Natural killer (NK) cells were first found in peripheral blood due to its function which could target and destroy leukemia cells [1]. The percentage of NK cells in all of lymphocytes was estimated about 10% - 15%, and its half-life was approximately 7 to 10 days in blood [2]. Later studies have found that NK cells are also located in distinct tissue sites, including liver, lung, bone marrow, thymus, intestine and lymph nodes, ect., where they exert unique roles in the particular tissues, and we named it tissue-resident NK cells (trNK) [3] [4] [5] [6]. Unlike B and T cells, NK cells in blood circulation could accurately recognize and lyse many kinds of cancer cells, senescent cells and virus-infected cells without prior sensitization [7] [8] [9]. We now know that mature NK cells are a heterogeneous population in humans which could be divided into two subsets: CD3⁻CD16⁺CD56^bright and CD3⁻CD16⁻CD56^dim [10], each subset performs different roles in vivo, like CD3⁻CD16⁺CD56^bright mainly for immune regulation through producing cytokines, such as interferon-γ (IFN-γ), tumor necrosis factor-α (TNF-α) ect., another subset CD3⁻CD16⁻CD56^dim mainly for cytotoxic effect which has been mostly applied for killing cancer cells in the immunotherapy against both hematologic and solid cancers in clinic [11]. However, in mice, mature NK cells are usually expressing CD27 and CD11b in circulation, and they are grouped into four subsets by surface density: CD11b^lowCD27^low, CD11b^lowCD27^high, CD11b^highCD27^high, and CD11b^highCD27^low as described in Laura et al. [12].

There are a number of receptors expressed in NK cell membrane, they could be divided into activating and inhibitory receptors according to their different functions. Those receptors can recognize ligand protein expressed in the target cells, thereby controlling the cytolytic function and cytokines-produced function. Uniquely, they can discriminate between normal and abnormal cells via “missing-self” or “induced-self” recognition models [13] [14]. Inhibitory receptors can transmit negative signals that block NK cell auto-reactivity, and many researchers have found major histocompatibility complex (MHC) class I molecules (also known human leukocyte antigens (HLA) class I in humans), whose absence may lead to NK activation, are tightly recognized by inhibitory receptors, which is the so-called “missing-self recognition” [14] [15]. The most important inhibitory receptors in humans are represented by killer cell immunoglobulin like receptors (KIRs) specific for allotypic determinants shared by groups of HLA-A, -B or -C alleles [16]. In the later studies, scientists found it was not sufficient to induce NK cell activation by the absence expression of MHC molecules, but rather signaling from activating receptors was necessary, such as NKG2D, which is the so-called “induced-self recognition” [17] [18] [19]. Now it has a consistency that the balance between signals from those two kinds of receptors governs the activation or inhibition of NK cells, these cells remain homeostasis in vivo due to the predominance of inhibitory signals over activating signals, but when signals from activating receptors prevail, NK cells will respond immediately for cytolytic effect [20].

Activating receptors expressed in NK cell membrane, such as NKG2D, NKp30 and NKp46, are considered the most relevant receptors in the process of activating NK cells. Once activated, NK cells can destroy tumor cells, senescent cells and virus-infected cells by releasing the cytotoxic molecules perforin and granzyme, leading to the increasing expression of Fas ligand (FASL), TNF-related apoptosis-inducing ligand (TRAIL) and inflammatory cytokines and by antibody-dependent cellular cytotoxicity (ADCC) [21] [22] [23]. In addition to their direct cytotoxic function against abnormal cells, NK cells also have a specific role in the homeostasis of the immune system, especially initiate other immune cells (T cell, B cell, DC cell, macrophage ect.) activation via secreting cytokines and chemokines [24]. NK cells indirectly increase the immune responses of T cells by facilitating DC maturation [25]. Many scientific findings have come to the conclusion that NK cells can promote the antiviral activity of CD8+ T cells effect and moderate T cells exhaustion [26] [27].

NK cells with the powerful killing capacity have been used in clinic to getting successful results in the treatment of hematological malignancies, such as acute myeloid leukemia [28] [29], malignant lymphoma [30], relapsed and refractory leukemia [31] et al., but the clinical outcomes in the solid tumor were not satisfied, later studies show that NK cells function, phenotype and activation are severe damaged by the tumor microenvironment, even gave rise to NK cells dysfunctional or exhausted [32] [33]. Another challenge is that NK cell is hard to migrate to and infiltrate into the tumor sites, as a result, NK cells cannot effectively recognize and lyse the target cancerous cells. Thus, it is really essential to develop effective methods to increase the homing of NK cell to the tumor stroma and the cytolytic function. In the review, we discuss the current strategies and further optimize these strategies to enhance the clinical outcome of NK cell-based immunotherapy against solid cancer.

Although owing the potent killing capacity, NK cells still face major challenges that dampen their clinical outcome in the complicated microenvironment in vivo. Many experiments have found that human NK cells, infiltrated into the tumor sites, have changed expression of activating and inhibitory receptors in surface and damaged its cytolytic functions [34] [35]. A number of signal pathways regulate NK cell suppression in the tumor microenvironment, several of which also give rise to impairing the T cells role [36]. It is thought that cancer cells have the capacity to evade immunosurveillance by shedding ligands which are expressed on the transformed cell membranes and can bind to the NKG2D activating receptors on NK cells. Recently, releasing of soluble NKG2D ligands have been drawn a great deal of attention due to its regulating process for NK cells activation. The soluble NKG2D ligands bind NKG2D receptors on NK cell surface, which could prevent their function to engage with the ligands on cancer cell membrane that would exert a cytolytic role [37] [38]. Using antibodies to target the soluble ligands has proved successful in multiple fully immunocompetent mouse models and reduced human melanoma metastases in a humanized mouse model [39]. However, Deng et al. found the opposite role of the soluble NKG2D ligands, which increase NK cells cytolytic function, as in the case of soluble MULT1, which prevented NK cell desensitization in the animal cancer model [40]. Those results demonstrate that soluble ligands show distinct role in a specific microenvironment and it needs to be more investigated.

Cells isolated from the tumor microenvironment in hepatocellular carcinoma [41] and cancer-associated fibroblasts(CAF) patients [42] [43] [44] may have suppressive functions against NK cells, such as myeloid-derived suppressor cell (MDSCs) and tumor-associated macrophages (TAM), which can constitute an intricate tumor microenvironment together that restrain NK cell activity by changing the balance between NK activating and inhibitory receptors in favors of impaired immunity [45] [46] [47]. As reported in recent years, the role of NK cells in tumor microenvironment in ovarian cancer patients is severely impaired, whereas their counterpart exhibit normal function compared to that of healthy controls [48]. There are a large number of immunosuppressive cytokines and the other soluble molecules that have a great influence on NK cell maturation, proliferation and functionality in tumor microenvironment, including TGF-β, TNF-α, IFN-γ, indoleamine 2,3-dioxygenase (IDO), IL-4, prostaglandin E2 (PGE2), lactate dehydrogenase A (LDHA) et al. [43] [49] - [55]. Those cytokines, especially TGF-β mainly secreted by CAF in tumor microenvironment [56], could induce down expression of NKG2D in membrane, and finally leading to a diminished cytotoxicity of NK cells [57], also CAFs can recruit monocytes into the TME to promote polarization of M2 macrophages and drive inhibition of NK cell function [58]. Apart from the impaired cytotoxicity, TGF-β can change the anti-metastatic function [59], metabolism [60] and mitochondrial role [61] in NK cells. Some experiment results show that NK cells can also transform to type 1 innate lymphoid cells by non-canonical TGF-β signaling, allowing for tumor growth and metastasis in mice [62] [63]. Cancerous cells have the ability to escape the attack by NK cell, due to the prevention of cell contact. An anti-adhesion glycoprotein mucin 16 (MUC16) are continuously expressing on the ovarian cancerous cell membrane, which can inhibit synapse formation between NK and cancerous cells [64].

For NK cell-based immunotherapy against human cancers, it is much easier for NK cell to get access to hematological malignancies compared with the immunosuppressive microenvironment of solid cancers [65]. and NK cells are the first subtype of innate lymphoid cells to recovery after human allogeneic hematopoietic stem cell transplantation (HSCT) [66]. So NK cells are mostly used as a “star cell” in the immunotherapy against hematological malignancies as well as T cells. But recent years, there are more than 80 clinical trials (https://www.clinicaltrials.gov/) to assess the NK cell-based immunotherapy in variety of solid cancers, either as monotherapy, or combined with other regents.

Autologous NK cells have been applied for immunotherapy against solid tumors, such as advanced and/or metastatic digestive cancer, metastatic melanoma, renal cell carcinoma, but clinical responses were not observed after intravenous infusion [67] [68]. It is also found that autologous NK cells are not able to expand ex vivo in numbers for clinical infusion [69], this could be due to the radiotherapy or chemotherapy treatments, which can destroy both cancerous and normal cells, received by the patients before NK cell isolation from the peripheral blood, which may provide the explain for poor clinical outcomes. Some studies also show that allogeneic NK cell infusion has the promising clinical results in the treatment of pediatric refractory solid tumors, advanced non-small cell lung cancer and advanced solid tumors [30] [70] [71], it is not only safe and feasible, but also exhibited clinically effective. The human cell line NK92 has also been a cell source of allogeneic NK cell-based immunotherapy in clinic, and have observed encouraging clinical responses in advanced lung carcinoma patients, but its lifespan in patients peripheral blood is rarely limited, and only persists for 48 h [72] [73] [74]. Considering the source of NK92, more investigations should be conducted to validate the safety in the clinic. As for off-the-shelf NK cells, pluripotent stem cell could be differentiated into NK cell in vitro, and exert effective cytolytic function against cancerous cells, both in vivo and in vitro in animal model [75] [76] [77]. But these pluripotent stem cell-derived NK cells have not been applied in the clinical treatment due to its unverified safety and effectiveness in human. With the encouraging success of CAR-T cells for hematological malignancies treatment, there is a growing interest in developing CAR-NK cells for immunotherapy in both hematological malignancies and solid cancer in vivo. The most common target antigens in solid cancerous cells were constructed including EGFR [78], EGFRvIII [78] [79], HER-2 [80], PSMA [81], ROBO1 [82], MUC-1 [83] [84], CS1 [85] and Mesothelin [86] et al.. Although with various target antigens in the CAR-NK cell field, but limited progress we have made in clinic so far, especially for the solid tumor immunotherapy. Only eight clinical trials were registered to assessing the safety and effectiveness of CAR-NK cells in solid tumor patients, all of those trials are ongoing currently.

Many studies have come to the conclusion that the tumor microenvironment was the major factor that resulting tumor escape from NK cell anti-tumor immunity against solid cancers [45] [46] [87]. Thus efforts must be concentrated on the following aspects to augmenting the clinical outcomes of NK cell-based immunotherapy: 1) increasing the number of functioning NK cells in tumor stroma; 2) recuperating damaged metabolism and effectors roles in order to boost the cytotoxicity of NK cells against cancerous cells.

As we described above, a high concentration of TGF-β secreted by TME blocks local NK cell anti-tumor immunity response. A study found that TGF-β receptor-I inhibitor SB505124 could give rise to incredible increased percentages and absolute numbers of NK cells in the tumor site in metastatic B16 melanoma murine models [88]. Another study showed that the use of TGF-β inhibitor, TGF-β receptor kinase inhibitor LY2157299, could increase NK cell infiltration into liver tissue in a model of colon cancer metastasis [89]. Young et al. [90] have found that both A2AR inhibitor and A2AR gene knockdown could target A2AR/ADO signaling to augment the accumulation of NK cells which have a highly cytolytic function, and finally leading to inhibition of tumor progression. Local administration of Car-T cell has been done in patients with solid cancers, such as breast cancer, malignant mesothelioma, medulloblastoma and colorectal cancer, and great success has been achieved in ClinicalTrials in recently years [91] [92] [93] [94]. Following those train of thoughts, an ongoing phase I clinical study, aiming at recurrent HER2-positive glioblastoma with the number of 1 × 10⁷ to 1 × 10⁸ by intracranial injection of NK-92/5.28.z cells, has been recruiting 30 participants (NCT03383978).

The lifespan of NK cells in blood circulation was really short, and they exhibit finite persistence in vivo [95], which can restrict their therapeutic efficiency in clinic. A Clinical Trial conducted by Bachanova shows that the persistence of NK cell has been greatly elevated by infusing IL-2 in adoptively transferred haploidentical NK cell in patients with refractory acute myeloid leukemia [96]. The similar study was finished in liver metastases of gastrointestinal carcinoma by infusing NK cells combined with IL-2 and Cetuximab [97], but the immune activation by IL-2 is not specific for NK cell, as it can also activate other immune cells, such as Tregs [98]. IL-15 has the significant effect on NK cell function [99], a number of studies have shown that NK cells with longer persistence, higher proliferative capacity and stronger cytotoxicity were got in vivo and in vitro by adding IL-15 or combining other cytokines, such as IL-12, IL-18 and IL-21 [100] [101] [102] [103]. the application of recombinant IL-15 in treating patients with different kinds of advanced solid cancers has been completed recently (NCT01727076), but no result was published until now. Apart from the cytokines stimulation, targeting metabolic pathways linked to survival could be another hopeful method to boost their persistence. Direct activation of citrate-malate shuttle by activating the transcription factor Srebp has been shown a promising strategy to promote glucose metabolism and cytolytic function of NK cell, and finally could have a role in enhancing their persistence [104]. Blocking glycolytic metabolism with 2-NBDG [105] or inhibiting Akt [106] could be another valuable therapeutic strategy which both have been elucidated on CD8+ T cells. All that being said, metabolic network about NK cell has not been fully understood, so more efforts should be devoted to clarify the mechanism about the overall network.

NK cells have been found for approximately 40 years, and they have become great potential killer against cancerous cells for its board cytotoxicity and good safety. Despite many challenges in clinic we have to meet, researchers still apply NK cells alleviating the patients’ pain to some extend which is suffering from the solid cancers. A number of preclinical and clinical trials have been developed with the precise available methodologies, such as engineered NK cell (TGF-β-resistant NK cell [107] [108], CAR-NK cell [109] [110] [111] ), NK-92 cell line [72] [73] [112], autologous or allogeneic NK cell combined with small molecules [103] [113] [114] [115]. The clinical therapeutic efficiency is not satisfied due to its persistence and difficulty to infiltration into the tumor sites. As we mentioned above, the key areas of developing strategies to improving clinical efficiency may mainly concentrate on homing of functioning NK cell in tumor sites and restoring activity to enhancing the cytolytic functions. It is most likely that the most beneficial therapeutic responses will be seen with the combination therapies.

This work is supported by CEO of Neology Zhu Yi and Business director of Neology Huang Guojun and Wang Ruoyu. The director of hospital Tao Qiqiang kindly gives advices to select the topic and supervise the manuscript. Li Wenrui assists to search papers related to NK cell immunotherapy.

The authors declare no conflicts of interest regarding the publication of this paper.

Lv, P.L., Li, W.R., Wang, R.Y., Huang, G.J., Zhu, Y. and Tao, Q.Q. (2021) Natural Killer Cell-Based Immunotherapy against Solid Cancer. Open Access Library Journal, 8: e7056. https://doi.org/10.4236/oalib.1107056