Effective cancer immunotherapy combining mRNA-encoded bispecific antibodies that induce polyclonal T cell engagement and PD-L1-dependent 4-1BB costimulation

Abstract

Background: Immune checkpoint inhibitors have revolutionized cancer therapy, but many patients fail to respond or develop resistance, often due to reduced T cell activity. Costimulation via 4-1BB has emerged as a promising approach to enhance the effector function of antigen-primed T cells. Bispecific T cell-engaging (TCE) antibodies are an effective way to provide tumor-specific T cell receptor-mediated signaling to tumor-infiltrating lymphocytes. mRNA-based delivery of bispecific antibodies, offer a novel approach to enhance tumor-specific immune responses while minimizing adverse effects.

Methods: Two bispecific antibodies were generated: the EGFR x CD3 TCE antibody (LiTE) and the PD-L1 x 4-1BB costimulatory antibody (LiTCo), which was further fused to a high FcRn albumin variant (Albu-LiTCo). The mRNA encoding these bispecific antibodies contains an N1-methylpseudouridine modified nucleoside and regulatory sequences to ensure proper expression and stability. A series of in vitro assays and cell-based analyses were performed to characterize both antibodies. The in vivo efficacy of the mRNA-encoded bispecific antibodies was evaluated in xenograft tumor models expressing EGFR.

Results: We investigated the combined effect of two mRNA-encoded Fc-free bispecific antibodies with complementary mechanisms of action: an EGFR-targeting TCE and a half-life extended PD-L1 x 4-1BB costimulatory antibody. The mRNAs encoding both bispecific LiTE^RNA and Albu-LiTCo^RNA, showed similar binding specificity and in vitro function to their protein analogues. Pharmacokinetic studies demonstrated sustained expression of both bispecific antibodies following intravenous administration of the mRNAs formulated using a polymer/lipid-based nanoparticle (LNP) but different pharmacokinetic profiles, shorter for the TCE and longer for the PD-L1 x 4-1BB. When administered as a mRNA-LNP combination (Combo^RNA), the growth of EGFR-positive tumors in immunocompetent mice was significantly inhibited, resulting in tumor regression in 20% of cases with no associated toxicity. Histological analysis confirmed increased T cell infiltration in the tumors treated with LITE^RNA and Combo^RNA. Repeated administration resulted in sustained production of bispecific antibodies with different exposure cycles and potent antitumor activity with a favorable safety profile.

Conclusions: These results highlight the potential of combining two mRNA-encoded bispecific antibodies with different mechanisms of action and programmable half-life for cancer immunotherapy.

Keywords: T-cell engager; cancer immunotherapy; combined RNA; costimulatory antibody; mRNA encoded bispecific antibodies.

Figure 1

Functional characterization and mechanisms of action of LiTE and Albu-LiTCo antibodies. Schematic representation of EGFR-targeted LiTE and Albu-LiTCo antibodies (A). LiTE-mediated EGFR-dependent activation of tumor cells (B). CT26 and CT26^EGFR were cocultured with mouse splenocytes (E:T ratio 5:1) in the presence of LiTE or anti-CD3 IgG at 3.34 nM equimolar concentration. Mouse CD69 expression was measured by flow cytometry after 24 hours (B). LiTE-mediated EGFR-dependent cytotoxicity of tumor cells (C). Luciferase-expressing CT26 and CT26^EGFR cells were cocultured with splenocytes (E:T ratio 5:1) and different concentrations of LiTE. Specific lysis was measured by bioluminescence at 48 hours. Percent specific lysis was calculated relative to control. Data are mean ± SD (n = 3). Significance was calculated by unpaired Student’s t test. Competition ELISA for PD-L1/PD-1 interaction inhibition by anti-PD-L1 IgG or Albu-LiTCo (D). Data are mean ± SD (n = 3). One representative experiment from two independent experiments is shown. LogIC50 is shown. Blocking activity of Albu-LiTCo in a cell-based bioassay (E). JurkatPD-1 cells were cocultured with CHO^PD-L1 cells and increasing concentrations of anti-PD-L1 IgG or Albu-LiTCo. Luminescence was measured after 6 hours. Data are expressed as fold induction relative to unstimulated JurkatPD-1 cells. Representative dose-concentration curves are shown as mean ± SD (n = 3). IC50 is shown. Agonistic activity of Albu-LiTCo (F). Wild type CHO cells or CHO^PD-L1/EGFR were cocultured with LiTE-activated mouse CD8a⁺ T cells (3.34 nM) in the presence of Albu-LiTCo, anti-4-1BB IgG or anti-PD-L1 IgG at 6.67 nM equimolar dose. IFNγ secretion was determined after 72 h Negative controls (–) consisted of CD8⁺ T cells cultured with either CHO or CHO^PD-L1/EGFR, and in the absence (- control in the left) or presence (- control in the right) of soluble LiTE. Data are mean ± SD (n = 3). One representative experiment of three independent experiments is shown. Significance was calculated by unpaired Student’s t test.

Figure 2

Functional characterization of LiTE^RNA and Albu-LiTCo^RNA. Western blot detection using a HRP-conjugated anti-V_HH mAb in the conditioned media of LiTE^RNA and Albu-LiTCo^RNA transfected HEK293 cells. Molecular mass (kDa) is indicated (A). Specific binding of LiTE^RNA (B) and Albu-LiTCo^RNA (C, D) to plastic immobilized specific antigens (hEGFR and m4-1BB, mPD-L1, respectively) demonstrated by ELISA using an HRP-conjugated anti-V_HH mAb cocktail (B, C) or HRP-conjugated anti-human serum albumin (HRP-HSA) mAb (D). Data are represented as a mean ± SD (n = 3). BSA, bovine serum albumin; NFDM, non-fat dry milk. Binding of LiTE^RNA and Albu-LiTCo^RNA to cell surface expressed antigens mCD3 (E), m4-1BB (F), hEGFR (G) and hPDL-1 (H) by flow cytometry. The y-axis shows the relative cell number, and the x-axis represents the R-phycoerythrin (PE)-fluorescence, expressed on a linear scale. The anti-CD3 IgG, anti-4-1BB IgG, anti-EGFR IgG and anti-PD-L1 IgG were used as controls. One representative experiment out of two independent experiments is shown. The number indicates the percentage of positive cells (%).

Figure 3

Effects of LiTE^RNA and Albu-LiTCo^RNA on tumor cell cytotoxicity, agonistic costimulatory activity and tumor growth inhibition. EGFR-dependent cytotoxicity by LiTE^RNA (A). Luciferase-expressing CT26 and CT26^EGFR cells were co-cultured with mouse splenocytes at an E:T ratio of 5:1. Specific tumor cell lysis was measured by bioluminescence. The percentage of specific lysis was calculated relative to the same number of tumor cells cultured with splenocytes. Data are expressed as mean ± SD (n = 3). Significance was calculated by unpaired Student’s t test. CHO or CHO^PD-L1/EGFR cells were plated with mouse CD8a⁺ T cells activated with LiTE^RNA in the presence of Albu-LiTCo (B) or with LiTE in the presence of Albu-LiTCo^RNA (C). IFNγ secretion was determined after 72 hours. Data are expressed as mean ± SD (n = 3). Significance was calculated by unpaired Student’s t test. Anti-4-1BB IgG and anti-PD-L1 IgG (B) or EGFP^RNA (C) were used as controls. Pharmacokinetic profile of LiTE^RNA (solid blue line) and Albu-LiTCo^RNA (solid red line) expressed as ng/mL (left Y-axis) after a single intravenous (i.v.) administration of mRNA-LNP in BALB/c mice (D). Ex vivo specific tumor lysis of CT26^EGFR cells mediated by LiTE-containing mouse serum (dashed blue line) and Albu-LiTCo-containing mouse serum (dashed red line) expressed as % of Lysis (right Y-axis) (D). Target cells modified for the expression of luciferase were co-cultured with splenocytes at the effector/target (E/T) ratio of 5:1 in the presence of mouse serum obtained at 4, 96 and 168 hours post-mRNA-LNP administration. After 48 hours, the percentage of specific tumor lysis was measured by bioluminescence (right Y-axis). Data are presented as mean ± SD (n = 3). BALB/c mice were subcutaneously (s.c.) inoculated with CT26^EGFR cells, randomized into n=5/group with similar mean tumor sizes and SDs and treated with 3 i.v. injections of LiTE^RNA, Albu-LiTCo^RNA (10 µg/mouse) as monotherapy or combined (Combo^RNA) (E). Average tumor volume growth of mice in each group is shown. Data are presented as mean ± SD. Significance was determined by one-way ANOVA with Tukey’s correction test for multiple comparisons. Quantitative analysis of intratumoral CD8⁺ T cells in mouse tumor tissue (n = 4/group) (F). Data were calculated as percentage of CD8⁺ versus total cell number and presented as mean ± SD. Significance was determined by one-way ANOVA with Tukey’s test correction for multiple comparisons.