Unusual Partners: γδ-TCR-Based T Cell Therapy in Combination with Oncolytic Virus Treatment for Diffuse Midline Gliomas

Abstract

Due to the minimal survival benefits of existing therapies for pediatric diffuse midline glioma (DMG) patients, new therapeutic modalities are being investigated. Immunotherapies such as CAR-T cells and oncolytic viruses (OVs) are part of these efforts, as evidenced by the increasing number of clinical trials. αβ T cells engineered with a high-affinity γ9δ2 T-cell receptor (TEGs) are immune cells designed to target metabolic changes in malignant or virally infected cells via BTN2A1 and BTN3A. Because the expression of BTN2A1 and BTN3A can be altered in tumor and infected cells, combining TEGs and OVs could potentially enhance the anti-tumor response. We investigated this hypothesis in the following study. We demonstrate that TEGs can indeed target DMG, which expresses BTN2A1 and BTN3A at varying levels, and that OVs can further enhance the expression of BTN3A-but not BTN2A1-in DMG. Functionally, TEGs killed DMG cell cultures, and this killing was further increased after OV infection of the DMGs with either adenovirus Δ24-RGD or reovirus R124 under suboptimal conditions. However, this additive effect was lost when γ9δ2 TCR-ligand interaction was boosted by pamidronate. This study demonstrates the additive effect of combining OVs and Vγ9Vδ2 TCR-engineered immune cells under suboptimal conditions and supports a combination strategy to enhance the efficacy of both therapeutic modalities.

Keywords: ?9?2TCR; D24-RGD; R124; TEGs; diffuse midline glioma; immune-oncology; immunotherapy; oncolytic viruses.

Figure 1

Patient-derived diffuse midline glioma cell cultures expressing BTN2A1 and BTN3A are targeted by TEGs. (A) BTN2A1 and BTN3A expression of SU-DIPG-IV and VUMC-DIPG-G, as normalized median fluorescent intensity (MFI) vs. their unstained counterparts. (B) The 48-h co-culture killing assays of SU-DIPG-IV and VUMC-DIPG-G with LM-1 and TEG cells in the absence and presence of 30μM PAM, (C) INFγ release after 48 h of SU-DIPG-IV and VUMC-DIPG-G with LM-1 and TEG cells in the absence and presence of 30 μM PAM. Data from multiple independent experiments are shown for BTN2A1 and BTN3A expression (n = 2) and for the killing and IFN-γ experiments (n = 3). Biological replicates are represented as dot points and the error bar as ±SD. BTN2A1 and BTN3A data were normalized on the MFI of unstained SU-DIPG-IV and VUMC-DIPG-G, respectively. Statistics were assessed with the Student’s two-tailed t-test. Killing data are normalized on the number of alive tumor cells without any treatment (tumor only). Statistics were assessed with one-way ANOVA and corrected for multiple comparisons with the Tukey test.

Figure 2

Combining oncolytic viruses with TEGs enhanced specific killing of patient-derived diffuse midline glioma cell cultures in a cell-specific manner and BTN2A1 and BTN3A-independent manner in the absence of PAM. Killing percentage of DMG cell SU-DIPG-IV and VUMC-DIPG-G 2 days post-infection with 5, 10, 20, and 50 MOI of D24-RGD (A) and with 20, 100, and 300 MOI of R124 (B), data are normalized on the number of alive tumor cells without any treatment (tumor only). BTN2A1 (C) and BTN3A (D) expression as normalized MFI vs. untreated SU-DIPG-IV and VUMC-DIPG-G, respectively, conditions tested included SU-DIPGIV and VUMC-DIPG-G infected with D24-RGD or R124 and in the presence or absence of 30 μM PAM for 48 h (C,D). The 2-day co-culture killing assays comparing the killing of DMGs infected with D24-RGD (E,G) or infected with R124 (F,H) in combination with TEG cells in the presence of 30 μM PAM (E,F) as well as in the absence of PAM (G,H). IFN-γ release fold change of co-cultures of TEG cells with SU-DIPG-IV and VUMC-DIPG-G infected with 5 MOI of D24-RGD (I) or 20 MOI R124 (J) in the absence of PAM. BTN2A1 and BTN3A experiments are represented as mean ± SD from n = 2 independent experiments of the normalized MFI to the untreated control. Statistics are assessed with a one-way ANOVA and corrected for multiple comparisons with the Tukey test. Data for killing assays from multiple independent experiments (n = 2) for SU-DIPG-IV and VUMC-DIPG-G for (A,B) while (n = 4) for (E–H) represented as dot points (biological replicates) and error bar as ± SD. Statistics were assessed with the one-way ANOVA and corrected for multiple comparisons with the Tukey test. Data for IFN-γ from n = 4 independent experiments are represented as (pg/mL) in dot points (biological replicates) and the error bar as ± SD. Statistics are assessed with the Student two-tailed t-test.