Genetically Engineered Biomimetic Nanovesicles Co-Deliveing a Checkpoint Inhibitor and Doxorubicin for Enhanced Cancer Chemo-Immunotherapy

Abstract

Background/Objectives: Despite the clinical success of immune checkpoint blockade (ICB), its efficacy remains limited in immunologically “cold” tumors, primarily due to poor immunogenicity and an immunosuppressive tumor microenvironment (TME). Chemo-immunotherapy offers a potential strategy to enhance ICB response, yet its application is often hindered by inadequate tumor-targeted delivery and systemic immunosuppressive side effects. Biomimetic nanotechnology represents a promising approach to overcoming these limitations by improving drug delivery and facilitating effective combination regimens. Methods: We developed a biomimetic nanosystem (NVs@DOX) through genetic engineering of cellular membranes and optimized nanoformulation techniques, enabling co-delivery of doxorubicin (DOX) and ICB agents. This design aims to maximize synergistic antitumor effects while minimizing adverse impacts. Results: In vitro studies demonstrated the potent cytotoxicity of NVs@DOX, including significant inhibition of cancer cell proliferation and complete suppression of colony formation. In a 4T1 murine breast cancer model, NVs@DOX treatment led to substantial tumor growth inhibition (approximately 72%) without notable body weight loss, underscoring a favorable safety profile alongside enhanced therapeutic efficacy. Conclusions: The NVs@DOX platform effectively integrates doxorubicin with ICB within a biomimetic nanocarrier, significantly improving chemo-immunotherapy outcomes. This strategy highlights the potential of genetically engineered cellular nanoparticles as a promising combinatorial approach for the treatment of breast cancer.

1. Introduction

Chemotherapy and immunotherapy are both advantageous monotherapeutic approaches in cancer treatment, employing distinct therapeutic strategies yet with daunting limitations. Immune checkpoint blockade (ICB) therapy, particularly targeting the programmed cell death-1 (PD-1)/PD-1 ligand (PD-L1) pathway, has achieved remarkable clinical success in malignancies such as melanoma, non-small cell lung cancer, colorectal cancer, and renal cell carcinoma by reinvigorating exhausted T cells [1,2,3,4,5]; however, its efficacy is often undermined in immunologically “cold” tumors which are characterized by poor immunogenicity and an immunosuppressive tumor microenvironment (TME) [6,7,8,9]. Similarly, conventional chemotherapy can induce rapid tumor responses but is limited by severe systemic toxicities and the frequent development of treatment resistance. For example, doxorubicin (DOX), with a potent anti-cancer effect through DNA intercalation and topoisomerase inhibition [10], is hampered by cardiotoxicity and myelosuppression [11,12,13,14], as well as chemoresistance [15,16]. These challenges underscore that relying solely on monotherapy is not the most effective strategy in the fight against cancer.

Recent studies have demonstrated that combining immunotherapy with chemotherapy can produce synergistic therapeutic effects [17,18,19]. In such combinations, the cytotoxic drugs act as immunomodulators by inducing immunogenic cell death (ICD) [20,21,22], which enhances the release of tumor antigens and their subsequent capture by dendritic cells (DCs) [23,24,25]. Additionally, certain chemotherapeutic agents can boost the adjuvanticity of tumors by depleting certain immunosuppressive factors such as Tregs and cytokines in the TME [26,27]. However, translating the combination strategy into clinical practice faces several challenges, including inefficient on-target delivery due to physiological barriers [28,29], temporal/spatial mismatches between immune activation and drug delivery [30], and off-target adverse effects on the immune system caused by nonspecific distribution of chemical agents. These limitations highlight the demand for developing innovative delivery strategies to maximize the combinatory effect of immunotherapy and chemotherapy.

Nanotechnology, especially biomimetic nanotechnology, has revolutionized the targeted and efficient delivery of cancer therapies in clinical practice, as exemplified by liposomal Doxil^® [31], and has also achieved efficient immunotherapeutic effects via cell membrane-derived vesicles in many preclinical studies [32]. In this context, tumor cell membranes have been successfully engineered as innovative biomimetic materials for drug delivery. Coating nanocarriers with such membranes confers homotypic targeting capability, mediated by adhesion molecules expressed on the membrane, which helps reduce systemic adverse responses [33,34]. Moreover, these biomimetic materials can be genetically modified to act as ICB agents with high levels of PD-1 [35,36]. For instance, Gu et al. engineered cell membrane-derived nanovesicles to display PD-1 receptors artificially. This design specifically blocks the PD-1/PD-L1 pathway to potentiate ICB therapy, offering enhanced efficacy with reduced toxicity. In contrast to systemic antibody injection, this targeted strategy minimizes off-target effects and mitigates the risk of excessive immune activation. Furthermore, the materials can encapsulate multiple therapeutic agents with high efficiency and provide sustained release profiles [37], similar to Vyxeos^®, the first liposomal combination nanomedicine. Furthermore, in contrast to many inorganic nanocarriers, which can raise long-term toxicity concerns and require complex surface modifications for targeting, engineered cell membrane-derived nanovesicles offer intrinsic biocompatibility, natural homotypic targeting, and an integrated checkpoint blockade. Therefore, biomimetic nanotechnology is a versatile and promising platform for the combined delivery of immunotherapy and chemotherapy, maximizing chemo-immunotherapeutic efficacy.

To this end, a biomimetic nanosystem (NVs@DOX) was developed through the genetic engineering of tumor cells. In this system, PD-1-enriched nanovesicles (NVs) derived from tumor cell membranes not only serve as highly efficient targeted-delivery vehicles for cancer therapies but also function as ICB agents. Additionally, the immunomodulatory drug DOX is encapsulated within the NVs to work synergistically with the PD-1 molecules on the vesicles (Figure 1A). In the design of NVs@DOX, DOX release from NVs@DOX is triggered in acidic environments, such as endosomes following cellular uptake or within the TME. Once released, DOX exerts its cytotoxic effect on tumor cells by intercalating DNA and inhibiting topoisomerase, which leads to tumor antigen release and subsequent activation of antitumor immune responses (Figure 1B). Simultaneously, the PD-1 present on NVs competes with PD-1 on T cells to neutralize PD-L1 on tumor cells, thereby effectively blocking the immunosuppressive PD-1/PD-L1 signaling pathway and enhancing the immune response initiated by DOX (Figure 1C). Through this dual mechanism (Figure 1B,C), NVs@DOX can deliver a combined chemo-immunotherapeutic effect by inducing ICD while promoting T-cell activation. Moreover, systemic toxicity can be minimized due to the targeted release of DOX in acidic biological environments (endosomes and TME) and specific tumor accumulation mediated by homologous targeting of the NVs. Proof-of-concept studies demonstrated that NVs@DOX exhibited superior cellular uptake and potent cytotoxicity in vitro, as well as enhanced tumor accumulation and improved therapeutic efficacy with no significant weight loss in vivo, compared to monotherapies. Overall, the biomimetic nanosystem NVs@DOX successfully achieved chemo-immunotherapeutic therapy through efficient on-target delivery and a synergistic combination approach.

2. Materials, Characterization, and Methods

2.1. Characterization

FEI-TALOS-F200X TEM (Thermo Fisher Scientific, Wilmington, DE, USA), Malvern Zetasizer Nano-ZS (Malvern PANalytical, Malvern, UK), UV-2600i spectrophotometer (Shimadzu, Kyoto, Japan) were used for characterization of nanoparticles in this study.

2.2. Cell Culture, Construction, and Characterization of Genetically Engineered Cells

4T1 cells were purchased from Servicebio (Wuhan, China) and maintained in high-glucose DMEM (Gibco) supplemented with 10% fetal bovine serum (Gibco, Thermo Fisher Scientific, Wilmington, DE, USA) and 1% penicillin/streptomycin at 37 °C in humidified 5% CO₂ incubator. PD-1/m Cherry-expressing 4T1 cells were generated via blasticidin-resistant plasmid transfection. Positive cells were sorted by flow cytometry (Sony SH800, Tokyo, Japan) and then placed in incubator to continue their expansion with fresh culture medium. PD-1 expression was confirmed by confocal laser scanning microscopy (CLSM; ZEISS LSM980, Carl Zeiss AG, Jena, Germany).

2.3. Preparation of NVs and NVs@DOX

The PD-1-overexpressed 4T1 cells were disrupted by ultrasound (40% power, pulse 1 s followed by 2 s interval, repeated 20 times, SCIENT Z), and then the cell membranes were collected by centrifugation (4000 rpm for 40 min; then 100,000× g for 2 h, Optima XE-90, Beckman Coulter, Brea, CA, USA). The obtained pellet was resuspended in PBS and extruded through 800, 400, and 200 nm polycarbonate membranes to obtain NVs (Avanti Polar Lipids). The protein concentration of NVs was quantified using a BCA assay (KeyGEN, BioTECH., Nanjing, China). Subsequently, 100 μg of NVs were incubated with 160 μg of DOX hydrochloride (DOX·HCl, Meilunbio, Dalian, China) in 1 mL PBS (pH 7.4) under continuous shaking at 4 °C. The mixture was then subjected to ice-bath sonication (5 min, 40 kHz), followed by overnight incubation at 4 °C to facilitate drug encapsulation. The NVs@DOX were pelleted via centrifugation (100,000× g, 2 h) and stored at 4 °C until further use. The mass of DOX was determined using calibration curves obtained at 488 nm in PBS buffer (pH 7.4). The loading content (LC) was calculated using the following equation: LC (wt%) = [(Mass of DOX in NVs@DOX)/Mass of NVs@DOX] × 100%

2.4. pH-Responsive Release of DOX

To evaluate the pH-responsive release of DOX, NVs@DOX was mixed with PBS solution (pH 7.4 or pH 5.5) and then stirred for different times (2, 4, 8, and 12 h) in the dark, respectively. At predetermined time intervals, aliquots were collected and centrifuged at 100,000× g for 2 h at 4 °C to separate the nanoparticles from the released DOX in the supernatant. The release kinetics were directly assessed by monitoring UV-Vis absorption (488 nm) of the resulting mixture. The content of DOX was determined using calibration curves obtained at 488 nm in different PBS buffers (pH 7.4 or pH 5.5), and the release of DOX was calculated using the following equation: Drug release (%) = [(DOX in supernatant after centrifugation)/DOX in NVs@DOX] × 100%

2.5. Cytotoxicity Validation and Cell Colony Formation

4T1 cells were maintained in high-glucose DMEM (Gibco) supplemented with 10% fetal bovine serum (Gibco, USA) and 1% penicillin/streptomycin at 37 °C in humidified 5% CO₂ incubator. Cell viability was quantified using CCK-8 assay (Beyotime Biotechnology, Shanghai, China). Briefly, 4T1 cells (5 × 10³ cells/well) were incubated with different treatments for 24 h or 48 h. Absorbance at 450 nm was measured after 2 h incubation with CCK-8 reagents. Cells were seeded at low density (4 × 10³ cells/well) and treated with the same concentration of NVs or NVs@DOX for 24 h. After 7–10 days incubation, colonies were fixed with methanol, stained with 0.5% crystal violet (Beyotime Biotechnology), and quantified using ImageJ 2.0 software.

2.6. Flow Cytometry Analysis on Apoptosis/Necrosis

Annexin V-FITC/propidium iodide (PI) double staining was performed on cells treated with NVs or NVs@DOX with an equivalent concentration of DOX (16 µg/mL) for 48 h. Flow cytometric analysis was conducted using a Beckman Coulter system (USA), with data analyzed using Flowjo v10.8.1 software. For each sample, 10,000 events were acquired. The gating strategy involved first excluding debris based on FSC-A and SSC-A, followed by the selection of single cells using FSC-A versus FSC-H. Apoptotic and necrotic populations were then analyzed in the Annexin V-FITC/PI channels.

2.7. SDS-PAGE Analysis and Western Blot Analysis

Total cellular proteins were extracted using RIPA buffer, quantified by BCA assay, and separated by 10% SDS-PAGE (80 V for 30 min followed by 120 V for 60 min). Gels were stained with Coomassie Brilliant Blue (Solarbio, Beijing, China) and destained for protein band visualization. Following electrophoresis, the gel was subjected to Coomassie blue staining (Solarbio, Beijing, China) for 30 min, followed by three DI water washes and overnight destaining prior to imaging. For Western blot analysis, the separated proteins were transferred to PVDF membranes (ice-cooled conditions). After blocking (5% non-fat milk, 1 h), the membranes were incubated sequentially with PD-1 primary antibody (overnight, 4 °C; Abcam, Cambridge, UK) and HRP-conjugated secondary antibody (1 h, Servicebio, Wuhan, China).

2.8. Cellular Uptake Assay

4T1 cells were seeded in 6-well plates at a density of 1 × 10⁵ cells per well and allowed to adhere prior to NVs@DOX treatment. Following incubation for different time points, cells were processed sequentially by PBS washing, trypsin digestion, and centrifugation. The harvested cells were then resuspended in PBS containing with 2% FBS, and DOX fluorescence was measured using flow cytometry (Beckman Coulter; B610-ECD channel, excitation 488 nm). For subcellular localization studies, cells cultured overnight in confocal dishes were exposed to NVs@DOX (1 or 8 h). Followed by being stained with LysoTracker Red (30 min) and Hoechst 33342 (10 min), fluorescence signals were acquired by confocal microscopy with specific excitation wavelengths, i.e., 647 nm (LysoTracker, Beyotime, Shanghai, China), 346 nm (Hoechst), and 488 nm (DOX).

2.9. Hemolysis Assay

Fresh blood was harvested from healthy Balb/c mice and collected into EDTA-anticoagulated tubes. After centrifugation and repeated washing until a colorless supernatant was obtained, the pelleted red blood cells (RBCs) were incubated with various concentrations of NVs@DOX for 5 h at 37 °C, using PBS and H₂O as negative and positive controls, respectively. Following incubation, the samples were centrifuged (15,000 rpm, 15 min), and the absorbance of the supernatants was measured at 570 nm. The hemolysis ratio (HR) was calculated as HR (%) = [(Asample − Anegative)/(Apositive − Anegative)] × 100%.

2.10. Therapeutic Evaluation in 4T1 Primary Tumor Mice Model

Female Balb/c mice (6–8 weeks old) were obtained from the Laboratory Animal Center of Shenzhen Bay Laboratory and maintained under standard conditions for in vivo studies. Shenzhen Bay Laboratory Animal Experimentation Regional Ethics Committee (Permit No. AERL202401). To establish the tumor model, 4T1 cells were inoculated subcutaneously in the dorsal region. Tumor dimensions were measured using digital calipers, and tumor volumes were calculated using the formula Length × Width²)/2. Humane endpoints (euthanasia) were implemented when tumors reached over 1500 mm³. The tumor model was established by right dorsal flank injection of 1 × 10⁶ 4T1 cells in Balb/c mice. Once tumors reached approximately 100 mm³, animals were randomly assigned to one of four groups: PBS, DOX, NVs, and NVs@DOX (n = 4). Intravenous administration was conducted on Days 0, 2, and 6 with an equivalent dose of DOX (100 μg per mouse). Tumor size and body weight were measured every two days throughout the study. On day 8, one mouse in each group was sacrificed, and the corresponding tumor was taken out for ICD marker staining. For immune profiling, tumors were harvested from the 4T1 models on day 8 post-treatment initiation, following the same therapeutic regimen as the tumor suppression study. Each tumor was dissected into 2–4 mm fragments and digested in a solution containing collagenase IV (0.2 mg mL⁻¹), DNase I (250 U mL⁻¹), and hyaluronidase (0.1 mg mL⁻¹) at 37 °C for 40 min. The digested tissue was then processed into a single-cell suspension for subsequent staining with the specified fluorescent antibodies (CD45, CD3, CD4, and CD8, BioLegend), and the result was analyzed by flow cytometry. Additionally, to evaluate the synergistic therapeutic advantages of NVs@DOX, mice bearing 4T1 tumors were randomly allocated to receive PBS (control), NVs+DOX, 4T1@DOX, or NVs@DOX (n = 4). Treatments were administered intravenously on days 0, 2, 4, and 6, with the DOX-loaded groups (NVs+DOX, 4T1@DOX, and NVs@DOX) each receiving 100 μg of DOX per mouse per dose. On day 8, one mouse in each group was sacrificed, and the corresponding tumor was taken out for immunofluorescence staining.

2.11. Statistical Analysis

All results are presented via the mean ± standard deviation. Two-group comparisons were conducted using paired two-tailed t-tests. Multiple-group comparisons were analyzed using ordinary one-way or two-way ANOVA followed by Tukey’s test. All statistical analyses were performed with Prism 8.0 software (GraphPad Prism 8.0 software.).

3. Results and Discussion

3.1. Characteristics of NVs@DOX

PD-1-modified nanovesicles (NVs) and their DOX-loaded counterparts (NVs@DOX) were systematically characterized. First of all, the expression of mCherry-PD-1 fusion protein of modified 4T1-PD-1 cells was verified by the strong red fluorescence observed through immunofluorescence microscopy, compared to untransfected controls (Figure 2A). This PD-1 expression was further validated by flow cytometric analysis (Figure 2B). Furthermore, Western blot analysis (Figure 2C) corroborated the successful incorporation of PD-1 in both engineered cells (4T1-PD-1) and derived NVs, while preserving native membrane protein profiles, as indicated by Na⁺/K⁺ ATPase as a membrane marker and GAPDH as an intracellular reference. Then, transmission electron microscopy (TEM) showed that NVs exhibited a spherical morphology with an intact bilayer and a closed vesicle structure (Figure 2D). The hydrodynamic size of NVs@DOX was approximately 165 nm with a polydispersity index of 0.161 (Figure S1), which indicated the good dispersion of NVs@DOX in the buffer.

A maximum DOX loading content of 45% was achieved at a minimal feed concentration of 160 μg/mL, given the concentration-dependent DOX-loading method (Figure 2E). Native NVs exhibited a zeta potential of −30 mV, which decreased to −35 mV after DOX loading, indicative of successful drug loading (Figure 2F). The colloidal stability of the nanoparticles was demonstrated by there being no significant change in hydrodynamic diameter over 7 days in physiological buffers (Figure 2G). Extending the stability evaluation period to 14 days, NVs@DOX showed no notable degradation or aggregation in physiological solution, exhibiting good stability (Figure S2). It is well-known that the cell membrane coating of NVs@DOX provides inherent biocompatibility and targeting. Supporting this, hemolysis tests revealed a low hemolysis rate (<5%, Figure S3), underscoring its hemocompatibility and foundational biosafety. Additionally, DOX was released in a pH-dependent manner, with less than 20% cumulative release at pH 7.4 versus more than 60% release in acidic buffer (pH 5.5). This feature is expected to enhance DOX release in acidic TME. SDS-PAGE analysis confirms that the loading of DOX did not affect the expression of surface proteins on the vesicles (Figure 2I).

3.2. Efficient Cellular Uptake and Tumor Targeting of NVs@DOX

To investigate the cellular uptake of NVs@DOX, the binding capability of NVs@DOX to murine breast carcinoma 4T1 cell membranes was first assessed. Confocal fluorescence imaging revealed the green fluorescent signal (NVs@DOX) overlapped with the red fluorescent signal (cell membrane), indicating successful cellular entry of NVs@DOX (Figure 3A). Moreover, as shown in Figure 3B, 4T1 cells exhibited a time-dependent uptake of NVs@DOX, with fluorescence intensity steadily increasing from 0 to 6 h, demonstrating efficient nanodrug internalization, as confirmed by the rising median fluorescence intensity and substantial cellular accumulation.

The acidic environment of lysosomes facilitates the release of DOX from NVs@DOX; therefore, we further studied the co-localization of DOX and lysosomes. NVs@DOX NPs were abundantly internalized into the cells, showing a gradual increase in green signal over time (Figure 3C and Figure S4). A strong co-localization of the signals of NVs@DOX and lysotrackers was observed in 8 h, indicating that NVs@DOX NPs entered cells via a certain endolysosomal pathway. Confocal imaging further confirmed this by revealing time-dependent accumulation of DOX-associated green fluorescence within lysosomal compartments labeled by LysoTracker, with maximal overlap in 8 h.

Since the anti-cancer efficacy of NVs@DOX depends heavily on tumor targeting and penetration within the tumor, we evaluated the tumor-targeting ability of NVs@DOX using fluorescent dyes. 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindotricarbocyanine iodide (DiR)-labeled NVs@DOX and free DiR were administered intravenously to 4T1 tumor-bearing Balb/c mice. Compared with free DiR, the fluorescence signal of DiR-labeled NVs@DOX at the tumor site demonstrated a progressive increase over time, indicating the effective tumor-targeting and tumor-penetrating capability of NVs@DOX (Figure S5).

3.3. Potent Cytotoxicity of NVs@DOX In Vitro

The cytotoxic effect of NVs and NVs@DOX was thoroughly evaluated through multiple approaches. CCK-8 assays revealed a concentration-dependent cytotoxicity profile (Figure 4A). Notably, NVs@DOX exhibited significantly enhanced toxicity over time, killing ~50% (24 h) and ~70% (48 h) of cells (Figure 4B,C), indicating an enhanced therapeutic effect with prolonged exposure. In contrast, NVs alone maintained high cell viability (>85%) even at high concentrations after 24 h (Figure 4A). Flow cytometric analysis using Annexin V-FITC/PI staining revealed significant differences in apoptotic response (Figure 4D,E). NVs@DOX caused a cell apoptosis rate of up to 50%, whereas it was below 10% for the treatment of PBS and NVs. This result indicates that the killing effect of NVs@DOX was mainly achieved through the activation of the cell apoptosis pathway. Clonogenic assays further confirmed the superior antitumor efficacy of NVs@DOX, demonstrating complete suppression of colony formation with a survival fraction, i.e., below 30% vs. 95% in PBS and NVs groups (Figure 4F,G). These results collectively demonstrate a dual anti-cancer mechanism, i.e., homologous targeting by NVs combined with the chemotherapeutic action of DOX, resulting in enhanced therapeutic outcomes.

3.4. Synergistic Antitumor Effect of NVs@DOX In Vivo

The antitumor efficacy of NVs@DOX was systematically evaluated using 4T1 tumor-bearing Balb/c mice. After the tumors reached approximately~100 mm³, the animals were randomized into four treatment groups (PBS, DOX, NVs, NVs@DOX; n = 4) and received intravenous injections on Days 0, 2, and 6 (Figure 5A). NVs@DOX achieved the greatest tumor volume inhibition (72.1%) and the lowest tumor weight (0.21 g) (Figure 5E and Figure S6). In comparison, DOX and NVs groups showed moderate growth inhibition of 32.9% and 50.6%, respectively (Figure 5B,C and Figure S6). It should be noted that no significant change in body weight was observed in the treatment with NVs and NVs@DOX, unlike with the DOX treatment (Figure 5D), indicating an improved safety profile for NVs@DOX. Then, to evaluate ICD induction by NVs@DOX in vivo, the release of damage-associated molecular patterns (DAMPs) in tumor tissue was examined via immunohistochemistry. Calreticulin (CRT) exposure and high mobility group protein B1 (HMGB1) translocation were markedly enhanced in the NVs@DOX group compared to all controls (Figure S7), demonstrating its efficacy in inducing ICD and remodeling the immunogenicity of the TME in solid tumors. Given the robust ICD induced by NVs@DOX, we next explored the influence of different treatments on T-cell activation. As expected, NVs@DOX induced the highest CD8⁺ T-cell infiltration (16.11%), a level 3.6 times greater than that in the PBS group (4.48%). (Figure S8). The enhanced therapy effect of NVs@DOX was attributed to the synergistic effect of PD-1 combined with DOX and the efficient biomimetic delivery system. Additionally, to evaluate the therapeutic advantages of the prepared nanocomposites, we also compared the immune activation effects of NVs@DOX, NVs combined with DOX (NVs+DOX) and non-engineered (wild-type) 4T1 cell-derived vesicle-loaded DOX (4T1@DOX) by immunofluorescence staining of tumor tissues. As shown in Figure S9, the NVs@DOX group showed the strongest and most extensive CD8⁺ T cell infiltration, suggesting its superior ability to activate antitumor immunity. This result demonstrated the therapeutic superiority of the NVs@DOX nanocomposite, an effect that stemmed from its dual capabilities in targeted PD-1 delivery and enhanced tumor accumulation of DOX.

4. Conclusions

By the rational engineering of tumor cell membranes in this study, a PD-1-enriched, DOX-loaded, biomimetic nanosystem (NVs@DOX) was successfully developed for enhanced cancer chemo-immunotherapy, with optimal physicochemical properties including a uniform spherical morphology (165 nm), high drug-loading capacity (45%), and pH-responsive release kinetics. In vitro experiments demonstrated efficient cellular uptake of NVs@DOX via endolysosomal trafficking, resulting in significant cytotoxicity through apoptosis and complete suppression of colony formation. In vivo evaluation using 4T1 tumor-bearing mice demonstrated that NVs@DOX achieved remarkably higher tumor growth inhibition (72.1%) than free DOX or NVs alone, without causing significant weight loss. The enhanced antitumor effect was attributed to the combination of PD-1 and DOX with distinctive immunotherapeutic mechanisms and efficient drug delivery by as-developed biomimetic nanosystem. These findings highlight the potential of NVs@DOX as a promising combination strategy for breast cancer with improved immunotherapeutic efficacy. Future research will focus on optimizing dosing regimens and evaluating long-term therapeutic outcomes in advanced tumor models.