Targeting Myeloid-Derived Suppressor Cells in Ovarian Cancer

Myeloid-derived suppressor cells (MDSCs) are a heterogeneous population of immature myeloid cells that exhibit immunosuppressive activity. They also directly stimulate tumor cell proliferation, metastasis, and angiogenesis. In ovarian cancer, there are increased numbers of circulating or tumor-infiltrating MDSCs, and increased frequencies of MDSCs are associated with a poor prognosis or an advanced clinical stage. Moreover, in murine models of ovarian cancer, MDSC depletion has shown significant growth-inhibitory effects and enhanced the therapeutic efficacy of existing anticancer therapies. In this review, we summarize the current knowledge on MDSC biology, clinical significance of MDSC, and potential MDSC-targeting strategies in ovarian cancer.


Introduction
Epithelial ovarian cancer is a leading cause of cancer-related death among women, accounting for 295,000 new cases and 185,000 deaths annually worldwide [1]. Due to its asymptomatic nature and lack of effective screening tests, most patients are diagnosed at advanced stages [2]. Although most advanced-stage ovarian cancers respond to the initial treatment, including primary debulking surgery and platinum-based chemotherapy, more than 70% of patients will ultimately relapse [3]. Therefore, it is imperative to overcome platinum resistance by identifying new therapeutic targets.
An increasing body of evidence suggests that the immunological microenvironment plays a significant role in the development/progression of cancer, and treatment options that activate the immune surveillance response have emerged as a promising cancer treatment strategy [4]. However, although immunotherapy has become a viable treatment for recurrent ovarian cancer, current immunotherapies face barriers that limit their clinical efficacy: only a limited number of patients have responded to checkpoint inhibitors, and antigen-specific active immunotherapy has demonstrated no survival benefit [5,6].
Various types of cancer can undergo intratumoral immunosuppression, including ovarian cancer [7,8]. This facilitates cancer cells to escape from destruction by the immune system, and could limit the therapeutic efficacy of current immunotherapies that target cytotoxic T lymphocyte-associated protein 4 (CTLA-4) or programmed death ligand 1 (PD-L1)/programmed death 1 (PD-1) [4].
In the early 20th century, it was noted that cancer progression was often accompanied by extramedullary hematopoiesis and resulting leukocytosis or neutrophilia [9]. These leukocytes or neutrophilia were further characterized by their suppressive activities and were called immature myeloid cells, myeloid suppressor cells, or natural suppressor cells. Eventually, they were named myeloid-derived suppressor cells (MDSCs) in 2007 [10]. Since then, an increasing amount of information regarding the biology and clinical significance of MDSCs in various pathological conditions has been reported. MDSCs are a heterogeneous population of immature myeloid cells (IMC) which can promote tumor growth by inducing tumor cell immunological anergy and tolerance; they block the proliferation and activity of both T cells and natural killer (NK) cells [11]. In addition, MDSCs can directly stimulate tumor cell proliferation, metastasis, and angiogenesis, all of which can lead to tumor progression and limit the potency of current therapeutic interventions [4,11]. As the acquired ability of cancer cells to escape from immune surveillance is a hallmark of ovarian cancer and increased MDSC levels have been demonstrated in ovarian cancer patients, MDSCs are now regarded as a promising therapeutic target and a predictive biomarker of treatment outcomes.
In this review, we summarize the current knowledge on MDSC biology and its role in ovarian cancer. We also discuss the utility of the number of MDSCs as a predictive marker and highlight how MDSCs can be targeted therapeutically in patients with ovarian cancer.

Definition of MDSC
MDSCs are a heterogenic population of IMC that differ in morphology and function from terminally differentiated mature myeloid cells (macrophages, dendritic cells (DC), or neutrophils). MDSC can be subdivided into two major subsets, monocytic MDSC (M-MDSC), which morphologically and phenotypically resemble monocytes, and polymorphonuclear (PMN) MDSC (also known as granulocytic MDSC), which are morphologically similar to neutrophils [10,11].

Immunosuppressive Functions of MDSCs
MDSCs (both PMN-MDSC and M-MDSC) can suppress both innate and adaptive immune responses. As shown in Figure 1, MDSCs mainly exert their suppressive effects by producing arginase-1 (Arg-1) which causes the removal of L-arginine, an essential amino acid for T cell differentiation, from the tumor microenvironment (TME). The depletion of L-arginine subsequently causes the downregulation of CD247 (the ζ-chain of the T cell receptor) expression in T cells. As CD247 is a subunit of the natural killer receptors NKp46, NKp30, and TcγIII in NK cells, the depletion of L-arginine leads to the inhibition of T cell and NK cell proliferation. MDSCs also produce reactive oxygen species (ROS) and nitric oxide (NO), which lead to the nitration of signaling molecules downstream of the FcγRIIIA, resulting in the inhibition of the activities of T cells and NK cells [10,11]. NO produced by MDSCs also nitrate signal transducers and activators of transcription (STAT)1, leading to the diminished interferon response in T cells and NK cells. Moreover, MDSCs induce regulatory T cell (Treg) expansion, which also acts to suppress the effector T cells [11]. Lastly, MDSCs have an increased expression of PD-L1, which leads to the downregulation of T cell function via engagement of cell surface PD-1 [13]. 2) Nonimmune activities of MDSC; MDSC promote cancer progression by inducing angiogenesis and tumor invasion via the production of VEGF, bFGF, Bv8, and MMP-9. Moreover, MDSC stimulate the metastatic activities of cancer cell by facilitating EMT and creating "premetastatic niches." MDSC also induce "stemness" in certain cancers, which might be associated with resistance to chemotherapy or radiotherapy.

Nonimmune Activities of MDSC
In addition to suppressing immune responses within the TME, MDSCs also promote cancer progression by stimulating tumor angiogenesis and enhancing tumor cell invasion and metastasis ( Figure 1). These processes are regulated by MDSC-derived mediators, including vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), Bv8, and matrix metalloproteinase (MMP)-9, all of which are essential for tumor angiogenesis and cancer cell invasion [11,14]. Moreover, recent reports suggested that MDSCs stimulate the metastatic activity of cancer cells by facilitating epithelial-to-mesenchymal transition (EMT) or by creating "premetastatic niches" [15,16]. Importantly, the evidence has indicated that MDSCs induce "stemness," which might be associated with resistance to existing anticancer treatments including chemotherapy or radiotherapy [17].

MDSC Generation and Recruitment
In healthy individuals, IMC develop from common myeloid progenitor cells that differentiate into mature nonsuppressive myeloid cells such as DCs, macrophages, or granulocytes. Under pathological conditions such as infection, inflammation, or cancer, their differentiation was directed away from mature nonsuppressive cells to suppressive cells (MDSCs) [10,11]. It has been reported that PMN-MDSC undergo expansion in most patients with solid tumors. However, in certain cancers, such as melanoma, multiple myeloma, and prostate cancer, the M-MDSC population is more prevalent [18]. 2) Nonimmune activities of MDSC; MDSC promote cancer progression by inducing angiogenesis and tumor invasion via the production of VEGF, bFGF, Bv8, and MMP-9. Moreover, MDSC stimulate the metastatic activities of cancer cell by facilitating EMT and creating "premetastatic niches." MDSC also induce "stemness" in certain cancers, which might be associated with resistance to chemotherapy or radiotherapy.

Nonimmune Activities of MDSC
In addition to suppressing immune responses within the TME, MDSCs also promote cancer progression by stimulating tumor angiogenesis and enhancing tumor cell invasion and metastasis ( Figure 1). These processes are regulated by MDSC-derived mediators, including vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), Bv8, and matrix metalloproteinase (MMP)-9, all of which are essential for tumor angiogenesis and cancer cell invasion [11,14]. Moreover, recent reports suggested that MDSCs stimulate the metastatic activity of cancer cells by facilitating epithelial-to-mesenchymal transition (EMT) or by creating "premetastatic niches" [15,16]. Importantly, the evidence has indicated that MDSCs induce "stemness," which might be associated with resistance to existing anticancer treatments including chemotherapy or radiotherapy [17].

MDSC Generation and Recruitment
In healthy individuals, IMC develop from common myeloid progenitor cells that differentiate into mature nonsuppressive myeloid cells such as DCs, macrophages, or granulocytes. Under pathological conditions such as infection, inflammation, or cancer, their differentiation was directed away from mature nonsuppressive cells to suppressive cells (MDSCs) [10,11]. It has been reported that PMN-MDSC undergo expansion in most patients with solid tumors. However, in certain cancers, such as melanoma, multiple myeloma, and prostate cancer, the M-MDSC population is more prevalent [18].

Recruitment of MDSC into TME
An important factor directing the migration of MDSC is chemokines. Studies have shown that multiple chemokines in the TME including C-X-C motif ligand (CXCL)1, CXCL8, CXCL12, C-C motif ligand (CCL)1, CCL2, CCL3, CCL5, CCL7, and their corresponding receptors on MDSCs (C-C chemokine receptor [CCR]2, CCR5, and C-X-C chemokine receptor [CXCR]4 differently regulate the recruitment of MDSCs [11]. These chemokines are not specific to particular types of cancer and have a high degree of redundancy.

The Frequency of MDSCs as a Prognostic Indicator or a Biomarker of Tumor Progression in Ovarian Cancer Patients
Increased numbers of circulating MDSCs have been detected in patients with various types of cancers [11,14,15]. Like in other cancer patients, as shown (Table 1), MDSCs were significantly increased in the peripheral blood mononuclear cells (PBMC), tumor or ascites in ovarian cancer patients [17,[22][23][24][25][26][27][28][29][30][31][32][33]. Obermajer et al. and Cui et al. are the first to demonstrate an increased MDSC in the ascites [33] and ovarian tumors [32] of patients with ovarian cancer, respectively. Cui et al. also found that an increased number of tumor-infiltrating MDSCs was significantly associated with a short survival in patients with high-grade serous ovarian cancer [32]. Since then, an increasing number of reports have consistently suggested a strong association between increased MDSCs and decreased survival in ovarian cancer patients. Although the increments of both PMN-and M-MDSC have been observed in ovarian cancer patients (Table 1), some reports have suggested that M-MDSC might be a more reliable predictor of the clinical stage or survival in ovarian cancer patients, compared to PMN-MDSC [26,27,31,33].  Interestingly, two recent reports have suggested an association between decreased MDSCs and favorable treatment outcomes in ovarian cancer patients. Lee et al. showed that germline BRCA1 and 2 mutation-associated ovarian cancer, which is believed to have higher response rates to platinum-based chemotherapy than BRCA wild-type [34], has fewer circulating MDSCs and higher CD8 + T cells in PBMC compared with BRCA wildtype ovarian cancer [24]. Second, Li et al. demonstrated that metformin treatment in diabetic patients with ovarian cancer was associated with reduced circulating MDSCs, a concomitant increase in the circulating CD8 + T cells, and longer survival [29].
In an effort to investigate the cause of increased MDSC production in ovarian cancer, some groups have found that increased MDSCs were associated with increased levels of IL-6 and IL-10 in ascites [31], and leukocytosis [17]. Interestingly, a recent investigation conducted by Komura et al. has suggested that increased MDSC is observed in ovarian cancer patients with leukocytosis or in those whose ovarian tumors exhibit increased G-CSF expression [17]. The G-CSF has been widely used clinically during the course of chemotherapy to reduce the risk of chemotherapy-induced neutropenia. Although there have been no clinical reports suggesting that the exogenous G-CSF treatment have a negative impact on the survival of cancer patients, the impact of G-CSF treatment on the survival or the progression of patients with ovarian cancer has been undetermined. Thus, future studies are needed to evaluate the impact of exogenous G-CSF treatment on the induction of MDSC and survival in ovarian cancer patients receiving chemotherapy.
Although all of these studies have suggested the prognostic significance of MDSC in ovarian cancer, there have been many limitations in these studies such as small sample sizes, the inconsistent the use of inconsistent MDSC surface markers, or limited clinical information (patient characteristics, clinical stage, response to treatment, or survival rates). Moreover, the association between the histological subtypes of ovarian cancer and the number of MDSC has never been investigated. To employ MDSC as a prognostic indicator in the clinical management of ovarian cancer, further investigations will be required using a comprehensive epidemiological model.
In addition to its suppressive activity against CD8 + T cells in ovarian cancer TME [17,32,37], previous studies have demonstrated that MDSC increase the stem cell-like properties of ovarian cancer cells via by producing PGE2 [17], inducing the microRNA101 [32] or CSF2/STAT3 pathway activation [23]. Moreover, by producing PGE2, MDSCs increase the PD-L1 expression in ovarian cancer cells by activating AKT/mTOR signaling, which may facilitate ovarian cancer cells to escape destruction by the immune system [17].

Preclinical Investigation of MDSC-Targeting Therapies
In murine ovarian cancer models, as shown in Table 2, various strategies aiming at eliminating MDSCs from the TME have been evaluated: anti-Gr-1 antibody [17,30], anti-GM-CSF antibody [35], CXCR2 or CXCR4 antagonists [28,36], PGE2 or COX-2 inhibition [33], metformin [29], thrombin inhibitor [38], or bis-benzylidine piperidone RA190 [39]. They showed significant antitumor activities when used as monotherapies or in combination with chemotherapy. Moreover, recent efforts have demonstrated that MDSC-inhibition therapies targeting IL-10 or CXCR4 enhanced the therapeutic efficacy of anti-PD-1 therapy, leading to prolonged survival [37,40]. However, as the association between the histological subtypes of ovarian cancer and the number of MDSC remains undetermined, we cannot tell which histological subtypes are the potential candidates for MDSC-inhibition therapy in ovarian cancer patients.

Strategy to Inhibit Human MDSCs
In murine studies, the anti-Gr-1 antibody has been widely used to eliminate MDSCs from TME. However, due to the absence of a Gr-1 homologue in humans, anti-Gr-1 antibodies cannot be used in the clinical setting, and no specific inhibitors of human MDSC are currently available. At present, as shown in Table 3, various strategies to target MDSC have been proposed: (1) depletion of MDSC; (2) inhibition of MDSC functions; (3) prevention of MDSC recruitment into TME; and (4) promotion of the differentiation of MDSC into mature, nonsuppressive cells. Some of these MDSC-targeting strategies have already been tested in solid cancer patients and demonstrated significant activity to reduce the number of circulating or tumor-infiltrating MDSCs . Table 3. Strategies for MDSC-targeting.

Clinical Trials Targeting MDSCs in Cancer Patients
Presently, biomarkers used to identify patients who might benefit from MDSCtargeting therapy have not been developed. However, as shown in Table 4, various clinical studies on MDSC-targeting therapies are currently underway either as a monotherapy, in combination with chemotherapy or as immune checkpoint inhibitors in patients with solid malignancies [72]. Hopefully, the next couple of years will bring exciting positive clinical data regarding MDSC-targeting therapies.

Conclusions
MDSCs are increased in ovarian cancer patients and play integral roles in disease progression. In order to inhibit their tumor-promoting effects, the efficacy of MDSC-targeting therapies (either as monotherapies or in combination with conventional treatments, including chemotherapy, radiotherapy, or anticancer immunotherapeutics) against ovarian cancer is currently being evaluated preclinically. We consider that increasing our understanding of MDSC biology will aid the development of optimal MDSC-targeting therapies, leading to the improvement of the prognosis of ovarian cancer patients.
Author Contributions: S.M., project development, manuscript writing/editing, funding acquisition. T.S., manuscript writing. N.K., manuscript writing. All authors have read and agreed to the published version of the manuscript.
Funding: This study was supported in part by Grants-in-aid for General Scientific Research 20K07596 from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.

Conflicts of Interest:
The authors declare no conflict of interest.