Fully human antibody VH domains to generate mono and bispecific CAR to target solid tumors

Abstract

Background: Chimeric antigen receptor (CAR) T cells are effective in B-cell malignancies. However, heterogeneous antigen expression and antigen loss remain important limitations of targeted immunotherapy in solid tumors. Therefore, targeting multiple tumor-associated antigens simultaneously is expected to improve the outcome of CAR-T cell therapies. Due to the instability of single-chain variable fragments, it remains challenging to develop the simultaneous targeting of multiple antigens using traditional single-chain fragment variable (scFv)-based CARs.

Methods: We used Humabody V_H domains derived from a transgenic mouse to obtain fully human prostate-specific membrane antigen (PSMA) V_H and mesothelin (MSLN) V_H sequences and redirect T cell with V_H based-CAR. The antitumor activity and mode of action of PSMA V_H and MSLN V_H were evaluated in vitro and in vivo compared with the traditional scFv-based CARs.

Results: Human V_H domain-based CAR targeting PSMA and MSLN are stable and functional both in vitro and in vivo. V_H modules in the bispecific format are capable of binding their specific target with similar affinity as their monovalent counterparts. Bispecific CARs generated by joining two human antibody V_H domains can prevent tumor escape in tumor with heterogeneous antigen expression.

Conclusions: Fully human antibody V_H domains can be used to generate functional CAR molecules, and redirected T cells elicit antitumoral responses in solid tumors at least as well as conventional scFv-based CARs. In addition, V_H domains can be used to generate bispecific CAR-T cells to simultaneously target two different antigens expressed by tumor cells, and therefore, achieve better tumor control in solid tumors.

Keywords: CAR-T; Humabody; MSLN; PSMA; immunotherapy; tumor escape.

Figure 1

Human antibody V_H domain-based CAR targeting PSMA is expressed and signals in T cells. (A) Schematic diagram of J591 and PSMA-V_H constructs. (B, C) Representative flow cytometry plots (B) and summary (C) illustrating J591 and PSMA-V_H expression in T cells. The CD19-specific CAR (CD19) and non-transduced T cells (NT) were used as positive and negative controls, respectively. ****P<0.0001, one-way ANOVA. (D) In vitro expansion of CD19, J591, PSMA-V_H and NT T cells; error bars represent SD, (n=4). P≥0.05 by one-way ANOVA. (E) T cell subset composition based on CD45RA and CCR7 expression in CD19, J591, PSMA-V_H and nt T cells at day 14 of culture; error bars represent SD, (n=4). P≥0.05 by one-way ANOVA. (F) Western blots detecting phosphorylation of CAR-CD3ζ, Akt and ERK in J591 and PSMA-V_H T cells activated via CAR cross-linking with an anti-FLAG ab followed by incubation with a secondary ab to induce the aggregation of car molecules. Total CAR.CD3ζ and endogenous CD3ζ were used as loading controls. Data are representative of two experiments. ANOVA, analysis of variance; CAR, chimeric antigen receptor; MSLN, mesothelin; PSMA, prostate-specific membrane antigen.

Figure 2

T cells expressing the human antibody V_H domain-based CAR targeting PSMA are functional in vitro. (A) Representative flow cytometry plots showing the expression of PSMA in C4-2, PC3 and PC3 cells engineered with a retroviral vector to express PSMA. (B, C) Rpresentative flow cytometry plots (B) and summary (C) illustrating Granzyme-B expression of T cells expressing either J591 or PSMA-V_H cocultured overnight with a tumor cell line expressing PSMA (PC3-PSMA-eGFP) at E:T ratio of 1:2; error bars represent SD, (n=4). P≥0.05 by t-test. (D, E) Representative flow cytometry plots (D) and summary (E) illustrating the kinetics of CD69 expression of T cells expressing either J591 or PSMA-V_H and cocultured overnight with a tumor cell line expressing PSMA (PC3-PSMA-eGFP) at E:T ratio of 1:2. Data are representative of 4 experiments. ***P<0.001 two-way ANOVA. (F) Representative flow cytometry plots showing coculture of CD19, J591 and PSMA-V_H T cells with C4-2-eGFP, PC3-eGFP and PC3-PSMA-eGFP. T cells were cocultured with tumor cells at an E:T ratio of 1:5 for 6 days. At day 6, all cells were collected and analyzed by flow cytometry to quantify tumor cells (GFP) and T cells (CD3), respectively. (G) Summary of coculture of CD19, J591 and PSMA-V_H T cells with tumor cells in (F); error bars represent SD, (n=4). ****P<0.0001, two-way ANOVA. (H, I) IFN-γ (H) and IL-2 (I) were detected by ELISA in the coculture supernatant of cocultures of CD19, J591 and PSMA-V_H T cells with tumor cells illustrated in (F); error bars represent SD, (n=5). *P<0.05, **p<0.01, ***p<0.001, ****p<0.0001, two-way ANOVA. (J) Representative flow cytometry plots showing the proliferation of J591 and PSMA-V_H T cells in response to tumor cells as assed by CFSE dilution. Data are representative of four experiments. ANOVA, analysis of variance; CAR, chimeric antigen receptor; E:T, effector to target ratio; IFN-γ, interferon-γ; IL-2, interleukin 2; MSLN, mesothelin; PSMA, prostate-specific membrane antigen.

Figure 3

T cells expressing the human antibody V_H domain-based CAR targeting PSMA are functional in vivo. (A) Schematic of the metastatic prostate cancer model using PC3-PSMA-FFluc-eGFP tumor cells in NSG mice and treatment with CD19, J591 and PSMA-V_H T cells (n=5 mice per group). (B) Representative images of tumor bioluminescence (BLI) at selected time points post T cell injections. (C) Kinetics of tumor BLI post T cell injections. (D) Schematic of the metastatic prostate cancer model using PC3-PSMA-FFluc-eGFP tumor cells in NSG mice and treatment with low dose of CD19, J591 and PSMA-V_H T cells (n ≥4 mice per group). (E) Representative images of tumor BLI at selected time points post-T cell injections for low dose of T cells. (F) Kinetics of tumor BLI post T cell injections low dose of T cells. (G, H) Percentage of T cells in the gate of live cells (G) and total cell numbers (H) in blood, spleen and bone marrow from PC3-PSMA-bearing mice treated with low doses of CAR-T cells. Mice were euthanized at day 58 after CAR-T cells infusion and T cells were identified as CD45⁺CD3⁺ cells by flow cytometry. J594 group (n=5), PSMA-V_H group (n=4). P≥0.05 by t-test. CAR, chimeric antigen receptor; BLI, bioluminescence; PSMA, prostate-specific membrane antigen.

Figure 4

T cells expressing the human antibody V_H domain-based CAR targeting MSLN demonstrate antitumor activity. (A) Schematic diagram of MSLN-scFv and MSLN-heavy-chain-only (MSLN-V_H) CAR constructs. (B) Summary of coculture of CD19, MSLN.scFv and MSLN-V_H T cells with Aspc-1-eGFP (MSLN⁺) and PC3-eGFP (MSLN^-) tumor cell lines. T cells were cocultured with tumor cells at an E:T ratio of 1:5 for 6 days. At day 6, all cells were collected and analyzed by flow cytometry to quantify tumor cells and T cells, respectively. Error bars represent SD, (n=4). ****P<0.0001, two-way ANOVA. (C) IFN-γ (upper panel) and IL-2 (lower panel) detected in the supernatants of the cocultures illustrated in (B) as measured by ELISA; error bars represent SD, (n=4). ****P<0.0001, two-way ANOVA. (D) Representative flow cytometry plots showing the proliferation of MSLN.scFv and MSLN-V_H T cells in response to tumor cells as assessed by CFSE dilution. Data are representative of three experiments. (E) Schematic of the metastatic pancreatic cancer model using Aspc-1-FFluc-eGFP tumor cells in NSG mice. (F, G) Representative tumor BLI (F) and BLI kinetics (G) of Aspc-1-FFluc-eGFP tumor growth at the representative time points post T cell injections. (n=5 mice per group). (H) Kaplan-Meier survival curve of mice in (E) (n=5 mice per group). Data are representative of two experiments. (I) Frequency of human CD45⁺CD3⁺cells in blood at 22 days (left) post-T-cell infusion and at euthanasia (right) of MSLN-scFv and MSLN-V_H T cells, respectively. Data are shown as individual values and the mean (n = 5 mice per group). P≥0.05 by t-test. ANOVA, analysis of variance; BLI, bioluminescence; CAR, chimeric antigen receptor; CFSE, carboxyfluorescein diacetate succinimidyl ester; E:T, effector to target ratio; IFN-γ, interferon-γ; IL-2, interleukin 2; i.v, intravenous; MSLN, mesothelin; scFv, single-chain fragment variable.

Figure 5

In vitro analysis of monospecific and bispecific Humabody V_H binding. (A) Schematic representation of monospecific (single V_H) or bispecific (double V_H) proteins. (B) Single cycle BIAcore kinetic analysis of PSMA binding. (C) BIAcore kinetic analysis of MSLN binding, threefold dilution series starting at 300 nM, except the control scFv protein which started at 33.3 nM. Data are representative of two experiments. MSLN, mesothelin; PSMA, prostate-specific membrane antigen; scFv, single-chain fragment variable.

Figure 7

T cells expressing two human antibody V_H domain-based CARs demonstrate dual specificity in vivo. (A) Schematic of the xenograft mouse model in which NSG mice were systemically engrafted with mixed FFluc-eGFP labeled PC3-PSMA (5×10⁵ cells) and PC3-MSLN (5×10⁵ cells) cells at 1:1 ratio, and treated with two doses of CAR-T cells at day 0 and day 7, respectively (6×10⁶ cells each dose, n=5 mice per group). (B, C) Representative tumor BLI images (B) and BLI kinetics (C) at selected time points post T cell injections. (D) Number of human CD45⁺CD3⁺ cells in the peripheral blood collected at day 21 post second T-cell infusion in mice treated as described in (A). Data are shown as individual values and the mean (n=5 mice per group) and are representative of two experiments, p≥0.05 by one-way ANOVA. (E) Representative antigen expression pattern in the tumor cells isolated from the mice with relapsed tumor in mice treated as described in (A). ANOVA, analysis of variance; BLI, bioluminescence; CAR, chimeric antigen receptor; MSLN, mesothelin; PSMA, prostate-specific membrane antigen.

Figure 6

T cells expressing two human antibody V_H domain-based CARs demonstrate dual specificity in vitro. (A) Schematic diagram of PSMA-V_H, MSLN-V_H, and PSMA/MSLN-V_H CAR constructs. (B, C) Representative flow cytometry plots (B) and summary (C) illustrating CAR expression in T cells. The CD19-specific CAR (CD19) was used as negative controls. P≥0.05 by one-way ANOVA. (D) Representative flow cytometry plots showing PC3-PSMA-eGFP (PSMA target), PC3-MSLN-eGFP (MSLN target) and mixture of PC3-PSMA-eGFP and PC3-MSLN-eGFP (1:1 ratio) cotultured with CD19.CAR, PSMA-V_H.CAR, MSLN-V_H.CAR and PSMA/MSLN-V_H.CAR T cells at the E:T ratio of 1:5 for 6 days. Tumor cells and T cells were quantified at day six by flow cytometry. (E) Summary of coculture experiments illustrated in (D); error bars represent SD, (n=5). *P<0.05, **p<0.01, ****p<0.0001, two-way ANOVA. (F, G) IFN-γ (F) and IL-2 (G) detected in the coculture supernatant of the coculture experiments described in (D) as measured by ELISA; error bars represent SD, (n=3) *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, two-way ANOVA. ANOVA, analysis of variance; CAR, chimeric antigen receptor; E:T, effector to target ratio; IFN-γ, interferon-γ; IL-2, interleukin; MSLN, mesothelin; NT, non-transduced; PSMA, prostate-specific membrane antigen.