A Novel Dual-Payload ADC Platform Integrating Exatecan and Triptolide to Enhance Antitumor Efficacy and Overcome Resistance

Abstract

Antibody–drug conjugates (ADC) that utilize DNA topoisomerase I inhibitors, such as deruxtecan and SN-38, have significantly enhanced the efficacy of cancer treatment. However, the development of more effective ADCs to combat drug-resistant tumors remains a critical challenge. ADCs covalently link the topoisomerase I inhibitor exatecan and the RNA polymerase II inhibitor triptolide to the same antibody, thereby establishing a dual-killing mechanism against tumor cells. Through optimization of the linker design, this dual-payload ADC achieved precise drug delivery in vivo. Specifically, triptolide is initially released to downregulate the expression of the stress response protein heat shock protein 70 and the efflux pump P-glycoprotein in tumor cells, thereby significantly enhancing drug sensitivity and effectively overcoming resistance, followed by gradual exatecan release, further inhibiting tumor cell proliferation and survival. Experimental results demonstrated that this dual-payload ADC exhibits significant antitumor activity in both in vitro cell models and in vivo xenograft models, including cell line–derived xenograft and patient-derived xenograft models, successfully addressing the challenge of drug resistance encountered in traditional monotherapy. Additionally, non-Good Laboratory Practice–compliant toxicity studies have confirmed the favorable safety profile of this ADC. In summary, this study provides an innovative strategy for overcoming tumor resistance, demonstrating substantial potential for clinical translation, and offering promising therapeutic outcomes and prognosis for patients with cancer.

Graphical Abstract

Introduction

Antibody–drug conjugates (ADC), as innovative therapies in the field of cancer treatment, have successfully achieved targeted drug delivery by conjugating highly specific monoclonal antibodies with potent cytotoxic small-molecule drugs, thus significantly transforming the landscape of cancer treatment (1). The core advantage of ADCs is their ability to precisely deliver cytotoxic drugs into tumor cells while minimizing damage to normal tissues, thereby effectively improving treatment efficacy and reducing systemic toxicity (2). This therapy has achieved remarkable results in clinical practice. ADCs with camptothecin (CPT)-based drugs such as DS-8201 and sacituzumab govitecan have been proven to significantly increase the survival rates of patients with various solid tumors and enhance tumor response rates to treatment (3, 4).

Despite the remarkable achievements of ADCs in clinical applications, many problems remain unsolved in the field of cancer treatment (5). Single-payload ADCs are prone to inducing drug resistance during treatment (6). Tumor cells can gradually develop resistance to single-payload ADCs through mechanisms such as upregulation of the expression of drug efflux pumps [e.g., P-glycoprotein (P-gp)] or downregulation of the expression of target antigens, which undoubtedly limits their long-term treatment efficacy (7). However, some tumor types have low sensitivity to ADCs, resulting in poor treatment responses, which further highlights the limitations of existing ADCs in clinical applications (8). Notably, overexpression of the efflux pump proteins ABCG2 or P-gp and upregulation of heat shock protein 70 (HSP70) are two major mechanisms of tumor cell resistance (9, 10).

Exatecan (EXA) mesylate has higher cell membrane permeability compared with deruxtecan (Dxd; ref. 11). This class of CPT-based drugs exert cytotoxic effects mainly by targeting topoisomerase I (TOP1), which is crucial for DNA replication, transcription, and repair processes. They inhibit the activity of TOP1, causing DNA damage and ultimately leading to cell death (12). However, similar to many anticancer drugs, the development of drug resistance severely restricts their long-term effectiveness. In recent years, a new resistance mechanism related to EXA, the HSP pathway, has emerged. The increased expression of the HSP70 protein family in tumor cells is one of the key potential mechanisms for the development of resistance (13). The high expression of the HSP70 protein in tumor cells can promote the development of resistance through multiple pathways (14). For example, treatment with CPT and its derivative topotecan leads to upregulated HSP70 expression and a significant increase in the cell survival rate, which fully demonstrates that the overexpression of HSP70 alone can make tumor cells resistant to these CPT-derived drugs (15).

Triptolide (TPL) is a natural compound derived from Tripterygium wilfordii. In traditional Chinese medicine, T. wilfordii has long been used in the treatment of autoimmune diseases and inflammation, with systemic lupus erythematosus and rheumatoid arthritis as typical examples (16). TPL exhibits significant antitumor activity in various cancer cell lines (17). Its mechanism of action is mainly achieved by interfering with multiple key signaling pathways, especially the HSP70 pathway (18). By disrupting these signaling cascades, TPL effectively inhibits cell proliferation and induces apoptosis, thereby suppressing the growth and spread of cancer cells. Specifically, TPL can covalently bind to human Xeroderma Pigmentosum Complementation Group B (a subunit of transcription factor II H), inhibit DNA-dependent ATPase activity, and further inhibit the transcription process mediated by RNA polymerase II (Pol II), which explains its strong and promising anticancer activity (19). However, the clinical application of TPL is limited by its toxicity and unfavorable pharmacokinetic (PK) characteristics. Although researchers have designed various delivery strategies to reduce the adverse reactions to TPL and enhance its efficacy, none of these strategies have been successfully applied in clinical practice. Minnelide, a water-soluble prodrug of TPL, has been used in several clinical studies on advanced gastrointestinal cancer and pancreatic cancer. In both monotherapy and combination therapy, minnelide has demonstrated clinically significant activity and controllable safety (20).

Against this background, this study innovatively designed a dual-payload ADC by conjugating EXA and TPL in parallel to a mAb, with an average of approximately four EXA and four TPL molecules conjugated to each antibody molecule. Through careful optimization of the linker design, this dual-payload ADC could achieve precise drug delivery in vivo. Specifically, TPL is first released to enhance the sensitivity of tumor cells to drugs and overcome drug resistance; then, EXA is released to further inhibit the proliferation and survival of tumor cells. In cell line–derived xenograft (CDX) and patient-derived xenograft (PDX) models, the Trop2 dual-payload ADC showed a dose-dependent inhibitory effect on tumor growth. Compared with single-payload TOP1 inhibitor ADCs, the Trop2 dual-payload ADC still exhibited superior antitumor activity, even at equivalent or lower doses. Moreover, in the DXd-resistant CDX model with increased P-gp expression and the PDX model with disease progression after SN-38–ADC treatment, this dual-payload ADC effectively inhibited tumor growth, fully highlighting its great potential to overcome ADC resistance. A non-Good Laboratory Practice (GLP) toxicity study in cynomolgus monkeys further revealed the safety of this dual-payload ADC and determined the highest nonseverely toxic dose (HNSTD) of hRS7-E+T at 50 mg/kg administered intravenously every 3 weeks for 7 weeks. In summary, the data from this study indicate that compared with previous ADCs, the EXA and TPL dual-payload ADCs are expected to overcome tumor resistance mechanisms and improve the sensitivity of various tumors to drugs, which may provide a new strategy for solving the problem of tumor resistance and bring potential clinical benefits to patients with cancer.

Materials and Methods

Antibodies and ADCs

The anti-TROP2 antibody (hRS7 or sacituzumab) and anti-HER3 antibody (patritumab) were generated in the CHO or 293F expression system and purified by MabSelect SuRe resin (GE HealthCare) in a gravity column. The sequences of these antibodies (Supplementary Fig. S1) were obtained from the corresponding patents and literature. hRS7-E+T and patritumab-E+T were made in-house with hRS7 and patritumab using linker–payloads customized and purchased from MedChemExpress. Datopotamab deruxtecan (Dato-DXd, HY-141598) and sacituzumab govitecan (HY-132254) were directly purchased from MedChemExpress.

Cell lines and culture conditions

NCI-H292 (RRID: CBP60187, female), NUGC-4 (RRID: CBP60493, female), and COLO205 (RRID: CBP60026, male) cell lines were purchased from Nanjing Cobioer Biosciences Co., Ltd. in 2023. NCI-H2170 (RRID: BNCC101664, male) and HCT-15 (RRID: BNCC100189, male) cell lines were obtained from Bena Culture Collection in 2023. The MDA-MB-231 (RRID: CL-0150, female) cell line was purchased from Pricella in 2024. High TROP2-expressing HCT-15 cells (HCT-15–TROP2) were generated via lentiviral-mediated transfection of a TROP2 overexpression vector (sequence based on NM_002353.3), using parental HCT-15 cells (RRID: BNCC100189, male) as the host.

All parental cell lines were accompanied by short tandem repeat authentication reports provided by their respective suppliers (latest authentication date: June 2023 for Cobioer-derived lines, July 2023 for Bena Culture Collection–derived lines, and January 2024 for Pricella-derived MDA-MB-231), which confirmed cell identity and excluded cross-contamination. For the engineered HCT-15–TROP2 cells, additional authentication of TROP2 overexpression was verified via flow cytometry (latest verification date: August 2023) prior to experimental use.

All cell lines were regularly tested for mycoplasma contamination using the Mycolor One-Step Mycoplasma Detector kit (cat. # D201-01, Vazyme), with the latest test conducted in September 2023. The detection procedure was performed as follows: 1 μL of cell culture supernatant was mixed with 19 μL of Mycolor LAMP Mix, followed by the addition of 20 μL of mineral oil. The mixture was incubated at 65°C for 1 hour, and results were determined by color change—red indicated mycoplasma negativity, whereas yellow indicated positivity. Only mycoplasma-negative cells were used for experiments.

All cell lines were cultured in appropriate media: RPMI 1640 medium for NCI-H292, NUGC-4, NCI-H2170, and COLO205 and DMEM for HCT-15, MDA-MB-231, and HCT-15–TROP2. All media were supplemented with 10% heat-inactivated FBS, and cells were maintained at 37°C in a humidified atmosphere with 5% CO₂.

Cell passaging and culture duration were standardized to ensure consistency: After receipt, all parental cell lines were thawed and passaged using 0.25% trypsin-EDTA (Gibco, cat. #25200-056) when reaching 80% to 90% confluence. Experiments were conducted using cells at passages 3 to 12 (corresponding to a culture period of < 6 months after thawing, in compliance with American Association for Cancer Research editorial policies) to avoid cellular senescence or phenotype drift. Prior to experimental assays (e.g., cell proliferation, drug treatment), cells were seeded in respective media and cultured for 24 to 48 hours to reach the logarithmic growth phase.

Analysis of dead and apoptotic cells

Cells treated under various conditions in 12-well plates were stained with 100 μL of Acridine Orange-Propidium Iodide (RE010212, Countstar) diluted 1:10 in PBS and incubated at room temperature for 20 minutes. Dead or primary cells were detected using Operetta CLS. After trypsinization, the cells were stained with an Annexin V-Alexa Fluor 647/PI apoptosis detection kit (FXP023, 4Abio) and analyzed by flow cytometry (BD FACSCelesta).

Western blot and qRT-PCR

Total protein for Western blot samples was extracted from the cells using RIPA lysis buffer containing phenylmethanesulfonyl fluoride (Beyotime, ST506; Beyotime, P0013B). Samples were denatured at 98°C for 15 minutes, followed by SDS-PAGE. Proteins were transferred to polyvinylidene difluoride membranes (Merck Millipore, IPVH00010) at 4°C with a current of 255 mA for 1,200 minutes. Polyvinylidene difluoride membranes were blocked in skim milk for 2 hours and then incubated overnight at 4°C with the following antibodies: anti-HSP70 (1:500, Beyotime, AF0189), anti-TOP1 (1:1,000, BD Biosciences, 556597), anti–Pol II (F-12; 1:500, Santa Cruz Biotechnology, sc-55492), and anti–α-tubulin (1:2,000, Immunoway, YT4777). After incubation with the corresponding horseradish peroxidase–conjugated secondary antibodies, the proteins were detected using chemiluminescent reagents. The protein band intensity on the films was quantified using Quantity One software (version 4.6.2, Bio-Rad).

Total RNA was extracted from the cells using TRIzol reagent (Thermo Fisher Scientific, 99089301). The concentration and purity of the total RNA were assessed using a UV5Nano spectrophotometer (Mettler Toledo). First-strand cDNA was synthesized using a cDNA Synthesis Kit (RR047A; Takara Bio). qRT-PCR was performed on a real-time fluorescence qPCR system (Applied Biosystems) using the SYBR Green Master Mix (Applied Biosystems, 4367659). The target gene expression levels were normalized to 18S RNA and calculated using the 2^−ΔΔCt method. The primer sequences were as follows: hTOP1: forward, 5′-CAG​CCG​TTT​CTG​GAG​TCT​CG-3′, and reverse, 5′-AAT​CCG​CTT​CGA​TCT​GGG​AA-3′; HSP70S: forward, 5′-ACC​AAG​CAG​ACG​CAG​ATC​TTC-3′, and reverse, 5′-CGC​CCT​CGT​ACA​CCT​GGA​T-3′; ABCB1: forward, 5′-GCT​ACA​TGA​GAG​CGG​AGG​AC-3′, and reverse, 5′-TCC​CCT​TGC​AAG​TTG​ACA​CT-3′; and 18S RNA: forward, 5′-CTC​GCT​CCT​CTC​CCA​CTT​G-3′, and reverse, 5′-TGA​CCG​GGT​TGG​TTT​TGA​TC-3′.

Detection of target protein expression on tumor cells

Expression of TROP2 or HER3 in tumor cell lines was detected using flow cytometry (BD FACSCelesta). Cells were incubated with the following antibodies on ice for 30 minutes: 1 mg/mL PE-conjugated mouse IgG2a, κ isotype control, and PE-conjugated anti-human TROP2 (hRS7) or HER3 (patritumab) antibodies. After washing, the labeled cells were analyzed using a flow cytometer (BD FACSCelesta). The following gating strategy was used for the flow cytometry: In the forward scatter (FSC)/side scatter plot, high-FSC cells were selected to exclude cell debris. In the forward scatter area/forward scatter height plot, single cells were selected using a single-slope gate. In the phycoerythrin area plot, nonspecific binding was excluded using a small phycoerythrin area cell gate through an isotype control.

Binding activity assay

Binding and dynamic affinity were evaluated using ELISA. Human TROP2 protein (1 μg/mL; ACROBiosystems, TR2-H5223) was coated onto 96-well plates overnight at 4°C. The plates were blocked with 3% BSA (Sigma-Aldrich, B2064) at 37°C for 2 hours. Diluted hRS7-E+T, hRS7, or isotype control (human IgG1 antibody) was added and incubated at 37°C for 2 hours. After washing, diluted (1:2,000) mouse anti-human IgG[jdc-10] horseradish peroxidase (Abcam, ab99759) was added and further incubated at 37°C for 1 hour. Finally, TMB (3,3′,5,5′-tetramethylbenzidine) solution was added, and the absorbance was measured at 450 nm using a SpectraMax i3X microplate reader (Molecular Devices).

Cytotoxic activity assay

Tumor cells were seeded in 96-well plates at the following densities: NCI-H292 (3,000 cells/well), NCI-H2170 (5,000 cells/well), and NUGC-4 (4,000 cells/well). After a 4-hour incubation, the diluted test samples were added. Cell viability was assessed 72 hours later using the CellCounting-Lite 2.0 luminescent cell cytotoxic activity assay kit (Novozymes Biotechnology Co., Ltd., DD1101) and a SpectraMax i3X microplate reader (Molecular Devices). The percentage cell viability was calculated as follows: cell viability percentage = RLU(X)/RLU(Ctrl) × 100%.

Human tumor xenograft (CDX) models

All animal experiments were conducted in a laboratory fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International. The content and procedures related to animal experiments in this study complied with relevant laws and regulations for the use and management of experimental animals and the requirements of the Institutional Animal Care and Use Committee (IACUC). The study protocol was approved (IACUC number: IACUC-A2018086-T014-01). CDX models were established by subcutaneously transplanting tumor cells into female NOD/SCID mice (Jiangsu Biocytogen GeneTech Co., Ltd.). When tumors grew to an appropriate volume, tumor-bearing mice were randomly divided into treatment and control groups based on tumor volume, and dosing was initiated. The antibodies and ADCs were administered intravenously. Tumor growth inhibition (TGI) percentage was defined as %TGI = [1 − (Ti − T0)/(Vi − V0)] × 100% to evaluate the in vivo antitumor efficacy, where T0 and Ti are the mean tumor volumes of the treatment group on the day of randomization and the measurement day, respectively, and V0 and Vi are the mean tumor volumes of the control group on the day of randomization and the measurement day, respectively. Changes in body weight and mortality were recorded to assess the tolerability of the test samples. When the mice reached the endpoint (tumor volume exceeding 3,000 mm³, body weight loss >10%, or clinical symptoms indicating ethical reasons for euthanasia), they were euthanized with carbon dioxide gas.

PDX models

Two lung cancer PDX models, LU-01-0813 and LU-01-0561, were provided by WuXi AppTec Co., Ltd., and a colorectal cancer xenograft model, BP0170-R6P8, was provided by Beijing Biocytogen GeneTech Co., Ltd. A HuPrime breast cancer xenograft model derived from a patient with primary BR9801 breast cancer was provided by Crown Bioscience, Inc. IHC analysis of paraffin-embedded PDX tissue sections was performed by the respective companies. Primary tumor tissues were harvested from “breeder mice” and implanted subcutaneously into the right flank of NOD/SCID (LU-01-0813, LU-01-0561, and BR9801) or B-NDG (BP0170-R6P8) mice as tumor fragments (2–3 mm in length). In the LU-01-0813 and LU-01-0561 models, when the average tumor volume reached approximately 167 mm³ (n = 6), mice were randomly divided into five groups. Tumor-bearing mice received i.v. injections of hRS7-E+T at doses of 2.5, 5, and 10 mg/kg once weekly for 4 weeks and datopotamab deruxtecan (Dato-DXd) at a dose of 5 mg/kg once weekly for 4 weeks. In the BP0170-R6P8 model, when the average tumor volume reached approximately 221 mm³ (n = 7), the mice were randomly divided into three groups. Tumor-bearing mice received i.v. injections of hRS7-E+T and Datotuximab at doses of 5 mg/kg and 10 mg/kg, respectively, twice weekly. In the BP0170-R6P8 model, when the average tumor volume reached approximately 179 mm³ (n = 6), the mice were randomly divided into four groups. Tumor-bearing mice received i.v. injections of hRS7-E+T, Datotuximab, or sacituzumab govitecan at 5 mg/kg once weekly. The tumor volume and body weight were measured twice per week.

Toxicity evaluation in cynomolgus monkeys

This study included four cynomolgus monkeys, with equal numbers of males and females. Before dosing, male animals weighed between 3.1 and 3.5 kg, whereas female animals weighed approximately 2.6 kg. Animals were randomly divided into two groups based on sex and weight, with one male and one female in each group. The two groups were hRS7-E+T 50 mg/kg and hRS7-E+T 40 mg/kg. The drugs were administered via intravenous infusion at a rate of approximately 30 minutes per animal. The dosing was repeated every 3 weeks for a total of three doses. All the animals were euthanized and necropsied on day 50 after the first dose.

In vitro stability of hRS7-E+T in monkey and human plasma

To evaluate the in vitro stability of hRS7-E+T in monkey and human plasma, the payload concentration in the plasma was measured using LC/MS-MS. The payload release rate of hRS7-E+T at 1,500 μg/mL was detected in monkey plasma over 504 hours at 37°C, and the payload release rate of hRS7-E+T at 250 μg/mL was detected in human plasma samples over 504 hours at 37°C.

Statistical analysis

PK parameters, including half-life (T_1/2), time to peak concentration (T_max), peak concentration (C_max), AUC from 0 to t (AUC_0–t), and AUC from 0 to infinity (AUC_0–∞), were calculated using a noncompartmental model. In vitro data were analyzed and displayed using PRISM 5.0 (GraphPad). The EC₅₀ and IC₅₀ were calculated using a nonlinear regression model with S-shaped fitting. The in vitro antitumor activities of the drugs were evaluated using a two-tailed paired t test.

Results

Design and preparation of the dual-payload ADC

To overcome the limitations of single-payload ADCs, we designed an EXA/TPL dual-payload ADC. Random linkage of EXA to the cysteine of hRS7 was facilitated by a maleimide linker and a dipeptide cleavable linker, whereas TPL was directed to the modified sugar group at the N301 site of hRS7 through a tetrapeptide linker with a dibenzocyclooctyne group. It is estimated that each antibody molecule is coupled to approximately four EXAs and four TPLs.

We successfully developed a novel dual-payload TROP2-targeting ADC, hRS7-E+T (KH815), composed of a humanized IgG1 antibody directed against TROP2, conjugated to a TOP1 inhibitor and an RNA Pol II inhibitor, based on cysteine and glycosite antibody coupling technologies (Fig. 1A). The critical quality attributes of the hRS7-E+T were also assessed. Peptide mapping is crucial for evaluating the primary structure of hRS7-E+T and the stability of its production process. A reduced peptide map, combined with LC-MS detection, was used to identify the characteristic peptides of hRS7-E+T (Fig. 1B). Reverse phase high-performance liquid chromatography (RP-HPLC) and LC-MS analyses revealed homogeneous drug distribution, and the drug-to-antibody ratio (DAR) of hRS7-E+T was approximately 8 (4 + 4; Fig. 1C and D). Molecular size heterogeneity and isoelectric point are critical parameters for controlling the preparation process and stability of hRS7-E+T. The size-exclusion chromatography HPLC results showed that the hRS7, hRS7-E+T, thiol-coupled, and glycosyl-coupled ADCs exhibited high purity, with size-exclusion chromatography purity above 95% (Fig. 1E). Compared with hRS7, the isoelectric point of hRS7-E+T shifted toward the acid peak although no significant differences were observed between the mono- and dual-toxin ADCs (Fig. 1F). Additionally, Uncle (Unchained Labs) and hydrophilic interaction liquid chromatography HPLC analysis revealed that hRS7-E+T displayed a homogeneous particle size distribution, with a hydrated particle size of approximately 10 nm (Fig. 1G), excellent thermal stability (Tm1 of 62.5°C; Fig. 1H), and a reasonable glycoform ratio (Fig. 1I). The study and control of these key quality attributes are essential to ensure the safety, efficacy, and quality control of hRS7-E+T.

Figure 1.

The schematic structure and critical quality attributes of dual-payload ADC. A, The schematic structure of the TROP2-targeted dual-toxin ADC (hRS7-E+T/KH815). B, The reduced peptide map analysis of hRS7, thiol-coupled, and glycosyl-coupled ADC by LC-MS. C, The DAR and the conjugated drug distribution analysis of hRS7-E+T by LC-MS. D, The DAR and the conjugated drug distribution analysis of hRS7-E+T, thiol-coupled and glycosyl-coupled ADC by RP-HPLC. E, The size variants analysis of hRS7, hRS7-E+T, thiol-coupled, glycosyl-coupled and deglycosylated ADC by size-exclusion chromatography HPLC. F, The isoelectric point analysis of hRS7, hRS7-E+T, thiol-coupled, and glycosyl-coupled ADC by imaged capillary isoelectric focusing (icIEF). G, The particle size distribution analysis of hRS7-E+T by dynamic light scattering. H, The thermal stability analysis of hRS7-E+T by the Uncle multifunctional protein stability analyzer. I, The glycoform analysis of hRS7 by hydrophilic interaction liquid chromatography HPLC.

Drug release characteristics

ADCs consist of a mAb, linker, and small-molecule toxin. The core mechanisms of ADCs include target recognition, endocytosis, toxin release, and subsequent cell killing (21). The unique structure of dual-payload ADCs results in a distinctive drug release pattern that is crucial for their therapeutic efficacy.

In vitro stability studies under simulated plasma conditions revealed that the release rate of TPL from the antibody backbone was significantly higher than that of EXA. Plasma incubation with KH815 showed that the concentration of TPL-containing ADC molecules gradually decreased over time: 72.32% remained at 8 hours, 38.22% at 24 hours, 11.43% at 72 hours, 1.34% at 168 hours, 0.04% at 336 hours, and below the detection limit at 504 hours. In contrast, the release of EXA-containing ADC molecules was relatively slow: 107.71% remained at 8 hours, 100.32% at 24 hours, 72.23% at 72 hours, 67.89% at 168 hours, 57.25% at 336 hours, and 46.45% at 504 hours (Fig. 2A and B).

Figure 2.

In vivo and in vitro drug release profiles. A and B, Stability of dual-payload ADC in vitro plasma. Observe the changes in the contents of ADC with TPL or EXA (TPL-ADC or EXA-ADC) and small-molecule toxins (TPL and EXA) over time in human plasma and cynomolgus monkey plasma. C and D, For cynomolgus monkeys (n = 6) administered a single dose of dual-payload ADC (10 mg/kg), observe the PK characteristics of the changes in the contents of ADC with TPL or EXA (TPL-ADC or EXA-ADC) and small-molecule toxins (TPL and EXA) in vivo.

Consistent results were observed in the in vivo experiments. After a single i.v. injection of 10 mg/kg of the dual-payload ADC in cynomolgus monkeys, the half-life of TPL-containing ADC molecules was 22.6 hours, whereas that of EXA-containing ADC molecules was approximately 56.2 hours (Fig. 2C). The release of the small-molecule toxins showed that EXA could be detected up to 168 hours after administration, whereas TPL levels were below the detection limit, possibly due to its shorter half-life (Fig. 2D). This dual asynchronous release mechanism may have contributed to the drug’s in vivo efficacy studies.

Mechanisms of action of the two small-molecule toxins

EXA, a derivative of CPT, acts as a TOP1 inhibitor and shows significant antitumor activity. TPL, on the other hand, induces the degradation of RNA Pol II subunit A (POLR2A) and inhibits its activity, thereby suppressing gene transcription and ultimately causing cell-cycle arrest and apoptosis.

NCI-H292 cells were treated with EXA or TPL for 1.5 hours and then cultured for 30 minutes under drug-free conditions. The results showed that EXA primarily induced the degradation of TOP1, whereas TPL primarily induced the degradation of POLR2A (Fig. 3A–C). Overexpression of ABCB1 and HSP70 is a key factor leading to acquired resistance to multiple chemotherapeutic drugs in tumor cells. After treating HCT-15 cells with these drugs, the mRNA expression levels of Abcb1, Hsp70, and Top1 were measured. The results indicated that TPL reduced the mRNA expression of HSP70, ABCB1, and TOP1, whereas EXA alone had no such effect. The combination of both drugs was most effective in reducing the transcription levels of these three genes (Fig. 3D).

Figure 3.

EXA and TPL act on target cells through different mechanisms. A, NCI-H292 cells were incubated with 1 μmol/L EXA or 1 μmol/L TPL for 1.5 hours, respectively. Then, the levels of TOP1 and POLR2A were determined by Western blotting. B and C, Quantitative analysis was performed on the total TOP1 and POLR2A bands in the cells. The intensity of the bands was analyzed using ImageJ software, and normalization was carried out with α-tubulin, which served as a loading control. D, HCT-15 cells were treated with 100 nmol/L EXA, 100 nmol/L TPL, or 100 nmol/L EXA + 100 nmol/L TPL for 12 hours. Then, the expression of Hsp70 and Abcb1 genes was evaluated by qPCR. The data were presented as fold changes relative to the vehicle (control) group. Statistical analysis: Unpaired t test was used, and the data were expressed as mean ± SE (n = 6 per group). E, The inhibitory activity of EXA, TPL, and EXA + TPL on the TOP1 enzyme was detected in NCI-H292 cells after 2 and 4 hours of treatment. Plasmid DNA was separated by agarose gel electrophoresis and stained with SYBR Safe DNA Gel Stain. F, Western blotting analysis of HSP70 in heat-shocked NCI-H292 cells was conducted. The cells were heat-shocked at 43°C for 2 hours and then incubated at 37°C for 0 to 4 hours, with or without 1 μmol/L TPL. The control lane contained proteins extracted from untreated cells. G, Representative images of cell states under different treatments were presented. H, Flow cytometry analysis of apoptosis in cells under different treatments was performed.

Heat shock has been proven to be a simple method to effectively increase HSP70 protein expression in cells. In this study, HSP70 expression in NCI-H292 cells was significantly upregulated after 2 hours of heat shock at 43°C, followed by 4 hours of culture at 37°C. However, in the presence of TPL, HSP70 expression did not increase significantly (Fig. 3F). Combined experiments using Acridine Orange-Propidium Iodide (RE010212, Countstar) reagent for cell death detection and flow cytometry for apoptosis analysis showed that heat-shocked cells were resistant to EXA-induced apoptosis, whereas the combination of TPL and EXA was highly effective in killing heat-shocked cells (Fig. 3G and H). Treatment of non–heat-shocked cells with 1 and 10 μmol/L EXA resulted in similar percentages of apoptotic cells (approximately 20%). In contrast, heat-treated cells showed reduced sensitivity to EXA, with apoptotic cell percentages of approximately 5% at 1 μmol/L EXA and 10% at 10 μmol/L EXA. The combination of EXA and TPL significantly affected both cell death and apoptosis. Flow cytometry analysis revealed that the percentage of apoptotic cells increased to approximately 15% (Fig. 3H). In NCI-H292 cells, each drug inhibited TOP1 activity. The level of relaxed DNA decreased after 2 hours of treatment with the EXA and EXA + TPL combination, and the difference became more pronounced after 24 hours. TPL did not cause significant changes in relaxed DNA levels after 2 hours of treatment but led to a significant reduction after 24 hours. Moreover, the EXA + TPL combination induced a more pronounced change compared with TPL alone (Fig. 3E).

hRS7-E+T induces tumor cell killing by inhibiting TOP1 and POLR2A

To thoroughly elucidate the cell-killing mechanism of the dual-payload ADC hRS7-E+T, we conducted a detailed investigation of its effects on the activity levels of POLR2A and TOP1. At the protein level, compared with single-payload ADCs, the same concentration of the dual-payload ADC (hRS7-E+T) significantly reduced the protein levels of POLR2A and TOP1 (Supplementary Fig. S2A–S2C).

This result was further validated at the mRNA level. The data showed that hRS7-E+T effectively downregulated the expression of TOP1, whereas hRS7-EXA did not reduce TOP1 expression at the mRNA level. Additionally, when cells were treated with hRS7-EXA, the expression of HSP70 was significantly upregulated; however, hRS7-E+T not only inhibited this upregulation but also reduced the transcriptional level of HSP70 to pretreatment levels (Supplementary Fig. S2D). Moreover, compared with single-payload ADCs, hRS7-E+T significantly reduced the production of relaxed DNA (Supplementary Fig. S2E), clearly indicating more effective inhibition of TOP1 activity.

In vitro tumor activity assessment of dual-payload ADC molecules

To verify whether the anti-TROP2 mAb exerts its targeting function through specific recognition of highly expressed TROP2 antigens on the surface of tumor cells, we conducted a series of experiments. First, at the protein level, we measured the affinity of hRS7-E+T and hRS7 for TROP2 protein, obtaining EC₅₀ values of 0.06 and 0.03 nmol/L, respectively (Fig. 4A). This result indicated that the conjugation of EXA and TPL to the antibody did not significantly affect the binding ability of the mAb to the TROP2 target.

Figure 4.

In vitro activities of hRS7-E+T. A, Binding to human TROP2 estimated by ELISA. B, hRS7-E+T binding to NCI-H292 was estimated by FACS. C, Internalization of hRS7-E+T in NCI-H292 cells. D–F,In vitro growth inhibitory activity (mean ± SEM) in NCI-H292, NCI-H2170, and NUGC-4. G–I, Patient-derived organoid response to hRS7-E+T and Dato-DXd. Lung cancer patient-derived organoids were treated with hRS7-E+T and Dato-DXd at different concentrations. HTC, highest tested concentration. OD, optical density; MFI, mean fluorescence intensity.

Additional cell-based experiments demonstrated that the binding affinity of hRS7-E+T in TROP2-positive NCI-H292 cells was comparable with that of hRS7, with respective EC₅₀ values of 17.83 and 4.70 nmol/L (Fig. 4B). These data strongly demonstrate that the targeting function of hRS7-E+T primarily relies on the specific binding of the anti-TROP2 antibody to TROP2-positive cells, and the introduction of payloads did not significantly interfere with the interaction between the antibody and the target. Additionally, endocytosis experiments showed that both hRS7-E+T and hRS7 exhibited a significant increase in the endocytosis rate over time in TROP2-positive cells (Fig. 4C). In summary, hRS7-E+T can successfully mediate targeted endocytosis by specifically binding to TROP2 proteins expressed on the cell surface.

To comprehensively evaluate the antitumor activity of hRS7-E+T and Dato-DXd, we used the CellCounting-Lite 2.0 luminescent cell cytotoxic activity assay to conduct experiments in cancer cell lines with high, medium, and low TROP2 expression (NCI-H292, NCI-H2170, NUGC-4) as well as in patient-derived organoids (LU0743B, LU1542B, LU6437B; Fig. 4D–I). The results clearly showed that hRS7-E+T exhibited significantly better inhibition of tumor cell growth than Dato-DXd, and its activity was evident across different levels of TROP2 expression (high, medium, and low). This indicates that hRS7-E+T, through specific binding to TROP2 on the cell surface and subsequent endocytosis, releases dual-payload small molecules to effectively inhibit tumor cell proliferation.

Comparison of efficacy between dual-payload ADCs (hRS7-E+T and patritumab-E+T) and DXd ADCs

Both hRS7-E+T and Dato-DXd are ADC drugs that target TROP2, but they use different antibodies. Our research found that hRS7-E+T demonstrated significantly better efficacy than Dato-DXd in tumors with low TROP2 expression.

We determined the minimum effective dose (MED) for inducing tumor regression in vivo using a drug titration method. In the CDX model derived from the NCI-H292 cell line with high TROP2 expression, when the average tumor volume reached 165 mm³, animals were precisely grouped and dosed on days 0 and 7. The results showed that at the same dose, hRS7-E+T had a significantly stronger inhibitory effect on tumor growth than Dato-DXd. Specifically, the TGI rate for the Dato-DXd 5 mg/kg group was 97.3%, whereas the TGI for hRS7-E+T at doses of 2.5, 5, and 10 mg/kg was 69.8%, 105.5%, and 108%, respectively. Notably, in the 10 mg/kg hRS7-E+T group, all five animals achieved complete tumor regression by day 31 (Fig. 5A). Thus, the MED for hRS7-E+T in NCI-H292 cells was less than 2.5 mg/kg.

Figure 5.

In vivo antitumor activity of dual-payload ADCs in CDX and PDX modes. A, Antitumor efficacy of hRS7-E+T in the NCI-H292 xenograft model. The tumor-bearing mice were intravenously administered hRS7-E+T (2.5, 5, and 10 mg/kg, respectively), Dato-DXd (5 mg/kg), or vehicle control at a dose of once every week for twice, N = 8. B, Antitumor efficacy of hRS7-E+T in the NCI-H2170 xenograft model. The tumor-bearing mice were intravenously administered hRS7-E+T (1.25, 2.5, and 5 mg/kg, N = 10, respectively), Dato-DXd (5 mg/kg, N = 30), or vehicle control (N = 10) at a dose of once every two weeks for twice. C, Antitumor efficacy of hRS7-E+T in the NUGC-4 xenograft model. The tumor-bearing mice were intravenously administered hRS7-E+T (2.5, 5, and 10 mg/kg, respectively), Dato-DXd (5 mg/kg), or vehicle control at a dose of once every week for three times, N = 8. D, Antitumor efficacy of patritumab-E+T in the NUGC-4 xenograft model. The tumor-bearing mice were intravenously administered patritumab-E+T (N = 8), patritumab-DXd (N = 16), or vehicle control (N = 8) at the indicated dose. E, Antitumor efficacy of patritumab-E+T in the NCI-H2170 xenograft model. The tumor-bearing mice were intravenously administered patritumab-E+T (N = 8), patritumab-DXd (N = 16), or vehicle control (N = 8) at the indicated dose. F, Antitumor efficacy of patritumab-E+T in the Colo205 xenograft model. The tumor-bearing mice were intravenously administered patritumab-E+T, patritumab-DXd, or vehicle control at the indicated dose, N = 10. G, Antitumor efficacy of hRS7-E+T in the lung cancer PDX (LU-01-0813) model. The tumor-bearing mice were intravenously administered hRS7-E+T (2.5, 5, or 10 mg/kg), Dato-DXd, or vehicle control at a dose of once every two weeks for four times (N = 6). H, Antitumor efficacy of hRS7-E+T in the lung cancer PDX (LU-01-0561) model, the tumor-bearing mice were intravenously administered hRS7-E+T (2.5, 5, or 10 mg/kg), Dato-DXd, or vehicle control at a dose of once every two weeks for four times (N = 6). I, Antitumor efficacy of hRS7-E+T in the colorectal cancer PDX (BP0170-R6P8) model, the tumor-bearing mice were intravenously administered hRS7-E+T, Dato-DXd, or vehicle control at the indicated dose (N = 7).

In contrast, in the NCI-H2170 model with medium TROP2 expression and the NUGC-4 model with low TROP2 expression, although the overall efficacy of ADC drugs decreased, hRS7-E+T still demonstrated superior tumor suppression compared with Dato-DXd and could achieve the same or even lower doses to inhibit tumor growth. Specifically, in NCI-H2170 cells, the MED for hRS7-E+T was approximately 2.5 mg/kg; in NUGC-4 cells, the MED for hRS7-E+T was approximately 5 mg/kg, whereas the same dose of Dato-DXd had a weaker effect on tumor growth (TGI = 13.7%; Fig. 5B and C).

Similar results were observed with another dual-payload ADC targeting HER3, patritumab-E+T. We established CDX models using three cell lines: NUGC-4 cells with high HER3 expression, H2170 cells with medium HER3 expression, and COLO205 cells with low HER3 expression. The experimental results showed that patritumab-E+T exhibited better efficacy than the single-payload DXd ADC patritumab-DXd in tumors with low HER3 expression (Fig. 5D–F).

Dual-payload ADC shows more persistent antitumor responses in PDX models

To further investigate the therapeutic effects of hRS7-E+T and Dato-DXd on TROP2-high-expressing tumors, we conducted a series of experiments using two lung cancer PDX models: LU-01-0813 and LU-01-0561. When the tumor volumes reached 156 and 167 mm³, respectively, we initiated a dosing regimen once weekly for four consecutive weeks.

In the LU-01-0813 model, 28 days after treatment, we observed that both 2.5 mg/kg of hRS7-E+T and Dato-DXd showed similar TGI, with TGI rates of 59.5% and 63.1%, respectively. As the dose of hRS7-E+T increased to 5 and 10 mg/kg, the antitumor efficacy became more pronounced, with TGI values reaching 74.4% and 106%, respectively (Fig. 5G). In the LU-01-0561 model, hRS7-E+T cells demonstrated significantly better TGI than Dato-DXd cells. At a dose of 2.5 mg/kg, hRS7-E+T achieved a TGI of 55%; at 5 and 10 mg/kg, TGI values were 101.7% and 113.6%, respectively. In contrast, Dato-DXd at 5 mg/kg achieved a TGI of only 68% (Fig. 5H).

In the colorectal cancer PDX model (BP0170-R6P8), the same dosing strategy was used for both Dato-DXd and hRS7-E+T. When the average tumor volume reached 221 mm³, the first dose of 5 mg/kg was administered, followed by additional doses of 10 mg/kg on days 7 and 14. Dynamic monitoring of tumor growth showed that both drugs had similar inhibitory effects after the first two doses. However, significant differences emerged after the final dose: Tumors in the Dato-DXd group resumed growth, whereas those in the hRS7-E+T group continued to shrink. This observation strongly suggests that repeated dosing may induce resistance to Dato-DXd, whereas hRS7-E+T maintains effective antitumor activity (Fig. 5I).

Dual-payload ADC overcomes resistance to Dato-DXd

Multidrug resistance (MDR) transporters, such as ABCG2 and ABCB1, have been identified as key factors in chemotherapy resistance, primarily through active drug efflux, ultimately leading to treatment failure (22, 23). Previous studies have shown that the combination of IMMU-132/gosatuzumab (TRODELVY) with the ABCG2 inhibitor YHO-13351 can effectively overcome resistance to SN-38 in breast and gastric cancer models (24).

In this context, we selected the P-gp–positive cell line HCT15, which overexpresses ABCB1, and generated an HCT15 cell line overexpressing human TROP2 (HCT15-TROP2) to evaluate the therapeutic differences between Dato-DXd and hRS7-E+T for targeting TROP2. The in vivo experimental results clearly showed that Dato-DXd had almost no therapeutic effect on HCT15-TROP2 cells, whereas hRS7-E+T exhibited good therapeutic activity. Under the experimental conditions of three doses of 5 mg/kg every 2 weeks until day 46, hRS7-E+T achieved a TGI of 89.5%, whereas Dato-DXd only achieved 31.4% (Fig. 6A).

Figure 6.

Dual-payload ADCs overcome resistance. A, hRS7-E+T overcomes multidrug resistance. The tumor-bearing mice were intravenously administered hRS7-E+T (5 mg/kg), Dato-DXd (5 mg/kg), or vehicle control at a dose of once every two weeks for three times (N = 7). B, hRS7-E+T overcomes Dato-DXd treatment resistance in NCI-H2170 tumors. The tumor-bearing mice were intravenously administered hRS7-E+T (10 mg/kg) or Dato-DXd (5 or 10 mg/kg) for one dose (N = 9). C, Patritumab-E+T overcomes patritumab-DXd treatment resistance in NCI-H2170 tumors. The tumor-bearing mice were intravenously administered patritumab-E+T (10 mg/kg) or patritumab-DXd (10 mg/kg) once every week for four times (N = 8). D, Patritumab-E+T overcomes patritumab-DXd treatment resistance in NUGC-4 tumors. The tumor-bearing mice were intravenously administered patritumab-E+T (10 mg/kg, N = 7) or patritumab-DXd (10 mg/kg, N = 8) once every week for three times. E, IHC analysis of TROP2 expression in the R9801 xenografted tumors. F, hRS7-E+T overcomes TRODELVY treatment resistance in TNBC PDX (BR9801). The tumor-bearing mice were intravenously administered hRS7-E+T (5 mg/kg), TRODELVY (5 mg/kg), Dato-DXd (5 mg/kg), or vehicle control at a dose of once every week for twice (N = 6).

In studies of low Trop2-expressing tumor models, the dual-payload ADC successfully overcame intrinsic resistance to DXd ADC. In the NUGC-4 low TROP2-expressing CDX model resistant to Dato-DXd, hRS7-E+T demonstrated strong antitumor activity (Fig. 5C). In the NCI-H2170 model with low HER3 expression, patritumab-DXd treatment was ineffective, showing no significant difference compared with the PBS control group; however, patritumab-E+T treatment led to tumor regression (Fig. 5E).

To further verify the ability of the dual-payload ADC to overcome resistance to DXd-ADC, we examined the growth of NCI-H2170 tumors treated with patritumab-DXd and Dato-DXd, as well as NUGC-4 tumors treated with patritumab-DXd (Fig. 6B–D). After two doses of Dato-DXd, NCI-H2170 tumors grew to approximately 500 mm³ (Fig. 5B). At this point, a single dose of hRS7-E+T (10 mg/kg) caused significant tumor regression, whereas single doses of Dato-DXd (5 and 10 mg/kg) had no noticeable effect (Fig. 6B). Similarly, in NCI-H2170 cells treated with patritumab-DXd, tumors grew to approximately 600 mm³ after two doses (Fig. 5D). Subsequent administration of four doses of patritumab-E+T (10 mg/kg) led to significant tumor regression, whereas one dose of patritumab-DXd (10 mg/kg) had no significant effect (Fig. 6C). In the NUGC-4 tumor experiment, after three doses of patritumab-DXd on day 42, the tumor volume reached approximately 1,000 mm³. Three doses of patritumab-E+T (10 mg/kg) caused tumor regression, whereas patritumab-DXd (10 mg/kg) did not (Fig. 6D).

Additionally, we used a high TOP2-expressing PDX model derived from a patient who received gosatuzumab and EXA combined with α-helical peptides. The data showed that the administration of Dato-DXd or gosatuzumab at 5 mg/kg twice weekly failed to effectively inhibit tumor growth. However, at the same dose, hRS7-E+T achieved a TGI of 56.93% (Fig. 6E and F). Collectively, these experimental results demonstrate that the dual-payload ADC has significant efficacy against tumors that are resistant to DXd/SN-38/EXA ADC treatment.

Non-GLP toxicity study in monkeys

To assess the safety of hRS7-E+T, we conducted a non-GLP toxicity study in cynomolgus monkeys. During the experiment, monkeys received repeated doses of 40 and 50 mg/kg via intravenous infusion every 3 weeks (on days 1, 22, and 43, with no recovery period). All the animals survived until necropsy was scheduled.

We found that doses of 40 mg/kg and above of hRS7-E+T caused hair loss, changes in skin color (darkening), and skin cracking in monkeys. Reversible effects  were also noted on the hematopoietic system (granulocytes and erythroid lineage) and liver function (alanine transaminase, aspartate transaminase, and lactate dehydrogenase). However, doses of 40 mg/kg or higher did not significantly affect body weight, temperature, blood pressure, electrocardiogram, coagulation function, or immune parameters (lymphocyte immunophenotyping and cytokines). No abnormalities related to test substance administration were observed in the bone marrow smears.

Histopathologic examination revealed that hRS7-E+T–related histopathologic changes were primarily localized to the skin. Perivascular pigmentation was observed in the lung tissues of female and male animals in the 50 mg/kg dose group, while a definitive link between this finding and the test article remains undetermined. Based on the comprehensive experimental data, the HNSTD for monkeys was determined to be 50 mg/kg.

Discussion

CPT derivatives, particularly TOP1 inhibitors, have gained significant attention as payloads for ADCs, largely owing to the clinical success of DS-8201a (ENHERTU) and sacituzumab govitecan (TRODELVY; ref. 25). Currently, most ADCs under development are single-payload ADCs that use CPT derivatives. Although ADC therapy often produces positive initial results, drug resistance frequently occurs (26). In recent years, various mechanisms underlying tumor resistance to ADCs have been elucidated, including downregulation of target antigens and upregulation of drug efflux pumps (27, 28). Combination chemotherapy, involving the use of ADCs with multiple drugs or multipayload ADCs, holds promise in overcoming ADC-related resistance (22, 29). To address this, our dual-payload ADC is engineered with linkers that undergo cleavage at distinct times, enabling staggered release of the two payloads. The linker conjugated to EXA exhibits higher stability, designed to be cleaved after the ADC is internalized by targeted cancer cells—this allows EXA to exert potent cytotoxic effects directly on the target cells. Additionally, EXA can be re-released by endosomes into the broader tumor microenvironment, thereby killing adjacent cancer cells. In contrast, the linker conjugated to TPL is designed to undergo cleavage in the tumor microenvironment prior to ADC internalization, which helps eliminate surrounding cancer cells and enhances the agent’s bystander effect on the broader tumor tissue.

EXA is one of the most potent TOP1 inhibitors used clinically. It exhibits 10 to 50 times greater cytotoxic activity against various tumor cells than the second most active TOP1 inhibitor, SN-38, potentially offering superior therapeutic outcomes (23). EXA may offer better therapeutic benefits than other clinical TOP1-targeted agents (24). Although it has been reported that EXA is a poor substrate for P-gp and remains effective against P-gp–mediated multidrug-resistant cells (30), recent studies have shown that SLFN11-negative tumor cells exhibit reduced sensitivity to EXA, suggesting potential mechanisms of EXA resistance. The upregulation of HSP70 is closely associated with chemoresistance in various cancers and resistance to CPTs (14, 31, 32). Our data also suggest the existence of potential mechanisms of EXA resistance, necessitating targeted interventions.

TPL is a promising natural antitumor agent. It suppresses gene transcription by inhibiting RNA Pol, demonstrating broad-spectrum antitumor activity against various cancers (33). TPL also exhibits cytotoxic activity against multidrug-resistant cancer cells and some cancer stem cells. However, its poor water solubility and instability significantly limit its in vivo activity, and it is associated with several severe adverse reactions, including hepatotoxicity, which limit its clinical application (34). TPL conjugated to antibodies represents a new breakthrough in cancer treatment. However, when used as a payload in ADCs, TPL typically requires higher doses to achieve the desired tumor suppression. According to the Cortellis Drug Discovery Intelligence database, no TPL-ADC drug has been approved for clinical use.

In our dual-payload ADC, the combination of EXA and TPL successfully overcomes the resistance of tumor cells to single-agent drugs. For EXA, our study confirmed that upregulation of HSP70 reduces cellular sensitivity to EXA. TPL, by reducing RNA Pol II, significantly lowers the transcription levels of genes, including HSP70, making cells more sensitive to the combined action of TPL and EXA. For example, in the PDX model R9801 derived from a patient whose disease progressed despite treatment with Alphalex-EXA, the tumor likely developed resistance to EXA-based ADCs. In contrast, hRS7-E+T cells, which carry both EXA and TPL, remained highly effective in suppressing this tumor. EXA selectively inhibits TOP1 activity, affecting DNA, whereas TPL selectively inhibits RNA Pol II, affecting the transcription of numerous genes. The synergistic integration of these two mechanisms is a distinctive feature of the dual-payload ADC. Compared with traditional ADCs, dual-payload ADCs offer significant advantages in overcoming tumor heterogeneity. However, their application faces numerous challenges. The PK properties of dual-payload ADCs are complex and involve the synergistic actions and PK of the two drug payloads. Ensuring the stability and activity of both payloads during conjugation is a key challenge. Given the potential complexity of the toxic responses, more extensive preclinical and clinical studies are required.

Our study utilized the GGFG linker to conjugate TPL to the target antibody via glycosylation, aiming to stabilize the drug preparation process and achieve a DAR of 4. Simultaneously, the MC-VA linker was used to randomly conjugate EXA to the cysteine of the antibody, achieving an average DAR of approximately 4. This method not only facilitates precise control over drug production but also allows for the successful conjugation of both drugs to a single antibody without compromising physical properties such as solubility. These factors ensure the controllability of the drug production process, enabling the rational loading of two drugs on a single antibody while avoiding the degradation of physical properties owing to the high payload. Studies have shown that in tumor homogenates at pH 5, the release rate of TPL is faster than that of EXA, allowing the dual-payload ADC to preferentially release TPL into the tumor microenvironment. This precise drug release pattern rapidly reduced the expression of EXA resistance genes by inhibiting RNA Pol II, followed by the relatively slower release of EXA, which achieved sustained and effective inhibition of tumor growth. This further highlights the advantages of precise drug delivery in cancer treatment.

In summary, our study demonstrated that the dual-payload ADC carrying EXA and TPL is more effective in cancer treatment. Current data show that it has excellent therapeutic efficacy and controllable toxicity, thus providing a highly promising new strategy for cancer treatment.

Conclusion

This study successfully developed a novel dual-payload ADC platform integrating EXA and TPL, achieving precise dual targeting of tumor cells and overcoming resistance issues associated with traditional single-agent therapies. Both in vitro cell and in vivo model experiments demonstrated that this dual-payload ADC has superior antitumor activity, outperforming the control drugs in various tumor models. Additionally, it exhibits good safety, with non-GLP toxicity studies in monkeys showing reversible effects on the skin and no significant adverse effects on key physiologic parameters, with the HNSTD of 50 mg/kg. This platform offers an innovative and promising strategy for cancer treatment, with the potential to provide more effective treatment options and better patient outcomes.

Supplementary Material

Supplementary Data

This supplemental file contains three core parts: Supplemental Materials and Methods, Supplemental Figure S1, and Supplemental Figure S2. Supplemental Materials and Methods Details the experimental protocols and technical parameters of all in vitro, in vivo, and analytical experiments in this study, laying a foundation for the reproducibility of the research. Supplemental Figure S1 Presents the amino acid sequences of Datopotamab, Patritumab and hRS7 (sacituzumab), along with the drug-antibody ratio (DAR) distribution data of the dual-payload ADC hRS7-E+T. Supplemental Figure S2: Shows the key mechanistic details of how the dual-payload ADC hRS7-E+T acts on target cells through different pathways.

Data Availability

The amino acid sequences of the antibodies used in this study—Dato (chain ID: 11218_H, 11218_L), hRS7 (chain ID: 10418_H, 10418_L), and patritumab (chain ID: 11093_H, 11093_L)—were retrieved from the international ImMunoGeneTics information system, a publicly available repository. These sequences can be accessed via the international ImMunoGeneTics information system using the respective chain IDs provided above. All relevant data supporting the findings of this study are included in this article and its supplementary material. Additional raw datasets generated and/or analyzed during the current study (not contained in the article or supplements) are available from the corresponding authors (Xiao Ke and Yonghao Zhao) upon reasonable request, subject to ethical and privacy considerations.

Authors’ Disclosures

No disclosures were reported.

Authors’ Contributions

P. Ren: Conceptualization, software, writing–original draft. M. Guan: Data curation, validation. J. Tang: Formal analysis, methodology. S. Yin: Resources, validation, investigation. L. Qi: Data curation, investigation, methodology. J. Yue: Data curation, software, formal analysis. Z. Li: Validation. X. Fan: Formal analysis. G. Lei: Formal analysis. T. Zuo: Formal analysis. J. Chen: Methodology. Y. Xu: Conceptualization, resources, writing–review and editing. X. Ke: Conceptualization, resources, funding acquisition. Y. Zhao: Conceptualization, resources, funding acquisition, writing–review and editing.