [12, 13]. Pushing the treatment boundary forward, the ASCENT-03 data indicated a 38% reduction in the risk of progression compared to chemotherapy in first-line TNBC patients who are ineligible for immunotherapy [78]. More recently, the indications for SG have expanded to HR-positive, HER2-negative disease in March 2025 (NMPA approval), supported by positive outcomes from the EVER-132-002 trial, achieving comprehensive coverage across major biological subtypes [79]. Distinct from SG, Datopotamab deruxtecan (Dato-DXd) employs the same tetrapeptide linker technology and payload (DXd) in T-DXd while targeting a different antigen. In the TROPION-Breast01 study, the agent showed a significant improvement in mPFS compared to chemotherapy (6.9 vs. 4.9 months), corresponding to a 37% reduction in the risk of progression or death [80]. Based on these results, the NMPA granted approval on August 22, 2025, for the treatment of unresectable or metastatic HR-positive, HER2-negative breast cancer in patients who have received endocrine therapy and at least one line of chemotherapy. In the first-line treatment of TNBC, Dato-DXd also presented exciting efficacy: the TROPION-Breast02 trial demonstrated that Dato-DXd was better than the investigator’s choice of chemotherapy in untreated patients who are not suitable for immunotherapy, either, extending mPFS to 10.8 months (vs. 5.6 months) and OS to 23.7 months (vs. 18.7 months) [81]. This research has suggested a potential shift in the first-line standard for this specific patient population. Sacituzumab tirumotecan (SKB264) is a structurally optimized agent developed within the domestic ADC landscape. As the first domestically developed TROP2-targeted ADC to receive full approval, SKB264 conjugates a stability-optimized CL2A linker with a proprietary novel topoisomerase I inhibitor (T030). The OptiTROP-Breast01 study, concentrating on advanced TNBC, demonstrated a significant improvement in mPFS compared to chemotherapy (6.7 vs. 2.5 months), reducing the risk of disease progression or death by 68%, which led to its NMPA approval in November 2024 [82]. Subsequent data from OptiTROP-Breast02 indicates consistent efficacy in earlier treatment lines, with median PFS reaching 8.3 months while improving overall survival trends [83]. Nectin-4 Nectin cell adhesion molecule 4 (Nectin-4) is a type I transmembrane cell adhesion protein as well as an important oncofetal antigen. It is aberrantly overexpressed in malignancy while its physiological distribution is limited to embryonic tissues and specific anatomical sites (e.g., skin, bladder) [84]. In TNBC, elevated Nectin-4 expression correlates with rapid tumor proliferation and poor survival [85]. For therapeutic targeting, the molecule’s kinetics are crucial: Nectin-4 mediates rapid internalization, efficiently ferrying bound ADC complexes to the lysosome. Utilizing this feature, agents such as enfortumab vedotin (EV) exploit this entry route to deposit cytotoxic payloads intracellularly, demonstrating robust anti-tumor activity in refractory breast cancer [84, 86]. Nevertheless, the physiological expression of this target in tissues, including the skin and bladder, results in distinct on-target toxicities, which still require further research. LIV-1 LIV-1 (also known as SLC39A6 or ZIP6) operates primarily as a zinc transporter, primarily localized to the endoplasmic reticulum and plasma membrane [87]. Its expression is positively regulated by estrogen and correlates closely with the epithelial-mesenchymal transition (EMT) [88]. While highly expressed in over 90% of HR-positive breast cancers and a significant proportion of TNBCs, LIV-1 has another specific advantage in antigen targeting due to its stability: antigen density persists and remains consistent during the progression from primary tumor to lymph node metastasis [89]. LIV-1-targeted ADCs (e.g., SGN-LIV1A) achieve precise payload release via receptor-mediated endocytosis by utilizing their physiological function in regulating intracellular zinc homeostasis [90]. Consequently, targeting this antigen provides a strategy to selectively eliminate aggressive cell populations associated with high metastatic potential. B7-H4 B7-H4 (VTCN1) operates as a negative immune checkpoint ligand within the B7 family, facilitating tumor immune escape by suppressing T-cell activation and proliferation. While expression in normal tissues is minimal, B7-H4 is aberrantly expressed in breast cancer, particularly in TNBC and HR-positive subtypes [91]. In addition to payload-induced cytotoxicity, ADCs directed at this target (e.g., AZD8205 and SGN-B7H4V) may promote synergistic anti-tumor immunity by eliminating antigen-positive cells to restructure the immune microenvironment [92]. Clinical trials in advanced solid tumors indicate preliminary efficacy of AZD8205 in breast cancer treatment; however, further targeted studies are warranted to confirm these findings [93]. Tumor microenvironment antigens Malignant breast tumor consists of malignant cells and stromal components, including cancer-associated fibroblasts (CAFs), vascular endothelial cells, immune cells, the extracellular matrix (ECM), and so on. These abnormal cells and the stroma are recognized as the TME [94]. In breast cancer (especially TNBC), dense desmoplasia creates a formidable physical barrier that excludes immune effector cells and impedes the penetration of conventional therapeutics [95]. Rather than functioning passively, stromal cells such as CAFs actively drive tumor growth, metastasis, and drug resistance by secreting growth factors and remodeling the ECM [96]. While traditional ADCs focus on killing target tumor cells themselves, the development of next-generation ADCs is shifting towards TME. Through permeable payloads and the bystander effect, these strategies could theoretically damage tumor cells surrounded by stroma, representing an innovative paradigm for overcoming tumor heterogeneity and stromal barriers. EDB-FN Fibronectin (FN) is a major component of the ECM. Its extra-domain B splice variant (EDB-FN) is a tumor-specific isoform only expressed during active angiogenesis and tissue remodeling [97]. Targeting EDB-FN offers a theoretical strategy to break the dense stromal barrier of desmoplastic tumors. Preclinical studies suggest that by releasing payloads extracellularly within the TME, these agents can potentially exploit the bystander effect to kill adjacent tumor cells regardless of their antigen expression levels, thus addressing the challenges posed by antigen heterogeneity and downregulation [98]. Additionally, it is hypothesized that EDB-FN-targeted ADCs could destroy Targeting EDB-FN-secreting CAFs and tumor vasculature to disrupt the physical matrix barrier, potentially facilitating the infiltration of immune effector cells and triggering “cold” tumors to an immunotherapy-responsive state. FAP Virtually absent in healthy adult tissues, fibroblast activation protein (FAP) is a serine protease mainly expressed on the surface of CAFs, which is upregulated in the stroma of over 90% of epithelial malignancies and is almost absent in healthy adult tissues [99]. In breast cancer, the expression level of FAP is associated with tumor aggressiveness, lymph node metastasis, and poor prognosis [100]. In preclinical models, ADCs targeting FAP specifically bind to CAFs in TME and release potent cytotoxic payloads. Through a robust bystander effect, they are hypothesized to bridge the stromal barrier to eliminate cancer cells encased within the dense desmoplasia, offering a way to overcome resistance driven by epithelial antigen heterogeneity [101]. Since FAP-positive CAFs are crucial to maintaining the immunosuppressive environment, eliminating these cells theoretically not only physically dismantles the tumor scaffold but may also reverse the immune-exclusion phenotype, thereby promoting tumor infiltration by effector T cells [102]. LRRC15 Distinguished from the aforementioned targets, Leucine-Rich Repeat Containing 15 (LRRC15), a transmembrane protein governed by TGF-β signaling, functions as a unique “dual tumor-stroma target”. It exhibits specific overexpression on both mesenchymal subtype tumor cells and on the surrounding CAFs in desmoplastic areas [103]. For stroma-rich, refractory breast cancers, early evidence suggests that LRRC15-targeted ADCs have the potential to transform the stroma into a drug depot, aiming to achieve sustained tumor eradication through the in situ persistent release of permeable payloads. Meanwhile, high LRRC15 expression is frequently associated with poor prognosis and an “immune-desert” phenotype [104]. Given these, targeting this antigen may help remodel the pro-tumorigenic stromal microenvironment, potentially sensitizing the tumor to subsequent immune checkpoint blockade. Management of toxicities Although the imaginative design to combine antibody and cytotoxicity of ADCs has significantly lowered the systemic toxicity compared to traditional chemotherapy, their unique hybrid structure has simultaneously introduced complex treatment-emergent adverse events (TEAEs). Beyond the common toxicities associated with structural design and cell-killing effect, the toxicity profiles across different agents exhibit large heterogeneity, which is largely attributed to target antigen distribution, chemical linker stability, and the biochemical characteristics of the cytotoxic payload [105]. Table 4 provides a summary of TEAEs associated with ADCs approved for breast cancer. Table 4Overview of Treatment-Emergent Adverse Events and Safety Warnings of Approved ADCs in Breast CancerAdverse ReactionT-DM1T-DXdRC48A166SGSKB264Dato-DXdSHR-1811Blood and lymphatic system disorders Neutropenia*+*▲+*+* Thrombocytopenia+*+*++ Anemia++*+++Cardiac disorders Left Ventricular Dysfunction*** Eye disorders*** Corneal diseasesa++ Dry eye++ Blurred vision+ Eye discomfortb+Gastrointestinal disorders Diarrhea+++▲++ Nausea++++*+++ Vomiting++++*+++ Constipation++++++ Abdominal Pain+++++ Stomatitis++++*+*+ Dry Mouth+ Dyspepsia+Hepatobiliary disorders Nodular regenerative hyperplasia*General disorders and administration site conditions Fatigue/asthenia++++++++ Pyrexia++ Infusion site extravasation* Hypersensitivity**+* Fever+Infections and infestations Urinary tract infection+++ Upper respiratory tract infection+Metabolism and nutrition disorders Decreased appetite++++++ Hypokalemia++ hyperlipemia++ Hyperglycemia+++ hyperuricemia+++ hypoproteinemia+Musculoskeletal and connective tissue disorders Back pain+ Arthralgia++ Musculoskeletal pain+++ Myalgia+Nervous system disorders Headache+++ Dizziness++++ Dysesthesia++* Peripheral neuropathy+*+*+Psychiatric disorders Insomnia++Reproductive system and breast disorders Embryo-fetal toxicity▲▲******Respiratory, thoracic, and mediastinal disorders Epistaxis++ Dyspnea+ Cough++++ Respiratory infection+ Interstitial lung disease▲+▲**Skin and subcutaneous tissue disorders Alopecia+++++++ Rash+++++ Pruritus++Vascular Disorders Hemorrhage+*Injury, poisoning, and procedural complications Infusion-related reaction*****Investigations Transaminases increased+*++*+++++* Decreased weight++++ Urinary protein increased+ Decreased hemoglobin+++++++ Decreased lymphocyte count+++++ Decreased leukocyte count++*+++++ Decreased neutrophil count++++++++ Decreased platelet count++++++ Increased glucose+ Decreased calcium++ Decreased magnesium+ Decreased potassium+++ Increased bilirubin+*+++ Increased albumin+* Increased creatinine++* Increased AST+*++++++* Increased ALT+*++++++* Increased ALP+++++ Decreased phosphate+ Decreased sodium+ Increased LDH+++ Decreased glucose+This table summarizes TEAEs with an incidence of ≥ 10% (all grades) reported in the prescribing information and key clinical trials for each agentSymbols: + indicates an incidence of ≥ 10% (all grades); ▲ indicates a Boxed Warning (FDA) in the prescribing information; * indicates a specific warning and precaution highlighted in the prescribing informationDefinition of Aggregated Terms: (a) Corneal diseases include corneal epithelial defect, corneal opacity, ulcerative keratitis, keratitis, corneal neovascularization, corneal scar, and corneal injury; (b) Eye discomfort includes photophobia, eye pruritus, eye pain, foreign body sensation in the eyes, eye irritation, eye swelling, eyelid pruritus, and asthenopiaData Sources: T-DM1: Phase III EMILIA trial (n = 490). T-DXd (Trastuzumab deruxtecan): Phase III DESTINY-Breast03 trial (n = 257). RC48 (Disitamab vedotin): Based on the NMPA prescribing information, summarizing pooled safety data from 544 patients across clinical trials (C001, C002, C003, C005, C006, C008, and C009). A166 (Trastuzumab botidotin): Based on the NMPA prescribing information, summarizing pooled safety data from 490 patients across 6 clinical trials (including KL166-III-06). SG (Sacituzumab govitecan): Phase III ASCENT trial (n = 258). SKB264 (Sacituzumab tirumotecan): Based on the NMPA prescribing information, summarizing pooled safety data from 806 patients across trials, including SKB264-01, SKB264-II-08, and SKB264-III-03 (OptiTROP-Breast01). Dato-DXd: Phase III TROPION-Breast01 trial (n = 360). SHR-A1811: Based on the NMPA prescribing information, summarizing pooled safety data from 439 patients across trials, including SHR-A1811-I-101, SHR-A1811-I-102, SHR-A1811-I-103, and SHR-A1811-II-202Abbreviations AST aspartate aminotransferase, ALT alanine aminotransferase, ALP alkaline phosphatase, FDA Food and Drug Administration, LDH lactate dehydrogenase, NMPA National Medical Products Administration Overview of Treatment-Emergent Adverse Events and Safety Warnings of Approved ADCs in Breast Cancer This table summarizes TEAEs with an incidence of ≥ 10% (all grades) reported in the prescribing information and key clinical trials for each agent Symbols: + indicates an incidence of ≥ 10% (all grades); ▲ indicates a Boxed Warning (FDA) in the prescribing information; * indicates a specific warning and precaution highlighted in the prescribing information Definition of Aggregated Terms: (a) Corneal diseases include corneal epithelial defect, corneal opacity, ulcerative keratitis, keratitis, corneal neovascularization, corneal scar, and corneal injury; (b) Eye discomfort includes photophobia, eye pruritus, eye pain, foreign body sensation in the eyes, eye irritation, eye swelling, eyelid pruritus, and asthenopia Data Sources: T-DM1: Phase III EMILIA trial (n = 490). T-DXd (Trastuzumab deruxtecan): Phase III DESTINY-Breast03 trial (n = 257). RC48 (Disitamab vedotin): Based on the NMPA prescribing information, summarizing pooled safety data from 544 patients across clinical trials (C001, C002, C003, C005, C006, C008, and C009). A166 (Trastuzumab botidotin): Based on the NMPA prescribing information, summarizing pooled safety data from 490 patients across 6 clinical trials (including KL166-III-06). SG (Sacituzumab govitecan): Phase III ASCENT trial (n = 258). SKB264 (Sacituzumab tirumotecan): Based on the NMPA prescribing information, summarizing pooled safety data from 806 patients across trials, including SKB264-01, SKB264-II-08, and SKB264-III-03 (OptiTROP-Breast01). Dato-DXd: Phase III TROPION-Breast01 trial (n = 360). SHR-A1811: Based on the NMPA prescribing information, summarizing pooled safety data from 439 patients across trials, including SHR-A1811-I-101, SHR-A1811-I-102, SHR-A1811-I-103, and SHR-A1811-II-202 Abbreviations AST aspartate aminotransferase, ALT alanine aminotransferase, ALP alkaline phosphatase, FDA Food and Drug Administration, LDH lactate dehydrogenase, NMPA National Medical Products Administration This heterogeneity in toxicity profiles presents distinct clinical management challenges, ranging from payload-specific hematological toxicities to life-threatening events such as interstitial lung disease. A comprehensive overview of these toxicity profiles, including FDA Boxed Warnings, is provided in Fig. 5. Overview of antibody-drug conjugate toxicity profiles. The distinct toxicity profiles of ADCs are dictated by a complex interplay of target antigen tissue distribution (on-target, off-tumor toxicity), premature linker cleavage, and the intrinsic biochemical vulnerabilities of the payload. Key clinical challenges include interstitial lung disease (ILD; notably linked to T-DXd), severe neutropenia and early-onset diarrhea (characteristic of topoisomerase I inhibitors), alongside peripheral neuropathy and thrombocytopenia (frequently driven by microtubule inhibitors). Bold text denotes FDA Boxed Warnings, highlighting severe, life-threatening, or fatal risks (e.g., embryo-fetal toxicity, pneumonitis). Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; ILD, interstitial lung disease. The schematic diagram was created with BioGDP.com [106] Mechanism of toxicities The toxicity of ADCs can be classified into two main mechanisms. One is “on-target/off-tumor” toxicity, which arises when ADCs inadvertently damage healthy cells due to low-level antigen expression in normal tissues [107]. For instance, the physiological distribution of TROP-2 across various epithelial tissues (e.g., gastrointestinal mucosa, hair follicles, bone marrow progenitors) induces toxicities such as diarrhea, alopecia, and neutropenia frequently observed with SG [108]. On the other hand, the more complicated “off-target” toxicity is the chief cause of the typical spectrum of adverse events seen for these drugs. This toxicity mainly results from the nonspecific interactions of the ADC components in non-neoplastic tissues [109]. Linker stability acts as a fundamental safety determinant; premature cleavage in the systemic circulation releases free cytotoxic payloads, mimicking the systemic toxicity of conventional chemotherapy. In addition, normal immune cells may nonspecifically uptake the ADCs through interactions with the Fcγreceptors or the mannose receptors. This process is particularly implicated in T-DXd-induced interstitial lung disease (ILD), where current evidence points to alveolar macrophage uptake and the subsequent inflammatory cascade as the central pathogenic mechanism [110]. Even the bystander effect, which plays an important role in the therapeutic efficiency, acts like a double-edged sword: Hydrophobic payloads (e.g., DXd) with high membrane permeability can diffuse from the targeted tumor cells to adjacent normal stromal cells or vascular endothelium surrounding the tumor, causing additional injury within the TME [111]. Beyond the above mechanisms, the particular target class of the cytotoxic moiety defines the pattern of prevalent toxicity [105]. For example, microtubule inhibitors (e.g., DM1, MMAE) exhibit inherent susceptibility to ocular, neurological, and hepatotoxicity, in addition to thrombocytopenia. Conversely, topoisomerase I inhibitors (e.g., DXd, SN-38) are more frequently associated with gastrointestinal adverse events, including nausea, vomiting, and diarrhea, as well as myelosuppression [34]. Among these, the unique pulmonary toxicity associated with T-DXd warrants extreme care in clinical management. Table 5 summarizes the mechanisms underlying ADC-related toxicities in breast cancer and the corresponding clinical management strategies. Table 5Potential Mechanism and Clinical Management for Treatment-Emergent Adverse Events Associated with ADCs in Breast CancerSystem Organ Class (SOC)Specific Adverse EventsMain Pathogenic ComponentPotential MechanismClinical ManagementBlood and lymphatic system disordersAnemiaNeutropeniaThrombocytopeniaPayload(All classes)Payload CharacteristicsIntrinsic Cytotoxicity: Most cytotoxic payloads exert direct antiproliferative effects on rapidly proliferating hematopoietic progenitor cells in the bone marrow.Metabolic Susceptibility: The metabolism of SN-38 is mediated by the UGT1A1 enzyme. Patients homozygous for the UGT1A1*28 allele experience reduced glucuronidation activity, leading to systemic accumulation of SN-38 and an increased risk of myelosuppression.Megakaryocyte Cytotoxicity: Microtubule inhibitors (e.g., DM1) specifically impair the differentiation and maturation of megakaryocytes, resulting in suppressed thrombopoiesis (reduced platelet production).Off-Target Toxicity: Megakaryocytes non-specifically internalize ADCs via FcγRIIa receptors, leading to cytotoxicity and subsequent platelet destruction.On-Target Toxicity: Potential expression of TROP2 on hematopoietic stem and progenitor cells (HSPCs) may mediate direct targeted toxicity.Monitor: CBC before each dose and as clinically indicated.Prophylaxis: G-CSF for high-risk neutropenia (e.g., SG); platelet transfusion for severe bleeding.Intervention: Withhold for Grade ≥ 3 neutropenia/thrombocytopenia or febrile neutropenia until recovery (typically Grade ≤ 1); resume at reduced dose for recurrent Grade 3 or any Grade 4 events.Caution: Avoid anticoagulants/antiplatelets if platelets < 50,000/mm^3 (esp. T-DM1).Cardiac disordersLeft ventricular dysfunctionTarget (HER2)On-Target Toxicity: Blockade of HER2 signaling pathways on the surface of cardiomyocytes impairs their survival mechanisms and capacity to respond to physiological stress.Payload Characteristics: Certain cytotoxic payloads may disrupt mitochondrial function or interfere with microtubule networks within cardiomyocytes.Monitor: LVEF assessment at baseline and regular intervals.Intervention: Withhold for LVEF < 40–45% or absolute decrease ≥ 10–20% from baseline; discontinue for Symptomatic CHF or confirmed persistent LVEF decline.Eye disordersKeratopathydry eyeblurred visionPayload (MTI)TargetOff-Target Toxicity: Human corneal epithelial cells (HCECs) non-specifically internalize ADCs via macropinocytosis, leading to intracellular accumulation and off-target toxicity.Payload Characteristics: Microtubule inhibitor payloads preferentially accumulate in corneal basal stem cells and induce apoptosis, resulting in the formation of characteristic microcyst-like epithelial changes (MECs).On-Target Toxicity: Basal physiological expression of TROP2 and HER2 in the corneal epithelium may mediate direct target-dependent toxicity.Monitor: Ophthalmic exams at baseline and periodically.Prophylaxis: Use preservative-free artificial tears daily; strictly avoid contact lenses.Intervention: Withhold for Grade ≥ 2 or symptomatic events until recovery; resume at reduced dose upon improvement; discontinue for Grade 4.Gastrointestinal disordersDiarrheaTarget (TROP2)Payload (SN38)On-Target Toxicity: TROP2 is physiologically expressed in the basal keratinocytes of the oral mucosa and intestinal epithelium. Upon penetration and uptake into the mucosal layer, ADCs induce apoptosis of basal cells and disruption of the mucosal barrier, resulting in late-onset diarrhea.Payload Characteristics: SN-38 inhibits acetylcholinesterase, leading to acetylcholine accumulation and subsequent early-onset cholinergic diarrhea (characterized by cramping and abdominal pain).Early-onset: Atropine (IV/SC) for cholinergic symptoms.Late-onset: Loperamide at onset + fluids/electrolytes; rule out infection.Intervention: Withhold for Grade ≤ 3; resume at reduced dose upon recovery (≤ G1).StomatitisTarget (TROP2)PayloadOn-Target Toxicity: ADC targeting TROP2 on the oral mucosa, causing ulceration.Payload Characteristics: Direct damage to rapidly dividing mucosal cells.Prophylaxis: Corticosteroid mouthwash (SWISH) starting Day 1; Cryotherapy (ice chips) during infusion.Care: Soft toothbrush, non-alcoholic mouthwash.Intervention: Withhold for Grade ≥ 2/3 (drug-specific); reduce dose for recurrence.Nausea / VomitingPayloadPayload Characteristics: Potent cytotoxic payloads (esp. DXd, SN-38) directly stimulate the Chemoreceptor Trigger Zone (CTZ) in the CNS.Prophylaxis: 2–3 drug antiemetic regimen (5-HT3 RA + Dex ± NK1 RA) prior to infusion.Support: Take-home antiemetics.Intervention: Withhold for Grade ≥ 3 events; resume with dose reduction & optimized antiemetics.Hepatobiliary disordersHepatotoxicityNodular regenerative hyperplasia (NRH)ADC Complex (Fc-Uptake)Payload (DM1)Off-Target Uptake: The liver serves as the primary organ for ADC clearance. Hepatic sinusoidal endothelial cells (LSECs) and Kupffer cells sequester ADCs via Fc receptor-mediated or non-specific endocytosis, leading to hepatic accumulation.Payload Characteristics: T-DM1 specifically disrupts the cytoskeletal architecture of LSECs, inducing Nodular Regenerative Hyperplasia (NRH) and subsequent non-cirrhotic portal hypertension.Monitor: Liver function tests (AST/ALT, TBIL) prior to each dose.Intervention: Withhold for Grade ≥ 3 AST/ALT elevation or Grade ≥ 2 hyperbilirubinemia; dose reduction upon recovery. Permanently stop for Hy’s Law cases (AST/ALT >3x ULN + Bilirubin >2x ULN) or confirmed NRH.Caution: Suspect NRH in T-DM1 patients presenting with portal hypertension features but normal transaminases.Metabolism and nutrition disordersHyperuricemiaPayload (Potent Cytotoxins)Payload Characteristics (Tumor Lysis): Potent cytotoxic payloads (e.g., Duo-5 in A166, KL610023 in SKB264) induce rapid and massive apoptosis of tumor cells, releasing significant amounts of intracellular nucleic acids and potassium, which are metabolized into uric acid.Off-Target Toxicity: Certain ADCs may cause subclinical renal tubular injury (manifested as proteinuria) via non-specific uptake, which secondarily compromises the renal excretion of uric acid, exacerbating hyperuricemia.Hydration: Oral hydration (> 2 L/day) to maintain urine output.Monitor: Serum uric acid and renal function (Cr/BUN) regularly.Intervention: Prophylactic or therapeutic urate-lowering agents (e.g., Febuxostat, Allopurinol) for high-risk patients or clinically significant elevations.HyperglycemiaConcomitant Meds (Steroids)PayloadMetabolic Dysregulation (Drug-Induced): Corticosteroid premedication required for antiemetic/hypersensitivity prophylaxis may induce transient insulin resistance and gluconeogenesis.Payload/Target Characteristics: Direct interference with glucose metabolic pathways by specific payloads or targets may also contribute to metabolic dysregulation.Screening: Monitor fasting blood glucose (FBG) and HbA1c at baseline and prior to infusions.Intervention: Initiate or adjust hypoglycemic therapy (e.g., metformin, insulin) if FBG > 13.9 mmol/L (> 250 mg/dL); withhold ADC for uncontrolled Grade ≥ 3 hyperglycemia until controlled.Nervous system disordersPeripheral NeuropathyPayload (MTI)Target (Nectin-4)Payload Characteristics: Microtubule inhibitors (e.g., MMAE, DM1, Duo-5) disrupt the microtubule dynamics within neuronal axons, impairing axoplasmic transport and leading to distal axonal degeneration.Off-Target Toxicity: Neurons may uptake circulating ADCs through non-specific mechanisms such as macropinocytosis or receptor-mediated endocytosis, resulting in intracellular payload accumulation and cytotoxicity.On-Target Toxicity: Basal expression of specific antigens (e.g., Nectin-4) in peripheral nerve tissues may mediate direct targeted toxicity.Monitor: Assess sensory and motor function prior to each dose.Support: Consider symptomatic treatment (e.g., duloxetine) for neuropathic pain.Intervention: Withhold for Grade ≥ 2/3 (drug-specific) peripheral neuropathy until resolution (to ≥ 1 or baseline); resume at reduced dose for recurrent Grade 2 or first occurrence of Grade 3 events; discontinue for Grade 4.Skin and subcutaneous tissue disordersAlopeciaRashTarget (Nectin-4)Payload (MMAE)On-Target Toxicity: Nectin-4 is physiologically expressed in human cutaneous structures, including epidermal keratinocytes, sweat glands, and hair follicles, mediating targeted uptake of ADCs into skin tissue.Payload Characteristics: The released MMAE payload accumulates in the skin, disrupting microtubules and inducing keratinocyte apoptosis, which histopathologically manifests as interface dermatitis. In severe cases, massive epidermal necrosis can precipitate life-threatening Stevens-Johnson Syndrome (SJS) or Toxic Epidermal Necrolysis (TEN).Monitor: Skin exam at baseline and each cycle.Intervention: Withhold for Grade ≥ 2/3 (drug-specific) until resolution (≥ G1); resume at reduced dose for recurrent Grade 2 or first Grade 3 event; discontinue for Grade 4.Respiratory, Thoracic, and Mediastinal DisordersInterstitial Lung Disease / PneumonitisADC Complex (Fc-Uptake)Payload (DXd)Off-Target Toxicity: Alveolar macrophages expressing Fcγ receptors (FcγRs) or mannose receptors non-specifically internalize ADCs. The subsequent intracellular release of cytotoxic payloads triggers an immune-inflammatory cascade.Payload Characteristics: Membrane-permeable, hydrophobic payloads (e.g., DXd) released by macrophages diffuse into adjacent normal alveolar epithelial cells via the “bystander effect,” causing widespread tissue damage.Monitor: SpO2 and Chest CT at baseline and regularly; prompt workup for new respiratory symptoms.Intervention: Withhold for Grade 1 (asymptomatic); resume only if resolved to Grade 0; permanently discontinue for Grade ≥ 2 (symptomatic), initiate systemic corticosteroids immediately.Reproductive systemEmbryo-Fetal ToxicityPayloadTarget (HER2)Payload Characteristics: Cytotoxic payloads serve as potent teratogens, indiscriminately targeting rapidly dividing embryonic cells, leading to embryo-fetal death or structural anomalies.On-Target Toxicity: The blockade of HER2 signaling in the fetal kidney impairs urine production, causing oligohydramnios, which secondarily induces fetal pulmonary hypoplasia and skeletal abnormalities (Potter sequence).Monitor: Verify pregnancy status prior to initiation.Prophylaxis: Use effective contraception (Females: 7 months post-dose; Males: 4–6 months post-dose); do not breastfeed during treatment and for 7 months post-dose.Intervention: Immediately discontinue if pregnancy confirmed; refer to high-risk obstetrics for oligohydramnios surveillance.Grading System: Toxicity grades refer to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) version 5.0 or 6.0, unless otherwise specified in individual drug labelsManagement Generalization: Clinical management strategies summarized here are based on FDA/NMPA prescribing information and major clinical guidelines. Physicians should consult the specific drug label for detailed dose modification protocols, as criteria may vary by agentAbbreviations ADC antibody-drug conjugate, ANC absolute neutrophil count, CNS central nervous system, CT computed tomography, CTZ chemoreceptor trigger zone, DXd exatecan derivative, ECHO echocardiogram, G-CSF granulocyte colony-stimulating factor, HbA1c glycated hemoglobin, HCECs human corneal epithelial cells, HER2 human epidermal growth factor receptor 2, ILD interstitial lung disease, LVEF left ventricular ejection fraction, MECs microcyst-like epithelial changes, MMAE monomethyl auristatin E, MUGA multigated acquisition scan, Nectin-4 nectin cell adhesion molecule 4, NRH nodular regenerative hyperplasia, SJS Stevens-Johnson syndrome, SN-38 7-ethyl-10-hydroxycamptothecin, SpO2 peripheral oxygen saturation, TEN toxic epidermal necrolysis, TLS tumor lysis syndrome, TME tumor microenvironment, Topo-I topoisomerase I, TROP2 trophoblast cell-surface antigen 2, UGT1A1 uridine diphosphate glucuronosyltransferase 1A1 Potential Mechanism and Clinical Management for Treatment-Emergent Adverse Events Associated with ADCs in Breast Cancer Anemia Neutropenia Thrombocytopenia Payload (All classes) Payload Characteristics Intrinsic Cytotoxicity: Most cytotoxic payloads exert direct antiproliferative effects on rapidly proliferating hematopoietic progenitor cells in the bone marrow. Metabolic Susceptibility: The metabolism of SN-38 is mediated by the UGT1A1 enzyme. Patients homozygous for the UGT1A1*28 allele experience reduced glucuronidation activity, leading to systemic accumulation of SN-38 and an increased risk of myelosuppression. Megakaryocyte Cytotoxicity: Microtubule inhibitors (e.g., DM1) specifically impair the differentiation and maturation of megakaryocytes, resulting in suppressed thrombopoiesis (reduced platelet production). Off-Target Toxicity: Megakaryocytes non-specifically internalize ADCs via FcγRIIa receptors, leading to cytotoxicity and subsequent platelet destruction. On-Target Toxicity: Potential expression of TROP2 on hematopoietic stem and progenitor cells (HSPCs) may mediate direct targeted toxicity. Monitor: CBC before each dose and as clinically indicated. Prophylaxis: G-CSF for high-risk neutropenia (e.g., SG); platelet transfusion for severe bleeding. Intervention: Withhold for Grade ≥ 3 neutropenia/thrombocytopenia or febrile neutropenia until recovery (typically Grade ≤ 1); resume at reduced dose for recurrent Grade 3 or any Grade 4 events. Caution: Avoid anticoagulants/antiplatelets if platelets < 50,000/mm^3 (esp. T-DM1). On-Target Toxicity: Blockade of HER2 signaling pathways on the surface of cardiomyocytes impairs their survival mechanisms and capacity to respond to physiological stress. Payload Characteristics: Certain cytotoxic payloads may disrupt mitochondrial function or interfere with microtubule networks within cardiomyocytes. Monitor: LVEF assessment at baseline and regular intervals. Intervention: Withhold for LVEF < 40–45% or absolute decrease ≥ 10–20% from baseline; discontinue for Symptomatic CHF or confirmed persistent LVEF decline. Keratopathy dry eye blurred vision Payload (MTI) Target Off-Target Toxicity: Human corneal epithelial cells (HCECs) non-specifically internalize ADCs via macropinocytosis, leading to intracellular accumulation and off-target toxicity. Payload Characteristics: Microtubule inhibitor payloads preferentially accumulate in corneal basal stem cells and induce apoptosis, resulting in the formation of characteristic microcyst-like epithelial changes (MECs). On-Target Toxicity: Basal physiological expression of TROP2 and HER2 in the corneal epithelium may mediate direct target-dependent toxicity. Monitor: Ophthalmic exams at baseline and periodically. Prophylaxis: Use preservative-free artificial tears daily; strictly avoid contact lenses. Intervention: Withhold for Grade ≥ 2 or symptomatic events until recovery; resume at reduced dose upon improvement; discontinue for Grade 4. Target (TROP2) Payload (SN38) On-Target Toxicity: TROP2 is physiologically expressed in the basal keratinocytes of the oral mucosa and intestinal epithelium. Upon penetration and uptake into the mucosal layer, ADCs induce apoptosis of basal cells and disruption of the mucosal barrier, resulting in late-onset diarrhea. Payload Characteristics: SN-38 inhibits acetylcholinesterase, leading to acetylcholine accumulation and subsequent early-onset cholinergic diarrhea (characterized by cramping and abdominal pain). Early-onset: Atropine (IV/SC) for cholinergic symptoms. Late-onset: Loperamide at onset + fluids/electrolytes; rule out infection. Intervention: Withhold for Grade ≤ 3; resume at reduced dose upon recovery (≤ G1). Target (TROP2) Payload On-Target Toxicity: ADC targeting TROP2 on the oral mucosa, causing ulceration. Payload Characteristics: Direct damage to rapidly dividing mucosal cells. Prophylaxis: Corticosteroid mouthwash (SWISH) starting Day 1; Cryotherapy (ice chips) during infusion. Care: Soft toothbrush, non-alcoholic mouthwash. Intervention: Withhold for Grade ≥ 2/3 (drug-specific); reduce dose for recurrence. Prophylaxis: 2–3 drug antiemetic regimen (5-HT3 RA + Dex ± NK1 RA) prior to infusion. Support: Take-home antiemetics. Intervention: Withhold for Grade ≥ 3 events; resume with dose reduction & optimized antiemetics. Hepatotoxicity Nodular regenerative hyperplasia (NRH) ADC Complex (Fc-Uptake) Payload (DM1) Off-Target Uptake: The liver serves as the primary organ for ADC clearance. Hepatic sinusoidal endothelial cells (LSECs) and Kupffer cells sequester ADCs via Fc receptor-mediated or non-specific endocytosis, leading to hepatic accumulation. Payload Characteristics: T-DM1 specifically disrupts the cytoskeletal architecture of LSECs, inducing Nodular Regenerative Hyperplasia (NRH) and subsequent non-cirrhotic portal hypertension. Monitor: Liver function tests (AST/ALT, TBIL) prior to each dose. Intervention: Withhold for Grade ≥ 3 AST/ALT elevation or Grade ≥ 2 hyperbilirubinemia; dose reduction upon recovery. Permanently stop for Hy’s Law cases (AST/ALT >3x ULN + Bilirubin >2x ULN) or confirmed NRH. Caution: Suspect NRH in T-DM1 patients presenting with portal hypertension features but normal transaminases. Payload Characteristics (Tumor Lysis): Potent cytotoxic payloads (e.g., Duo-5 in A166, KL610023 in SKB264) induce rapid and massive apoptosis of tumor cells, releasing significant amounts of intracellular nucleic acids and potassium, which are metabolized into uric acid. Off-Target Toxicity: Certain ADCs may cause subclinical renal tubular injury (manifested as proteinuria) via non-specific uptake, which secondarily compromises the renal excretion of uric acid, exacerbating hyperuricemia. Hydration: Oral hydration (> 2 L/day) to maintain urine output. Monitor: Serum uric acid and renal function (Cr/BUN) regularly. Intervention: Prophylactic or therapeutic urate-lowering agents (e.g., Febuxostat, Allopurinol) for high-risk patients or clinically significant elevations. Concomitant Meds (Steroids) Payload Metabolic Dysregulation (Drug-Induced): Corticosteroid premedication required for antiemetic/hypersensitivity prophylaxis may induce transient insulin resistance and gluconeogenesis. Payload/Target Characteristics: Direct interference with glucose metabolic pathways by specific payloads or targets may also contribute to metabolic dysregulation. Screening: Monitor fasting blood glucose (FBG) and HbA1c at baseline and prior to infusions. Intervention: Initiate or adjust hypoglycemic therapy (e.g., metformin, insulin) if FBG > 13.9 mmol/L (> 250 mg/dL); withhold ADC for uncontrolled Grade ≥ 3 hyperglycemia until controlled. Payload (MTI) Target (Nectin-4) Payload Characteristics: Microtubule inhibitors (e.g., MMAE, DM1, Duo-5) disrupt the microtubule dynamics within neuronal axons, impairing axoplasmic transport and leading to distal axonal degeneration. Off-Target Toxicity: Neurons may uptake circulating ADCs through non-specific mechanisms such as macropinocytosis or receptor-mediated endocytosis, resulting in intracellular payload accumulation and cytotoxicity. On-Target Toxicity: Basal expression of specific antigens (e.g., Nectin-4) in peripheral nerve tissues may mediate direct targeted toxicity. Monitor: Assess sensory and motor function prior to each dose. Support: Consider symptomatic treatment (e.g., duloxetine) for neuropathic pain. Intervention: Withhold for Grade ≥ 2/3 (drug-specific) peripheral neuropathy until resolution (to ≥ 1 or baseline); resume at reduced dose for recurrent Grade 2 or first occurrence of Grade 3 events; discontinue for Grade 4. Alopecia Rash Target (Nectin-4) Payload (MMAE) On-Target Toxicity: Nectin-4 is physiologically expressed in human cutaneous structures, including epidermal keratinocytes, sweat glands, and hair follicles, mediating targeted uptake of ADCs into skin tissue. Payload Characteristics: The released MMAE payload accumulates in the skin, disrupting microtubules and inducing keratinocyte apoptosis, which histopathologically manifests as interface dermatitis. In severe cases, massive epidermal necrosis can precipitate life-threatening Stevens-Johnson Syndrome (SJS) or Toxic Epidermal Necrolysis (TEN). Monitor: Skin exam at baseline and each cycle. Intervention: Withhold for Grade ≥ 2/3 (drug-specific) until resolution (≥ G1); resume at reduced dose for recurrent Grade 2 or first Grade 3 event; discontinue for Grade 4. ADC Complex (Fc-Uptake) Payload (DXd) Off-Target Toxicity: Alveolar macrophages expressing Fcγ receptors (FcγRs) or mannose receptors non-specifically internalize ADCs. The subsequent intracellular release of cytotoxic payloads triggers an immune-inflammatory cascade. Payload Characteristics: Membrane-permeable, hydrophobic payloads (e.g., DXd) released by macrophages diffuse into adjacent normal alveolar epithelial cells via the “bystander effect,” causing widespread tissue damage. Monitor: SpO2 and Chest CT at baseline and regularly; prompt workup for new respiratory symptoms. Intervention: Withhold for Grade 1 (asymptomatic); resume only if resolved to Grade 0; permanently discontinue for Grade ≥ 2 (symptomatic), initiate systemic corticosteroids immediately. Payload Target (HER2) Payload Characteristics: Cytotoxic payloads serve as potent teratogens, indiscriminately targeting rapidly dividing embryonic cells, leading to embryo-fetal death or structural anomalies. On-Target Toxicity: The blockade of HER2 signaling in the fetal kidney impairs urine production, causing oligohydramnios, which secondarily induces fetal pulmonary hypoplasia and skeletal abnormalities (Potter sequence). Monitor: Verify pregnancy status prior to initiation. Prophylaxis: Use effective contraception (Females: 7 months post-dose; Males: 4–6 months post-dose); do not breastfeed during treatment and for 7 months post-dose. Intervention: Immediately discontinue if pregnancy confirmed; refer to high-risk obstetrics for oligohydramnios surveillance. Grading System: Toxicity grades refer to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) version 5.0 or 6.0, unless otherwise specified in individual drug labels Management Generalization: Clinical management strategies summarized here are based on FDA/NMPA prescribing information and major clinical guidelines. Physicians should consult the specific drug label for detailed dose modification protocols, as criteria may vary by agent Abbreviations ADC antibody-drug conjugate, ANC absolute neutrophil count, CNS central nervous system, CT computed tomography, CTZ chemoreceptor trigger zone, DXd exatecan derivative, ECHO echocardiogram, G-CSF granulocyte colony-stimulating factor, HbA1c glycated hemoglobin, HCECs human corneal epithelial cells, HER2 human epidermal growth factor receptor 2, ILD interstitial lung disease, LVEF left ventricular ejection fraction, MECs microcyst-like epithelial changes, MMAE monomethyl auristatin E, MUGA multigated acquisition scan, Nectin-4 nectin cell adhesion molecule 4, NRH nodular regenerative hyperplasia, SJS Stevens-Johnson syndrome, SN-38 7-ethyl-10-hydroxycamptothecin, SpO2 peripheral oxygen saturation, TEN toxic epidermal necrolysis, TLS tumor lysis syndrome, TME tumor microenvironment, Topo-I topoisomerase I, TROP2 trophoblast cell-surface antigen 2, UGT1A1 uridine diphosphate glucuronosyltransferase 1A1 Overcoming strategy To fundamentally enhance the safety of ADCs, contemporary research focuses on the precise optimization of their structural architecture and components. Among which, the optimization of linker technology is of primary importance for systemic stability [32]. Research is now concentrating increasingly on the design of linkers, which remain inactive during circulation but can be efficiently cleaved in response to specific tumor enzymes or TME, such as the GGFG tetrapeptide linker used in T-DXd [27]. This design effectively minimizes the premature release of the cytotoxic payload in the blood circulation. Furthermore, third-generation ADCs are taking advantage of advances in site-specific conjugation technologies, including engineered cysteines and unnatural amino acid insertion [112]. These allow homogeneous conjugates of specific DAR, thereby avoiding the formation of high DAR species, as is conventional in traditional random conjugate reactions. By eliminating these heterogeneous fractions, such strategies reduce aggregation tendencies and toxicity risks induced by high DAR, particularly pulmonary toxicity. The exploration of novel payloads now goes beyond traditional microtubule inhibitors and topoisomerase I inhibitors. Emerging classes such as pyrrolobenzodiazepines (PBDs), capable of inducing irreversible DNA crosslinks, hold much promise of retaining therapeutic efficacy while improving safety [113]. Beyond these innovative strategies associated with structural stabilization and component optimization, prospective toxicity mitigation strategies such as “reverse targeting” are under observation. This approach involves simultaneous infusion of specific antibody fragments capable of selective binding of unconjugated drug moieties. By neutralizing cytotoxic agents prematurely released into the circulation, this method protects normal tissues and has shown the ability to increase the maximum tolerated dose in preclinical models [114]. As these technologies continue to mature and advance into clinical translation, ADCs are expected to maintain the efficacy while demonstrating improved safety profiles, ultimately benefiting a greater number of patients. Challenges and strategies ADCs have profoundly reshaped the therapeutic landscape of breast cancer, marked by an expansion into earlier treatment settings (from advanced to early-stage disease) along with a notable diversification of molecular targets beyond HER2 to include TROP2, HER3, etc. Further exploration focuses on continuous innovations in ADC technologies, including advancements in linker stability, novel payload mechanisms of action, and site-specific conjugation techniques. However, the development and use of ADCs also face some obstacles because of the complexity of tumors and the pharmacokinetics involved with these drugs. The mechanisms of various forms of drug resistance, optimal treatment regimen, absolute need for more reliable predictive biomarkers, and ways to properly address the TEAEs associated with ADCs, including ILD, are some of the factors that have to be addressed. Future work will be needed both to understand and overcome these resistances and also to optimize patient selection [115]. Tumor antigen heterogeneity In breast cancer, the concept of tumor antigen heterogeneity refers to the significant differences in molecular phenotypes, gene expression, and functional characteristics among cancer cells within the primary tumor or metastatic lesions [4]. Since ADCs rely on binding to specific antigens on the tumor cell surface to initiate internalization and killing processes, tumor cells that are antigen-negative or express low antigen levels cannot effectively internalize ADCs, thereby evading the cytotoxic effects and persisting as residual lesions leading to relapse [116]. Traditional ADCs like T-DM1 are only effective in tumor cells with high antigen expression that successfully internalize; their efficacy is significantly compromised when treating tumors with low antigen expression or heterogeneity. Lots of efforts are made to enhance therapeutic efficacy and overcome tumor heterogeneity. Improvement of ADC components The TME is characterized by specific features such as high concentrations of proteases (e.g., cathepsins) or an acidic pH. Designing linkers that can specifically cleave within this unique TME enables precise localized drug release at the tumor site. For instance, the TROP-2-targeting agent SHR-A1921 and the B7-H3-targeting agent ICP-B794 both use cleavable linkers designed to be cleaved by the tumor-specific proteases, whereas V66-exatecan targets extracellular DNA uniquely present in necrotic tumor areas [30]. Cytotoxic payloads with membrane-permeable properties can kill the antigen-positive tumor cells while diffusing into the surrounding microenvironment and eliminating adjacent antigen-negative tumor cells. This “bystander effect” allows for more effective clearance of heterogeneous tumor cell populations. Among the approved ADCs, payloads such as DXd (used in T-DXd) and SN-38 (used in SG) have this property that can utilize the potent bystander effect to eliminate tumor cells, significantly broadening the patient population. Next-generation ADCs In addition to component improvement, the structural evolution of ADCs has advanced toward multi-targeted and multi-functional architectures to fundamentally overcome tumor antigen heterogeneity and acquired resistance (Fig. 1). Biparatopic ADCs represent a mechanistically distinct and vital subclass, particularly in the HER2 field [117]. Unlike conventional monoclonal ADCs, biparatopic ADCs simultaneously target two non-overlapping epitopes on the same antigen (e.g., domains II and IV of HER2) [118, 119]. This dual binding promotes extensive receptor clustering and cross-linking on the cell surface, which accelerates rapid lysosomal internalization and degradation, ensuring efficient payload delivery [120]. A prime translational example is JSKN003, a novel biparatopic HER2-targeting ADC. In recent studies, JSKN003 demonstrated highly encouraging anti-tumor activity in heavily pretreated patients, coupled with a highly favorable tolerability profile and extremely low rates of interstitial lung disease, highlighting the vast clinical potential of this design [121, 122]. Furthermore, bispecific ADCs targeting two completely different antigens (e.g., HER2 and TROP2) are currently under intensive development. By requiring the co-expression of two antigens for optimal binding, these agents enhance tumor specificity and combat resistance driven by the downregulation of a single pathway [123, 124]. Additionally, dual-payload ADCs have emerged as an innovative strategy. By conjugating two different cytotoxic payloads (e.g., a topoisomerase I inhibitor and a microtubule inhibitor) onto the same antibody, these agents aim to prevent cross-resistance, induce synergistic cell death, and effectively eradicate highly heterogeneous tumor cell populations that may be refractory to a single mechanism of action [125]. Combinations of ADC and immune checkpoint inhibitors Beyond directly killing the tumor cells, combination strategies involving ADCs and Immune checkpoint inhibitors (ICIs) hold the theoretical potential to overcome the challenge of breast cancer heterogeneity by remodeling the tumor immune microenvironment [126]. Immunologically, breast cancer is usually classified as a “cold tumor”, characterized by insufficient T-cell infiltration and enrichment of immunosuppressive cells (e.g., tumor-associated macrophages, immunosuppressive tumor-associated neutrophils, regulatory T cells, myeloid-derived suppressor cells) in the TME, which stems from its low tumor mutational burden, immunosuppressive microenvironment, and physical barrier [49, 127]. Tumor cells achieve immune escape by inhibiting T-cell activity through expressing an immune checkpoint ligand (e.g., PD-L1). ICIs, such as Pembrolizumab and Atezolizumab, could block these inhibitory signals to restore T-cell-mediated tumor recognition and cytotoxicity [128]. However, the extent of immune infiltration varies across different subtypes of breast cancer: TNBC typically exhibits the most abundant lymphocytic infiltration, followed by HER2-positive disease; in contrast, the immune microenvironment in HR-positive/HER2-negative breast cancer is highly immunosuppressive [49]. Paradoxically, TNBC, which was previously considered the most challenging subtype to manage, now demonstrates the highest sensitivity to ICI interventions. Overall, due to the immunosuppressive status of most breast cancer subtypes, ICIs exhibit low response rates in breast cancer treatment and thus have relatively limited applications. When ADCs carrying potent payloads kill cancer cells, they induce immunogenic cell death (ICD), releasing large amounts of tumor-associated antigens and damage-associated molecular patterns (e.g., HMGB1, ATP), which may in turn activate the immune system and turn “cold tumors” into “hot tumors"[129]. These released signals are hypothesized to promote dendritic cell (DC) maturation and antigen presentation, ultimately enhancing CD8 + T cell-mediated tumor recognition and cytotoxicity. Furthermore, ADCs can recruit and activate natural killer (NK) cells and macrophages via ADCC and antibody-dependent cellular phagocytosis (ADCP). Through this ICD-driven immune infiltration, ADCs may theoretically facilitate the transition of immunologically “cold” tumors into a more inflamed “hot” phenotype. This innate immune activation can subsequently induce the upregulation of immune checkpoints, including PD-L1, creating a more suitable environment for ICI therapy [130]. The combination strategy has already shown early signals of efficacy in the KATE2 trial: Although T-DM1 plus atezolizumab yielded no PFS advantage in the intention-to-treat population, clinical benefit emerged within the PD-L1-positive subgroup [131]. Moving to the first-line setting for advanced TNBC, the BEGONIA trial tested Dato-DXd combined with durvalumab. Preliminary results present a high ORR of 79%, with efficacy observed regardless of the PD-L1 status [132, 133]. However, the combination regimens are accompanied by the risk of particular toxicity. For example, the incidence of ILD is elevated under T-DXd plus PD-1 inhibitor combinations compared to either agent alone [110]. Most importantly, the optimal sequencing of these agents remains elusive. While early preclinical models suggest that a sequential strategy—administering ADC prior to ICI—might be advantageous for pre-conditioning the immunosuppressive TME, robust clinical validation is still required. It remains to be determined whether sequential priming or concurrent administration yields the optimal balance between synergistic immune activation and cumulative toxicity [129]. Beyond combination strategies, emerging drug development focuses on integrating the advantages of ADCs and ICIs: Bispecific antibodies targeting both tumor antigens and immune-modulatory molecules that can promote the killing effect towards tumor cells by immune cells [134]; immune-checkpoint-targeting drug conjugates and immunostimulatory antibody conjugates, which aim to achieve targeted tumor elimination while activating innate immune responses directly within the tumor site [128, 135]. Resistance The therapeutic efficacy of ADCs relies on a precise, interlinked cascade: stability in systemic circulation, specific antigen recognition, efficient internalization, lysosomal payload release, apoptosis induction, and bystander killing [136]. While this “magic bullet” design aims to maximize efficacy while minimizing toxicity, any disruption within the process can influence the therapeutic effect and induce resistance. Elucidating these mechanisms is critical for developing next-generation ADCs and rational combination strategies. Table 6 elaborates the principal mechanisms of ADCs and the relevant overcoming strategies [137–149]. Table 6Mechanisms of Resistance to ADCs in Breast Cancer and Overcoming StrategiesCategorySpecific MechanismImpact on EfficacyOvercoming StrategiesAntigen-DependentAntigen downregulation/lossLoss of specific binding and targeting capabilityBispecific ADCs: Target dual antigens or distinct epitopes to mitigate single-antigen lossBystander effect: Exploit payload diffusion to kill antigen-negative cellsAntigen mutation/truncationLoss or alteration of epitopes impairs recognition and internalizationCombination with TKIs: Agents like neratinib induce degradation of truncated receptors (e.g., p95HER2)Internalization and ProcessingImpaired internalizationSurface retention or mis-sorting of ADC-antigen complex prevents payload releaseEnhance internalization: Target clathrin-independent pathways (e.g., modulate SNX10-RAB11A axis)Non-internalizing ADCs: Design linkers for extracellular cleavage within the tumor matrixLysosomal dysfunctionElevated lysosomal pH or reduced protease activity inhibits linker cleavageNovel linkers: Use pH-independent linkers or substrates for alternative enzymes (phosphatases, glucuronidases)Co-administration: Combine with lysosomal acidifiersPayload-RelatedEfflux pump upregulationMDR1 (P-gp/ABCB1) transporters efflux hydrophobic payloads, reducing intracellular concentrationHydrophilic modification: PEGylation of linkers/payloads to evade efflux; combine with efflux inhibitorsNovel payloads: Use cytotoxic agents not recognized by MDR1 transportersTarget alterationStructural changes in targets (tubulin/topoisomerase) reduce payload potencyMechanism switching: Sequential use of ADCs with distinct cytotoxic mechanismsDual-payload ADCs: Conjugate dual agents with complementary mechanismsEnhanced anti-apoptotic signalingUpregulation of BCL-2 or MCL-1 blocks apoptosis initiationCombination therapy: Co-administer BCL-2 inhibitors or immune checkpoint inhibitors (ICIs)Tumor MicroenvironmentPhysical barriersDense extracellular matrix restricts drug penetrationStroma-targeting ADCs: Target cancer-associated fibroblasts (CAFs) or tumor vasculature antigensImmunosuppressionAn immunosuppressive “cold” environment inhibits immune-mediated killingImmunotherapy combination: Combine with ICIs to exploit ADC-induced immunogenic cell death (ICD) for “cold-to-hot” tumor conversion Mechanisms of Resistance to ADCs in Breast Cancer and Overcoming Strategies Lessons from failed ADC programs While the recent approvals of novel ADCs have revolutionized breast cancer treatment, critically evaluating discontinued programs and terminated programs (as detailed in Tables 1, 2 and 3) provides indispensable translational lessons for future drug design. For instance, within the lineage and oncofetal antigens category, agents targeting PTK7 (PF-06647020) and Mesothelin (BAY94-9343) were discontinued despite strong preclinical rationales, largely due to a narrow therapeutic window [150, 151]. Although highly expressed in tumors, these antigens often exhibit baseline expression in normal tissues like skin, lungs, or liver, leading to prohibitive “on-target/off-tumor” toxicities before therapeutic efficacy is achieved. Similarly, TME-targeting antigens like LRRC15 (ABBV-085) and lineage/oncofetal antigens, including ROR1 (MK-2140), B7H4 (SGN-B7H4V), and B7H3 (MGC018), have also encountered substantial hurdles in breast cancer (Tables 1, 2 and 3). Beyond target selection, structural components significantly impact clinical viability. Many of these discontinued ADCs utilized early-generation microtubule inhibitors or DNA alkylators, where premature linker cleavage in systemic circulation released free cytotoxic payloads, triggering severe off-target toxicities that mimicked conventional chemotherapy and limited the maximum tolerated dose [43]. In addition, the failure of agents like glembatumumab vedotin (targeting gpNMB) in triple-negative breast cancer highlights the critical challenge of antigen heterogeneity and the limitation of treating unselected patient populations [152]. Without precise, biomarker-driven patient selection to identify those with truly high target reliance, therapeutic benefits are easily diluted by tumor heterogeneity. These failures emphasize that the success of next-generation ADCs hinges not only on biological druggability but also on high tumor specificity, optimal linker stability, a potent bystander effect, and rigorous biomarker stratification to secure a viable therapeutic window [153]. Conclusions and future perspectives ADCs have already revolutionarily redefined the treatment of breast cancer, evolving from a therapeutic concept to the cornerstones of current clinical practice. The therapeutic landscape of breast cancer ADCs has witnessed progressive expansion, with indications gradually evolving to encompass HER2-low and HER2-ultralow subtypes, as well as TNBC, ultimately extending survival benefits to previously underserved patient populations. Nevertheless, the journey is far from complete. The clinical management of ADCs is increasingly confronted by the emergence of resistance mechanisms, the need for optimal sequencing, and the management of distinct toxicities. Beyond clinical practice, the advent of ADCs is reshaping basic and translational research paradigms. While traditional targeted therapies focus on blocking oncogenic signaling pathways, the ADC paradigm utilizes cell-surface antigens primarily as specific conduits for intracellular drug delivery. Consequently, basic research is pivoting toward the intricate dynamics of receptor-mediated endocytosis, lysosomal trafficking, and the bystander effect within the tumor microenvironment. Translationally, this shift requires a departure from traditional binary biomarker assessments (e.g., positive versus negative expression) toward multidimensional profiling. Future translational models must integrate quantitative antigen density, lysosomal enzymatic activity, and tumor-stroma spatial interactions to develop predictive biomarkers and guide rationale-based combination strategies. Moving forward, the next phase of ADC development depends on a deeper mechanistic understanding of tumor biology. By transitioning into an era of precision ADC therapy, these agents hold the promise to not only prolong survival but ultimately achieve curative outcomes for patients with breast cancer. Acknowledgements Not applicable. Abbreviations Advanced breast cancer Antibody-drug conjugate Adverse events Breast cancer Brain metastases Drug-to-antibody ratio Durable disease control Dose-limiting toxicity DNA alkylating agent Early breast cancer Endocrine therapy Hormone receptor Monoclonal antibody Metastatic breast cancer Maximum tolerated dose Microtubule inhibitor Metastatic triple-negative breast cancer National Medical Products Administration Objective response rate Overall survival Pathologic complete response Progression-free survival Pharmacokinetics Recommended phase 2 dose Triple-negative breast cancer Topoisomerase I inhibitor Treatment of physician’s choice Authors’ contributions SYL designed the framework of this review, conducted evidence synthesis and visualization, and wrote the original draft. ZMX, WHR, XQW, and HMY contributed to scientific knowledge interpretation anddiscussion. MY and PFF supervised the study and revised the manuscript. All authors approved the final version of the manuscript. Funding This work was supported by the National Natural Science Foundation of China (Grant No. 82503827) and the Science and Technology Project of Zhejiang Province (Grant No. 2025C01107). 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. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Ming Yi, Email: mingyi_onco@outlook.com, Email: mingyi2022@zju.edu.cn. Peifen Fu, Email: fupeifen@zju.edu.cn. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.