M1 polarization enhances the antitumor activity of chimeric antigen receptor macrophages in solid tumors

The study addresses whether inducing an M1 polarization state can enhance the antitumor activity of chimeric antigen receptor macrophages (CAR-Ms) against solid tumors. While CAR-T therapies have transformed treatment of some hematologic malignancies, their efficacy in solid tumors is hindered by poor trafficking, immunosuppressive tumor microenvironments (TMEs), and antigen heterogeneity. Macrophages are innate immune cells with strong trafficking and infiltration into tumors, phagocytic and cytotoxic capabilities, and immunomodulatory functions, making them promising alternative CAR carriers. Tumor-associated macrophages (TAMs) are heterogeneous and often skew toward an M2-like, protumor phenotype; conversely, M1-like macrophages are tumoricidal. Given the plasticity of macrophages, reprogramming toward an M1-like phenotype could improve antitumor immunity. The authors hypothesized that M1 polarization would enhance CAR-M proinflammatory phenotype, phagocytosis, cytotoxicity, and ability to prime adaptive immunity, thereby improving therapeutic efficacy against solid tumors.

Prior work shows CAR-T efficacy in hematologic cancers but limited impact in solid tumors due to TME barriers and antigen heterogeneity. Alternative CAR-engineered cells (NK cells, γδ T cells, macrophages) are under investigation. CAR-Ms can directly phagocytose tumor cells, present antigens, promote epitope spreading, and remodel TMEs via cytokine/chemokine secretion. Adenoviral vector (Ad5f35)-engineered human CAR-Ms have shown durable M1-like phenotypes and antitumor activity; however, Ad5f35 transduces human but not murine cells due to low CD46 expression in mice. Strategies to reprogram TAMs to M1-like states (e.g., TLR/CD40 activation, MPLA + IFN-γ, CD47-SIRPα blockade, exoASO-STAT6) have demonstrated antitumor benefits. Administration of polarized M1 macrophages has shown therapeutic effects in other disease models and may maintain phenotype in vivo for extended periods. These findings support testing whether ex vivo M1 polarization can augment CAR-M function.

- CAR design and production: A HER2-targeting CAR comprising a humanized anti-HER2 scFv, a CD28 hinge, and FcγRI transmembrane and intracellular signaling domains (containing ITAMs) was synthesized and cloned into pRRLSIN.cPPT.RFPL4b lentiviral vector, with GFP as a transduction marker. Lentivirus was produced in HEK293T cells using PEI, harvested at 48/72 h, filtered, PEG6000-precipitated, and concentrated. - Cells and culture: Murine MC38 and B16F10 tumor cell lines cultured in RPMI-1640; ID8, 293T, and J774A.1 in high-glucose DMEM; all with 10% FBS, NEAA, L-glutamine, and penicillin-streptomycin. MC38 and ID8 were transduced to express luciferase and mCherry, then transduced with truncated human HER2 (extracellular/transmembrane domains) to generate HER2+ targets. - Macrophage generation: Bone marrow-derived macrophages (BMDMs) differentiated with M-CSF (25 ng/ml) for 7 days. Primary macrophages cultured in DMEM + 10% FBS. Macrophages transduced with lentivirus at MOI 50 with polybrene (10 μg/ml). Transduction efficiency assessed by GFP and by recombinant His-tagged HER2 binding with anti-His-APC. - M1 polarization: BMDMs treated with IFN-γ (20 ng/ml) and LPS (50 ng/ml) for 24 h before assays or infusion; washed prior to use. - Flow cytometry: Standard staining performed; CAR expression confirmed by His-tagged HER2 binding. Activation markers CD80 and CD86 measured after coculture with tumor cells. Phagocytosis assay quantified double-positive (GFP+ mCherry+) events after 1 h coculture (E:T 5:1) with mCherry+ targets. - In vitro cytotoxicity: Luciferase-based viability assay with Luc+ MC38-HER2 targets at varying E:T ratios, measured at 24–48 h using IVIS. Specific lysis calculated relative to tumor-only controls. YOYO3 dye used to visualize dead tumor cells. - Gene expression and cytokines: qRT-PCR (normalized to GAPDH) for Il-6, Il-12, Tnf-α; ELISAs for IL-12p70, IL-1β, and TNF-α in coculture supernatants after 24 h. - Animal models: Female C57BL/6 mice (6–8 weeks), SPF conditions, IACUC approved. • Intraperitoneal (ovarian) model: 2×10^6 Luc+ ID8-HER2 cells i.p.; macrophages administered i.p. on days 14 and 21; tumor burden monitored by IVIS; survival assessed. • Subcutaneous melanoma: 2×10^5 B16F10-HER2 cells s.c.; when tumors ~50 mm^3 (day 7), single i.v. infusion of engineered macrophages; tumor volumes measured biweekly; euthanasia at 2000 mm^3. GFP+ CAR-M infiltration assessed by anti-GFP staining. • Lung metastasis: 1×10^6 B16F10-HER2 cells i.v.; macrophages infused i.v. on days 7 and 10; lungs analyzed on day 14 for metastatic nodules, lung/body weight ratio, and H&E histology. - Statistics: GraphPad Prism 9.4; ANOVA with multiple comparisons; Kaplan–Meier via log-rank Mantel–Cox; significance thresholds: * P<0.05, ** P<0.01, *** P<0.001.

- Efficient CAR expression: >80% of primary murine macrophages were GFP+ and CAR+ by 72 h post-transduction; CAR detected via His-tagged HER2 binding. - Antigen specificity: CAR-Ms phagocytosed and killed HER2-expressing targets more effectively than HER2-negative cells. - Phagocytosis enhanced by M1 polarization: Flow cytometry-based assays showed GFP-Ms engulfed ~1% of tumor cells, increasing to 4.2% after M1 polarization; CAR-Ms increased from 6.75% to 11.7% upon M1 polarization. Fluorescence microscopy corroborated increased phagocytic events after M1 treatment. - Cytotoxicity improved with M1 polarization: In luciferase-based assays against Luc+ MC38-HER2, CAR-M cytotoxicity was E:T dependent; at E:T 10:1 after 24 h, killing was ~50% for CAR-Ms and ~70% for M1-polarized CAR-Ms, whereas GFP-Ms showed minimal killing without dose dependence. YOYO3 staining confirmed more dead tumor cells in the M1-polarized CAR-M group. - Activation and cytokines: CD80 and CD86 expression increased on CAR-Ms after coculture with MC38-HER2 and further increased after M1 polarization, indicating enhanced antigen-presenting capacity. qRT-PCR showed higher Il-6 and Tnf-α in CAR-M vs GFP-M and increased Il-6, Il-12, and Tnf-α after coculture in both groups. ELISAs confirmed elevated IL-12p70, IL-1β, and TNF-α secretion in M1-polarized macrophages. - Intraperitoneal ovarian cancer model: Compared with GFP-Ms, CAR-Ms significantly reduced tumor burden and delayed bloody ascites formation; M1-polarized CAR-Ms produced the strongest suppression of tumor progression, reduced ascites, limited weight gain due to ascites, and prolonged survival. - Subcutaneous melanoma model: Single i.v. infusion of CAR-Ms reduced tumor burden and prolonged survival versus controls; M1-polarized CAR-Ms (CAR-M1) demonstrated the most potent antitumor effects and significantly extended overall survival. GFP-positive CAR-Ms were detected in tumor tissue at 24 h post-injection. - Lung metastasis model: Systemic CAR-Ms and M1-polarized CAR-Ms significantly reduced the number of lung metastatic nodules, decreased lung/body weight ratio, and diminished metastatic areas by H&E. - Safety: No significant differences in body weight post-treatment among groups; histopathology of major organs (lungs, liver, kidney, spleen) showed no significant abnormalities, indicating favorable tolerability.

The findings validate the hypothesis that M1 polarization enhances the antitumor activity of CAR-Ms. The engineered HER2-targeting CAR-Ms specifically recognized and killed HER2+ tumor cells and, when polarized to an M1 phenotype, displayed augmented phagocytosis, cytotoxicity, and proinflammatory cytokine secretion. These properties translated into superior in vivo efficacy across multiple solid tumor models, with improved tumor control and survival. The work underscores macrophages’ advantages as CAR carriers for solid tumors, including inherent tumor infiltration, antigen presentation, and the ability to remodel the TME through cytokines and chemokines. Unlike adenoviral transduction approaches that intrinsically induce M1-like activation in human macrophages, lentiviral transduction did not activate murine macrophages, highlighting the necessity and benefit of deliberate M1 polarization (LPS + IFN-γ). Importantly, robust proinflammatory activity did not result in detectable systemic toxicity in mice within the study conditions, supporting the therapeutic window of CAR-M1 therapy. These results suggest that ex vivo polarization can potentiate CAR-M function, potentially addressing key barriers faced by CAR-T in solid tumors, such as immunosuppression and poor infiltration, and may promote epitope spreading and adaptive immunity priming.

This study introduces a novel HER2-targeting CAR-M platform using an FcγRI signaling domain and demonstrates that ex vivo M1 polarization significantly enhances CAR-M phagocytosis, cytotoxicity, proinflammatory activation, and antitumor efficacy in vivo across intraperitoneal, subcutaneous, and metastatic solid tumor models. The therapy was well tolerated in mice. Future research should focus on strategies to achieve stable and durable M1-like polarization in vivo, optimize CAR constructs for macrophages, combine with approaches addressing antigen heterogeneity, and further evaluate safety and efficacy in translational settings.

- The M1 polarization induced ex vivo is transient; the duration of the M1 state in vivo was not measured and may wane in the immunosuppressive TME. - The M1/M2 dichotomy is an oversimplification of macrophage heterogeneity; in vivo macrophage phenotypes are dynamic and functionally diverse, which may not be fully captured by this paradigm. - While increased cytokine production was observed in vitro, systemic cytokine levels and potential for cytokine-related toxicities were not extensively profiled beyond weight and histopathology endpoints. - Antigen heterogeneity in solid tumors remains a challenge and was not directly addressed beyond HER2 targeting.