3.1. TIGIT as a Potential Prognostic Marker for Cancer

Accumulating data from the immune monitoring of cancer patients have revealed that TIGIT expression is elevated in T and NK cells, and it often appears to be associated with advanced disease status and poor clinical outcomes [34,35,48,49,50,51,52,53,54,55,56,57,58]. In follicular lymphoma (FL) patients, TIGIT is highly expressed on intratumoral T_reg and late-stage memory CD8⁺T cells, and increased numbers of TIGIT-expressing tumor infiltrating T cells reveal a correlation with poor survival rate [48]. Multidimensional flow cytometric analysis of intratumoral T cells obtained from FL patients before and after anti-PD-1 therapy has revealed that TIGIT⁺ T_ex cells majorly respond to this therapy. It has been observed that TIGIT⁺ exhausted T cell populations are downregulated and TIGIT⁺ effector cells are upregulated by anti-PD-1 therapy [48]. Increase in the proportion of highly suppressive tumor-infiltrating T_reg cells following TIGIT expression is associated with poor clinical outcomes in patients with hepatocellular carcinoma (HCC) and metastatic melanoma [47,57]. Moreover, upregulation of TIGIT indicates unfavorable disease status. High-risk patients with myelodysplastic syndrome (MDS) express higher levels of TIGIT and PD-1 in peripheral blood T and NK cells than low-risk patients [58]. High TIGIT expression renders CD4⁺T, CD8⁺T, and NK cells hypo-responsive to stimulation in high-risk MDS patients. Several studies have reported that TIGIT upregulation after treatment is correlated with recurrence. In patients with high-grade serous carcinoma, NanoString analysis of tumor tissues has indicated that recurrent tumors acquire a more inflamed phenotype with increased expression of TIGIT, CTLA4, Lag-3, and Tim-3 compared to primary tumors [59]. The proportion of TIGIT⁺CD8⁺T cells is increased in peripheral blood collected from acute myeloid leukemia (AML) patients, and it becomes more evident in patients with primary refractory disease and leukemia relapse post-allogeneic stem-cell transplantation [49]. Furthermore, TIGIT and/or PD-1 expression in CD8⁺T cells is increased in patients with gastric cancer relapse after treatment with SOX (S-1 and oxaliplatin) regimen, whereas no notable increase in the proportion of TIGIT⁺ and/or PD-1⁺CD8⁺T cells was found in relapse-free patients [60]. The compensatory increase in TIGIT expression post-treatment has also been observed in high-grade neuroendocrine neoplasms upon anti-PD-1 therapy [61].

3.2. TIGIT Blockade in Anti-Tumor Immunity

Based on the mechanism underlying TIGIT-mediated regulation of anti-tumor immune responses, efforts have been made to enhance T or NK cell activity by blocking TIGIT binding to its ligands, PVR and nectin-2, with monoclonal antibodies (mAbs) for therapeutic interventions. Several preclinical mouse models have been used to assess the anti-tumor efficacy of anti-TIGIT blocking mAbs. In CT26 colon carcinoma, EMT6 breast carcinoma, MC38 colon carcinoma, and GL261 glioblastoma models, treatment with anti-TIGIT-blocking mAbs combined with anti-PD-1 or PD-L1-blocking mAbs leads to nearly complete remission of tumor growth, whereas the treatment of anti-TIGIT mAbs as a single agent presents limited efficacy [42,62,63]. CD8⁺T cell depletion using anti-CD8α-depleting mAbs in CT26-bearing mice has revealed that the synergistic effect of dual blockade of TIGIT and PD-1 is mainly driven by the promotion of CD8⁺T cell responses. A triple combination of anti-TIGIT mAbs, anti-PD-L1 mAbs, and radiotherapy elicits almost complete remission of tumor growth in CT26-bearing mice [64].

Sufficient tumor regression by treatment with anti-TIGIT mAbs alone has been reported in different mouse tumor models. In multiple myeloma (MM) mouse tumor model, TIGIT blockade leads to reduced tumor growth and increased survival compared with mice receiving control IgG or anti-PD-1 mAbs [35]. Moreover, TIGIT blockade presents anti-tumor efficacy in Tgfbr1/Pten2 conditional knock-out (KO) mouse model that spontaneously develops head and neck squamous cell carcinoma (HNSCC) upon tamoxifen injection [55,65]. Both studies suggest that TIGIT is highly expressed on CD8⁺T and T_reg cells in MM or HNSCC TILs and that anti-TIGIT mAbs reverse TIGIT-mediated suppression of CD8⁺T cell effector functions; however, the potent anti-tumor effect of anti-TIGIT mAbs as a single agent may not be fully guaranteed simply by the increased expression of TIGIT in TILs, since high TIGIT expression is also observed in CD8⁺ TILs in CT26-bearing mice that are not responsive to TIGIT blockade [42]. A recent study by Chiu et al. provided additional insights into the mechanism through which TIGIT blockade mitigates tumor immune evasion and resistance to PD-1 blockade [66]. They found that anti-PD-1 mAb treatment induced the upregulation of TIGIT in CD8⁺ TILs in Trp53 KO/C-Myc^OE mice, which is a highly aggressive HCC model; however, the compensatory expression of TIGIT upon PD-1 blockade was not observed in Hepa1-6-bearing mice that are known to be an anti-PD-1-sensitive orthotopic HCC model. PVRL1, which does not directly bind to TIGIT, contributed to TIGIT-mediated suppression of CD8⁺T cells by stabilizing PVR in HCC cells, and PVRL1 deficiency rendered HCC to be more sensitive to anti-PD-1 mAb treatment. In accordance with this finding, a differential level of the ligand expression, such as PVR and PD-L1, or an increase in the binding affinity of TIGIT to PVR under an acidic tumor microenvironment has been recently identified to contribute toward the sensitivity of tumor cells to TIGIT blockade [67,68].

Although TIGIT blockade is known to mainly act on CD8⁺T and T_reg cells, NK cell dependent efficacy of anti-TIGIT mAbs is also suggested. A recent study by Zhang et al. reported that treatment with anti-TIGIT mAbs 3 days after tumor cell implantation prevented tumor-infiltrating NK cell exhaustion in CT26 or methylcholanthrene (MCA)-induced fibrocarcinoma-bearing mice, which resulted in the enhancement of CD8⁺T cell responses and tumor rejection [34]. However, the mechanism through which TIGIT blockade has an impact primarily on NK cells compared to T cells and the mechanism through which NK cells promote CD8⁺T cell responses need to be further elucidated, since these results are contradictory to those of previous studies, revealing the CD8⁺T or the T_reg cell-mediated effect of TIGIT blockade using temporary depletion of these populations with specific antibodies [35,42,55]. A more recent study reported that anti-TIGIT mAbs enhanced IL-15-driven NK cell cytotoxicity in both B16F10 and LWT1 metastatic melanoma-bearing mice [69].

The potency of human anti-TIGIT blocking mAbs on CD8⁺T cells has been demonstrated in cancer patients. Cancer testis antigen NY-ESO-1-specific CD8⁺T cell responses are increased by the addition of blocking mAbs against TIGIT and/or PD-1 when peripheral blood mononuclear cells (PBMCs) from melanoma patients are stimulated with NY-ESO-1^157–165 peptide. Furthermore, TIGIT blockade increases the capacity for proliferation and degranulation of CD8⁺TILs from advanced melanoma patients upon TCR stimulation using autologous non-CD3 cells and anti-CD3 mAbs [50]. Upon TCR stimulation with anti-CD2/anti-CD3/anti-CD28 microbeads, bone marrow (BM) CD8⁺T cells in MM patients show increased CD107a expression and cytokine production in response to TIGIT blockade [35]. When anti-TIGIT mAbs are added to ex vivo co-culture of CD3⁺TILs and Mel-624 cells expressing membrane-bound anti-CD3 scFv (Mel-624 OKT3), IFN-γ and IL-2 production by CD3⁺TILs from patients with endometrial, ovarian, kidney, head and neck, and lung cancers is promoted [14]. A recent study reported that antigen specific responses to CEF (CMV, EBV, flu) peptide are augmented by TIGIT blockade in peripheral blood CD8⁺T cells derived from pancreatic ductal adenocarcinoma (PDAC) patients after mFOLFIRINOX therapy [43].

3.3. Mode of Action of Anti-TIGIT Therapy

Competitive binding of TIGIT and CD226 to PVR has been known as a key mechanism of TIGIT-driven immune suppression, and anti-TIGIT blocking mAbs are presumed to reverse the suppression by inhibiting TIGIT binding to PVR. This may occur as a mode of action; however, several questions need to be addressed for its clinical success and further translation of other members of TIGIT family receptors into cancer immunotherapy.

Despite the importance of understanding the molecular interplay between TIGIT, CD226, and PVR, the mechanism through which extracellular signals from the receptor-ligand binding/receptor dynamics are integrated into the intracellular regulation, particularly in the context of anti-TIGIT therapy, remains unclear.

A recent study by Jin et al. reported that the effect of TIGIT blockade depends on tyrosine phosphorylation at Y322 of CD226, which was the first study to define the molecular requirements for anti-TIGIT blocking mAbs [43]. They showed that TIGIT-mediated intracellular inhibition of CD226 phosphorylation at Y322 was restored by TIGIT blockade. Moreover, CD226 mutant at Y322 (CD226^Y322A) expressing CD8⁺T cells did not respond to TIGIT blockade, whereas CD226^WT or CD226^Y329A expressing CD8⁺T cells produced increased IFN-γ upon treatment with anti-TIGIT mAbs, suggesting that TIGIT blockade promotes T cell activation via CD226 phosphorylation at Y322 (Figure 2). CD226 dependent effect of anti-TIGIT mAbs was further shown in effector memory CD8⁺T cells expressing a low level of CD226 (CD226^loCD8⁺T_em) not responsive to both antigen stimulation and anti-TIGIT mAb treatment. CD226 activation using anti-CD226 agonist mAbs renders CD226^loCD8⁺T_em responsive to TIGIT blockade.

Figure 2. Role of CD226 in anti-TIGIT immunotherapy. TIGIT has a direct effect on intracellular regulation of CD226 activation in response to PVR binding. (Left) When TIGIT expression is absent or low, engagement of CD226 with PVR induces the phosphorylation of tyrosine 322 (Y322) on CD226, which leads to the activation of intracellular signaling cascades. (Middle) PVR preferentially binds to upregulated TIGIT over CD226. Upon interaction with PVR, the cytoplasmic tail of TIGIT is phosphorylated. This PVR-induced TIGIT phosphorylation inhibits T cell responses by promoting CD226 dephosphorylation. (Right) TIGIT blockade suppresses PVR-induced TIGIT phosphorylation and restores the impaired Y322 phosphorylation of CD226, thereby leading to T cell activation.

Recently, several studies have highlighted the importance of selecting appropriate fragment crystallizable (Fc) region for therapeutic antibodies. To date, the approved human therapeutic IgG antibodies belong to IgG1, IgG2, or IgG4 subclasses [70]. It is increasingly clear that binding of the Fc region of antibody to Fc gamma receptors (FcγRs) can elicit various immunomodulatory functions, including antibody dependent-cellular cytotoxicity (ADCC), complement-dependent cytotoxicity (CDC), and antibody-dependent cellular phagocytosis (ADCP) [71]. In addition, FcγR binding was reported to enhance agonistic activity of mAbs targeting tumor necrosis factor receptor superfamily members, such as CD28, CD137, CD40, and OX40 (CD134) [72].

The importance of the Fc domain of anti-TIGIT mAb is emphasized by the findings that anti-TIGIT mAb with Fc devoid of effector functions, which was intended to solely block TIGIT binding to its ligands, fails to exert any of anti-tumor efficacies in preclinical models [36,73,74]. It may be due to the loss of its depleting activity against TIGIT-expressing intratumoral T_reg cells, which has been considered as a potential mechanism of anti-TIGIT mAb-mediated anti-tumor effect [74]; however, it is still not clear whether the anti-tumor efficacy of anti-TIGIT mAbs depends on T_reg depletion, since there are recent reports that anti-TIGIT mAbs on mIgG2a isotype induce anti-tumor responses without evidence of T_reg depletion in mouse tumor models [36,73]. It may possible that FcγR on APC could act as a scaffold to crosslink anti-TIGIT mAb bound to TIGIT on immune cells, which may enhance the effect of TIGIT antagonism independent of T_reg cells. In addition, Han et al. recently have shown that the antibody-FcγR engagement induced activation of myeloid cells, leading to pro-inflammatory chemokine and cytokine secretion (Figure 3) [36]. Comparison of clinical activities of anti-human TIGIT mAbs with different Fc scaffolds could provide insight into whether FcγR binding is required for optimal anti-tumor responses of TIGIT blockades.

Figure 3. Proposed mechanisms of anti-TIGIT monoclonal antibodies (mAbs) in cancer immunotherapy. (1) Blockade of TIGIT could reverse the exhaustion of T and NK cell-mediated anti-tumor immunity. (2) Intratumoral regulatory T cells (T_{reg_}) cells expressing high levels of TIGIT could be preferentially depleted by anti-TIGIT mAbs, presumably through antibody-dependent cellular phagocytosis (ADCP) by macrophages and/or antibody-dependent cellular cytotoxicity (ADCC) by NK cells. (3) The TIGIT mAb- fragment crystallizable (Fc) gamma receptors (FcγR) engagement could activate myeloid cells, leading to enhanced antigen presentation function and proinflammatory chemokine and cytokine secretion.