Improved efficacy against malignant brain tumors with EGFRwt/EGFRvIII targeting immunotoxin and checkpoint inhibitor combinations

Abstract

Background: D2C7-IT is a novel immunotoxin (IT) targeting wild-type epidermal growth factor receptor (EGFRwt) and mutant EGFR variant III (EGFRvIII) proteins in glioblastoma. In addition to inherent tumoricidal activity, immunotoxins induce secondary immune responses through the activation of T cells. However, glioblastoma-induced immune suppression is a major obstacle to an effective and durable immunotoxin-mediated antitumor response. We hypothesized that D2C7-IT-induced immune response could be effectively augmented in combination with αCTLA-4/αPD-1/αPD-L1 therapies in murine models of glioma.

Methods: To study this, we overexpressed the D2C7-IT antigen, murine EGFRvIII (dmEGFRvIII), in established glioma lines, CT-2A and SMA560. The reactivity and therapeutic efficacy of D2C7-IT against CT-2A-dmEGFRvIII and SMA560-dmEGFRvIII cells was determined by flow cytometry and in vitro cytotoxicity assays, respectively. Antitumor efficacy of D2C7-IT was examined in immunocompetent, intracranial murine glioma models and the role of T cells was assessed by CD4+ and CD8+ T cell depletion. In vivo efficacy of D2C7-IT/αCTLA-4/αPD-1 monotherapy or D2C7-IT+αCTLA-4/αPD-1 combination therapy was evaluated in subcutaneous unilateral and bilateral CT-2A-dmEGFRvIII glioma-bearing immunocompetent mice. Further, antitumor efficacy of D2C7-IT+αCTLA-4/αPD-1/αPD-L1/αTim-3/αLag-3/αCD73 combination therapy was evaluated in intracranial CT-2A-dmEGFRvIII and SMA560-dmEGFRvIII glioma-bearing mice. Pairwise differences in survival curves were assessed using the generalized Wilcoxon test.

Results: D2C7-IT effectively killed CT-2A-dmEGFRvIII (IC₅₀ = 0.47 ng/mL) and SMA560-dmEGFRvIII (IC₅₀ = 1.05 ng/mL) cells in vitro. Treatment of intracranial CT-2A-dmEGFRvIII and SMA560-dmEGFRvIII tumors with D2C7-IT prolonged survival (P = 0.0188 and P = 0.0057, respectively), which was significantly reduced by the depletion of CD4+ and CD8+ T cells. To augment antitumor immune responses, we combined D2C7-IT with αCTLA-4/αPD-1 in an in vivo subcutaneous CT-2A-dmEGFRvIII model. Tumor-bearing mice exhibited complete tumor regressions (4/10 in D2C7-IT+αCTLA-4 and 5/10 in D2C7-IT+αPD-1 treatment groups), and combination therapy-induced systemic antitumor response was effective against both dmEGFRvIII-positive and dmEGFRvIII-negative CT-2A tumors. In a subcutaneous bilateral CT-2A-dmEGFRvIII model, D2C7-IT+αCTLA-4/αPD-1 combination therapies showed dramatic regression of the treated tumors and measurable regression of untreated tumors. Notably, in CT-2A-dmEGFRvIII and SMA560-dmEGFRvIII intracranial glioma models, D2C7-IT+αPD-1/αPD-L1 combinations improved survival, and in selected cases generated cures and protection against tumor re-challenge.

Conclusions: These data support the development of D2C7-IT and immune checkpoint blockade combinations for patients with malignant glioma.

Keywords: EGFR; Immune checkpoint inhibitors; Immunotoxin; Malignant gliomas; T cells.

Fig. 1

Flow cytometric analysis of D2C7 mAb and cytotoxicity of D2C7-IT against malignant glioma cell lines. a and c CT-2A-dmEGFRvIII-Luc (a) and SMA560-dmEGFRvIII-Luc (c) cell lines were stained with D2C7-AF488/−APC (pink open peaks) or IgG1-AF488/−APC isotype control Ab (filled purple peaks). b and d Cytotoxicity of D2C7-IT against CT-2A-dmEGFRvIII-Luc (b) and SMA560-dmEGFRvIII-Luc (d) cell lines was assessed by the WST-1 assay. The data are given as IC₅₀ values, the concentration of D2C7-IT that causes 50% tumor cell death after a 48-h incubation

Fig. 2

Dose-response comparison of two different D2C7-IT concentrations in orthotopic glioma models. Treatment schedule, survival curves, and median survival are shown for C57BL/6 J mice bearing intracranial CT-2A-dmEGFRvIII-Luc tumors (a) and VM/Dk mice bearing intracranial SMA560-dmEGFRvIII-Luc tumors (b) infused with either vehicle control, 0.1μg D2C7-IT, or 0.3μg D2C7-IT. The p-values generated from the generalized Wilcoxon test are provided for both tumor models

Fig. 3

Antitumor effects of D2C7-IT therapy post-CD4+ and CD8+ T cell depletion in orthotopic glioma models. Treatment schedule, survival curves, and median survival are shown for C57BL/6 J mice bearing intracranial CT-2A-dmEGFRvIII-Luc tumors (a) and VM/Dk mice bearing intracranial SMA560-dmEGFRvIII-Luc tumors (b). The p-values generated from the generalized Wilcoxon test are provided for both tumor models

Fig. 4

In vivo efficacy of D2C7-IT+αCTLA-4/αPD-1 combination therapies in subcutaneous CT-2A-dmEGFRvIII-Luc tumor-bearing C57BL/6 J mice. (a) Survival curves for groups 1–6 (G1–6) followed up to Day 35 and (b) Groups 4–6 (G4–6) followed up to Day 62 after initial tumor inoculation are presented as SEM. (c) Survival curves are presented as SEM for mice surviving symptom-free to Day 72 post-tumor implantation from different treatment groups and were re-challenged initially in the left flank with 1 × 10⁶ CT-2A parental cells. C57BL/6 J mice (N = 5) were used as naïve controls. (d) Mice surviving symptom-free from different treatment groups after first re-challenge underwent a second re-challenge with 3 × 10⁵ CT-2A-dmEGFRvIII-Luc cells in the brain on Day 126. C57BL/6 J mice (N = 5) were used as naïve controls. Median survival estimates are presented

Fig. 5

In vivo efficacy of D2C7-IT+αCTLA-4/αPD-1 combination therapies in bilateral subcutaneous CT-2A-dmEGFRvIII-Luc tumor-bearing C57BL/6 J mice. (a) The tumor growth curves are presented as SEM for the right treated tumors (upper panel) and left untreated tumors (lower panel). b and c Differences in left tumor volumes were assessed between the vehicle control group (G1) and all other treatment groups (G2–6) on Day 35 (b) and between D2C7-IT monotherapy group (G2) and D2C7-IT+αCTLA-4/αPD-1 combination therapy groups (G5–6) on Day 43 (c) and are not adjusted for multiple testing

Fig. 6

Treatment of CT2A-dmEGFRvIII-Luc intracranial tumors with D2C7-IT and αPD-1/αCTLA-4 combinations. (a) Survival curves and median survival estimates for CT2A-dmEGFRvIII-Luc tumor-bearing mice treated with vehicle control, D2C7-IT, αPD-1, and αCTLA-4 mono or combination therapies. The p-values were generated from the generalized Wilcoxon test and are not adjusted for multiple testing. (b) Median survival was estimated for mice surviving symptom-free to Day 100 post-tumor implantation from different treatment groups and were re-challenged in the contralateral hemisphere of the brain with 1.5 × 10⁵ CT-2A-dmEGFRvIII-Luc cells. C57BL/6 J mice (N = 5) were used as naïve controls

Fig. 7

Anti-tumor efficacy of D2C7-IT and immune checkpoint inhibitor combinations in orthotopic glioma models. a and c Survival curves and median survival estimates data are shown for C57BL/6 J mice bearing intracranial CT-2A-dmEGFRvIII-Luc tumors (a) and VM/Dk mice bearing intracranial SMA560-dmEGFRvIII-Luc tumors (c) treated with vehicle control, D2C7-IT monotherapy, or D2C7-IT+αPD-1/αPD-L1/αTim-3/αLag-3/αCD73 combinations. The p-values generated from the generalized Wilcoxon test are provided for both tumor models and are not adjusted for multiple testing. b and d Median survival for mice that survived to Day 77, and were re-challenged in the contralateral hemisphere of the brain with 2 × 10⁵ CT-2A (b), or 3 × 10⁴ SMA560 (d) parental cells. C57BL/6 J mice (N = 5) or VM/Dk (N = 5) were used as naïve controls