Programmable nanomicelles rewire myeloid immunity for durable control of primary and metastatic breast cancer

Yang, Jie; Chang, Di; Li, Yingbo; Xu, Zuoyu; Li, Daoshuang; Xu, Xiaoxuan; Zhang, Zhiqi; Li, Yize; Zhang, Jianqiong; Xie, Jinbing; Xiao, Zebin; Ju, Shenghong

doi:10.1038/s41467-026-70859-5

Download PDF

Article
Open access
Published: 17 March 2026

Programmable nanomicelles rewire myeloid immunity for durable control of primary and metastatic breast cancer

Nature Communications volume 17, Article number: 4082 (2026) Cite this article

3263 Accesses
8 Altmetric
Metrics details

Subjects

Abstract

Immunosuppressive myeloid cells, including immature dendritic cells (DC) and tumor-promoting macrophages, drive tumor progression and immune evasion. Reprogramming these cells into immune effectors remains challenging. Here, we present a programmable nanomicelle platform that reconfigures myeloid immunity to combat both primary and metastatic breast cancer (BC). The nanomicelles co-deliver photosensitizers (IR825), anti-PD-L1 antibodies, and antagomiR-182 with spatiotemporal precision. Upon activation, they induce immunogenic cell death, promoting DC maturation to enhance antigen presentation. Concurrently, miR-182 inhibition repolarizes M2-like macrophages into M1-like phenotypes via the TLR4/MYD88/NF-κB pathway. This dual reprogramming enhances CD8⁺ T cell infiltration and functionality, driving durable anti-tumor responses. In multiple BC models, the platform suppresses tumor growth, prevents metastasis, and induces long-term immune memory. Its efficacy and immunomodulatory effects were further validated in a patient-derived tumor fragment assay and patient-derived xenograft models, highlighting translational potential. This work offers a clinically relevant strategy to rewire tumor-promoting myeloid cells for cancer immunotherapy.

Targeting myeloid-derived suppressor cells in combination with tumor cell vaccination predicts anti-tumor immunity and breast cancer dormancy: an in silico experiment

Article Open access 11 April 2023

Machine learning-based spatial characterization of tumor-immune microenvironment in the EORTC 10994/BIG 1-00 early breast cancer trial

Article Open access 07 March 2025

Therapeutic utility of engineered myeloid cells in the tumor microenvironment

Article Open access 28 February 2023

Introduction

Breast cancer (BC) is the most prevalent malignancy and a leading cause of cancer-related mortality among women globally^1,2. Among its subtypes, triple-negative BC (TNBC) poses the greatest therapeutic challenge due to its aggressive nature, lack of hormone receptor expression, and absence of human epidermal growth factor receptor 2 amplification, leaving patients with limited treatment options and dismal outcomes^3,4. The 5-year survival rate for patients with recurrent or metastatic triple-negative breast cancer remains below 15%, despite advances in systemic therapies⁵. Disappointingly, even landmark trials such as IMpassion132, which combined anti-programmed death ligand 1 (αPD-L1) antibody atezolizumab with chemotherapy, have achieved only modest efficacy, with a 40% overall response rate and no significant improvement in survival⁶. These outcomes highlight an urgent unmet need for innovative therapeutic strategies that can overcome the barriers posed by the immunosuppressive tumor microenvironment (TME) and provide a durable response.

A hallmark of the TME in TNBC is its dense infiltration of myeloid cells, including immature dendritic cells (DCs) and abundant tumor-associated macrophages (TAMs), which exhibit remarkable plasticity and dual functionality^7,8. In early tumorigenesis, myeloid cells can support tumor elimination by orchestrating innate and adaptive immune responses^9,10,11,12. However, in established tumors, these same cells are often co-opted into immunosuppressive roles, secreting cytokines and growth factors that drive immune escape, tumor progression, and metastasis^13,14,15. This duality presents a unique therapeutic opportunity: reprogramming myeloid cells toward an anti-tumor phenotype could dismantle the immunosuppressive TME and unlock robust anti-tumor immunity.

DCs are central orchestrators of immune surveillance and T cell activation^16,17,18. However, their anti-tumor potential remains limited by their functional states^19,20, which are often skewed toward immaturity within the TME²¹. While immunogenic cell death (ICD) inducers can promote DC maturation and antigen presentation²², they are insufficient to overcome immune evasion due to the upregulation of immune checkpoints, such as PD-L1²³. Recent advances in nanotechnology have enabled the development of photosensitizer immunoconjugates^23,24, which synergize immune activation and tumor destruction by simultaneously inducing ICD and potentiating immune checkpoint blockade (ICB)²⁵. These hybrid systems hold promise for achieving spatiotemporally precise activation of DCs and effector T cells, addressing critical challenges of poor tumor penetration and immune-related adverse events²⁶.

Similarly, TAMs, a major myeloid population within the TME, play pivotal roles in shaping immune responses^27,28. M2-like TAMs, often referred to as the “protumor” phenotype, drive immune suppression, angiogenesis, and tumor metastasis²⁹. In contrast, M1-like TAMs are potent effector cells that secrete proinflammatory cytokines and activate cytotoxic T cells³⁰. Despite progress in macrophage-reprogramming therapies³¹, clinical outcomes remain inconsistent, largely due to the complexity of macrophage polarization pathways³². MicroRNAs (miRNAs) have emerged as powerful regulators of macrophage function^33,34, offering an innovative approach to fine-tune TAM phenotypes and enhance ICB therapies³⁵.

Here, we introduce a ROS-responsive polymer micelle nanoplatform for precise co-delivery of IR825, αPD-L1, and antagomiR-182 to reprogram myeloid cells and improve immunotherapy outcomes in TNBC. IR825 conjugated with αPD-L1 promotes dendritic cell maturation and antigen presentation, while antagomiR-182 repolarizes myeloid cells toward the pro-inflammatory phenotype, collectively enhancing cytotoxic T cell responses and effectively inhibiting tumor progression and metastasis in both murine TNBC models and patient-derived specimens. The coordinated remodeling of innate immunity and adaptive immunity establishes sustained systemic anti-tumor activity. Our findings indicate that this intelligent spatiotemporal nanoplatform not only has the potential to overcome the challenges of TNBC, but also sets a new benchmark for immunotherapy design. By achieving precise multimodal immune reprogramming, this approach provides a solid foundation for future clinical translation and the pursuit of durable cancer remission.

Results

Immune landscape of primary and metastatic breast cancer cohort

To comprehensively characterize the immune status of the TME in BCs, we performed single-cell RNA sequencing (scRNA-seq) on six fresh tissue samples from three BC patients, including three surgically resected primary BC tumors (P1-P3) and their paired normal tissues (N1-N3). A total of 45,620 single-cells passed stringent quality control in the scRNA-seq dataset and were annotated based on canonical lineage markers (Supplementary Fig. 1). These cells were grouped into 14 clusters representing six main cell types, including endothelial cells, fibroblasts, lymphoid cells, epithelial cells, myeloid cells, and pericytes. Notably, scRNA-seq analysis revealed a pronounced enrichment of myeloid cells within tumor tissues. Among these major myeloid subpopulations, lower levels of mature dendritic cells (mDCs, identified as LAMP3⁺ DCs) and higher proportions of M2-like macrophages were observed in tumor tissues compared to adjacent normal tissues. Flow cytometry analysis further revealed significantly higher proportions of immature DCs in tumor tissues, which could impair antigen processing and presentation (Supplementary Fig. 2a, b). Furthermore, total macrophages constituted 25.48 ± 3.82% of cells in tumors, compared to 5.22 ± 1.36% in adjacent normal tissue. M2-like macrophages were 2.8-fold more abundant in tumors (57.86 ± 9.76%) than in adjacent normal tissue (20.44 ± 2.59%) (Supplementary Fig. 2c). This observation is consistent with their established roles in antigen presentation and immunosuppression within the TME. Moreover, the infiltration score of M2-like macrophages negatively correlated with CD8⁺ T cell population (Supplementary Fig. 3). We further explore the relationship between M1/M2-like macrophages and CD4⁺/CD8⁺ T cell infiltration, as well as the correlation between PD-L1 expression and CD8⁺ T cell infiltration in human BC (Supplementary Fig. 4). Collectively, these findings highlight the stage- and tissue-specific expression of DCs and M2-like macrophages as pivotal therapeutic targets in BC.

Breast cancer’s high mortality rate is primarily driven by metastasis³⁶, with the lung being one of the most common metastatic sites^37,38. The median survival time after lung metastasis diagnosis is less than 2 years, posing a significant clinical challenge. Understanding the cellular composition of primary and metastatic lesions and elucidating mechanisms underlying metastasis remain critical research priorities. To address this, we analyzed fat pad, primary tumor, and lung metastases tissues from MMTV-PyMT transgenic mice, a widely used mouse model of invasive BC, using flow cytometry (Supplementary Fig. 5a). Immature DCs were highly enriched in both primary tumors and metastatic tumors (Supplementary Fig. 5b). Similarly, total macrophage and M2 macrophage populations were significantly elevated in both primary and metastatic tumors (Supplementary Fig. 5c, d), consistent with clinical observations in BC patients³⁹. Together, these findings underscore the pivotal role of myeloid cell reprogramming in remodeling the TME and inhibiting metastasis.

Meanwhile, while previous studies have identified miR-182 as an oncogene in cancer cells³⁴, we found it to be significantly overexpressed in M2 macrophages within BC tissues (Supplementary Fig. 6). These results suggested that miR-182 may play an important role in M2 macrophage reprogramming of breast cancer.

Characterization of IPANP

Given the critical role of myeloid cells in modulating the TME, they represent a promising target for BC immunotherapy. In this study, we developed ROS-responsive polymer micelles IR825-αPD-L1@antagomiR-182@PEG-PLL (IPANP) to reprogram the immunosuppressive TME. IPANP was synthesized by co-encapsulating IR825, αPD-L1, and antagomiR-182, enabling precise BC diagnosis and therapy (Fig. 1a). Optimization of drug encapsulation efficiency for IR825 and αPD-L1 was achieved by evaluating various mass ratios of PEG-PLL, IR825, and αPD-L1 (Supplementary Table 1). Based on these results, a 12:1:5 ratio was selected for further experiments. AntagomiR-182 was subsequently adsorbed via electrostatic interaction with the polymer micelles. Transmission electronic microscopy (TEM) confirmed that IPANP exhibited a uniform spherical morphology with an average particle size of 108.5 ± 4.6 nm and good distribution (Fig. 1b and Supplementary Fig. 7a). IPANP displayed a characteristic IR825 absorption peak at 825 nm, indicating successful encapsulation (Fig. 1c). IPANP demonstrated excellent stability, with no significant changes in size or absorption spectra observed over 7 days under different conditions (Supplementary Fig. 7b, c). The binding ability of IR825-αPD-L1@PEG-PLL (IPNP) for antagomiR-182 was assessed using mass ratios ranging from 5:1 to 100:1. We found that the zeta potentials of IPANP rose with the increase of IPNP: antagomiR-182 mass ratio, which revealed the IPNP has excellent ability to bind negatively charged miRNA through electrostatic interaction (Fig. 1d). Agarose gel electrophoresis further validated complete binding of antagomiR-182 at ratios exceeding 30:1, consistent with zeta potential results (Fig. 1e). A 30:1 mass ratio was therefore selected for subsequent experiments, yielding a zeta potential of 1.5 ± 0.1 mV (Supplementary Fig. 7d). To minimize IR825-induced photothermal effect, we systematically investigated the photothermal properties of IPANP under near-infrared (NIR) laser irradiation (808 nm). Compared with free IR825, IPANP demonstrated significantly enhanced photothermal conversion efficiency (Supplementary Fig. 8a), with temperature increases correlating positively with IPANP concentrations (Supplementary Fig. 8b). IPANP solutions also exhibited power density-dependent temperature elevations under the 808 nm laser irradiation (Supplementary Fig. 8c). Notably, IPANP showed superior photothermal stability compared to IR825 or ICG solutions, maintaining consistent performance during multi-cycle laser on/off irradiation (Supplementary Fig. 8d). Based on these findings, an NIR irradiation protocol (808 nm, 1.0 W/cm², 5 min) was established for subsequent in vivo studies. This parameter was selected to sufficiently trigger the PDT effect of IPANP, as quantitatively confirmed by its singlet oxygen quantum yield (0.37%) and ROS generation assays (Supplementary Fig. 9a). Concurrently, it generates a moderate and controlled photothermal effect, ensuring a balanced therapeutic outcome. The measured photothermal conversion efficiency of 21.3% is therapeutically beneficial yet deliberately moderate (Supplementary Fig. 9b, c). This careful balance prevents the PTT effect from dominating the process and consuming excessive energy at the expense of PDT. This approach allows for an effective combination therapy while prioritizing treatment safety by minimizing potential thermal damage to surrounding healthy tissues.

**Fig. 1: Immunomodulatory effect of IPANP on myeloid dendritic cells and the induction of T cells in vitro.**

Anti-tumor efficacy and ICD effect in vitro

To evaluate the biocompatibility of IPANP, CCK-8 assays were performed on HC11 mouse mammary epithelial cells and 4T1 breast tumor cells treated with varying concentrations of the formulations at 24 h. IPANP exhibited negligible cytotoxicity even at concentrations up to 10 μg/mL, demonstrating excellent biosafety (Fig. 1f and Supplementary Fig. 10). Next, the cytotoxic effects of various formulations under NIR laser irradiation were assessed in 4T1 cells. Upon laser exposure, IPANP demonstrated significantly higher cytotoxicity than other groups at equivalent IR825 concentration, with a clear concentration-dependent effect (Fig. 1f). IPANP was specifically designed to incorporate the NIR photothermal agent IR825, capable of inducing ROS production and triggering immunogenic cell death (ICD) upon laser irradiation. To confirm this, the ROS generation capacity of IPANP was evaluated using the ROS detection probe DCFH-DA in 4T1 cells. The strongest fluorescence intensity was observed in the IPANP, followed by the laser irradiation group, highlighting its superior ROS production ability (Fig. 1g, h). To test if IR825-induced ICD could promote the release of danger-associated molecular patterns (DAMPs), the levels of adenosine triphosphate (ATP), high mobility group box 1 (HMGB1), and calreticulin (CRT) were measured in 4T1 cells following treatments with various formulations. The IPANP + Laser group showed significant increases in ATP release and HMGB1 secretion compared to other groups (Fig. 1i, j). In addition, confocal laser scanning microscopy (CLSM) images revealed markedly higher CRT exposure level in 4T1 cells treated with IPANP followed by laser irradiation (Fig. 1k, l). To further assess the immunostimulatory effect of IPANP-induced ICD and their effect on T cells, bone marrow-derived dendritic cells (BMDCs) and T cells were extracted from C57BL/6 mice (Fig. 1m). BMDCs were incubated with conditioned media from 4T1 cells pretreated with different formulations, and subsequently, T cells were added into the culture dish. Then cells were harvested and analyzed by flow cytometry. Flow cytometric analysis showed that the IPANP + Laser group significantly enhanced the expression of CD80 and CD86, promoting dendritic cell maturation (Fig. 1n–p). Additionally, flow cytometry analyses for IFN-γ and Granzyme B (GrzB) demonstrated that conditioned media-treated BMDCs in the IPANP + Laser group significantly increased cytokine production in CD8⁺ T cells (Fig. 1q, r). These results demonstrate that IR825-mediated ICD effectively promotes DC maturation, potentially enhancing CD8⁺ T cell-triggered anti-tumor immune responses.

IPANP reprograms M2-like macrophages into the M1 phenotype via activation of the TLR4/MYD88/NF-κB signaling pathway

To investigate the potential of reprogramming M2-like macrophages into M1 phenotype, BMDMs were isolated from mice (Supplementary Fig. 11), stimulated with interleukin-4 (IL-4) to induce an M2-like phenotype, and treated with various formulations (Fig. 2a). We found that miR-182 expression was markedly higher in M2-like macrophages compared to M0- or M1-like macrophages (Fig. 2b). Transcriptomic analysis of published datasets further confirmed the elevated expression of miR-182 in human M2-like macrophages (Fig. 2c). Notably, high miR-182 expression was negatively correlated with BC patient survival (Fig. 2d). These findings highlight miR-182 as a potential marker for M2-like macrophages in BC. The IPANP + Laser group exhibited significant downregulation of the M2 marker (Arg1) and upregulation of M1 marker (iNOS) compared to the polarized M2 group (Fig. 2e). ELISA results further demonstrated a decrease in M2-associated cytokines [transforming growth factor-β (TGF-β) and interleukin-10 (IL-10)] and an increase in M1-associated cytokines [tumor necrosis factor (TNF) and interleukin-12 (IL-12)] in the IPANP + Laser group compared to either antagomiR-182 inhibitor or IPANP alone (Fig. 2f, i). Flow cytometry and confocal microscopy analyses revealed that IPANP + Laser treatment significantly reduced the expression of the M2 marker CD206 while increased the expression of the M1 marker CD86 (Fig. 2j–l and Supplementary Fig. 12). The M1/M2 macrophages ratio was significantly increased (Fig. 2m), underscoring the potential of IPANP followed by laser irradiation in reprogramming M2 TAMs toward an anti-tumor M1 phenotype.

**Fig. 2: IPANP reprograms myeloid cells toward the pro-inflammatory macrophage phenotype.**

Macrophage repolarization involves complex changes in multiple signaling pathways. We further explored the underlying effect and mechanisms of IPANP, followed by laser irradiation on macrophage polarization using transcriptomic sequencing analysis. Volcano plot analysis revealed that IPANP + Laser treatment upregulated pro-inflammatory cytokine-encoding genes (e.g., RORC, IL17RA, CXCL1) and downregulated anti-inflammatory cytokine-associated genes (e.g., ARG1, MGL2, IL-33) compared to the control group (Fig. 2n). RNA sequencing analysis further showed increased expression of M1-macrophage marker genes and suppressed expression of M2-macrophage marker genes in the group of IPANP followed by laser irradiation (Fig. 2o). Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis highlighted the enrichment of genes regulated by IPANP through the Toll-like receptor (TLR) and downstream NF-κB signaling pathways (Fig. 2p–r). Western blot analysis confirmed the up-regulation of TLR4, MYD88, and p65 phosphorylation, providing direct evidence that IPANP + Laser treatment induces macrophage polarization from M2-like to M1-like phenotype through activation of the TLR4/MYD88/NF-κB pathway (Fig. 2s). Notably, this activation is not driven by extrinsic microbial ligands such as LPS; instead, IPANP acts through an endogenous gene regulatory circuit in which delivered antagomiR-182 relieves miR‑182‑mediated suppression of TLR4 expression, thereby sensitizing macrophages to endogenous signals within the tumor microenvironment and promoting MYD88‑dependent M1 polarization (Supplementary Fig. 13). Gene Ontology (GO) enrichment analysis further demonstrated that immune response-associated biological processes were significantly influenced by IPANP + Laser treatment (Supplementary Fig. 14). Collectively, these data elucidate the underlying mechanism by which IPANP reprograms macrophages and highlights its potential for modulating the TME in BC.

Biodistribution and biosafety of IPANP

The biodistribution of IPANP was evaluated in 4T1 breast tumor models following intravenous administration of different formulations. Fluorescence imaging revealed significantly stronger tumor-specific signals in mice treated with IPNP and IPANP compared to those receiving IR825 alone, with the peak signal intensity observed at 24 h post-injection (Fig. 3a). Notably, IPANP-treated mice maintained strong fluorescence signals in tumors even at 72 h, whereas IR825-treated mice exhibited consistently low tumor fluorescence, highlighting the superior tumor-targeting capability and in vivo accumulation of IPANP. Imaging of major organs and tumors 72 h post-administration further confirmed the enhanced tumor accumulation and prolonged retention of IPANP (Fig. 3b). Next, we evaluated whether IPANP could achieve spatially specific targeting within the TME using a single carrier for multi-payload delivery (Supplementary Fig. 15). αPD-L1 was predominantly associated with tumor cells, with 32.3 ± 6.9% positivity at 24 h, whereas antagomiR-182 was efficiently internalized by tumor-associated macrophages, reaching 36.5 ± 3.9% at 24 h. These data demonstrate that IPANP enables spatially specific co-delivery of distinct payloads to their intended cellular targets within the tumor microenvironment.

**Fig. 3: Biodistribution and anti-tumor efficacy of IPANP combined with laser irradiation in vivo based on primary BC tumor models.**

To assess the biosafety of IPANP in vivo, histopathological and biochemical analyses were conducted. The results showed no pathological changes were observed during the 90-day observation period, confirming the excellent safety profile of IPANP (Supplementary Fig. 16). Additionally, no inflammatory reactions or tissue damage were observed in H&E-stained sections of major organs from all groups at day 15 (Supplementary Fig. 17). Hematological examinations revealed no significant differences among groups in serum biochemical parameters (Supplementary Fig. 18). Moreover, no body weight loss was observed during the 14-day treatment period among all groups (Supplementary Fig. 19). Collectively, these results demonstrate the superior tumor-targeting capability, prolonged retention, and excellent biocompatibility of IPANP, underscoring its potential for safe and effective in vivo applications.

IPANP followed by laser irradiation inhibits cancer progression in orthotopic BC models

The therapeutic efficacy of IPANP was evaluated in 4T1 tumor-bearing mice following the treatment schedule outlined in Fig. 3c. Tumor volume in the αPD-L1 and antagomiR-182 groups increased rapidly during treatment, whereas IPANP combined with laser irradiation significantly suppressed tumor growth compared to all other groups (Fig. 3d and Supplementary Fig. 20a). In a long-term survival study, mice treated with IPANP + Laser demonstrated a 40% survival rate over a 60-day observation period, significantly outperforming other treatments (Fig. 3e and Supplementary Fig. 20b). Tumors were excised, imaged, and weighed to further evaluate treatment efficacy (Fig. 3f and Supplementary Fig. 20c). Tumor weights in the IPANP group were reduced to approximately 38.02% of the PBS group, and laser irradiation further enhanced efficacy, reducing tumor weight to 9.7% of the PBS group (Fig. 3g, h and Supplementary Fig. 20d). Histological analyses, including H&E, Ki67, CD31, and TUNEL staining, provided further evidence of treatment efficacy. Extensive tumor necrosis was observed in the IPANP + Laser group via H&E staining (Supplementary Fig. 21). Ki67 and CD31 expression were significantly reduced, indicating suppressed proliferation and angiogenesis (Fig. 3i–k). And TUNEL staining revealed markedly increased tumor apoptosis in the IPANP + Laser group (Supplementary Fig. 21). The therapeutic efficacy of IPANP + Laser was also confirmed in EMT6 tumor-bearing mice. Compared to the PBS group, IPANP combined with laser irradiation achieved complete tumor remission and significantly improved survival, with 60% treated mice fully cured (Fig. 3l–n, Supplementary Fig. 22). The tumor photographs provided direct evidence of this robust efficacy. The visible shrinkage of tumors in the IPANP + Laser group stood in stark contrast to those in the control groups (Fig. 3o). In addition, weight measurements demonstrated superior tumor regression effects (Fig. 3p, q). Compared to other groups, IPANP combined with laser irradiation showed a significant loss of Ki67-positive cells, confirming the suppression of proliferation (Fig. 3r, s). In addition, it showed a marked decline in CD31 expression, demonstrating impaired angiogenesis (Fig. 3t). Moreover, H&E staining revealed extensive tumor necrosis, and the TUNEL assay confirmed a pronounced increase in apoptotic cells (Supplementary Fig. 23). These results underscore the robust anti-tumor efficacy of IPANP + Laser in multiple orthotopic BC models. Furthermore, the broad applicability of our nanoplatform was validated in the CT26 tumor-bearing mouse model, demonstrating its potential to overcome the immunosuppressive TME in poorly immunogenic cancers (Supplementary Fig. 24, 25).

IPANP followed by laser irradiation reprograms myeloid cells and enhances anti-tumor immune responses in a BC primary tumor model

This study introduces a novel strategy to reprogram the immunosuppressive TME and achieve synergistic immunotherapy for BC (Fig. 4a). To elucidate the anti-tumor immunity mechanisms induced by IPANP combined with laser irradiation in orthotopic BC models, immune cell dynamics were analyzed using flow cytometry. Treatment with IPANP and laser irradiation significantly enhanced DC maturation in lymph nodes, a critical step in promoting CD8⁺ T cell-mediated anti-tumor immunity (Fig. 4b–e and Supplementary Fig. 26). Moreover, this treatment markedly decreased CD206 expression in both 4T1 and EMT6 tumors, with reductions of 6.8-fold and 4.5-fold, respectively, compared to the control group (Fig. 4f and Supplementary Fig. 27). Concurrently, the expression of the M1 marker CD86 and the M1/M2 ratio was significantly elevated, indicating effective reprogramming of M2-like TAMs toward the anti-tumor M1 phenotype (Fig. 4g, h). To investigate the effects on T cell immunity, T cell subtypes were analyzed using flow cytometry and immunohistochemical staining in 4T1 and EMT6 orthotopic BC models. In tumors treated with IPANP + Laser, CD8⁺ T cells increased by 14.1-fold, while regulatory T cells (T_regs) decreased by 5.9-fold compared to PBS-treated controls (Fig. 4i, j and Supplementary Fig. 28). These changes led to a markedly elevated CD8⁺/T_reg ratio, a key determinant of robust anti-tumor immune responses (Fig. 4k). To explore the mechanism behind the increased proportion of CD8⁺ T cells, we detected the expression of Ki67 and key chemokines to evaluate the proliferative and tumor-recruiting ability of CD8⁺ T cells (Supplementary Fig. 29). Flow cytometry analysis revealed significantly increased expression of Ki67 on CD8⁺ T cells. Concurrently, we observed a marked upregulation in the tumor tissue of the T cell-related chemokines CXCL9, CXCL10, and CXCL11, which are critical for directing CD8⁺ T cells to the tumor site. The above results indicate that the expansion of the CD8⁺ T cell population was driven by both enhanced local proliferation and increased recruitment. Importantly, IPANP + Laser treatment also prevented T cell exhaustion within the TME. The frequency of antigen-specific CD8⁺ T cells producing cytokines such as IFN-γ and Granzyme B increased, while the expression of inhibitory molecules decreased, including PD-1 and TIM-3 (Fig. 4l–o). Additionally, the treatment enhanced natural killer (NK) cell activation (Supplementary Fig. 30). In summary, this study demonstrates the potential of IPANP followed by laser irradiation to reprogram myeloid cells, restore T cell functionality, and remodel the immunosuppressive TME. These findings highlight the therapeutic promise of this novel strategy for synergistic BC immunotherapy.

**Fig. 4: IPANP followed by laser irradiation reprograms myeloid cells and enhances anti-tumor immune responses in a 4T1 primary tumor model.**

IPANP followed by laser irradiation induces a potent anti-tumor response in 4T1 distant tumor and lung metastasis models

To evaluate the systemic anti-tumor immune effects of IPANP, we established 4T1 distant tumor models. The therapeutic schedule is shown in Fig. 5a. Treatment with IPANP followed by laser irradiation almost completely inhibited the growth of primary tumors and significantly reduced the size of distant tumors (Fig. 5b, c and Supplementary Fig. 31). These findings were consistent with tumor photographs showing notable reductions in tumor size in the IPANP + Laser group (Supplementary Fig. 32). Flow cytometry analysis revealed that IPANP + Laser treatment led to an 11.6-fold increase in CD8⁺ T cells in primary tumors and a 10-fold increase in distant tumors compared to PBS-treated controls (Fig. 5d, e and Supplementary Fig. 33). Histological analysis using H&E staining confirmed superior treatment efficiency in the IPANP + Laser group for both primary and distant tumors (Supplementary Fig. 34). These results demonstrate that IPANP followed by laser irradiation induces robust systemic anti-tumor immune responses.

**Fig. 5: IPANP + Laser treatment induced systemic and durable anti-tumor immunity.**

Considering the high metastasis potential and poor prognosis associated with BC, we further investigated the ability of IPANP + Laser to inhibit lung metastasis using a 4T1 lung metastasis model. In brief, on day 7 post-primary tumor implantation, mice were intravenously injected with 5 × 10⁵ luciferase-labeled 4T1 cells to simulate lung metastasis and monitor its progression (Fig. 5f). Treatment with IPANP + Laser significantly prolonged survival over 60-day observation period (Supplementary Fig. 35). Lung metastases were moderately inhibited in mice treated with IPANP alone, whereas no visible metastatic lesions were detected in the lungs of mice treated with IPANP + Laser (Fig. 5g). Ex vivo bioluminescence imaging of lung tissue confirmed these findings, showing the complete absence of metastatic signals in the IPANP + Laser group (Fig. 5h, i). Additionally, photographs and H&E staining of lung tissues further corroborated the potent anti-metastatic activity of IPANP + Laser treatment (Fig. 5h, j). These results demonstrate that IPANP followed by laser irradiation effectively inhibits the growth of primary tumors, suppresses distant tumor progression, and prevents lung metastasis, highlighting its potential as a powerful systemic immunotherapeutic strategy for BC.

Tumor rechallenge study in orthotopic BC models

To evaluate the long-term effects of IPANP + Laser, tumor rechallenge models were established by inoculating tumor cells into contralateral fat pads 60 days after primary tumor treatment (Fig. 5k). Tumor growth was completely suppressed in IPANP + Laser-treated mice, whereas rapid tumor progression occurred in naïve mice (Fig. 5l and Supplementary Fig. 36a). All mice in the IPANP + Laser group survived the 60-day observation period (Fig. 5m and Supplementary Fig. 36b). Furthermore, a strong immune memory response was observed in the IPANP + Laser group, contributing to its durable anti-tumor effect (Fig. 5n, o and Supplementary Fig. 37). These results demonstrate that IPANP combined with laser irradiation exhibits superb anti-tumor efficacy, achieving long-term tumor remission and immune memory in orthotopic BC models.

IPANP followed by laser irradiation achieved complete tumor regression and inhibits lung metastasis in MMTV-PyMT transgenic mice

To closely mimic the clinicopathological characteristics of BC development and metastasis, we evaluated the therapeutic efficacy of IPANP combined with laser irradiation in MMTV-PyMT spontaneous BC mice. Once the total tumor volume across all lesions reached ~100 mm³, mice were randomly divided into six groups and administrated with indicated treatments (Fig. 6a). Tumor volume and the number of lung metastatic nodules were evaluated over an 88-day observation period after birth. Treatment with IPANP + Laser significantly reduced tumor burden compared to all other groups (Fig. 6b, c). Furthermore, the combination therapy significantly decreased the number of metastatic nodules (Fig. 6d, e). Importantly, IPANP + Laser treatment prolonged survival, achieving a 90% survival rate by the end of the study (Fig. 6f). These results confirmed the potent anti-tumor and anti-metastatic efficacy of IPANP followed by laser irradiation in MMTV-PyMT transgenic mice.

To explore the immune-modulatory effects of this combination therapy, we analyzed immune responses within the TME. Immunofluorescence staining revealed a 7.5-fold reduction in M2-like macrophages and a 6.6-fold increase in M1-like macrophages in the IPANP + Laser group compared to controls (Fig. 6g–j). Additionally, the highest levels of mature DCs were observed in both tumor and lymph node tissues following IPANP + Laser treatment, highlighting its potential to enhance T cell activation (Fig. 6k–r). Furthermore, this therapy led to a 3.1-fold increase in CD8⁺ T cell infiltration within tumors compared to the PBS group (Fig. 6s, t). Collectively, these results demonstrate that IPANP combined with laser irradiation achieves long-term tumor remission and robust inhibition of metastasis in a clinically relevant BC model. This intelligent, spatiotemporal polymer micelle system holds potential for future clinical evaluation in patients with metastatic BC.

A human tumor fragment platform validates the therapeutic efficacy and immunological response of IPANP

To evaluate the clinical translational potential of IPANP, we assessed its anti-tumor efficacy and immune-modulatory effects using a patient-derived tumor fragment (PDTF) platform⁴⁰. Briefly, BC tissues obtained from surgical resections were sectioned into small fragments (PDTFs) (~5 × 5 mm) (Fig. 7a). To minimize the influence of tumor heterogeneity, PDTFs were treated with PBS, IPANP, or IPANP + Laser for 48 h. IPANP + Laser treatment significantly enhanced the frequency of mature DCs, facilitating efficient antigen presentation compared to PBS-treated controls (Fig. 7b, c). This therapy also promoted macrophage repolarization from the M2-like to M1-like phenotype, as evidenced by decreased expression of the M2 marker CD206 and increased expression of the M1 marker CD86, resulting in a marked increase in the M1/M2 ratio (Fig. 7d–f). Furthermore, the levels of immunosuppressive cytokines TGF-β and IL-10 were significantly reduced, while proinflammatory cytokines TNF and IL-12 were upregulated following IPANP + Laser treatment (Fig. 7g–j). In addition to enhancing innate immunity, IPANP combined with laser irradiation induced higher levels of immune-stimulatory cytokines IFN-γ and Granzyme B within the TME (Fig. 7k, l). Flow cytometry analysis revealed a significant increase in CD8⁺ T cell infiltration in PDTF tumors treated with IPANP + Laser (Fig. 7m). RNA sequencing further demonstrated that this treatment upregulated anti-tumor gene expression while downregulating protumor genes (Fig. 7n). Histological analysis corroborated these findings. H&E staining revealed significant tumor cell necrosis in IPANP + Laser-treated PDTFs, confirming its potent therapeutic efficacy (Fig. 7o). Additionally, Ki67 expression, a marker of proliferation, was markedly decreased in IPANP + Laser-treated tumors compared to PBS controls. TUNEL staining showed increased tumor cell apoptosis in the IPANP + Laser group, further supporting its robust anti-tumor activity (Fig. 7o). To further validate these findings in an in vivo model that preserves TME heterogeneity and enables assessment of systemic immune interactions, we employed patient-derived xenograft (PDX) models (Fig. 7p). Consistent with the observations from patient-derived tumor fragment (PDTF) assays, IPANP combined with laser irradiation in breast cancer PDX models induced significant tumor growth inhibition (Fig. 7q, r), and promoted dendritic cell maturation, macrophage repolarization toward an M1 phenotype, enhanced CD8⁺ T cell infiltration (Supplementary Fig. 38). Collectively, these results demonstrate that IPANP plus laser treatment effectively reprograms myeloid immunity and amplifies both innate and adaptive immune responses across complementary ex vivo and in vivo platforms, thereby strengthening the translational relevance of this therapeutic strategy.

Discussion

TNBC is a clinically aggressive subtype characterized by a highly immunosuppressive TME and a strong tendency toward therapeutic resistance⁴¹. Even landmark clinical trials, such as IMpassion132, which combined the αPD-L1 antibody (atezolizumab) with chemotherapy, achieved only modest responses with a 40% overall response rate and no significant improvement in survival⁶. TNBC is now recognized as a heterogeneous group of histologically and molecularly distinct diseases driven by diverse oncogenic programs⁴². Exposure to therapy may trigger multifaceted resistance mechanisms⁴³, and this underlying complexity poses a major challenge to improve immunotherapy outcomes in patients with metastatic disease⁴⁴. These limitations underscore an urgent unmet need for innovative therapeutic strategies capable of overcoming the immunosuppressive TME and acquiring durable responses.

Traditionally, nanomedicines have primarily focused on improving the pharmacokinetics and tumor accumulation of drugs via the enhanced permeability and retention (EPR) effect. While this represents a valuable strategy, its therapeutic benefits are often constrained by physiological barriers such as high interstitial fluid pressure and dense extracellular matrix, which limit deep tumor penetration and uniform drug distribution⁴⁵. Furthermore, many systems function as passive drug carriers without the capacity to actively remodel the TME. In our study, the IPANP platform is fundamentally designed to overcome these limitations through its unique “on-demand” drug release profile and its active involvement in TME modulation. IPANP is engineered to be specifically activated by the elevated levels of ROS, a hallmark of the TME. This ROS-responsive design ensures a more tumor-specific drug release, minimizing premature leakage in circulation and enhancing the spatial precision of drug action at the target site.

Mechanistically, IPANP functions as an active orchestrator of the immunosuppressive TME rather than a passive carrier. IR825-mediated phototherapy induces immunogenic cell death, promoting dendritic cell maturation and antigen presentation, while αPD-L1 relieves adaptive immune suppression at the tumor cell interface. In parallel, antagomiR-182 reprograms M2-like tumor-associated macrophages towards an M1-like phenotype by relieving miR-182-mediated repression of the TLR4/MYD88/NF-κB axis, thereby restoring pro-inflammatory signaling. This coordinated engagement of innate and adaptive immune pathways enables synergistic immune reprogramming. Therapeutically, this integrated design translates into superior efficacy in triple-negative breast cancer models, including robust suppression of primary tumors, inhibition of metastatic progression, and durable remodeling of the immune landscape. Collectively, these features distinguish IPANP from existing anti-tumor nanomedicines and establish it as a programmable immunomodulatory platform that bridges innate and adaptive immunity to achieve sustained anti-tumor responses.

Recent studies suggest that tumor-associated inflammation fosters myeloid immunosuppression⁴⁶. Myeloid cells, which represent the most abundant hematopoietic lineage in the human body, exhibit broad functional diversity⁴⁷. Among these, DCs and tumor-promoting macrophages (such as M2-like macrophages) are terminally differentiated subsets that play pivotal roles in maintaining immune homeostasis and bridging innate and adaptive immunity⁴⁸. Our scRNA-seq and flow cytometry analyses revealed a marked enrichment of immature DCs and M2-like macrophages in both primary and metastatic tumors in breast cancer patients and MMTV-PyMT transgenic mice. These findings align with previous studies highlighting the tumor-promoting roles of myeloid populations, including in angiogenesis, invasion, and metastasis⁴⁹. Although these cells are classically involved in pathogen defense, tissue remodeling, and apoptotic cell clearance, the TME can reprogram them into potent immunosuppressive phenotypes⁴⁷. This functional plasticity presents a unique therapeutic opportunity to redirect the fate of tumor-supportive myeloid cells toward immunostimulatory phenotypes, potentially eliminating immunosuppressive barriers and unleash robust anti-tumor immunity.

To date, most studies have examined myeloid subsets in isolation, leading to fragmented mechanistic insight and suboptimal therapeutic efficacy^50,51,52. In contrast, our study demonstrates a programmable nanomicelle platform (IPANP) that functionally reconfigures myeloid immunity through dual-lineage reprogramming, simultaneously promoting DC maturation and polarizing M2-like macrophages toward an M1 phenotype. This coordinated remodeling of the immune microenvironment enhances CD8⁺ T cell infiltration, activation, and effector function, ultimately yielding durable anti-tumor responses.

In tumors, DCs respond to DAMPs such as ATP, HMGB1, and CRT released during ICD⁵³. These signals promote DC maturation and trafficking to lymphoid organs, where they initiate T cell priming. As such, DCs play a central role in anti-tumor immunity and in mediating the efficacy of ICB. Advances in nanotechnology have enabled the development of photosensitizer immunoconjugates⁵⁴, which synergize immune activation and tumor destruction by simultaneously inducing ICD and potentiating ICB²⁵. These hybrid systems hold promise for achieving spatiotemporally precise activation of DCs and effector T cells. Our study revealed that immature DCs stimulated by photo-activable immunoconjugates transitioned from antigen-sampling to antigen-presenting modes. Combined with immune checkpoint blockade, this nanoplatform activated effector T cells and induced robust anti-tumor immunity.

TAMs, closely related to DCs, also exert profound effects on immune suppression and disease progression⁵⁵. Their polarization from an M2-like to an M1-like phenotype has emerged as a compelling strategy to reprogram the TME. Although inhibitors targeting signaling pathways such as PI3Kγ are being evaluated clinically³³, their efficacy and safety remain to be fully established. This is due, in part, to the complexity and redundancy of macrophage regulatory circuits⁴⁷. Emerging evidence points to the critical roles of microRNAs in governing macrophage polarization and function⁵⁶. For instance, miR-182 is among the most consistently upregulated oncogenic miRNAs across a spectrum of cancers, including breast, lung, glioma, melanoma, and ovarian malignancies⁵⁷. While previous studies have primarily investigated miR-182 in tumor cells, its immunomodulatory role within the TME remains poorly characterized. Our findings identify miR-182 as a highly expressed suppressive factor in M2-like TAMs in breast cancer tissues. We designed a ROS-responsive polymeric nanocarrier to co-deliver antagomiR-182, which reprograms M2-like macrophages towards an M1-like phenotype primarily through activation of the TLR4/MYD88/NF-κB axis. Notably, this signaling pathway operates within a broader regulatory network and engages in crosstalk with other pathways implicated in macrophage polarization. TLR4 activation can indirectly engage the PI3K/AKT pathway, while functional interactions between TLR4/MYD88/NF-κB and STAT3 signaling have been reported, with sustained STAT3 activity promoting M2 polarization. Such pathway convergence may contribute to the robustness and stability of the observed macrophage repolarisation. Functionally, this reprogramming alleviates immunosuppression within the TME and enhances CD8⁺ T cell infiltration, thereby reinforcing cytotoxic anti-tumor immunity.

While our long-term biosafety assessments in murine models, including histopathological examination and longitudinal blood biochemistry over 90 days, did not reveal detectable tissue damage or systemic toxicity following IPANP administration. Although the results indicate favorable safety profiles under the tested conditions, systemic immune modulation carries inherent risks of delayed or context-dependent adverse effects, such as sustained inflammation or off-target immune activation, which may not be fully captured in mouse studies. Therefore, further extended safety evaluations in larger animal models and ultimately in carefully designed clinical trials remain essential to comprehensively assess long-term immune homeostasis and potential translational risks.

To assess clinical relevance, we employed a PDTF platform, a novel ex vivo system that retains the architectural and immunological features of freshly resected tumors⁴⁰. Unlike traditional organoid models that require extensive dissociation and regrowth, PDTFs preserve native tissue structures and immune infiltrates under short-term culture. This enables comparative testing of multiple immunotherapies within matched patient-derived samples and facilitates mechanistic readouts through transcriptomic and proteomic profiling. In this study, IPANP plus laser activation reprogrammed myeloid cells, boosted innate and adaptive immune responses, and produced superior therapeutic effects in PDTFs. While this model does not recapitulate systemic immune interactions, it provides a valuable intermediate system bridging preclinical efficacy with translational relevance. To further validate these findings in a model that preserves the tumor microenvironmental heterogeneity and allows for the study of systemic immune interactions, we employed patient-derived xenograft (PDX) models. Consistent with the PDTF results, IPANP + Laser treatment in breast cancer PDX models promoted dendritic cell maturation, macrophage repolarization toward an M1 phenotype, enhanced CD8⁺ T cell infiltration, and induced significant tumor growth inhibition.

Of note, the clinical translation of nanoplatforms faces several major barriers, including suboptimal targeting efficiency, concerns regarding nanoparticle safety, limitations in scalability, and the insufficient use of preclinical models that faithfully recapitulate human tumors. In particular, scalable and reproducible manufacture represents a critical hurdle for the clinical adoption of complex, multi-component nanosystems. Accordingly, several advanced strategies have been developed to facilitate large-scale production while maintaining functional integrity. For instance, modular approaches such as the two-vial system allow separate preparation of universal carrier components and active pharmaceutical ingredients, enabling bedside self-assembly and simplifying mass production⁵⁸. Meanwhile, techniques such as microfluidics and flash nano-precipitation offer precise control over nanoparticle characteristics and are highly adaptable to various payloads, including hydrophobic drugs and nucleic acids⁵⁹. To ensure consistent product quality, the implementation of a Quality by Design (QbD) framework early in development can systematically optimize critical process parameters and facilitate robust, clinically viable manufacturing⁶⁰. Therefore, it is essential to integrate scalability considerations at the initial stages of nanoplatform design to successfully bridge the gap between laboratory innovation and clinical application.

In conclusion, we present a rationally designed, multifunctional nanomedicine platform that reprograms immunosuppressive myeloid cells to overcome the immunological barriers of the TME (Supplementary Fig. 39). The ROS-responsive polymeric nanomicelles integrate phototherapy, immune checkpoint inhibition, and epigenetic modulation via antagomiR-182 to simultaneously activate DCs and repolarize TAMs. This dual reprogramming reshapes the immune landscape and elicits durable anti-tumor immunity across multiple preclinical models, including MMTV-PyMT mice and patient-derived xenograft models. Furthermore, the efficacy of this approach is validated in a clinically relevant ex vivo human tumor model. These findings establish IPANP as a promising platform for breast cancer immunotherapy and suggest broader potential in targeting immunosuppressive TMEs in other solid tumors.

Methods

Patient samples

All experiments using human samples have received approval from the Zhongda Hospital Southeast University Ethics Committee and the Fourth Hospital of Harbin Medical University Ethics Committee according to related ethical guidelines. All patients consented to use biospecimens for research. Detailed patient characteristics are summarized in (Supplementary Table 2). Paraffin-embedded BC samples were acquired from Zhongda Hospital Southeast University for immunofluorescence staining. To compare immune cell infiltration in BC with non-tumor tissues, fresh BC specimens and their adjacent normal tissues were obtained from the Fourth Hospital of Harbin Medical University. In addition, we also evaluated the effects of nanoformulations in resected human BC samples. Tumor tissues were processed and incubated as described previously^61,62,63. In brief, freshly surgically resected human BC specimens were cultured in ice-cold RPMI-1640 medium containing 10% FBS and 1% penicillin-streptomycin, and then separated into small tumor fragments (PDTFs) of approximately 5 × 5 mm length. Each part was placed in a 6-well plate and incubated with PBS, IPANP, or IPANP + Laser at the same concentration of antagomiR-182 (25 μM) and treated for 48 h. Laser (808 nm, 1.0 W/cm², 5 min) was applied after administrations with IPANP for 1 h. Subsequently, flow cytometry, ELISA, histological analysis, and transcriptomic sequencing were performed to assess the therapeutic efficacy and anti-tumor immune response.

Cell culture and animals

The BC cell lines 4T1 (catalog no. CL-0007) and EMT6 (catalog no. CL-0573) were purchased from Procell Life Science & Technology Co., Ltd (Wuhan, China). HC11 cells (catalog no. ZQ0533) were purchased from Zhong Qiao Xin Zhou Biotechnology Co., Ltd (Shanghai, China). Each cell line was tested as mycoplasma-free. Furthermore, 4T1 cells and EMT6 cells were cultured and maintained in RPMI 1640 culture medium, while HC11 cells were cultured and maintained in DMEM. Both media were supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin. All cells were incubated in an atmosphere at 37 °C with 5% CO₂.

BALB/c mice (5–7 weeks old, female) and C57BL/6 mice (8 weeks old, male) were purchased from the Comparative Medical Center of Yangzhou University (Yangzhou, China). MMTV-PyMT mice (9–10 weeks old, female, strain NO.C001212) with spontaneous tumors were purchased from Cyagen (Suzhou, China). 6-week-old female NCG mice (Strain NO.T001475) and huHSC-NCG-M female mice (Strain NO.T056601) were purchased from GemPharmatech (Nanjing, China). The mice were co-housed in the specific pathogen-free (SPF) care facility with free access to food and water under a 12 h light/dark cycle. All mice were maintained at 22–26 °C with humidity levels between 30 and 50%. All experimental and control animals were co-housed. All animal studies were approved by the Animal Care & Welfare Committee of Southeast University. Tumor size was measured daily using calipers, and tumor volume was calculated using the formula: length×width²/2. Mice were euthanized by cervical dislocation in accordance with the approved animal protocol. In accordance with institutional ethical guidelines, the maximal tumor burden permitted was 1500 mm³ for subcutaneous and orthotopic tumors or a palpable diameter exceeding 1.5 cm. No experimental animals exceeded this limit during the study.

Flow cytometry

The single-cell suspensions from tumors, tumor-draining lymph nodes (TDLNs), or spleens of the BALB/c mice and PyMT mice were prepared as previously mentioned⁶⁴. Similarly, PDTFs were manually took out from the matrix after incubation for 48 h, and prepared as the single-cell suspensions⁶¹. Briefly, tissues were shredded and digested at 37 °C for 1 h. The cells were then filtered with 70-μm cell strainers. Subsequently, Fc receptors were blocked with CD16/CD32 antibody (eBioscience) for 10 min on ice, and incubated with fixable viability dye (eBioscience) for 20 min at 4 °C. Next, samples were stained with fluorescent dye-labeled antibodies purchased from eBioscience, including anti-mouse CD45-EF506 (30-F11, 69-0451-82, 0.5 μg/10⁶ cells), anti-mouse CD11b-PerCP/Cy5.5 (M1/70, 45-0112-82, 0.25 μg/10⁶ cells), anti-mouse F4/80-FITC (BM8, 11-4801-82, 0.5 μg/10⁶ cells), anti-mouse CD206-PE (MR6F3, 12-2061-82, 0.125 μg/10⁶ cells), anti-mouse CD86-APC (GL1, 17-0862-82, 0.06 μg/10⁶ cells), anti-mouse CD11c-FITC (N418, 11-0114-81, 0.25 μg/10⁶ cells), anti-mouse MHC-II-SB436 (M5/114.15.2, 62-5321-82, 0.125 μg/10⁶ cells), anti-mouse MHC-I-PE-Cy7 (28-14-8, 25-5999-82, 1.0 μg/10⁶ cells), anti-mouse CD40-PE-Cy5 (1C10, 15-0401-82, 0.25 μg/10⁶ cells), anti-mouse CD80-PE (16-10A1, 12-0801-81, 0.06 μg/10⁶ cells), anti-mouse CD3-PerCP/Cy5.5 (145-2C11, 45-0031-82, 1 μg/10⁶ cells), anti-mouse CD4-PE (GK1.5, 12-0041-81, 0.125 μg/10⁶ cells), anti-mouse CD8-FITC (53-6.7, 11-0081-82, 0.5 μg/10⁶ cells), anti-mouse CD25-APC (PC61.5, 17-0251-82, 0.125 μg/10⁶ cells), anti-mouse Foxp3-EF450 (FJK-16s, 48-5773-80, 0.06 μg/10⁶ cells), anti-mouse CD11b-BV421 (M1/70, 404-0112-80, 0.125 μg/10⁶ cells), anti-mouse Ly6G-PE (1A8-Ly6g, 12-9668-80, 0.125 μg/10⁶ cells), anti-mouse Ly6C-APC (HK1.4, 17-5932-82, 0.125 μg/10⁶ cells), anti-mouse CD44-APC (IM7, 17-0441-81, 0.06 μg/10⁶ cells), anti-mouse CD62L-PE (MEL-14, 12-0621-81, 0.125 μg/10⁶ cells), anti-mouse PD-1-APC (RMP1-30, 17-9981-82, 1 μg/10⁶ cells), anti-mouse TIM-3-PE (RMT3-23, 12-5870-82, 0.5 μg/10⁶ cells), anti-mouse IFN-γ-PE (XMG1.2, 12-7311-82, 0.25 μg/10⁶ cells), anti-mouse Granzyme B-EF660 (NGZB, 50-8898-82, 0.125 μg/10⁶ cells), anti-mouse CD49-PE/Cy7 (DX5, 25-5971-82, 0.5 μg/10⁶ cells), anti-mouse CD80-FITC (16-10A1, 11-0801-82, 0.25 μg/10⁶ cells), anti-human CD45-EF506 (HI30, 69-0459-42, 0.5 μg/10⁶ cells), anti-human CD3-PerCP/Cy5.5 (OKT3, 45-0037-42, 0.25 μg/10⁶ cells), anti-human CD4-PE (RPA-T4, 12-0049-42, 0.5 μg/10⁶ cells), anti-human CD8-FITC (RPA-T8, 11-0088-42, 1 μg/10⁶ cells), anti-human CD11b-FITC (ICRF44, 11-0118-41, 0.5 μg/10⁶ cells), anti-human CD206-PE (19.2, 12-2069-42, 0.125 μg/10⁶ cells), anti-human CD86-APC (IT2.2, 17-0869-42, 0.25 μg/10⁶ cells), anti-human CD68-FITC (Y1/82 A, 11-0689-42, 0.125 μg/10⁶ cells), anti-human CD80-PE (2D10.4, 12-0809-42, 0.5 μg/10⁶ cells), anti-human HLA-DR-SB436 (LN3, 62-9956-42, 0.125 μg/10⁶ cells), anti-human CD11c-FITC (3.9, 11-0116-42, 1 μg/10⁶ cells). Antibodies were applied for staining according to the manufacturer’s suggestions. Finally, cells were washed and analyzed by Attune NxT flow cytometry (Thermo Fisher Scientific). The data were analyzed and quantified by the FlowJo V10 software (BD Biosciences). For intracellular staining, cells were fixed, permeabilized, and then cultured with antibodies, such as CD206-PE or Foxp3-EF450. For detecting anti-tumor cytokine production (IFN-γ and granzyme B) from T cells, cells were stimulated with a cell stimulation cocktail (00-4975-93, eBioscience) for 6 h. Cells were then stained with the specific antibodies. The gating strategies were displayed in Supplementary Fig. 40.

Single-cell RNA-seq library preparation and sequencing

To assess the immunological landscape of the TNBC microenvironment, scRNA-seq was conducted on six fresh tissue samples obtained from three TNBC patients. These comprised three primary tumors surgically removed along with their paired adjacent normal tissues. After quality filtering, transcriptomic data were derived from 45,620 high-quality cells, which included endothelial cells, fibroblasts, lymphoid cells, epithelial cells, myeloid cells, and pericytes. Library preparation was carried out in accordance with the manufacturer’s protocol using the Single Cell 3′ Library and Gel Bead Kit V3.1. Subsequently, sequencing was performed on an Illumina Novaseq6000 sequencer with a sequencing depth of at least 100,000 reads per cell with a pair-end 150 bp (PE150) strategy (performed by CapitalBio Technology, Beijing).

Fluorescence in situ hybridization (FISH)

The level of miR-182 was analyzed by the Cy3-miR-182 probe (F07101, GenePharme) and miR-182 FISH Kit (F04501, GenePharme) in paraffin-embedded BC tissues according to the manufacturer’s protocol.

ELISA

Enzyme linked immunosorbent assay (ELISA) kits were applied to evaluate the cytokine levels of murine TGF-β (KE10005, Proteintech), murine IL-10 (E-EL-M0046c, Elabscience), murine TNF (E-EL-M3063, Elabscience), murine IL-12 (KE10014, Proteintech), murine HMGB1 (E-EL-M0676, Elabscience), human TGF-β (E-EL-0162, Elabscience), human IL-10 (E-EL-H6154, Elabscience), human IL-12 (E-EL-H0150, Elabscience), human TNF (E-EL-H0109, Elabscience), human IFN-γ (E-EL-H0108, Elabscience) and human granzyme B (E-EL-H1617, Elabscience), in accordance with the manufacturer’s instructions.

Macrophage polarization

Bone marrow-derived macrophages (BMDMs) were extracted from the tibias and femurs of C57BL/6 mice (8 weeks old, male). BMDMs were incubated in DMEM with 10% FBS and recombinant murine macrophage colony-stimulating factor (M-CSF, 10 ng/mL, AF-315-02, PeproTech) for 7 days. Next, BMDMs were seeded into 6-well plates at a density of 2 × 10⁶ cells/mL and stimulated with IL-4 (20 ng/ml) to M2 phenotype for 48 h. In addition, to explore the potential of IPANP + Laser for reprograming M2 macrophages to M1, M2 BMDMs were treated with prepared formulations for 24 h. Next, M2-associated markers (CD206, Arg1, IL-10, TGF-β) and M1-associated markers (CD86, iNOS, IL-12, TNF) were analyzed by flow cytometry, immunofluorescence, western blot (WB), and ELISA.

RNA extraction and quantitative real-time PCR

Total RNA was extracted with TRIzol reagent (Invitrogen). Normalized RNA was reverse transcribed by the HiScript II Q RT SuperMix for qPCR Kit (Vazyme). Then, qRT-PCR was conducted by the ChamQ Universal SYBR qPCR Master Mix (Vazyme) according to the manufacturer’s instructions. The data were normalized to U6 expression and analyzed through the 2^−ΔΔCt method. The primers were generated by Genewiz (Suzhou, China), and the sequences were provided in Supplementary Table 3.

Cytotoxicity evaluation

The CCK-8 assay was used to evaluate the cytotoxicity of various formulations. Briefly, 4T1 and HC11 cells were seeded into 96-well plates at a density of 1 × 10⁴ cells/well for 12 h. The cells were treated with the formulations at different concentrations and incubated for 24 h. For the laser groups, the NIR irradiation was applied 4 h after different treatments (808 nm, 1.0 W/cm², 5 min). Next, 20 µL CCK-8 solution (Vazyme) was added and cultured for 4 h. Finally, the absorbance at 450 nm was determined with a microplate reader (Thermo Fisher Scientific) to evaluate the cell viability.

Intracellular ROS detection

4T1 cells were seeded into 12-well plates and incubated overnight. Cells were treated with different formulations for 4 h. The cells were then irradiated (808 nm, 1.0 W/cm², 5 min) and cultured overnight. Next, cells were cultured with 2.7-dichlorofluoresceindiacetate (DCFH-DA) to monitor the intracellular ROS level following the manufacturer’s protocol. Finally, cells were washed with PBS and imaged by CLSM.

In vitro ICD induction

To assess the ICD effects after treatment with different formulations, 4T1 cells were seeded in a confocal dish at a density of 1 × 10⁵ cells/mL overnight. Then the cell culture medium was replaced with fresh medium containing various formulations at a final concentration of 5 μg/mL of IR825. After 4 h, cells were irradiated with 808 nm laser (1.0 W/cm²) for 5 min and cultured for 24 h. Next, the supernatant was removed. And the cells were washed with cold PBS and then cultured with anti-CRT monoclonal antibody (12238, D3E6, Cell Signaling Technology, 1:200) overnight at 4 °C and stained with fluorescent-labeled secondary antibody for 1 h at room temperature. The cell nuclei were stained with DAPI for 5 min and further observed by CLSM. In addition, to detect the secretion of HMGB1 and ATP, HMGB1 ELISA kits and ATP assay kits were used according to the manufacturer’s instructions. To further assess the ICD effects on DCs and their impact on T cell activation, bone marrow-derived dendritic cells (BMDCs) were generated from C57BL/6 mice. In brief, bone marrow was flushed from femurs and tibias, and the cells were cultured in RPMI 1640 medium supplemented with 20 ng/mL GM-CSF for 7 days. The spleen was mechanically dissociated and enzymatically digested, followed by isolation of naïve T cells via magnetic bead-based sorting. Next, BMDCs were treated with different formulation-pretreated 4T1 cell culture mediums for 24 h. Subsequently, T cells were collected and co-cultured for an additional 72 h. The co-culture was performed in recombinant mouse IL-2 and incubated in a 37 °C incubator with 5% CO₂. Then the cells were harvested and stained with anti-CD11c, anti-CD86, anti-CD80, anti-CD3, anti-CD8, anti-IFN-γ, and anti-Granzyme B antibodies. Finally, the data were measured and analyzed by flow cytometry.

Western blot

Cells were lysed by the RIPA lysis buffer containing protease and phosphatase inhibitors. The extracted proteins were separated by SDS-PAGE and then transferred onto PVDF membranes. After blocking with 5% fat-free milk, the membranes were incubated with primary antibodies at 4 °C overnight, followed by corresponding HRP-conjugated secondary antibodies. Finally, the protein bands were visualized by the enhanced chemiluminescence (ECL) detection reagent. The antibodies used were as follows: anti-iNOS antibody (ab205529, EPR16630, Abcam, 1:1000), anti-Arg1 antibody (93668, D4E3M, Cell Signaling Technology, 1:1000), anti-TLR4 antibody (14358, D8L5W, Cell Signaling Technology, 1:1000), anti-MYD88 antibody (ab219413, EPR21824, Abcam, 1:1000), anti-P65 antibody (8242, D14E12, Cell Signaling Technology, 1:1000), anti-p-P65 antibody (3033, Ser536, Cell Signaling Technology, 1:1000), anti-GAPDH antibody (2118, 14C10, Cell Signaling Technology, 1:1000).

Preparation and characterization of IPANP

IPANP was synthesized using PEG-PLL, ROS-cleavable linker TK, IR825, αPD-L1, and antagomiR-182. Firstly, IR825-αPD-L1 was prepared through mixing αPD-L1 (1 mg, 1 mg/mL, BE0101, 10 F.9G2, BioXcell) and IR825-NHS (200 µg, 1 mg/mL, Xi’an Ruixi Biological Technology Co., Ltd), and kept at 4 °C overnight. Then the mixture solution was purified using a dialysis bag (MWCO: 3 kDa). Similarly, anti-rat IgG conjugated with IR825 was formulated as a control. Next, PEG-PLL (0.24 mg, 1 mg/mL) and antagomiR-182 or negative control (0.12 mg, 15 nmol) were stirred overnight at 4 °C. The ROS-cleavable linker TK (20 µg, 1 mg/mL) was then added to the resulting solution. Subsequently, IR825-αPD-L1 was added to the above mixture and then stirred for 4 h at room temperature. Finally, the unreacted drug was removed using a dialysis bag (MWCO: 100 kDa) with PBS (pH = 7.4). The preparation methods of IPNP and ANP were the same as described above, except that the lack of other drugs. The morphology of the IPANP was characterized using transmission electron microscopy (HITACHI HT7800, Hitachi High Technologies, Japan). The hydrodynamic diameter and zeta potential were measured by dynamic light scattering (DLS) using a 90Plus Particle Size Analyzer (Malvern Instruments Ltd., UK).

Biodistribution of IPANP

To assess the biodistribution, various formulations (at an IR825-equivalent dose of 0.5 mg/kg) were administered intravenously via the tail vein to 4T1 tumor-bearing BALB/c mice (5–7 weeks old, female, n = 3 mice per group). The mice were then imaged at multiple time points using an IVIS-Spectrum system (Perkin Elmer, Santa Clara, CA). For ex vivo analysis, the animals were euthanized 72 h post-injection. Tumors and major organs were harvested for imaging. To investigate the cellular populations conferred by each payload within the IPANP delivery system, IPANP was prepared using Cy5-labeled αPD-L1 or Cy5-labeled antagomiR-182, respectively. 4T1 tumor-bearing mice were administered each probe via intravenous injection. Tumors were harvested at 4 and 24 h post-injection, dissociated into single-cell suspensions, and stained with antibodies for flow cytometry analysis.

Anti-tumor effect in orthotopic BC mice

The orthotopic BC models were established in female BALB/c mice (5–7 weeks old) as described above. In brief, 4T1 or EMT6 cells were implanted into the mammary fat pad of mice (2 × 10⁶ cells/ml). When the tumor volume reached 50–100 mm³, mice were randomly divided into 10 groups (n = 16). The mice were then treated with PBS, IR825, αPD-L1, antagomiR-182, IPNP, ANP, IPANP, IR825 + Laser, IPNP + Laser, and IPANP + Laser. The mice were intravenously injected with different formulations on day 1, 5, and 9. For the laser groups, tumors were irradiated with an 808 nm laser (1.0 W/cm²) for 5 min on days 2, 6, and 10. Free IR825 was intravenously injected at 0.5 mg/kg. Free αPD-L1 was injected intravenously at 2.5 mg/kg. Free antagomiR-182 was injected intravenously at 2 mg/kg. IPNP, ANP, and IPANP were intravenously administered at the αPD-L1 and antagomiR-182 dosages of 2.5 mg/kg and 2 mg/kg, respectively. The tumor volume and body weight were monitored and recorded every 2 days. Meanwhile, the survival rate of mice was monitored throughout the experiment. On the 15th day after treatment, six mice of each group were euthanized, and the tumors and lymph nodes were resected. Additionally, the tumors were weighed and imaged. The tumor growth inhibition (TGI) was calculated based on the tumor weight. The tumors were collected for further analysis, including hematoxylin and eosin (H&E) staining, TUNEL assay, immunohistochemical staining, immunofluorescence staining, and flow cytometry analysis. And lymph nodes were used for flow cytometry analysis.

In vivo synergistic therapeutic against lung metastasis

When the tumor volume of 4T1 tumor-bearing mice reached 50–100 mm³, female BALB/c mice (5–7 weeks old) were randomly divided into PBS, IR825, αPD-L1, antagomiR-182, IPNP, ANP, IPANP, IR825 + Laser, IPNP + Laser, and IPANP + Laser groups (n = 13). Then, 2 × 10⁵ 4T1-luc cells were intravenously injected to form lung metastases. The mice were treated with different formulations on days 1, 5, and 9. For the laser groups, primary tumors were irradiated with 808 nm laser (1.0 W/cm²) for 5 min on days 2, 6, and 10. Free IR825 was intravenously injected at 0.5 mg/kg. Free αPD-L1 was injected intravenously at 2.5 mg/kg. Free antagomiR-182 was injected intravenously at 2 mg/kg. IPNP, ANP, and IPANP were intravenously administered at the αPD-L1 and antagomiR-182 dosages of 2.5 mg/kg and 2 mg/kg, respectively. The mice were monitored for 3 weeks. On day 21, lungs were harvested, and the metastatic foci were counted. The survival rate of mice was monitored throughout the experiment.

Tumor rechallenge study

To evaluate the long-term effects of IPANP + Laser, tumor rechallenge models were established by inoculating tumor cells into contralateral fat pads 60 days after primary tumor treatment. At the same time, naïve mice were injected with the same dose of 4T1 cells as a control. Tumor volume and survival were recorded. Additionally, to assess the immune memory response, the frequencies of central memory (CD44^highCD62L^high) and effector memory (CD44^highCD62L^low) CD8⁺ T cells in the TDLN, spleen, and blood were analyzed by flow cytometry.

Anti-tumor effect in MMTV-PyMT transgenic mice

MMTV-PyMT mice (9–10 weeks old, female) with spontaneous tumors were randomly assigned to six groups (n = 13), and the total tumor size of each group was in the range of 80–110 mm³. The mice were then treated with PBS, αPD-L1, antagomiR-182, IPNP, IPANP, and IPANP + Laser. Free αPD-L1 was injected intravenously at 2.5 mg/kg three times on days 71, 75, and 79 after birth. IPNP and IPANP suspended in sterile saline were intravenously administered at αPD-L1 and antagomiR-182 dosages of 2.5 and 2 mg/kg, respectively. And irradiated with 808 nm laser (1.0 W/cm²) for 5 min on days 72, 76, and 80 after birth. Body weight, tumor number, and tumor volume were measured and recorded every 2 days. 9 days after the last administration, five mice from each group were euthanized, and the tumor and lymph nodes were collected for flow cytometry analysis. On day 88, five mice from each group were euthanized, and the lungs were harvested, washed, and fixed in the paraformaldehyde. After repeated rinsing, the lung organs with metastatic tumor nodules were photographed. Then, the excised lung organs were fixed, sectioned, and stained with H&E, followed by imaging using an inverted fluorescence microscope to investigate further inhibition of tumor lung metastasis.

Patient-derived xenograft (PDX) generation

Patient-derived xenograft (PDX) tumor models were established using 6-week-old female NCG mice. Fresh tumor specimens were dissected into approximately 2 × 2 ×2 mm³ fragments under sterile conditions. Following local hair removal with depilatory cream, tumor fragments were surgically implanted into the dorsal flank of recipient mice. After wound closure by suturing and disinfection, the resulting tumors were designated as the first passage (F1). When F1 tumors reached a volume of approximately 600-800 mm³, mice were euthanised, and tumors were excised for serial passaging. Necrotic regions were removed, and viable tumor tissue was sectioned into small fragments for implantation into a new cohort of NCG mice to generate F2 tumors. This process was repeated to obtain F3 tumors. For immunological evaluation, F3 tumors were transplanted into 6-week-old humanized Hu-HSC mice (NCG mice reconstituted with human CD34⁺ haematopoietic stem cells), which harbor multi-lineage human immune cells. Following tumor establishment, therapeutic interventions were administered, and tumor growth and immune responses were monitored as described.

Statistical analysis

All statistical analyses were performed by GraphPad Prism 8.0. Statistical evaluation was performed with the Student’s t test and One-way or two-way analysis of variance (ANOVA). Survival was analyzed using log-rank tests. All values were described as the means ± standard deviation (SD), and differences were considered as statistically significant at P < 0.05.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The miRNA-seq datasets analysed in this study were obtained from the GEO database under accession code GSE51307. The single-cell RNA-sequencing data have been deposited in Genome Sequence Archive (GSA) under the accession number PRJCA040777 [https://ngdc.cncb.ac.cn/gsa-human/browse/HRA011711]. All data are included in the Article and Supplementary Information. The raw numbers for charts and graphs are available in the Source Data file whenever possible. Source data are provided with this paper.

References

Siegel, R. L., Giaquinto, A. N. & Jemal, A. Cancer statistics, 2024. CA A Cancer J. Clin. 74, 12–49 (2024).
Google Scholar
Early Breast Cancer Trialists’ Collaborative Group Reductions in recurrence in women with early breast cancer entering clinical trials between 1990 and 2009: a pooled analysis of 155 746 women in 151 trials. Lancet 404, 1407–1418 (2024).
Article Google Scholar
Barbie, T. U. et al. Targeting an IKBKE cytokine network impairs triple-negative breast cancer growth. J. Clin. Investig. 124, 5411–5423 (2014).
Article PubMed PubMed Central Google Scholar
Bachelot, T. et al. Durvalumab compared to maintenance chemotherapy in metastatic breast cancer: the randomized phase II SAFIR02-BREAST IMMUNO trial. Nat. Med. 27, 250–255 (2021).
Article PubMed CAS Google Scholar
Jiang, Z. et al. Toripalimab plus nab-paclitaxel in metastatic or recurrent triple-negative breast cancer: a randomized phase 3 trial. Nat. Med. 30, 249–256 (2024).
Article PubMed CAS Google Scholar
Dent, R. et al. IMpassion132 double-blind randomised phase III trial of chemotherapy with or without atezolizumab for early relapsing unresectable locally advanced or metastatic triple-negative breast cancer. Ann. Oncol. Off. J. Eur. Soc. Med. Oncol. 35, 630–642 (2024).
Article CAS Google Scholar
van Vlerken-Ysla, L., Tyurina, Y. Y., Kagan, V. E. & Gabrilovich, D. I. Functional states of myeloid cells in cancer. Cancer cell 41, 490–504 (2023).
Article PubMed PubMed Central Google Scholar
Harris, M. A. et al. Towards targeting the breast cancer immune microenvironment. Nat. Rev. Cancer 24, 554–577 (2024).
Article PubMed CAS Google Scholar
Güç, E. & Pollard, J. W. Redefining macrophage and neutrophil biology in the metastatic cascade. Immunity 54, 885–902 (2021).
Article PubMed Google Scholar
Wculek, S. K. et al. Dendritic cells in cancer immunology and immunotherapy. Nat. Rev. Immunol. 20, 7–24 (2020).
Article PubMed CAS Google Scholar
Wang, Y., Johnson, K. C. C., Gatti-Mays, M. E. & Li, Z. Emerging strategies in targeting tumor-resident myeloid cells for cancer immunotherapy. J. Hematol. Oncol. 15, 118 (2022).
Article PubMed PubMed Central Google Scholar
Dunn, G. P., Bruce, A. T., Ikeda, H., Old, L. J. & Schreiber, R. D. Cancer immunoediting: from immunosurveillance to tumor escape. Nat. Immunol. 3, 991–998 (2002).
Article PubMed CAS Google Scholar
Cassetta, L. et al. Human tumor-associated macrophage and monocyte transcriptional landscapes reveal cancer-specific reprogramming, biomarkers, and therapeutic targets. Cancer cell 35, 588–602.e510 (2019).
Article PubMed PubMed Central CAS Google Scholar
Nian, Z. et al. Interleukin-34-orchestrated tumor-associated macrophage reprogramming is required for tumor immune escape driven by p53 inactivation. Immunity 57, 2344–2361.e2347 (2024).
Article PubMed CAS Google Scholar
Tie, Y., Tang, F., Wei, Y. Q. & Wei, X. W. Immunosuppressive cells in cancer: mechanisms and potential therapeutic targets. J. Hematol. Oncol. 15, 61 (2022).
Article PubMed PubMed Central CAS Google Scholar
Gogoleva, V. S., Mundt, S., De Feo, D. & Becher, B. Mononuclear phagocytes in autoimmune neuroinflammation. Trends Immunol. 45, 814–823 (2024).
Article PubMed CAS Google Scholar
Mempel, T. R., Henrickson, S. E. & Von Andrian, U. H. T-cell priming by dendritic cells in lymph nodes occurs in three distinct phases. Nature 427, 154–159 (2004).
Article ADS PubMed CAS Google Scholar
Jin, S. M. et al. A nanoadjuvant that dynamically coordinates innate immune stimuli activation enhances cancer immunotherapy and reduces immune cell exhaustion. Nat. Nanotechnol. 18, 390–402 (2023).
Article ADS PubMed CAS Google Scholar
Bosteels, V. & Janssens, S. Striking a balance: new perspectives on homeostatic dendritic cell maturation. Nat. Rev. Immunol. 25, 125–140 (2024).
Article PubMed Google Scholar
Liu, G. et al. Bacteria-derived nanovesicles enhance tumour vaccination by trained immunity. Nat. Nanotechnol. 19, 387–398 (2024).
Article ADS PubMed CAS Google Scholar
Bošnjak, B., Do, K. T. H., Förster, R. & Hammerschmidt, S. I. Imaging dendritic cell functions. Immunol. Rev. 306, 137–163 (2022).
Article PubMed Google Scholar
Kroemer, G., Galassi, C., Zitvogel, L. & Galluzzi, L. Immunogenic cell stress and death. Nat. Immunol. 23, 487–500 (2022).
Article PubMed CAS Google Scholar
Papalexi, E. et al. Characterizing the molecular regulation of inhibitory immune checkpoints with multimodal single-cell screens. Nat. Genet. 53, 322–331 (2021).
Article PubMed PubMed Central CAS Google Scholar
Spring, B. Q. et al. Selective treatment and monitoring of disseminated cancer micrometastases in vivo using dual-function, activatable immunoconjugates. Proc. Natl. Acad. Sci. USA 111, E933–E942 (2014).
Article PubMed PubMed Central CAS Google Scholar
Guo, Y. et al. Copackaging photosensitizer and PD-L1 siRNA in a nucleic acid nanogel for synergistic cancer photoimmunotherapy. Sci. Adv. 8, eabn2941 (2022).
Article PubMed PubMed Central CAS Google Scholar
Chen, T. et al. Peptide-based and small synthetic molecule inhibitors on PD-1/PD-L1 pathway: a new choice for immunotherapy? Eur. J. Med. Chem. 161, 378–398 (2019).
Article ADS PubMed CAS Google Scholar
Christofides, A. et al. The complex role of tumor-infiltrating macrophages. Nat. Immunol. 23, 1148–1156 (2022).
Article PubMed PubMed Central CAS Google Scholar
Li, M. et al. Metabolism, metabolites, and macrophages in cancer. J. Hematol. Oncol. 16, 80 (2023).
Article PubMed PubMed Central Google Scholar
Koo, S. et al. Ceria-vesicle nanohybrid therapeutic for modulation of innate and adaptive immunity in a collagen-induced arthritis model. Nat. Nanotechnol. 18, 1502–1514 (2023).
Article ADS PubMed CAS Google Scholar
Wang, Y. N. et al. Vinblastine resets tumor-associated macrophages toward M1 phenotype and promotes antitumor immune response. J. Immunother. Cancer 11, e007253 (2023).
Article PubMed PubMed Central Google Scholar
Jaynes, J. M. et al. Mannose receptor (CD206) activation in tumor-associated macrophages enhances adaptive and innate antitumor immune responses. Sci. Transl. Med. 12, eaax6337 (2020).
Article PubMed PubMed Central CAS Google Scholar
Pittet, M. J., Michielin, O. & Migliorini, D. Clinical relevance of tumour-associated macrophages. Nat. Rev. Clin. Oncol. 19, 402–421 (2022).
Article PubMed Google Scholar
Song, Y. et al. Albumin nanoparticle containing a PI3Kγ inhibitor and paclitaxel in combination with α-PD1 induces tumor remission of breast cancer in mice. Sci. Transl. Med. 14, eabl3649 (2022).
Article PubMed PubMed Central CAS Google Scholar
Mohapatra, S., Pioppini, C., Ozpolat, B. & Calin, G. A. Non-coding RNAs regulation of macrophage polarization in cancer. Mol. Cancer 20, 24 (2021).
Article PubMed PubMed Central CAS Google Scholar
Winkle, M., El-Daly, S. M., Fabbri, M. & Calin, G. A. Noncoding RNA therapeutics—challenges and potential solutions. Nat. Rev. Drug Discov. 20, 629–651 (2021).
Article PubMed PubMed Central CAS Google Scholar
Yofe, I. et al. Spatial and Temporal Mapping of Breast Cancer Lung Metastases Identify TREM2 Macrophages as Regulators of the Metastatic Boundary. Cancer Discov. 13, 2610–2631 (2023).
Article PubMed PubMed Central CAS Google Scholar
Jin, L. et al. Breast cancer lung metastasis: molecular biology and therapeutic implications. Cancer Biol. Ther. 19, 858–868 (2018).
Article PubMed PubMed Central CAS Google Scholar
Dalla, E. et al. Lung-resident alveolar macrophages regulate the timing of breast cancer metastasis. Cell 187, 6631–6648.e6620 (2024).
Article PubMed PubMed Central CAS Google Scholar
Kaewkangsadan, V. et al. The differential contribution of the innate immune system to a good pathological response in the breast and axillary lymph nodes induced by neoadjuvant chemotherapy in women with large and locally advanced breast cancers. J. Immunol. Res. 2017, 1049023 (2017).
Article PubMed PubMed Central Google Scholar
Versluis, J. M., Thommen, D. S. & Blank, C. U. Rationalizing the pathway to personalized neoadjuvant immunotherapy: the Lombard Street Approach. J. Immunother. cancer 8, e001352 (2020).
Article PubMed PubMed Central Google Scholar
Wang, X. et al. In vivo CRISPR screens identify the E3 ligase Cop1 as a modulator of macrophage infiltration and cancer immunotherapy target. Cell 184, 5357–5374.e5322 (2021).
Article PubMed PubMed Central CAS Google Scholar
Lehmann, B. D. et al. Identification of human triple-negative breast cancer subtypes and preclinical models for selection of targeted therapies. J. Clin. Investig. 121, 2750–2767 (2011).
Article PubMed PubMed Central CAS Google Scholar
Zheng, Y., Li, S., Tang, H., Meng, X. & Zheng, Q. Molecular mechanisms of immunotherapy resistance in triple-negative breast cancer. Front. Immunol. 14, 1153990 (2023).
Article PubMed PubMed Central CAS Google Scholar
Bareche, Y. et al. Unravelling triple-negative breast cancer molecular heterogeneity using an integrative multiomic analysis. Ann. Oncol. Off. J. Eur. Soc. Med. Oncol. 29, 895–902 (2018).
Article CAS Google Scholar
Xiao, Z. & Puré, E. The fibroinflammatory response in cancer. Nat. Rev. Cancer 25, 399–425 (2025).
Article PubMed CAS Google Scholar
Nakamura, K. & Smyth, M. J. Myeloid immunosuppression and immune checkpoints in the tumor microenvironment. Cell. Mol. Immunol. 17, 1–12 (2020).
Article PubMed CAS Google Scholar
Goswami, S., Anandhan, S., Raychaudhuri, D. & Sharma, P. Myeloid cell-targeted therapies for solid tumours. Nat. Rev. Immunol. 23, 106–120 (2023).
Article PubMed CAS Google Scholar
Bassler, K., Schulte-Schrepping, J., Warnat-Herresthal, S., Aschenbrenner, A. C. & Schultze, J. L. The myeloid cell compartment-cell by cell. Annu. Rev. Immunol. 37, 269–293 (2019).
Article PubMed CAS Google Scholar
Biswas, S. K. & Mantovani, A. Macrophage plasticity and interaction with lymphocyte subsets: cancer as a paradigm. Nat. Immunol. 11, 889–896 (2010).
Article PubMed CAS Google Scholar
Han, S. et al. d-lactate modulates M2 tumor-associated macrophages and remodels immunosuppressive tumor microenvironment for hepatocellular carcinoma. Sci. Adv. 9, eadg2697 (2023).
Article PubMed PubMed Central CAS Google Scholar
Zheng, P. et al. Engineered bacterial biomimetic vesicles reprogram tumor-associated macrophages and remodel tumor microenvironment to promote innate and adaptive antitumor immune responses. ACS Nano 18, 6863–6886 (2024).
Article PubMed CAS Google Scholar
Uno, A. et al. A novel β-glucan-oligonucleotide complex selectively delivers siRNA to APCs via Dectin-1. J. Control. Release. Off. J. Control. Release. Soc. 338, 792–803 (2021).
Article CAS Google Scholar
Krysko, D. V. et al. Immunogenic cell death and DAMPs in cancer therapy. Nat. Rev. Cancer 12, 860–875 (2012).
Article PubMed CAS Google Scholar
Savellano, M. D., Pogue, B. W., Hoopes, P. J., Vitetta, E. S. & Paulsen, K. D. Multiepitope HER2 targeting enhances photoimmunotherapy of HER2-overexpressing cancer cells with pyropheophorbide-a immunoconjugates. Cancer Res. 65, 6371–6379 (2005).
Article PubMed CAS Google Scholar
Gabrilovich, D. I., Ostrand-Rosenberg, S. & Bronte, V. Coordinated regulation of myeloid cells by tumours. Nat. Rev. Immunol. 12, 253–268 (2012).
Article PubMed PubMed Central CAS Google Scholar
Basavarajappa, D. et al. Dicer up-regulation by inhibition of specific proteolysis in differentiating monocytic cells. Proc. Natl. Acad. Sci. USA 117, 8573–8583 (2020).
Article ADS PubMed PubMed Central CAS Google Scholar
Spitschak, A., Meier, C., Kowtharapu, B., Engelmann, D. & Pützer, B. M. MiR-182 promotes cancer invasion by linking RET oncogene activated NF-κB to loss of the HES1/Notch1 regulatory circuit. Mol. Cancer 16, 24 (2017).
Article PubMed PubMed Central Google Scholar
Zuckerman, J. E. et al. Correlating animal and human phase Ia/Ib clinical data with CALAA-01, a targeted, polymer-based nanoparticle containing siRNA. Proc. Natl. Acad. Sci. USA 111, 11449–11454 (2014).
Article ADS PubMed PubMed Central CAS Google Scholar
Mahmoudi, M. Debugging nano-bio interfaces: systematic strategies to accelerate clinical translation of nanotechnologies. Trends Biotechnol. 36, 755–769 (2018).
Article PubMed CAS Google Scholar
Adityan, S., Tran, M., Bhavsar, C. & Wu, S. Y. Nano-therapeutics for modulating the tumour microenvironment: Design, development, and clinical translation. J. Control. Release. Off. J. Control. Release. Soc. 327, 512–532 (2020).
Article CAS Google Scholar
Voabil, P. et al. An ex vivo tumor fragment platform to dissect response to PD-1 blockade in cancer. Nat. Med. 27, 1250–1261 (2021).
Article PubMed CAS Google Scholar
Peranzoni, E. et al. Ex vivo imaging of resident CD8 T lymphocytes in human lung tumor slices using confocal microscopy. J. Vis. Exp. (2017).
Peranzoni, E. et al. Macrophages impede CD8 T cells from reaching tumor cells and limit the efficacy of anti-PD-1 treatment. Proc. Natl. Acad. Sci. USA 115, E4041–e4050 (2018).
Article PubMed PubMed Central CAS Google Scholar
Ma, S. et al. YTHDF2 orchestrates tumor-associated macrophage reprogramming and controls antitumor immunity through CD8(+) T cells. Nat. Immunol. 24, 255–266 (2023).
Article PubMed PubMed Central CAS Google Scholar

Download references

Acknowledgements

This work was supported by National Key R&D Program of China (No.2021YFF0501504 to S.J., 2021YFA1101304 to S.J.), the National Natural Science Foundation of China (NSFC, No. 82402353 to J.Y., 82372024 to D.C., 82330060 to S.J., 92359304 to S.J., 82427803 to S.J., 82402355 to X.X.), Jiangsu Province Frontier Technology Research and Development Project (No. BF2025609 to S.J.), the China Postdoctoral Science Foundation (2023M740617 to J.Y., 2025T180648 to J.Y.), Research Personnel Cultivation Programme of Zhongda Hospital, Southeast University (CZXM-GSP-RC120 to J.Y.), the Fundamental Research Funds for the Central Universities (“Zhishan’” Youth Scholar of Southeast University, No. 2242024RCB0056 to D.C.), the Jiangsu Province High-Level Hospital Pairing Assistance Construction Funds (Zhongda Hospital Affiliated to Southeast University, No.zdlyg20 to D.C.), the Medical Imaging Artificial Intelligence Special Research Fund Project by Nanjing Medical Association Radiology Branch (No.10 to D.C.), Jiangsu Excellent Postdoctoral Program (2024ZB365 to J.Y.), SEU Innovation Capability Enhancement Plan for Doctoral Students (CXJH_SEU 24233 to Yingbo Li).

Author information

These authors contributed equally: Jie Yang, Di Chang.

Authors and Affiliations

Department of Radiology, Zhongda Hospital, School of Medicine, Nurturing Center of Jiangsu Province for the State Laboratory of AI Imaging and Interventional Radiology, Southeast University, Nanjing, China
Jie Yang, Di Chang, Yingbo Li, Xiaoxuan Xu, Zhiqi Zhang, Yize Li, Jianqiong Zhang, Jinbing Xie, Zebin Xiao & Shenghong Ju
Department of Nuclear Medicine, The Fourth Hospital of Harbin Medical University, Harbin, China
Zuoyu Xu
Department of Ultrasound, The Fourth Hospital of Harbin Medical University, Harbin, China
Daoshuang Li
Department of Microbiology and Immunology, School of Medicine, Southeast University, Nanjing, China
Jianqiong Zhang

Authors

Jie Yang
View author publications
Search author on:PubMed Google Scholar
Di Chang
View author publications
Search author on:PubMed Google Scholar
Yingbo Li
View author publications
Search author on:PubMed Google Scholar
Zuoyu Xu
View author publications
Search author on:PubMed Google Scholar
Daoshuang Li
View author publications
Search author on:PubMed Google Scholar
Xiaoxuan Xu
View author publications
Search author on:PubMed Google Scholar
Zhiqi Zhang
View author publications
Search author on:PubMed Google Scholar
Yize Li
View author publications
Search author on:PubMed Google Scholar
Jianqiong Zhang
View author publications
Search author on:PubMed Google Scholar
Jinbing Xie
View author publications
Search author on:PubMed Google Scholar
Zebin Xiao
View author publications
Search author on:PubMed Google Scholar
Shenghong Ju
View author publications
Search author on:PubMed Google Scholar

Contributions

J.Y. conceived the study, performed the experiments, analyzed the data, collected patient samples, and wrote the manuscript. D.C. conceived the study and contributed to review, edit and approval the manuscript. Y.L. (Yingbo Li) performed the experiments and analyzed the data. Z.X. (Zuoyu Xu) and D.L. contributed to the collection of patient samples and data. X.X., Z.Z., and Y.L. (Yize Li) helped in the animal experiments. Z.X. (Zebin Xiao) guided the project and contributed to review, edit and approval the manuscript. S.J., J.Z., and J.X. guided the project. S.J. supervised the whole research.

Corresponding author

Correspondence to Shenghong Ju.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Guangjun Nie and the other anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information (download PDF )

Reporting Summary (download PDF )

Transparent Peer Review file (download PDF )

Source data

Source Data (download XLSX )

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Yang, J., Chang, D., Li, Y. et al. Programmable nanomicelles rewire myeloid immunity for durable control of primary and metastatic breast cancer. Nat Commun 17, 4082 (2026). https://doi.org/10.1038/s41467-026-70859-5

Download citation

Received: 11 May 2025
Accepted: 09 March 2026
Published: 17 March 2026
Version of record: 06 May 2026
DOI: https://doi.org/10.1038/s41467-026-70859-5