Nanoscale artificial antigen presenting cells for cancer immunotherapy

Abstract

Exciting developments in cancer nanomedicine include the engineering of nanocarriers to deliver drugs locally to tumors, increasing efficacy and reducing off-target toxicity associated with chemotherapies. Despite nanocarrier advances, metastatic cancer remains challenging to treat due to barriers that prevent nanoparticles from gaining access to remote, dispersed, and poorly vascularized metastatic tumors. Instead of relying on nanoparticles to directly destroy every tumor cell, immunotherapeutic approaches target immune cells to train them to recognize and destroy tumor cells, which, due to the amplification and specificity of an adaptive immune response, may be a more effective approach to treating metastatic cancer. One novel technology for cancer immunotherapy is the artificial antigen presenting cell (aAPC), a micro- or nanoparticle-based system that mimics an antigen presenting cell by presenting important signal proteins to T cells to activate them against cancer. Signal 1 molecules target the T cell receptor and facilitate antigen recognition by T cells, signal 2 molecules provide costimulation essential for T cell activation, and signal 3 consists of secreted cues that further stimulate T cells. Classic microscale aAPCs present signal 1 and 2 molecules on their surface, and biodegradable polymeric aAPCs offer the additional capability of releasing signal 3 cytokines and costimulatory molecules that modulate the T cell response. Although particles of approximately 5–10 microns in diameter may be considered the optimal size of an aAPC for ex vivo cellular expansion, nanoscale aAPCs have demonstrated superior in vivo pharmacokinetic properties and are more suitable for systemic injection. As sufficient surface contact between T cells and aAPCs is essential for activation, nano-aAPCs with microscale contact surface areas have been created through engineering approaches such as shape manipulation and nanoparticle clustering. These design strategies have demonstrated greatly enhanced efficacy of nano-aAPCs, endowing nano-aAPCs with the potential to be among the next generation of cancer nanomedicines.

Graphical Abstract

Introduction

Recent decades have seen major advances in the use of nanomedicine to treat cancer. Cancer nanomedicine is the application of nanoscale tools, especially nanoscale particulates, to diagnose, treat, and prevent the disease (1). Current cancer nanomedicine therapeutics utilize nanoparticles engineered to deliver drug cargos locally to tumor sites, reducing negative systemic side effects, such as those associated with chemotherapies (2). These nanocarriers can encapsulate and deliver poorly soluble drugs, stabilize drugs, and slow their degradation and elimination (3). These vectors also take advantage of the enhanced permeation and retention (EPR) of nanoparticles in the leaky vasculature associated with tumors, allowing for passive tumor targeting and a decrease in off-target toxicity (2). In addition to passive targeting, nanoparticles can be actively targeted to tumors through the conjugation of ligands including monoclonal antibodies, peptides, antibody fragments, or small molecules that bind to specific markers on the surface of tumor cells (2, 4, 5). Nanocarriers can be fabricated from various biomaterials, including ones that are lipid-based (3, 6), albumin-bound (7), inorganic (2), or polymeric (8–12), the latter of which are advantageous because they can be constructed from biocompatible, biodegradable materials that encapsulate and release drugs in a controlled manner.

Despite nanocarrier advances, metastatic cancer is still considered largely incurable and is responsible for the vast majority of cancer deaths. Metastases are often poorly vascularized compared to primary tumors, and cannot be effectively accessed by nanoparticles via the EPR effect (13). Additionally, metastases can present in multiple organ environments, each with a unique set of challenges that must be considered when designing particles to target and penetrate the tumor (13, 14). To potentially cure metastatic cancer and prevent recurrence, a nanocarrier therapy must overcome these barriers and deliver its drug cargo to kill every tumor cell. Immunotherapies may be a more effective strategy than traditional nanomedicines because they harness the power of the host’s immune system to fight the tumor. Instead of relying on nanoparticles to destroy every tumor cell, immunotherapeutic approaches activate the adaptive immune system against tumor-specific antigens, leading immune cells to mount an amplified response to recognize and destroy tumor cells (15). One promising technology for cancer immunotherapy is the artificial antigen presenting cell (aAPC), a micro- or nanoparticle-based biomimetic system that functions like a natural antigen presenting cell (APC) to activate the adaptive immune system against cancer. Through their ease of production and their ability to efficiently activate and expand antigen-specific T cell populations, aAPCs show exciting potential as an “off-the-shelf” cancer immunotherapy (16).

The aAPC: A Biomimetic Platform for Cancer Immunotherapy

aAPCs are synthetic constructs that mimic the natural process of T cell activation by biological APCs through the presentation of essential T cell-stimulating signal proteins conjugated to their surface. When aAPCs interact with T cells, these signal proteins activate the T cells and modulate their response. Cellular aAPC platforms to expand antigen-specific CD8+ T cells for adoptive T cell therapies are under clinical investigations to treat a variety of cancers including advanced melanoma (17), but scale-up is expensive and labor-intensive. Acellular aAPC platforms are more easily produced, can be stored as an off-the-shelf therapy, and can be readily adapted to accommodate a variety of antigen specificities and surface ligands. As synthetic systems, they also eliminate the risk of tumorgenicity associated with cellular aAPCs and are unaffected by the immunosuppressive tumor microenvironment (18).

Signals 1, 2, and 3: Antigen presentation, Costimulation, and Cytokine Release

Classic aAPCs present signal 1 and 2 molecules on their surface that are critical for T cell activation (Fig. 1). Signal 1 is used in APC/T cell recognition and provides antigen-specificity. The majority of nonspecific aAPCs bear an antibody that stimulates the CD3 T cell co-receptor, which associates with the TCR complex to produce an activation signal within the T cell. Nonspecific targeting with anti-CD3 is often employed in aAPC systems, but antigen specificity is necessary in a therapeutic setting for efficient expansion of effector T cells against tumor-specific antigens (19). Antigen-specific signal 1 can be provided by MHC single chain complexes, but MHC multimers offer more efficient T cell activation and peptide-loading flexibility (19, 20). Signal 1 density on the aAPC surface affects T cell activation. Matic et al. demonstrated that an anti-CD3 density of 1000 proteins/um² increased T cell proliferation and IL-2 production compared to 120 and 60 proteins/um² (21).

Figure 1.

Strategies for aAPC design. A) A classic micro-aAPC with surface-bound signal 1 (anti-CD3, MHC multimer, etc.) and signal 2 (anti-CD28, anti-4-1BB, etc.) molecules. B) Newer aAPC systems are constructed from biodegradable materials capable of releasing costimulatory and signal 3 cytokines (IL-12, IFN-α/β, etc.). C) Nano-aAPCs are less efficient T cell activators, but may outperform micro-aAPCs in vivo due to their transport properties. D) Recent findings suggest that ellipsoidal nano-aAPCs activate T cells more efficiently compared to spherical nano-aAPCs due to increased contact surface area. E) Magnetic field-induced clustering of paramagnetic nano-aAPCs has been shown to increase T cell induction and expansion.

Signal 2 is a costimulatory signal required for T cell activation. Most aAPC platforms (Fig. 1A) utilize an antibody against CD28 as a costimulatory molecule, although 4-1BB can also stimulate antigen-specific T cells in the absence of CD28 costimulation (22, 23). Interestingly, a 75:25 blend of anti-4-1BB and anti-CD28 on the surface of polystyrene micro-aAPCs also presenting an HLA MHC Class I molecule as Signal 1 led to a five-fold increase of antigen-specific CD8+ T cells compared to each individual signal (24).

In addition to the type of costimulatory ligand, the strength and density of costimulation affect the nature of cell activation (25). Stimulation of CD4+ cells with anti-CD28, which has a greater affinity for the CD28 receptor than natural costimulatory ligands B7-1 and B7-2, was shown to contribute to a Th1 phenotype (26). Meanwhile, B7-1 costimulation led to a Th2 response in CD4+ T cells (26) and was insufficient to sustain a CD8+ T cell response (27). CD8+ cells were shown to require a higher density of costimulatory molecules than CD4+ cells to achieve the same level of activation (27). In a liposomal aAPC, a density of 2000 molecules/um² was found to be optimal for stimulation with a 100:1 liposome to target cell ratio (28).

Although not necessary for activation, secreted cytokines provide a third signal that influences T cell effector phenotype (29). Biodegradable polymeric aAPCs are capable of releasing signal 3 cytokines and costimulatory molecules (Fig. 1B). Steenblock et al. showed that release of IL-2 from PLGA aAPCs with stable surface ligand presentation resulted in a 3–4-fold increase in T cell expansion compared to an equivalent amount of exogenous IL-2, demonstrating the enhancement in effective concentration caused by local delivery and close proximity between a drug releasing particle and a cell (30). These aAPC systems have been reproduced and shown to be effective in multiple studies (31, 32). Zhang et al. demonstrated co-release of IL-2 and anti-CTLA-4 from PLGA micro-aAPCs presenting surface-bound MHC dimer and anti-CD28 (33). These aAPCs significantly slowed tumor growth and enhanced survival in a murine melanoma model over a combination of aAPCs releasing each signal individually, and non-releasing aAPCs with IL-2 and anti-CLTA-4 injected systemically (33).

Micro-aAPCs for Cancer Immunotherapy

aAPC particles with a diameter of 5–10 microns, approximately the same size as biological APCs, have been shown to be most efficient at T cell stimulation for tumor immunotherapy (34). In one study, Mescher found that particles at least 4–5 microns in diameter, a biomimetic length scale, were necessary to activate CD8+ T cells. Equivalent activation could not be reached with increased doses of smaller particles (35). Similarly, Steenblock et al. demonstrated that micro-aAPCs induced a 3-fold increase in CD8+ T cell IL-2 production compared to nano-aAPCs bearing the same protein dose (34). Liposomal (36, 37), paramagnetic (38), and polymeric materials (24, 34, 39, 40) presenting a variety of signal 1 and 2 molecules have all been used to successfully expand functional CD8+ T cells in vitro. In one example, Lu et al. utilized polystyrene beads coated with an MHC class I dimer, anti-CD28, 4-1BBL, and CD83 molecules to expand CD8+ T cells ex vivo for adoptive transfer into murine melanoma and pulmonary metastasis models. Adoptively transferred antigen-specific CD8+ cells led to complete regression of established B16 melanoma tumors in a model and significant tumor reduction in the murine metastasis model (41). In an early example of aAPC in vivo efficacy, Shalaby et al. showed that a combination of poly-glycolic acid (PGA) microparticles with surface bound anti-CD3 and anti-CD28 and PGA particles releasing granulocyte macrophage colony-stimulating factor (GM-CSF) from their surface prevented tumor implantation in 100% of mice, as well as complete tumor regression in 57% of mice with established fibrosarcoma tumors (40). Shen et al. demonstrated that polystyrene aAPCs bearing MHC class 1 tetramer and a combination of anti-CD28 and anti-4-1BB on their surface inhibited tumor growth and delayed tumor progression in mice with established B16 melanoma tumors (42). More recently, biodegradable PLGA aAPCs administered in vivo have been shown to prevent implantation, reduce tumor burden, and prolong survival in murine melanoma models (33, 43, 44).

Nano-aAPCs for Cancer Immunotherapy

Microparticulate, and to a lesser extent, nanoparticulate systems (45) have successfully expanded and activated T cells in vitro. However, nanoscale aAPCs (Fig. 1C) may be better suited for in vivo administration than micro-scale aAPCs, due to improved transport properties and a reduced risk of embolism (18). Several nano-aAPC systems have outperformed micro-aAPCs at stimulating antigen-specific CD8+ cells in vivo, their success likely linked to superior biodistribution and drainage properties (45, 46). Despite these successes, nanoscale aAPCs have a much smaller surface area for contact with T cells and are recognized as generally less efficient T cell activators compared to microscale aAPCs. Due to the fact that sufficient contact between the T cell and aAPC is essential for activation, nano-aAPCs with microscale contact surface areas have been created through strategies including shape manipulation (Fig. 1D) and nanoparticle clustering (Fig. 1E).

Anisotropic Particle Shape

Nonspherical, anisotropic nanoparticles can possess a micron length scale radius of curvature on a long axis to approximate the interfacial geometry of micro-aAPCs (Fig. 2A), and display enhanced pharmacokinetic properties compared to spherical particles (47, 48). Ellipsoidal particles demonstrated reduced non-specific uptake (49), increased specific uptake, and enhanced cancer cell killing (50). Compared to spherical and flattened disc-shaped particles, ellipsoidal particles showed the most efficient particle attachment and lowest in vitro internalization rates (51). Sunshine et al. showed that ellipsoidal micro-aAPCs had more frequent interactions with T cells and greater contact area of interaction than corresponding spherical aAPCs and the ellipsoidal aAPCs also led to improved anti-cancer T cell activity in vivo in a B16 melanoma mouse model compared to spherical aAPCs (43). Using the same materials and film stretching method, Meyer et al. created ellipsoidal PLGA nano-aAPCs presenting an MHC Class 1 Ig dimer signal 1 protein and anti-CD28 as signal 2 (48). In vitro, 14% of ellipsoidal nano-aAPCs were phagocytosed by macrophages compared to 78% of spherical nano-aAPCs. When administered intravenously, higher concentrations of ellipsoidal particles remained in the bloodstream for an hour post-injection and were better dispersed (Fig. 2B). Following intravenous administration with adoptively transferred PMEL CD8+ cells, ellipsoidal nano-aAPCs led to 2-fold greater CD8+ T cell expansion in a B16 melanoma model (Fig. 2C). Overall, ellipsoidal nano-aAPCs demonstrated superior immunogenicity and pharmacokinetic properties in vivo compared to spherical nanoparticles (48).

Figure 2.

Ellipsoidal and magnetically clustered nano-aAPC enhance antigen-specific T cell activation and expansion in vivo. A) Ellipsoidal nanoparticles were created by mechanical stretching. (Scale bar is 500 nm). B) Fluorescent ellipsoidal nano-aAPCs circulate with a longer half-life in the bloodstream compared to spherical nano-APCs. C) Ellipsoidal nano-aAPCs led to a statistically significant increase in the percentage of antigen-specific T cells in vivo compared to spherical nano-aAPCs. Reproduced with permission from Small (48). D) Schematic of magnetic field-induced clustering of TCRs bound to paramagnetic nanoparticles. E) T cells stimulated with nano-aAPCs in a magnetic field and adoptively transferred were present at higher frequencies seven days post-injection compared to controls. F) Treatment with magnetically enhanced T cells led to increased survival when T cells and particles were adoptively transferred on Day 10. Reproduced with permission from ACS Nano <http://pubs.acs.org/doi/10.1021/nn405520d> (52), further permissions related to the material excerpted should be directed to the ACS.

Magnetic Field-Induced Nanoparticle Clustering

Another method to increase the surface contact area between nano-aAPCs and T cells and strengthen activation is through magnetic-field induced nanoparticle clustering (Fig. 2D) (52). Application of an external magnetic field has been shown to induce magnetization and aggregation of superparamagnetic particles and resultant clustering of nanoparticle-bound surface receptors on target cells (53, 54). When T cells interact with APCs, MHC-TCR complexes aggregate at the contact site to form an immunological synapse, which strengthens signaling and subsequent T cell activation. aAPCs with pre-clustered MHC complexes on their surface improved T cell activation compared to those with a uniformly distributed MHC complexes (37). Perica et al. utilized iron-dextran nanoparticles directly conjugated with surface-bound MHC Ig dimer signal 1 and anti-CD28 signal 2 to generate reproducible magnetic aAPCs that have also been used in other studies (45). They demonstrated that nano-clustering in response to an applied magnetic field led to a two-fold increase in TCR/CD3 aggregate size and a two-fold decrease in the number of TCR clusters per T cell compared to a control group with no applied magnetic field (52). The magnetically clustered nano-aAPCs induced 15-fold antigen-specific expansion of CD8+ T cells compared to non-clustered aAPCs (Fig. 2E). Adoptive transfer of magnetically enhanced T cells in mice expressing B16 melanoma tumors significantly reduced tumor burden, prolonged survival, and led to complete tumor regression in 50% of treated mice compared with no treatment or non-enhanced T cell controls (52) (Fig. 2F).

Conclusions and Future Perspective

Despite ongoing nanocarrier development, there are still many challenges associated with the local delivery of nanotherapeutics to treat cancer, particularly in advanced stages. aAPCs are a promising technology that can allow a host to generate an immune response to fight cancer systemically. Micro, and to a lesser extent, nano-aAPCs have shown success in expanding antigen-specific T cell populations ex vivo and successful adoptive transfer into murine cancer models. Nano-aAPCs have the potential for antigen-specific T cell stimulation in vivo and could become a translatable, off-the-shelf cancer therapy without the need for adoptive T cell transfer. In the design of future nano-aAPCs, special consideration must be given to the type, density, organization, and mobility of signals presented on the surface, as well as the encapsulation and controlled release of cytokines from the particles in vivo. Engineering a biomimetic length scale for the aAPC-T cell contact interface is particularly important when designing aAPCs on the nanoscale. The next generation of nano-aAPCs may utilize strategies such as manipulation of physical shape and aspect ratio and application of external triggers to facilitate more accurate approximation of interfacial geometry between biological APCs and T cells. Besides length-scale considerations, other physical and biological properties of the immunological synapse must be optimized in nano-aAPC systems. In addition to particle shape and aspect ratio, parameters such as particle rigidity may be tuned to in future designs to exercise more control over aAPC-T cell interactions. The biological nature of the immunological synapse can be recapitulated through surface engineering strategies such as patterning signal proteins in precise spatial arrangements, or coating aAPCs with supported lipid bilayers to allow signal protein and receptor reorganization in response to T cell contact. Future nano-aAPC systems may benefit from a combination of optimized physical and biological parameters that enhance their biomimicry and efficacy. The use of nano-aAPCs as a biomimetic platform for cancer immunotherapy has strong potential to contribute to the advancement of cancer nanomedicine.

HIGHLIGHTS.

• Artificial antigen presenting cells (aAPC) are particle-based systems that mimic an antigen presenting cell by presenting signal 1 and 2 molecules on their surface to activate T cells.

• Nanoscale aAPCs can have desirable pharmacokinetic properties that make them suitable for systemic injection.

• A micron sized length scale of contact between T cells and aAPCs is desirable for T cell activation.

• Nano-aAPCs with microscale contact surface areas have been created through engineering approaches such as shape manipulation and nanoparticle clustering.

• The use of nano-aAPCs as a biomimetic platform for cancer immunotherapy has strong potential to contribute to the advancement of cancer nanomedicine.