Regulatory cells and the effect of cancer immunotherapy

Tumors can modify the microenvironment by releasing extracellular molecules, inducing tumor angiogenesis and promoting peripheral immune tolerance, while the immune cells in the microenvironment can affect the growth and evolution of cancerous cells. Increasing amounts of T CD3 + , cytotoxic CD8 + and memory CD45RO + T cells are associated with greater disease-free survival and overall survival (OS) in most studies [1‐3]. Histological analysis of tumors has highlighted the importance of immunological infiltrates, including macrophages, dendritic cells, polymorphonuclear cells, natural killer (NK) cells, B cells and T cells, and revealed a wide diversity of these among patients [4]. Although the extent of the immune infiltrate can be a good prognostic indicator in some cancers, the anti-tumor response is clearly insufficient to prevent disease progression. In inflamed tumors, negative immune regulatory factors tend to be dominant due to the chronic nature of the immune infiltrate. For instance, in a study aimed at identifying biomarkers associated with clinical outcome in melanoma patients, the decrease in FOXP3 + /regulatory T cells (Tregs) was associated with better clinical responses in the group treated with ipilimumab [5].

Therefore, patients with tumors containing infiltrates could be induced to respond to immunotherapy if immune cells within the microenvironment are reactivated. On the other hand, the presence of regulatory cells such as Tregs, MDSCs (myeloid-derived suppressor cells) and TAMs (tumor-associated macrophages) has generally been associated with a poor clinical prognosis. The actions on the vast majority of these components that have been reviewed here are not as advanced in head and neck cancer (HNC) as in other neoplasms such as melanoma.

Although therapy based on immune checkpoint blockade has achieved outstanding results in terms of OS [6], a high percentage of patients still show intrinsic resistance [7]. Some features are required to achieve a successful response to programmed cell death protein 1 (PD-1) blockade, such as a high frequency of tumor neoantigens, the amount of infiltrating effector T cells, a high level of expression of PD-L1, or an IFN-related gene signature [8]. Due to their potent suppressive activities against effector lymphocytes and their abundance in the tumor microenvironment, immunosuppressive cells act as a major barrier to cancer immunotherapy. A variety of therapeutic approaches directed towards immunosuppressive cells are actively being tested in preclinical and clinical studies [9‐14].

Since regulatory cells are characterized by their high level of expression of immune checkpoints, it is also expected that immune checkpoint inhibitors perform part of their function by blocking these molecules and enhancing the immune responses. This suggests that immunotherapy does not only act by activating specific effector T cells but also can directly or indirectly attenuate the suppressive activity of regulatory cells in tumor tissues. The aim of this review is to draw together the current knowledge about the effect of immunotherapy on the various types of regulatory cells, and how these effects may be beneficial in the response to immunotherapy (Fig. 1).

The activities of Tregs are the most widely studied of the mechanisms regulating the immune response. They were discovered several decades ago [15] and are known to have a strategic role in the maintenance of immune homeostasis [16]. Their function has been closely linked to the development of diverse pathologies, including autoimmunity [17] and cancer [18]. As Tregs are known to express many immune checkpoint inhibitor (ICI) targets, the effect of these immunotherapies could alter Treg numbers and function. For this reason, a comprehensive understanding of their action in cancer settings could help increase the efficacy and reduce the incidence of immune-related adverse effects after ICI treatment (Fig. 2).

In the clinical setting, the use of daclizumab (a humanized neutralizing monoclonal antibody against the α-chain of the interleukin-2 receptor) in patients with metastatic melanoma resulted in depletion of Tregs and effector T cells, so that augmented T cell responses in the tumor are not induced [19]. In contrast, in a study of breast cancer patients, administration of daclizumab followed by vaccination with multiple tumor-associated peptides reduced the frequency of Tregs and gave rise to a durable, stable disease with little progression [20‐22]. Denileukin diftitox (a ligand–toxin fusion protein that consists of full-length IL-2, and which has been approved for the treatment of CD25 + cutaneous T-cell leukemia and lymphoma) was used for lung, ovarian and breast cancer, demonstrating that a single dose reduced the prevalence and absolute numbers of peripheral Tregs and increased effector T-cell activation in all patients [23, 24]. Considering all the previous results, it was inferred that therapeutic success may depend on the phase and state of the anti-tumor immune responses [25].

Several reports have suggested that the effect of ipilimumab (anti CTLA-4, human antibody IgG1 isotype) on Tregs is mediated only through antibody-dependent cell-mediated cytotoxicity, by which their suppressive functions are maintained [26‐28]. On the other hand, tremelimumab (anti CTLA-4 antibody IgG2 isotype) has been shown to be able to replenish effector and memory CD4 + and CD8 + T-cell numbers by suppressing Treg activity without influencing the proportion of Tregs [29]. A recent study showed that ipilimumab and tremelimumab both increase infiltration of intratumoral CD4 + and CD8 + cells without depleting FOXP3 + cells in human tumors, suggesting that their efficacy could be enhanced by modifying the Fc portions of the monoclonal antibodies (mAbs) to enhance Fc-mediated depletion of intratumoral regulatory T cells [30]. Also, it has been described that a combined treatment with anti B7x may help to overcome the B7x-mediated resistance to anti-CTLA-4 [31].

However, the mechanism underlying the action of these antibodies is unknown. The discrepancies could be explained by the fact that the number of peripheral Tregs depends on the genetic background of the patients and the time since immunotherapy that the analysis is performed.

PD1 blockade can induce CD8 + T-cell proliferation and cytokine production, thereby negatively regulating Treg numbers by increasing the CD8 + Teff-to-Treg ratio [32]. The suppressive capacity of effector Tregs was enhanced in some advanced gastric cancer patients treated with PD1 blockade and was associated with rapid cancer progression [33]. This pattern has also been described by other authors [34]. Considering these findings together, the reports suggest that a combination of anti-CTLA4-depleting Tregs with PD1 blockade could be used to enhance CD8 activation and Treg depletion. It would also be interesting to study the effects of the timing and duration of such a Treg-targeting antibody treatment, to determine whether it is critical for the differential control of Treg and effector T cells, in other words, whether the longer the treatment, the more likely effector T cells are to be depleted, thereby hindering the generation of effective tumor immunity [18, 35]. The anti-CCR4 antibody, mogamulizumab, has demonstrated an effective reduction in the frequency of effector Tregs that selectively augment the induction of tumor antigen-specific CD4 + and CD8 + T cells in vivo [36, 37].

Moreover, anti CCR8 monoclonal antibodies were shown to cure tumors in mice by selectively depleting tumor Tregs and increasing CD8 + effector T cells [38]. Van Damme et al. also demonstrated that anti-CCR8 antibodies had antitumor effects and were seen to display synergistic antitumor effects combined with anti-PD-1 mAbs [39]. More recent studies describe the use of microRNAs -15a/16–1 to regulate immunosuppression in hepatocellular carcinoma by reducing CCL22 binding to C–C chemokine receptor type 4 on Tregs [40].

New therapeutic strategies are being developed to target the major co-inhibitory and co-stimulatory molecules of Tregs. Some TIGIT monoclonal antibodies (BMS-986207) are under clinical trial in combination with nivolumab for the treatment of advanced solid tumors (NCT02913313) and as monotherapy or in combination with atezolizumab (anti PD-1 antibody) (MTIG7192A) [41‐43]. Studies of patients with melanoma show that LAG-3 + Tregs selectively expand in PBMCs and TILs, bestowing potent suppressive activity on them in a cell-to-cell, contact-dependent manner [42].

Other approaches, such as blockade of TIM-3 [44], V-domain Ig suppressor of T cell activation (VISTA) [45, 46], treatment with glucocorticoid-induced TNFR-related (GITR) protein agonistic antibody [47, 48], anti-OX40 antibodies [49‐51], IDO-1 inhibitors [52, 53], TGF-β inhibitors [54]VEGF-targeting therapy receptor 2 (VEGFR2) [55, 56], PI3K inhibitor and HSP inhibitor (phosphoinositide 3-kinase (PI3K) pathway, or heat shock protein (HSP) [57], have also yielded promising results. Some drugs in combination with immune checkpoint inhibitors are currently the subject of ongoing clinical trials for cancer therapy [58].

Others approaches such as vaccines, nanodrugs [59], the generation of chimeric antigen receptor (CAR-)T cells, [60] or directly targeting FOXP3 in Tregs with an antisense oligonucleotide are also under investigation with the aim of reprogramming Tregs [61].

MDSCs are recognized as one of the major cell components in the tumor microenvironment, where they promote tumor growth by exerting their immunosuppressive functions. MDSCs have emerged as major regulators of immune responses in cancer and key targets for treating cancer [62] (Fig. 3).

Suk Lee et al. reported that anti-thymocyte globulin (ATG)-treated mice showed a reduction in MDSCs frequency and function, suggesting that ATG could be used to suppress MDSCs [63]. On the other hand, the role of mitogen-activated protein kinase (MAPK) activation is known to be crucial in the regulation of pathways involved in the expansion of MDSCs [64]. The beneficial effects mediated by AMPK activation could therefore be induced by the inhibition of MDSCs functions, and the authors proposed that AMPK activators could be promising drug candidates for cancer therapy [64, 65].

In a murine model of neuroblastoma, the adoptive recipient leukocyte infusion (RLI) triggered a potent cytotoxic T cell response against the tumor. Conversely, the RLI also provoked a systemic expansion of MDSCs that weakened the CD8 anti-tumoral effect. When these MDSCs were depleted with 5-FU the inhibition of tumor growth and OS were improved significantly, suggesting that MDSCs targeting is a novel approach for increasing RLI effectiveness [66]. In another study, Eriksson et al. reported reduced levels of Tregs and PMN-MDSCs in pancreatic cancer patients treated with gemcitabine [67]. Weiss et al. demonstrated, in two murine models, the importance of targeting MDSCs and Tregs in anti-tumor immunotherapy. They described how this MDSCs and Treg reduction was Fas-dependent and, consistent with the conclusions of other authors, noted the importance of targeting these regulatory cells in order to ensure that the therapies produced successful anti-tumoral activity [68]. Dang et al. investigated the expression of TLR8 on MDSCs and the effect of motolimod, a TLR8 agonist, on MDSCs survival and function. They found that motolimod treatment reduced MDSCs levels in healthy donors and cancer patients, and concluded that TLR8 agonists could be deployed in conjunction with cancer immunotherapeutic approaches in order to enhance the anti-tumor effects of the adaptive immune response [69]. Davis et al. demonstrated functional inhibition of MDSCs with IPI-145, an inhibitor of PI3Kδ and PI3Kγ isoforms that enhances responses to PD-L1 blockade. These results offer a proof of concept for the low-dose use of isoform-specific PI3Kδ/γ inhibitors to suppress MDSCs and thereby enhance responses to the immune checkpoint blockade [70].

Other approaches focus on promoting MDSCs differentiation. A few years ago, Gabrilovich´s group demonstrated how all-trans-retinoic acid (ATRA) induces MDSCs differentiation into macrophages and dendritic cells (DCs) [71]. In 2018, Tobin et al. conducted a randomized phase II clinical trial in which they treated advanced melanoma patients with ipilimumab monotherapy, or ipilimumab plus ATRA. They showed that treatment with ATRA reduced MDSCs function in mixed lymphocyte reactions. ATRA also reduced the expression of immunosuppressive genes by MDSCs. Finally, ATRA significantly decreased the frequency of circulating MDSCs relative to ipilimumab treatment alone in advanced-stage melanoma patients [9].

Results published by Xin Lu et al. shed light on the field of metastatic castration-resistant prostate cancer (mCRPC) in relation to the use of immune checkpoint blockade for the purpose of blocking MDSCs. They developed a chimeric mouse model of mCRPC with which they demonstrated that combined therapies were capable of increasing IL-1ra and suppressing MDSCs-promoting cytokines. The authors combined PDL1 blockade and the kinase inhibitors cabozantinib and BEZ235, which induce a reduction in MDSCs function. This combined therapy proved to be more effective than single therapy [72].

These findings highlight the necessity of including MDSCs neutralization in novel strategies of combined cancer treatment.

The combination of immunoregulatory treatments with immune checkpoint blockade might represent a novel beneficial approach. The regulatory effects of diarylheptanoid curcumin on STAT3 and JAK2 signaling involved the decrease of IL-6 production by MDSCs but with no adverse effects detected [73]. Conversely, when STAT3 inhibitors were applied in clinical settings for targeting tumor-associated myeloid cells, the studies were disrupted by the side effects [74]. Additional approaches aimed at blocking STAT3 exploit the administration of siRNA and decoy oligonucleotides. For instance, AZD9150 is a STAT3 oligonucleotide inhibitor under investigation in combination with immune-checkpoint inhibitors in phase I/II clinical trials [10].

Various approaches that aim to target MDSCs have explored the blockade of MDSCs mobilization from the bone marrow. In this regard, bisphosphonates, drugs administered in cancer patients with bone metastases, have an MMP-9 inhibitory effect that is related to reduced MDSCs expansion in peripheral blood and bone marrow [75]. This reduction overcame the immune suppression and potentiated the anti-tumor response induced by immunization against the p185/HER-2. In spite of these results, most clinical trials of MMP inhibitors in recent decades have failed because of side effects: MMP regulates multiple signaling targets and the inhibition of some MMPs could have pro-tumorigenic effects that impair the benefits of target inhibition. These effects might be responsible for the failure of MMP inhibitors in clinical trials [76]. More recently, more specific MMP inhibitors with improved toxicity have been developed [77]. To date, several studies have indicated that modulating MMP activity can improve immunotherapy. As a consequence, several MMP inhibitors are the subject of clinic trials [62, 77].

More strategies aimed at decreasing the recruitment of MDSCs such as targeting tumor glutamine metabolism [78] and targeting chemokines [79] have been described.

Huber et al. described microRNAs associated with MDSCs features and shorter progression-free survival (PFS) and OS in melanoma patients treated with immune checkpoint inhibitors. These miRs were responsible for the conversion of monocytes into MDSCs mediated by melanoma extracellular vesicles (EVs) [80]. The authors argued that the role of the identified MDSC-miRs may reflect functional features that MDSCs display upon immune activation triggered by ICIs, because resistance to ICIs can be reverted by myeloid cell depletion. This highlights the need for drugs that are capable of blocking myeloid cells dysfunction. Hence, MDSC-related miRs are potential biomarkers for assessing systemic immunosuppression and immunotherapy outcomes in cancer patients [80].

Limagne et al. reported the outcomes of a prospective immunomonitoring study of 25 metastatic colorectal cancer (mCRC) patients treated with a first-line combination regimen of 5-fluorouracil 5-FU, oxaliplatin and bevacizumab (FOLFOX–bevacizumab), compared with healthy volunteers. FOLFOX–bevacizumab treatment produced a drop in gMDSCs in most patients and was associated with better survival. Moreover, those gMDSCs expressing higher levels of PD-L1, CD39 and CD73 showed potent immunosuppressive activity that could be reversed by blocking the CD39/CD73 and PD-1/PD-L1 axes [81].

The inhibition of prostaglandin E2 receptor 4 (EP4) in MDSCs may offer a strategy for enhancing the efficacy of immunotherapy in colorectal cancer [82]. Li T et al. showed that pharmaceutical inhibition of c-Rel in mice markedly inhibited cancer growth and suggest c-Rel as a myeloid checkpoint that may be targeted for treating cancer [83]. Similarly, S100A9-derived peptides conjugated to antibody Fc [84] and targeting the TNF-related apoptosis-induced ligand (TRAIL) receptor could be a potent and selective method for MDSCs depletion [85]. Several drugs against CSF1R have shown promising antitumor efficacies by inhibiting the survival of M-MDSCs and tumor associated macrophages (TAMs) and are being tested in in cancer patients [86].

In summary, several candidate drugs have been proposed for reducing MDSCs frequency and function; other approaches aim to modulate MDSCs differentiation or mobilization. A combination of immunoregulatory drugs with ICBs might be more effective in targeting MDSCs as anti-tumor immunotherapy.

The extent of the immune infiltrate can be a good prognostic indicator in some cancers, but the anti-tumor immune response is, in most cases, insufficient to prevent disease progression. Several mechanisms and cell types are involved in negatively regulating the immune response, and the infiltration of large numbers of regulatory cells into tumor tissues is associated with poor prognosis.

In inflamed tumors, negative immune regulatory factors tend to dominate due to the chronic nature of the immune infiltrate. In cold tumors, it has been postulated that resistance to treatment may be due to low antigenicity, which causes less lymphocytic infiltration and is associated with an increase in the frequency of immunosuppressive cells and other factors.

Therapeutic success may depend on the phase and state of anti-tumor immune responses as the timing of the treatments seems to be critical to the differential control of these regulatory cells. When an immunosuppressive tumor microenvironment is already established, it will be more difficult to overcome and the immunotherapy will be less effective.

Although there are some current indications of the use of ICIs as first-line treatments, one of the main difficulties in clinical trials is that they were carried out in patients with advanced-stage tumors that were refractory to chemotherapy. Therefore, in tumors with low immunotherapeutic efficacy, it is plausible to consider that their failure to reach clinical end-points was due to the trial design.

Another challenge is that, in order to have validated drugs, it is necessary to study the possible anti-target effects of these molecules. The improvement in the design of clinical trials and the deep knowledge of the target and off-target effects of the treatments could lead to improved toxicity profiles and more selective drugs with less severe side-effects.

The discovery of new molecular targets could also help predict the tumors that are most immunogenic and most likely to respond to treatment, thereby making it possible to detect responders to immunotherapy.

In tumors characterized by a highly immunosuppressive environment, the development of new treatment options should certainly explore the efficacy of different drug combinations.

Numerous ongoing studies and clinical trials are exploring chemo-immunotherapeutic combinations that aim not only to eradicate the tumor mass but also to neutralize tumor-induced immunosuppression, thereby facilitating the effect of concurrent immunotherapy. Results obtained from these studies will shed light on the field and enable established approaches to be modified in order to better treat cancer patients in the next few years.