In the last decade, immunotherapies have revolutionised anticancer treatment. However, there is still a number of patients that do not respond or acquire resistance to these treatments. Despite several efforts to combine immunotherapy with other strategies like chemotherapy, or other immunotherapy, there is an ‘urgent’ need to better understand the immune landscape of the tumour microenvironment. New promising approaches, in addition to blocking co-inhibitory pathways, such those cytotoxic T-lymphocyte-associated protein 4 and programmed cell death protein 1 mediated, consist of activating co-stimulatory pathways to enhance antitumour immune responses. Among several new targets, glucocorticoid-induced TNFR-related gene (GITR) activation can promote effector T-cell function and inhibit regulatory T-cell (Treg) function. Preclinical data on GITR-agonist monoclonal antibodies (mAbs) demonstrated antitumour activity in vitro and in vivo enhancing CD8+ and CD4+ effector T-cell activity and depleting tumour-infiltrating Tregs. Phase I clinical trials reported a manageable safety profile of GITR mAbs. However, monotherapy seems not to be effective, whereas responses have been reported in combination therapy, in particular adding PD-1 blockade. Several clinical studies are ongoing and results are awaited to further develop GITR-stimulating treatments.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, any changes made are indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
In the last decade, immunotherapies, mainly through antiprogrammed cell death protein 1 (anti-PD-1)/programmed death-ligand 1 and anticytotoxic T-lymphocyte-associated protein 4 (anti-CTLA-4) monoclonal antibodies (mAbs), have revolutionised anticancer treatment. However, there is still a number of patients that do not respond or acquire resistance to these treatments. According to recent tumour classification by their immune infiltration, some types of cancer potentially respond to immune checkpoint inhibitors (highly immune-infiltrated or ‘hot tumour’), while in other tumours available immunotherapies appear not to be effective (non-immune-infiltrated or ‘cold tumour’). Despite several efforts to combine immunotherapy with other strategies like chemotherapy, radiotherapy or other immunotherapy aiming to convert ‘cold’ to ‘hot’ tumour, there is an ‘urgent’ need to better understand the immune landscape of the tumour microenvironment and to find alternative approaches to modulate immune function.1
New promising approaches, in addition to blocking co-inhibitory pathways, such those CTLA-4 and PD-1 mediated, consist of activating co-stimulatory pathways to enhance antitumour immune responses.2 One such strategy includes the development of agonist antibodies to target members of the tumour necrosis factor receptor superfamily (TNFRSF) with key role on immune activation and antitumour immune response, like 4-1BB, OX40, CD27 and glucocorticoid-induced TNFR-related gene (GITR).3 Several data demonstrate that GITR activation can promote effector T-cells function and inhibit regulatory T-cells (Treg) function.3 4
In this review, we focus on the GITR/GITR ligand (GITRL) axis.
GITR and GITRL expression
GITR (TNFRSF18/CD357/AITR) is a type 1 transmembrane protein belonging to the TNFRSF including OX40, CD27, CD40 and 4-1BB. Human GITR is constitutively expressed at high level on CD4+CD25+FoxP3+ Tregs and at low levels on naïve and memory T-cells.4–7 On activation of CD8+ and CD4+ effector T-cells, GITR expression increases rapidly on both Tregs and effector T-cells, reaching the highest level on activated Tregs.4 5
GITR is also expressed on natural killer (NK) cells and at low levels on B cells, macrophages and dendritic cells, and can be upregulated by activation, especially on NK.8 9
GITRL is a type 2 transmembrane protein and is also a member of the TNFRSF. It is commonly identified as a trimer, although it can also be present as a monomer or assemble into others multimeric forms.10
GITRL is predominantly expressed by activated antigen-presenting cells, including macrophages, B cells, dendritic cells and endothelial cells.4 8 Notably, GITR and GITRL expression is not restricted to haematopoietic cells. GITR expression has been described on epidermal keratinocytes and osteoclast precursors and GITRL expression on endothelial cells, especially after type I interferon (IFN) exposure.6
Recently, another GITR endogenous ligand has been described: SECTM1A, which is expressed both as a transmembrane protein and as a secreted protein. In mouse, SECTM1A is able to activate both GITR and CD7, but its role is not yet defined.11
GITR signalling and function
GITR, as other molecules of the TNFRSF, can act as a co-stimulatory receptor, thus representing a potential target to enhance immunotherapy, in particular immune checkpoint inhibitors.
All TNFR are characterised by their ability to bind TNF ligand and activate the transcription nuclear factor-κB (NF-κB) pathways via TNF receptor-associated factors (TRAFs), a family of six proteins that are recruited to further transduce signals within the cell. In particular, the activation of GITR signalling pathways, mediated by TRAF2/5-NF-κB, results in reduced T-cell apoptosis and promotes T-cell survival, at least in part by upregulating the expression of the Bcl-xL prosurvival molecule.12
In the periphery, after T-cell receptor (TCR) stimulation, the GITRL or agonist antibodies on conventional T-cells increases T-cell activation by inducing interleukin (IL)-2 and IFN-γ expression, enhancing CD25 expression and stimulating cell proliferation (figure 1).12–14 Furthermore, GITR co-stimulation enhances CD8+ T-cell cytotoxic function, and promotes survival of bone marrow CD8+ memory T-cell (figure 2).15
In vitro and in vivo, GITR signalling, especially mediated by agonist mAb, can inhibit Treg ability to suppress effector T-cells, either by rendering effector T-cells less susceptible to Treg immunosuppressive activities or by directly inhibiting Tregs.16 17 This last mechanism could be due to the transient loss of FoxP3 on Tregs, although it has been observed only in Tregs from tumour-bearing mice and not in Tregs from naïve mice.18
Interestingly, the GITR/GITRL axis effect on Treg seems to be inhibitory in the short-term, while the long-term over stimulation in vivo favours the expansion and the activity of Treg in mice.16
In addition, GITR co-triggering of conventional T-cells stimulates IL-10 production, favouring differentiation of conventional CD4+ T-cells into T-helper 2 and Treg cells, these findings sustain the role of GITR in the balancing between T-helper and Treg cells.19
In summary, while commonly Treg cells antagonise effector T-cells, thereby limiting antitumour activity, GITR activation on effector T-cells increase effector function by limiting the sensitivity of these cells to Treg suppression.
Modulation of GITR in preclinical tumour models
Antitumour activity of GITR mAb
In recent years, GITR has been largely studied as a pharmacological target.
Co-activation of GITR by agonist mAbs can increase immune response, inflammation and thereby antitumour response.9 Differently, GITR inhibition, through antagonist mAbs could inhibit T-cell activation and immune response.6 Consequently, GITR agonist mAbs has been further developed as antitumour agents.
In tumour models, the antitumour activity of GITR mAbs is mainly based on the ability to enhance CD8+ and CD4+ effector T-cell activity and on the inhibition/depletion of tumour-infiltrating Tregs.21–24
Importantly, GITR is not expressed on the tumour itself, but it is expressed on tumour-infiltrating lymphocytes (TILs) of several human cancer types including lung cancers, renal cell carcinoma, head and neck carcinoma and melanoma.25
The most widely used molecules to trigger GITR are agonist antibodies like DTA-1 (a rat IgG2b)5 or recombinant version of GITRL, like GITR-Fc.
The DTA-1 mAb has demonstrated in vivo antitumour activity in multiple syngeneic mouse tumour models (eg, melanoma,24 cervical26) enhancing CD8+ and CD4+ T-cell proliferation and cytokine induction. A recent study reported that GITR agonists can also increase cellular metabolism to support CD8+ T-cell effector function and proliferation.27
The intermediate role of CD8+ and CD4+ T-cells in tumour rejections seems to be crucial.
Regressing tumour-bearing mice, treated with DTA-1, were found infiltrated by a large number of CD4+ and CD8+ T-cells, including those secreting IFN-γ. However, the treatment resulted in tumour regression only in IFN-γ-intact mice but not IFN-γ-deficient mice.28 29 The effect of DTA-1 was lost/decreased in the absence of CD8+ T and NK cells.4
Moreover, GITR engagement by DTA-1 promoted the differentiation of IL-9-producing CD4+ T-helper cells, thus enhancing immune-mediated tumour response.30
The additional crucial concomitant mechanism to inhibit tumour growth, following DTA-1—GITR triggering is the reduction of Treg activity and number. Such a reduction can occur via Treg-specific and tumour-specific antibody-dependent cell cytotoxicity (ADCC): GITR+ Tregs specific for tumour antigens, through the Fc domain of anti-GITR mAbs, are recognised and killed by myeloid and NK cells present in the tumour.22 23
GITR has a higher expression in tumour infiltrating Treg compared with peritumoral region in several tumour like renal, colorectal and hepatocarcinoma.31–33
FoxP3+ Treg reduced accumulation in tumours has been also hypothesised as a result of reduced trafficking or loss of FoxP3 expression in intratumour Treg and their ‘conversion’ into activated T-cells.24
However, Mahne et al reported that mDTA-1 depletes rather than converts intratumour Tregs. In tumour-bearing mice, Treg depletion together with GITR triggering were necessary to revert intratumour CD8+ T-cell exhaustion, thus improving antitumour efﬁcacy.34
Vence et al confirmed that tumours with high expression of CD8+ and CD4+, after GITR mAb treatment, have the better response, mainly lung cancer, renal cancer and melanoma.25
Moreover, preliminary results showed a better suppression of tumour growth with intratumour compared with intravenous injection. In fact, the intratumour injection was able to induce a systemic antitumour immune reaction, exerting its effect on injected and on un-injected tumours.35
Combination of GITR mAb with immune-modulating therapies
GITR, like other co-stimulating molecules, has a key role on T-cell activation and its activity can potentiate, in a synergic effect, other anticancer therapies.
Combined treatment with anti PD-1 and GITR-agonist mAbs was able to achieve long-term survival in mouse model of ovarian and breast cancer, stimulating IFN-γ producing conventional T-cells and inhibiting immunosuppressive Tregs and myeloid-derived suppressor cells.4 36 The treatment combination manages to rescue CD8+ T-cell dysfunction and to induce proliferation of precursor effector memory T-cell phenotype in a CD226-dependent manner.37 Durable responses were also reported adding cytotoxic chemotherapy or radiotherapy to anti-PD-1/GITR mAbs.36 38 39
Co-administration of GITR mAbs and anti-CTLA-4 resulted in an 80% tumour-response in CT26 (colon carcinoma) and CMS5a (fibrosarcoma) mice tumour models reducing intratumour Treg (via GITR) and stimulating CD8+ T-cells (via CTLA-4).37
Targeting GITR together with an OX40 agonist (OX40 ligand fusion protein), showed unexpectedly a synergistic antitumour effect on CT26 tumour-bearing mice, although the toxic profile of the combination could represent a limit to clinical development.40
The synergistic and complimentary antitumour effect obtained combining GITR mAbs and vaccines was reported13 in cervical cancer41 and in melanoma.42 Moreover, adding chemotherapy (gemcitabine) to the combination of vaccine and GITR mAb was able to decrease tumour-suppressive environment and to induce a long-lasting memory immune response.43
In conclusion, in preclinical tumour models co-activating GITR through agonist mAb was able to induce antitumour responses. In particular DTA-1 mAb demonstrated in vivo antitumour activity in multiple mouse tumour models, enhancing CD8+ and CD4+ T-cell proliferation/cytokine induction, and reducing Treg activity and number, especially via ADCC. Moreover, GITR agonist mAbs best antitumour responses were achieved in combination with other immune-modulating therapies.
Clinical trials with GITR monoclonal antibodies
MEDI1873, a GITR-ligand/IgG1 agonist fusion protein, was tested in a phase I trial reporting G3 treatment-related adverse events (TRAEs) in the 22.5% of patients and no G4-5 TRAEs (table 1). Pharmacodynamics analysis confirmed that MEDI1873 increased CD4+Ki67+ T-cells and induced a >25% decrease in GITR+/FoxP3+ T-cells in the evaluable patients. Stable disease (42.5%), durable in the 17.5% of patients, was the best response in this heavily pretreated population, supporting further clinical trials.44
The phase I trial with AMG 228, an agonistic human IgG1 GITR-mAb, reported a favourable safety profile, but no evidence of T-cell activation or antitumour activity, at least as monotherapy.45
BMS-986156, a fully human IgG GITR-mAb, has been tested as monotherapy and in combination therapy with nivolumab in a phase I/IIa trial. None of the 34 patients in the monotherapy arm experienced a dose-limiting toxicity (DLT) or grade G3-5 TRAEs, a patient out of 258 had a DLT in combination with nivolumab 240 mg. No responses were seen with monotherapy, although an objective response rate (ORR) of 9% (18 out of 200 evaluable patients) across all tumour types was achieved in the combination arm.46
No responses were reported in the phase I trial with TRX518, a fully humanised Fc-dysfunctional aglycosylated IgG1κ GITR-mAb, in monotherapy. Pharmacodynamics data and subsequent in vitro and in vivo investigation highlighted the possible mechanisms of tumour resistance to anti-GITR monotherapy and its possible overcome combining anti PD-1/PD-L1 therapy. In a murine model, DTA-1 early treatment delayed tumour growth, preventing intratumour Treg accumulation and CD8+-not exhausted T-cell upregulation. Differently, in advanced tumours microenvironment, high Treg expression increases dysfunctional CD8+ T-cells that shows an exhausted profile and fail to upregulate markers of activation and cytotoxicity. Thus, adding PD-1 blockade was able to counteract CD8+ T-cells exhaustion, resulting in better tumour control.47 Preliminary evaluations of tumour response among the first patients enrolled in the phase I combinational trial were encouraging (NCT02628574).
MK-4166, a humanised IgG1 agonist GITR mAb, in combination with pembrolizumab, an anti PD-1 mAb, demonstrated a good safety profile and potential activity, in particular among patients with melanoma naïve to treatments.48
Others compounds under investigation (table 2) are ASP1951 (PTZ-522),49 a tetravalent monospecific (TM) anti-GITR agonist antibody (NCT03799003); INCAGN01876, a humanised IgG1 mAb (NCT03126110) and GWN323 (NCT02740270).
Conclusions and future perspectives
GITR can act as a co-stimulatory receptor, representing a potential target to enhance immunotherapy efficacy. Preclinical data confirmed GITR triggering could increase CD8+ and CD4+ effector T-cell activity and reduce tumour-infiltrating Tregs. GITR mAbs have a manageable safety profile. However, they seem not to be effective as monotherapy, whether responses have been reported in phase I/II trials combination therapy with immune checkpoint inhibitors. In particular, adding PD-1 blockade may have a synergistic and complimentary antitumour effect, by converting CD8+ T-cells exhaustion.
Several clinical studies are ongoing, especially in combination with other treatments and results are awaited to further develop GITR-stimulating treatment.
Contributors All authors contributed on writing this review and provided critical feedback and final approval of the version to publish.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests LDM reports personal fees from Roche, personal fees from Pfizer, personal fees from Ipsen, personal fees from Eli Lilly, personal fees from Novartis, personal fees from Takeda, personal fees from MSD, personal fees from Genomic Health, personal fees from Celgene, personal fees from Seattle Genetics, outside the submitted work.
Patient consent for publication Not required.
Provenance and peer review Commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.