Combining CD47 blockade with trastuzumab eliminates HER2-positive breast cancer cells and overcomes trastuzumab tolerance

Trastuzumab, a targeted anti-human epidermal-growth-factor receptor-2 (HER2) monoclonal antibody, represents a mainstay in the treatment of HER2-positive (HER2⁺) breast cancer. Although trastuzumab treatment is highly efficacious for early-stage HER2⁺ breast cancer, the majority of advanced-stage HER2⁺ breast cancer patients who initially respond to trastuzumab acquire resistance to treatment and relapse, despite persistence of HER2 gene amplification/overexpression. Here, we sought to leverage HER2 overexpression to engage antibody-dependent cellular phagocytosis (ADCP) through a combination of trastuzumab and anti-CD47 macrophage checkpoint immunotherapy. We have previously shown that blockade of CD47, a surface protein expressed by many malignancies (including HER2⁺ breast cancer), is an effective anticancer therapy. CD47 functions as a “don’t eat me” signal through its interaction with signal regulatory protein-α (SIRPα) on macrophages to inhibit phagocytosis. Hu5F9-G4 (magrolimab), a humanized monoclonal antibody against CD47, blocks CD47’s “don’t eat me” signal, thereby facilitating macrophage-mediated phagocytosis. Preclinical studies have shown that combining Hu5F9-G4 with tumor-targeting antibodies, such as rituximab, further enhances Hu5F9-G4’s anticancer effects via ADCP. Clinical trials have additionally demonstrated that Hu5F9-G4, in combination with rituximab, produced objective responses in patients whose diffuse large B cell lymphomas had developed resistance to rituximab and chemotherapy. These studies led us to hypothesize that combining Hu5F9-G4 with trastuzumab would produce an anticancer effect in antibody-dependent cellular cytotoxicity (ADCC)-tolerant HER2⁺ breast cancer. This combination significantly suppressed the growth of ADCC-tolerant HER2⁺ breast cancers via Fc-dependent ADCP. Our study demonstrates that combining trastuzumab and Hu5F9-G4 represents a potential new treatment option for HER2⁺ breast cancer patients, even for patients whose tumors have progressed after trastuzumab.

Overexpression of human epidermal-growth-factor receptor-2 (HER2) occurs in ∼16% of breast cancers in the United States (1–3) and has been associated with a number of adverse prognostic factors (summarized in ref. 4). Prior to the advent of HER2-targeted therapeutics, HER2 overexpression was associated with increased risk of recurrence and poor survival rates (1, 2). Trastuzumab is a humanized monoclonal antibody that selectively binds HER2. Clinical use of trastuzumab has dramatically improved the outcomes of patients with HER2⁺ breast cancer and remains the foundational component of modern standard of care treatment regimens for HER2⁺ breast cancer in the neoadjuvant, adjuvant, and metastatic settings (5, 6). Early studies of trastuzumab’s mechanism of action focused on trastuzumab’s inhibition of protumor growth HER2 signaling pathways (7–9). Subsequent research revealed that trastuzumab also coopts a patient’s immune system to promote an antitumor response (7–11). This later body of research initially elucidated trastuzumab’s ability to engage Fc-receptors on natural killer cells (NKs) to promote antibody-dependent cellular cytotoxicity (ADCC) (12, 13). Recent reports have further illuminated trastuzumab’s ability to engage Fc-γ receptors (FcγR) on macrophages and promote antibody-dependent cellular phagocytosis (ADCP) (14).Administering trastuzumab to early-stage HER2⁺ breast cancer patients significantly increases disease-free survival and overall survival rates (15–17). Treating advanced-stage HER2⁺ breast cancer patients with the most efficacious trastuzumab-based regimens, however, produces less hopeful outcomes. For example, the Food and Drug Administration (FDA)-approved regimen studied in the CLEOPATRA clinical trial for HER2⁺ metastatic breast cancer in the first line of treatment utilized trastuzumab in combination with docetaxel and pertuzumab; this regimen resulted in a median progression-free survival of 18.7 mo (18, 19). In the same study, 19.8% of patients did not achieve an objective clinical response to trastuzumab-based treatment (20). And, of the advanced-stage HER2⁺ breast cancer patients who initially responded to trastuzumab, pertuzumab, and docetaxel, the median duration of response was 20.2 mo; thereafter, the majority of patients experienced objective disease-progression, defining acquired clinical resistance to trastuzumab-based therapy (18).A myriad of potential mechanisms of trastuzumab resistance have been reported, such as: 1) perturbation of HER family receptors or binding of therapeutic antibodies to HER2 (e.g., shedding of the HER2 extracellular domain, expression of the Δ16HER2 splice isoform, overexpression of MUC4/MUC1 resulting in steric hindrance to trastuzumab binding to the HER2 extracellular domain, and increased phosphorylation of HER3); 2) parallel receptor pathway activation (e.g., overexpression of other HER family members, up-regulation of IGF1 receptor, erythropoietin receptor, AXL receptor, or MET receptor); and 3) activation of downstream signaling events distal to HER2 receptor (e.g., hyperactivation of the PI3 kinase/Akt pathway by loss of PTEN or PIK3CA mutational activation, cyclin E amplification/overexpression, up-regulation of miR-21, and expression of the estrogen receptor) (21). Impairments in trastuzumab-mediated ADCC may also lead to relative resistance to trastuzumab (22, 23). Interestingly, it has been shown that even after HER2⁺ breast cancers relapse or progress after trastuzumab, resistant cells most often still overexpress HER2 (24). Given the many ways in which trastuzumab resistance develops, there is an urgent clinical need for novel treatment approaches that provide HER2 specificity without eliciting the same mechanisms of trastuzumab resistance that are currently seen in the clinic (25). This led us to hypothesize that engagement of macrophages and activation of ADCP could still be effective even in the face of various resistance mechanisms—as long as the HER2 ectodomain target epitope is present.Our laboratory has previously shown that many different cancers overexpress the CD47 surface protein to convey a “don’t eat me” signal to macrophages (26–29) and to counteract “eat me” signals (30–32), thereby resulting in immune evasion through inhibition of macrophage phagocytosis. This led to the development of a new type of immunotherapy based on macrophage checkpoint inhibition through blockade of CD47. Hu5F9-G4 (magrolimab) is a humanized monoclonal antibody against CD47 that blocks CD47''s interaction with signal regulatory protein-α (SIRPα), thereby diminishing the inhibition of macrophages by cancer cells (33). As a monotherapy, Hu5F9-G4’s anticancer activity works by blocking CD47’s antiphagocytic signaling. The combination of Hu5F9-G4 and tumor-targeting antibodies, moreover, promotes ADCP by increasing the amount of “eat me” signals provided by the interaction of cancer-targeting antibody Fc domains and macrophage Fc-receptors (33–36). We have previously demonstrated this principle of ADCP enhancement in the context of rituximab-resistant CD20-expressing diffuse large B cell non-Hodgkin’s lymphoma. These studies showed that combining Hu5F9-G4 with rituximab (an anti-CD20 tumor-targeting antibody) in human xenograft models produced an anticancer ADCP response (34). In the clinical trials that followed, about half of the patients who are relapsed and refractory to rituximab plus or minus chemotherapy nevertheless responded to magrolimab plus rituximab (35). This clinical study suggested that combining Hu5F9-G4 with rituximab may resensitize refractory lymphoma cells to rituximab via an ADCP mechanism (35, 37).Based on these previous studies, we hypothesized that administering a combination of trastuzumab and Hu5F9-G4 to ADCC-tolerant HER2⁺ breast cancer cells would resensitize these cells to trastuzumab. We found that this combination was more efficacious than either treatment alone. We also found that this combinatorial treatment augments ADCP via Fc-receptor-mediated phagocytosis. Taken together, our study suggests that combining Hu5F9-G4 and trastuzumab may represent an alternative or complementary approach to the current standard of care for advanced HER2⁺ breast cancer that expands trastuzumab’s efficacy through engaging ADCP while preserving and utilizing trastuzumab’s HER2-targeting capabilities.