Recent advances in bispecific antibody-drug conjugates for breast cancer therapy

Abstract

Purpose

Bispecific antibody-drug conjugates (BsADCs) represent a promising strategy to overcome limitations of conventional ADCs in breast cancer, such as tumor heterogeneity and inefficient internalization. This review summarizes recent advances and the therapeutic potential of BsADCs.

Methods

We conducted a literature review, focusing on BsADC candidates selected for clinical relevance (e.g., ZW49, BL-B01D1), mechanistic innovation (e.g., biparatopic targeting, engaging fast-internalizing receptors), and potential in challenging subtypes like triple-negative breast cancer (TNBC).

Results

BsADCs enhance drug delivery through dual targeting. Biparatopic HER2-targeting agents (e.g., ZW49, JSKN003) induce receptor clustering and robust internalization. BsADCs co-engaging rapidly internalizing receptors (e.g., HER2×CD63) hijack efficient endocytic pathways, showing activity even in low HER2-expression models. Furthermore, BsADCs targeting compensatory pathways, such as EGFR×HER3 (BL-B01D1) and TROP2×HER3 (JSKN016), have demonstrated breakthrough efficacy in TNBC. Optimization of linker technology and drug-to-antibody ratio (DAR) has improved stability and the therapeutic window, enabling the progression of several BsADCs into Phase III trials.

Conclusion

BsADCs are a transformative therapeutic modality for breast cancer. Their ability to enhance tumor selectivity, overcome heterogeneity, and target resistant pathways positions them as key players in the future oncology landscape, with ongoing trials poised to define their clinical role.

Introduction

Breast cancer is one of the most common malignant tumors among women worldwide, and the incidence rate is increasing year by year [1, 2]. The treatment landscape for breast cancer has profoundly evolved into an era of precise molecular subtyping and individualized therapy [3]. Despite these advances, formidable challenges such as intrinsic and acquired drug resistance, profound tumor heterogeneity, and target escape remain the major bottlenecks to achieving durable clinical responses [4].

Antibody-drug conjugates (ADCs) are immunoconjugates comprised of a monoclonal antibody tethered to a cytotoxic drug (known as the payload) via a chemical linker [5]. ADC is a vector-based chemotherapy that allows the selective delivery of a potent cytotoxic agent within a tumor [6]. Traditionally, ADCs were produced through stochastic conjugation, leading to a heterogeneous mixture of molecules with varying DARs and conjugation sites. However, there is a strong trend towards site-specific conjugation technologies (e.g., engineering cysteines, incorporating non-natural amino acids, or utilizing enzymatic conjugation). These methods generate homogeneous ADCs with defined DARs, which often exhibit improved pharmacokinetics, enhanced stability, and a broader therapeutic window compared to their heterogeneous counterparts [7].

Currently, ADCs are an important class of cancer therapeutics, with seventeen ADCs approved so far by the US Food and Drug Administration [8]. Despite all the advancements in cancer therapy, inherent and acquired drug resistance continues to be major obstacles to successful treatment. Multiple mechanisms of resistance to ADCs have been reported, including antigen-related resistance, failure in internalization, impaired lysosomal function, drug-efflux pumps, and alterations of targets [9]. A prospective approach to address the aforementioned challenges involves the conjugation of a linker-payload complex to Bispecific antibodies (BsAbs), giving rise to the concept of Bispecific antibody drug conjugates (BsADCs). The clinical rationale for this dual-targeting approach is powerfully demonstrated by the success of the trastuzumab and pertuzumab combination in HER2-positive breast cancer. By targeting domains IV and II of HER2 respectively, this standard-of-care regimen synergistically inhibits HER2 signaling and has become a standard-of-care in both metastatic and early-stage HER2-positive breast cancer, providing a strong proof-of-concept [10]. In contrast to traditional ADCs, the unique dual epitope/target binding modes of BsADCs not only enable binding to co-expressed antigens in solid tumors to enhance selectivity but also significantly improve internalization. These distinctive advantages position BsADCs as a substantial force in the realm of next-generation ADCs. The rationale for enhanced internalization by BsADCs is multifaceted and represents a core advantage over conventional ADCs. As illustrated in Fig. 1, this enhancement can occur through several mechanisms: (1) Cooperative Internalization and Clustering: Biparatopic BsADCs binding two non-overlapping epitopes on the same antigen can cross-link and cluster receptors, forming large aggregates that are more efficiently internalized via endocytosis. (2) Leveraging Rapidly Internalizing Receptors: BsADCs targeting a tumor-associated antigen (e.g., HER2) and a second, rapidly internalizing receptor (e.g., CD63, PRLR) can exploit the efficient endocytic machinery of the latter, effectively ‘shuttling’ the ADC into the cell even if the primary tumor antigen has poor internalization kinetics.

Fig. 1.

Mechanisms of enhanced internalization by Bispecific antibody-drug conjugates (BsADCs)

(A) A conventional monospecific ADC shows limited internalization efficiency. In contrast, BsADCs enhance payload delivery through two primary strategies: (B) A biparatopic BsADC induces receptor clustering and robust internalization by binding two distinct epitopes on the same tumor-associated antigen. (C) A Bispecific ADC co-engages a tumor-specific antigen and a rapidly internalizing “shuttle” receptor (e.g., CD63), hijacking the latter’s efficient endocytic pathway for improved cellular uptake. The internalized complexes are trafficked through early and late endosomes to lysosomes, where cytotoxic payloads are released.)

This review primarily focuses on BsADCs in breast cancer that exemplify key strategic advances and have reached at least the stage of preclinical proof-of-concept. We have selected representative candidates based on the following criteria: (1) Clinical relevance: BsADCs that have entered clinical trials and generated reported data (e.g., ZW49, MEDI4276, BL-B01D1, JSKN016). (2) Mechanistic innovation: BsADCs that illustrate distinct targeting paradigms, such as biparatopic targeting for enhanced internalization (e.g., JSKN003) or engaging fast-internalizing receptors (e.g., HER2×CD63). (3) Therapeutic potential for challenging subtypes: Agents showing breakthrough efficacy in triple-negative breast cancer (TNBC) or against tumors with low antigen expression. While this selection is not exhaustive, it provides a comprehensive overview of the current landscape and the most promising design strategies in the field.

Structure of BsADC

The core structure of BsADC includes a Bispecific antibody (bsAb), a linker, and a payload. Following the binding to cancer cell-surface antigen, BsADC is internalized by receptor-mediated endocytosis, and the payload is released by the degradation of the linker or the antibody in the endolysosomal compartment [11]. Therefore, the development of BsADCs requires careful consideration of multiple factors, including target antigen biology, antibody specificity, payload cytotoxicity and mechanism of action, linker stability and cleavage, and conjugation sites. Beyond these factors, an inherent consideration in Bispecific targeting is the potential requirement for co-expression of both target antigens on the same tumor cell for optimal activity. An inherent consideration in Bispecific targeting is the potential requirement for co-expression of both target antigens on the same tumor cell for optimal activity. This concept could, in theory, restrict the target cell population compared to a monospecific ADC. However, this potential limitation is counterbalanced by two key factors. First, the enhanced internalization and potency on double-positive cells may more effectively eradicate this critical subpopulation. Second, and crucially, the use of cleavable linkers enables the ‘bystander effect’, whereby membrane-permeable payloads released in double-positive cells can diffuse and kill adjacent tumor cells expressing only one or even none of the target antigens [12]. This effect is vital for overcoming tumor heterogeneity and ensuring comprehensive tumor cell killing. The structure of BsADCs integrates several critical design components that collectively determine their therapeutic efficacy and safety profile. As summarized in Fig. 2, these include the Bispecific antibody format, linker chemistry, and cytotoxic payload, each offering distinct strategic advantages.

Fig. 2.

Key design components and strategies of Bispecific antibody-drug conjugates (BsADCs)

• (A)General architecture of a BsADC, comprising a Bispecific antibody, chemical linker, and cytotoxic payload.
• (B)Classification of Bispecific antibody formats, distinguishing between Bispecific (dual-antigen) targeting (e.g., HER2 × CD63) and biparatopic (dual-epitope) targeting (e.g., HER2 ECD2 × ECD4).
• (C)Linker technologies, detailing cleavable linkers including protease-cleavable (e.g., Val-Cit), reduction-sensitive (e.g., disulfide), and acid-labile (e.g., hydrazone) types, as well as non-cleavable linkers (e.g., SMCC).
• (D)Diverse payload classes, encompassing microtubule inhibitors (e.g., auristatins, maytansinoids), DNA-damaging agents (e.g., calicheamicin, duocarmycins, PBDs), topoisomerase I inhibitors (e.g., DXd, SN-38), and other emerging payloads (e.g., RNA polymerase II inhibitors such as amanitins).

Bispecific antibodies(BsAb)

Bispecific antibodies (bsAbs) bind two different epitopes on the same or different antigens. Through this dual specificity for soluble or cell-surface antigens, bsAbs exert activities beyond those of natural antibodies, offering numerous opportunities for therapeutic applications [13]. A critical distinction must be made between Bispecific and biparatopic antibodies. Bispecific antibodies (BsAbs) constitute a broad category defined by their ability to bind two different antigens or epitopes. This category can be divided into two primary classes: those that engage two distinct antigens (e.g., HER2 and CD63), and biparatopic antibodies, which are a specialized subset that bind two distinct, non-overlapping epitopes on the same antigen (e.g., two different domains of HER2). Consequently, while all biparatopic antibodies are Bispecific, not all Bispecific antibodies are biparatopic. Both strategies are leveraged in BsADC design, with biparatopic targeting primarily aimed at enhancing receptor clustering and internalization, while Bispecific targeting across different antigens can engage distinct biological pathways or exploit differential internalization rates [14]. Antibodies consist of two heavy chains and light chains, encompassing the Fc region and Fab region. When constructing ADCs, the fundamental requirements for the antibody skeleton include a suitable circulating half-life and low immunogenicity. More than 100 different bsAb formats have been reported according to a recent review. While their molecular architectures are different, virtually all of them can be grouped into one of the two large categories: fragment-based bsAbs and Fc-based bsAbs [11]. Fragment-based bsAbs, consisting exclusively of antibody variable domains, are a major class of bsAb formats. They are in general smaller in size than Fc-based bsAbs and hence, show better tissue penetration in vivo [15]. Meanwhile, the loss of binding to Fc gamma receptor (FcγR) reduces Fc mediated nonspecific drug uptake [16]. Conversely, the loss of binding to the neonatal Fc receptor (FcRn) leads to a relatively short half-life in plasma [17]. To circumvent this limitation, strategies such as PEGylation or the introduction of albumin-binding motifs are actively explored to prolong the plasma half-life of these smaller constructs [18, 19]. This creates a fundamental design trade-off: optimizing for superior tissue penetration often comes at the cost of plasma persistence, and vice versa. While there are a variety of different fragment-based bsAb formats, three of them have been more extensively investigated in preclinical and clinical studies: tandem scFvs, dual affinity retargeting (DART) proteins, and tandem diabodies (TandAbs) [5]. Fc-based bsAbs have homo- or heterodimeric Fc domains (and less frequently CH3 domains only) to which Fab, scFv or sdAb are attached through a peptide linker [20]. Fc-containing BsADCs bring additional advantages such as antibody-dependent cell-mediated cytotoxicity (ADCC), complement-dependent cytotoxicity (CDC), immune phagocytosis, and cytokine release, which collectively contribute to tumor killing [21]. However, the binding to FcγR during endocytosis may lead to the internalization of BsADCs in normal tissues, resulting in off-target toxicity [22]. This risk can be mitigated through Fc engineering, such as introducing point mutations (e.g., L234A/L235A, or N297A) to reduce or ablate FcγR binding, thereby minimizing Fc-mediated, antigen-independent uptake [23]. While this challenge is inherent to all Fc-containing ADCs, the dual-targeting nature of BsADCs necessitates particularly careful optimization of Fc function to balance potential efficacy (e.g., ADCC) against the risk of off-target toxicity. Fc-based bsAbs can be categorized into two large groups: symmetric and asymmetric. Symmetric Fc-based bsAbs typically have additional Fv or scFv moieties at the N- and/or C-termini of the polypeptide chains, making them larger than conventional IgG antibodies. On the other hand, asymmetric Fc-based bsAbs are produced by the preferential heterodimerization of two engineered Fcs, making them identical in size and shape to conventional IgG and each of the two arms of the bsAb recognizing a different antigen [24].

The paramount considerations in the construction of BsADCs lie in the judicious selection of a suitable target combination. Target selection serves as a fundamental prerequisite for the successful development of ADCs, exerting a pivotal influence on the ultimate therapeutic window and systemic toxicity. Ideally, the target antigen for ADC needs to be overexpressed on cancer cell surfaces with no or negligible expression on normal healthy tissues [25]. Owing to the unique dual targeting characteristics of BsADCs, a comprehensive consideration of the deep-level effects of antigen combinations is essential. This encompasses factors include internalization, recycling, turnover rates, lysosomal degradation, and obligate mechanisms [26].

Linkers

The linker connects the antibody and the cytotoxic payload and is a key component in the function of BsADCs. The linker imparts the following characteristics to BsADCs: (1) high stability in the circulation, and (2) specific release of payload in the target tissue [27].

The first type is the cleavable linker, which is designed to be selectively cleaved in the tumor microenvironment or inside the cancer cell to release the cytotoxic payload. The most prevalent and clinically validated strategy utilizes enzymatically cleavable linkers, which leverage the high activity of specific lysosomal proteases (e.g., cathepsin B). Other important mechanisms include acid-labile linkers that hydrolyze in the markedly lower pH within endolysosomes (pH ~ 4.5–5.0) compared to blood (pH ~ 7.4), and reducible disulfide linkers that are cleaved by elevated concentrations of reducing agents like glutathione (GSH) in the cytosol [28]. Based on these mechanisms, cleavable linkers are categorized into three major classes [29]. Protease-cleavable linkers, exemplified by the Valine-Citrulline (Val-Cit) dipeptide, are stable in plasma and are selectively cleaved by lysosomal proteases such as cathepsin B following ADC internalization [30]. This strategy is a cornerstone of many modern ADCs, including brentuximab vedotin [31]. In contrast, acid-labile linkers (e.g., hydrazone) are designed to hydrolyze in the acidic compartments of the endo-lysosomal system; however, their susceptibility to hydrolysis at neutral pH can contribute to premature payload release and off-target toxicity, a limitation observed in the first-generation ADC gemtuzumab ozogamicin [32]. Finally, disulfide linkers rely on the high reducing potential within the cell cytosol for cleavage by glutathione. The in vivo stability of disulfide linkers can be enhanced through structural engineering to introduce steric hindrance [33]. Beyond targeted release, a pivotal advantage of cleavable linkers is their ability to generate membrane-permeable payload catabolites. This property enables the “bystander effect,” whereby the cytotoxic agent can diffuse from an antigen-positive tumor cell to kill adjacent antigen-negative cancer cells. This effect is crucial for overcoming tumor heterogeneity [34]. Ultimately, the central challenge for all cleavable linker technologies lies in balancing this potent efficacy against the risk of off-target toxicity from premature cleavage.

The second type of linker is non-cleavable. In contrast to the cleavable linker, there are no chemical triggers in this structure, and the linker is part of the payload. Non-cleavable linkers (e.g., SMCC), characterized by high stability in plasma, can only undergo degradation in the lysosome to release the payload. This linker design typically results in superior plasma stability, which can contribute to an improved therapeutic index by minimizing premature payload release [35]. However, it is important to note that non-cleavable linkers are not devoid of off-target toxicity. For instance, ADCs utilizing non-cleavable linkers have been associated with significant ocular toxicity, as seen with tisotumab vedotin, indicating that the released payload metabolite can still affect sensitive tissues [36].

Cytotoxic payloads

The cytotoxic payload is the warhead that exerts cytotoxicity after internalization of BsADCs into cancer cells. Cytotoxic payloads, integral to BsADCs, play a crucial role in determining the overall antitumor effect and potential adverse reactions. The activity and physicochemical properties of the payload have a direct impact on the antitumor efficacy of ADC drugs. The mechanism of action of the payload is an important factor determining the performance of the ADC (e.g., adverse reactions). Besides, certain other characteristics of ADC payloads, such as cytotoxicity, immunogenicity, stability of storage during preparation and circulation, water solubility, and modifiability are also important [37].

The ideal payloads should have the following characteristics. First, they should have adequately high cytotoxicity, typically with half-maximal inhibitory concentration (IC₅₀) values in the picomolar (pM) to low nanomolar (nM) range (e.g., 0.01–1.01 nM) against target tumor cell lines [38]. The fundamental reason ADC payloads require extreme potency is that the actual number of molecules reaching the cytosol of each tumor cell is exceedingly low. This is driven by two factors: the limited repertoire of safe and specific antigen targets, and the overall low delivery efficiency–collectively caused by the poor permeability of antibodies in solid tumors and the intrinsically inefficient internalization process [12]. Second, payloads must be sufficiently hydrophilic or incorporate structural features (e.g., polar groups, solubilizing linkers) to achieve adequate water solubility (often targeting > 0.1–1.1 mg/mL) to prevent aggregation during ADC formulation and in circulation [39]. Third, ADC payloads should have sufficiently low immunogenicity. Protein drugs have the risk to induce immunogenicity, which may negatively affect the ADC efficacy or even lead to mortality of the treated patients [40]. Fourth, high stability is paramount. The linker-payload complex should be stable in the systemic circulation to minimize off-target toxicity from premature release, ideally exhibiting a long half-life (> 7 days) in plasma. An ideal linker must achieve a delicate balance between sufficient stability during circulation and efficient release of the cytotoxic payload specifically within the tumor. Furthermore, the conjugated ADC must demonstrate high chemical stability in formulation, with minimal degradation or aggregation, which can be assessed by techniques such as size-exclusion chromatography (SEC) and sedimentation velocity analytical ultracentrifugation (SV-AUC) [41]. Fifth, ADC payloads should have functional groups that can be modified without significantly affecting their potency. The payload must have a modifiable functional group or a site that can conjugate to the monoclonal antibody. The site of the modification must be carefully selected to preserve the potency of the parental drug. More importantly, when using non-cleavable linkers, the payload must retain its potency even after the antibody degrades [42]. Finally, the ADC payloads should have bystander effects. Some ADC drugs are internalized and release small, uncharged, permeable membrane hydrophobic molecules, which spread through the cell membrane and kill tumor cells with negative expression of adjacent antigens. This process is known as the “bystander effect” and has important implications on tumor cells with uneven antigen-expression [43].

Currently, cytotoxic payloads used in ADCs mainly include potent microtubule inhibitors, DNA-damaging agents, and immunomodulators [44]. Microtubules are an important part of the cytoskeleton in eukaryotic cells. Microtubules play an important role in maintaining various cellular functions such as cell morphology, signal transduction, organelle transportation, cell motility, cell division, and mitosis, and are important targets for tumor therapy. Tubulin is the composition of microtubules, and tubulin inhibitors interfere with the dynamic combination of microtubules by binding to tubulin, arresting cells in the G2/M phase of the cell cycle, and ultimately leading to apoptosis [45]. Compared to tubulin inhibitors, DNA-damaging agents (e.g., calicheamicin, duocarmycins) can act throughout the cell cycle and have demonstrated efficacy in solid tumors [46]. It was historically hypothesized that due to their extreme potency, fewer molecules of DNA-damaging payloads would be required per cell to achieve cell kill. However, the clinical development of some ultra-potent DNA-intercalators like the pyrrolobenzodiazepine (PBD) dimers has been challenged by significant off-target toxicities, potentially related to their persistent DNA damage and the bystander effect [47]. In contrast, topoisomerase I inhibitors (e.g., DXd, SN-38) have emerged as a highly successful class of DNA-damaging payloads, offering a favorable balance of high potency and manageable safety profiles, as evidenced by the success of ADCs like trastuzumab deruxtecan (T-DXd) [48].

Drug-to-antibody ratio (DAR)

Drug-to-antibody ratio (DAR) represents the number of linker-drugs attached to a given antibody. There are substantial differences in the in vivo pharmacokinetics of various drug-carrying forms of ADCs (DAR = 0–8). For BsADCs, the DAR typically ranges between 2 and 4. This range strikes an optimal balance between potency and pharmacokinetic properties [49]. Concrete preclinical and clinical data substantiate the significant impact of DAR on pharmacokinetic profiles. For instance, in a murine model, a site-specific trastuzumab-based ADC with a DAR of 6 exhibited a faster clearance and a shorter half-life compared to its DAR 4 counterpart when measuring conjugated antibody (the intact ADC) [50]. This phenomenon, often referred to as “DAR-dependent clearance,” is attributed to increased hydrophobicity and non-specific uptake by the mononuclear phagocyte system at higher DARs. Clinical observations further support this; population PK modeling of the DM4-based ADC, tusamitamab ravtansine (DAR ~ 3.8), characterized the irreversible deconjugation from higher to lower DAR species as a key determinant of its overall disposition [51]. Consequently, for less potent payloads, higher DAR values may be necessary to achieve sufficient cell-killing effects. However, a simple increase in payload conjugation at high DAR can lead to conjugate hydrophobicity, leading to faster clearance by the mononuclear phagocyte system and accelerated plasma elimination, rather than renal clearance which is predominant for small molecules [52].

The mechanism of action of BsADCs, from cell surface binding to payload release and induction of cell death, including the bystander effect, is depicted in Fig. 3.

Fig. 3.

Mechanism of action and bystander effect of Bispecific antibody-drug conjugates (BsADCs). A: The antibody binds to the antigen1 and antigen2 and initiates the internalization process. B, C: BsADC-antigen complex gradually forms an endosome. (drugs release from cleavable linkers). D: BsADC-antigen complex is degraded within a lysosome. (drugs release from cleavable and non-cleavable linkers). E: The payload produces various cytotoxic effects, such as the following; E1, microtubule disruption; E2, DNA damage. F: Bystander effect in neighbor tumor cells not expressing the antigens

Progress of BsADCs in breast cancer research

BsADCs on HER2

Breast cancer is the most common cancer type among women worldwide, with HER2-positive breast cancer accounting for 15% to 20% of all breast cancer cases [53]. Clinically approved HER2-targeting monoclonal antibodies (mAbs) and ADCs have significantly transformed treatment options for HER2-positive breast cancer, yielding clinical benefits. However, the expression heterogeneity of HER2 poses a limitation to the efficacy of current anti-HER2 therapies in cases of relatively low HER2 levels. This not only narrows the range of indications for HER2 therapy but also fosters drug resistance under therapeutic pressure81. Additionally, the potential resistance of HER2 to internalization diminishes drug efficacy. Fortunately, the dual binding model of BsADCs offers a promising solution to the current challenges of poor internalization and drug resistance in HER2-targeting for breast cancer.

Dual-epitope BsADC targeting HER2

ZW49:

ZW49 is generated from the conjugation of a novel N-acyl sulfonamide auristatin payload to the inter-chain disulfide bond cysteines of the Bispecific anti-HER2 IgG1 antibody ZW25 (now known by its generic name, zanidatamab), via a protease cleavable linker. Auristatin-based payloads, such as the conventional monomethyl auristatin E (MMAE), are exceptionally potent microtubule inhibitors with IC₅₀ values typically ranging from 10 to 100 pM against various cancer cell lines. The strategic physicochemical modifications in this novel variant are designed to fine-tune its membrane permeability, thereby aiming to control the bystander effect and expand the therapeutic window. Zanidatamab is an asymmetric IgG1-like molecule produced using the Azymetric™ platform, which incorporates engineered CH3 domains (“knobs-into-holes” and electrostatic steering) to ensure heavy-chain heterodimerization and correct assembly. This site-specific conjugation strategy, targeting the hinge cysteine residues, enables the generation of a highly homogeneous ADC population with a controlled DAR of 2–3, as characterized by hydrophobic interaction chromatography (HIC), minimizing the heterogeneity and aggregation risks common with stochastic conjugation methods [54]. The conjugate exhibits high stability in circulation, as the protease-cleavable linker–which incorporates a valine-citrulline (Val-Cit) dipeptide sequence–remains intact in plasma. However, upon internalization and trafficking to the lysosome, the Val-Cit motif is efficiently cleaved by cathepsin B, enabling rapid payload release kinetics. This efficient intracellular release mechanism, combined with the membrane-permeable nature of the payload, is critical for mediating a potent bystander effect, allowing it to diffuse into and kill adjacent cancer cells that may have heterogeneous or low HER2 expression, thereby overcoming a key mechanism of tumor resistance [55]. In cellular binding assays, it was confirmed that the payload conjugation to ZW25 did not affect the antibody’s binding to HER2-expressing cells [56]. Zanidatamab (formerly ZW25) was constructed using the IgG1-like heterodimeric Azymetric™ Fc platform, with an anti-HER2-ECD4 single chain variable fragment (scFv) linked to heavy chain 1 and an anti-HER2-ECD2 fragment antigen-binding (Fab) domain on heavy chain 2. Zanidatamab can bind two HER2 via a trans receptor binding model, inducing the formation of large, polarized aggregate clusters. This not only improves the internalization effect but also promotes antibody hexamerization, achieving enhanced CDC effects [57]. ZW49 incorporates a novel payload, N-acyl sulfonamide auristatin, making it well-tolerated. The favorable tolerability profile of ZW49 is attributed to strategic physicochemical modifications conferred by the N-acyl sulfonamide group on the auristatin payload. This moiety acts as a carboxylic acid bioisostere, which significantly reduces the payload’s intrinsic lipophilicity compared to conventional auristatins like monomethyl auristatin E (MMAE) [58]. The resultant decrease in membrane permeability confines the cytotoxic activity more effectively within target cells, thereby mitigating the potent bystander effect–a known contributor to on-target, off-tumor toxicity in healthy tissues expressing low levels of HER2. This controlled bystander activity helps expand the therapeutic window [59]. Furthermore, the enhanced hydrophilicity of the N-acyl sulfonamide auristatin likely improves the solubility of the conjugate and contributes to superior plasma stability by minimizing aggregation and non-specific interactions, which collectively reduce the risk of premature payload release and systemic exposure [38]. The Bispecific antibody nature of ZW49 contributes to superior internalization compared to trastuzumab. Its Fc region imparts ADCC, ADCP, and CDC effects, while the hexamer mode of HER2 enhances CDC and internalization [56].

Preclinical data indicates that ZW49 exhibits a potent tumor-killing effect and good tolerance without compromising HER2 affinity (highest non-severely toxic dose = 18 mg/kg). Currently, ZW49 is undergoing a phase I clinical trial (NCT03821233). Preliminary results have suggested promising efficacy in various types of HER2-positive tumors. In eight breast cancer patients with a median of six prior therapies treated at the cohort expansion recommended dose, an ORR of 13% was observed (1 of 8 patients).Toxicity analysis revealed two cases of G2 keratitis lasting more than 14 days; approximately 43% of patients exhibited keratitis, but all events decreased to G1 or eventually resolved. There were no interstitial lung disease (ILD) events or deaths related to treatment [60]. However, subsequent clinical development of ZW49 appears to have been deprioritized in favor of other assets within the Zymeworks pipeline, and its future development strategy remains to be clarified. This strategic decision likely stems from a confluence of factors beyond the clinical data itself. Firstly, the competitive landscape for HER2-targeting ADCs became increasingly crowded, most notably with the approval and immense success of trastuzumab deruxtecan (T-DXd), which set a very high bar for efficacy [48]. Secondly, the observed limited efficacy (ORR of 13% in the phase I breast cancer cohort), coupled with a significant rate of ocular toxicity (keratitis), resulted in a potentially narrow therapeutic index for ZW49 [60]. Within this context, the parent antibody, zanidatamab (the Bispecific antibody component of ZW49), demonstrated compelling clinical activity both as a monotherapy and in combinations, presenting a broader and more immediate developmental path [61]. Consequently, resources were strategically reallocated to prioritize zanidatamab and other next-generation candidates, reflecting a pipeline realignment based on commercial viability and a clearer path to registration.

MEDI4276:

MEDI4276 was constructed by attaching the single-chain variable fragment (scFv) of trastuzumab to the N terminus of the heavy chain of the anti-HER2 fully human mAb 39 S (IgG1κ).Trastuzumab and 39 S bind to nonoverlapping distant epitopes on HER2. Thus, MEDI4276 contains four antigen-binding sites, two on each arm, that are capable of interacting with HER2. The presence of four high-affinity binding sites promotes efficient recruitment and clustering of HER2 molecules, even in tumor models with low HER2 expression. Such a biparatopic, tetravalent, monospecific antibody was shown to induce robust receptor clustering, internalization, lysosomal trafficking, and subsequent degradation of the complex [62]. It is plausible that this dense clustering enhances the local concentration of lysosomal proteases, thereby facilitating more efficient cleavage of the protease-labile linker and subsequent payload release. The payload, AZ13599185 is a tubulysin variant that developed by AstraZeneca/MedImmune. Tubulysins are a class of ultra-potent cytotoxics, often exhibiting IC₅₀ values in the low nanomolar to picomolar range, which is commensurate with the requirements for effective ADC therapy. Its efficient release within the cell, coupled with its inherent membrane permeability, is crucial for mediating the potent cytotoxicity observed, even in low HER2-expressing models. The mode of action of this small-molecule toxin is to inhibit microtubule polymerization during mitosis to induce cell death. The biparatopic antibody contains three site mutations in the Fc region, L234F, S239C, and S442C. The two engineered cysteine residues per heavy chain (S239C and S442C) enable site-specific conjugation of AZ13599185 to the antibody via a maleimidocaproyl linker, resulting in a biparatopic ADC with a DAR of 4. The mutation L234F in combination with the S239C mutation reduced FcγR binding. It is thought that these mutations may minimize the FcγR-mediated, HER2-independent uptake of ADC by normal tissues, thereby reducing off-target toxicity [63]. It is noteworthy that these mutations are strategically located outside the FcRn binding interface; therefore, they are not anticipated to compromise FcRn-mediated recycling, thus preserving the antibody’s extended plasma half-life [64]. This design exemplifies a sophisticated strategy to decouple undesirable Fc effector functions from favorable pharmacokinetic properties.

A first-in-human, phase 1, multicenter, open-label, dose-escalation study(NCT02576548) was conducted to evaluate the safety, pharmacokinetics (PK), immunogenicity, and antitumor activity of MEDI4276 in patients with HER2-positive advanced breast cancer or gastric cancer. This study showed that despite some clinical activity (albeit limited) in breast cancer, further clinical development of MEDI4276 is challenged by an unfavorable PK profile (insufficient to overcome a potential antigen sink) and high toxicity. MEDI4276 doses escalated from 0.05 to 0.9 mg/kg (60- to 90-minute intravenous infusion every 3 weeks). Primary endpoints were safety and tolerability; secondary endpoints included antitumor activity (objective response, progression-free survival (PFS), and overall survival), pharmacokinetics, and immunogenicity. Forty-seven patients (median age 59 years; median of seven prior treatment regimens) were treated. The maximum tolerated dose was exceeded at 0.9 mg/kg with two patients experiencing dose-limiting toxicities (DLTs) of grade 3 liver function test (LFT) increases, one of whom also had grade 3 diarrhea, which resolved. Two additional patients reported DLTs of grade 3 LFT increases at lower doses (0.4 and 0.6 mg/kg). The most common (all grade) drug-related adverse events (AEs) were nausea (59.6%), fatigue (44.7%), aspartate aminotransferase (AST) increased (42.6%), and vomiting (38.3%). The most common grade 3/4 drug-related AE was AST increased (21.3%). Five patients had drug-related AEs leading to treatment discontinuation. In the as-treated population, there was one complete response (0.5 mg/kg; breast cancer), and two partial responses (0.6 and 0.75 mg/kg; breast cancer)–all had prior trastuzumab, pertuzumab, and ado-trastuzumab emtansine (T-DM1). MEDI4276 has demonstrable clinical activity but displays intolerable toxicity at doses > 0.3 mg/kg [65].

The failure of MEDI4276 underscores the critical challenge of balancing efficacy and toxicity in ADC development. The unfavorable therapeutic index was likely attributable to a combination of factors rather than a single flaw. The tetravalent, biparatopic design, while driving exceptional internalization, may have contributed to an ‘antigen sink’ and rapid clearance, undermining its pharmacokinetics. Furthermore, the payload (tubulysin variant AZ13599185) and the relatively high DAR of 4 might have played a significant role [66]. Even with Fc engineering to reduce FcγR binding, the potent payload and efficient delivery system may have led to unacceptable on-target/off-tumor toxicity in tissues with basal HER2 expression. Therefore, the failure of MEDI4276 should not be viewed solely as a failure of the Bispecific strategy, but as a lesson in the intricate optimization of antibody format, payload, and DAR [67].

JSKN003:

JSKN003 is a Bispecific HER2-targeting ADC that harnesses the dual epitope binding (domains II/IV) of KN026. The Fc glycan-directed conjugation platform allows for site-specific attachment of four DXd molecules per antibody, resulting in enhanced structural homogeneity compared to maleimide-based stochastic cysteine conjugation. This approach also improves hydrophilicity, reducing aggregation, and enhances serum stability, leading to prolonged payload retention. These properties contribute to a favorable pharmacokinetic profile and reduced off-target toxicities, which is anticipated to extend the therapeutic window. The improved structural homogeneity and stability may, in theory, contribute to a reduced risk of certain off-target toxicities. However, the comparative safety profile, particularly regarding critical adverse events like ILD, awaits direct clinical validation in ongoing and future studies. This stable architecture delivers the topoisomerase I inhibitor DXd. DXd is a highly potent derivative of exatecan, demonstrating IC₅₀ values in the low nM range (e.g., 0.5–5.5 nM) in multiple cancer cell lines. Furthermore, DXd is designed to be membrane-permeable, which is a key factor underlying the robust bystander effect that has become a hallmark of DXd-based ADCs like trastuzumab deruxtecan (T-DXd), allowing it to effectively target heterogeneous tumors. In addition to its enhanced structural stability, JSKN003 retains the superior HER2-binding and internalization properties of KN026, ensuring efficient payload delivery to trastuzumab-resistant tumor cells. This enables JSKN003 to potentially suppress the proliferation of trastuzumab-resistant tumors [68].

JSKN003 entered a first-in-human, dose-escalation, multicenter, open-label, phase I study (NCT05494918).The phase 1 clinical trial of 32 patients against various solid tumors with different HER2 expression levels yielded promising results. Treatment was well-tolerated, with Grade 3 or higher TRAEs occurring in 6.3% of patients (2/32), manifested as anemia (hematologic) and fatigue (constitutional). The most common TRAEs of any grade were diarrhea (62.5%) and nausea (50.0%). Importantly, no dose-limiting toxicities (DLTs) were observed, and the maximum tolerated dose (MTD) was not reached up to 8.4 mg/kg. Pharmacokinetic analysis revealed dose-proportional exposure and a half-life of approximately 5–6 days at doses of 6.3 mg/kg and above, with minimal accumulation, supporting a Q3W dosing schedule [69]. These efficacy outcomes can be contextualized against historical data from the DESTINY-PanTumor02 trial, where DS-8201a (T-DXd) reported an ORR of 37.1% across all HER2-expressing solid tumors and 61.3% in the IHC 3 + subgroup [70]. While direct cross-trial comparisons have inherent limitations, the notably higher DCR observed with JSKN003 (90.6%) may indicate a potential for sustained disease control. Based on these encouraging preliminary efficacy and safety signals, JSKN003 entered phase 3 trials in December 2023 (NCT06079983) [71].

Dual-antigen BsADC targeting HER2

Dual-targeted BsADCs that simultaneously target HER2 and fast-turnover receptors (such as CD63, PRLR, and APLP2) can further optimize the internalization and lysosomal transport of HER2-targeted ADCs, with the potential to enhance cytotoxicity and reduce the HER2 threshold for clinical application. It is noteworthy that these innovative strategies, while demonstrating compelling preclinical proof-of-concept, have not yet advanced into clinical trials, highlighting the challenges in translating these designs into therapeutics.

BsADC on HER2×CD63:

CD63, a member of the tetraspanin superfamily, exhibits widespread but not ubiquitous expression. It is primarily localized on the cell surface, late endosomes, and lysosomes [72]. The cytoplasmic domain of CD63 contains a YXXØ consensus motif, enabling its interaction with Aps and the recruitment of a clathrin coat. This interaction with AP-2/AP-3 and clathrin positions CD63 as an adaptor protein, facilitating the extended trafficking of associated proteins from the cell surface to the late endosome–lysosomal compartment44,45 [73]. The presence of CD63 in these cellular compartments makes it a potential target for BsADCs aiming to enhance internalization and lysosomal transport, ultimately improving drug delivery and therapeutic efficacy.

To ensure the tumor specificity of bsAb, the CD63 arm of bsHER2xCD63 should not bind and internalize in absence of a tumor-specific arm. Therefore, a panel of mutated CD63 antibodies with variable CD63 affinity is required. It has been demonstrated that single amino acid histidine substitutions can be used to alter antibody affinity. Therefore, site-directed mutagenesis was applied to introduce histidine substitutions in the variable heavy and light chain domains of the anti-CD63 monoclonal antibody (clone 2192) to reduce its affinity for CD63. The histidine substitution strategy leverages the unique chemical property of the histidine side chain, which becomes progressively protonated as the pH drops from the neutral extracellular environment (pH ~ 7.4) to the acidic endo-lysosomal compartment (pH ~ 4.5–5.0). This protonation introduces positive charges that can disrupt electrostatic interactions or cause steric clashes within the antigen-antibody binding interface, thereby reducing binding affinity specifically at low pH [74]. To quantitatively characterize these engineered antibodies, their binding properties are typically assessed using surface plasmon resonance (SPR) [75]. This technique measures key kinetic parameters such as the association rate (k_on), dissociation rate (k_off), and equilibrium dissociation constant (K_D) across a range of pH buffers (e.g., from pH 7.4 to 5.5). A successful pH-sensitive mutant would exhibit a K_D comparable to the wild-type antibody at neutral pH to ensure cell surface binding, but a significantly weakened K_D (markedly increased k_off rate) at acidic pH to facilitate rapid dissociation in the lysosome and enable efficient receptor recycling [76]. Furthermore, epitope binning assays (e.g., using competitive SPR or ELISA) are employed to confirm that the histidine mutations do not alter the epitope specificity, ensuring that the antibody still targets the desired region on CD63. This comprehensive characterization is crucial to validate that the affinity reduction is both pH-dependent and does not compromise the intended targeting function. By combining this low-affinity mutant arm with another Fab arm from the anti-HER2 antibody (clone 153), a BsAb targeting HER2 × CD63 is created [77, 78]. The arm targeting HER2 is responsible for specific binding to tumor cells, and the arm targeting CD63 is responsible for promoting internalization and lysosomal delivery of cytotoxic payloads. Subsequently, bsHER2 × CD63 was conjugated to anti-mitotic agent duostatin-3 via the VC linker [79]. Potent cytotoxicity of bsHER2xCD63-ADC against HER2-positive tumors, which was not observed with monovalent HER2- and CD63-specific ADCs. For instance, in the high-HER2 expressing SK-BR-3 cell line, the bsHER2xCD63-ADC achieved an IC₅₀ of 0.15 nM, representing a > 10-fold increase in potency compared to the monospecific HER2-ADC (IC₅₀ > 1.5 nM). However, the observed lack of efficacy in low-HER2 tumors (e.g., MCF-7 cells, IC₅₀ > 10 nM) indicates the necessity for further optimization.

BsADC on HER2×PRLR:

The Prolactin Receptor (PRLR) is a type 1 cytokine receptor that is expressed in a subset of breast cancers and may contribute to its pathogenesis. It is relatively overexpressed in approximately 25% of human breast tumors while expressed at low levels in some normal human tissues including the mammary gland. In contrast to HER 2, PRLR is rapidly and constitutively internalized and efficiently translocated to the lysosome, where it is degraded. The PRLR cytoplasmic domain facilitates its own rapid constitutive internalization and lysosomal degradation. Critically, when HER2 is cross-linked to PRLR at the cell surface by the Bispecific antibody, this property is leveraged to co-opt HER2 into the same rapid internalization and degradation pathway. This forced partnership dramatically enhances the degradation of HER2, thereby improving the delivery and efficacy of the conjugated cytotoxic payload. Low levels of cell surface PRLR are sufficient to mediate effective killing by PRLR ADCs [80].

A bsAb with HER2 and PRLR arms (HER2(T)xPRLR) was generated using the “knobs-into-holes” approach [41]. The HER2(T) arm was derived from the primary sequence of trastuzumab. The PRLR arm was derived from a fully human VelocImmune PRLR antibody designated H1H7672P2. HER2xPRLR bsADC is composed of HER2xPRLR bsAb conjugated to maytansine derivative DM1 via a noncleavable, hetero-bifunctional linker (succinimidyl trans-4-[maleimidylmethyl] cyclohexane-1-carboxylate, SMCC; Supplementary) with average DAR of 3.3. Noncovalently cross-linking HER2 to PRLR at the cell surface, using a Bispecific antibody that binds to both receptors, dramatically enhances the degradation of HER2 as well as the cell killing activity of a noncompeting HER2 ADC. Furthermore, in breast cancer cells that coexpress HER2 and PRLR, a HER2xPRLR Bispecific ADC kills more effectively than HER2 ADC. This was demonstrated by an approximately 10-fold lower IC₅₀ in BT-483 cells (0.15 nM for bsADC vs. 1.5 nM for HER2 ADC) and markedly enhanced potency in T47D/HER2 and MDA-MB-361 cell lines, underscoring the consistent benefit of this Bispecific approach [81].

BsADC on HER2×APLP2:

APLP2 is a ubiquitously expressed member ofthe amyloid precursor protein family. The tyrosine contained in the cytoplasmic tail of APLP2 contains overlapping tyrosine-based NPXY and YXXØ motifs. Similarly, APLP2 could bind to AP-2, mediate effective internalization and direct to lysosomal degradation after clathrin-mediated endocytosis [82].

APLP2 could deliver MHCI complexes and antigen‒antibody complexes into the lysosomal degradation pathway via full-length Bispecific antibody technology, the BsAb with one αHER2 (Herceptin) arm and one arm of a αAPLP2 antibody was designed, which can bound to recombinant ECD of both APLP2 and HER2 simultaneously. Through the site-specific coupling strategy of glutamine transaminase to ensure stability, AmPEG6-MMAD was conjugated on the BsAb with DAR = 2 or 4. APLP2 targeting could effectively redirect recycled targets to lysosome to release payloads [83]. In vitro, the HER2xAPLP2 BsADC with a DAR of 4 exhibited potent cytotoxicity against HER2-positive cells (e.g., IC₅₀ of 0.8 nM in BT-474), significantly outperforming a conventional HER2-targeting ADC. This in vitro potency translated into in vivo efficacy, where a single dose of the BsADC (5 mg/kg) achieved 85% tumor growth inhibition in a BT-474 xenograft model, compared to 45% inhibition by the conventional ADC at the same dose.

BsADCs on HER2 and HER3

Members of human epidermal growth factor receptors (ErbB) are potent mediators of normal cell growth and development. The ErbB family consists of four closely related type I trans-membrane tyrosine kinase receptors: EGFR, HER2, HER3 and HER4 [84]. Receptor dimerization (hetero- or homo-dimerization) is an essential requirement for ErbB function and the signaling activity of these receptors. The HER2-HER3 heterodimer is considered the most potent ErbB pair with respect to strength of interaction, ligand-induced tyrosine phosphorylation and downstream signaling and functions as an oncogenic unit [85]. Some researchers reported that HER3 might be a necessary partner for the oncogenic activity of HER2 in tumors overexpressing HER2. In addition, co-overexpression of HER2 and HER3 was a predictor of impaired survival of breast cancer patients [86]. Higher expression and activation of HER3 was observed in HER2 + breast cancer cell lines with resistance to T-DM1 [87]. Taking into consideration of the drug resistance of T-DM1, a BsADC targeting HER2/HER3 heterodimer would be a promising strategy to overcome the limitation of T-DM1 therapy.

Researchers constructed a HER2/HER3-targeting Bispecific ADC (BsADC) and characterized its physiochemical properties, target specificity and internalization in vitro, and assessed its anti-tumor activities in breast cancer cell lines and in animal models. The HER2/HER3-targeting BsADC had a DAR of 2.89, demonstrated potent and selective antitumor activity, with quantitative data supporting each key claim. Specificity was confirmed in co-culture models, where the BsADC exhibited 99.21% specific binding to JIMT-1 tumor cells even in the presence of a 20-fold excess of MCF10A normal breast epithelial cells. Cellular internalization was slightly enhanced, with the BsADC achieving 57% internalization within 4 h–a 1.4-fold and 2.1-fold increase over the HER2- and HER3-monospecific ADCs, respectively. The conjugate demonstrated potent, target-dependent cytotoxicity, with EC₅₀ values of 80.25 ng/mL in T-DM1-resistant JIMT-1 cells, representing a > 15-fold and ~ 2-fold increase in potency compared to the monospecific HER2- and HER3-targeting ADCs, respectively. This enhanced in vitro potency translated into in vivo efficacy, where a single 3 mg/kg dose of the BsADC achieved 30% tumor growth inhibition in JIMT-1 xenograft models, matching the efficacy of the combined monospecific ADCs. Furthermore, a comprehensive in vivo characterization included pharmacokinetic, dose-response, and safety analyses. The BsADC exhibited a favorable pharmacokinetic profile with a terminal half-life of 4.65 days in mice. A clear dose-response relationship was established, wherein single doses of 0.3 and 1 mg/kg showed minimal efficacy, while significant tumor growth inhibition (TGI ~ 30%) was achieved at 3 mg/kg. Critically, a higher and repeated dosing regimen (10 mg/kg, weekly for two weeks) demonstrated superior and potent tumor growth inhibition (TGI of 92%), underscoring the compound’s dose-dependent efficacy. Regarding safety and tolerability, a key translational metric, no significant body weight loss was observed in any BsADC-treated group across all studies. This favorable safety profile was in stark contrast to the significant toxicity (evidenced by substantial weight loss) induced by the small-molecule control agent canertinib. Additionally, treatment with the BsADC significantly extended the survival of tumor-bearing mice, further supporting its therapeutic potential [88]. The superior efficacy of the HER2/HER3 BsADC compared to the combination of monospecific ADCs, achieving comparable tumor growth inhibition at half the total ADC dose, strongly suggests a synergistic effect. This synergy is likely mediated by enhanced internalization resulting from the co-engagement and heterodimerization of HER2 and HER3 by the single Bispecific molecule, which aligns perfectly with the core rationale for developing BsADCs.

BsADC targeting EGFR x HER3 (BL-B01D1)

EGFR (also known as ERBB1 or HER1) is a member of the ERBB receptor tyrosine kinase family, which also includes HER2 (ERBB2), HER3 (ERBB3), and HER4 (ERBB4). EGFR plays a key role in regulating the basic functions of epithelial malignancies, thereby inducing disturbances in homeostasis in vivo. However, due to therapeutic stress-induced acquired genomic alterations, targeting EGFR monoclonal antibodies and TKIs (tyrosine kinase inhibitors) often leads to the emergence of clinical resistance [89]. BsADCs are expected to address the mechanisms of anti-EGFR resistance, including sensitizing mutations and activation of the bypass pathway.

HER3 is encoded by the ERBB3 gene. HER3 itself is not kinase active, but it is capable of forming heterodimers with HER2 (and/or EGFR), which significantly increases transphosphorylation and activation of downstream signaling cascades [90]. Resistance to EGFR-targeted therapies may also occur through compensatory signaling at the ERBB receptor, involving receptors such as HER2 and HER3, suggesting the potential of combination therapy [91]. However, clinical data suggest that the combination of patuximab and cetuximab leads to overlapping toxicity, making the combination intolerable. HER3 is significantly overexpressed in EGFR-resistant cancer cell lines, promoting both intrinsic and acquired resistance to EGFR along with activation of PI3K [92].

By utilizing both an anti-EGFR Fab and an anti HER3 scFv, a BsAb targeting EGFR and HER3 was constructed, named SI-B001. topoisomerase I inhibitor (TOPO1i) ED04 was coupled to a cysteine site on SI-B001 via a novel cleavable AC linker, generating a novel BsADC, named BL-B01D1, with a reported average DAR of approximately 8. ED04 belongs to the class of topoisomerase I inhibitors, which are known for their high potency (IC₅₀ often < 10 nM). Similar to other TOPO1i payloads like DXd and SN-38, ED04 is expected to possess adequate membrane permeability to facilitate a bystander effect, which is likely contributing to the encouraging efficacy observed in HER2-heterogeneous breast cancer populations. This high DAR is achieved through cysteine-based conjugation, which typically produces a heterogeneous mixture of DAR species (e.g., DAR 0, 2, 4, 6, 8). While maximizing the payload load, such a high and heterogeneous DAR profile is associated with increased hydrophobicity, elevated risk of aggregation, and accelerated plasma clearance, necessitating rigorous analytical control during development. BL- B01D1 achieved targeted killing of EGFR-dependent tumors and attenuated HER3-induced drug resistance [93].

Recent clinical data have provided robust, quantitative evidence for its efficacy and safety profile. In a phase I study (NCT05470348), patients with locally advanced or metastatic HER2- (IHC 0, 1+, or 2+/ISH-) breast cancer who were enrolled in the study were treated with BL-B01D1 at a dose of 2.5 mg/kg every three weeks (D1D8 Q3W). Enrolled patients were not screened based on EGFR or HER3 expression levels. As of September 30, 2024, among 121 evaluable patients (median prior lines of therapy = 3), the objective response rate (ORR) was 42.1%, with a confirmed ORR (cORR) of 36.4% and a disease control rate (DCR) of 80.2%. The median progression-free survival (mPFS) was 6.9 months (95% CI: 5.5, 8.4), and the median duration of response (mDOR) was 9.7 months (95% CI: 5.8, 11.7). Encouragingly, efficacy was observed across HER2 expression subgroups, including HER2-zero disease (ORR 41.8%) [94]. Despite the encouraging efficacy, the safety profile of EGFR×HER3-targeting agents like BL-B01D1 warrants careful consideration. HER3 is expressed in various healthy tissues, including the heart and nervous system, raising theoretical concerns for on-target, off-tumor toxicity [95]. Notably, the Phase I clinical data reveal that the most frequent treatment-related adverse events (TRAEs) were predominantly hematologic (e.g., neutropenia [86.8%], leukopenia [90.1%], thrombocytopenia [71.9%]) and gastrointestinal (e.g., nausea [61.2%], stomatitis [52.9%], vomiting [47.9%]), which are typical of topoisomerase I inhibitor-based ADCs and manageable with supportive care [96, 97]. Grade 3 and above TRAEs which were predominantly hematologic in nature, were able to be effectively managed with standard supportive measures including dose reductions, as demonstrated by the TRAE leading to discontinuation rate of 5.0%. Importantly, no signal of cardiac toxicity or cytokine release syndrome was prominently reported, and no incidents of ILD were observed. This clinical profile suggests that while theoretical risks exist, the realized toxicity of BL-B01D1 may be primarily driven by the potent payload and can be clinically managed. Continued vigilance remains essential to fully characterize its long-term safety [94].

BsADC targeting TROP2 x HER3 (JSKN016)

Trop2(Trophoblast cell surface antigen 2) is a type I surface glycoprotein that has garnered attention in oncology. Trop2 is critical in embryonic organ development, and its expression is limited in normal tissues [98]. Accumulating evidence suggests that Trop2 is overexpressed in a range of solid tumors, significantly impacting tumor growth, invasion, and metastasis [99]. Consequently, Trop2 has become an attractive prognostic marker and therapeutic target for solid tumors. The approval of the first Trop2-targeted antibody-drug conjugate (ADC), TrodelvyTM, which enhances survival in metastatic breast cancer and metastatic urothelial cancer, has encouraged further research on Trop2 biology and the expansion of Trop2-targeted therapeutic strategies [100]. TROP2 and HER3 are both highly expressed in TNBC and associated with worse survival, providing a strong rationale for dual targeting.

JSKN016 is a Bispecific TROP2/HER3-targeting ADC developed with novel linker to conjugate the payload, a topoisomerase Ⅰ inhibitor with a DAR of 4. While the specific compound is not disclosed, topoisomerase I inhibitors used in ADCs (e.g., DXd, SN-38) are universally characterized by their high potency (IC₅₀ in the nM range) and ability to diffuse across cell membranes. This combination of extreme cytotoxicity and membrane permeability is the fundamental driver behind the remarkable preliminary efficacy (ORR = 80%) and potential bystander effect observed in the mTNBC cohort. Emerging clinical data from an ongoing first-in-human phase I study (JSKN016-101, NCT06592417) have begun to quantify its promising profile. As of December 23, 2024, in a cohort of heavily pre-treated patients with metastatic triple-negative breast cancer (mTNBC), JSKN016 demonstrated substantial antitumor activity. Among 5 efficacy-evaluable mTNBC patients enrolled across multiple dose levels (4–8 mg/kg Q3W), the ORR was 80.0% (4/5 patients achieving partial response), with the remaining patient achieving stable disease accompanied by 29.5% tumor shrinkage. The maximum tolerated dose (MTD) had not been reached at the time of reporting, and PFS data were not yet mature [101]. Despite the compelling antitumor activity, the safety profile of this TROP2×HER3 Bispecific ADC warrants ongoing evaluation. Preliminary data from the JSKN016-101 trial reveal a distinct safety profile: the most frequent treatment-related adverse events (TRAEs) were nausea, vomiting, and oral mucositis. A dose-limiting toxicity (DLT) of Grade 3 dermatitis acneiform occurred at the 8 mg/kg dose level. Grade ≥ 3 TRAEs were infrequent, observed in only 10.5% of patients, and included neutropenia, lymphopenia, and oral mucositis. Crucially, no signal of cardiac toxicity, cytokine release syndrome, or treatment-related ILD was observed, and no adverse events led to treatment discontinuation or death, supporting the clinical tolerability of JSKN016 within the studied dose range [101]. This early clinical profile suggests that the realized toxicity of JSKN016 may be primarily attributable to its topoisomerase I inhibitor payload and the TROP2 target [102]. Continued investigation in larger cohorts is essential to fully characterize its long-term safety. These early data suggest a potentially differentiated profile for JSKN016. When compared to the approved TROP2-targeting ADC, sacituzumab govitecan (Trodelvy), the preliminary ORR of 80% in a heavily pre-treated mTNBC cohort appears numerically higher than the ~ 35% ORR reported in the ASCENT trial, though cross-trial comparisons require caution due to the small initial cohort size of JSKN016. Furthermore, JSKN016’s safety profile shows a distinct pattern, with a notably lower incidence of severe (Grade ≥ 3) neutropenia (5.3%, 1/19 patients) reported to date compared to the ~ 51% incidence commonly associated with Trodelvy [103]. The Bispecific engagement of TROP2 and HER3 may not only enhance efficacy by targeting two pathways implicated in TNBC but also potentially mitigate the resistance mechanisms that can limit the activity of monovalent ADCs. While the clinical data for HER3-targeted ADCs like patritumab deruxtecan are still emerging in TNBC, the robust initial activity of JSKN016 underscores the promise of this dual-targeting strategy. The structural features, mechanisms of action, and current developmental status of the BsADCs discussed in this section are comprehensively summarized in Table 1.

Table 1.

Summary of key bispecific Antibody-Drug conjugates (BsADCs) in development for breast cancer therapy

Target(s)	Example BsADC	Construction	Key Features/Findings	Development Status/Comment
Dual-Epitope HER2(ECD2/ECD4)	ZW49 (based on ZW25)	Payload: N-acyl sulfonamide auristatin DAR: 2 Linker: cleavable linker	Biparatopic binding induces HER2 clustering, enhances internalization/CDC; ORR:13% in Phase I; Manageable toxicity (e.g., keratitis).	Phase I ongoing; potent in low- HER2 tumors.(NCT03821233)
Dual-Epitope HER2	MEDI4276	Payload: AZ13599185(microtubule inhibitor) DAR:4 Linker: cleavable linker	Tetravalent biparatopic; robust clustering/lysosomal trafficking. Limited activity due to PK/toxicity issues.	Phase I discon- tinued; high tox- icity (ILD).
HER2 (Do- mains II/IV)	JSKN003 (based on KN026)	Payload: DXd(Topoisomerase I inhibitor) DAR:4 Linker: cleavable tetrapeptide linker	Improved stability/homogeneity over T-DXd. Enhanced binding/internalization.	Preclinical/early clinical; reduced ILD risk.
HER2/CD63	bsHER2xCD63 ADC	Payload: duostatin-3(microtubule inhibitor) DAR: not specified Linker: cleavable VC linker	Low-affinity CD63 arm pro- motes lysosomal delivery. Po- tent in HER2 + tumors but needs optimization for low- HER2.	Preclinical; ad- dresses internal- ization barriers.
HER2/PRLR	HER2xPRLR BsADC	Payload: DM1(microtubule inhibitor) DAR:3.3 Linker: noncleavable SMCC linker	Cross-links receptors for enhanced degradation; Improved killing in co- expressing cells.	Preclinical; targets PRLR- overexpressing subsets (~ 25% breast tumors).
HER2/APLP2	HER2xAPL2P2 BsADC	Payload: AmPEG6-MMAD DAR:2/4 Linker: cleavable linker	Redirects to lysosomal degradation; Effective redirection of recycled targets.	Preclinical; leverages APLP2’s en- docytosis motifs.
HER2/HER3	HER2/HER3 BsADC	Linker-Payload: MC-vc-PAB-MMAE DAR:2.89	Targets heterodimer for re- sistance; inhibits viability in resistant cell lines/xenografts.	Preclinical; com- parable to com- bined monospe- cific ADCs.
EGFR/HER3	BL-B01D1 (based on SI-B001)	Payload: TOPIi ED04(topoisomerase I inhibitor) DAR:8 Linker: cleavable AC linker	Attenuates HER3-induced resistance. Encouraging ef- ficacy in HER2- BC (Phase I/II).	Phase I/II; man- ageable safety in pretreated patients.
TROP2/HER3	JSKN016	Payload: Topoisomerase I inhibitor DAR:4 Linker: cleavable linker	Antitumor activity in TNBC. Manageable safety in FIH study.	Phase I ongoing; breakthrough in TNBC survival.

Conclusions and perspectives

In conclusion, Bispecific antibody-drug conjugates represent a promising and rapidly evolving frontier in breast cancer therapy, holding the potential to reshape treatment paradigms… While ADCs like trastuzumab deruxtecan (T-DXd) have undeniably revolutionized care, BsADCs are still in earlier stages of clinical validation. Preclinical and early clinical studies underscore their potential: HER2-targeting BsADCs (e.g., ZW49) show activity in resistant models, internalization-enhancing combinations (e.g., HER2×CD63) may overcome low antigen density, and TROP2×HER3 BsADCs (e.g., JSKN016) have demonstrated promising efficacy in triple-negative breast cancer. However, the path to clinical success is paved with interconnected challenges. First, the clinical toxicity profile is transformed rather than eliminated, as seen in the hepatotoxicity of MEDI4276 and keratitis with ZW49. Second, the complex engineering and stringent manufacturing of Bispecific formats create production bottlenecks and high costs, directly impacting patient access. Finally, fundamental scientific hurdles persist, including managing immunogenicity, overcoming tumor antigen heterogeneity, and ensuring linker stability to prevent off-target toxicity. Therefore, the future of BsADCs hinges on a concerted effort to address this triad of challenges. Success will require the continuous biological optimization of targets, linkers, and biomarkers to be seamlessly integrated with innovations in manufacturing and a clear-eyed understanding of the clinical safety profile. Only through such a holistic approach can the full promise of more effective and tolerable individualized therapy be realized.

Author contributions

X.J. wrote the original manuscript. All authors reviewed and approved the final version of the manuscript.

Data availability

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Consent to participate

Not applicable.