Abstract 3219: Assessing strategies to activate DNAM1 signaling and inhibit the TIGIT/PVRIG axis using mathematical modeling

Abstract PVR (CD155) and CD112 are ligands from the Nectin/Nectin-like family, frequently expressed on various tumor cells. PVR binds to the inhibitory receptors TIGIT and CD96, triggering a direct inhibitory signal in T and NK cells, and binds to the co-stimulatory receptor DNAM1 (CD226), leading to its downregulation. CD112 binds to the inhibitory receptor PVRIG and to DNAM1, stimulating T and NK cells. Late-stage clinical trials of anti-TIGIT monoclonal antibodies (mAbs) have shown mixed results regarding efficacy (Cho et al. Lancet Oncol. 2022 Jun;23(6):781-792; Rudin et al. J Clin Oncol. 2024 Jan 20;42(3):324-335). Numerous anti-PVR, anti-CD96, and anti-PVRIG mAbs, along with their combinations with anti-TIGIT mAbs and other therapies, are currently in early clinical development. This study aimed to compare how the complete inhibition of TIGIT, PVRIG, PVR, or their combinations affects the engagement of TIGIT, CD96, PVRIG, and DNAM1. We developed a mechanistic model describing the surface expression of PVR and CD112 on cancer cells, and TIGIT, CD96, DNAM1, and PVRIG on CD8 T-cells. This model captures the interactions of these ligands and receptors within the immunological synapse (IS) between CD8 T-cells and cancer cells in the tumor. In vitro data and expression data of these ligands and receptors were used to identify parameters. The model was validated against data on the downmodulation of DNAM1 and TIGIT on CD8 T-cells in the tumor compared to the blood. We simulated total DNAM1 on the surface of CD8 T-cells in the tumor and the engagement of TIGIT, CD96, PVRIG, and DNAM1 in the IS (i.e., the number of PVR:TIGIT, PVR:CD96, CD112:PVRIG, PVR:DNAM1, and CD112:DNAM1 complexes) under 100% inhibition of TIGIT, PVR, PVRIG, TIGIT and PVRIG, or PVR and PVRIG. TIGIT blockade did not increase DNAM1 engagement. Moreover, 100% inhibition of TIGIT led to the release of additional free PVR, resulting in the downmodulation of total DNAM1 and an increase in CD96 engagement. Inhibition of PVRIG resulted in an approximately 1.3-fold increase in bound DNAM1, but not in total DNAM1. The combination of TIGIT and PVRIG blockade did not increase engaged DNAM1 beyond the levels observed with anti-PVRIG monotherapy. Inhibition of PVR led to a significant increase in total DNAM1 and a moderate increase in bound DNAM1 (approximately 1.3-fold), along with the disruption of TIGIT and CD96 engagements. The combination of PVR and PVRIG blockade resulted in the strongest increase in engaged DNAM1 (approximately 2-fold), in addition to disrupting TIGIT, CD96, and PVRIG engagements. The inability to increase DNAM1 engagement during TIGIT inhibition may be one of the reasons for the lack of efficacy observed in the phase 3 trial of tiragolumab (anti-TIGIT mAb) used in combination with atezolizumab. Targeting PVR alone or in combination with PVRIG appears to be a better strategy than targeting TIGIT or TIGIT in combination with PVRIG. Citation Format: Oleg Demin Jr, Galina Kolesova, Dmitry Shchelokov. Assessing strategies to activate DNAM1 signaling and inhibit the TIGIT/PVRIG axis using mathematical modeling. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3219.