Biomaterials-Mediated Engineering of the Immune System

Abstract

Modulation of the immune system is an important therapeutic strategy in a wide range of diseases, and is fundamental to the development of vaccines. However, optimally safe and effective immunotherapy requires precision in the delivery of stimulatory cues to the right cells at the right place and time, to avoid toxic overstimulation in healthy tissues or incorrect programming of the immune response. To this end, biomaterials are being developed to control the location, dose, and timing of vaccines and immunotherapies. Here we discuss fundamental concepts of how biomaterials are used to enhance immune modulation, and evidence from preclinical and clinical studies of how biomaterials-mediated immune engineering can impact the development of new therapeutics. We focus on immunological mechanisms of action and in vivo modulation of the immune system, and we also discuss challenges to be overcome to speed translation of these technologies to the clinic.

INTRODUCTION

Biomaterials are synthetic materials injected, implanted, or applied to tissue surfaces in biomedical applications that elicit therapeutic responses. Traditional applications of biomaterials include the use of metal alloys as hip implants and the housing of pacemakers, resorbable polymer fibers as synthetic sutures, and hydrogels as contact lenses (1). In recent years, clinical biomaterials have been developed with increasingly complex functionality, such as nanoparticles that deliver nucleic acids into host cells and/or tissues, implantable scaffolds as niches for cell delivery and cell therapy, and materials that respond to chemical cues in the tissue microenvironment (2–4). One of the new roles for synthetic materials in medicine under intensive investigation in preclinical and clinical studies is the use of biomaterials for immunomodulatory therapies. Modulation of the immune system is an important approach in modern medicine, with treatment applications ranging from infectious disease to cancer to autoimmunity to graft tolerance. Seven of the top ten highest-selling drugs in 2021 act on the immune system, and unsurprisingly this list includes the Pfizer/BioNTech and Moderna SARS-CoV-2 vaccines (5). Over the last ten years, immunotherapy has become an important pillar of oncology treatment across many cancers, and intensive preclinical and clinical efforts are exploring new therapies aiming to build on these successes. However, a common challenge across diverse therapeutic areas is developing new immunotherapies that effectively direct the immune system to treat disease without inducing unacceptable toxicity.

Diverse biomaterials, including hydrogels, nanoparticles and microparticles, nanoscale fibers, and porous foams, are being developed to enable new types of immunomodulatory interventions and immunotherapies. The most prominent recent example is the development of SARS-CoV-2 mRNA vaccines, which are encapsulated in highly engineered lipid nanoparticle formulations that protect the mRNA during storage and injection into tissue, promote mRNA uptake by cells in vivo, and induce endosomal disruption to deliver the nucleic acid into the cytoplasm of host cells (6). While these mRNA vaccine development campaigns were an amazing feat—from the publication of the viral genome to the emergency authorization of the first vaccines in less than a year—the critical materials technology underlying mRNA delivery has been in development for more than 20 years. Nevertheless, the mRNA vaccine success is an example of how biomaterials are being developed that enable new ways to modulate the immune system to prevent or treat disease.

There are several fundamental mechanisms by which biomaterials can enable immunotherapies. A key issue with safe and effective immunomodulation is providing correct cues to the right immune cells with the proper timing, dose, and location. This could mean immune cells at a site of disease, in disease site–draining lymphoid organs, at mucosal portals of pathogen entry, or in systemic immune compartments. Biomaterials are being developed to deliver immunostimulatory cues for all of these scenarios, with the goal of promoting effective immune stimulation while avoiding uncontrolled systemic immune activation. In this review, we provide an overview of how biomaterials are being designed and used to enhance immunotherapies, highlighting particularly biomaterials-based therapies that are in or entering clinical translation. We organize the discussion by five major application areas—vaccines, cancer immunotherapy, adoptive cell therapy (ACT), tolerance and wound healing, and controlling innate immunity and inflammation—aiming to focus on major conceptual ideas explored in the field to date. Finally, we also discuss challenges and considerations for enabling clinical translation of these approaches, which are by definition more complex than production of a systemically administered small-molecule or antibody-based drug.

PROGRAMMING VACCINE-ELICITED IMMUNITY

Vaccines are one of the most important interventions for modulating the immune system to promote protection from disease, and currently there are more than 100 vaccines licensed by the US Food and Drug Administration (FDA) (7). Despite these successes, there remain many challenges in the development of vaccines against difficult pathogens such as HIV, Mycobacterium tuberculosis, dengue virus, and Plasmodium spp. (to name a few), which establish chronic infections, have high mutation rates, complex life cycles, and/or multilayered mechanisms of immune evasion and immunosuppression (8). In addition, universal vaccines for influenza capable of eliciting cross-seasonal protection and effective therapeutic cancer vaccines remain important unmet needs (8). Finally, the COVID-19 pandemic has illustrated very clearly the threat posed by newly emerging pathogens (9). Biomaterials are being developed to aid in engineering vaccine immunity and gain fine control over the timing, duration, and magnitude of antigen and inflammatory cues during immunization (10). The potential for transformative impact is clearly illustrated by the advent of the Pfizer/BioNTech and Moderna COVID-19 mRNA vaccines, which are based on stabilized mRNA encoding the SARS-CoV-2 S protein encapsulated in lipid nanoparticles (LNPs) for intramuscular delivery. These nanoparticle vaccines were manufactured and entered phase 1 human testing in approximately 2 months following the release of the SARS-CoV-2 genome sequence, achieved more than 90% protective efficacy in phase 3 trials, and were licensed within less than 12 months, an astounding feat (11). While this application of biomaterials alone has the potential to forever change vaccine development, there are numerous ways biomaterials are being engineered to enhance diverse vaccine modalities.

Promoting Vaccine Delivery to Lymphoid Organs

A fundamental requirement for vaccines is effective delivery of antigen and inflammatory cues to lymphoid organs (most often lymph nodes draining a muscle or skin injection site). This issue is often taken for granted, but low-molecular-weight vaccine components (e.g., molecular adjuvants or peptide antigens) show very poor uptake into lymphatics, instead clearing into the bloodstream (12). Antigens that enter lymphatics from the injection site must be captured at the draining lymph node (dLN); material not captured in the dLN continues on through the lymphatic chain to the thoracic duct and then enters the systemic circulation (13). This combination of factors means that in a typical immunization, only a small percentage of the vaccine dose accumulates in dLNs (14). As antigen availability has manifold effects on the immune response, any strategy to augment vaccine delivery to dLNs can have a major impact. Several biomaterials-based approaches to enhance vaccine accumulation in dLNs have been demonstrated.

A first fundamental strategy for lymph node targeting is to formulate vaccines as particulates with a size between ~5 nm and ~50 nm; proteins/particles <5 nm in size tend to rapidly clear into the bloodstream, and particles >50–100 nm in size become trapped in the extracellular matrix at the injection site (15) (Figure 1a). Although leukocytes can internalize such particles and actively transport them to the dLN, this process is generally less efficient in promoting lymph node antigen accumulation compared to direct lymphatic trafficking (16, 17). There are hundreds of preclinical studies exploring this approach; prominent examples include decorating lipid nanodiscs (bicelle-like lipid assemblies stabilized by apolipoprotein-derived peptides) with neoantigen peptides (18), linkage of antigens to albumin-binding lipid tails (to promote association of the vaccine with endogenous albumin, which constitutively traffics out of tissues via lymph) (19, 20), and synthetic protein nanoparticles displaying surface-arrayed antigens (21, 22). Size-based targeting of lymph nodes has also been effectively demonstrated with vaccine adjuvants formulated as polymeric nanoparticles, LNPs, or designed with albumin-binding lipid tails (20, 23, 24). Recently, an elegant strategy for two-stage delivery of small-molecule compounds using polymer nanoparticles ~30 nm in diameter achieved efficient rapid delivery to the subcapsular sinus of dLNs, and these nanoparticles subsequently released small-molecule payloads that permeated the entire lymph node from the sinuses (25). This approach is attractive for the delivery of small-molecule adjuvants such as imidazoquinoline Toll-like receptor 7 and 8 (TLR7 and 8) agonists.

Figure 1.

Biomaterials approaches to enhance vaccines. (a) Materials/therapeutics injected parenterally into connective tissue smaller than ~3–4 nm (~40 kDa for globular proteins) clear directly into the bloodstream, whereas larger particles are instead convected into lymph to reach draining lymph nodes. (b) Lipid nanoparticles encapsulating DNA or RNA are used for delivery of nucleic acids into cells in vivo for vaccines and other applications. These nanoparticles protect the nucleic acid from nucleases, promote uptake into cells, and typically incorporate a pH-responsive ionizable lipid that promotes disruption of endosomes as they acidify, releasing the nucleic acid cargo into the cytoplasm. (c) Nanoparticles can promote immune responses on a single cell level by delivering antigen and innate immune agonists (such as TLR ligands) directly to the same cell. (d) Implantable or injectable polymer scaffolds are being designed that create a microenvironment to chemoattract APCs and APC precursors, where they undergo differentiation, activation, and antigen loading within the biomaterial, followed by migration to draining lymph nodes to prime immune responses. Abbreviations: APC, antigen-presenting cell; TLR, Toll-like receptor. Figure adapted from images created with BioRender.com.

Beyond particle size, a number of additional mechanisms for engineering-enhanced vaccine accumulation in lymph nodes have been recently described (17). Saponin and the TLR4 agonist monophosphoryl lipid A, innate immune stimulants, assembled with phospholipids into ISCOM (immune-stimulating complexes)-like nanoparticles that were shown to both enhance lymph flow and increase permeation of antigen into lymph nodes from the subcapsular sinus, greatly increasing early antigen uptake by cognate B cells after immunization (24). Heavily glycosylated nanoparticles are recognized by mannose-binding lectin, which activates complement, leading to complement-mediated transport of the nanoparticles to follicular dendritic cells (DCs) (26, 27). Follicular DC localization depends on particle size, with nanoparticles <15 nm in size being cleared quickly, while particles ~30–100 nm can exhibit prolonged retention on follicular DCs (27, 28). Amphiphile vaccines, peptide or protein antigens conjugated to an albumin-binding lipid tail via a water-soluble poly(ethylene glycol) chain, have recently been shown to promote uptake across the nasal and respiratory mucosae, by binding to albumin in the airways and transcytosis into the tissue via the neonatal Fc receptor (29, 30). Following transcytosis, amphiphile vaccines are efficiently trafficked to mucosa-draining lymphoid tissues, leading to a strong amplification of cellular and humoral immunity. Each of these material examples represent generalizable strategies that could be applied to diverse subunit vaccine candidates.

Delivery of Nucleic Acid Vaccines

DNA, mRNA, siRNA (small interfering RNA), and other nucleic acids have been under intense investigation for use as vaccines and therapeutics for more than 30 years, with dramatic acceleration in the last few years after the first siRNA therapy was approved by the FDA in 2017 (31) and two parenterally administered mRNA-based SARS-CoV-2 vaccines received emergency use authorization during the COVID-19 pandemic in 2020 (11). Additionally, LNPs carrying mRNA for cancer vaccines have been developed that are intravenously administered for targeted uptake by DCs in systemic lymphoid tissues (32). This formulation elicits a systemic type I interferon response in tandem with antigen presentation, leading to potent T cell priming in mice, and clinical trials of this technology for cancer vaccines are ongoing (ClinicalTrials.gov NCT02410733, NCT04526899). As nucleic acid therapeutics all require delivery into the cytosol or nucleus, the most common strategy is to formulate them in lipid or polymer nanoparticles. Nanoparticle formulations protect nucleic acids from attack by nucleases, promote endocytosis into target cells, and promote release of the nucleic acid out of endosomes into the cytoplasm (endosomal escape) (33) (Figure 1b). Recent studies have further revealed that LNPs and lipoplexes used for mRNA delivery have intrinsic adjuvant activity that is dependent on the ionizable lipid component but independent of the innate immune sensors TLR2, 3, 4, 5, and 7 or inflammasome activation (34–36). Ionizable lipids have also been discovered that can activate other innate sensing pathways such as STING (stimulator of interferon genes) (37). Thus, the innate stimulatory activities of lipids provide another axis for tuning the immune response using LNPs. Hundreds of clinical trials of mRNA vaccines and therapeutics are ongoing, and thus this is a particularly exciting area where biomaterials play a critical role.

Controlling Vaccine Kinetics

Optimized spatiotemporal localization of vaccines in lymphoid organs can have a major impact on both cellular and humoral immunity (38). For example, recent studies have shown that prolonging exposure to antigen via repeated injections or implantable osmotic pumps over a period of two to four weeks can greatly augment T follicular helper and germinal center B cell responses in both mice and nonhuman primates, and prolonged vaccine exposure increased the number of unique B cell clones contributing to the germinal center response—an effect that correlated with enhanced induction of neutralizing antibody responses to HIV immunogens (39, 40). Further, escalating-dose immunization through repeated injections was recently demonstrated to initiate germinal centers that were continuously active for six months, leading to enhanced breadth of neutralizing responses to an HIV Env immunogen (41). Although the classic vaccine adjuvant alum has long been thought to act as a depot for sustained antigen release following administration, in many instances, antigens are rapidly released and cleared within a few days in vivo (42). These kinetics can be altered, however, by chemically engineering immunogens for tight binding to alum (42). Recently, bioresorbable hydrogels have also been designed that allowed release rates of both antigens and coadministered adjuvants to be tuned over days to months, and sustained release from these biomaterials drove stronger germinal centers and induced 1,000-fold increases in affinity maturation against model antigens (43). Skin patches that implant tiny microneedles fabricated from the natural biomaterial silk fibroin protein have also been described that release vaccine over a period of weeks and elicit augmented germinal center and antibody responses (44).

Technologies are also in development that can facilitate single-shot immunization, via biomaterials carriers that release vaccine at a preprogrammed interval following injection. Recent exciting examples of this approach are the use of biodegradable polymer microboxes, which open to release vaccine at designated intervals up to more than one month (45), and freeze-dried vaccine particles coated with nanometer-thick layers of alumina, which dissolve over a time period determined by the layer thickness (46). Further, new fabrication approaches based on 3D printing and advanced microfabrication are enabling the generation of microneedle patches that implant biodegradable polymer tips into the skin that can carry multiple cargos and release bursts of vaccine at defined times (47, 48). While all of the examples in this section are still in the early stages of translational development, these technologies offer promise for enhancing immune priming and may be critical for successful translation of sequential immunogen exposure strategies in development for vaccines against HIV and other difficult pathogens (49).

Optimizing Immune Stimulation at the Single-Cell Level

A number of biomaterials-based vaccine strategies are being developed that aim to program immunity at the single-cell level. Nanoparticles or soluble polymers functionalized with vaccine components are one important class of materials in this context: Nanoparticles displaying many copies of an antigen enhance B cell activation in vitro and in vivo by efficiently engaging B cell receptor signaling (21, 22, 50). Recent studies using DNA origami nanoparticles revealed that antigen spacing and 3D geometry directly impact the potency of B cell activation (51). An intriguing approach is the use of peptide nanofibers that multivalently present T cell and/or B cell epitopes; this platform has shown the capability of breaking tolerance to self-antigens (52, 53). Nanoparticles and soluble polymers that carry antigen together with innate immune stimuli such as TLR agonists have been designed to activate antigen-presenting cells (APCs) and B cells that acquire antigen, thereby amplifying both cellular and humoral immunity (23, 54–56) (Figure 1c). Antigen/TLR agonist–codelivering nanoparticles can elicit distinct T cell priming as a function of route of immunization, with intravenously administered nanoparticles priming a larger proportion of stem-like memory T cells compared to subcutaneous immunization and correlating with enhanced antitumor activity when applied as a cancer vaccine (57).

Another strategy for immune cell programming is designing biomaterial scaffolds that mimic a localized infection by attracting, differentiating, antigen-loading, and activating DCs, which subsequently migrate to the dLN to prime potent immune responses (Figure 1d). Subcutaneously implanted poly(lactide-co-glycolide) (PLGA) disc-like sponges loaded with GM-CSF to attract DC precursors, tumor lysate as a source of antigen, and CpG to activate recruited DCs have been shown to effectively prime antitumor immunity in mouse models of melanoma (58, 59) and are in a phase 1 clinical trial (ClinicalTrials.gov NCT01753089). Furthermore, this strategy has been adapted to injectable biomaterial scaffolds and can effectively drive humoral immunity and T cell responses (60). Although the most common clinical vaccine adjuvant alum in principle acts to promote these same processes (61), engineering scaffolds to explicitly promote each step in this cascade can considerably amplify immune responses relative to this traditional adjuvant (60).

INCREASING THE SAFETY AND EFFICACY OF CANCER IMMUNOTHERAPY

Immunotherapy has arguably had its greatest clinical impact in the treatment of cancer. The development of antibodies that block immune checkpoint signaling through CTLA-4 (cytotoxic T lymphocyte antigen 4) or PD-1 (programmed death 1) pathways and chimeric antigen receptor (CAR) adoptive T cell therapies have had dramatic impacts on patient survival in a subset of cancers (62, 63). Despite these successes major challenges remain, as only a minority of patients respond to checkpoint blockade therapy in most cancers, and CAR T cell therapies have so far had limited success in the treatment of common solid tumors (64). Biomaterials-mediated immune modulation may offer approaches to overcome some of the challenges in therapeutically targeting immunomodulatory pathways and offer unique ways to stimulate antitumor immunity.

Engineering Immunotherapeutic Drug Localization

A major hurdle to developing additional effective cancer immunotherapies is that administration of immunostimulatory drugs intravenously, a standard drug treatment practice in traditional oncology, very often leads to unacceptable toxicity. Hence, broad classes of immunotherapy agents such as cytokines, innate immune receptor agonists, and agonist antibodies against costimulatory receptors have so far faced significant challenges in clinical development (65, 66). Intratumoral administration of these drugs is one obvious strategy that is being explored to lower systemic exposure and focus the action of immunotherapy agents on tumors and/or tumor-draining lymph nodes (TDLNs). Since the FDA approval of the oncolytic virus–based talimogene laherparepvec as the first intratumoral immunotherapy in 2015, intratumoral treatment trials have steadily expanded (67, 68). However, intratumorally injected drugs typically rapidly leak into the systemic circulation, aided by high intratumoral fluid pressure (69). Such systemic dissemination lessens the therapeutic effect and can lead to toxicity that is indistinguishable from systemic immune stimulation. To mitigate these issues, immunotherapy agents have been fused to antibodies or recombinant protein domains that bind to antigens expressed by tumor cells (70) or to ubiquitous components of the tumor extracellular environment such as collagen (71–73). When injected intratumorally, these therapeutics anchor to tumor cells or the tumor extracellular matrix, promoting tumor retention. However, such approaches are limited by the lifetime of binding and endocytic clearance by tumor cells, leading to retention lasting only a few days after a single injection. Injectable or implantable biomaterials can provide more prolonged local drug exposure, by slowly releasing encapsulated drugs or presenting immobilized molecules to immune cells in the tumor. For example, polymer microspheres, injectable hydrogels, and nanoparticles are being developed that become trapped in the tumor extracellular matrix on injection (due to their size) and release cytokines, chemokines, or combinations of factors over a period of days or weeks in the tumor site (Figure 2a). Such approaches have been shown to drive potent remodeling of the tumor microenvironment (TME), prime systemic tumor immunity, and synergize with other localized therapies such as radiation treatment (74–78). Recently, an approach to create local cytokine factories in tumors was reported based on the delivery of engineered cells releasing IL-2 encapsulated in hydrogel microspheres, which were injected locoregionally to treat ovarian cancer models (79). This technology is already moving toward first-in-humans testing. Similar approaches are being developed to provide sustained delivery of immunomodulatory drugs to tumor resection sites using injectable or implantable hydrogels carrying small-molecule or protein drugs (80–82).

Figure 2.

Enhancing cancer immunotherapy. (a) Biomaterials that can be injected into a tumor (such as bioresorbable polymer microspheres) and release immunomodulatory drugs over a period of days or weeks are being developed to focus immunotherapy in tumors. (b) Nanoparticles ~10–100 nm in diameter and carrying immunotherapy agents when administered intravenously accumulate in tumors through their dysfunctional vasculature, allowing immunotherapy delivery to disseminated metastases. (c) Nanomaterials are being developed that promote various forms of inflammatory cancer cell death to promote antitumor immunity. Figure adapted from images created with BioRender.com.

While these examples of drug delivery to tumors above have all demonstrated enhanced safety, efficacy, or both, biomaterials that release immuno-oncology agents into the TME must have their rate of drug release precisely matched to the rate of local drug consumption by target cells, or the system reverts to the case of free intratumoral drug injection and systemic leakage of the payload. An alternative strategy is to engineer a delivery matrix that can be injected into the tumor to present immunotherapy drugs rather than releasing them. For example, spherical nucleic acids, nanoparticles with surface-anchored DNA or RNA ligands for TLR9 and TLR3, have been developed for intratumoral administration to activate innate immune cells (83). Agarwal et al. (84) recently developed an approach to anchor protein drugs to the surfaces of alum particles. A single intratumoral injection of alum loaded with IL-2 or IL-12 led to substantial complete responses in multiple syngeneic murine tumor models and induced systemic antitumor immunity without toxicity. This simple alum-anchoring immunotherapy approach is currently in manufacturing for first-in-humans testing in 2023. A final key approach using synthetic biomaterials to augment localized immunotherapy is in the setting of gene delivery to tumors. Intratumoral injection of LNPs or polymer particles carrying mRNA encoding cytokines or costimulatory receptors has been demonstrated to safely promote remodeling of the TME, and certain cytokine/receptor gene combinations have shown dramatic tumor rejection responses in mouse models (85–88). Clinical trials of intratumorally delivered mRNA are ongoing.

While direct intratumoral injection is very effective in stimulating a targeted lesion, it remains of great interest to achieve targeted immunomodulation in disseminated tumors. Nanoparticles administered intravenously have been intensively studied in the biomaterials community for more than 30 years for delivery of drugs to metastatic tumors, based on the ability of particles in the ~10–100-nm size range to exhibit low-level accumulation in tumors through their dysfunctional blood and lymphatic vasculatures, which permit nanoparticle entry into tumors without lymphatic clearance (89) (Figure 2b). This approach has met with only modest clinical success in the delivery of chemotherapy agents, due in part to the inability of nanoparticles to effectively concentrate drug in all cells throughout a tumor bed, and rapid phagocytosis by tumor-associated macrophages and myeloid cells surrounding the tumor vasculature (90). However, this latter limitation is now being pursued as a strategy to target tumor-associated myeloid cells and reprogram the TME—for example through delivery of TLR agonists to tumor-associated macrophages (91). An alternative approach is the design of nanoparticles functionalized with antibodies or cytokines to target circulating lymphocytes in the blood, which can then transport drug-loaded biomaterials into tumors (92, 93). New biomaterials are also being developed to enhance the delivery of immunotherapy agents deep into tumors. For example, lipid nanodiscs carrying a STING agonist prodrug were recently shown to efficiently penetrate deep into solid tumors following intravenous administration in syngeneic mouse models of colon carcinoma, leading to profound antitumor immune responses (94). Nanoparticles are also being designed to release immunomodulatory payloads in response to specific environmental properties of the TME. Li et al. (95) developed nanoparticles that released encapsulated annexin A5 in response to the oxidative microenvironment in tumors, thereby blocking phagocytosis of tolerance-inducing apoptotic tumor cells and shifting the TME toward an immunity-inducing state. These approaches appear promising for enabling systemic targeting of immunotherapy agents to tumors.

Nanoparticle Formulations Promoting Immunogenic Cancer Cell Death

In order to initiate the cancer-immunity cycle, cancer cells must undergo immunogenic cell death (ICD), releasing tumor antigens in tandem with damage-associated molecular patterns (DAMPs) (96–102). DAMP signals—such as annexin A1, surface-presented calreticulin, and extracellular ATP or high-mobility group protein B1 (HMGB1)—recruit and activate DCs to migrate to TDLNs for effective antigen presentation to T cells. Hence, therapeutic approaches to amplify ICD in tumors are of great interest for cancer immunotherapy.

A number of traditional chemotherapeutic drugs such as doxorubicin, mitoxantrone, and oxaliplatin have been found to promote ICD (97), but the efficiency of ICD induction is often poor, and systemic chemotherapy administration can cause collateral immunosuppression by acting on immune cells in both the tumor and other tissues. Biomaterials such as nanoparticle formulations of chemotherapy agents are being explored to enhance ICD induction, particularly through combination drug delivery and focusing the action of chemotherapy agents on cancer cells (Figure 2c). For example, lipid vesicles have been designed to encapsulate a conjugate of doxorubicin and an indoleamine 2,3-dioxygenase-1 (IDO-1) inhibitor while also carrying a camptothecin derivative chemically linked to each lipid. These nanoparticles accumulated in tumors following systemic administration and elicited strong antitumor immunity through the combined effects of ICD induction and inhibition of immunosuppressive IDO activity (103). As a second example of synergistic drug delivery, LNPs have been developed to codeliver oxaliplatin and dihydroartemisinin, a small-molecule drug that produces reactive oxygen species (ROS), leading to potent induction of ICD in murine colon carcinomas (104). Nanoparticles are also being developed to engineer specific pathways of inflammatory cell death: Wang et al. (105) developed nanoparticles surface-conjugated with an N-terminal fragment of gasdermin (GSDM), which was attached to the particles via a linker cleavable by the PET imaging agent phenylalanine trifluoroborate (Phe-BF₃). GSDM is a pore-forming protein that induces pyroptosis (106, 107); when tumor-bearing animals were administered the GSDM-linked nanoparticles followed by Phe-BF₃, the PET agent concentrated in the tumor and released GSDM from the particles taken up by tumor cells, leading to ICD and strong upregulation of inflammation in tumors. The ability of nanoparticles to efficiently traffic into lymphatic vessels has also been exploited to promote ICD and simultaneously inflame TDLNs: Lipid nanodiscs carrying the antimicrobial molecule melittin induced tumor cell death following intratumoral injection and concurrently trafficked to the TDLN, where they induced high levels of DC activation, leading to enhanced T cell priming and systemic immunity (108).

Cancer phototherapy makes use of biomaterials to deliver photosensitizers to the tumor. Following localized stimulation by tissue-penetrating near-infrared light applied to tumors, photosensitizers generate ROS (photodynamic therapy) or heat (photothermal therapy), promoting ICD while minimizing systemic toxicity (109–111). Nanoparticle agents are being developed to enable combinatorial immunomodulation in tumors during phototherapy: In one recent example, polymeric photosensitizer nanoparticles were conjugated with a proteolysis-targeting chimera (PROTAC) designed to degrade IDO-1. When administered to mice bearing syngeneic breast tumors, the particles accumulated in the tumors; reduced IDO-1 levels; and following near-infrared irradiation of the tumor, induced potent cancer cell ICD (112). Another approach is to utilize phototherapy to activate immunotherapy agents in the TME. For example, the TLR7/8 agonist resiquimod was linked to photothermally responsive polymer nanoparticles via a heat-sensitive linker. Injection of these nanoparticles in mice followed by photoirradiation of 4T1 breast tumors led to cancer cell ICD accompanied by release of the TLR agonist, which led to potent inflammation in the tumors and systemic antitumor immunity (113). Hence, nanoparticle biomaterials can be used to augment ICD in a variety of ways.

AUGMENTING ADOPTIVE CELL THERAPY

ACT employs ex vivo–expanded therapeutic immune cells as a treatment for cancer and infectious disease. The most successful form of ACT to date is CAR T cell therapy, wherein autologous T cells from patients with cancer are transduced to express a synthetic antigen receptor (the CAR) that redirects these cells to recognize and destroy cancer cells bearing the target antigen. This treatment has been successful for a range of hematologic malignancies, leading to six FDA-approved CAR T cell products (114–116). Beyond CAR T cells, ACT is being developed using a variety of other immune cells, including T cell receptor (TCR)-transgenic T cells, tumor-infiltrating lymphocytes, natural killer cells, and macrophages for treatment of cancer and infectious diseases (63, 64). Adoptive therapy using regulatory T cells (Tregs) and engineered regulatory immune cells is also being pursued for treatment of autoimmune disorders and graft-versus-host disease and to promote transplant tolerance (117). Despite the success of CAR T cell treatments in leukemias and lymphomas and promising early data from trials with other immune cell therapies, a number of challenges remain. Several biomaterials-based approaches appear promising for addressing limitations of current adoptive therapy treatments and enhancing the function of these living drug products.

While CAR T cells have been shown to be very effective in treating blood cancers, their efficacy against common solid tumors has thus far been disappointing. Treatment of solid tumors with TCR-transgenic T cells or tumor-infiltrating lymphocytes has shown efficacy in some small trials but has not yet translated into an approved therapy (63, 118). This is believed to result from a combination of factors including the immunosuppressive solid-tumor microenvironment (63, 119). Genetically engineering T cells to express supporting factors such as cytokines to maintain their function in solid tumors has been shown effective in preclinical models but has been challenging to translate to humans due to difficulties in controlling the level of gene expression, as continuous cytokine expression can cause transformation of T cells (120, 121). To overcome these issues, nanoparticles releasing supporting drugs have been chemically conjugated to the plasma membrane of T cells (Figure 3a). Nanoparticles coupled to T cells can remain on the cell surface even during cell division, providing sustained autocrine-like release of proteins or small-molecule drugs to T cells as they traffic to tumors in vivo (122–125). This strategy can also be used to deliver therapeutics meant to act on other cells (e.g., tumor cells) in a paracrine manner (126), and it has been used to augment the function of Tregs in addition to effector T cells (127). By engineering nanoparticles that release protein drugs in response to changes in the oxidative state at the cell surface following TCR triggering, drug release can be confined to sites of T cell activation (i.e., tumors and TDLNs), further increasing the safety and efficacy of adjuvant drug delivery (125). This approach is currently being evaluated in the setting of antigen-primed polyclonal T cell therapy in a phase 1 clinical trial (ClinicalTrials.gov NCT03815682). “Backpacking” of immune cells with biomaterials to augment cell function is also being tested with innate immune cells. Discoidal polymeric backpacks loaded with IFN-γ and attached to macrophages have been demonstrated to provide autocrine stimulation, promoting M1 polarization and improving their antitumor activity in models of breast cancer metastasis (128) (Figure 3a).

Figure 3.

Biomaterials approaches to augment adoptive cell therapy. (a) Biomaterials loaded with stimulatory signals such as cytokines or small-molecule drugs have been developed to “backpack” on immune cells, including effector T cells, Tregs, and macrophages, to provide continuous exposure to supportive drugs in an autocrine-like manner. (b) Delivery of CAR ligands to APCs using lipid-CAR ligand conjugates or lipid nanoparticles carrying mRNA encoding the CAR have been demonstrated to provide vaccine-like priming of CAR T cells in vivo. (c) Polymeric scaffolds loaded with patient-derived T cells, CAR-encoding viral vectors, and proliferation-driving stimuli have been designed that upon implantation transduce T cells with a CAR and trigger their expansion in vivo. (d) Nanoparticles carrying CAR-encoding nucleic acids that target T cells are being developed for in vivo engineering of CAR T cells. Abbreviations: APC, antigen-presenting cell; CAR, chimeric antigen receptor; TCR, T cell receptor; Treg, regulatory T cell. Figure adapted from images created with BioRender.com

A second approach to enhance the function of T cell therapies is to utilize vaccines to expand and differentiate the donor cells directly in the patient instead of relying on nonphysiologic ex vivo stimulation in cell culture. This is straightforward for TCR-transgenic T cells, leading to the combination of T cell epitope vaccines and ACT being tested in small clinical trials (129, 130). Two recently reported strategies have been used to engineer vaccine-like stimulation for CAR T cells as well (Figure 3b). One approach utilized a CAR ligand linked to an amphiphilic albumin-binding lipid tail via a water-soluble PEG polymer spacer. This “amph-ligand” bound to albumin on injection, which ferried the compound to dLNs, where it then transferred its lipid tails into the plasma membrane of macrophages and DCs (131). CAR T cells encountering ligand-decorated DCs were activated and received native costimulatory signals in addition to being stimulated through the CAR, leading to substantially enhanced CAR T cell activity in vivo. In a second approach, intravenous injection of LNPs carrying mRNA encoding a CAR ligand was used to transfect DCs in vivo in lymph nodes and spleen, leading to a similar vaccinal stimulation of CAR T cells in tandem with DC-derived native costimulation (132). This latter technology has already moved into a first-in-humans clinical trial in combination with claudin-targeting CAR T cells (133).

Biomaterials are also being developed as implantable matrices to locoregionally deliver ACT T cells to support their engraftment and antitumor activity. Implantable hydrogels and mesh-like scaffolds made of the same materials used in arterial stents have been synthesized that carry CAR T cells or TCR-transgenic T cells together with a variety of supporting stimuli, including checkpoint blockade antibodies, anti-CD3/CD28/CD137 antibodies, and/or cytokines such as IL-15 (134, 135). In addition to supporting donor cell delivery directly to tumors, hydrogel implants have also been used to support engraftment of CAR T cells at primary tumor resection sites (136). These approaches have been shown to substantially boost the persistence and antitumor efficacy of transferred T cells.

Despite the success of CAR T cell therapy in hematologic malignancies, the complexity and cost of current engineered cell therapy products based on ex vivo modification of autologous patient cells remain an important issue (118). Motivated by this challenge, researchers are developing viral vectors targeted to T cells to enable production of CAR T cells directly in vivo (137). However, antivector immunity and toxicities from viral vectors are challenges for such approaches, as is the traditional weeks-long processing time for ex vivo CAR T cell generation. To mitigate these issues, implantable sponge-like polymer scaffolds are being developed that carry a viral vector carrying a CAR gene together with stimulatory anti-CD3/CD28 antibodies and IL-2. When loaded with human T cells and implanted into immunodeficient mice, CAR T cells were produced that were effective as traditional ex vivo–transduced lymphocytes (138, 139) (Figure 3c). Scaffold-based transduction reduces the traditional two- to four-week-long CAR T cell production timeline down to a single day by enabling T cell activation, CAR transduction, and cell expansion steps all to occur in vivo. An alternative approach under intense investigation is to deliver CAR-encoding DNA or mRNA directly to T cells in the blood using targeted nanoparticles (Figure 3d). Proof of principle for such an approach was first demonstrated using anti-CD3-conjugated polymer nanoparticles encapsulating plasmid DNA encoding a CAR and carrying transposon sequences for nuclear integration. These particles transduced T cells in a mouse model and showed effective tumor control in a syngeneic mouse model of leukemia (140). As nuclear integration poses safety risks, follow-up studies showed that anti-CD3-targeted polymer nanoparticles carrying mRNA could also deliver a CAR or transgenic TCR to mouse and human T cells in vivo, leading to effective antitumor immune responses in models of leukemia and solid tumors (141). In a notable recent example, LNPs targeting the T cell surface receptor CD5 were used to deliver a CAR that recognizes fibroblast activation protein (FAP). These targeted LNPs effectively generated CAR T cells in vivo that eliminated FAP⁺ fibroblasts driving inflammation in models of cardiac fibrosis (142). Given the proven safety and efficacy of LNP-mediated mRNA delivery in humans demonstrated by the COVID-19 mRNA vaccines, there appears to be a clear path to test these concepts in humans soon.

PROGRAMMING IMMUNE TOLERANCE AND WOUND HEALING

The immune system is intrinsically designed to maintain a functional unresponsiveness toward self-tissues, known as tolerance. If tolerance is broken, autoreactive T and B cells are activated by self-antigens, often leading to autoimmune disease accompanied by sustained lymphocyte activation and damaging effector functions, despite the persistent presence of the self-antigen (143). Unwanted immune activation also underlies allergy and tissue transplant rejection and can inhibit wound healing. Biomaterials-based treatments provide significant opportunities to improve upon current therapies for promoting tolerance or suppressing inflammation in wound healing.

Biomaterials Delivery of Biologics and Drugs for Immune Tolerance

Current therapies to reduce inflammation and autoreactivity include treatment with anti-inflammatory drugs and biologics that activate immunosuppressive pathways. However, systemic administration of clinically available corticosteroids and anti-inflammatory drugs is marked by a wide range of side effects due to the untargeted suppression of immune cells and increases susceptibility to infections (144).

One approach to increase the efficacy and/or safety of immunoregulatory agents is to use biomaterials to concentrate these drugs at a disease/transplant site. For example, bioresorbable polymer microspheres or porous discs prepared from PLGA that release dexamethasone or the immunoregulatory cytokine IL-33 over time were shown to greatly increase the survival of MHC-mismatched pancreatic islets when implanted together with the β cells, providing localized immunosuppression correlated with increased Treg infiltration in the implant (145, 146). Biomaterials can also present immunoregulatory proteins to promote tolerance. Coimplantation of mismatched pancreatic islets with polymer microgels surface-conjugated with recombinant Fas ligand or PD-L1 promotes substantially enhanced graft survival by promoting Treg accumulation and drives tissue-infiltrating effector cells toward an exhausted PD-1⁺Tim3⁺Lag3⁺ phenotype (147, 148). Karabin et al. (149) developed an elegant block copolymer–based biomaterial that forms an injectable nanofiber hydrogel; upon injection these nanofiber matrices loaded with small-molecule drugs undergo a slow morphological transition, releasing micelles that carry the drug efficiently into the local dLNs. When they were loaded with bioactive vitamin D₃, injection of these nanofibers into atherosclerosis-prone mice led to expansion of Tregs in dLNs, which correlated with an increased appearance of Tregs in atherosclerotic lesions (148).

Alternatively, nanoparticles loaded with tolerogenic agents can be targeted to specific immune cell populations to focus drug action: Targeted delivery of a DNA methyltransferase inhibitor to CD4⁺ and CD8⁺ T cells using antibody-conjugated polymer/lipid hybrid nanoparticles promoted Treg expansion and a reduction in activity of cytotoxic T lymphocytes in models of lupus (150).

Importantly in the context of autoimmunity, the breakdown products of biodegradable materials, especially biodegradable polymers that undergo hydrolysis in vivo, can themselves have immunoregulatory effects that might potentiate or work counter to tolerogenic strategies. For example, poly(β-amino esters) activate APCs via the IRF inflammatory pathway independently of TLR and NF-κB signaling (151), and the natural polysaccharide biomaterial chitosan activates STING (152). Conversely, it has recently been shown that PLGA microparticles contribute to an immunosuppressive environment in a molecular-weight-dependent manner, with higher-molecular-weight PLGA resulting in dampening of maturation molecules and reduced NF-κB activation in DCs (153). The intrinsic immunoregulatory properties of biomaterials remain understudied and provide opportunities for rationally engineered immune activation or immunosuppression.

Generating Antigen-Specific Tolerance

In many disease settings, the development of antigen-specific tolerance would be the ideal approach to safely control pathologic immune responses while avoiding harmful generalized immunosuppression. However, efficient means for the therapeutic induction of antigen-specific tolerance remain an important unmet need. Two general strategies using biomaterials to meet this challenge have been explored, with promising results both preclinically and in early-stage clinical trials: (a) targeting of antigens to key APCs implicated in tolerance in the absence of inflammatory cues and/or with tolerance-driving drugs, and (b) direct stimulation of autoreactive lymphocytes using nanoparticles or soluble polymers that present antigen in the absence of costimulatory cues.

APCs in the spleen and liver have been implicated in promoting tolerance to self-antigens during homeostasis, for example in the clearance of dying cells (154, 155). One promising approach is to use biomaterials to target antigens to APC populations for tolerogenic antigen presentation. Apoptotic autologous cells conjugated with autoantigens and administered intravenously have shown both prophylactic and therapeutic tolerizing effects in mouse models and phase 1 human trials (156). Mimicking homeostatic clearance of self-cells through antigen presentation, DCs in the spleen capture and present these delivered self-antigens in a tolerogenic manner involving IL-10 production and high PD-L1 expression (156). To achieve similar effects in a more easily implemented off-the-shelf therapy, PLGA particles coupled with autoantigens have been employed; intravenous injection of these particles restored tolerance in mouse models of type 1 diabetes through negative regulatory pathways (CTLA-4, PD-1) and expansion of Tregs (157, 158). Intriguingly, nanoparticle-delivered autoantigens have been reported to be more efficacious in mouse models of multiple sclerosis when administered into the lungs rather than intravenously, mimicking convenient aerosol delivery in humans, where they loaded lung DCs with antigen in the absence of costimulatory receptor expression and caused redirection of T cells to the lung (159). Liver Kupffer cells and sinusoidal endothelial cells have also been implicated as tolerogenic APCs, and using polymer nanoparticles or soluble polymers to target antigens to these cells promotes effective tolerance in models of type 1 diabetes and allergy (160, 161).

Combination strategies that target antigens together with immunoregulatory drugs to APCs have also been developed to ensure induction of a tolerogenic state through priming of Treg populations. For example, PLGA particles have been designed for injection that locally attract DC precursors through the release of GM-CSF and simultaneously promote a tolerogenic state in the recruited cells through release of factors such as TGF-β and vitamin D₃ together with autoantigens, leading to antigen-specific tolerance in mouse models of multiple sclerosis and arthritis (162–164). Intravenous administration of PLGA nanoparticles encapsulating rapamycin together with selected antigens or simply admixed with antigens of interest elicits strong prophylactic and therapeutic tolerance, based on the induction of a tolerogenic state in splenic DCs (165–167). This technology is now in clinical testing and has shown promising safety and efficacy results in a phase 1b trial for inducing tolerance to recombinant uricase in gout patients (168).

An alternative strategy is to directly stimulate lymphocytes with nanoparticles or water-soluble polymer chains presenting antigen in the absence of costimulatory signals. Santamaria and colleagues first demonstrated that intravenous administration of class I peptide-MHC (pMHC)-conjugated nanoparticles led to development of CD8⁺ T cells with a regulatory phenotype that could induce therapeutic disease control in mouse models of type 1 diabetes (169). Notably, these pMHC-displaying particles are ~50 nm in diameter and hence engage with T cells very differently from a natural APC. Nanoparticles presenting class II pMHCs stimulate cognate antigen-specific CD4⁺ T cells to become type 1 Tregs that also suppress autoimmune disease (170). Perhaps even more strikingly, repeated administration of nanoparticles displaying the noncanonical MHC molecule CD1d complexed with the invariant natural killer T (iNKT) cell ligand α-galactosyl ceramide was recently shown to induce tolerogenic iNKT cells that suppressed liver and pancreatic autoimmunity (171). Finally, direct antigen delivery to tolerize B cells is being pursued via conjugation of peptide antigens to water-soluble polymers. Intravenous administration of such soluble antigen arrays promoted disease control in mouse models of multiple sclerosis (172–174).

Biomaterials That Modulate Immune Responses to Promote Wound Healing

Biomaterials have long been studied as scaffolds or synthetic ECMs to create a physical supportive niche in vivo and promote wound healing or tissue regeneration (175). Recently, the discovery that T cells, particularly Th2-polarized CD4⁺ T cells, play a critical role in the pro-regenerative effects of biomaterial scaffolds (176) has energized studies where synthetic scaffolds are being designed to engineer such responses for therapeutic impact in sites of inflammation and wound healing. This discovery was made in studies focused on understanding the role of the immune system in the pro-regenerative functions of scaffolds prepared from natural decellularized xenogeneic ECMs, which have shown promising regenerative effects in humans (177). In this setting, T cells are believed to be primed against foreign ECM-derived peptides and polarized to a Th2 state following scaffold implantation and thereafter to infiltrate the scaffold microenvironment and promote development of an M2-like state in macrophages that further drives tissue repair (176, 178). This prohealing Th2 response has been shown to be activated by allogeneic ECM-derived scaffolds, which are now being translated into human testing (179). As another approach building on these findings, PLGA foam discs releasing a model antigen implanted into the ischemic hindlimb of mice that received an alum/antigen immunization (to induce formation of Th2-biased memory T cells) attracted cognate CD4⁺ T cells to the implant site, which promoted local vascularization and repair of the muscle tissue (180). While the roles of T cells in directing wound healing remain under investigation, immune-directive scaffolds to enhance tissue regeneration appear quite promising.

CONTROLLING INNATE IMMUNE RESPONSES AND INFLAMMATION

Innate immune cells play critical roles in generating and regulating inflammation. Dysfunction of these innate cells is implicated in cancer and diverse inflammatory diseases including atherosclerosis, inflammatory bowel disease, arthritis, nonalcoholic fatty liver disease, and diabetic retinopathy (181–184). To this end, biomaterials that can therapeutically modulate innate cells including macrophages, neutrophils, and myeloid cell precursors are being studied for their potential to safely and effectively abrogate damaging inflammatory responses (185).

Recently, certain microbial stimuli have been discovered to imprint metabolic and epigenetic changes in myeloid cells and their precursors, a phenomenon termed trained immunity (186, 187). This rewiring of myeloid cells can lead to enhanced or suppressed responses to innate immune stimulation, and when imprinted in myeloid precursors can have a long-lasting immunomodulatory effect persisting for weeks or months that could be relevant for therapeutic modulation of disease states. Accordingly, synthetic lipid nanodiscs formed by the self-assembly of apolipoprotein A-1 with phospholipids have been shown to be preferentially taken up by monocytes, macrophages, and neutrophils systemically following intravenous injection, as well as macrophages in tumors and myeloid cell precursors in the bone marrow (188, 189). By loading nanodiscs with compounds to either suppress or promote trained immunity, these nanomaterials have been shown to inhibit inflammation associated with transplant rejection (188) or augment immunotherapy responses to checkpoint blockade in tumor models (189), respectively. As these nanoparticles are taken up by myeloid cells but not lymphocytes, they offer an approach to achieve targeted immunomodulation without systemic toxicity.

Macrophages play important roles in the progression of inflammatory atherosclerotic lesions that give rise to cardiovascular disease and strokes; they are therefore another important therapeutic target. Hyaluronan nanoparticles administered intravenously have been shown to enter atherosclerotic plaques through endothelial junctions and accumulate in plaque-associated macrophages, leading to modulation of inflammatory and metabolic activities in the plaque (190, 191). Intriguingly, using block copolymers that self-assemble into different nanostructures (spherical micelles, cylindrical micelles, or larger spherical vesicles), it was shown that intravenously administered micelles predominantly targeted liver macrophages and monocytes in the blood, while vesicles were efficient in targeting splenic and aortic plaque DCs (192). In a therapeutic setting, ROS-scavenging antioxidant nanoparticles inhibited macrophage foam cell formation by decreasing internalization of oxidized low-density lipoproteins and thereby reducing atherogenicity (193). Biomaterials-based approaches are also being developed to modulate inflammatory pathologies in other disease settings such as ischemic stroke (194).

The gut microbiome plays a significant role in modulating host innate immunity and is also implicated in local insults such as inflammatory bowel disease and colon cancer (195). For example, orally administered hyaluronan-bilirubin nanoparticles are taken up by macrophages in the inflamed gastrointestinal tract in models of colitis, polarizing them to an anti-inflammatory state and diversifying the microbiota (196). Oral administration of hydrogels of the natural polysaccharide inulin promoted growth of gut microorganisms producing short-chain fatty acids that promote T cell function and enhanced the antitumor effects of checkpoint blockade treatment in a colon cancer model (197). Similarly, promotion of Peptostreptococcus coupled with antimicrobial killing of competing bacteria using silver nanoparticles also synergized with checkpoint blockade therapy to induce higher T cell responses (198). Hence, biomaterials appear well-suited to help enact therapeutic effects on the microbiome that resolve disease and promote antitumor immunity.

Targeting innate immune cells can also be done in the vasculature prior to their homing to inflamed/damaged tissue. Intravenously administered negatively charged poly(lactide-co-glycolide) (PLG) nanoparticles were able to reprogram innate immune cell trafficking and phenotype to limit deleterious inflammatory responses, thus promoting regeneration in the context of traumatic primary spinal cord injury (199). Negatively-charged particles (made from diverse materials including PLGA, polystyrene, and microdiamonds) have been shown to be taken up by inflammatory monocytes in the blood, triggering their sequestration in the spleen and reducing inflammatory tissue damage in mouse models of myocardial infarction, multiple sclerosis, ulcerative colitis, peritonitis, and lethal flavivirus encephalitis (200). In the context of chronic inflammatory diseases, such as psoriasis, where biologic drugs are often ineffective and are a costly treatment option, nanofibers containing an engineered fragment of complement protein C3dg, with or without the presence of T and B cell epitopes, raised strong antibody responses against both TNF and C3dg and also expanded a population of autoreactive C3dg-specific T cells, suppressing inflammation and disease symptoms (52).

CONCLUSIONS

As evidenced by the studies discussed here, biomaterials-mediated approaches to immune engineering are important in the development of new vaccines and therapeutics, with both successful clinical examples and a large range of new technologies in preclinical and clinical development. While these systems are generally more complex than simple recombinant protein or small-molecule drugs, biomaterials provide effective means to control the timing, dose, and location of immune stimuli in a manner that may be critical for safe and effective immunotherapy in many settings. Several issues are important for the field to continue to progress and reach its full potential: First, biomaterials designed for therapeutic immune engineering need to be developed with translation in mind from the start. Too many bioengineered systems are still being developed that show striking effects in preclinical models but are too hopelessly complex to have any realistic potential for human translation. Second, immune engineering is a highly interdisciplinary field. Exciting new ideas will come from strong partnerships of immunologists working in tandem with colleagues from the physical and engineering sciences. Even more powerful is the training of engineers who are also facile in immunology and immunologists who are comfortable with the possibilities and potential of approaches from chemistry, materials science, and bioengineering convergence science (201). Driven by such interdisciplinary teams, new biomaterials-based therapies may hold the key to solve some of the most challenging pathologies and diseases.