Potential applications of nanoparticles in cancer immunotherapy

ABSTRACT

In recent years considerable progress has been made in the field of cancer immunotherapy whereby treatments that modulate the body's own immune system are used to combat cancer. This has the potential to not only elicit strong anti-cancer immune responses which can break pre-existing tolerance and help promote tumor regression, but could also induce immunological memory which may help prevent tumor recurrence. In order to ensure effective delivery of immunotherapeutic agents, such as vaccines, checkpoint inhibitors, chemotherapeutic agents and nucleic acids, a safe and effective delivery system is often required. One such approach is the use of multifunctional nanoparticles (NPs), such as liposomes, polymers, micelles, dendrimers, inorganic NPs, and hybrid NPs, which have the potential to combine the delivery of a diverse range of therapeutic immunomodulators thereby increasing the efficacy of tumor cell killing. This review focuses on recent progress in NP-mediated immunotherapy for the treatment of cancer.

KEYWORDS:

• cancer
• delivery system
• immunotherapy
• immunomodulation
• nanoparticles

Introduction

In recent years there has been considerable progress made in the prevention and treatment of cancer; nevertheless it remains one of the main causes of morbidity and mortality worldwide, with approximately 14 million new cases and 8.2 million cancer related deaths in 2012; a number expected to rise in the coming decades.¹ Cancer refers to a series of diseases characterized by the uncontrollable division of abnormal cells which may invade or spread to nearby healthy tissue and organs resulting in the formation of malignant tumors. Tumor specific antigens (TSA) are exclusively expressed on tumor cells and not on any other cells whereas tumor associated antigens (TAA) are endogenous antigens present on both tumor and normal cells. Although many of these antigens should be immunogenic, cancers that are detected clinically have evaded anti-tumor immune responses by a combination of immune evasion, induction of tolerance and downregulation of T cell signaling which has enabled tumors to progressively grow.² Even with improvement of patient's survival by conventional therapies (surgery, radiation, chemotherapy etc.), many types of cancer are still incurable or are associated with severe therapy-related adverse effects.^3,4 However, in recent years we have gained an increased knowledge of tumor immunology that has allowed the development of novel therapies with minimal toxicity.

Immunotherapies against existing cancer either stimulate the activities of specific effector mechanisms of the immune system or counteract inhibitory or suppressive mechanisms. The generation of tumor-specific immunity involves multiple sequential immune actions: the release of antigens from tumor beds, presentation of tumor antigens by antigen-presenting cells (APCs), priming and activation of T cells by activated antigen-presenting cells (APCs), migration and infiltration of effector T cells back to the tumor, and finally the recognition and killing of tumor cells by effector T cells (Fig. 1).^5,6 It has been recognized that T-lymphocytes, particularly cytotoxic T Lymphocytes (CTL) that can selectively target and destroy malignant cells, play a pivotal role in immune cancer therapy. However, their isolated response is not sufficient enough to mount an effective stimulation and killing, but instead it is the cooperation between CTL, T helper (Th) cells, B cells and mature dendritic cells (DC) that has the potential to reduce the tumor mass and induce immunological memory to control tumor relapse.^7,8

Figure 1. The generation of tumor-specific immunity involves multiple sequential immune actions.

A number of different immunotherapeutic approaches against cancer have been proposed. These can broadly be divided into active and passive immunotherapies. Active immunotherapies involve stimulation of the body's own immune response whether it be specifically via the administration of vaccines containing tumor-specific antigens and/or tumor-associated antigens or non-specifically, for example via the administration of cytokines such as IL-2 or IFN-α. In contrast, passive immunotherapy, for example via the administration of monoclonal antibodies or adoptive T cell transfer, targets the disease itself rather than activating the immune system to do so.^9,10

A number of novel immunotherapeutics against existing cancer have recently been approved which highlight the potential of these approaches. The first therapeutic cancer vaccine product, sipuleucel-T (Provenge®), a cell-based cancer immunotherapy for prostate cancer, was approved in the US in 2010, and greatly expedited the clinical development of other novel cancer immunotherapeutics.^11,12 More recently, an increasing number of monoclonal antibodies, including those termed checkpoint inhibitors which target the immune-inhibitory pathways activated by cancer cells, have now been approved. For example Ipilumumab (Yervoy), an anti-CTLA-4 monoclonal antibody (mAb) was approved by the US Food and Drug Administration (FDA) in 2011 as a first line therapy for melanoma patients with metastatic disease.¹³

While the majority of cancer therapies have focused on the treatment of existing cancer, there have been some successes in the prevention of cancers, in particular those caused by certain viral infections such as the hepatitis B virus (HBV), a risk factor of hepatocellular cancer, and human papillomavirus (HPV), a risk factor of cervical cancer, for which vaccines are currently available. In Taiwan, where universal childhood HBV vaccination was introduced in 1984, there has been more than a 80% decline in liver cancer incidence rates among youth and young adults in the last 30 y.¹⁴ While it is still too early to determine the effects of HPV vaccination on incidence of cervical cancer, results so far are very promising.¹⁵

Despite these successes there still remain some major hurdles to the development of cancer immunotherapies. In many cases, response rates are low, responses are short-lived, side-effects can be severe and treatments can be prohibitively expensive.¹³ In order to overcome this, a number of combinatorial approaches have been proposed including dual-checkpoint inhibitors, checkpoint inhibitors and agonists of co-stimulatory receptors, and checkpoint inhibitors and therapies that enhance tumor cell recognition or immunogenicity.^16,17 Another approach has been to enhance delivery of immune therapies, such as vaccine antigens and/or adjuvants, cytotoxic drugs, monoclonal antibodies, small interfering RNA (siRNA) etc., via the use of nanoparticles (NPs). Over the past 3 decades, there has been considerable advance in the use of NP delivery systems for cancer therapy and several NP-based compounds delivering encapsulated or conjugated cytotoxic drugs targeted to solid tumors have been approved by the FDA. There are multiple types of NPs used including lipid-based particles such as liposomes, lipoplexes and lipid NPs; polymeric NPs and micelles; pH and thermosensitive nanoparticles; and metal and inorganic NPs such as gold NPs.¹⁸ Many of these particles act as a multi-functional drug delivery platform due to their ability to target delivery of tumor antigens and adjuvants,^19-22 enhance immune activation,^23-25 augment the efficacy of cell therapies,^26-28 accommodate multiple therapeutic moieties with different physicochemical properties, modify pharmacokinetics and pharmacodynamics of therapeutic agents,²⁹ and to improve the stability and resistance issues.³⁰

The first generation antitumor nanomedicines, such as DOXIL® and Abraxane®, relied on enhanced permeability and retention (EPR) effects arising due to the unique vascular characteristics of tumor tissue.³¹ However, EPR effects are influenced by the interplay between the NP and tumor microenvironment which may vary considerably among different tumor types.³² Therefore, EPR dependent passive targeting often results in unpredictable clinical outcomes, particularly in metastatic cancer, where tumor cells grow in different vascular beds.³³ Overall, clinical outcome therefore depends not only on the immunomodulatory effect of NPs but also their localization, retention, cell binding, internalization, and toxicity and this is in turn dependent on a number of factors including the NP composition, surface chemistry, biophysical properties, affinity to protein, and the route of administration.³⁴ One approach to help improve the efficacy of NP delivered therapies is through active targeting whereby NPs are associated with ligands selected to target surface molecules or receptors overexpressed on tumor cells.

Here, we review the current NP platform technologies for rational design of targeted next-generation carriers for cancer immunotherapy, ranging from antigen/adjuvant delivery vehicles to target professional antigen-presenting cells, to direct tumor antigen-specific T-lymphocyte-targeting compounds and their combinations thereof.

Immune regulation, tumor microenvironment and tumor evasion mechanisms

Cancer immunotherapy has made considerable progress in recent years largely due to an increase in our understanding of the complex tumor microenvironment (TME). The success of cancer immunotherapy has been dependent on strategies to optimize tumor-specific immune responses while overcoming tumor-evasion mechanisms. The key role that the immune system plays in tumor development is demonstrated by the association of tumor growth with chronic inflammation,³⁵ immune deficiencies such as malfunctions of IFN-γ, perforin, T cells and natural killer (NK) cells,^36,37 and an increased cancer occurrence in immune compromised or immunosuppressive recipients.³⁸ In the early stages of tumor growth, the presence of tumor cells can induce inflammatory signals leading to the recruitment and infiltration of immune cells such as NK cells, neutrophils, innate lymphoid cells, macrophages and DCs. This can lead to the direct killing of tumor cells by NK cells as part of the innate immunosurveillance process,^39,40 and also to the initiation of adaptive immune responses via APCs, in particular DCs, which process and present tumor derived antigens in the context of MHC class I molecules to activate CD8⁺ T cells. The subsequent release of IFN-γ may also impede angiogenesis and prevent tumor progression.⁴¹ In some cases, tumor cells can be completely killed by such antitumor immune response and this may protect us from many early stage cancers. However, in many other cases this innate immunosurveillance is not sufficient and cancer cells do proliferate and form tumors. The TME is composed of a broad variety of cell types in addition to malignant cells, including lymphocytes, tumor associated macrophages, NK cells, DCs, myeloid-derived suppressor cells (MDSCs), the tumor vasculature endothelial cells, as well as fibroblasts, pericytes and sometimes adipocytes.⁴² In addition, osteopontin, galectin-3, transforming growth factor-β and matrix metalloproteinases (MMP) are important secreted proteins closely associated with cancer development. The interaction among these cells, plus proteins, creates an extracellular matrix (ECM) which forms a physical skeleton to support the evolution and spreading of cancers. Moreover, cancer-associated fibroblasts are present in high numbers with an aberrant phenotype and have a potent effect on tumorigenesis and can impede anti-tumor responses. Furthermore, collagen deposition in the TME may also hinder T cell entry and in many patients, the tumor vasculature prevents the trafficking and function of T cells, particularly in those tumors which lack a type 1 interferon signature. The complexity of the TME therefore posts a significant challenge to the entry of adaptive immune cells into tumor sites.

Tumors form for a variety of reasons. Due to genetic instability, tumor cells constantly divide and can become less immunogenic over time so that they can evade immune elimination. Indeed, tumors can stay dormant without mass expansion for long periods of time until vasculature is established.⁴³ The continuous stimulation of innate receptors such as Toll-like receptors (TLRs) by necrotic cancer cells can create a chronic inflammatory condition that can promote immune tolerance, maintain tumor microenvironment, and promote tumor angiogenesis that actually supports tumor progression.⁴⁴ In addition, immunosuppressive cells in the TME such as CD4⁺/CD25⁺/FoxP⁺ regulatory T cells (T_regs) play a crucial role in tumor escape.^45,46 It has been demonstrated that blocking CD25⁺ by anti-CD25 mAb promotes tumor rejection.⁴⁷ Immune suppressive cytokines produced in the TME such as interleukin (IL)-10 and transforming growth factor-β (TGF-β), can facilitate the conversion of CD4⁺ T cells into T_reg in situ, resulting in the direct inhibition of cytotoxic T lymphocyte proliferation and thereby contribute to tumor growth and progression.^48,49 Tumor cells can also recruit other inhibitory immune cells such as alternatively-activated macrophages (M1 and M2) and MDSCs to dampen cytotoxic functions of CTLs and to aid in tumor initiation, angiogenesis and metastasis.⁵⁰ Another mechanism by which tumors evade immune surveillance is by lowering the expression of major histocompatibility complex (MHC) Class I. As a result of this, tumor antigen expression is down regulated and effective killing of tumor by cytotoxic tumor infiltrating lymphocytes fails to occur.⁵¹ Moreover, immunosuppressive enzymes like indoleamine 2,3-dioxygenase (IDO), arginase and nuclear factor kappa-B kinase may act on T cell tolerance leading to tumor cell proliferation.⁵² Tumor cells also over express ligands that bind to inhibitory receptors and dampen T cell function within the TME. For example, programmed death-ligand 1 (PD-L1) is commonly expressed on the surface of macrophages and DCs and binds to programmed cell death protein 1 (PD-1) on activated T cells thereby modulating T cell response and minimizing over activation. However, PD-L1 is commonly overexpressed on tumor cells as a mechanism to evade detection and inhibit the immune response. Our understanding of inhibitory checkpoint molecules such as PD-1 and cytotoxic T lymphocyte associated protein 4 (CTLA-4) has led perhaps to the biggest breakthrough is cancer immunotherapy in recent years with the development of checkpoint inhibitors, which are mAbs targeting these receptors. Indeed, the FDA approved an anti-CTLA4 and an anti-PD1 antibody for the treatment of melanoma in 2011 and 2014, respectively, with subsequent approvals for other cancers whether used alone or in combination.^13,16

Use of NPs for targeted delivery

NPs were initially developed as a protective vehicle for incorporated material, such as antigens, proteins, peptides nucleic acids, etc., thereby allowing the use of lower doses and a more sustained release. However, by altering their biophysical features and through the incorporation of appropriate functional ligands, they can also be used to target distinct cell types such as tumor cells, DCs, T cells or the tumor microenvironment.

These advanced NP systems therefore provide a unique opportunity to develop therapeutic compounds that can more specifically promote tumor cell death and subsequent antitumor immunity.

Influence of biophysical features of NPs on tumor targeting

The uptake of TAAs, carried within NPs, by DCs is largely influenced by physicochemical properties of the NP including size, shape, surface charge and hydrophobicity. DCs preferentially engulf virus-sized particles (20 ∼200 nm) while macrophages (Mφ) preferentially uptake larger particles (0.5 ∼5 µm). For example, following intradermal administration to mice of small core-shell NPs (20, 45 or 100 nm) consisting of poly-ethylene glycol cross-linked to a core of polypropylene sulfide, particles with a size of 20 or 45 nm were retained in lymph nodes for up to 5 d.⁵³ An increase in NP-containing DCs was seen, suggesting these cells trafficked the NPs to the lymph nodes. Interestingly, NPs were internalized by up to 50 % of lymph node DCs without the use of a targeting ligand, indicating the key role of particle size.⁵³ Particle size will influence not only cellular uptake but also distribution, retention and clearance of NPs. Following subcutaneous, intradermal or intramuscular administration, the majority of small NPs (< 20 nm) will preferably drain to blood capillaries and subsequently be eliminated by the organism; NPs between 20 and 100 nm will penetrate the extracellular matrix, enter directly into the lymphatic vessels, travel to the lymphatic nodes where they will be taken up by immune cells, in particular DCs; larger NPs (> 100 nm) will most likely stay at the injection sites until they are captured by peripheral DC and then migrate to lymphatic nodes to initiate immune response.⁵⁴

However, due to their physical characteristics, NPs can also be internalized by scavenger cells, such as Mφ, which have impaired T cell priming capacity compared to DC. Kim et al. showed that, at least with NPs composed of poly-γ-glutamic acid and L-phenylalanine ethyl ester, small NPs having diameter in a range of approximately 30-130 nm were less accumulated into DCs in vitro, compared with larger NPs having diameter in range of 190-360 nm, although DC activation was strong with both types.⁵⁵ This suggests that DC activation occurs not only based on size-dependent uptake of the NPs, but also additional factors such as the interactions between the NPs and DCs, which may also be responsible for the induction of DC maturation.⁵⁵

The endocytosis by DCs of NPs is predominantly by the classic receptor-mediated pathway for NPs less than 150 nm, but by the caveolae-mediated pathway for smaller NPs in the range of 50-80 nm.⁵⁶ Studies found that the uptake of Simian Virus 40 (SV40), which is the most extensively studied ligand for caveolar endocytosis with a particle diameter of 50nm, occurs specifically by the caveolar pathway. Even when an excess of SV40 particles were incubated with cells, still less than 5% of internalized SV40 were found to pass through clathrin-coated pits and vesicles.⁵⁷ In contrast, when negative mutants of caveolin were expressed, internalization or infection was lost.⁵⁸

Since cell membranes are typically negatively charged, they attract positively charged NPs with high affinity. Therefore, while particles having diameters of less than 0.5 µm are usually optimal for DC uptake; larger positively charged particles can also be efficiently taken up.⁵⁹ Once internalized by APCs, cationic NPs are found in the perinuclear area, whereas negatively charged or neutral NPs tend to localize within lysosomes.⁶⁰ Thus, both the surface charge of NPs as well as the charge of the material to be delivered are clearly important considerations for influencing particle selection and their cellular uptake. For example, K-ras, the most common mutated oncogenic protein in solid tumors, has +8 surface charge. Therefore, a biodegradable negative charged polyester-based vector (BCPV) targeting to K-ras was used to effectively transport siRNA into pancreatic cancer cells, leading to apoptosis of tumor cells.⁶¹ In contrast, cationic liposomes composed of 1,2-dioleoyl-3-trimethylammonium-propane (DOTAP) were able to activate mouse bone marrow DCs (BMDCs) in a concentration-dependent manner resulting in the generation of reactive oxygen species (ROS) which led to activation of extracellular signal-regulated kinase (ERK) and p38, cytokine/chemokine production, and expression of the B7 costimulatory molecules CD80 and CD86. However, the presence of higher doses of DOTAP led to cytotoxicity and subsequent cell death,⁶² highlighting the need for appropriate dosing levels when using cationic NPs.

Particle shape is another key parameter that can influence NP biodistribution, cellular uptake, and toxicity.⁶³ For example, non-spherical particles composed of diblock copolymers of poly(ethylene glycol) (PEG) and poly(ethylethylene) or biodegradable poly(ε-caprolactone) present as filamentous micelles,⁶⁴ showed a prolonged circulation time, increased accumulation in tumor tissue with a significantly greater anti-tumor effect than its spherical counterparts when used to deliver paclitaxel to A549 tumor-bearing mice.^64,65 Worm-shaped micelles made of polylactic acid (PLA) and inert block copolymer amphiphiles showed more effective penetration into nanoporous gel (a tissue model) than conventional 100 nm vesicles.⁶⁶

Innate immune activation can also be influenced by texture of microparticles. For example, polystyrene-block-poly(ethylene oxide) (PS-PEO) microparticles having a budding texture could be more readily phagocytosed, and induced a faster neutrophil recruitment to the injection site than smooth particles, leading to stronger IL-1β secretion through activation of the NLRP3 inflammasome.⁶⁷

The biodistribution, retention and clearance of NPs can be influenced by multiple properties including size and shape, charge, composition and presence of ligands. While certain NPs (typically > 400 nm) may accumulate in tumor sites as a result of extravasation from tumor vasculature enhanced by angiogenesis or via various targeting mechanisms,⁶⁸ other NPS will be phagocytosed and cleared from the bloodstream, in particular by the mononuclear phagocyte system (MPS) in the liver and spleen.⁶⁹ A better understanding of the impact that biophysical properties of NPs have on their biodistribution can directly impact their efficacy. Pegylated or STEALTH liposomes have been used for many years to improve circulation time and avoid the MPS. They can circulate for long periods as stable constructs and slowly extravasate into tumors, providing passive targeting to tumor tissue. More recently, mixed-charge gold NPs (AuNPs) have been shown to have a much longer blood half-life and accumulate far less in the liver and spleen than PEG-coated AuNPs which can result in higher accumulation and slower clearance in tumors. A number of other approaches including coating of AuNPs with biocompatible polymers have also been shown to improve the stability of the NPs and the payloads, improve biocompatibility and promote longer systemic circulation.

NPs target DCs

Dendritic cells have been called nature's adjuvant because they are the most potent antigen-presenting cells and are capable of inducing T cell responses in vivo without other adjuvants.⁷⁰ Therefore, DCs are central to the generation of antitumor responses and have been the target of immunotherapy for a number of years. The first approved autologous cellular immunotherapy, sipuleucel-T (Provenge®) for the treatment of asymptomatic or minimally symptomatic metastatic castrate-resistant (hormone-refractory) prostate was approved by the FDA in 2010, based on an approximate 4 month median improvement in overall survival in Phase 3 trials.⁷¹ Despite this success, there remains room for improvement in DC-based therapeutics and direct DC targeting in vivo using NPs may be one potential option.⁷²

Although small NPs (< 500 nm) can be efficiently internalized by local DCs, active targeting of designed NPs through molecular recognition, i.e., ligand-receptor interaction or antibody-antigen recognition, has been shown to be efficient in co-delivering Ags and other stimulatory molecules to specific DC population.^73,74 Particulate delivery systems including virus vectors, whole-cell vaccines, virosomes, immunostimulating complexes (ISCOMs), virus-like particles (VLPs) and other biodegradable nanocarriers can carry specific ligands and thereby increase interaction with target cells as compared to soluble antigens. For example, Cruz et al. investigated three distinct cell-surface receptors expressed on DCs as targets for pegylated poly(lactic-coglycolic acid) (PLGA) NPs: i) CD40, a TNF-α family receptor with known DC activating properties after binding to its specific ligand; ii) DEC-205, a C-type lectin receptor, and iii) CD11c, integrin receptor. These receptors were targeted by means of specific mAbs coupled to the NP, with co-encapsulation of OVA and TLR3/7 ligands. All three mAb-targeted NPs were efficiently internalized by DC in vitro compared to non-targeted NPs.

In addition, all targeted NPs could equally stimulate IL-12 production and induce strong proliferation and IFN-γ production by T cells in vitro. Moreover, all targeted NPs showed higher efficacy than non-targeted NP in stimulating antigen specific CD8⁺ T cell responses after subcutaneous administration to mice.^75-77

It is noteworthy that distinct immune responses can be induced by targeting different DC populations. For example, by using chimeric monoclonal antibodies to target antigen to the 2 major types of DCs found in spleen, it was shown that delivery of antigen to CD8+DEC205+ DCs led to preferential antigen presentation by MHC class I molecules, whereas targeting to CD8–33D1+ DCs led to presentation by MHC class II molecules.⁷⁸ Subsequent studies have used this approach of targeting different DC subpopulations in a number of different animal models. A similar approach may be feasible by using different functional groups to allow targeting to NPs to distinct DC subpopulations.

It is worth noting that while DC targeting appears to be an attractive form of immunotherapy for cancer, DC malfunction and failure of DC maturation often occurs in cancer patients.⁷⁹ Indeed, it is believed that immature or partially matured DCs may induce T cell anergy or T_reg expansion, resulting in antigen-specific tolerance.⁸⁰ The maturation of DCs is associated with increased expression of various surface markers including CD40, CD80, CD83, CD86 and MHC-I/-II and can be induced by inflammatory factors such as LPS, bacterial DNA or inflammatory cytokines such as TNF-α. LPS contamination in gold NPs has been shown to stimulate DC maturation, leading to an increased expression of co-stimulatory and MHC class I and II molecules and production of cytokines.⁸¹ Signal transducer and activator of transcription 3 (STAT3) inhibits DC differentiation and when down-regulated can result in improved DC function.⁸² Poly (ethylene glycol)-b-poly (L-lysine)-b-poly (L-leucine) (PEG-PLL-PLLeu) polypeptide micelles loaded with Poly I:C (PIC, a TLR3 agonist), STAT3 siRNA and OVA antigen could effectively overcome DC dysfunction in a murine B-16 melanoma model by deleting STAT3 gene in situ.⁸³ Similar results have been shown by other groups using NP based delivery of STAT3 siRNA alone or as part of an anti-cancer vaccine.^84,85

While the focus of cancer immunotherapy is on activation of DCs, for other therapeutic areas down-regulation of DCs is required. For example, in a murine autoimmune encephalomyelitis model, the administration of polystyrene or biodegradable PLGA microparticles (500-nm diameter) bearing encephalitogenic peptides induced long-term T-cell tolerance, highlighting the potential use of microparticles in autoimmune disease.⁸⁶ Therefore, it is important to note that DC targeting is not as simple as adding a DC targeting ligand to NPs, and that the polarization of DCs, either stimulation or suppressive, is essential to the success of immunotherapy. Thus, the challenge to achieve a desired immune outcome, particularly in an immune compromised population, remains. For cancer immunotherapy, the key to successful DC targeting is to select the appropriate DC surface target and the appropriate payload, whether be it antigen and adjuvant to elicit a protective and long-lasting immune response or mAbs or siRNA to downregulate inhibitory mechanisms.

NPs target tumor microenvironment

Tumor-associated macrophages (TAM), MDSCs and T_regs are a heterogeneous group of cells that are major components of the TME and play a critical role in enhancing tumor cell invasion and metastasis, promoting angiogenesis and extracellular matrix remodeling, while inhibiting the anti-tumor immune surveillance. Elimination or reprogramming of the immune suppressive TME is one of the major current challenges in immunotherapy of cancer.⁵⁷ Perhaps the most successful cancer immunotherapy to date has been immune checkpoint inhibition, which targets regulatory pathways in T cells to enhance antitumor immune responses. Four new immune checkpoint agents have now been approved by the US FDA for the treatment of melanoma. A mAb against CTLA-4 (ipilimumab) was approved in 2011, and 2 antibodies against PD-1 (pembrolizumab and nivolumab) were approved in 2014.⁸⁷ The therapeutic benefit of checkpoint inhibitors seems to be most optimal in patients with a pre-existing T-cell response against their tumor.⁸⁸ Li et al., used NPs, composed of poly(ethylene glycol)-block-poly(D,L-lactide) (PEG–PLA) that encapsulated CTLA-4 siRNA (siCTLA-4) and showed direct T cell activation both in vitro and in vivo. These cationic lipid-assisted PEG–PLA-based NPs efficiently delivered siRNA into T cells in vitro and significantly increased the percentage of anti-tumor CD8⁺ T cells, while also decreasing the ratio of CD4⁺ FOXP3⁺ T_regs among tumor infiltrating lymphocytes (TILs), resulting in inhibition of tumor growth and prolonged survival time.⁸⁹ Other approaches include the use of Papaya mosaic virus NPs (PapMV) as both an immunostimulatory molecule and as a vaccine platform to activate the innate immune response in an IFN-α-dependent manner. A synergistic effect of PapMV treatment in combination with PD-1 blockade was demonstrated in a murine melanoma model in triggering CD8⁺ T cells responses specific for the tumor antigens gp100 and TRP2, compared to anti-PD-1 treatment which when used alone, did not significantly increase tumor-specific T-cell numbers.⁹⁰ An increased therapeutic effect of anti-PD-1 blockade has been shown in a murine melanoma model when used in combination with immunostimulatory RNA, supporting the hypothesis that a synergistic therapeutic effect is likely the result of increased immune-cell infiltration and tumor-specific T-cell priming.⁹¹

The tumor vasculature is abnormal in terms of its heterogeneous and chaotic branching structure, uneven vessel lumen, and leakiness. These features result in increased interstitial fluid pressure and uneven blood flow, oxygenation, nutrient and drug distribution which in turn increases hypoxia and promotes metastasis mediated via vascular endothelial growth factor (VEGF).⁴² Although tumor vasculature is highly accessible and can therefore enhance delivery of drugs to tumors, most therapeutic agents, especially chemotherapy, have no intrinsic affinity for tumor vessels, and can cause severe drug resistance and systemic side effects. Vascular targeting using NPs is another inviting strategy for cancer therapy requiring effective margination to bring drugs/vaccines etc. to the tumor site. A number of TME targeting NP formulations have been designed for the prevention of angiogenesis, tumor growth and metastasis, for example by activating endogenous angiogenesis inhibitors via phage display,⁵⁸ silencing proangiogenic factors via delivery of CXCR4 antagonists using lipid-based NP,⁹² or by directly destroying endothelial cells via active Integrin targeting.⁹³

Tumors possess a dense extracellular matrix (ECM) composed of fibrous proteins such as collagen and elastin, as well as a highly viscous polysaccharide-containing fluid.⁹⁴ NPs have been developed to target various ECM components thereby remodelling ECM dynamics and destroying the matrix needed for tumor growth. Several studies have utilized hyaluronan (HA)-conjugated nanocarriers to target CD44, an adhesion/homing molecule, which acts as receptor for glycosaminoglycan hyaluronan, one of the major components of the tumor extracellular matrix.⁹⁵ HA-conjugated NPs designed to target CD44 in the ECM have been used to deliver anti-cancer drugs (e.g., epirubicin, doxorubicin, paclitaxel, and mitomycin c), as well as siRNA to CD44 over-expressing cells.⁹⁵ Similarly, Tan et al. used gadolinium-containing dendrimers conjugated with a peptide (CLT-1), which bind specifically to fibronectin–fibrin complexes in the tumor ECM, to target breast tumor support structure and showed the potential of this approach for magnetic resonance cancer molecular imaging.⁹⁶ Several studies have used NPs coated with HA to target CD44 in combination with other targeting moieties as dual-functional drug carriers. For example, PLGA NPs conjugated with HA and anti-HER2/neu peptide (AHNP) have been used to deliver SN38, the active metabolite of the topoisomerase 1 inhibitor Irinotecan, and was shown to inhibit tumor growth in murine models to a greater extent than Irinotecan alone.⁹⁷ Indeed, even complex multi-functional drug carriers have now been proposed. For example, a multifunctional liposomal nanocarrier has been tested in vitro that combined: i) a nanoscale size for passive tumor targeting, ii) an anti-nucleosome mAb (mAb 2C5) for active targeting, iii) PEG tails to prevent non-specific interactions and prolong particle circulation time, iv) a matrix metalloprotease 2 (MMP2)-sensitive bond between PEG and lipid that is cleaved in the tumor by the highly expressed extracellular MMP2 to remove the PEG chains and v) a cell-penetrating peptide (TATp) that enhances intracellular delivery of the system after long-chain PEG removal and exposure of the previously hidden surface-attached TATp. This complex NP has been shown to enhance the targetability and internalization of nanocarriers in cancer cells.^98,99

T cell targeting

Adoptive T cell therapy (ACT) is based on the isolation and ex vivo expansion of tumor specific T cells from fresh patient biopsy specimens typically in the presence of tumor antigen and IL-2. The tumor specific T cells are then infused into the same patient where they can target and kill tumor cells. A number of different forms of adoptive T cell therapy have been used for cancer treatment; culturing tumor infiltrating lymphocytes (TILs), isolating and expanding one particular T cell or clone, and using engineering T cells to recognize and attack tumors. While initial studies with ACT were limited by weak immune responses, attributed to the decline in viability and function of transplanted cells, insufficient expansion of transferred T cells, inefficient trafficking to tumor regions and dose-limiting toxicities,^100,101 considerable improvement in this technology have been made in recent years including the use of chimeric antigen receptors (CAR)-T cell technology whereby T cells are genetically engineered to express tumor-specific antigen receptors to enable targeting to tumor cells.^102-104

One promising approach to improve the efficacy of ACT is by combining CAR-T with blockade of co-inhibitory immune checkpoints such as anti-PD-1, anti-PD-L1 or anti-CTLA-4 monoclonal antibodies.^105,106,107 Another approach to enhance cell therapy is to modify the surfaces of therapeutic cells with adjuvant drug-loaded NPs. For example, liposome particles with encapsulated cytokines (IL-15, IL-21) or drugs (glycogen synthase kinase-3 β inhibitor TWS119) were coupled to the surface of living T cells via thiol-reactive maleimide headgroups on the lipid bilayer surface of the particles. These surface decorated NPs were non-toxic to their carrier T-cells, did not interfere with intrinsic cell function or migration patterns, yet gave their carrier cells substantially enhanced function using low drug doses that were ineffective when used alone by traditional systemic routes. Of note, after crossing the endothelial barrier, 83% (± 3%) of their original NP cargo was still physically attached to the carrier CD8⁺ T-cells.²⁷ Several other studies have demonstrated that T cells may serve as efficient delivery vehicles for the treatment of cancer. For example, gold NPs (AuNP) have been extensively studied for photothermal cancer therapy because AuNPs can generate heat upon near-infrared irradiation. When human T cells were loaded with 45 nm AuNPs and administered in a human tumor xenograft mouse model, 4-fold higher levels of AuNP were observed in tumors 24 hr post-injection compared to injection of free PEGylated AuNPs and overall NP biodistribution was also changed.¹⁰⁸ The cytotoxic antibiotic doxorubicin has also been delivered using Target MAG-doxorubicin NPs loaded onto T cells.¹⁰⁹

A selective drug carrier system involving antibody-coupled NPs is also an attractive drug-targeting system. Not only would they allow incorporation of cytotoxic antitumor drugs into the NP matrix, but they can also be routed by attachment of antibodies to a specific tumor cell type. This will allow a higher drug carrier capacity combined with improved specificity of drug targeting. Antibodies specific for the CD3 antigen on T cells were conjugated to the surface of gelatin NPs and were shown to be internalized into CD3-positive human T-cell leukemia cells and primary T lymphocytes. An uptake rate of ∼84% into T-cell leukemia cells was observed demonstrating the feasibility of receptor-mediated cellular uptake into a given target cell type.¹¹⁰

Hybrid NPs

Hybrid NPs are typically composed of a synthetic polymer core with a lipoid shell. Originally developed to overcome the problem of delivering water-soluble drugs in hydrophobic lipid particles, they can be composed of a wide selection of biocompatible polymers and lipids, are easy to synthesize and possess the ability to deliver multiple therapeutic agents to the same target cell including both hydrophobic and hydrophilic molecules.¹¹¹ Co-delivery approaches include the delivery of multiple chemotherapeutic drugs, chemotherapeutic drugs and nucleic acids (e.g., DNA, RNAi, siRNA), chemotherapeutic drugs and mAbs, chemotherapeutic and epigenetic drugs.¹¹² In addition, as with other types of NPs, they can be surface coated with targeting moieties, to enable targeted drug delivery to tumor cells.

One of the tumor evasion mechanisms which tumor cells implement is the secretion of transforming growth factor-β (TGF-β), which restricts local tumor immune responses, limiting conventional cytokine therapy. In order to overcome this, a core-shell delivery platform was established combining features of both liposomal and solid cyclodextrin systems. This nanoscale liposomal polymeric gel was composed of drug-complexed cyclodextrins and cytokine-encapsulating biodegradable polymers, which enables sustained and simultaneous release of a TGF-β inhibitor and hydrophilic IL-2 to the TME, resulting in significantly delayed tumor growth and increased survival in a murine B16 melanoma model.¹¹³ Such hybrid multifunctional drug-delivery systems offer the potential for combinatorial delivery and synergistic action in cancer therapy.

STAT3 inhibits DC differentiation and when downregulated can result in improved DC function.⁸² It therefore imposes a major limitation for DC-based cancer therapies and becomes an attractive target for immunotherapeutic applications.¹¹⁴ Activation of STAT3 in DCs can inhibit the expression of immunostimulatory molecules triggered by TLR ligands, such as cytosine-phosphate guanine oligonucleotides (CpG ODN), that regulate T cell activation.¹¹⁵ Kim et al., designed an immunomodulatory hybrid nanoconjugate system aimed at concomitantly silencing STAT3 by siRNA and activating TLR9 by CpG ODN. Polymeric PLGA nanocomposites containing quantum dots (QDs) conjugated to CpG ODNs and STAT3 siRNA using a cleavable disulfide linker were able to simultaneously deliver STAT3 siRNA and CpG ODN to the target immune cells thereby inhibiting STAT3 and activating DCs by CpG ODNs giving synergistic antitumor effects in a murine B16 melanoma model.¹¹⁵ Another similar approach was developed using PLGA NPs containing both siRNA for the knock-down of immune-suppressor gene STAT3 of DCs and imiquimod for the activation of DCs through TLR 7. Again, these NPs were efficiently taken up by DCs and various cytokines were expressed in matured DCs. Immunization of mice with these NP-treated DCs induced OVA-specific CTL activity in the EG7-OVA tumor model and tumor growth was efficiently delayed.¹¹⁶

In an attempt to mimic the mechanism whereby immune effector cells crosslink surface receptors on target tumor cells and induce apoptosis and to circumvent the use of potentially toxic chemotherapeutic drugs, Chu et al. designed a biomimetic material platform composed of self-assembling hybrid nanoconjugates. This system comprised an anti-CD20 Fab’ antibody fragment, a pair of complementary phosphorodiamidate morpholino oligomers (MORF1 and MORF2), and a linear polymer (P) of N-(2-hydroxypropyl)methacrylamide (HPMA). Malignant CD20⁺ B-cells were first treated with anti-CD20 Fab’ conjugated to MORF1, resulting in the decoration of the cell surfaces with MORF1. Further addition of copolymer grafted with multiple copies of MORF2 resulted in MORF1 and MORF2 hybridization at the cell surface with concomitant CD20 cross-linking, which triggered cell apoptosis, resulting in improved long term survival in a murine model of human non-Hodgkin's lymphoma.¹¹⁷ Although CD20 mAbs (e.g., obinutuzumab) can also induce direct apoptosis, this approach could also potentially reduce adverse effects.

Combination therapy

The strength of combinatorial approaches to cancer immunotherapy was demonstrated by the FDA's decision to grant accelerated approval to the combination of the PD-1 inhibitor nivolumab and the CTLA-4 inhibitor ipilimumab to treat advanced melanoma cancer based on strong Phase 3 results.^118,119 In addition, a number of clinical studies have shown promising data with combinations of checkpoint inhibitors and agonists of costimulatory receptors such as CD70, 4-1BB (CD137/TNF superfamily 9), OX40 (CD134), combinations of checkpoint inhibitors and therapies that promote tumor cell recognition by T cells such as tumor vaccines and IDO inhibitors, and combinations of checkpoint inhibitors and therapies that enhance tumor immunogenicity such as inducers of ICD, oncolytic viruses and epigenetic therapy,¹⁶ as well as the combination of different immunotherapies with conventional chemotherapy.¹²⁰ The use of NP technology may help some of these combinatorial approached achieve their full potential.

In an attempt to increase the efficacy of chemotherapeutic approaches without causing systemic toxicity due to high dose levels required, combinatory chemo-immunotherapeutic approaches have been evaluated. Sequential delivery of paclitaxel (PTX) and TLR9 agonists (CpG ODNs) plus siRNA to silence IL-10 using 2 types of PLGA NPs has shown promising results in murine tumor models. It is hypothesized that the primary injection of PTX caused tumor cell death and the subsequent release of large amounts of antigenic debris resulted in vaccination in situ. The released tumor-associated antigen was thought to subsequently be internalized by tumor-recruited bone marrow-derived dendritic cells (BMDCs). The secondary injection with immunomodulating CpG ODN-encapsulated in PLGA NPs and IL-10 siRNA-encapsulated PLGA NPs triggered the activation and migration of BMDCs to the tumor-draining lymph nodes. As a result, tumor growth was inhibited and animal survival rate was increased.¹²¹

To date, the main limitation of immunotherapy is the lack of reduction of tumor load in clinical patients. Additionally, tumor progression and metastasis usually occurs, frequently after a slightly extended period of remission. Targeted therapy appears to obtain a higher response rate in a greater number of patients. Nevertheless, despite a lower overall response rate, responses to immunotherapy, tend to be more durable than those seen with targeted therapy.^122-125 Therefore, it is attractive to combine immunotherapy with targeted therapy to try to achieve sustained anti-tumor responses with a long-lasting tumor remission. A number of tumor antigen targets have been identified, for example, overexpressed proteins on tumor cells (e.g., human epidermal growth factor receptor 2 protein in breast cancer (HER-2)), mutant proteins that drive cancer progression (e.g., an altered form of BRAF in melanomas) and fusion genes that produce fusion proteins (e.g., BCR-ABL fusion protein in leukemia cells).¹²³ The folate receptor has also been recognized as a tumor marker as it is overly overexpressed on the surface of different cancers making it a good target for folate nanoconjugates.¹²⁶ For example, various folic acid functionalized polyethylenimine (PEI) polymers were evaluated to deliver PD-L1 siRNA to SKOV-3-Luc epithelial ovarian cancer cells. Folic acid modified PEI polymers increased siRNA uptake into SKOV-3-luc cells and decreased non-specific uptake into monocytes resulting in enhanced T cell killing.¹²⁷ In many cases, although tumor cells exhibit abnormal antigens due to their genetic defects, the immune responses to these tumor antigens is not strong because they are recognized as self. One way to overcome this is by the use of effective delivery sytems. For example, archaeosomes, which are liposome like vesicles made with polar membrane lipids of Archaea, possess receptor-targeting sugar headgroups which can facilitate targeted and more efficient uptake of archaeosome vesicles by macrophages.¹²⁸ Archaeosome have been shown to overcome immunologic tolerance to melanoma self antigens and effectively evoke CD8⁺ T cells.¹²⁹ The use of mAbs, Ab fragments (Fab, single chain variable fragment, scFv) and single domain antibodies, (sdAbs) can also act as highly specific probes when they are attached to NPs to aid in targeted delivery of various antitumor cytotoxic agents.^130,131 Therefore, overall combinatorial approaches including antibody targeting, chemotherapy and immunotherapy are growing in popularity due to their ability to combine specific targeting to cancer cells and potent antitumor effects.^130,131

Conclusion

The goal of the current review was to highlight the potential application of NPs in the field of immunotherapy. Great efforts have been applied to the development of novel vaccine adjuvants and nanoparticulate delivery systems using our improved knowledge of cancerous cellular dynamics and the interactive mechanism between NPs and tumors. Current nanotechnology allows for the simultaneous delivery of synergistic drug combinations and the engagement of the patient's innate and adaptive immune systems to combat cancer. By rationally designing NPs based on active targeting of immune cells, solid tumors and the tumor microenvironment, an improved efficacy can be achieved. The application of nanotechnology to cancer immunotherapy has already produced some exciting results in controlling tumor growth and holds even greater promise for cancer patients in the future.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.