Triple-negative breast cancer (TNBC) is a highly aggressive and heterogeneous disease that often relapses following treatment with standard radiotherapies and cytotoxic chemotherapies. Combination therapies have potential for treating refractory metastatic TNBC. In this study, we aimed to develop an antibody–drug conjugate with dual payloads (DualADC) as a chemoimmunotherapy for TNBC. The overexpression of an immune checkpoint transmembrane CD276 (also known as B7-H3) was associated with angiogenesis, metastasis, and immune tolerance in more than 60% of patients with TNBC. Development of a mAb capable of targeting the extracellular domain of surface CD276 enabled delivery of payloads to tumors, and a platform was established for concurrent conjugation of a traditional cytotoxic payload and an immunoregulating Toll-like receptor 7/8 agonist to the CD276 mAb. The DualADC effectively killed multiple TNBC subtypes, significantly enhanced immune functions in the tumor microenvironment, and reduced tumor burden by up to 90% to 100% in animal studies. Single-cell RNA sequencing, multiplex cytokine analysis, and histology elucidated the impact of treatment on tumor cells and the immune landscape. This study suggests that the developed DualADC could represent a promising targeted chemoimmunotherapy for TNBC.

Significance: An anti-CD276 monoclonal antibody conjugated with both a cytotoxic drug and an immune boosting reagent effectively targets triple-negative breast cancer by inducing tumor cell death and stimulating immune cell infiltration.

Triple-negative breast cancers (TNBC) represent a highly aggressive, metastatic, and heterogeneous group of tumors with distinct subtypes, including basal-like 1, basal-like 2 (BL2), mesenchymal-like, mesenchymal stem–like, luminal androgen receptor (LAR), immunomodulatory, and unstable (1). Unfortunately, patients with advanced TNBC encounter a daunting prognosis, with a 5-year survival rate of 59% at stage III and a stark 11% at stage IV, and a high rate of recurrence following standard treatment involving surgery, radiotherapy, and cytotoxic chemotherapies like doxorubicin and paclitaxel (2, 3). The FDA has approved the anti-trophoblast cell surface antigen 2 antibody–topoisomerase I inhibitor SN-38 conjugate (sacituzumab govitecan) to treat refractory metastatic TNBC in patients who have undergone at least two systemic therapies (4). However, the majority of patients with TNBC fail to benefit from conventional antibody–drug conjugates (ADC) that carry a single cytotoxic payload, due to the limited clinical efficacy in the recurrent and metastatic setting, cancer phenotypic heterogeneity (5, 6), and/or development of drug resistance (7). A recent phase III trial showed that combining pembrolizumab with chemotherapies, such as paclitaxel, nanobody–paclitaxel, or gemcitabine–carboplatin, significantly improved the progression-free survival in patients with metastatic TNBC (8). These advancements underscore the great potential of targeted therapies and therapeutic combinations in TNBC treatment.

TNBCs are characterized by the absence of estrogen receptor, progesterone receptor, and HER2 expression. To develop effective targeted therapies, numerous surface receptors have been explored. Recently, the transmembrane protein CD276 (B7-H3, UniProt: Q5ZPR3; ref. 9), comprised of two Ig-like V-type and two Ig-like C2-type extracellular domains, was detected in more than 80% of breast cancer tissues (10). Consistent with the literature (11), The Cancer Genome Atlas (TCGA) and a recent clinical trial of enoblituzumab (10, 12), our study revealed high expression of CD276 in more than 60% of patients with TNBC (126 cases) and various cell lines representing different subtypes, whereas minimal to low expression levels were observed in 33 normal human organs examined. Importantly, CD276 has been implicated to associate with angiogenesis, invasion, metastasis, and poor prognosis in patients with cancer. Furthermore, CD276 is recognized as an immune checkpoint molecule that inhibits the secretion of effector cytokines (IFNγ, TNFα, and IL4), as well as the immune function of NK and T cells (13). Recently, the combined anti-CD276 enoblituzumab/anti-PD-1 retifanlimab, vobramitamab duocarmazine, and bispecific antibody targeting CD276/CD3 are evaluated to treat head/neck cancer in phase I trial (NCT02475213) or multiple cancers in phase I/II trial (MGD009 and MGC018; ref. 12). The clinical data, literature reports, and our study collectively indicated that targeting CD276 could cover the majority of patients with TNBC and upregulate tumoral immunity, rendering it a promising therapeutic strategy for aggressive TNBCs. We, therefore, developed and engineered a new CD276 mAb to construct a combined chemoimmunotherapy in this study.

The agonists of Toll-like receptors (TLR), typically expressed in T cells, NK cells, dendritic cells, macrophages, and other monocytes (14), have been developed and actively investigated in both preclinical studies (15) and clinical trials (16) for the treatment of cancers and various diseases. The literature and our study have demonstrated that TLR7/8 agonists play a pivotal role in recruiting and activating immune cells within the immunologically “cold” tumor microenvironment (TME; ref. 17). Importantly, these agonists exhibit lower toxicity than immune checkpoint blockers, such as anti-PD-1/PD-L1 mAbs (18). Furthermore, TLR7/8 agonists have been found to impede cancer cell proliferation (17), induce apoptosis (19), and stimulate the release of cytokines (e.g., IL2/6/8/10/12/18, IFNα/γ, and TNFα) by immune cells (20, 21). Despite these promising immunotherapeutic effects, the administration of free TLR agonists lacking tumor selectivity may induce a potentially lethal cytokine storm and other adverse side effects. To overcome this challenge, the present study used anti-CD276 mAb to precisely deliver TLR7/8 agonists, thereby fostering the upregulation of tumoral immunity in TNBC.

Traditional ADCs that carry a microtubule-targeting agent [e.g., emtansine (DM1), ravtansine, and monomethyl auristatin E/monomethyl auristatin F (MMAF)] or DNA-damaging substrate (e.g., SN-38) have been actively investigated in clinical trials or approved by the FDA (22, 23). Whereas these ADCs offer notable advantages in terms of tumor specificity and minimal adverse effects (24), the clinical effectiveness of single-payload ADCs can be compromised by the unpredictable compensatory mechanisms, emergence of drug resistance during prolonged treatment, and inherent heterogeneity of cancers (25). To overcome these limits of traditional ADCs and address the off-target–induced immune toxicity of free TLR agonists, thereby improving tumor treatment efficacy, we established an advanced conjugation platform of dual-payload ADC (named DualADC), in which one mAb carries both highly cytotoxic chemotherapy and immunotherapy.

The objective of this study was to develop and evaluate an innovative immune checkpoint CD276-targeted DualADC for chemoimmunotherapy of TNBCs. It is hypothesized that our DualADC integrating cancer proliferation inhibition, tumoral cytokine enhancement, immune cell reactivation, and TME modulation could effectively eradicate TNBC cells in vivo. A unique mAb with cross activity was developed and further engineered to target CD276+ TNBC patients. A new platform was established to conjugate the engineered mAb with synergistic dual therapies through two linkers. Our all-in-one ADC reduced tumor burden by 90% to 100% in TNBC xenografted mouse models including patient-derived xenograft (PDX) models. Furthermore, the underlying anticancer mechanisms identified from various posttreatment analyses confirmed our hypothesis. Taken together, the DualADC developed in this study represents a viable strategy for the treatment of aggressive TNBCs.

The animal studies were conducted according to the Institutional Animal Care and Use Committee Protocols 2023A00000039 and 2022A00000071, which were approved by the Institutional Biosafety Committee at the Ohio State University.

Cell lines and culture media

The human TNBC cell lines, including MDA-MB-231 (ATCC, cat. #HTB-26, RRID: CVCL_0062), MDA-MB-468 (ATCC, cat. #HTB-132, RRID:CVCL_0419), MDA-MB-231-FLuc (GenTarget, cat. #SC059-Puro, RRID:CVCL_YZ80), and MDA-MB-468-FLuc (GeneCopoeia, cat. #SL027, RRID: CVCL_C8XW) were maintained in DMEM medium supplemented with 10% FBS (v/v) and 1% pen/strep (v/v). The normal breast epithelium 184B5 cell line (ATCC, cat. #CRL-8799, RRID: CVCL_4688) was maintained in MEGM BulletKit growth medium (Lonza) supplemented with 5% FBS. The mouse TNBC cell lines 4T1 (ATCC, cat. #CRL-2539, RRID:CVCL_0125) and 4T1-FLuc (ATCC, cat. #CRL-2539-LUC2, RRID:CVCL_5I85) were cultivated in RPMI-1640 medium supplemented with 10% FBS and 1% pen/strep. The mice carrying TNBC PDX (The Jackson Laboratory, cat. #J000103917) were purchased from The Jackson Laboratory; then PDX was harvested, passaged, and maintained in NOD/SCID gamma (NSG) mice or freshly frozen and stored in a liquid nitrogen tank. The Expi293 cells for chimeric CD276 mAb production were maintained in Expi293 Expression medium supplemented with 4 mmol/L GlutaMAX and 6 g/L glucose. The Chinese hamster ovary (CHO) cells producing humanized anti-CD276 mAb were kept in Dynamis medium supplemented with 4 mmol/L L-glutamine and 6 g/L glucose. All cell lines were incubated at 37°C and 5% or 8% CO2 in a humidified incubator (Eppendorf). All media supplements and bioreagents used in this study were purchased from Thermo Fisher Scientific unless otherwise specified. All the cell lines or PDX lines were purchased commercially, authenticated with genetics profiling for polymorphic short tandem repeat analysis at University Genomics Core and confirmed with in-house Mycoplasma test using PCR primers that amplify sequences of 16S rRNA genes. The latest test date of all cell banks with stock vials of 30 to 100 was November 21, 2022. The length of time between cell thaw of the tested cell bank and use in our experiment was 2 to 3 weeks.

Anti-CD276 mAb development, engineering, and production

The peptide cloned from the extracellular domain (Leu29-Pro245) of human CD276 was used to stimulate immune response in BALB/cJ mice (The Jackson Laboratory). As we described previously (26, 27), blood samples were collected from the tail vein to titrate serum concentration of CD276 mAb using ELISA at 14 to 21 days after immunization. Once mAb was detected, splenocytes were harvested and fused with myeloma cells Sp2/0-Ag14 (ATCC) to generate hybridoma, followed by limiting dilution with a seeding density of 1 to 4 cells/well in 96-well plates. The top hybridoma clones with high mAb titer and CD276 binding were sequenced. To minimize the immunogenicity, improve serum stability, and enhance Fc-mediated antibody effector functions, we first engineered the murine antihuman CD276 mAb by constructing a chimeric CD276 mAb, which grafts the complementary-determining region (CDR) with a truncated Fc region of human IgG1. Then, a humanized CD276 mAb was constructed by combining the murine framework regions and three human CDRs.

The transient production system was used to generate chimeric CD276 mAb from Expi293F cells in a 2-L stirred-tank bioreactor (temperature of 37°C, Agt of 140 rpm, DO of 40%, and pH of 7.2) or shaker flask culture (Temp of 37°C, Agt of 130 rpm, and CO2 of 8%). The stable CHO production cells were developed to produce humanized CD276 mAb under similar conditions as above. A liquid chromatography system (Bio-Rad), which is equipped with Bio-Scale Mini UNOsphere SUPrA affinity chromatography cartridges (protein A column, Bio-Rad), was utilized for mAb purification following our established procedure (2632). The mobile phase A of 0.02 mol/L sodium phosphate and 0.02 mol/L sodium citrate (pH 7.5) and phase B (elution buffer) of 0.1 mol/L sodium chloride and 0.02 mol/L sodium citrate (pH 3.0) were used.

Conjugation of CD276-targeting single-payload and DualADCs

Single-payload ADC (CD276 mAb–MMAF)

A bifunctional dibromomaleimide (DBM) linker was used to cross-link mAb interchain cysteines and carry four MMAF drugs. The 5 mmol/L TCEP, dissolved in deionized water at pH 7 and 5 mg/mL mAb in PBS were mixed with a molar ratio of 44:1 and performed at 37°C for 0.5 hours to completely reduce the disulfide bonds in mAb. The 10 mmol/L commercial DBM–MMAF payload was prepared in DMSO, and 7 molar equivalents were added to the completely reduced mAb with 1-hour incubation at room temperature. The crude ADC was purified using a protein A column with a liquid chromatography system. The LC-purified ADC was buffer-exchanged into PBS using a 2 kDa Slide-A-Lyzer Dialysis Cassette and concentrated to a higher concentration with 10 kDa MWCO PES concentrators. The ADC purity, drug–antibody ratio (DAR) and homogeneity were tested using high-performance liquid chromatography (HPLC; Shimadzu) equipped with a MAbPac HIC-Butyl column (5 μm, 4.6 × 100 mm). The mobile phase A of 2 mol/L ammonium sulfate and 100 mmol/L sodium phosphate at pH 7.0 and mobile phase B of 100 mmol/L sodium phosphate at pH 7.0 were used in HPLC analysis. The UV/Vis spectroscopy was used to calculate DAR as an alternative approach. The purified ADC was filtered through a 0.2 μm PES syringe filter (basix) before i.v. injection in mice and was subsequently stored at 4°C for short-term storage.

Single-payload ADC (CD276 mAb–IMQ)

In this conjugation platform, NHS-Azide and NHS-Phosphine reagents (Thermo Fisher Scientific) were used to conjugate mAb and imiquimod (IMQ). Mixture A is created by combining 5 mg/mL of mAb in PBS with 10 mmol/L of NHS-Phosphine linker at a molar ratio of 1:14. In mixture B, 10 mmol/L of NHS-Azide linker and 10 mmol/L of IMQ were combined in 500 μL PBS at a molar ratio of 14:22.4. Mixture A and mixture B were performed at 37°C for 2 hours, respectively. A 10 kDa MWCO PES concentrator was used to purify phosphine-labeled mAb in mixture A from excess NHS-Phosphine. The phosphine-labeled mAb was mixed with 500 μL of mixture B and incubated at 37°C for 2 hours to synthesize mAb–IMQ ADC. The purification and concentration steps are identical to those used for the mAb–MMAF ADC.

DualADC (CD276 mAb–MMAF/IMQ)

The 5 mg/mL of synthesized mAb–MMAF in PBS was mixed with 10 mmol/L of NHS-Phosphine linker at a molar ratio of 1:14, forming mixture A. Mixture B was created by combining 10 mmol/L of NHS-Azide linker and 10 mmol/L of IMQ in 500 μL PBS at a molar ratio of 14:22.4. Mixture A and mixture B were incubated at 37°C for 2 hours each. Mixture A was then applied to 10 kDa MWCO PES concentrators to remove the free linker. Mixture B was added to the phosphine-labeled mAb–MMAF ADC and incubated at 37°C for 2 hours. The crude DualADC was subsequently purified, buffer-exchanged to PBS, and concentrated as described above.

Characterization of ADCs

HPLC characterization

The analysis of purity and the DAR of the ADC was conducted using a HPLC system from Shimadzu, utilizing a MAbPac HIC-Butyl column with dimensions of 5 μm, 4.6 × 100 mm. Two mobile phases were applied in the analysis: mobile phase A, containing 1.5 mol/L ammonium sulfate and 50 mmol/L sodium phosphate at a pH of 7.0, and mobile phase B, containing 50 mmol/L sodium phosphate at a pH of 7.0. The procedure was carried out at a temperature of 25°C and a consistent flow rate of 1.0 mL/minutes.

Matrix-assisted laser desorption/ionization time-of-flight mass spectrometer characterization

Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry was applied to characterize and confirm the structure of our ADCs. Both ADC and mAb samples were diluted with HPLC-grade water to 2 mg/mL. Ten mg of sinapinic acid was dissolved in 1 mL HPLC-grade water containing 50% acetonitrile and 0.1% trifluoroacetic acid; 10 μL of the samples were mixed with 10 μL of the sinapinic acid matrix solution, and then 4 μL of the mixture was spotted onto an MSP 96 target and air-dried. The samples were analyzed using a matrix-assisted laser desorption/ionization time-of-flight mass spectrometer (microflex LRF, Bruker) in a linear positive mode. The data were processed using flexAnalysis software.

Flow cytometry analysis of cell surface binding

The surface binding of our anti-CD276 mAb in TNBC cell lines (MDA-MB-231, MDA-MB-468, and 4T1) was tested by flow cytometry analysis following our reported protocols (2628, 31, 3337). The CD276 mAb was labeled with a fluorescent Alexa Fluor 647 labeling kit (Life Technologies, part of Fisher). One million human or mouse TNBC cells were stained with 5 µg of anti-CD276 mAb-AF647 at 37°C for 60 minutes. After washing three times with PBS, the stained cells were analyzed using a BD LSRII flow cytometer (BD Biosciences), with FlowJo software for data processing. In the analysis of the surface binding rate of our anti-CD276 mAb, gating was set where TNBC cells without mAb staining have <0.5% fluorescent population. In the analysis of TNBC specificity by anti-CD276 mAb, the standard forward and side scatter gating was applied with anti-HER2 mAb as negative control.

Live-cell confocal microscopy

The surface binding and internalization of CD276 mAbs were evaluated using live-cell confocal imaging, following our established protocols (2628, 35, 37). TNBC MDA-MB-468 cells were cultured in 35 mm glass bottom dishes (Cellvis) at a density of 1 × 104 cells per dish in 1.5 mL of medium. To visualize the cytoplasm and nucleus, BacMam GFP Transduction Control (Invitrogen) and NucBlue Live ReadyProbes Reagent (Invitrogen) were used for staining following manufactures protocols, respectively. Next, the Cy5.5-labeled CD276 mAb was added to the cells at a final concentration of 1 μg per mL. Live-cell images were captured at 2 to 12 hours after the addition of mAb using a Nikon A1R-HD25 confocal microscope (Nikon USA).

In vitro cytotoxicity assay

The human and mouse TNBC cells were seeded in 96-well plates with a density of 10,000 cells/well for MDA-MB-468, 1,000 cells/well for MDA-MB-231 and 500 cells/well for 4T1. The free MMAF, mAb–MMAF, and mAb–MMAF/IMQ were added into each well to reach the final dosages of 0 to 300 or 400 nmol/L. Higher concentrations, i.e., 0 to 20 µmol/L, of free IMQ and mAb–IMQ, were tested. The treated cells were incubated at 37°C and 5% CO2 for 5 days, and the cell viability was analyzed using the MTT Cell Proliferation Assay kit (38, 39).

PDX model and in vivo treatment

The CD276+++ TNBC PDX, identified from The Jackson Laboratory PDX lines through transcript analysis and IHC staining, was used following our published procedure (36). Briefly, the PDX tumors were minced into small fragments (1 × 1 × 1 mm3), loaded into a 1-mL sterile syringe connected with a 13G needle (BD), and subcutaneously injected into the rear flank of 5 to 7-week-old NSG female mice with 40 to 50-μL implantation for each mouse. The anti-TNBC efficacy of humanized CD276 mAb–derived single-payload and DualADCs was evaluated in the PDX model when tumor volume reached 25 to 50 mm3 by intravenously injecting saline, 16 mg/kg of mAb–MMAF, or 16 mg/kg mAb–MMAF/IMQ on an injection every 7 days (total of 3 injections; Q7Dx3) schedule (n = 5–7) until tumor volume reached >1,500 mm3.

Cell line–derived xenografts and in vivo treatment

4T1-FLuc–xenografted immunocompetent models

The 4T1-FLuc cells (2 × 106 cells per mouse) were subcutaneously injected into 6-week BALB/cJ female mice. When the average tumor volume reached 20 to 50 mm3, mice were randomized into multiple groups (n = 5, or 7 or 8) and intravenously injected with saline (control), 8 mg/kg of mAb, mAb–IMQ, mAb–MMAF (controls), or 8, 16 or 24 mg/kg of mAb–MMAF/IMQ following schedules of an injection every 7 days (total of 4 injections; Q7Dx4). Tumor size was measured by an electric caliper or in vivo imaging system (IVIS) imaging, and tumor volume was calculated as (length × width2)/2. Tumor and mice body weight were monitored twice a week. Mice were sacrificed when tumor volume in the control group reached >1,000 mm3 or ulceration >2 mm. All tumors, main organs (brain, heart, lung, kidneys, liver, and spleen), whole blood, or serum were collected for further analysis.

MDA-MB-231-FLuc–xenografted immunocompromised models

The 5 × 106 of MDA-MB-231-FLuc cells were subcutaneously injected into 6-week-old NSG (NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ) female mice. When tumor volume reached 50 to 100 mm3, mice were randomized (n = 5) and intravenously administrated with saline (control), 16 mg/kg mAb–MMAF, or 16 mg/kg mAb–MMAF/IMQ following the schedule of Q5Dx5. Tumor volume was monitored with electric caliper or IVIS imaging and mice body weight was measured twice a week. Mice were sacrificed to collect tumors, organs, blood, or serum for posttreatment analysis when tumor volume reached >1,000 mm3 or ulceration was >2 mm.

IVIS

When the tumor volume reached 50 to 100 mm3, 40 µg of engineered anti-CD276 mAbs, which were labeled with cyanine-5.5 fluorescent dye (Lumiprobe), were intravenously injected into the mouse via tail vein. The live-animal images were captured at 24 hours after injection with i.p. injection of FLuc substrate. Then, the mice were sacrificed to harvest major organs, including the brain, heart, lung, spleen, liver, and kidneys, and tumors for ex vivo imaging. The excitation/emission wavelength of 660/710 nm and exposure time of 5 seconds were used in IVIS imaging.

Whole blood analysis

To evaluate the possible peripheral toxicity of our developed anti-CD276 mAb, free TLR7/8 agonist, and ADCs, we performed whole blood analysis using BALB/cJ mice. Specifically, 8 mg/kg mAb, 8 mg/kg IMQ, and 8 to 24 mg/kg single-payload ADC or DualADCs were intravenously injected into each mouse through the tail vein with saline as control. Ten or 21 days after injection, the blood samples were collected via cardiac puncture for whole blood analysis using Hemavet 950FS (Drew Scientific). The blood levels of leukocytes (white blood cell, neutrophil, lymphocyte, monocyte, and eosinophils), erythrocytes (red blood cell, hemoglobin, and mean corpuscular hemoglobin), and thrombocytes (platelet) were measured and analyzed.

Luminex assay

To analyze the tumoral or serum cytokines, a Luminex-based multiplexing assay (Luminex Corporate) was employed. The preconfigured (EPX070-20835-901, EPX260-26088-901) and customized (PPX-03 and PPX-13) chemocytokine assay kits purchased from Thermo Fisher Scientific covering 3, 7, 13, or 26 plex, and all assay reagents were prepared following manufacturer’s procedure. The Luminex assays of tissue or serum samples (n = 3) were performed in 96-well plates provided with the kits, and the raw data of mean fluorescence intensity were read using the Luminex MAGPIX under the xPONENT software.

Western blotting

The TNBC cells were washed three times with cold PBS and lysed using RIPA buffer to extract cellular proteins. The protein concentration was quantified using the bicinchoninic acid method with a Pierce Protein Assay kit (Pierce). Next, 30 µg of proteins were loaded per lane onto a gradient SDS-PAGE, NuPAGE 4% to 12% gradient gel (Invitrogen), along with a protein size marker (Bio-Rad Precision Plus) to separate the proteins by molecular weight. The separated proteins were then transferred onto a polyvinylidene difluoride membrane using a Bio-Rad power supply (Bio-Rad Laboratories) at a constant voltage of 100 V for 90 minutes. After transfer, the polyvinylidene difluoride membrane was blocked with 5% nonfat milk in TBST buffer and agitated at room temperature for 1 hour. For primary antibodies, we used CD276 (dilution 1:1,000, AB134161, Abcam) and β-actin (1:2,000, sc-47778). The membrane was then incubated overnight at 4°C with continuous agitation. On the following day, the primary antibodies were discarded, and the membrane was washed three times with TBST buffer on a shaker, with each wash lasting 5, 5, and 10 minutes, respectively. After the washing steps, horseradish peroxidase–conjugated secondary antibodies (dilution 1:2,000) specific to mouse or rabbit from Cell Signaling Technology in 3% nonfat milk was applied to the membrane accordingly for 1 hour at room temperature. After discarding the secondary antibody, the membrane was washed three times with TBST for 5, 5, and 10 minutes, respectively. Finally, the protein bands were visualized and quantified using an Odyssey Fc imaging system (LI-COR Biosciences). The expression of CD276 in TNBC cells was compared with the internal control, β-actin.

IHC staining

The TNBC (estrogen receptor negative/progesterone receptor negative/HER2) patient tissue microarray (TMA; cat. #BR1303, 126 cores) and 33 human normal organs tissue microarray (cat. #FDA662c) were purchased from US Biomax to detect CD276 expression or the potential nonspecific binding of our humanized CD276 mAb with IHC staining (26, 27). The harvested tumor tissues were either fresh frozen or fixed in 4% formalin. The fixed tissues were dehydrated through graded ethanol solutions, embedded in paraffin blocks, and sectioned with 4 to 5 μm thickness using a microtome and mounted onto glass slides.

During IHC staining, the TMA slide or tumor parafilm sectioned slide was first baked overnight at 60°C, deparaffinized with xylene, and hydrated in ethanol and deionized water. Then, the sectioned tissues were subjected to antigen retrieval in 0.01 mol/L sodium citrate buffer (pH 6.0) for 5 minutes, washed gently in deionized water, and transferred to a solution comprised of 0.05 mol/L Tris, 0.15 mol/L NaCl, and 0.1% Triton-X-100 (TBST, pH 7.6). The endogenous peroxidase was blocked with 3% H2O2 for 15 minutes, and slides were incubated with 5% normal goat serum for 45 minutes to reduce nonspecific background staining. All slides were incubated at 4°C overnight with anti-CD276 antibody (Abcam, rabbit monoclonal, cat. #ab226256, RRID:AB_3069232, 1/500 dilution) or our humanized CD276 mAb. After washing with TBST, the slide was incubated with goat antirabbit secondary antibody conjugated with horseradish peroxidase (Abcam, cat. #ab6721, RRID:AB_856214, 1:1,000). Finally, the stained TMA slide was scanned with Lionheart FX automated microscope (BioTek), and images were processed using Gen5 software. Following our previously established method (26), ImageJ was used to score the CD276 expression in the stained TMA. Briefly, the expression score of CD276 was calculated as (redintensity − blueintensity)/blueintensity × 100 with the definition of high expression with score of >10, medium expression with score of 6 to 10, low expression with score of 3 to 6, and no or minimal expression with score of 0 to 3.

Hematoxylin and eosin staining

The slides of paraffin sectioned organs were dewaxed with xylene and hydrated with gradient ETOH (100%–50%) and dH2O, followed by hematoxylin staining, dipping in 1% HCl and 70% ETOH, immersing in 1% NH4OH for blue color development, and staining with eosin for 30 seconds. Finally, the stained slides were dehydrated in 95% and 100% ethanol and cleared in xylene. The hematoxylin and eosin (H&E)–stained slides were imaged using a Lionheart FX automated microscope (BioTek).

Single-cell RNA sequencing and data analysis

Single-cell sample preparation

The flash-frozen tissue samples (25 mg) were collected and stored at −80°C until processing. Chromium Next GEM RNA Profiling Sample Fixation Kit (PN-1000414, 10× Genomics) was used to fix tissue by mixing 1 mL of fixation buffer with 25 mg of tissue, followed by fine mince and incubation at 4°C for 16 to 24 hours without agitation. After fixation, tissue samples were centrifuged, washed with chilled PBS, and resuspended in 1 mL of tissue resuspension buffer. The fixed tissues supplemented with prewarmed dissociation buffer were dissociated using an Octo Dissociator according to the manufacturer’s instructions. The dissociated tissue samples were filtered through a 30 μm filter, centrifuged, and resuspended in 1 mL of chilled quenching buffer. Cell concentration was determined using an automated cell counter with fluorescent nucleic acid staining. The dissociated tumor samples were stored in 50% glycerol with enhancer at −80°C until single-cell library generation.

Single-cell library construction

A 10× single-cell library was prepared using the Chromium Fixed RNA Kit (cat. #1000497, 10× Genomics). Briefly, ∼16,500 cells were hybridized with probe hybridization to target the polyadenylated RNA and loaded onto Chromium X (PN-1000326, 10× Genomics) to generate barcode-labeled single-cell gel beads in emulsion with a target cell count of ∼10,000. Gel beads in emulsion were then lysed using a recovery agent to release the hybridized RNA. cDNA was synthesized, amplified, and followed by sample index PCR for indexing. The quality control of cDNA was evaluated using an Agilent Bioanalyzer aiming for the final library molecules with P5 and P7 priming sites used in Illumina sequencers. For each cDNA sample, single-cell library construction was started using the 10× barcoded ligated probe products and sequenced using the NovaSeq 6000 flow cell 100-cycle kit (Illumina) in a Read1:i7:i5:Read2 format of 28:10:10:90 bp at 10× Genomics. Demultiplexed fastq files were subsequently utilized for analysis.

Data preprocessing and cell annotation

Three FASTQ files, which stored raw reads information of index reads, forward reads from the paired-end sequencing, and reverse reads from paired-end sequencing, were processed using 10× Genomics Cloud Analysis embedded in Cell Ranger Multi version 7.1.0, and the mouse reference genome mm10 2020-A was used for the alignment. The sequencing quality control was performed using Windows and Linux based FastQC (version 0.12.1). The count-by-gene matrix was analyzed using “Seurat” (version 4.3.0.1, PMID: 34062119) package in R (version 4.3.0). The SCTransform function developed by Christoph Hafemeister and Rahul Satija was used to normalize and scale data. CellMarker 2.0, PanglaoDB, and Tabula Muris database were used to annotate different cell types.

Differential expression genes and Gene Ontology pathway enrichment

Differential expression genes were identified between groups among all the cell types and the integrated cells using Seurat built-in function PrepSCTFindMarkers and FindMarkers. All the tests were operated based on Wilcoxon sum rank test. Notably, the parameters of min.pct and logfc.threshold were set to 0 for further gene set enrichment analysis pathway enrichment. The gseGO function embedded in clusterProfiler package (version 4.8.2) was used to operate gene set enrichment analysis pathway enrichment with the adjusted P value of 0.05, sourced from the Gene Ontology database (https://geneontology.org/).

Pharmacokinetics

Five doses of CD276 mAb–MMAF/IMQ, including 4, 8, 16, 24, and 32 mg/kg, were intravenously injected to 4T1 xenografted BALB/cJ mice. About 10 to 15 μL of blood samples were collected from tail vein or submandibular or submental route at 0.5, 2, 8, 24, 48, 72, 120 hours after injection. The supernatants were used to titrate DualADC using ELISA with human CD276 receptor and anti-MMAF antibody (Thermo Fisher Scientific, Cat. # PIMA542537, RRID:AB_2687990). The reported PK model was used to calculate the important parameter: clearance (CL) = DF2aAUC=Vdke, volume of distribution (Vd) = CL(t2-t1)lnC1-lnC2, half life t1/2 = 0.693VdCL, recommended dose (D) = Cmax.desiredkeVdT1-e-keτ1-ekeT, and dosing interval (τ) = lnCmax.desired-lnCmin.desiredke+T.

Statistical analysis

All experimental data were presented as mean ± SEM. The significance of difference among groups was analyzed using two-tailed t test, and the difference between the two groups was analyzed using one-way ANOVA followed by post hoc (Dunnett) analysis. Statistical analysis was performed using GraphPad Prism, and P value of <0.05 was used for all tests.

Data availability

The single-cell RNA sequencing (scRNA-seq) data generated in this study are publicly available in Gene Expression Omnibus at accession number GSE274871 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?&acc=GSE274871). The transcript data of surface receptor CD276 in breast cancer analyzed in this study were obtained from Breast Cancer data of TCGA database at https://portal.gdc.cancer.gov/. All other raw data generated in this study are available upon request from the corresponding author.

Overview of the concept of DualADC

Our CD276-targeted DualADC was designed as chemoimmunotherapy. The engineered CD276 mAb carrying potent molecule MMAF, or immune boosting reagent TLR7/8 agonist IMQ, or both payloads, named as CD276 mAb–MMAF/IMQ in this study, targets the overexpressed surface receptor CD276 on TNBCs. After internalization via receptor-mediated endocytosis and lysosomal degradation, the payloads MMAF and IMQ are released into the cytoplasm. The intracellular MMAF leads to direct cancer cell death by blocking microtubulin polymerization, and IMQ inhibits cell proliferation via signaling pathway MyD88/IRF7 after targeting TLR7/8 in endosome as previously reported (17). The IMQ (free drug released from TNBC cell death or drug conjugated in ADCs) upregulates the cytokines’ secretion by immune cells in the TME. Additionally, the inhibition of CD276 immune checkpoint by mAb reactivates NK and T cells in the tumor.

CD276 expression in TNBC patient tissues and subtypes

The expression of CD276 receptor was first evaluated using TNBC patient TMA (n = 126) with IHC staining (Fig. 1A). The receptor expression analysis revealed that 17 of 126 TNBC tissues (14%) had high level of CD276 expression with a score of >10, 58 (46%) samples had medium expression with a score of 6 to 10, 38 (30%) samples had low expression with a score of 3 to 6, and 13 (10%) samples had no or minimal expression with a score of <3. A total of 60% of TNBC patient tissues had high or medium CD276 expression. The representative IHC images of TMA cores with different levels of CD276 are described in Fig. 1B. The normal breast tissues (n = 2; control) showed minimal expression of CD276, and the microarray of 33 normal human organs or tissues that were stained with polyclonal antibody showed low CD276 expression in most tissues including the normal breast tissues (Supplementary Fig. S1A). One of the two adrenal glands, heart, and kidney showed positive staining (probably false-positive), but the Human Protein Atlas database does not report high CD276 expression in these organs. Then the CD276 expressions in TNBC cell lines representing various subtypes, including nontumorigenic breast epithelial lines 184B5 (control), BT549 (LAR), MDA-MB-468 (basal-like 1 and BL2), BT-20 (LAR and BL2), MDA-MB-231 (mesenchymal stem–like, BL2, and LAR), and 4T1 (mouse TNBC, similar to human subtype BL), were tested using Western blotting (Supplementary Fig. S1B). Most of the human TNBC subtypes have high CD276 expression, mouse TNBC has medium expression, and normal breast tissue has low or minimal expression. TCGA dataset also showed higher CD276 mRNA levels in breast cancers than normal breasts. These data indicated that CD276 receptor is an ideal target for developing anti-TNBC therapies.

Figure 1.

Evaluations of CD276 and characterization of humanized CD276 mAb in TNBCs. A, IHC staining of TMA with anti-CD276 antibody. B, Representative images of low, medium, and high CD276 expressions. C, TNBC surface binding by engineered CD276 mAbs. D, TNBC MDA-MB-468 cell surface binding and internalization of humanized CD276 mAb (Hu276 mAb) labeled with Cy5.5 fluorescent dye (red). E, Live-animal and ex vivo IVIS to confirm TNBC-specific targeting by Hu276 mAb-Cy5.5 at 24 hours after tail vein injection. F, IHC staining of 33 human normal organs with our humanized CD276 mAb.

Figure 1.

Evaluations of CD276 and characterization of humanized CD276 mAb in TNBCs. A, IHC staining of TMA with anti-CD276 antibody. B, Representative images of low, medium, and high CD276 expressions. C, TNBC surface binding by engineered CD276 mAbs. D, TNBC MDA-MB-468 cell surface binding and internalization of humanized CD276 mAb (Hu276 mAb) labeled with Cy5.5 fluorescent dye (red). E, Live-animal and ex vivo IVIS to confirm TNBC-specific targeting by Hu276 mAb-Cy5.5 at 24 hours after tail vein injection. F, IHC staining of 33 human normal organs with our humanized CD276 mAb.

Close modal

Anti-CD276 mAb development, engineering, and production

The extracellular domain (Leu29-Pro245) of human CD276 (UniProtKB Q5ZPR3/H0YL10) with 93% similarity to mouse CD276 (UniProtKB Q8VE98/A0A8C5XZR2) was synthesized and used as an immunogen to develop anti-CD276 mAb. The first-generation murine antihuman CD276 mAb was developed using hybridoma technology. The top hybridoma clone with high CD276 mAb expression and surface receptor binding was screened using ELISA with extracellular peptide as the coating antigen (Supplementary Fig. S2A). The isotype analysis revealed that the developed CD276 mAb was IgG2b/kappa. To minimize immunogenicity, improve serum stability, and enhance Fc-mediated antibody effector functions, we first engineered the murine antihuman CD276 mAb (Supplementary Fig. S2B, left) by constructing a chimeric CD276 mAb, which grafts the CDR (red and green) with a truncated Fc region (blue) of human IgG1 (Supplementary Fig. S2B, middle). Then, a humanized CD276 mAb was further developed by combining the murine framework regions (red and green) and three human CDRs (yellow; Supplementary Fig. S2B, right). The CHO cells produced humanized CD276 mAb with final titers of 20 to 40 mg/L (Supplementary Fig. S2C).

In vitro TNBC cell surface binding

The TNBC surface binding by engineered CD276 mAbs (chimeric and humanized) was analyzed in cell lines MDA-MB-231, MDA-MB-468, and 4T1 by flow cytometry (Fig. 1C). The human and mouse CD276 surface receptors have the same topology including one extracellular, one helical transmembrane, and one cytoplasmic domain. Protein BLAST analysis showed that human and mouse CD276 receptors have a similarity of 93% in the targeted extracellular domain. Flow cytometry showed that the binding rates of chimeric mAb in MDA-MB-231, MDA-MB-468, and 4T1 cells were 98.9%, 99.9%, and 72.8%, respectively (Fig. 1C). The humanized CD276 mAb also had high surface binding to MDA-MB-231 (99.9%) and MDA-MB-468 (99.8%) and similar medium binding to 4T1 cells (71.2%) as chimeric CD276 mAb. These data showed that our CD276 mAbs can efficiently bind human TNBCs while having relatively lower binding to mouse TNBC. The mouse TNBC 4T1 cell line represents stage IV human breast cancer with high metastasis and invasion and has high similarity to the subtype basal-like and immune suppressed of human TNBC (40). Therefore, the synergism of potent payload, immune regulating TLR7/8 agonist, and immune reactivating CD276 mAb was evaluated in TNBC xenograft immunocompetent mouse models using the engineered CD276 mAb with cross activity.

The cell surface binding by our CD276 mAbs was compared between human TNBC lines and normal breast line with HER2 mAb as negative control by flow cytometry (Supplementary Fig. S2D). The results showed that TNBC binding of the humanized mAb to TNBC MDA-MB-468 was significantly (69–89 folds) higher than that to the normal breast epithelium 184B5 line. Furthermore, the live-cell confocal microscopy imaging demonstrated that the humanized anti-CD276 mAb-Cy5.5 (Fig. 1D, red) and chimeric anti-CD276 mAb (Supplementary Fig. S3A) bound to the cell surface of TNBC MDA-MB-231 or -468 (green, BacMam GFP expressed in cytoplasm). The CD276 mAbs effectively internalized into TNBC cells via receptor-mediated endocytosis after forming the complex. These results demonstrated that our CD276 mAbs can effectively target TNBC cells and release payload intracellularly.

In vivo TNBC targeting and minimal binding to human normal organs

Both MDA-MB-231-FLuc xenografted NSG mice and 4T1-FLuc xenografted BALB/cJ mice were used to evaluate the TNBC-specific targeting and biodistribution of the engineered CD276 mAbs-Cy5.5 with live-animal and ex vivo IVIS imaging. The tumor bioluminescent signal (FLuc) was colocalized with the mAb fluorescent signal (Cy5.5) within 24 hours after tail vein injection (Fig. 1E; Supplementary Fig. S3B), indicating the in vivo TNBC specificity of our mAbs. The ex vivo images showed strong fluorescent signals in tumors whereas no or undetectable signals in the major organs such as the brain, heart, lung, spleen, liver, and kidneys (Fig. 1E; Supplementary Fig. S3C). These IVIS images indicated that the anti-CD276 mAbs were able to specifically bind and accumulate in both human and mouse TNBC tumors and effectively deliver payloads. Furthermore, the 33 normal human tissues (n = 2) stained with our humanized CD276 mAb did not detect obvious positive staining except for weak signal in a few tissues such as breast tissue (Fig. 1F). This result further confirmed that our humanized CD276 mAb can target the CD276 overexpressing TNBC with minimal off-target to normal organs.

Development of DualADC

A new sequential conjugation procedure was developed to construct DualADC by linking mAb with MMAF using a re-bridging DBM linker (26, 27, 31, 35, 37), followed with IMQ conjugation using phosphine-azide linker designed in this study (Fig. 2A and B). HPLC characterizations showed the successful conjugations of both single payload (IMQ or MMAF) and double payloads (IMQ and MMAF; Fig. 2C). Mass spectrometry evaluation confirmed the right structures of mAb, mAb–MMAF, mAb–IMQ, and mAb–MMAF/IMQ with the expected MZ values (Fig. 2D). The integrity of mAb and ADCs was confirmed in SDS-PAGE (Fig. 2E). The site-specific conjugation of MMAF had a homogenous structure, purity of 95% to 100%, and DAR of 3 to 4, whereas random conjugation of IMQ showed a heterologous structure and DAR of 7 to 14. We took advantage of random conjugation to optimize the condition and DAR (812) of IMQ to achieve high solubility and stability and tumoral immunity.

Figure 2.

Construction of DualADC via cysteine and lysine. A, Structure of DualADC CD276 mAb–MMAF/IMQ. B, Structure of DualADC. C, HPLC confirmed the conjugation of single payload and dual payloads with chimeric anti-CD276 mAb. D, Matrix-assisted laser desorption/ionization time-of-flight mass spectrometer to validate the molecular weight of mAb, single ADC (mAb–MMAF, mAb–IMQ), and DualADC (mAb–MMAF/IMQ). E, SDS-PAGE of ADCs. M, marker; 1, CD276 mAb; 2, mAb–MMAF; 3, mAb–IMQ; 4, mAb–MMAF/IMQ.

Figure 2.

Construction of DualADC via cysteine and lysine. A, Structure of DualADC CD276 mAb–MMAF/IMQ. B, Structure of DualADC. C, HPLC confirmed the conjugation of single payload and dual payloads with chimeric anti-CD276 mAb. D, Matrix-assisted laser desorption/ionization time-of-flight mass spectrometer to validate the molecular weight of mAb, single ADC (mAb–MMAF, mAb–IMQ), and DualADC (mAb–MMAF/IMQ). E, SDS-PAGE of ADCs. M, marker; 1, CD276 mAb; 2, mAb–MMAF; 3, mAb–IMQ; 4, mAb–MMAF/IMQ.

Close modal

In vitro anti-TNBC cytotoxicity of ADCs

The cytotoxicity of free MMAF and TLR7/8 agonists (control), CD276 mAb–MMAF and mAb–IMQ (controls), and dual-payload mAb–MMAF/IMQ was tested using MDA-MB-231, MDA-MB-468, and 4T1 cells. MMAF exhibited IC50 values of 151.0, 143.0, and 103.7 nmol/L for these three cell lines, respectively (Fig. 3A). IMQ with high TNBC cytotoxicity was identified from the tested TLR7/8 agonists, including two IMQs, AXC715 and R848 (Supplementary Fig. S4A–S4D). The IMQ1 showing IC50 values of 10.4, 6.2, and 11.4 μmol/L in three TNBC lines (Fig. 3B) was used in single-payload ADCs and DualADCs. IMQ had EC50 values of 24.5 and 6.7 μmol/L in TLR8 overexpressing human and murine HEK 293 cells (Fig. 3C). The IC50 values of humanized CD276 mAb (Hu276 mAb)–MMAF were 175.0 nmol/L (MDA-MB-231), 50.7 nmol/L (MDA-MB-468), and 125.4 nmol/L (4T1) as described in Fig. 3D. Similar to free IMQ, Hu276 mAb–IMQ (Fig. 3E) revealed IC50 values of 7.7 µmol/L (MDA-MB-231), 6.9 μmol/L (MDA-MB-468), and 13.1 μmol/L (4T1). The DualADC Hu276 mAb–MMAF/IMQ (Fig. 3F), with IC50 values of 18.8 nmol/L (MDA-MB-231), 16.9 nmol/L (MDA-MB-468), and 40.8 nmol/L (4T1), had the highest and synergistic cytotoxicity or potency to all TNBC cells as compared with free drugs and single-payload ADCs.

Figure 3.

In vitro evaluations of DualADC using humanized CD276 mAb. The TNBC MDA-MB-231, MDA-MB-468, and 4T1 cells were used in cytotoxicity studies with free drugs and single-payload ADC as controls. A, Anti-TNBC cytotoxicity and IC50 of free DM1 and MMAF drugs. B, Cytotoxicity and IC50 of IMQ. C, EC50 of IMQ on human and murine TLR8+ HEK cells. D, Anti-TNBC cytotoxicity and IC50 of single-payload ADC (mAb–MMAF). E, Cytotoxicity and IC50 of single-payload (mAb–IMQ). F, Cytotoxicity and IC50 of DualADC mAb–MMAF/IMQ. n = 3.

Figure 3.

In vitro evaluations of DualADC using humanized CD276 mAb. The TNBC MDA-MB-231, MDA-MB-468, and 4T1 cells were used in cytotoxicity studies with free drugs and single-payload ADC as controls. A, Anti-TNBC cytotoxicity and IC50 of free DM1 and MMAF drugs. B, Cytotoxicity and IC50 of IMQ. C, EC50 of IMQ on human and murine TLR8+ HEK cells. D, Anti-TNBC cytotoxicity and IC50 of single-payload ADC (mAb–MMAF). E, Cytotoxicity and IC50 of single-payload (mAb–IMQ). F, Cytotoxicity and IC50 of DualADC mAb–MMAF/IMQ. n = 3.

Close modal

In vivo anti-TNBC efficacy of DualADC in TNBC PDX models

The PDX models capitulating tumor heterogeneity and TME are crucial to fully evaluate the newly developed targeted therapy. As shown in Fig. 4A, PDX tumor volume without treatment (saline group) reached ∼1,700 mm3 on day 29 after treatment. Both Hu276 mAb–MMAF and Hu276 mAb–MMAF/IMQ completely inhibited tumor growth with a final volume of ∼0 mm3 on day 7, and no recurrence was observed after stopping treatment during days 15 to 29 (Fig. 4A and C). The body weight profiles showed no difference between ADCs and saline groups (Fig. 4B). The IHC staining of TNBC PDX without treatment confirmed the positive expression of CD276 (Fig. 4D). These data indicated that our humanized CD276 mAb-directed ADCs can effectively target and treat TNBC PDX.

Figure 4.

Antitumor efficacy of DualADC in the TNBC PDX xenograft model. A, Tumor volume changes after treatment with humanized CD276 mAb–derived ADC following the schedule of Q7Dx3 as indicated by the black arrow. Data were presented as mean ± SEM, n = 5–7. Saline (), 16 mg/kg of mAb–MMAF (▲), and 16 mg/kg of mAb–MMAF/IMQ (●). *, P < 0.05 vs. saline using ANOVA followed by Dunnett t test. B, Body weight. C, White light images at 14 days after treatment stopped. D, IHC staining of TNBC PDX tumor tissue. Scale bar, 20 μm.

Figure 4.

Antitumor efficacy of DualADC in the TNBC PDX xenograft model. A, Tumor volume changes after treatment with humanized CD276 mAb–derived ADC following the schedule of Q7Dx3 as indicated by the black arrow. Data were presented as mean ± SEM, n = 5–7. Saline (), 16 mg/kg of mAb–MMAF (▲), and 16 mg/kg of mAb–MMAF/IMQ (●). *, P < 0.05 vs. saline using ANOVA followed by Dunnett t test. B, Body weight. C, White light images at 14 days after treatment stopped. D, IHC staining of TNBC PDX tumor tissue. Scale bar, 20 μm.

Close modal

In vivo anti-TNBC efficacy of DualADC in TNBC immunocompetent models

The synergistic chemoimmunotherapy of DualADC was further evaluated in 4T1-FLuc xenografted female BALB/cJ mouse models (n = 5–8). It is found that 16 and 24 mg/kg of CD276 mAb–MMAF/IMQ reduced tumor burden to 142 mm3 25 days after the first injection, as compared with the saline group with final tumor volume of 642 mm3 (Fig. 5A). The 8 mg/kg of mAb–IMQ, mAb–MMAF, and mAb–MMAF/IMQ showed the medium treatment effect with final tumor volume of 241 to 376 mm3. Body weight had no obvious difference among the DualADC, single ADC, and saline groups (Supplementary Fig. S5A). A further evaluation of single and dual ADCs at dose of 8 mg/kg using 4T1 xenograft (n = 4) showed that the relative tumor volumes were 58.39% (mAb–MMAF), 64.68% (mAb–IMQ), and 43.62% (mAb–MMAF/IMQ) as compared with the saline group, validating the synergistic effects of DualADC (Supplementary Fig. S5B). The two engineered mAb (chimeric and humanized)–directed DualADCs (16 mg/kg) had similar anti-TNBC efficacy (Fig. 5B). Further pathology assessment with H&E staining of major organs (brain, heart, liver, kidney, lung, and spleen) revealed no inflammation, apoptosis or necrosis, damage, or toxicity in the 24 mg/kg DualADC group (Fig. 5C), although the enlarged spleen was observed at the endpoint.

Figure 5.

Anti-TNBC efficacy of 276 mAb–MMAF/IMQ in immunocompetent models. The mouse TNBC 4T1-FLuc xenografted female BALB/cJ mice were treated with DualADC (8, 16, 24 mg/kg mAb–MMAF/IMQ), mAb–MMAF, mAb–IMQ, or saline (controls) via i.v. injection through tail vein. n = 5–8. A, Tumor volume after treatment following the schedule of Q7Dx4 as indicated by black arrow. Data were presented as mean ± SEM. *, P < 0.05 vs. saline using ANOVA followed by Dunnett t test. B, Weight of terminal wet tumors treated with 16 mg/kg of DualADCs using chimeric CD276 mAb and humanized 276 mAb. *, P < 0.05 vs. saline using ANOVA followed by Dunnett t test. C, H&E staining of major organs. Scale bar, 70 μm. D, IHC staining of harvested tumors using markers of cell proliferation (Ki67), apoptosis (CCasp3), immune checkpoint inhibition (PD-1), and filtration and activation of CD8+ T, NK and macrophage cells (CD8, CD45, F4/80). Scale bar, 20 μm. E, H&E staining to analyze TNBC cell death in the treatment group. Scale bar, 40 μm.

Figure 5.

Anti-TNBC efficacy of 276 mAb–MMAF/IMQ in immunocompetent models. The mouse TNBC 4T1-FLuc xenografted female BALB/cJ mice were treated with DualADC (8, 16, 24 mg/kg mAb–MMAF/IMQ), mAb–MMAF, mAb–IMQ, or saline (controls) via i.v. injection through tail vein. n = 5–8. A, Tumor volume after treatment following the schedule of Q7Dx4 as indicated by black arrow. Data were presented as mean ± SEM. *, P < 0.05 vs. saline using ANOVA followed by Dunnett t test. B, Weight of terminal wet tumors treated with 16 mg/kg of DualADCs using chimeric CD276 mAb and humanized 276 mAb. *, P < 0.05 vs. saline using ANOVA followed by Dunnett t test. C, H&E staining of major organs. Scale bar, 70 μm. D, IHC staining of harvested tumors using markers of cell proliferation (Ki67), apoptosis (CCasp3), immune checkpoint inhibition (PD-1), and filtration and activation of CD8+ T, NK and macrophage cells (CD8, CD45, F4/80). Scale bar, 20 μm. E, H&E staining to analyze TNBC cell death in the treatment group. Scale bar, 40 μm.

Close modal

Anticancer mechanisms of DualADC

The underlying anticancer mechanisms were investigated with multiple research approaches, including H&E staining for tumor cell death, IHC staining with various antibodies to analyze tumoral immunity, Luminex assay to titrate tumoral cytokine secretion and test cytokine storm, whole blood analysis, and scRNA-seq to analyze immune cell infiltration, immune responses, and mitotic activities in the TME. All our data suggested a combined chemoimmunotherapy of DualADC for TNBC treatment.

First, IHC staining of tumor tissues demonstrated obvious infiltration of activated cytotoxic CD8 T cells (CD8 and CD45), activated NK cells (CD45), and phagocytosis of macrophage (F4/80) in the 24 mg/kg DualADC group (Fig. 5D). DualADC also reduced tumor intensity and slightly downregulated PD-1 expression. The expression of the proliferation marker (Ki67) was downregulated, and the apoptosis marker (CCasp3) was upregulated significantly in TNBC tumor treated with DualADC. These data indicated that our CD276-targeted DualADC mAb–MMAF/IMQ effectively modulated TNBC tumoral immunity. The H&E staining of tumor tissues showed healthy TNBC cells in the saline group, a certain level of cell death in single-payload ADC (mAb–MMAF and mAb–IMQ) groups, and severe cell death in DualADC (mAb–MMAF/IMQ) group (Fig. 5E).

Second, Luminex assay of tumor tissues harvested at the end of treatment identified several tumoral cytokines, such as TFNγ, TNFα, and IL6, with obvious enhancement, which confirmed the targeted delivery of IMQ by mAb and its immunotherapy in TNBC tumor (Fig. 6A). Additional tumoral cytokines (TFNγ, TNFα, IL6, IL10, IL2, IL4, and MCP-1) profiles of TNBC from the DualADC group are summarized in Supplementary Fig. S6. These results confirmed that TLR7/8 agonist boosted cytokine secretion in the TME. About 4.2 to 6.0 μg of IMQ was detected in the TME (NOT cell lysis) of 1 mg of tumor samples at the end of the animal study, but an advanced analysis of the drug’s dynamic distribution in tumor is needed in future studies.

Figure 6.

Analysis of tumoral cytokines and general toxicity after treatment of DualADC. The same mice as in Fig. 5 were used here. A, Luminex assay identified several enhanced cytokines and downregulated PD-1 in the TME. B, The complete blood cell counts. 1, Saline (control); 2, 8 mg/kg mAb–MMAF (control); 3, 8 mg/kg mAb–IMQ (control); 4, 8 mg/kg mAb–MMAF/IMQ; 5, 16 mg/kg mAb–MMAF/IMQ; 6, 24 mg/kg mAb–MMAF/IMQ. Data were presented as mean ± SEM, n = 4. *, P < 0.05 vs. saline using ANOVA followed by Dunnett t test. BA, basophil; EO, eosinophil; Hb, hemoglobin; HCT, hematocrit; LY, lymphocyte; MCH, hexachlorocyclohexane; MCHC, mean corpuscular hemoglobin concentration; MO, monocytes; MCV, mean corpuscular volume; MPV, mean platelet volume; NE, neutrophils; PDW, platelet distribution width; PTL, platelet; PCT, procalcitonin; RBC, red blood cell; RDW, red cell distribution width; WBC, white blood cell.

Figure 6.

Analysis of tumoral cytokines and general toxicity after treatment of DualADC. The same mice as in Fig. 5 were used here. A, Luminex assay identified several enhanced cytokines and downregulated PD-1 in the TME. B, The complete blood cell counts. 1, Saline (control); 2, 8 mg/kg mAb–MMAF (control); 3, 8 mg/kg mAb–IMQ (control); 4, 8 mg/kg mAb–MMAF/IMQ; 5, 16 mg/kg mAb–MMAF/IMQ; 6, 24 mg/kg mAb–MMAF/IMQ. Data were presented as mean ± SEM, n = 4. *, P < 0.05 vs. saline using ANOVA followed by Dunnett t test. BA, basophil; EO, eosinophil; Hb, hemoglobin; HCT, hematocrit; LY, lymphocyte; MCH, hexachlorocyclohexane; MCHC, mean corpuscular hemoglobin concentration; MO, monocytes; MCV, mean corpuscular volume; MPV, mean platelet volume; NE, neutrophils; PDW, platelet distribution width; PTL, platelet; PCT, procalcitonin; RBC, red blood cell; RDW, red cell distribution width; WBC, white blood cell.

Close modal

Third, complete blood count  analysis of whole blood samples showed neither mAb–MMAF and mAb–IMQ nor mAb–MMAF/IMQ significantly changed erythrocytes (red blood cell, hemoglobin, and hematocrit) and thrombocyte (platelet and plateletcrit), as summarized in Fig. 6B. The leukocyte cell counts (white blood cell, neutrophil, lymphocyte, monocyte, eosinophils, and basophile) of mice treated with 24 mg/kg of DualADC were higher than those in mice treated with 16 mg/kg of DualADC. There was no anemia or blood clot observed during the treatment with ADCs. The chemocytokine analysis of serum samples demonstrated that most of the titrated 25 chemocytokines had no obvious differences among all groups except an increase of CCL4 in the mice treated with 16 and 24 mg/kg of DualADC and CXCL1 in 24 mg/kg of the DualADC group (Supplementary Table S1). Moreover, 8 mg/kg of CD276 mAb, 8 mg/kg of free TLR7/8 agonist IMQ, or saline did not obviously change the cell counts of leukocytes, erythrocytes, and thrombocytes (Supplementary Fig. S7A and S7B). All these data indicated the safety and minimal systemic toxicity of the CD276-targeted mAb, single-payload ADCs, and DualADC with doses of 8 to 24 mg/kg.

Finally, scRNA-seq of TNBC identified and quantitated all cell types in tumor tissues, including tumor cells, myoepithelial cells, endothelial cells, fibroblasts, stromal cells, CD8+ T cells, dendritic cells, neutrophils, macrophages, and leukocytes (Fig. 7A; Supplementary Table S2). We found that the percentage of immune cells infiltrated in the TME of 16 mg/kg DualADC group was 43.1%, much higher than 15.3% in the saline group. These data were consistent with the IHC staining, as presented in Fig. 5D, but provided an accurate count of the immune cells. The counts and gene ratio of tumoral cells involved in various immune functions, such as spindle midzone assembly, mast cell activation, adaptive immune response, immune response–regulating cell surface receptor signaling pathway, activation of immune response, and immunity mediated by B cells, immunoglobulin, leukocyte, myeloid leukocyte, and lymphocyte, are summarized in Fig. 7B. The immune regulation functions of macrophages (Fig. 7C), neutrophils, CD8+ T cells, dendritic cells, leukocytes, and other cells were observed in the tumor. The analysis of tumor cell distribution in different mitotic stages confirmed the MMAF caused inhibition of cell proliferation (Fig. 7D).

Figure 7.

Analysis of immune cell infiltration and immune functions in the TME using scRNA-seq. The tumor tissues were harvested from the same animal study in Fig. 5. A, Overview of all cell types in TNBC tumors. B, Immune functions in the TME. C, Immune responses of macrophage. D, Analysis of mitotic activities. The adjusted P value is 0.05. UMAP, Uniform Manifold Approximation and Projection.

Figure 7.

Analysis of immune cell infiltration and immune functions in the TME using scRNA-seq. The tumor tissues were harvested from the same animal study in Fig. 5. A, Overview of all cell types in TNBC tumors. B, Immune functions in the TME. C, Immune responses of macrophage. D, Analysis of mitotic activities. The adjusted P value is 0.05. UMAP, Uniform Manifold Approximation and Projection.

Close modal

Validation of anti-TNBC efficacy of DualADC in immunocompromised models

Anti-TNBC efficacy of 276 mAb–MMAF/IMQ was validated in immunocompromised models using MDA-MB-231-FLuc xenografted female NSG mice. Supplementary Figure S8A shows that TNBC tumor volume was reduced to 15 to 17 mm3 on day 22 in 16 mg/kg of ADC treatment groups and 660 mm3 in the saline group. No TNBC recurrence was observed 14 days after treatment was stopped. The change in body weight had no obvious difference among the treatment and saline groups (Supplementary Fig. S8B). The endpoint IVIS imaging [Supplemenary Fig. S8C (left)] and white light imaging [Supplemenary Fig. S8C (right)] validated the high anti-TNBC efficacy of CD276-targeted ADC, i.e., reduced tumor volume and burden.

Pharmacokinetics

The PK parameters were assessed using 4T1 xenografted BALB/cJ models. The serum concentration profiles of CD276 mAb–MMAF/IMQ are presented in Supplementary Fig. S9. The calculated PK values include t1/2 = 1.21 to 2.99 days, Cmax = 21.39 to 93.37 μg/mL, D = 6.48 to 16.66 mg/kg, and τ = 5.13 to 7.10 days. None of the tested DualADC doses apparently affected mouse body weight, survival, and general health, including water intake, breathing, and locomotion, although maximal toxicity dosage was not reached.

Targeted therapies, such as mAb, ADC, and small-molecule inhibitors, have been developed to treat solid tumors, but none has been offered to treat primary and metastatic TNBCs due to a lack of promising targets. This study confirmed the overexpression of CD276 receptor in the majority (more than 60%) of TNBC patient tissues and multiple TNBC subtypes, consistent with TCGA transcript analysis and literature report of CD276 in 80% of breast cancers (10). Moreover, CD276 inhibits the immune functions of NK and T cells and reduces the secretion of effector cytokines as an immune checkpoint. This study developed an innovative CD276-targeted therapy by establishing a new platform to conjugate dual payloads with mAb, aiming to eliminate TNBC cells in vivo through synergistic anticancer mechanisms. The cross activity of CD276 mAb allowed us to evaluate the anti-TNBC efficacy and mechanism of DualADCs in mouse models. This therapy had high specificity to CD276+ tumor, effective tumor cell death, obvious immune cell infiltration, and cytokine secretion in the TEM. The tumor burden in all animal studies was significantly reduced, highlighting its great potential as a targeted therapy for TNBC treatment.

Different from traditional ADC using mAb to carry a single payload, our DualADC is comprised of CD276 mAb, a highly potent drug (MMAF), and an immune booster TLR agonist (IMQ). The concept of using DualADC to target and treat TNBCs (and other cancers) is innovative due to several advantages. First, DualADC combines different drugs with synergistic cancer therapeutic mechanisms, upregulating tumoral immunity while simultaneously inducing direct cancer cell death. The integration of chemotherapy and immunotherapy in one molecule could improve circulation stability, reduce side effects, and improve anticancer efficacy, especially for highly aggressive and heterogeneous cancers. Second, the established DualADC platform enables conjugating two different drugs at two sites (i.e., cysteine and lysine). Different from site-selective conjugation or attaching two payloads together, our conjugation strategy enables us to optimize the ratio of two payloads. The cancer cells and immune cells might have different responses to the dual payloads, so the flexibility to adjust the ratio between two payloads could achieve optimal anticancer efficacy. In the future, we need to evaluate the effects of conjugation strategies, linkers, combination of different payloads, and the ratio of two payloads on circulation stability, potential toxicity, and cancer treatment efficacy. Third, targeting CD276 could cover 60% of patients with TNBC (and other CD276+ cancers), improve therapeutic efficacy by targeted delivery of drugs to TME, and reduce dose and dosage. Fourth, the engineered anti-CD276 mAbs could improve its plasma stability, biological function, and thereby translational potential, which need further evaluations in advanced animal models.

Importantly, the investigation of anticancer mechanisms of our DualADC using scRNA-seq, Luminex assay, histology analysis, whole blood analysis, and others identified multiple anti-TNBC mechanisms (as detailed below): (i) direct cancer cell killing and proliferation inhibition through potent payload, (ii) combined tumoral immunity by CD276 mAb and TLR7/8 agonist, and (iii) TLR agonist–mediated tumoral cytokine enhancement.

First of all, our anti-CD276 mAb can effectively and specifically target the TNBC xenografts with minimal off-target and deliver payloads in vivo. Moreover, the clinical data and literature reports reported that blockade of CD276 can neutralize the inhibitory signaling, reactivate immune cells, and restore effector immune functions. This study showed that anti-CD276 mAb reactivated the immune functions in TME by increasing the infiltration of activated NK and T cells. In addition to treating cancers as monotherapy, the combination of CD276 mAb with other therapies, such as enoblituzumab/retifanlimab and vobramitamab duocarmazine, shows great potential in preclinical or clinical studies (12). We will further investigate the dually targeting CD276 and PD-1 with our unique ADCs in future studies to benefit the most (75%–90%) patients with TNBC without response to immune checkpoint blockers due to primary resistance, acquired resistance, and relapse during treatment (41).

Pattern recognition receptor agonists, such as TLR agonists, have been developed to treat cancer and other diseases, which target innate immune systems and stimulate immune responses via the MYD88 and other pathways (15). Pattern recognition receptor agonists, including TLR7/8/9 agonists, RIG-I, MDA-5, and STING (19), have been investigated in preclinical studies or clinical trials. Administration of free TLR agonists, which lack tumor selectivity, may cause fatal cytokine storm and other side effects (42). The conjugation linkers and strategies developed in this study enable delivering TLR agonists to tumors directly, overcoming the challenge of systematic toxicity. Consistent with the literature, the present study shows that TLR 7/8 agonists, targeting delivered with our anti-CD276 mAb, inhibit TNBC cancer proliferation (17), induce tumor cell apoptosis (19), and stimulate the production of multiple cytokines by immune cells that are activated in the TME (17). In addition, the tumor-associated antigen released by dead tumor cells can lead to a cascade of adaptive immune responses through antibody-dependent cellular phagocytosis (43).

In addition to immunotherapy functions, both potent payload (MMAF) and immune boosting reagent (IMQ) showed cytotoxicity to TNBC cells. It is well known that MMAF can effectively block microtubulin polymerization and inhibit the mitotic process and proliferation of TNBC cells. This study confirmed the cytotoxicity and direct cell death of free MMAF and single-payload ADC (mAb–MMAF). In addition to upregulating immune function, the TLR7/8 agonist and single-payload ADC (mAb–IMQ) also showed cytotoxicity to TNBC cells by inhibiting proliferation, reducing survival, and increasing apoptosis. The combination of MMAF and IMQ in one DualADC is more efficient in killing TNBC cells by integrating different anticancer mechanisms, which has great potential to bypass drug resistance, overcome cancer recurrence, and improve survival. Alternatively, the heterogeneous TNBCs with low CD276 expression could be targeted with other mAbs via alternative receptors such as EGFR, trophoblast cell surface antigen 2, LSR, or GRP56 as combined therapies, which could be investigated in future studies.

This study demonstrated high plasma stability of the DualADC in PK study and high TNBC specificity in biodistribution study. These features could benefit the TNBC treatment efficacy in future preclinical or clinical studies. Although maximal tolerated dosage was not reached, all tested doses of DualADC did not indicate toxicity in major organs, blood cell count, serum cytokine, body weight, survival, and general health. A future maximal tolerated dosage study is needed to define the safe dose for cancer treatment. A toxicology study using nonhuman primate is also highly desired to fully evaluate the potential toxicity including the effect on antigen-presenting cells with low CD276 expression in future.

In summary, this study developed an innovative therapy, CD276-targeted DualADC, which combines chemotherapy and immunotherapy into one molecule aiming to eliminate the aggressive and heterogeneous TNBCs. The promising anti-TNBC efficacy and minimal side effects were validated in multiple mouse models and posttreatment analyses. The concept of targeted delivery of dual payloads with synergistic functions is novel and readily translationable. In the new future, we will (i) improve the productivity of humanized CD276 mAb by optimizing expression vectors, (ii) further optimize the dual payloads and conjugation strategies to assure high circulation stability and biological function in vivo, (iii) combine this CD276-targeted DualADC with other therapy such as PD-1–targeted therapy, and (iv) develop or seek more clinically relevant models, such as PDX in humanized mouse models or nonhuman primate models, to collect preclinical data and pre-IND toxicology for launching future clinical trials.

L. Zhou reports grants from NIH and Department of Defense during the conduct of the study, as well as a patent for PCT/US2024/037838 pending and a patent for PCT/US2024/037844 pending. X.M. Liu reports grants from NIH and Department of Defense and nonfinancial support from the Ohio State University during the conduct of the study, as well as a patent for PCT/US2024/037838 pending and a patent for PCT/US2024/037844 pending. No disclosures were reported by the other authors.

Z.Z. Zhou: Resources, data curation, formal analysis, validation, methodology, writing–review and editing. Y. Si: Resources, data curation, formal analysis, validation, methodology, writing–review and editing. J. Zhang: Data curation, software, visualization, writing–review and editing. K. Chen: Resources, data curation, formal analysis, validation, methodology, writing–review and editing. A. George: Resources, data curation, formal analysis, validation, writing–review and editing. S. Kim: Resources, data curation, formal analysis, validation, visualization, methodology, writing–review and editing. L. Zhou: Conceptualization, supervision, funding acquisition, investigation, visualization, writing–review and editing. X.M. Liu: Conceptualization, resources, supervision, funding acquisition, investigation, writing–original draft, writing–review and editing.

The authors thank the Comparative Pathology and Digital Imaging Shared Resource, Clinical and Translational Science Shared Resource, Genomics Shared Resource, Small Animal Imaging Shared Resource, Flow Cytometry Shared Resource, and Campus Microscopy and Imaging Facility at the Ohio State University. The authors also thank Dr. Ju Hwan Cho’s kind support in Western blotting. This work is supported by NIH NCI 1R01CA281980-01 (X.M. Liu), 1R01CA262028-01A1 (X.M. Liu and L. Zhou), and DoD BCRP W81XWH2110066/67 (X. M. Liu and L. Zhou).

Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

1.
Hwang
S-Y
,
Park
S
,
Kwon
Y
.
Recent therapeutic trends and promising targets in triple negative breast cancer
.
Pharmacol Ther
2019
;
199
:
30
57
.
2.
Wein
L
,
Loi
S
.
Mechanisms of resistance of chemotherapy in early-stage triple negative breast cancer (TNBC)
.
Breast
2017
;
34
(
Suppl 1
):
S27
30
.
3.
Nedeljković
M
,
Damjanović
A
.
Mechanisms of chemotherapy resistance in triple-negative breast cancer-how we can rise to the challenge
.
Cells
2019
;
8
:
957
.
4.
Bardia
A
,
Messersmith
WA
,
Kio
EA
,
Berlin
JD
,
Vahdat
L
,
Masters
GA
, et al
.
Sacituzumab govitecan, a Trop-2-directed antibody-drug conjugate, for patients with epithelial cancer: final safety and efficacy results from the phase I/II IMMU-132-01 basket trial
.
Ann Oncol
2021
;
32
:
746
56
.
5.
Espinosa Fernandez
JR
,
Eckhardt
BL
,
Lee
J
,
Lim
B
,
Pearson
T
,
Seitz
RS
, et al
.
Identification of triple-negative breast cancer cell lines classified under the same molecular subtype using different molecular characterization techniques: implications for translational research
.
PLoS One
2020
;
15
:
e0231953
.
6.
Mayer
IA
,
Abramson
VG
,
Lehmann
BD
,
Pietenpol
JA
.
New strategies for triple-negative breast cancer–deciphering the heterogeneity
.
Clin Cancer Res
2014
;
20
:
782
90
.
7.
Roncato
F
,
Rruga
F
,
Porcù
E
,
Casarin
E
,
Ronca
R
,
Maccarinelli
F
, et al
.
Improvement and extension of anti-EGFR targeting in breast cancer therapy by integration with the Avidin-Nucleic-Acid-Nano-Assemblies
.
Nat Commun
2018
;
9
:
4070
.
8.
Loison
R
,
Loirat
D
.
[Pembrolizumab plus chemotherapy combination - first line in PD-L1 positive (CPS≥10) metastatic and advanced triple-negative breast cancer]
.
Bull Cancer
2022
;
109
:
387
9
.
9.
Chen
R
,
Jiang
X
,
Sun
D
,
Han
G
,
Wang
F
,
Ye
M
, et al
.
Glycoproteomics analysis of human liver tissue by combination of multiple enzyme digestion and hydrazide chemistry
.
J Proteome Res
2009
;
8
:
651
61
.
10.
Bachawal
SV
,
Jensen
KC
,
Wilson
KE
,
Tian
L
,
Lutz
AM
,
Willmann
JK
.
Breast cancer detection by B7-H3-targeted ultrasound molecular imaging
.
Cancer Res
2015
;
75
:
2501
9
.
11.
Castellanos
JR
,
Purvis
IJ
,
Labak
CM
,
Guda
MR
,
Tsung
AJ
,
Velpula
KK
, et al
.
B7-H3 role in the immune landscape of cancer
.
Am J Clin Exp Immunol
2017
;
6
:
66
75
.
12.
Seaman
S
,
Zhu
Z
,
Saha
S
,
Zhang
XM
,
Yang
MY
,
Hilton
MB
, et al
.
Eradication of tumors through simultaneous ablation of CD276/B7-H3-positive tumor cells and tumor vasculature
.
Cancer Cell
2017
;
31
:
501
15.e8
.
13.
Maeda
N
,
Yoshimura
K
,
Yamamoto
S
,
Kuramasu
A
,
Inoue
M
,
Suzuki
N
, et al
.
Expression of B7-H3, a potential factor of tumor immune evasion in combination with the number of regulatory T cells, affects against recurrence-free survival in breast cancer patients
.
Ann Surg Oncol
2014
;
21
(
Suppl 4
):
S546
54
.
14.
O’Donovan
DH
,
Mao
Y
,
Mele
DA
.
The next generation of pattern recognition receptor agonists: improving response rates in cancer immunotherapy
.
Curr Med Chem
2020
;
27
:
5654
74
.
15.
Xu
J
,
Li
X
,
Du
Y
.
Antibody-pattern recognition receptor agonist conjugates: a promising therapeutic strategy for cancer
.
Adv Biol (Weinh)
2022
;
6
:
e2101065
.
16.
Sun
H
,
Li
Y
,
Zhang
P
,
Xing
H
,
Zhao
S
,
Song
Y
, et al
.
Targeting toll-like receptor 7/8 for immunotherapy: recent advances and prospectives
.
Biomark Res
2022
;
10
:
89
.
17.
Patinote
C
,
Karroum
NB
,
Moarbess
G
,
Cirnat
N
,
Kassab
I
,
Bonnet
PA
, et al
.
Agonist and antagonist ligands of toll-like receptors 7 and 8: ingenious tools for therapeutic purposes
.
Eur J Med Chem
2020
;
193
:
112238
.
18.
Wu
SY
,
Xu
Y
,
Chen
L
,
Fan
L
,
Ma
XY
,
Zhao
S
, et al
.
Combined angiogenesis and PD-1 inhibition for immunomodulatory TNBC: concept exploration and biomarker analysis in the FUTURE-C-Plus trial
.
Mol Cancer
2022
;
21
:
84
.
19.
Chi
H
,
Li
C
,
Zhao
FS
,
Zhang
L
,
Ng
TB
,
Jin
G
, et al
.
Anti-tumor activity of toll-like receptor 7 agonists
.
Front Pharmacol
2017
;
8
:
304
.
20.
Zúñiga
LA
,
Leßmann
T
,
Uppal
K
,
Bisek
N
,
Hong
E
,
Rasmussen
CE
, et al
.
Intratumoral delivery of TransCon TLR7/8 agonist promotes sustained anti-tumor activity and local immune cell activation while minimizing systemic cytokine induction
.
Cancer Cell Int
2022
;
22
:
286
.
21.
Lee
SN
,
Jin
SM
,
Shin
HS
,
Lim
YT
.
Chemical strategies to enhance the therapeutic efficacy of toll-like receptor agonist based cancer immunotherapy
.
Acc Chem Res
2020
;
53
:
2081
93
.
22.
Teicher
BA
,
Morris
J
.
Antibody-drug conjugate targets, drugs, and linkers
.
Curr Cancer Drug Targets
2022
;
22
:
463
529
.
23.
Hafeez
U
,
Parakh
S
,
Gan
HK
,
Scott
AM
.
Antibody-drug conjugates for cancer therapy
.
Molecules
2020
;
25
:
4764
.
24.
Polakis
P
.
Antibody drug conjugates for cancer therapy
.
Pharmacol Rev
2016
;
68
:
3
19
.
25.
Mansoori
B
,
Mohammadi
A
,
Davudian
S
,
Shirjang
S
,
Baradaran
B
.
The different mechanisms of cancer drug resistance: a brief review
.
Adv Pharm Bull
2017
;
7
:
339
48
.
26.
Si
Y
,
Kim
S
,
Ou
J
,
Lu
Y
,
Ernst
P
,
Chen
K
, et al
.
Anti-SSTR2 antibody-drug conjugate for neuroendocrine tumor therapy
.
Cancer Gene Ther
2021
;
28
:
799
812
.
27.
Si
Y
,
Zhang
Y
,
Guan
J-S
,
Ngo
H
,
Totoro
A
,
Singh
A
, et al
.
Anti-CD47 monoclonal antibody-drug conjugate: a targeted therapy to treat triple-negative breast cancers
.
Vaccines (Basel)
2021
;
9
:
882
.
28.
Ou
J
,
Si
Y
,
Gah
KY
,
Song
J
,
Patrick
E
,
Flanigan
D
, et al
.
Process development of antibody-drug conjugation production for cancer treatment
.
PLoS One
2018
;
13
:
e0206246
.
29.
Xu
N
,
Ma
C
,
Ou
J
,
Sun
WW
,
Zhou
L
,
Hu
H
, et al
.
Comparative proteomic analysis of three Chinese Hamster Ovary (CHO) host cells
.
Biochem Eng J
2017
;
124
:
122
9
.
30.
Xu
N
,
Ou
J
,
Si
Y
,
Goh
K
,
Flanigan
D
,
Han
X
, et al
.
Proteomics insight into the production of monoclonal antibody
.
Biochem Eng J
2019
;
145
:
177
85
.
31.
Chen
K
,
Si
Y
,
Ou
J
,
Guan
JS
,
Kim
S
,
Ernst
P
, et al
.
Antibody-drug conjugate to treat meningiomas
.
Pharmaceuticals (Basel)
2021
;
14
:
427
.
32.
Xu
N
,
Ou
J
,
Gilani
AK
,
Zhou
L
,
Liu
M
.
High-level expression of recombinant IgG1 by CHO K1 platform
.
Front Chem Sci Eng
2015
;
9
:
376
80
.
33.
Si
Y
,
Guan
J
,
Xu
Y
,
Chen
K
,
Kim
S
,
Zhou
L
, et al
.
Dual-targeted extracellular vesicles to facilitate combined therapies for neuroendocrine cancer treatment
.
Pharmaceutics
2020
;
12
:
1079
.
34.
Si
Y
,
Kim
S
,
Zhang
E
,
Tang
Y
,
Jaskula-Sztul
R
,
Markert
JM
, et al
.
Targeted exosomes for drug delivery: biomanufacturing, surface tagging, and validation
.
Biotechnol J
2020
;
15
:
e1900163
.
35.
Si
Y
,
Xu
Y
,
Guan
J
,
Chen
K
,
Kim
S
,
Yang
ES
, et al
.
Anti-EGFR antibody-drug conjugate for triple-negative breast cancer therapy
.
Eng Life Sci
2021
;
21
:
37
44
.
36.
Si
Y
,
Zhang
Y
,
Ngo
HG
,
Guan
JS
,
Chen
K
,
Wang
Q
, et al
.
Targeted liposomal chemotherapies to treat triple-negative breast cancer
.
Cancers (Basel)
2021
;
13
:
3749
.
37.
Ou
J
,
Si
Y
,
Tang
Y
,
Salzer
GE
,
Lu
Y
,
Kim
S
, et al
.
Novel biomanufacturing platform for large-scale and high-quality human T cells production
.
J Biol Eng
2019
;
13
:
34
.
38.
Yang
S-T
,
Liu
X
.
Cell culture processes for biologics manufacturing: recent developments and trends
.
Pharm Bioprocess
2013
;
1
:
133
6
.
39.
Wu
L
,
Yi
B
,
Wei
S
,
Rao
D
,
He
Y
,
Naik
G
, et al
.
Loss of FOXP3 and TSC1 accelerates prostate cancer progression through synergistic transcriptional and posttranslational regulation of c-MYC
.
Cancer Res
2019
;
79
:
1413
25
.
40.
Schrörs
B
,
Boegel
S
,
Albrecht
C
,
Bukur
T
,
Bukur
V
,
Holtsträter
C
, et al
.
Multi-omics characterization of the 4T1 murine mammary gland tumor model
.
Front Oncol
2020
;
10
:
1195
.
41.
Topalian
SL
,
Sznol
M
,
McDermott
DF
,
Kluger
HM
,
Carvajal
RD
,
Sharfman
WH
, et al
.
Survival, durable tumor remission, and long-term safety in patients with advanced melanoma receiving nivolumab
.
J Clin Oncol
2014
;
32
:
1020
30
.
42.
Huck
BR
,
Kötzner
L
,
Urbahns
K
.
Small molecules drive big improvements in immuno-oncology therapies
.
Angew Chem Int Ed Engl
2018
;
57
:
4412
28
.
43.
Galluzzi
L
,
Buqué
A
,
Kepp
O
,
Zitvogel
L
,
Kroemer
G
.
Immunogenic cell death in cancer and infectious disease
.
Nat Rev Immunol
2017
;
17
:
97
111
.
This open access article is distributed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) license.