Considering the role of MDSCs in conferring resistance to ICIs and other anticancer therapies, reduction or prevention of the migration of MDSCs into the TME and/or their differentiation into TAMs is of significant interest as a therapeutic strategy in patients with cancer; however, results of studies investigating this approach have been mixed to date. For example, chemotaxis inhibition, via blockade of the chemokine receptors type 1 (CXCR1), type 2 (CXCR2), or C-C chemokine receptor type 2 (CCR2), was shown to reduce the migration of MDSCs into the TME, and thereby enhance the efficacy of PD-1 targeting therapy in animal models.30–32 However, in patients with pancreatic ductal adenocarcinoma (PDAC), inhibition of CCR2 with small molecule antagonists had limited clinical efficacy to date.33 Preclinical models suggested that resistance may be mediated by influx or expansion of alternate CCR2-negative MDSC subsets, compensating for the loss of the targeted myeloid population.31 32 To counteract these compensatory mechanisms, CCR2/CCR5 dual inhibitors that simultaneously target multiple chemotaxis pathways on different myeloid subsets are currently being evaluated in the clinic.34–36 Nonetheless, the potential for compensatory action by an untargeted chemokine axis or myeloid subset may limit this approach.31 32

Colony-stimulating factor 1 receptor (CSF-1R),37 phosphoinositide 3-kinase-gamma (PI3K_γ),38 and integrin39 signaling pathways also play an important role in MDSC migration and/or TAM polarization in the TME. Inhibition of CSF-1R reduced TAM infiltration and survival in the TME in preclinical and clinical studies.40–43 Although signs of efficacy have been seen with inhibition of CSF-1R in phase 1 (NCT01346358, NCT02265536),40 44 45 phase 1b/2 (NCT02713529),46 phase 2 (NCT01349036, NCT01217229, NCT03336216),41 47–49 and phase 3 (NCT02371369)43 clinical trials, activity was limited to very select indications. These outcomes suggested that targeting only TAM polarization may be insufficient across broader solid tumor indications due to other compensatory immunosuppressive mechanisms in the TME (eg, PMN-MDSCs).41 42 As such, exploration of additional combination strategies for CSF-1R inhibitors is currently underway.41 In addition, the inhibition of PI3K_γ was shown to reduce MDSC migration and accumulation in the TME,50 reverse TAM polarization, and was synergistic with PD-1 blockade in preclinical studies.38 50–53 These preclinical results have led to the investigation of PI3K_γ inhibitors in several phase 2 clinical studies (NCT03980041, NCT03795610, NCT03961698), and the results are forthcoming.

The CD11b/CD18 integrin heterodimer (Mac-1; α_Mβ₂), which is expressed on MDSCs, TAMs, and subsets of dendritic cells (DCs), mediates myeloid cell adhesion to the vasculature, transendothelial migration, tissue recruitment under inflammatory conditions, phagocytosis, and survival.54 55 Blocking of CD11b/CD18 and its ligands, or deletion of the CD11b or CD18 genes, was shown to decrease the severity of inflammatory response in animal models of vascular injury.56–58 However, blocking of CD11b/CD18 with monoclonal antibodies such as rovelizumab,59 erlizumab,60 or efalizumab,61 has had limited success in clinical trials to date, likely since saturation (>90% occupancy) of CD11b is not achievable at clinically tolerable doses.59 62 63 Interestingly, stabilizing integrins α₄β₁, α_Lβ₂, and α₄β₇ in an activated state has been shown to increase cell adhesion and decrease eosinophil migration,64–67 suggesting that stabilizing CD11b/CD18 in an activated state may have a similar effect on cell adhesion and migration. Indeed, leukadherin-1, a small molecule allosteric modulator of CD11b, was shown to increase cell adhesion and reduce migration in vitro and effectively reduce macrophage and neutrophil migration in several animal models of inflammatory conditions.63 68–71 These findings warranted further investigation of CD11b modulation on MDSC migration and function in the TME.

GB1275 (formerly ADH-503) is the salt form of leukadherin-1 designed to stabilize CD11b in an active state.63 68 72 73 Unlike inhibition, activation of CD11b does not require saturation of the integrin and, therefore, it is hoped that use of GB1275 will be able to overcome the dosing challenges of earlier CD11b-targeted strategies that focused on inhibition. GB1275 binds to the allosteric pocket in the αA domain of CD11b and stabilizes CD11b in the active conformation required for its ligand binding63 68 74 (figure 1). By altering CD11b conformation, GB1275 enhances CD11b-mediated adhesion to intercellular adhesion molecule 1 (ICAM-1, CD54) on vascular endothelium and thereby impairs extravasation and transendothelial myeloid cell migration into inflamed tissues.63 68 71 In addition, by modulating CD11b signaling in TAMs, GB1275 increases expression levels of interferon-related factors (IFNα1, IFNβ, CXCL9 and CXCL10) and antigen presentation/co-stimulation complexes (MHC-I, MHC-II, CD80, and CD86), while reducing the expression of immunosuppressive genes (IL10, TGFβ, and arginase-1).72 Thus, GB1275 allosteric modulation of CD11b, leads to a shift in TAM polarization towards pro-inflammatory and antigen-presenting phenotypes, potentiating an interferon response and facilitating T-cell recruitment and activation at the tumor site (figure 2).72 74 75

Based on these initial mechanistic findings, tumor types that are characterized by abundant infiltration by CD11b⁺ MDSCs and display dependence on MDSCs in checkpoint inhibitor resistance may potentially benefit from GB1275, including PDAC,72 lung cancer,74 breast cancer,76 prostate cancer,13 77 colorectal cancer (CRC),78 79 gastric cancer,80 and esophageal cancer.81 82 Elevated levels of Itgam (gene coding for the CD11b subunit of Mac-1 complex) messenger RNA were confirmed across a variety of tumors types in The Cancer Genome Project PanCancer profiling data set, where PDAC and lung adenocarcinoma displayed especially increased relative expression. Assessment of CD11b+ infiltrates by immunohistochemistry across PDAC, CRC and gastric cancer demonstrated heterogenous infiltration of CD11b+MDSCs across representative tumor sections, in line with prior published data.72 Importantly, subsets of CRC and gastric tumors displayed elevated presence of CD11b+ infiltrate, in some cases representing greater than 25%–50% of the cell population (figure 3).

Antitumor activity of GB1275 as a single agent, and in combination with chemotherapy or ICI therapies, was evaluated in several syngeneic tumor models, including PDAC72 and lung cancer74 75 and is reviewed below.

The effect of GB1275 in PDAC was studied in syngeneic orthotopic murine PDAC cell lines (KP2 or KI) or in the genetically engineered KPC (p48-CRE/LSL-Kras^G12D/p53^flox/flox) mouse models that closely resembles the pathology and genomic features of human PDAC.72 83 In all of these PDAC models, GB1275 significantly improved survival compared with a vehicle control, demonstrating efficacy as a single agent.72 In the same mouse models, GB1275 reduced tumor infiltration of CD11b⁺ MDSCs and increased influx of activated CD103⁺ DCs and CD8⁺ T cells. GB1275 was found to upregulate the expression of PD-1/PD-L1 checkpoint molecules on CD8⁺ T-cells, TAMs, and PDAC cells, suggesting that GB1275 may act concurrently with PD-1-blocking antibodies or other ICI therapies.72 Indeed, in syngeneic orthotopic PDAC mouse models a combination treatment of GB1275 with a PD-1-blocking antibody improved survival to >120 days in all animals, while survival in the vehicle control was 0% after 40 days.72 Notably, single-agent PD-1 blocking antibody failed to improve survival outcomes relative to the vehicle control, which aligns with clinical reports of limited efficacy of PD-1 pathway targeted therapies in patients with PDAC,84 providing further evidence for the relevance of the PDAC mouse model. Similar results were obtained in the KPC model for PDAC.72 Furthermore, a combination of GB1275 with SOC chemotherapy (gemcitabine plus paclitaxel) displayed improved survival when compared with SOC alone or with single-agent GB1275 treatment.72 These data suggest that GB1275 may be effective for the treatment of patients with PDAC as a single agent, or in combination with ICIs or SOC, and warrant further investigation in clinical studies.

Similar to findings in the PDAC model, GB1275 was shown to reduce TAM infiltration to the TME, increase the proportion of pro-inflammatory M1 TAMs, and inhibit tumor growth in a CD8 T-cell dependent manner in the Lewis lung carcinoma syngeneic lung cancer model. These effects were demonstrated to be dependent on CD11b modulation, as neither leukadherin-1 nor GB1275 were effective against LLC tumors propagated in CD11b knockout mice (Itgam –/–).74 75 Combinations with ICI were not evaluated in these studies; however, observed impact of GB1275 monotherapy on macrophage polarization and induction of T-cell mediated immunity, suggests that further studies may be warranted. Although treatment with ICIs significantly improves survival in patients with lung cancer, these agents are not effective in all patients.85–87 These preclinical studies suggest CD11b modulation with GB1275 may be a potential strategy for the treatment of lung cancer subsets, where TAM immune suppression may be contributing to ICI resistance.

The effect of leukadherin-1 on breast cancer was evaluated in two breast cancer models: a syngeneic orthotopic breast cancer model (CL-66) and a human triple negative breast cancer (TNBC) xenograft model (MDA-MB-231) in mice that lack T-cells. In the immune-competent model, leukadherin-1 displayed single agent antitumor activity and demonstrated improved outcomes in combination with either irradiation or paclitaxel.75 Furthermore, leukadherin-1 treatment led to a reduction in tumor CD11b⁺ MDSCs, and a repolarization of macrophages from the CD206⁺ to the MHCII⁺ subtype, which was further potentiated in combination with SOC chemotherapy (paclitaxel).75 Leukadherin-1 also reduced tumor growth in the human xenograft model, suggesting that modulation of the innate immunity alone was sufficient to delay tumor growth and that further studies to understand effects of GB1275 in breast cancer are warranted.75

Although GB1275 has not been assessed to date in animal models of prostate, colorectal, gastric, or esophageal cancers, there is a strong rationale for its investigation in these tumor types. Patients with prostate cancer often acquire resistance to androgen deprivation therapy and move to a castration-resistant prostate cancer (CRPC) phenotype.3 13 Resistance to androgen deprivation therapy is attributed to the activation of the androgen receptor pathway by IL23, which is secreted by PMN-MDSCs.3 Elevated levels of CD11b⁺ MDSCs or CD163⁺ TAMs in the TME are associated with poor prognosis in patients with prostate cancer.16 77 88 In addition, patients with elevated pretreatment levels of CD11b⁺CD33⁺HLA-DR^–CD14⁺ mMDSC were less responsive to combination treatment.16

In patients with CRC, the accumulation of CD11b⁺ PMN-MDSCs in the TME is associated with resistance to ICIs and poor outcomes.21 79 89 The involvement of CD11b⁺ MDSC in the development of CRC was studied in the Apc^Min/+ mouse model for the development of CRC. The mice were crossed with CD11b^–/– mice to generate Apc^Min/+ CD11b knockout mice (Apc^Min/+/CD11b^–/–).54 In the Apc^Min/+/CD11b^–/– mice, the lack of CD11b⁺ cells resulted in reduced PMN-MDSC infiltration into the TME, peripheral blood, and spleen compared with the Apc^Min/+ mice.54 Furthermore, the levels of IFNγ and CXCL9 were upregulated in the Apc^Min/+/CD11b^–/– mice tumor tissues, compared with tumor tissues in Apc^Min/+ mice.54 Therefore, targeting the infiltration of PMN-MDSCs may render the tumors more susceptible to SOC therapy and given poorer outcomes associated with disease progression and the dependence on cytotoxic chemotherapy, it would seem prudent for investigation of GB1275 in CRPC and CRC.

Similarly, elevated levels of CD14^–CD11b⁺ PMN-MDSCs in the peripheral blood of patients with gastric and esophageal cancers may be associated with reduced overall survival.80 81 In patients with esophageal squamous cell carcinoma (ESCC), infiltrating CD68⁺CD163⁺ TAMs were associated with resistance to chemotherapy and radiation.82 The role of these infiltrating TAMs was validated in a murine model for ESCC, where incidence and number of tumors were reduced with genetic approaches targeting TAM recruitment or polarization.90

Taken together, these preclinical studies demonstrated that CD11b modulation may render tumors with elevated MDSC infiltration more sensitive to ICIs and other SOC therapies. In general, most of the tumor types investigated in animal models to date have a poor response to anti-PD-1 therapy, and other effective treatment options are limited; consequently, outcomes are poor and there is a large unmet need for new strategies in these cancers.

Based on the promising preclinical evidence described above, GB1275 is currently being tested in a first-in-human, open-label, phase 1/2 clinical study (NCT04060342) as monotherapy, and in combination with pembrolizumab (Keytruda) in patients with specified advanced solid tumor types that are known to be resistant or less likely responsive to immuno-oncology therapies. The rationale for the selection of the tumor types in the dose escalation phase (phase 1) included the following: tumor types with high unmet need for effective treatment, or with moderate or poor response to ICIs therapy; tumor types with preclinical evidence of GB1275 antitumor efficacy; tumor types in which myeloid suppressor biology is known to be involved based on literature and The Cancer Genome Atlas analyses; and tumor types in which preclinical data suggested mechanistic rationale for combination with chemotherapy.

Adults (age ≥18 years) with Eastern Cooperative Oncology Group performance status 0 or 1 and prespecified relapsed or refractory tumors are eligible to enroll in the study. Patients with history of another malignancy within 2 years prior to enrollment, known history of testing positive for HIV, positive for hepatitis B or hepatitis C infection, unable to take oral medication, and women who are pregnant or nursing are excluded from the study. The study is divided into two phases: phase 1 (dose escalation and expansion) and a phase 2 basket expansion. The dose escalation phase, based on a standard 3+3 design, enrolled patients with microsatellite-stable colorectal cancer (MSS-CRC), gastric cancer, metastatic PDAC (mPDAC), TNBC, metastatic CRPC, or esophageal cancer. Oral GB1275 is given at escalation dose levels alone (regimen A) or with intravenous pembrolizumab 200 mg every 3 weeks (regimen B). Once a recommended phase 2 dose (RP2D) of regimen B is determined, up to 40 patients will be enrolled into the phase 1 expansion to be treated at the RP2D in combination with pembrolizumab to further assess the safety, tolerability, clinical and biological activities at the selected RP2D. Phase 2 basket expansion will use a Simon’s two-stage optimal design to evaluate GB1275 in combination with pembrolizumab.

The objective of the dose escalation phase is to evaluate safety and tolerability to determine the maximum tolerated dose or RP2D and to examine the pharmacokinetic and pharmacodynamic profile of GB1275 alone and in combination. The objective of the expansion phase is to determine the safety of the selected dose and schedule of GB1275 with intravenous pembrolizumab (regimen B) in patients with previously treated specified advanced solid tumors. The objective of phase 2 basket expansion is to evaluate the efficacy and safety/tolerability of GB1275 in combination with pembrolizumab.

Exploratory objectives for the study are to evaluate changes in biomarkers to assess target engagement, including mMDSCs and PMN-MDSCs in blood and tumor, immune cell distribution by flow cytometry, CD11b⁺ cells at the tumor site, and CD8⁺ cells at the tumor site.

To date, enrollment has commenced in six sites in the USA and one site in the UK, and additional sites are being recruited in the USA. The study is ongoing and is estimated to complete in March 2023.