Dual blockade of CD47 and HER2 eliminates radioresistant breast cancer cells

Radiotherapy (RT) is widely used across breast cancer (BC) subtypes and reduces recurrence and mortality compared with no RT. However, adaptive radioresistance can promote recurrence and metastasis, and combining RT with targeted immunotherapy (TI) is increasingly explored to improve outcomes. Radiation induces multiple proteins and cytokines within the tumor microenvironment, including immune regulators such as PD-1/PD-L1, but other radiation-induced immunosuppressive factors on tumor cells remain less defined. CD47 is a myeloid-specific immune checkpoint expressed by many cancers; binding to SIRPα on macrophages inhibits phagocytosis. Anti-CD47 therapies enhance phagocytic clearance and can have antitumor efficacy with tolerable toxicity. Beyond phagocytosis, CD47 blockade can also improve antigen presentation by dendritic cells and suppress aggressive phenotypes via EGFR signaling inhibition. The interplay between CD47 and growth factor pathways in radioresistant cancer cells is unclear. This study investigates the coordinated transcriptional regulation and functional crosstalk between CD47 and HER2 in radioresistant BC cells, irradiated tumors, and patient samples, and evaluates whether dual blockade can eliminate radioresistant cancer cells and synergize with RT.

Background work shows: (1) RT reduces recurrence and mortality but can trigger adaptive tumor resistance and alters the tumor immune microenvironment; (2) CD47 is an innate immune checkpoint limiting macrophage phagocytosis via SIRPα signaling and anti-CD47 antibodies enhance tumor clearance; (3) HER2 signaling activates PI3K–AKT and NF-κB, contributes to proliferation and radioresistance, and can suppress antitumor immunity (e.g., via STING pathway disruption); (4) Prior studies suggested CD47 blockade can augment trastuzumab efficacy through antibody-dependent cellular phagocytosis. Collectively, these data motivated investigating whether radiation induces a CD47–HER2 axis that promotes immune evasion and proliferation, and whether dual targeting enhances RT efficacy.

- Data mining: Analyzed TCGA via GEPIA to compare CD47 and HER2 expression across cancers, including 1085 breast tumors vs 291 normal tissues. - Clinical samples: Immunoblotting and IHC on HER2-positive and -negative BC patient tumors; paired primary vs recurrent samples assessed for CD47 and HER2 expression; survival analyses (RFS, DMFS, OS) stratified by HER2–CD47 signature using SurvExpress. - Cell models: Established radioresistant breast cancer lines (MDA-MB-231/C5, MCF7/C6) by repeated irradiation; isolated HER2+ radiation-derived BCSCs (RD-BCSCs). Additional lines included SKBR3, BT474, BT549, U251, HCT116, and normal MCF10A. - Protein expression: Western blots and flow cytometry assessed CD47 and HER2 levels and double-positive populations; radiation treatments included single 5–10 Gy or fractionated IR (e.g., 2 Gy × 5). - Mechanism assays: HER2 inhibition with lapatinib or Herceptin; NF-κB inhibition (IMD-0354, MLN120B, BMS-345541); CD47 promoter luciferase reporters with wild-type or NF-κB motif mutation; NF-κB (p65) ChIP–qPCR at the CD47 promoter; CRISPR/Cas9 KO of HER2 and/or CD47 in MCF7/C6 and RD-BCSCs. - Functional assays: Macrophage-mediated phagocytosis using THP1-derived human macrophages or RAW264.7 mouse macrophages; DIO/DDAO dual-label flow cytometry quantification. Aggressiveness assays included mammosphere formation, Matrigel transwell invasion, and wound gap-filling. Clonogenic survival after IR with or without antibodies (anti-CD47 B6H12.2; Herceptin) or inhibitors. - In vivo models: Syngeneic orthotopic BALB/c 4T1 tumors, including a radioresistant 4T1/C2 line; generated CRISPR CD47−/−, HER2−/−, and double KO tumors. Local tumor RT (e.g., 5 Gy single dose; 5 Gy × 2) combined with intratumoral anti-CD47 F(ab’) and/or Herceptin; tumor volume/weight endpoints. IHC and immunofluorescence for CD11b+ macrophage infiltration and GFP+ tumor phagocytosis; necrosis assessment (H&E). Statistics: t-tests/ANOVA; significance at P<0.05.

- Co-expression and prognosis: CD47 and HER2 are co-overexpressed across multiple cancers, including breast cancer (TCGA: 1085 tumors vs 291 normals). CD47 levels are higher in HER2+ vs HER2− BC cells and patient tumors. Both receptors are more frequently elevated in recurrent vs primary BC. Patients with dual high CD47–HER2 expression have worse recurrence-free and distant metastasis-free survival (log-rank P<0.001). - Radioresistant models: Radioresistant MDA-MB-231/C5 and MCF7/C6 cells exhibit markedly elevated CD47 and HER2 protein and increased CD47+/HER2+ populations. RT increases both receptors in syngeneic 4T1 tumors and elevates CD47 in multiple human cancer cell lines and xenografts. - Mechanism—HER2–NF-κB drives CD47 transcription: HER2 inhibition (lapatinib; Herceptin) reduces CD47+ populations and blocks radiation-induced CD47 upregulation in HER2+ cells and RD-BCSCs. HER2 CRISPR KO abrogates basal and radiation-induced CD47 expression and suppresses CD47 promoter activity. CD47 promoter activity depends on an NF-κB motif and is inhibited by NF-κB inhibitors (e.g., IMD-0354); ChIP confirms p65 recruitment to the CD47 promoter after IR or TNF-α. Conversely, CD47 KO lowers HER2+ cell fraction (HER2+ reduced from 20.2% to 3.66% in MCF7/C6). - Functional impact—phagocytosis and aggressiveness: Anti-CD47 increases macrophage phagocytosis of both MCF7 and MCF7/C6 cells; anti-CD47 also suppresses mammosphere formation, invasion, and wound closure in MCF7/C6. HER2 pathway inhibition (lapatinib, Herceptin) enhances phagocytosis; combining anti-CD47 with Herceptin yields synergistic phagocytosis. - Clonogenic survival and radiosensitization: Dual antibody blockade (anti-CD47 + Herceptin) with IR most strongly reduces clonogenic survival versus single agents. CRISPR CD47−/− or HER2−/− each increases phagocytosis; double KO does not further increase phagocytosis but confers greater dose-dependent radiosensitization and reduced clonogenicity than single KOs. - In vivo efficacy—genetic and antibody blockade: CD47−/−/HER2−/− 4T1/C2 tumors grow slowest; RT (5 Gy) further synergizes with double KO to inhibit growth, associated with more CD11b+ macrophage infiltration and tumor necrosis. In antibody studies, RT (5 Gy × 2) combined with dual anti-CD47 F(ab’) + Herceptin produces the greatest reduction in tumor volume/weight, with increased macrophage infiltration and in situ phagocytosis versus RT plus single antibody. Overall, results support a radiation-induced HER2–NF-κB–CD47 axis enabling proliferative and immune-evasive phenotypes; dual blockade best overcomes resistance.

The study addresses how radiation-induced adaptive responses in tumor cells overlap with immune evasion to limit RT–immunotherapy synergy. Findings demonstrate that HER2-driven NF-κB activation transcriptionally upregulates CD47, coupling intrinsic proliferation (HER2) with extrinsic anti-phagocytosis (CD47) in radioresistant BC cells and irradiated tumors. This crosstalk diminishes macrophage-mediated clearance and sustains clonogenic potential after RT. Disrupting either node partly restores antitumor immunity and radiosensitivity, but dual blockade—genetically or with antibodies—achieves superior control and enhances macrophage infiltration/phagocytosis in vivo. These insights rationalize combining RT with coordinated targeting of proliferative and immune checkpoint pathways to eliminate resistant tumor subpopulations, including HER2+ cancer stem-like cells. The work also contextualizes variable RT–CD47 interactions across tumor types and underscores the central role of NF-κB as a stress-responsive mediator integrating growth and immune signals.

CD47 and HER2 are coordinately upregulated in radioresistant breast cancer via a HER2–PI3K/AKT–NF-κB axis that drives CD47 transcription. This coupling endows radiation-surviving cells with enhanced proliferation and anti-phagocytic capacity. Dual blockade of CD47 and HER2—through CRISPR deletion or combined antibodies—synergizes with RT to maximize macrophage-mediated clearance, reduce clonogenic survival, and inhibit tumor growth in syngeneic models. These results support a therapeutic strategy of combining RT with concurrent inhibition of HER2 and CD47 to eradicate radioresistant, recurrent HER2-expressing BC. Future studies should evaluate clinical translation, optimize dosing/scheduling with RT, and explore additional radiation-inducible immune regulators for multi-target combinations.

- Preclinical scope: Conclusions are based on cell lines, radioresistant derivatives, and mouse syngeneic models; clinical validation is lacking. - Tumor heterogeneity and context: Responses to RT and CD47/HER2 modulation may vary by tumor subtype and microenvironment (e.g., HPV-positive cancers can show RT-induced CD47 decrease), limiting generalizability. - Antibody modality and delivery: In vivo studies used intratumoral dosing and F(ab’) fragments for anti-CD47 in mice; systemic pharmacokinetics, safety (e.g., anemia), and optimal regimens were not evaluated. - Mechanistic breadth: While NF-κB-mediated transcriptional control is supported, broader networks (other receptors/cofactors) induced by RT may contribute and were not comprehensively dissected. - Survival analyses: Prognostic associations were derived from public datasets and IHC cohorts; standardized prospective validation is needed.