Impact of the selective A2AR and A2BR dual antagonist AB928/etrumadenant on CAR T cell function

Abstract

Background: Chimeric antigen receptor (CAR) T cell therapy has been successfully translated to clinical practice for the treatment of B cell malignancies. The suppressive microenvironment of many malignancies is a bottleneck preventing treatment success of CAR T cells in a broader range of tumours. Among others, the immunosuppressive metabolite adenosine is present in high concentrations within many tumours and dampens anti-tumour function of immune cells and consequently therapeutic response.

Methods: Here, we present the impact of the selective adenosine A2_A and A2_B receptor antagonist AB928/etrumadenant on CAR T cell cytokine secretion, proliferation, and cytotoxicity. Using phosphorylation-specific flow cytometry, we evaluated the capability of AB928 to shield CAR T cells from adenosine-mediated signalling. The effect of orally administered AB928 on CAR T cells was assessed in a syngeneic mouse model of colon carcinoma.

Results: We found that immunosuppressive signalling in CAR T cells in response to adenosine was fully blocked by the small molecule inhibitor. AB928 treatment enhanced CAR T cell cytokine secretion and proliferation, granted efficient cytolysis of tumour cells in vitro and augmented CAR T cell activation in vivo.

Conclusions: Together our results suggest that combination therapy with AB928 represents a promising approach to improve adoptive cell therapy.

Fig. 1. Adenosine inhibits CAR T cell activation.

a Schematic illustration of the rationale of combining CAR T cell therapy and AB928. b, c In all, 2.5 × 10⁴ murine anti-EpCAM CAR T cells were cocultured with 2.5 × 10⁴ Panc02-EpCAM tumour cells for 24 h in the presence or absence of b NECA (titration from 0.1 nM to 10 µM) or c adenosine (titration from 1 µM to 100 µM + EHNA 2.5 µM). b IFN-γ, IL-2 and TNF-α concentrations in the coculture supernatant were determined by ELISA. Data were normalised to the vehicle control condition. c IFN-γ concentrations in the coculture supernatant were determined by ELISA. b, c Data are shown as mean ± SEM of n = 3 independent experiments with each dot representing the mean value of an individual experiment. *P < 0.05 by two-sided t test.

Fig. 2. AB928 protects CAR T cells from adenosine mediated suppression.

a In all, 2.5 × 10⁴ murine anti-EpCAM CAR T cells were cocultured with 2.5 ×10⁴ Panc02-EpCAM tumour cells for 24 h in the presence or absence of NECA (100 nM) and AB928 (titration from 0.1 nM to 10 µM). IFN-γ, IL-2 and TNF-α concentrations in the coculture supernatant were determined by ELISA. Data were normalised to the vehicle control condition. b Also, 2.5 ×10⁴ murine anti-EpCAM CAR T cells were cocultured with 2.5 × 10⁴ 4T1, LL/2-EpCAM, T110299-EpCAM and CT26-EpCAM tumour cells for 24 h cells in the presence or absence of NECA (100 nM) and AB928 (100 nM). Cytokine concentrations in the coculture supernatant were determined by IFN-γ ELISA. c Intracellular staining of murine anti-EpCAM CAR T cells activated with plate bound recombinant EpCAM for 18 h in the presence or absence of NECA (1 µM) and AB928 (1 µM). a–c Data are shown as mean ± SEM of a, c n = 3 or b n = 4–5 independent experiments with each dot representing the mean value of an individual experiment. *P < 0.05, **P < 0.01, ***P < 0.001 by one-way ANOVA.

Fig. 3. AB928 enables efficient CAR T cell effector responses in vitro.

a RTCA of coculture with 5 × 10⁴ murine anti-EpCAM CAR T cells and 2.5 × 10⁴ Panc02-EpCAM tumour cells in the presence or absence of NECA (1 µM) and AB928 (1 µM). CAR T cells and treatments were added at the timepoint indicated by the arrow. b In all, 2.5 × 10⁴ murine anti-EpCAM CAR T cells were cocultured with 2.5 × 10⁴ Panc02-EpCAM tumour cells for 24 h in the presence or absence of NECA (100 nM) and AB928 (100 nM). GzmB concentrations in the coculture supernatant were determined by ELISA. c, e In all, 10⁵ murine anti-EpCAM CAR T cells were activated with plate-bound recombinant EpCAM for c 18 h or e 48 h in the presence or absence of NECA (1 µM) and AB928 (1 µM). c Phenotypic characterisation of CAR T cells by flow cytometry. d Also, 10⁵ murine anti-EpCAM CAR T cells were cocultured with 2.5 × 10⁵ CT26-EpCAM tumour cells. Expression of inhibitory receptors on CD8⁺CAR⁺ T cells by flow cytometry. e CAR T cell proliferation was determined by flow cytometry with counting beads at the beginning and end of the experiment. Data were normalised to baseline proliferation of unstimulated cells. a, c Representative experiment of n = 3 independent experiments. Data represents mean of technical replicates. b, e Data are shown as mean ± SEM of b n = 7 or e n = 3 independent experiments with each dot representing the mean value of an individual experiment. d Representative experiment of n = 3 independent experiments. Data are shown as mean ± SEM of technical replicates. *P < 0.05, **P < 0.01, ***P < 0.001 by one-way ANOVA.

Fig. 4. AB928 shields CAR T cells from immunosuppressive signalling in response to adenosine.

a, b Phospho-specific flow cytometry of CREB of murine anti-EpCAM CAR T cells after treatment with NECA (5 µM) and/or AB928 (titration from 10 nM to 10 µM). Data are shown as a representative histogram and b mean ± SEM of n = 3 independent experiments with each dot representing the mean value of an individual experiment. *P < 0.05, **P < 0.01 by one-way ANOVA.

Fig. 5. Orally administered AB928 augments CAR T cell activation in vivo.

a Schematic illustration of the experimental setup. b–d CAR⁺ T cells in the spleen were tracked and phenotyped by flow cytometry at the end of the experiment. Data are shown as mean ± SEM of control n = 10 mice and AB928 n = 8 mice. *P < 0.05, ***P < 0.001 by two-sided t test.

Fig. 6. AB928 ameliorates activation of human CAR T cells in the presence of adenosine.

Coculture of 2.5 × 10⁴ human a anti-MSLN-CD28z CAR T cells or b anti-MSLN-4-1BBz CAR T cells and 2.5 × 10⁴ SUIT-2-MSLN tumour cells in the presence or absence of NECA (1 µM) and AB928 (titration from 10 nM to 1 µM). a, b After 24 h, IFN-γ and IL-2 concentrations in the coculture supernatant were determined by ELISA. c Coculture of 10⁵ anti-MSLN-CD28z CAR T cells and 2.5 × 10⁴ SUIT-2-MSLN tumour cells in the presence or absence of NECA (1 µM) and AB928 (1 µM). After 48 h, phenotype of CD8⁺ CAR⁺ T cells was determined by flow cytometry. a–c Data are shown as mean ± SEM of n = 4–8 independent experiments with each dot representing the mean value of an individual experiment. *P < 0.05, **P < 0.01 by one-way ANOVA.