Introduction

Various microorganisms occupy niches within the human body, some of which are beneficial, and some are hazardous, considering their asymptomatic existence. The immune system responds to foreign pathogens through innate and adaptive immune responses; the former responding rapidly to pathogens to eliminate them, the latter requiring the recognition and formation of immunological memory before initiating an immune response (Flajnik and Du Pasquier 2004). As the first line of defense, natural killer (NK) cells, a prototypical innate lymphoid cell lineage, defend the host against microbial infections and tumors through potent cytolytic effects (Hamerman et al. 2005).

NK cells are considered the most cytotoxic cells against tumors in vitro (Bassani et al. 2019). NK cells reported serving as a promising candidate for adoptive cell therapy for in vivo cancer treatment (Ogbomo et al. 2011; Kim et al. 2019). In this field of adoptive cell therapy, chimeric antigen receptor (CAR)-NK cell therapy offers numerous advantages over CAR-T cell therapy. First, CAR-NK cells recognize the cancer cells through their receptors, thus targeting multiple antigens and mitigating tumor immune evasion by downregulating CAR-targeted antigens (Souza-Fonseca-Guimaraes et al. 2019). Second, unlike CAR-T cells, CAR-NK cells do not undergo clonal expansion in vivo, which can cause the “cytokine storm” resulting in severe side effects (Shimabukuro-Vornhagen et al. 2018; Lee and Kim 2019). Lastly, stringent HLA matching is not required for the cytotoxic function of NK cells, thus ameliorating the risk of graft-versus-host disease, which has been observed in numerous cases of CAR-T cell therapies (Bollino and Webb 2017). Moreover, apart from their cytolytic function, NK cells reportedly promote both innate and adaptive immune responses by secreting cytokines including interferon gamma (IFN)-γ and tumor necrosis factor (TNF)-α (Reefman et al. 2010; Oth et al. 2018). However, numerous issues limited its application in cancer treatment, including the sources of NK cells (Hu et al. 2019), time-intensiveness, methodological complexities for ex vivo expansion (Tanaka et al. 2019), the limited in vivo life span (Nayar et al. 2015), poor infiltration to solid tumors (Phung et al. 2019b), and the immunosuppressive tumor microenvironment (TME) (Tormoen et al. 2018).

This review summarizes various approaches using nanosystems to overcome the limitations of NK cell-based therapy and enhance the toxicity of NK cells towards cancer cells (Fig. 1). Prior to these approaches, the biological properties of NK cells are discussed, including tumor cell recognition, the interplay between NK cells and other immune cells, and immunosuppressive factors within the TME. Finally, the prospects for improving cancer treatment based on NK cell activity are discussed herein.

Fig. 1
figure 1

The intervention of nanoparticles for improved NK cell activity against cancers. Nanoparticles can promote the cytotoxic function of NK cells directly via targeting their activating or inhibitory receptors and block immune checkpoint components. Distinctly, nanoparticles can also enhance NK cell activation indirectly via targeting dendritic cells and T cells, which result in the elevated NK cell stimulatory cytokines or inhibiting the immunosuppressive cells, such as regulatory T cells (Tregs), and myeloid-derived suppressor cells (MDSCs) as well as their secreted cytokines, such as transforming growth factor-beta (TGF-β), IL-10, adenosine. MICA/B major histocompatibility complex (MHC) class I chain-related protein A/B, HLA human leukocyte antigen, DNAM-1 DNAX Accessory Molecule-1, NKG2A natural killer group 2, member A, NKG2D natural killer group 2, member D, KIR killer cell immunoglobulin-like receptor, PD-1 programmed cell death protein 1, PD-L1 programmed death-ligand 1, CTLA-4 cytotoxic T-lymphocyte-associated protein 4, A2AR adenosine A2A receptor

Full size image

“Show me your identity”: NK cells detect cell surface abnormalities before exerting cytotoxicity

NK cells recognize stress cells and tumor cells in a major histocompatibility complex (MHC)-independent manner (Paul and Lal 2017), and thus spontaneously eradicating various cancers without requiring prior sensitization and HLA or MHC matching. Through immune receptors, NK cells detect abnormal cells either on the basis of the lack of identifying molecules such as MHC class I, which bind to inhibitory receptors, or the upregulation of ligands for activating NK cell receptors (Morvan and Lanier 2016). They also directly eliminate cancer cells by triggering antibody-dependent cellular cytotoxicity (ADCC) (Wang et al. 2015).

Moreover, the interaction between mature NK cells and target cells induces the release of perforin (a membrane-disrupting protein) and granzymes (a family of proteolytic enzymes) from NK cells, followed by target cell lysis (Voskoboinik et al. 2015). In addition, NK cell activation upregulates TNF family members on the cell surface to mediate tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)- or Fas ligand (FasL)-induced apoptosis pathways (Kumar 2018).

In the past few decades, the identification of various activating signaling pathways that regulate NK cell activation has yielded promising targets to enhance NK cell activity towards cancer cells (Table 1). Furthermore, several cytokines, including type I IFN, IL-2, IL-12, IL-15, IL-18, and IL-21 are crucial for NK activation and proliferation (Freeman et al. 2015). However, IL-2 plays a contrasting role in immunomodulation. First, it stimulates effector immune cells that impede tumor progression (Mortara et al. 2018; Coyne and Narayanan 2019). However, it also functions as an immunosuppressor by maintaining the inhibitory function of Foxp3+ regulatory T cells (Tregs) and negatively regulating activated effector T cells and NK cells via activation-induced cell death (AICD), resulting in immune tolerance (Marks-Konczalik et al. 2000; Nguyen et al. 2019a; Ou et al. 2019). IL-15 distinctly inhibits AICD and does not activate Tregs (Waldmann et al. 2001), while inducing the activation and proliferation of cytotoxic T cells (CTLs) and NK cells (Zhang et al. 2018). IL-21, a member of 4-helical cytokines closest to IL-15, reportedly elevate NK cell progenitor maturation and activate the tumor lytic effects of NK cells through interaction with natural killer group 2, member D (NKG2D) (Takaki et al. 2005). Surface molecules, NKG2D and CD16 activate NK cells, thus facilitating onco-immunotherapy (Morvan and Lanier 2016). CD16 is reportedly one of the most important receptors mediating NK cell effector function by enabling them to detect antibody-coated target cells and exert ADCC (Naeim 2008). However, CD16 has been considered low-affinity Fc receptor, which limits the efficacy of the therapeutic anti-CD16 antibody, e.g., Rituximab (Snyder et al. 2018). NK Group 2D (NKG2D) binds MHC class I polypeptide-related sequence A, B (MICA, MICB) which are abundantly expressed in human (Dhar and Wu 2018).

Table 1 Natural killer cell-activating and inhibitory receptors
Full size table

The expression of CARs helps NK cells effectively eliminate solid tumors (Yong et al. 2018). The single-chain variable fragment (scFv) is recruited to develop NKG2D-based CARs with co-stimulation by Dap10, 4-1BB, or CD28, being clinically promising (Barrow and Colonna 2019). In another approach, bi- and tri-specific killer engagers (BiKEs and TriKEs) are encoded by a scFv against CD16 and tumor antigens. Owing to their small size (50–75 kDa), BiKEs and TriKEs have an enhanced biodistribution and further enhance NK cell-mediated tumor rejection (Davis et al. 2017).

“Dance with me”: NK cells in the context of the immune system and tumor microenvironment

The cross-talk of dendritic cells (DCs) with various immune effector cells, including NK cells has been previously reported. DCs, upon maturation, directly enhance NK cell activation by interacting with several NK cell-activating receptors including NKG2D, NKp30, and NKp46, and/or indirectly via DC-produced IL-2, IL-12, IL-15, and IL-18 (Van Elssen et al. 2014; Ashraf et al. 2019). NK cells in turn recruit conventional type 1 DCs by expressing DC chemoattractants such as chemokine (C–C motif) ligand 5 (CCL5) and chemokine (C motif) ligand (XCL1) to enhance the anti-cancer efficacy (Bottcher et al. 2018). Regarding the association between NK cells and T cells, NK cells potentially regulate T cell immunity via cytokine secretion or direct cytolytic activity of NK cells. Moreover, NK cells potentially induce IFN-γ signaling on CD4+ T cells, which results in their differentiation into TH1 helper cells. However, NK cells can reduce CD8+ T cell activation or even eliminate CD4+ and CD8+ T cells (Crouse et al. 2015).

A major factor contributing to the failure of cancer immunotherapy is the abundance of immune suppressors in the TME (Sau et al. 2018; Cho 2019; Park and Youn 2019; Phung et al. 2019a). The tumor elicits several mechanisms to alter the responses of effector immune cells, such as recruitment of immunosuppressive cells, upregulation of immune checkpoint molecules, and increased production of immunosuppressive cytokines including transforming growth factor (TGF)-β (Choi and Moon 2018; Mahjub et al. 2018). TGF-β potently suppresses both innate and adaptive immune responses, playing an important role in tumor progression via effector T cell and NK cell exclusion while increasing the Treg population in the TME (Heo et al. 2019). TGF-β suppresses NK cell effector function by downregulating transcription factors and activating receptors including CD16, NKG2D, and NKp30, decreasing cytokine production, and preventing NK activation via IL-12, IL-15, and IL-18 (Foltz et al. 2018). Besides activating receptors, NK cells express various inhibitory markers that maintain NK cells in the inactive state and limit their cytotoxic function (Table 1). By inhibiting these pathways, the anti-tumor activity of NK cells is induced.

“BiKE or CAR? Doesn’t matter, I need a road”: the integration of nanotechnology in boosting NK cell-based immunotherapies

Enhancement of NK cell activity by targeting activating receptors

NK cells are activated and expanded upon induction of cytokines or stimulating molecules. Nanocarriers have been used to incorporate various such molecules including membrane-bound IL-15 (Oyer et al. 2015), membrane-bound IL-21 (Foltz et al. 2018), IL-15 plasmids (Liu et al. 2018), and anti-CD16 antibody (Loftus et al. 2018). Oyer et al. proposed a particle-based strategy using plasma membrane (PM) particles derived from genetically engineered leukemia K562 cells co-expressing membrane-bound IL-15 and 4-1BB ligands (K562-mbIL15-41BBL) to selectively expand NK cells from peripheral blood mononuclear cells (PBMCs) (Oyer et al. 2015). PM particles led to markedly greater NK cell expansion than those stimulated with free soluble IL-15 and 1-4BB ligands. Furthermore, these expanded NK cells exhibited high cytotoxicity against several leukemia cell lines. However, long-term expansion of NK cells stimulated by feeder cells expressing membrane-bound IL-15 was reportedly limited via senescence potentially resulting from telomere shortening. Moreover, NK cells activated by IL-15 decreased the level of effector receptor CD16 via proteolysis of metalloprotease-17 (Romee et al. 2013). To overcome this limitation, Denman et al. reported that genetically modified artificial antigen-presenting cells (aAPCs) expressing membrane-bound IL-21 (mbIL21) prolong NK cell expansion up to 6 weeks without senescence and produce more cytokines than mbIL15-expressing aAPCs (Denman et al. 2012). Accordingly, Oyer et al. generated particles isolated from K562-mb21-41BBL cells (PM21) to effectively expand NK cells from the PBMCs ex vivo and in vivo (Oyer et al. 2016). In a murine model, administration of PM21 particles increased the population of peripheral blood NK cells; this study further investigated the effect of K562-mb21-41BBL cell-derived exosomes (EX21) on NK cell expansion in vivo and found that EX21 efficiently expanded NK cells and in vivo administration of NK cells expanded via EX21 displayed equivalent antitumor efficacy to those stimulated with PM21 (Oyer et al. 2017). Furthermore, Liu et al. encapsulated IL15-harboring plasmids into DOTAP and MPEG-PLA (DMA) nanoparticles. The DMA-pIL15 complex significantly inhibited tumor growth by inhibiting angiogenesis, promoting apoptosis, and reducing proliferation by activating host immunity (Liu et al. 2018). Moreover, Loftus et al. used the nanoscale graphene oxide (NGO) as a template to mimic the signaling receptor nanoclusters to activate NK cells by targeting the CD16 receptor. In particular, NGO was functionalized with the 8-arm star, amine-terminated poly(ethylene glycol) (PEG) via an EDC coupling reaction. Thereafter, biotin was conjugated with the free amine groups of the PEGylated NGO, followed by the streptavidin coating. Finally, streptavidin-coated PEGylated NGO-biotin was linked to the anti-CD16 antibody to achieve the antibody-functionalized nanoscale NGO clusters (NGO-α-hCD16). Their study reported that NGO-α-hCD16 specifically binds to human NK cells through the CD16 receptor. Moreover, the biomimetic nanocluster successfully promoted the cytolytic function of NK cells via the production of IFN-γ and CD107a upregulation (Loftus et al. 2018).

Mediation of NK cell toxicity via APC regulation

Several nano-vaccines targeting DCs not only augment cytotoxic T cells but also promote NK cell activation and proliferation. DC-targeted Poly(g-glutamic acid)-based vaccines co-delivering antigenic OVA and poly(I:C), a toll-like receptor-3 (TLR) stimulator, markedly increased the NK cell population and their activation in vivo and promoted NK cell homing to the tumor tissue (Kim et al. 2017). Furthermore, Kang et al. proposed necroptic tumor cell-mimetic vaccines (αHSP70p-CM-CaP) constructed by a phospholipid bilayer and a phosphate calcium (CaP) core to target both DCs and NK cells at lymph nodes for melanoma treatment (Kang et al. 2018b) (Fig. 2). The exterior surface of the bilayer of the vaccines was reconstituted with membrane protein B16OVA and peptide functionalized-αHSP70 (αHSP70), and the targeting ligand activating DCs and NK cells upon interaction with the CD49 receptor (Specht et al. 2015), CpG, and TLR-9 agonist were encapsulated in the CaP core. Vaccines with artificial PMs were efficiently delivered to the lymph node and induced the expansion of IFN-γ-producing CD8+ T cells and NKG2D+ NK cells. Furthermore, a combination of the αHSP70p-CM-CaP vaccine with anti-PD-1 therapy resulted in significant tumor regression in B16OVA tumor-bearing mice.

Fig. 2
figure 2

(Reprinted with permission from Kang et al. 2018a, b. Copyright© 2018 Elsevier Ltd.)

a Mode of action of necroptic tumor cell-mimetic vaccines (αHSP70p-CM-CaP) for cancer immunotherapy; and b fabrication of αHSP70p-CM-CaP.(Kang et al. 2018a; Kumar 2018).

Full size image

In another approach, extracellular vesicles, especially exosomes, have emerged as cancer therapy platforms owing to their biocompatibility, long circulation property, the reflection of their relevant parent cell identities, and easy modification to harbor therapeutic agents (Bach et al. 2017; Manandhar et al. 2018). In both preclinical and clinical settings, exosomes derived from matured DCs directly activate NK cells, and the binding of TNF molecules on the exosome membrane to their corresponding receptors results in elevated IFN-γ production in NK cells (Reiners et al. 2014). In this context, several studies have attempted to engineer DC-derived exosomes (DEX) to promote in vivo NK cell activation. Viaud et al. reported that DEX-expressing membrane-bound NKG2D and IL-15Rα ligand directly mediate NK cell activation and proliferation in humans and mice. A phase I clinical study reported that DEX immunization significantly increased the population of circulating NK cells and restored their NKG2D-dependent function (Viaud et al. 2009). In another study, DEX generated from poly(I:C) and tumor antigens-pulsed DCs robustly stimulated and recruited antigen-specific CTLs and NK cells to the tumor, significantly inhibiting tumor growth in a mouse model of melanoma (Damo et al. 2015).

Co-delivery of NK cells and nanoparticles via direct conjugation

Chandrasekaran et al. adorned liposomes with TRAIL and anti-NK1.1 proteins via maleimide-thiol chemistry, mediating the conjugation of liposomes with natural killer cells. Targeting of NK cells with TRAIL liposomes prolonged their presence within the tumor-draining lymph nodes, further inhibiting the spread of cancer cells from the primary site (Fig. 3) (Chandrasekaran et al. 2016). Siegler et al. further induced the expression of chimeric antigen receptors to redirect their antitumor specificity. Furthermore, paclitaxel-loaded cross-linked multilamellar liposomal vesicles were conjugated with the NK cell surface. This combination of immunotherapy and chemotherapeutic boosted the antitumor efficacy in Her2- and CD19-overexpressing cancer models (Siegler et al. 2017).

Fig. 3
figure 3

(Reprinted with permission from Chandrasekaran et al. 2016. Copyright © 2016 Elsevier Ltd.)

Schematic representation of liposome formulations used herein. Liposomes were adorned with TRAIL and anti-NK1.1 antibodies via maleimide-thiol chemistry. Liposomal anti-NK1.1 mediated the conjugation of liposomes to NK1.1-expressing natural killer cells to form “super” natural killer cells (Chandrasekaran et al. 2016).

Full size image

Nanoparticles to direct NK cells to tumor tissue

The ideal strategy for enhanced tumor infiltration of NK cells is to upregulate the NK cell-activating ligands and trigger the production of NK cell stimulatory cytokines in the TME. Tan et al. reported an approach to enhance the migration of activated T cells and NK cells to tumors through delivery of a DNA fragment encoding NKG2D ligand and IL-21 (dsNKG2D–IL-21) to the TME. The fused dsNKG2D–IL-21 gene was effectively delivered to the tumor through chitosan-based nanoparticles, resulting in enhanced secretion of the NKG2D ligand and IL-21 by tumor cells, thereby promoting T and NK cell activation and accumulation in the tumor tissue. The authors observed that administration of dsNKG2D–IL-21-chitosan nanoparticles considerably delayed tumor growth and prolonged the life span of treated mice (Tan et al. 2017). Furthermore, Meraz et al. developed cationic liposomes delivering tumor suppressor candidate 2 (TUSC2) plasmid DNA to the tumor tissue to investigate its potential to activate and increase the NK cell population in the TME (Meraz et al. 2018b). TUSC2 is a potent tumor suppressor gene in lung cancer (Rimkus et al. 2017). However, TUCS2 mRNA is reportedly downregulated or suppressed in approximately 80% of tumors, which is associated with low overall survival (Rimkus et al. 2017). TUCS2 majorly contributes to cancer therapy, e.g., in inducing tumor cell apoptosis, inhibiting signaling for drug resistance, regulating crucial cytokines to maintain homeostasis, and stimulating innate and adaptive immunity (e.g., IL-2, IL-15). NK cell-mediated cancer eradication is also an important arm of TUCS2 gene therapy. A pervious study generated a cationic liposome comprising DOTAP and cholesterol to form a complex with the TUCS2 and reported that intravenous administration of the lipoplexes increased the number of NK cells in the TME of tumor-bearing mice, probably owing to the TUCS-2 upregulation-mediated release of proinflammatory cytokines and IL-15 while decreasing the population of Tregs and myeloid-derived suppressor cells (MDSCs) (Fig. 4). In the same context of nanoparticle-induced cytokine production by NK cells in the TME facilitating NK cell delivery to tumors, a PEGylated liposome co-encapsulated cdGMP, a STING agonist, and MPLA, a TLR-4 agonist, potently induced type I interferons in the TME, thus increasing the delivery of NK cells to the tumor (Atukorale et al. 2019). NK cells could also be directed to the tumor by sensing chemokine attraction. NK cells could be directed to tumors via an external simulator, actively migrating through chemokine signaling. Park et al. developed immunomodulatory Degradex® poly(lactic-co-glycolic acid) microspheres containing recombinant IFN-γ. Their delivery induced the release of chemokines in hepatocellular carcinoma, thus increasing NK cell infiltration (Park et al. 2017).

Fig. 4
figure 4

(Reprinted with permission from Meraz et al. 2018a, b. Copyright© 2018, American Association for Cancer Research)

Combinatorial delivery of liposomes harboring tumor suppressor candidate 2 (TUSC2) with anti-PD-1 therapy increased the number of tumor-infiltrating NK and CD8 T cells while decreasing the number of myeloid-derived suppressor cells (MDSCs) and regulatory T cells (Tregs) (Meraz et al. 2018a).

Full size image

Using a different approach, several studies have reported the potential of an external magnetic field to guide NK cells to the tumor. NK cells were engineered with magnetic nanoparticles ex vivo, infused in the body, and traced via external magnetic exposure (Jang et al. 2012; Oh et al. 2012; Wu et al. 2018; Burga et al. 2019). Jang et al. loaded silica-decorated superparamagnetic iron oxide (Fe3O4/SiO2) conjugated with fluorophore (Cyanine 5.5) into NK cells (Jang et al. 2012). Thereafter, the magnetic nanoparticle-modified NK cells were intravenously injected into B cell lymphoma-bearing mice and exposed to a magnetic field (Fig. 5). Consequently, NK cell infiltration in tumors was increased by 17-fold when applying the magnetic field, and nanoparticle labeling did not affect NK cell function. However, brief intratumor retention of NK cells was observed upon removal of the magnetic field. To overcome this limitation, Wu et al. implanted a magnetic plate to the hypodermic tumor, followed by injection of NK cells loaded with magnetic nanoparticles comprising an Fe3O4 core and polydopamine layer (Wu et al. 2018). This strategy showed improved accumulation and retention of NK in tumors, resulting in significantly enhanced therapeutic efficacy.

Fig. 5
figure 5

(Reprinted with permission from Jang et al. 2012. Copyright© 2012 Elsevier Ltd.)

Schematic representation of the preparation of silica-decorated superparamagnetic iron oxide (Fe3O4/SiO2) conjugated with Cyanine 5.5 for incorporation in NK cells for cancer therapy (Jang et al. 2012).

Full size image

Direction of nanoparticles to tumor sites using NK cell materials

Owing to the natural immunosurveillance properties of diseased/stress cells, NK cell membrane-associated components are recruited to form the NKsome. NK-NPs selectively accumulate in the tumor and eliminate primary tumor growth (Deng et al. 2018; Pitchaimani et al. 2018, 2019; Zhu et al. 2018). Pitchaimani et al. isolated the receptor proteins from activated NK-92 cells and infused them into liposomes to form NKsomes. Doxorubicin loaded NKsomes inhibited the tumor growth up to 78.5% after marked accumulation of vesicles at the tumor area (Fig. 6) (Pitchaimani et al. 2018).

Fig. 6
figure 6

(Reprinted with permission from Pitchaimani et al., 2018. Copyright© 2018 Elsevier Ltd.)

Natural killer cell membrane-infused biomimetic liposomes for targeted tumor therapy (Pitchaimani et al. 2018).

Full size image

Release NK cell activity by targeting immunosuppressive pathways

Park et al. reported a strategy to overcome TGF-β-mediated immune inhibition in the TME to reverse the anti-tumor activity of CTLs and NK cells. Using liposomal polymeric gels (nLGs), they co-delivered TGF-β inhibitor and IL-2 to the tumor site, reporting marked tumor inhibition and enhanced survival of tumor-bearing mice (Park et al. 2012). In particular, nLGs were generated by extruding the complex of TGF-β inhibitor and acrylated cyclodextrin, IL-2, PLA-PEG crosslinker, IL-2, DSPE-PEG-NH2, phosphatidylcholine, and cholesterol, followed by ultraviolet irradiation, resulting in polymerization of the crosslinker and cyclodextrin complex, subsequently encapsulating the TGF-β inhibitor and IL-2 within the hydrogel core. After intravenous administration, nLGs effectively accumulated at the tumor site owning to the enhanced permeation and retention (EPR) effect. The proportion of activated CD8+ T cells and NK cells in the tumor tissue was markedly increased, while the number of Tregs was reduced.

Immunosuppressive cells are critical for inhibiting NK cell function and inducing tumorigenesis. Tumor-associated dendritic cells (TADCs) have recently been considered potential targets for cancer immunotherapy. Although DCs are essential for initiation of anti-tumor immune responses, they are typically inactivated and dysfunctional under cancerous conditions; this is associated with effector T cell deletion and an increase in Tregs, thereby leading to immune tolerance (Truong et al. 2019). Thus, the redirection of dysfunctional TADCs to fully functional DCs is important for promoting immunotherapeutic effects against cancer. TLR agonists including poly(I:C), imiquimod, and CpG are commonly used to activate DCs, subsequently inducing CTL responses towards cancer (Bastola and Lee 2019; Tran et al. 2019; Wang et al. 2019). However, accumulating evidence indicates that TADCs respond poorly to TLR stimulators, thus deterring the development of effective cancer immunotherapy (Idoyaga et al. 2007). These studies have identified the crucial role of microRNAs (miRs) in regulating TLR simulation. Inhibition of miR-148a efficiently restores the sensitivity of TADCs to TLR-3/4 agonists. These findings suggest a rational approach to combine TLR stimulation with miR-148a inhibitor (miR-148ai) to synergistically reprogram TADCs to robustly facilitate anticancer immune responses. Recently, Liu et al. co-entrapped miR-148ai with poly(I:C), a the TLR-3 agonist, and OVA antigen into polypeptide micelles to formulate the polypeptide micelle/poly (I:C)/OVA/148ai (PMP/OVA/148ai) nanovaccines and reported that vaccination with the PMP/OVA/148ai led to increased mature DCs in the spleen and tumor, reduced Tregs and MDSCs in the tumor. Furthermore, upon administration of PMP/OVA/148ai, the cell population in the TME of effector immune cells was significantly elevated (2- to 3-times for CD4+ and CD8+ T cells, and 3- to 4-times for NK cells), thus markedly reducing the tumor burden (Liu et al. 2016).

“Where do we go from here? Start from the end”: conclusions and future directions

Advancements in materials science have placed nanotechnology at the frontier of onco-immunotherapy with several properties including: (1) the incorporation of multiple therapeutic agents in the same platform, thus facilitating effective combination therapies (Nguyen et al. 2017; Hwang et al. 2018; Le et al. 2018; Phung et al. 2019c), (2) chemical or biological modification of nanoparticles to specifically deliver payload agents to NK cells through surface functionalization with targeting ligands (Choi and Han 2018; Al-azzawi and Masheta 2019), (3) nanoparticle-mediated protection of bio-macromolecular components including peptides, nucleic acids, and proteins from in vivo degradation, resulting in increased therapeutic efficacy (Ghosh et al. 2018; Nguyen and Jeong 2018; Park et al. 2018; Jang et al. 2019; Lee 2019). NK cell-based therapy has attracted increasing attention in studies on cancer treatment. However, despite the marked potential of NK cells to eliminate tumor cells, the clinical translation of NK cell therapy remains challenging, probably owing to difficulties in the mass production of NK cells, their poor persistence and delivery at the tumor site, and a reduction in their activity upon administration to patients, together with numerous immunosuppressive factors in the TME. In this context, mobilization of host NK cells by nanoparticles might be a potential alternative to NK cell therapy. In several preclinical and clinical studies, engineered nanoparticles delivering therapeutic agents including antibodies, stimulatory cytokines, genes, or adjuvants have increased NK cell activity and proliferation and promoted NK cell migration to tumor sites, thus markedly inhibiting tumor progression.

However, few studies have thus far evaluated the potential of nanoparticles in modulating the anti-tumor activity of NK cells and the degree of therapeutic efficacy of NK cells, probably owing to their close proximity with the other lymphocytes and thus deterring treatment via specifically targeting NK cells (Nutt and Huntington 2019). However, simultaneous expression of several markers in NK cells and the other effector immune cells including T lymphocytes might provide an interesting therapeutic approach to synergistically augment the activity of both cell types to treat cancers. Similar to T cells, NK cells express numerous immune checkpoint molecules on the surface (e.g., CTLA-4, TIM-3, and PD-1) (Stojanovic et al. 2014; Huang et al. 2015; Xu et al. 2015), along with immunosuppressive intracellular signaling components (e.g., Cbl-b, CDk8) (Guillerey et al. 2016). Thus, therapies targeting these shared components not only elevate T cell activity but also improve NK cell toxicity towards the tumor. These findings suggest a promising strategy to design combinatorial treatment-based nanoplatforms to stimulate both T cell and NK cell responses to cancers.

Until recently, NK cell therapy has exhibited higher efficacy against hematological cancers rather than solid tumors, probably owing to their low tumor-infiltrating potential (Habif et al. 2019). The generation of nanoparticles targeting the TME to upregulate NK cell-activating ligands and stimulatory cytokines might be a potential approach to improve the delivery of NK cells to solid tumors. The discovery of the immunomodulatory effects of numerous approved chemotherapeutic drugs has been paradigm-shifting in cancer therapy. Indeed, chemotherapies upregulate NK cell-activating factors, thus improving NK cell-mediated recognition and eradication of tumor cells (Carotta 2016). Furthermore, nanoparticles reportedly increased the therapeutic efficacy and safety of chemotherapy (Lee et al. 2018; Nguyen et al. 2018a, b; Nguyen et al. 2019b; Pham et al. 2019; Soe et al. 2019). Therefore, regarding the elevation in ligand-mediated NK cell activation in tumor tissues, nanoparticles delivering chemotherapeutic agents to the tumor could be an effective strategy, especially in combination with different therapeutic agents to enhance anticancer activity.

Finally, numerous studies on NK cell biology and the efficacy of cancer treatment in both preclinical and clinical settings provide insights into the anti-tumor effects of NK cells, along with the interplay between the immune system and tumors. Furthermore, advancements in high-throughput technologies and big data sciences might further elucidate NK cell biomarkers crucial for the development of therapeutic agents and nanosystems specifically targeting NK cells for effective cancer immunotherapy.