Vδ2 T-cell engagers bivalent for Vδ2-TCR binding provide anti-tumor immunity and support robust Vγ9Vδ2 T-cell expansion

Abstract

Background: Vγ9Vδ2 T-cells are antitumor immune effector cells that can detect metabolic dysregulation in cancer cells through phosphoantigen-induced conformational changes in the butyrophilin (BTN) 2A1/3A1 complex. In order to clinically exploit the anticancer properties of Vγ9Vδ2 T-cells, various approaches have been studied including phosphoantigen stimulation, agonistic BTN3A-specific antibodies, adoptive transfer of expanded Vγ9Vδ2 T-cells, and more recently bispecific antibodies. While Vγ9Vδ2 T-cells constitute a sizeable population, typically making up ~1-10% of the total T cell population, lower numbers have been observed with increasing age and in the context of disease.

Methods: We evaluated whether bivalent single domain antibodies (VHHs) that link Vδ2-TCR specific VHHs with different affinities could support Vγ9Vδ2 T-cell expansion and could be incorporated in a bispecific engager format when additionally linked to a tumor antigen specific VHH.

Results: Bivalent VHHs that link a high and low affinity Vδ2-TCR specific VHH can support Vγ9Vδ2 T-cell expansion. The majority of Vγ9Vδ2 T-cells that expanded following exposure to these bivalent VHHs had an effector or central memory phenotype and expressed relatively low levels of PD-1. Bispecific engagers that incorporated the bivalent Vδ2-TCR specific VHH as well as a tumor antigen specific VHH triggered antitumor effector functions and supported expansion of Vγ9Vδ2 T-cells in vitro and in an in vivo model in NOG-hIL-15 mice.

Conclusion: By enhancing the number of Vγ9Vδ2 T-cells available to exert antitumor effector functions, these novel Vδ2-bivalent bispecific T cell engagers may promote the overall efficacy of bispecific Vγ9Vδ2 T-cell engagement, particularly in patients with relatively low levels of Vγ9Vδ2 T-cells.

Keywords: Vγ9Vδ2 T-cells; bispecific T-cell engager; cancer; expansion; immunotherapy; single domain antibody.

Figure 1

Bivalent Vδ2-TCR-specific VHHs support enrichment and expansion of Vγ9Vδ2 T-cells. (A) Binding of bivalent Vδ2-VHHs to Vγ9Vδ2 T-cells assessed using flow cytometry (n=3) or ELISA (n=2). (B) Overview of experimental design to assess expansion. (C) Representative gating strategy to asses enriched and expanded Vγ9Vδ2 T-cells within PBMC. (D) Enrichment (left panel) and fold expansion (right panel) of Vγ9Vδ2 T-cells during an 8 day culture of healthy donor PBMC in the presence or absence of 1 nM bivalent Vδ2-VHHs or 10 µM pamidronate (n=3-6). (E) Binding of the 6H4-5C7 bivalent VHH with either a 5, 10 or 20 AA linker to Vγ9Vδ2 T-cells (n=3). (F) Enrichment (left panel) and fold expansion (right panel) of Vγ9Vδ2 T-cells during an 8 day culture of healthy donor PBMC in the presence or absence of 0.001, 0.1, 1 or 100 nM of the 6H4-5C7 bivalent VHH with either 5, 10 or 20 AA linker or 10 µM pamidronate. Data in (A) (left panel) and (D–F) assessed using flow cytometry. Data in A (right panel) assessed using ELISA. Data represent mean and SEM (A, E) or individual data-points are indicated using open circles and box and whisker plots indicate the median, 25th to 75th percentiles and minimum to maximum (D, F). Two-way ANOVA with Tukey’s multiple comparisons test (F) were used for statistical analysis; P=< 0.05: *.

Figure 2

TAA-Vδ2^hi-lo bsVHH and TAA-Vδ2^hi-lo bsVHH-Fc support Vγ9Vδ2 T-cell expansion and trigger Vγ9Vδ2 T-cell degranulation and tumor cell lysis. (A) Illustration of the TAA-Vδ2^hi-lo bsVHH with and without Fc domain. (B) Binding of the TAA-Vδ2^hi-lo bsVHH with and without Fc to Vγ9Vδ2 T-cells. Data represent mean and SEM (n=3). (C) Enrichment (left panel) and fold expansion (right panel) of Vγ9Vδ2 T-cells during an 8 day culture of healthy donor PBMC in the presence or absence of 1 nM TAA-Vδ2^hi-lo bsVHH, 100nM TAA-Vδ2^hi-lo bsVHH-Fc or 10 µM pamidronate. Individual data-points are indicated using open circles and box and whisker plots indicate the median, 25th to 75th percentiles and minimum to maximum (n=15-30). (D) Binding of the TAA-Vδ2^hi-lo bsVHH and TAA-Vδ2^hi-lo bsVHH-Fc to SW480 (EGFR⁺), 22Rv1 (PSMA⁺) or MM.1s.CD1d (CD1d⁺) tumor cells. Data represent mean and SEM (n=3). (E) Vγ9Vδ2 T-cell CD107a expression after 24 hr co-cultures of Vγ9Vδ2 T-cells and SW480, 22Rv1 or MM.1s.CD1d tumor cells (1:1 E:T ratio) ± concentration range of the TAA-Vδ2^hi-lo bsVHH and TAA-Vδ2^hi-lo bsVHH-Fc. Data represent mean and SEM (n=3). (F) Lysis of SW480, 22Rv1 or MM.1s.CD1d tumor cells after 24 hr incubation with Vγ9Vδ2 T-cells (1:1 E:T ratio) ± concentration range of the TAA-Vδ2^hi-lo bsVHH and TAA-Vδ2^hi-lo bsVHH-Fc. Data represent mean and SEM (n=3-6). Data generated using flow cytometry. One-way ANOVA with Tukey’s multiple comparisons test was used for statistical analysis and asterisks are shown compared to IL-2 control; P=< 0.01: **, P=< 0.001: ***, P=< 0.0001: ****.

Figure 3

Vγ9Vδ2 T-cells expanded using TAA-Vδ2^hi-lo bsVHH and TAA-Vδ2^hi-lo bsVHH-Fc are activated and induce tumor lysis when exposed to TAA-expressing tumor cells. (A) Schematic overview of method used for PBMC culture, subsequent assessment of Vγ9Vδ2 T-cell expansion, as well as co-culture of enriched Vγ9Vδ2 T-cells and tumor cells. (B–E) Vγ9Vδ2 T-cell CD25 expression (A; n=4-12), CD107a expression (B; n=4-12), specific tumor cell lysis (C; n=4-12) and levels of IL-2, TNF and IFN-γ (pg/ml, D; n=4-5) in 24hr co-cultures of SW480 (EGFR⁺), 22Rv1 (PSMA⁺) or MM.1s.CD1d (CD1d⁺) tumor cells with Vγ9Vδ2 T-cells enriched (purity > 60%) from 8 day cultures of healthy donor PBMC in the presence or absence of 1 nM TAA-Vδ2^hi-lo bsVHH, 100 nM TAA-Vδ2^hi-lo bsVHH-Fc or 10 µM pamidronate (1:1 E:T ratio). Data generated through flow cytometry (B–D) or CBA (E). Individual data-points are indicated using open circles and box and whisker plots indicate the median, 25th to 75th percentiles and minimum to maximum. One-way ANOVA with Tukey’s multiple comparisons test was used for statistical analysis; P=< 0.05: *, P=< 0.01: **, P=< 0.001: ***, P=< 0.0001: ****.

Figure 4

Bivalent Vδ2^hi-lo VHH, TAA-Vδ2^hi-lo bsVHH and TAA-Vδ2^hi-lo bsVHH-Fc support Vγ9Vδ2 T-cell expansion in cancer patient PBMC and expanded Vγ9Vδ2 T-cells display a memory dominated phenotype. (A) Enrichment (left panel) and fold expansion (right panel) of Vγ9Vδ2 T-cells during an 8 day culture of cancer patient PBMC in the presence or absence of 1nM bivalent Vδ2^hi-lo VHH, 1 nM TAA-Vδ2^hi-lo bsVHH, 100nM TAA-Vδ2^hi-lo bsVHH-Fc or 10 µM pamidronate (n=5-10). (B, C) Proportion (%) of central memory (CD27⁺ CD45RA^-), effector memory (CD27^- CD45RA^-), terminally differentiated (CD27^- CD45RA⁺) or naïve (CD27⁺ CD45RA⁺) cells within total Vγ9Vδ2 T-cell population before (baseline) and after an 8 day culture in the presence or absence of 1 nM TAA-Vδ2^hi-lo bsVHH, 100nM TAA-Vδ2^hi-lo bsVHH-Fc or 10 µM pamidronate (n=5-16) using healthy donor PBMC (B) or cancer patient PBMC (C). (D, E) Expression of CD25, HLA-DR, DNAM-1, NKG2D, NKG2A, PD-1, CTLA-4 and TIGIT on Vγ9Vδ2 T-cells expanded for 8 days in the presence or absence of 1 nM TAA-Vδ2^hi-lo bsVHH, 100nM TAA-Vδ2^hi-lo bsVHH-Fc or 10 µM pamidronate (data reflect % positive of total Vγ9Vδ2 T-cell fraction; n=5-17) using healthy donor PBMC (D) or cancer patient PBMC (E). Data generated using flow cytometry. Individual data-points are indicated using open circles and box and whisker plots indicate the median, 25th to 75th percentiles and minimum to maximum (A, D, E). One-way ANOVA with Tukey’s multiple comparisons test was used for statistical analysis and asterisks are shown compared to IL-2 control; P=< 0.05: *, P=< 0.01: **, P=< 0.001: ***, P=< 0.0001: ****.

Figure 5

CD1d-Vδ2^hi-lo bsVHH induces Vγ9Vδ2 T-cell expansion in NOG-hIL-15 mice. NOG-hIL-15 mice were sub-lethally irradiated and i.v. inoculated with human PBMC on day 0 and treated i.p. with PBS (control group) or 0.5 mg/kg CD1d-Vδ2^hi-lo bsVHH on day 0 and 4 (n=4 mice per group). (A) Timeline of the in vivo study. (B) Enrichment (left panel) and fold expansion (right panel) of human PBMC-derived Vγ9Vδ2 T-cells in vivo 8 days after PBMC inoculation and two doses of i.p. injection with PBS or CD1d-Vδ2^hi-lo bsVHH. (C) Enrichment of human PBMC-derived Vγ9Vδ2 T-cells in spleen, liver and lungs 8 days after PBMC inoculation and two doses of i.p. injection with PBS or CD1d-Vδ2^hi-lo bsVHH. (D) Phenotype of expanded human PBMC-derived Vγ9Vδ2 T-cells in vivo from blood, spleen, liver and lungs 8 days after PBMC inoculation and two doses of i.p. injection with PBS or CD1d-Vδ2^hi-lo bsVHH. Data generated using flow cytometry. Individual data-points are indicated using open circles and box and whisker plots indicate the median, 25th to 75th percentiles and minimum to maximum. Unpaired t tests were used for statistical analysis; P=< 0.01: **, P=< 0.001: ***.