Antibody-Nanoparticle Conjugates in Therapy: Combining the Best of Two Worlds

Abstract

Monoclonal antibodies and antibody fragments have revolutionized medicine as highly specific binding agents and inhibitors. At the same time, several types of nanomaterials, including liposomes, lipid nanoparticles, polymersomes, metal and metal oxide nanoparticles, and protein nanostructures, are increasingly utilized and explored for therapeutic potential due to their versatility, chemical and physical properties, and tunability. However, nanomaterials alone often lack specificity, leading to relatively low efficacy and/or high toxicity. To address this problem, a rapidly emerging area is antibody-nanomaterial conjugates (ANCs), which combine the precise targeting specificity of antibodies with the effector functionality of the nanomaterial. In this review, we give a brief introduction to mAbs and major conjugation techniques, describe major classes of nanomaterials being studied for therapeutic potential, and review the literature on ANCs of each class. Special focus is given to emerging applications including ANCs addressing the blood-brain barrier, ANCs delivering nucleic acids, and light-activated ANCs. While many disease targets are related to cancer, ANCs are also under development to address autoimmune, neurological, and infectious diseases. While important challenges remain, ANCs are poised to become a next-generation therapeutic technology.

Graphical Abstract

Antibody-nanomaterial conjugates (ANCs) are an emerging therapeutic class combining the precise targeting specificity of antibodies with the versatility and functionality of nanomaterials. This review summarizes ANC design, synthesis, and potential therapeutic applications in this rapidly advancing field.

Introduction

The concept of a ‘magic bullet’ (‘Zauberkugel’) was proposed in 1907 by biologist Paul Ehrlich [1, 2] to capture an ideal therapy that would target diseased cells without interacting with healthy cells. Such therapies would be characterized by the essential qualities of high target specificity, robust therapeutic efficacy and low toxicity. An emerging strategy to create such therapeutic agents is antibody-nanomaterial conjugates (ANCs). Like antibody-drug conjugates (ADCs), ANCs combine the highly specific binding properties of antibodies with the efficacy achieved by molecular or nanoscale cargo. As the targeting agent, the antibody delivers the conjugate to specific cells or tissues by binding to particular, often over-expressed, antigens on the surface. This precise targeting ensures that the therapeutic payload is delivered directly to the intended site, minimizing off-target effects. At the same time, the nanomaterial can act as the payload itself (e.g., for photothermal therapy) or as a carrier for transporting a payload of therapeutic agents, such as small-molecule drugs, genes, or imaging markers.[3–6] As a carrier, the nanomaterial may also improve various aspects of delivery, such as increasing the stability and solubility of the cargo, facilitating transport across physiological barriers, increasing drug bioavailability, prolonging circulation time, enhancing permeability and retention (EPR), and reducing systemic toxicity.[7–10]

The literature on antibody-targeted nanoparticles has been previously reviewed specifically for cancer therapy.[11, 12] Here, we review the key considerations for designing ANCs for therapeutics in general, including autoimmune and neurodegenerative disorders, emphasizing recent technological developments incorporating ANCs into T-cell therapy, CRISPR-Cas9 treatments, and central nervous system targeting.

1. Antibodies for ANCs

1.1. Antibodies.

Human immunoglobulins (Igs) or antibodies (Abs) exist in five distinct isotypes : IgG (~80%), IgA (~10%), IgM (~5%), IgD (<1%), and IgE (<1%).[13] Of these, IgG, particularly the subtypes IgG1 and IgG4, and associated antibody fragments, are the most widely utilized for therapeutic applications.[14] The human IgG molecule, a ~150 kDa glycoprotein, comprises two identical heavy chains linked to each other by disulfide bonds, and two identical light chains, each of which is linked to a heavy chain by disulfide bonds (Figure 1a). Each chain consists of four (heavy chain) or two (light chain) immunoglobulin domains with anti-parallel β-sheet structure (110 amino acids).[15, 16] The heavy chain features one variable domain (V_H) at the N-terminus that interacts with the antigen (i.e., antigen-binding site), and three constant domains (C_H1, C_H2, and C_H3) that play structural roles and immune effector functions, such as antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC).[17] The light chain comprises a variable domain (V_L) for antigen interaction and a single constant domain (C_L). The light chain C_L domain is bound to the heavy chain C_H1 domain by disulfide bonds, and the two heavy chains are bound by disulfide bonds in the hinge between C_H1 and C_H2. The resulting assembly presents two binding sites (bivalent).[13, 17] Based on the cleavage of the hinge region by protease, i.e., papain, the Ig structure can be broken down into multiple fragments: two Fab (antigen-binding fragment comprising V_L, C_L, V_H, and CH₁) fragments, and one Fc (crystallizable interaction domain comprising CH₂ and CH₃ for ligands associated with subcellular transport and clearance) fragment.[15] Each of the fragments hosts multiple conserved N-linked glycosylation sites, that play a crucial role in maintaining structural stability and optimal binding to receptors.[18]

Figure 1. Monoclonal antibody (mAb) structures.

(a) IgG structure consists of two heavy (H) chains and two light (L) chains linked with disulfide bonds. Each heavy chain contains three constant domains (C_H1, C_H2 and C_H3) and one variable domain (V_H), while each light chain consists of one constant (C_L) and one variable (V_L) domain. V_L, C_L, C_H1, and V_H together form the Fab region, enabling specific antigen binding. The binding of antigen is specifically mediated by six loops, known as complementarity-determining regions (CDRs), with three loops located in each of the V_H and V_L domains. The remaining constant regions (C_H2 and C_H3) form the Fc domain that regulates effector functions, subcellular transport and clearance. The N-linked glycan (modulates effector function and pharmacokinetics) attached to the conserved asparagine (Asn) residue at position 297 consists of a core structure of N-acetylglucosamine and mannose. ADCC: Antibody-Dependent Cell-Mediated Cytotoxicity; ADCP: Antibody-Dependent Cellular Phagocytosis; CDC: Complement-Dependent Cytotoxicity. (b) Types of mAbs to reduce immunogenicity, from murine mAbs, to chimeric mAbs (murine variable (V) regions grafted onto human constant (C) regions), to humanized mAbs (human immunoglobulin scaffold and murine-derived CDRs), to human mAbs. (c) mAb fragments used in therapeutics. Fab (Fragment antigen-binding): monovalent fragment with antigen-binding capability but lacking immune effector functions. scFv (Single-chain variable fragment): monovalent fragment engineered as a single-chain made of fused variable regions, retaining antigen binding without an Fc region. scFvC (single-chain variable fragment with truncated Fc): mAb fragment combining scFv domains with the Fc region, enabling antigen binding along with Fc-mediated immune effector functions. Diabody: bivalent antibody fragment consisting of two scFv units. Bis-scFv: engineered fragment capable of binding two distinct antigens, for dual targeting. F(ab’)₂: bivalent antibody fragment composed of two Fab units, with bivalent antigen binding without immune effector regions. Nanobody (V_HH): minimal antibody fragment derived from camelid antibodies, which have a single variable region.

1.2. Therapeutic Antibodies.

Monoclonal antibodies (mAbs) are derived from a single B cell, providing a single antibody sequence preparation in contrast to a polyclonal mixture. mAbs engineered to bind selectively and tightly to a specific antigen have greatly impacted medicine, including the treatment of infectious diseases, cancer, and autoimmune disorders.[19] mAb pharmacokinetics are characterized by relatively slow tissue penetration as well as extended serum half-lives, typically lasting days to weeks. The long circulation time is due to a combination of factors, including the large size of mAbs, which exceeds the renal filtration threshold [20]. In addition, lysosomal degradation is reduced by recycling of mAbs by the neonatal Fc receptor (FcRn) in vascular endothelial cells, which returns mAbs to the circulation. These factors appear to maintain titers in patients for extended periods.[21, 22] To enhance mAbs as therapeutics, various molecular engineering strategies are employed for extending serum half-life [23–25], improving target affinity (affinity maturation) [26, 27], boosting effector functions [25], minimizing immunogenicity [28–30], and enhancing developability (e.g., resistance to aggregation).[31–33] Several critical features are discussed below.

1.3. Binding Affinity.

A crucial property of a mAb is its binding affinity for its receptor to ensure effective targeting and competition for receptor binding sites in the presence of competing ligands. In most cases, strong antigen-binding affinity improves the potency and pharmacokinetic profile of an antibody and lowers the required dosage and cost associated with the therapy. High affinity antibody variants can be produced through a range of methods. Aside from affinity maturation in B lymphocytes, several in vitro display technologies (including phage display), as well as in silico mutagenesis techniques, have been developed to achieve this [34]. However, sometimes excessive binding affinity may cause a ‘binding-site barrier’ [35], which occurs when the mAbs or ANCs bind strongly to the surface antigen and fail to diffuse further into the tissue. Particularly, in multicellular tumor spheroids, high-affinity ANCs preferentially saturate peripheral stromal cells near the vasculature, depleting the pool of ANCs available for extravasation and deeper tumor penetration, leading to poor tumor coverage and reduced therapeutic efficacy.[36] Thus, optimization of binding affinity might need consideration during ANC design.

1.4. Antigen Specificity.

The ability of an antibody to selectively bind its intended antigen without off-target interactions is critical for its effectiveness in targeted therapy. Binding is determined by spatially and energetically optimized interactions between the antigen-binding site (paratope) and the corresponding antigenic determinant (epitope), akin to a lock-and-key mechanism [37]. However, natural antibodies can exhibit promiscuity in target specificity due to the conformational flexibility of the paratope, molecular mimicry by other antigens, or allosterically induced changes in other antibody regions. While this poly-specificity can be advantageous in some applications, such as cross-species binding, targeting multiple chemokines, or recognizing toxin isoforms, promiscuity may also reduce potency, cause unwanted side effects, decrease stability and/or cause faster clearance from the body [37]. Additionally, antibody engineering must balance affinity and specificity. To engineer specificity, the selection for antigen binding (a positive selection, in which bound molecules are eluted and collected) may be coupled with a selection against binding to off-target antigens (a negative selection, in which the unbound molecules in the flow-through are collected). Alternatively, the antigen-binding selection can be conducted in the presence of off-target molecules [38]. These methods can modulate specificity to minimize off-target effects.

1.5. Antigen Expression.

Ideally, the antigen targeted by the mAb should be abundantly and specifically expressed on the surface of the target cells (i.e., not expressed on non-target tissue). The existence and identification of such antigens can be an important limiting factor in cancer therapy. Nevertheless, for example, CD133 is a cell-surface antigen expressed on cancer stem cells.[39] Nanoparticles loaded with a topoisomerase inhibitor (SN-38) and conjugated to anti-CD133 antibodies showed enhanced binding and cytotoxicity against a human colon cancer cell line (HCT116), compared to non-mAb-targeted nanoparticles.[40] Homogeneity in antigen expression by the target cells is also usually preferred. However, in case of the ‘bystander effect’ (i.e., the therapeutic agent causes signal release by target cells, which affects non-target or antigen-negative cells), some degree of heterogeneity might be tolerated.[41] Other important considerations include downregulation or shedding of the target antigen. For example, circulating antigens may compete with target cells for binding to the ANCs, eventually resulting in their clearance from the circulation.

1.6. Internalization.

Whether internalization of the ANC-antigen complex is desirable, undesirable, or incidental depends on the mechanism of action of the ANC. For payloads that act inside the cell (e.g., interact with intracellular structures), the antibody should facilitate the internalization of the nanoparticle into the cell. For payloads that can be effective regardless of localization (e.g., photothermal cargo), internalization may not be an important concern. In some cases, the ANC should specifically not be internalized. An example is antibody-directed enzyme-prodrug therapy (ADEPT), in which a cell surface enzyme converts the prodrug into the active form, and thus the prodrug-loaded ANC should remain external.[42, 43]

1.7. Antibody Stability.

The stability of the antibody against denaturation, degradation and aggregation is crucial for ensuring that the conjugate maintains structural integrity and binding affinity to the target antigen over time. To increase antibody stability, several strategies can be employed, including site-directed mutagenesis to improve structural integrity, modifying glycosylation patterns to enhance folding and resistance to degradation, and conjugating antibodies with stabilizing molecules like polyethylene glycol (PEG) to prevent aggregation. Engineering antibodies into simpler formats, particularly antibody fragments (see below), can also improve stability.

1.8. Immune Response to Antibodies.

A significant challenge in using antibodies is the potential for immune responses. For example, the development of human anti-mouse antibodies can neutralize the therapeutic effect, increase clearance from the circulation, decrease accumulation at the target site, and potentially lead to adverse reactions. Reduced immunogenicity can be engineered using chimeric, humanized, or fully human mAbs (Figure 1b), although adverse immune responses (e.g., anti-idiotypic antibodies) are still possible.

1.9. Whole vs Antibody Fragments.

Another consideration is whether to use whole antibodies or antibody fragments (such as Fab, scFv, or diabodies; Figure 1c). Whole antibodies generally offer higher binding affinity and stability, as the two binding sites contribute to avidity. Furthermore, Fc receptor binding from whole antibodies can cause cytotoxicity through ADCC and CDC, which may enhance anti-cancer activity. However, the presence of the Fc domain can lead to non-specific binding to normal tissues, particularly by macrophages which can cause expedited clearance by the liver and spleen. In comparison, antibody fragments can provide advantages like reduced size, better tissue penetration, single-chain engineering, and low immunogenicity due to the lack of the Fc domain and complement-activating region. In the case of F(ab’)2, containing two antigen binding sites joined by disulfide bonds, reduction of the disulfide bonds generates two Fab’ fragments containing free thiol (-SH) groups that may aid coupling to the nanomaterials. Single-chain constructs, such as scFv fragments, are advantageous for their straightforward identification (e.g., using phage display) and production (e.g., E. coli fermentation). However, their stability is generally low compared to Fab’ fragments or whole mAbs. The avidity of whole mAbs is typically considered to be advantageous compared to antibody fragments. However, nanomaterials may compensate for the loss of avidity in ANCs. A recent study demonstrated that targeted delivery by anti-CD4 Fab′-NCs was as efficient as whole antibody-NCs, suggesting that antibody avidity was effectively restored for the Fab′ when presented in the context of a nanoparticle conjugate.[44] The properties of whole antibodies and their fragments may thus be differentially affected by nanoparticle conjugation, indicating an area that deserves additional investigation.

2. General Nanoparticle Characteristics Influencing ANC Activities

2.1. Antibody Density.

The density of antibodies bound on the nanoparticle (NP) surface significantly impacts its biological activity and targeting efficacy. Higher density can enhance binding between the ANC and its receptor (e.g., through avidity effects), potentially increasing therapeutic efficacy. However, in addition to being difficult to achieve, excessive antibody density may lead to steric hindrance or changes in colloidal properties, reducing the accessibility of binding sites and affecting the overall performance of the ANC. Thus, optimizing the density is crucial to balance effective targeting with maintaining functional activity.[45, 46]

2.2. Nanoparticle Shape and Size.

The shape and size of NPs greatly influence their pharmacokinetics, biodistribution, and cellular uptake. Spherical NPs are generally synthesized in a more uniform distribution, and are typically more efficiently internalized by cells compared to non-spherical shapes.[47, 48] Additionally, nanoparticle shape can affect the interaction with cellular membranes and the efficiency of endocytosis.[49, 50] Several bioactivities of the NPs also depend on size. Size influences the efficiency with which NPs extravasate through the endothelial barrier and accumulate in target tissues, with smaller NPs often exhibiting improved penetration and distribution but higher cytotoxicity. On the other hand, intermediate-size NPs (50–200 nm) tend to have longer circulation times, making them ideal for exploiting the enhanced permeability and retention (EPR) effect in tumors. The route of elimination is also size-dependent. For renal clearance, NPs need to be smaller than 6 nm. Larger NPs are cleared by the reticuloendothelial system (RES) or the hepatobiliary system, and NPs over 500 nm require alveolar clearance.

2.3. Surface Charge and Hydrophobicity.

The charge and hydrophobicity of the NP surface and its ligands affects colloidal stability, cellular uptake, and interaction with biological membranes (Figure 2).[51] Positively charged NPs tend to have enhanced cellular internalization due to electrostatic interaction with negatively charged cell membranes but show higher cytotoxicity, while negatively charged NPs exhibit reduced cellular uptake but are more biocompatible.[51, 52] Further, small positively charged particles are rapidly cleared from the circulation, limiting their applicability in systemic applications. Hydrophobic surfaces can enhance the NP’s ability to interact with lipid bilayers, potentially increasing cellular uptake. However, excessive hydrophobicity may lead to aggregation in biological fluids or non-specific interactions, reducing targeting specificity.[53, 54] Thus, the both charge and hydrophobicity must be tuned to achieve optimal delivery and interaction with target cells and minimize non-specific interactions and maintain biocompatibility.[51, 55, 56]

Figure 2. Trends of nanoparticle (NP) size, surface charge, and hydrophobicity on cytotoxicity, blood circulation time, and clearance pathways.

Small, positively charged NPs exhibit high cytotoxicity but are rapidly cleared, reducing their systemic efficacy. Larger, neutral or slightly negatively charged NPs generally show lower cytotoxicity and longer circulation times. Positively charged, hydrophobic NPs are usually cytotoxic, and prone to aggregation and immune clearance. NPs with extended circulation times (e.g., hydrophilic, neutral/negatively charged, 50–200 nm) are ideal for exploiting the enhanced permeability and retention (EPR) effect, while short circulation time typically corresponds to rapid clearance by the reticuloendothelial system (RES) or kidneys. Cytotoxicity is based on qualitative biocompatibility trends observed during in vivo screening of ~130 nanoparticles designed for therapeutic applications [57, 58].

3. Binding and Loading Techniques

Conjugating antibodies and other ligands to various nanoparticles is a critical step in developing ANCs. Conjugation methods can be categorized as either non-covalent or covalent.[59] Chemical conjugation strategies have been recently reviewed comprehensively.[60–62] Here we introduce the major techniques used for ANCs and their advantages and limitations [Figure 3].

Figure 3. Selected conjugation processes and their advantages and limitations.

(a) Non-covalent binding (physical adsorption and electrostatic interactions); (b) carbodiimide chemistry (EDC/NHS mediated coupling of -COOH or -NH₂ functionalized nanoparticles with antibodies); (c) maleimide chemistry; (d) Cu-mediated click chemistry; (e) copper-free click chemistry (azide-DBCO reaction and inverse electron-demand Diels-Alder (iEDDA) reaction).

3.1. Non-Covalent Binding

Non-covalent binding of antibodies onto the nanoparticle surface can be simple to execute and less disruptive to the antibody’s structure and function compared to covalent bond formation.[63, 64] However, the stability of the conjugate can be influenced by several environmental factors like ionic strength and pH.[12, 65, 66] Non-covalent binding primarily occurs through three mechanisms: electrostatic interactions, hydrophobic associations, and specialized interactions.

Electrostatic Interactions:

Antibodies can be adsorbed onto nanoparticles via electrostatic attractions between oppositely charged groups when the nanoparticle surface is charged [Figure 3a]. This method is spontaneous, rapid, and straightforward to execute, but the bond strength is relatively weak, making it susceptible to desorption under certain conditions, such as changes in pH, ionic strength, or the presence of competing ions.[67–69]

Hydrophobic Association:

Interaction of hydrophobic regions on antibodies with hydrophobic NP surfaces facilitates non-covalent attachment without chemical modifications [Figure 3a]. This method is advantageous because it often preserves the native conformation and functionality of the antibody. However, because the hydrophobic effect is relatively nonspecific, the orientation of antibodies on the nanoparticle surface may sterically hinder the antigen-binding sites, potentially reducing the effectiveness of the conjugate. In addition, hydrophobic interactions can be disrupted by the presence of lipids and amphiphiles.[70, 71]

Specialized Interactions:

The specific binding interactions of proteins like Protein A, G, or L can be used to immobilize antibodies on nanoparticle surfaces by binding to the Fc region. This ensures the antibodies are oriented with their Fab regions outward, preserving functionality and enhancing binding efficiency and specificity. However, this method can be complex compared to other non-covalent loading techniques, due to the need for pre-coating (e.g., with Protein A) and purification of the pre-coated NPs.[72, 73] Another common approach is the His-tag, a sequence of histidine residues, which bind with high affinity and selectivity to metal ions like nickel or cobalt on the nanoparticle surface.[74] Genetic fusion of His-tag sequences to antibodies allow control over the site of attachment. Optimization of the tag position may be required to prevent steric hindrance or non-specific binding.[75, 76]

3.2. Covalent Conjugation

Covalent cross-linking creates stable chemical bonds between the antibody and the nanoparticle. Most techniques are relatively nonspecific with regard to site, which may partially block the antibody’s binding sites and reduce efficacy.[60, 77, 78] However, specificity of positioning can be improved by using low abundance amino acids, such as cysteine, or incorporation of non-natural or chemically modified amino acids.[79, 80]

EDC Coupling:

EDC (1-ethyl-3-(3-dimethylaminopropyl)carbodiimide) is employed to activate -COOH groups present on the nanoparticle surface, which then reacts with the amine groups on the antibody to form amide bonds [Figure 3b]. This method is straightforward but may result in a heterogeneous population of conjugates with varying orientations, as well as unintended cross-linking.[81–83] Amidation techniques other than EDC coupling are also available [84, 85], although EDC is widely used.

Thiol Chemistry:

In this approach, -SH groups on the antibody are targeted for conjugation with maleimide-functionalized nanoparticles, forming stable thio-ether bonds [Figure 3c]. This method enables more controlled and site-specific conjugation by reducing disulfide bonds, particularly in the Fc region in the antibody, to expose free thiol groups. This precision enhances the functional orientation of the antibody, leading to improved performance in applications.[86, 87] However, the process requires attention to oxidation state, and specificity will be influenced by the presence of cysteines in the variable regions.[88, 89] Disulfide rebridging chemistry targets the hinge-region disulfides of antibodies for site-specific nanoparticle attachment.[90] In this approach, the disulfide bonds are cleaved and reformed into stable linkages, preserving the antibody’s structural integrity[91] and improving conjugation homogeneity.

Click Chemistry:

Click chemistry (e.g., copper-catalyzed azide-alkyne cycloaddition (CuAAC)), provides a highly efficient bioorthogonal approach for conjugation.[92, 93] In this method, azide-functionalized nanoparticles react with alkyne-modified antibodies in the presence of a copper catalyst, resulting in a stable 1,2,3-triazole linkage [Figure 3d]. The reaction proceeds with high specificity and yield, thanks to the strain-promoted nature of the cycloaddition. CuAAC is advantageous for its minimal impact on antibody functionality and its compatibility with complex biological environments, making it a popular choice for precise and effective conjugation in various applications. A disadvantage is the requirement to introduce the alkyne functional group into the antibody, which may be technically challenging.[94, 95] Although rare, in certain scenarios, CuAAC reactions might lead to unintended antibody modifications or degradation.[60]

Cytotoxicity and other issues with copper can be avoided using an azide-DBCO (dibenzocyclooctyne) click reaction, [96] which also forms a stable triazole linkage. In this approach, antibodies are functionalized with an azide group via a bifunctional linker, while nanoparticles are modified with DBCO through a surface coating or linker [Figure 3e]. Although this reaction is relatively slow, a faster copper-free alternative is the inverse electron-demand Diels-Alder (iEDDA) reaction.[91, 97, 98] In this reaction, antibodies are functionalized with either a diene (Tz) or a dienophile (e.g., cyclooctene), typically by attachment to antibody amines, and nanoparticles are functionalized with the complementary reactive group [Figure 3e]. These and other click reactions yield covalent bonds that are stable under physiological conditions.

4. Notable ANCs and Their Therapeutic Potential

4.1. Liposomes

Perhaps the most well-studied ANC nanodelivery agents are liposomes. These spherical self-assembled structures may have one (i.e., unilamellar) or several (i.e., multilamellar) phospholipid bilayers surrounding an aqueous core.[99] Liposomes are used to encapsulate a wide variety of therapeutic agents, both hydrophobic (soluble in the lipid membrane) and hydrophilic (dissolved in the aqueous core), including small molecules, peptides, and macromolecules (proteins and nucleic acids). The ability to entrap both hydrophilic and hydrophobic drug moieties within the same system makes liposomes an ideal carrier in combination therapy where two or more drugs are intended for targeting the same cell or tissue, particularly in cancer therapeutics. As a delivery system, liposomes are attractive because of their simple structure, self-assembly, biocompatibility, bioavailability, and potential to carry large payload.[8, 100] Furthermore, liposome size, charge, surface properties and stability (both in vitro and in vivo) can be modulated by adjusting the chemical composition of the lipid mixture and/or varying the preparation process.[99, 100]

When antibodies are conjugated to the surface of liposomes (i.e., immunoliposome), they become highly specific delivery vehicles capable of targeting particular cells or tissues that express the corresponding antigens. Most studies in this field focus on cancer targeting, employing a wide range of antibodies. For example, anti-Her2 PEGylated liposomes encapsulating a toxin protein have shown receptor-specific binding, internalization and high cytotoxicity for Her2-overexpressing SK-BR3 cancer cells.[101] In advanced solid tumors, liposomes containing doxorubicin have been conjugated with antigen-binding fragments (Fab’) of cetuximab, an anti-EGFR mAb. These nanoparticles have shown significant potential in targeting EGFR-overexpressing tumor cells, enhancing the cytotoxic effects of doxorubicin while minimizing its systemic toxicity.[102] Gene delivery by ANC liposomes is also a promising approach. For example, an anti-transferrin receptor scFv-targeted liposome delivering the p53 gene was well-tolerated and effective against solid tumors at low DNA doses in a Phase I clinical trial.[103, 104]

Conventional liposomes often face the problem of rapid elimination from the systemic circulation, as they are absorbed and cleared by the reticuloendothelial system (RES).[105, 106] Coating the liposomal surface with relatively non-interacting polymers (typically PEG) can reduce opsonization and subsequent clearance by RES.[107, 108] For example, PEGylated doxorubicin liposomes (i.e., Doxil^®) have a significantly higher elimination half-life (T_1/2 = 55 h) compared to the non-PEGylated counterpart (i.e., Myocet^®) (T_1/2 = 2.5 h), or doxorubicin alone (T_1/2 = 0.5 h).[109] This principle also applies to immunoliposomes. For example, in an in vivo rat liver metastasis model, 25% of PEGylated immunoliposomes targeting a tumor antigen remained in the bloodstream 24 h post-injection, indicating a substantial circulation time.[110] However, a potential problem is that when antibodies and PEG are co-immobilized on the same liposome surface, the PEG layer can sterically hinder target binding by the antibody or antibody fragment.[111] This issue can be mitigated by attaching the antibody via a PEG spacer arm that extends beyond the dense PEG layer. For instance, a doxorubicin-loaded liposome conjugated to an anti-Her2-scFV (trastuzumab) via a PEG spacer (PEG-DSPE) targets Her2-overexpressing tumors.[112] This formulation remains in circulation for extended periods, allowing it to accumulate in tumors, and undergoes receptor-mediated endocytosis upon binding to Her2-expressing cells, releasing doxorubicin directly into the tumor cells. Similarly, conjugation of anti-c-Met scFv to PEGylated liposomes, via a maleimide-PEGylated lipid, enabled efficient delivery of doxorubicin into lung cancer cells expressing c-Met.[113] A similar approach enabled Doxil^® to be targeted by a nucleosome-specific mAb to target various tumors.[114] For larger cargo such as protein toxins, conjugation of endosome-disruptive peptides can improve release of the cargo from PEGylated immunoliposomes. For instance, co-encapsulation of a fusogenic peptide activated at low endosomal pH significantly improved release of diphtheria toxin A chain and increased cytotoxicity against ovarian carcinoma cells.[115]

Antibody-conjugated liposomes have been explored for the treatment of several diseases beyond cancer. For infectious diseases, liposomes displaying antibodies against Plasmodium falciparum, the causative agent of malaria, have been used to specifically target infected red blood cells.[116] Similarly, erythrocyte-based liposomes conjugated with anti-E. coli antibodies can deliver polymyxin B, a potent antibiotic, to resistant E. coli bacterial strains.[117] For neurodegenerative disease, liposomes conjugated to two different mAbs, anti-amyloid β and anti-transferrin receptor, were designed to cross the blood brain barrier (BBB) and target amyloid β.[118] To target damaged cells (e.g., from ischemic disease), liposomes can be designed to bind to exposed intracellular components, such as myosin. This strategy has been explored for myocardial infarction by targeting cardiac myosin [111], where the liposomes appear to help seal membrane damage and reduce cell death.[119, 120] Similarly, anti-actin-targeted liposomes reduced hemorrhage following focal embolic stroke in rats.[121] Cytoskeleton-specific immunoliposomes, capable of fusing with damaged cells, have also been utilized as carriers for gene delivery into hypoxic cells.[122] Thus, antibody-targeted liposomes represent a major application for ANCs.

4.2. Lipid Nanoparticles (LNPs)

Unlike liposomes, lipid nanoparticles (LNPs) lack a significant aqueous interior and typically form a micelle-like arrangement within their core, depending on formulation and synthesis parameters.[123] LNPs are currently used primarily for nucleic acid delivery.[124] Key components of LNPs include: phospholipids as a structural basis; cationic or ionizable lipids that complex with negatively charged nucleic acids and also facilitate endosomal escape; cholesterol to increase membrane fluidity, thus enhancing stability, and promote membrane fusion; and PEGylated lipids that increase stability and circulation time, as discussed previously [Figure 4].[125, 126] LNPs shield encapsulated nucleic acids from degradation and prevent the activation of RNA-sensing pathways and innate immune responses.[124] LNPs may be advantageous over liposomes in these cases, as illustrated by a comparison of gene delivery by anti-EGFR liposomes vs. immunolipoplexes (LNPs). The immunolipoplexes showed ~100x greater transfection efficiency and greater anti-tumor activity in a mouse model.[127] The effectiveness of LNPs for delivering nucleic acids to the cytosol, along with straightforward synthesis, compact size, and stability in serum, makes LNPs a major delivery vehicle for DNA and RNA therapies.[128, 129] Ionizable LNPs are particularly well-suited for nucleic acid delivery. These LNPs are nearly uncharged at physiological pH, but become positively charged during endosomal acidification. The cationic charge promotes release from the endosome, delivering the nucleic acids into the cytosol.[130, 131] For example, ionizable LNPs conjugated to an antibody that binds to an endothelial antigen, plasmalemma vesicle-associated protein, can deliver mRNA to lung tissue.[132]

Figure 4. Lipid nanoparticle (LNP) structure and antibody-targeted LNP-mediated nucleic acid delivery.

(a) Lipid nanoparticles typically consist of an ionizable lipid, a helper lipid, cholesterol and polyethylene glycol (PEG)–lipid. (b) Upon binding to the target cells, the Ab-targeted LNPs are internalized by receptor-mediated endocytosis followed by endosomal escape and payload delivery. (c-e) Endosomal escape mechanisms. PEGylated lipids facilitate fusion with endosomes (c); anionic lipids promote endosomal disruption through osmotic destabilization and charge reversal from proton influx (d); cationic lipids promote disruptive colloid osmotic effects (e).[136, 137]

Despite the advantages, LNPs face challenges such as low drug-loading capacity as well as biodistribution issues. Like many NPs, LNPs tend to accumulate in the liver and spleen.[133, 134] In this context, antibody conjugation to LNPs provides major benefits. For instance, conjugating anti-CD4 antibodies to LNPs enables targeting and greater mRNA delivery to CD4+ T cells.[135] Several Ab-targeted LNPs have been developed for use in targeted gene therapies, including siRNA, CAR T cell therapies, and CRISPR-Cas9. [vide infra]. Although less well-established than liposomes, LNPs represent a rapidly advancing subfield for lipid-based ANCs.

4.3. Polymeric NPs

Polymeric nanoparticles for therapy are generally composed of biocompatible and biodegradable polymers, including poly(lactic-co-glycolic acid) (PLGA), polyethylene glycol (PEG), and polycaprolactone (PCL). These nanoparticles can be engineered to control drug release rates and improve stability in the bloodstream. In analogy to the distinction between liposomes and LNPs, the two primary forms of polymeric NPs are nanocapsules, which consist of a cavity surrounded by a polymeric shell, and nanospheres, which are solid matrix systems. One type of nanocapsule is polymersomes, artificial vesicles formed by amphiphilic block copolymers, similar to liposomes but with enhanced stability and cargo retention. A recent advance is shell-crosslinked polymeric NPs, which can encapsulate payloads within a hollow nanocage core.[138] This design may enhance stability and loading capacity and allows adjustable cargo release rates through crosslinking density.[139, 140] Various synthesis techniques, such as emulsification, nanoprecipitation, ionic gelation, and microfluidics, produce polymeric NPs with different properties. The drug delivery capabilities of these NPs are highly adaptable. Therapeutic agents can be encapsulated within the nanocapsule, entrapped in the polymer matrix of nanospheres, chemically conjugated to the polymer itself, or attached to the NP surface. As with liposomes, a variety of cargo are possible, including hydrophobic and hydrophilic compounds, from small molecules to biomolecules. Also like liposomes, the versatility means that polymeric NPs are a possible choice for combination therapies. On the other hand, the fact that polymeric NPs are made of nonbiological components raises the possibility of particle aggregation and toxicity, requiring attention to colloidal properties, particle degradation and clearance.

Antibodies can be used to target polymeric NPs. For example, an anti-H-ferritin mAb was conjugated to polylactide NPs via thiol-maleimide-chemistry. These immunoNPs were loaded with paclitaxel palmitate and shown to target cancer cell lines.[141] Likewise, paclitaxel-loaded anti-Her2 immunoNPs made from poly(D,L-lactic acid) could target tumor cells that overexpress Her2 receptors.[142] Docetaxel-loaded PLGA nanoparticles, also targeted by an anti-Her2 mAb, showed improved internalization and cytotoxicity in breast cancer cells.[143] PLGA-PEG nanoparticles conjugated to a CEA-targeting antibody were loaded with paclitaxel and interacted with CEA-expressing cells, toward colorectal cancer therapy.[144] To cross the BBB, methoxy-poly(ethylene glycol)-poly(lactide) (Met-PEG-PLA) and maleimide-poly(ethylene glycol)-poly(lactide) (Mal-PEG-PLA) nanoparticles conjugated to anti-transferrin receptor antibodies can be used [145], analogous to the liposomal delivery strategy discussed above.

Apart from cancer, polymeric ANCs are being studied to treat other diseases. For instance, a novel vaginal gel was designed using an anti-integrin mAb – conjugated polymeric NP (composed of PEI and PLGA-PEG), formulated in a 1% hydroxyethylcellulose (HEC) gel, for preventing attachment of HIV-1.[146] In this case, no cargo was delivered, and the function of the ANCs was to prevent integrin-assisted viral binding to CD4⁺ T cells. In a rhesus macaque model, the ANCs bound to T cells expressing high levels of α₄β₇ integrin, potentially blocking a substantial fraction of attachment sites. For bacterial infections, an anti-S. aureus antibody was conjugated to polymeric NPs (composed of PLGA and PLGA-b-PEG) to deliver antibiotics (i.e., rifampin) to S. aureus biofilm in an infected mouse model.[147] Alternatively, the targeting molecule need not be an antibody. For example, intercellular adhesion molecule 1 (ICAM-1) targeted polymeric NPs can facilitate retention and delivery of drugs to treat peptic ulcer, gastrointestinal cancer, and other diseases of the GI tract.[148] In another example, SDF-1 is a small peptide cytokine that was conjugated to PLGA NPs to deliver agents to metastatic lymph nodes.[149]

Thus, polymeric NPs can be designed for targeted delivery using antibodies as well as other high-affinity proteins or peptides.

4.4. Gold Nanoparticles and Nanorods

Gold nanoparticles (AuNPs) are often utilized for nanomedicine applications due to their strong optical absorption, chemical stability, functionalization potential, and relative biocompatibility.[83] AuNPs exhibit two phenomena – surface plasmon resonance (SPR) and enhanced permeability and retention (EPR) – that are of special interest for biomedical applications. SPR arises from the collective oscillation of conduction band electrons of metal nanoparticles, which occurs at a resonant wavelength typically in the visible or near-visible wavelengths, resulting in strong absorbance and therefore coloration.[150–152] The SPR frequency depends on NP size (e.g., redshift with larger or aggregated NPs), shape (e.g., two peaks in nanorods, depending on the longitudinal and transverse dimensions), and the dielectric constant of the medium.[153–155] Efficient conversion of the energy of absorbed light to heat energy (i.e., gold lattice vibrations), and/or triggering the formation of reactive oxygen species (ROS) via a photosensitizer, enables AuNP use in photothermal or photodynamic therapy, respectively. AuNPs are also used for the EPR effect, in which nanoparticles with extended circulation times preferentially accumulate in tumor tissue.[156] This occurs due to the leaky nature of tumor vasculature, which allows these agents to pass through more easily, coupled with reduced lymphatic drainage in tumors that leads to their retention within the tumor bed.[157] The EPR effect is a type of ‘passive’ tumor targeting, and can be used to supplement ‘active’ targeting strategies, such as antibody-conjugation.

Antibody-based targeting of AuNPs confers specificity and efficiency. Examples include AuNPs conjugated to anti-epidermal growth factor receptor (EGFR) antibodies for targeting EGFR-overexpressing malignant oral [158–160] and cervical cells [161, 162], and anti-Her2 targeting for overexpressed Her2 on malignant breast cells.[163] Antibody-conjugated, drug-loaded AuNPs demonstrated greater cytotoxicity towards target cells compared to their nonconjugated counterparts due to the active targeting capability provided by the antibodies, which also facilitated the internalization of AuNPs into the targeted cancer cells. Moreover, the extent of cytotoxicity was directly linked to the number of targeted receptors on the cancer cells and the quantity of antibodies attached to the AuNPs.[164–166] A higher number of antibodies resulted in greater nanoparticle internalization and, consequently, higher cytotoxicity, indicating that the therapeutic efficacy of these nanoparticles can be tuned by adjusting the amount of conjugated antibodies. Importantly, the presence of antibodies did not hinder the ability of the drug-loaded AuNPs to generate ROS during photodynamic therapy to induce cancer cell death.[164, 167] Lower antibody concentrations led to reduced cytotoxicity, with no cell death observed below a certain threshold.[165] Targeting could be further augmented by conjugation with multiple antibodies (e.g., dual-antibody-conjugated AuNPs), especially in cases where cancer cells are resistant to one of the antibodies.[168] Interestingly, dual-antibody-conjugated AuNPs exhibited higher cytotoxicity than single-antibody-conjugated AuNPs, even when the cancer cells expressed receptors for both antibodies, indicating a synergistic effect.[169]

AuNPs may be targeted and delivered indirectly, as in an approach using antibody-targeted liposomes to encapsulate AuNP cargo.[170] In one example using anti-EpCam for targeting, the anti-EpCam antibody was linked to erythrocyte membranes, which in turn encapsulated paclitaxel-loaded AuNPs. These liposomes showed enhanced cellular uptake, with the paclitaxel cargo released upon irradiation via the photothermal effect. This approach combined two therapeutic modalities, namely hyperthermia treatment with chemotherapy. While antibody targeting is effective, local (as opposed to systemic) administration can further enhance efficacy by avoiding AuNP distribution throughout the volume. Intratumorally administered Her-2-targeted AuNPs were found to be more effective for delivery compared to intravenous injection, as it led to more localized cytotoxicity.[171]

As for ANCs in general, the success of antibody-conjugated AuNPs relies on maintaining the integrity of the antibody’s antigen-binding sites during conjugation. Alterations to these sites can reduce selectivity, so techniques like using protein G to correctly orient antibodies during conjugation have been explored, though these strategies may add potential immunogenicity issues.[172] Optimization in specific systems is needed to ensure preservation of the specific structure and function of the antigen-binding sites without inducing adverse immune responses.

4.5. Metal Oxide Nanoparticles

Antibody-targeted metal oxide nanoparticles represent a promising frontier in therapeutic applications, combining the unique properties of metal oxides with the specificity of antibody targeting. These nanomaterials are synthesized by conjugating metal oxide cores, such as iron oxide (Fe₃O₄), zinc oxide (ZnO), or silica (SiO₂), with specific antibodies against cell surface markers. The unique intrinsic properties of metal oxide nanoparticles – magnetic, optical, and/or catalytic – can be used for therapeutic interventions or diagnostic imaging. For example, iron oxide nanoparticles have been extensively utilized in magnetic hyperthermia, where exposure to an alternating magnetic field induces NP heating, resulting in cell ablation.[173] The same iron oxide nanoparticles can also function as contrast agents in magnetic resonance imaging (MRI), illustrating theranostic potential.

Targeting specificity of metal oxide NPs can be conferred by conjugation of antibodies. Iron oxide nanoparticles have been particularly well-studied in this regard. For example, superparamagnetic iron oxide nanoparticles (IONPs) coated by an amphiphilic polymer conjugated to the anticancer agent bortezomib, targeted by an anti-vascular endothelial growth factor (VEGF) antibody.[174] The agent exhibited both enhanced MRI contrast and effective anti-cancer activity, highlighting the multifunctional potential. Similarly, PEI-coated Fe₃O₄ nanoparticles were conjugated with anti-Fas antibodies to target HeLa cells.[175] Exposure to an alternating magnetic field induced hyperthermia and effectively delivered the anticancer drug cryptotanshinone. Iron oxide can be combined with other materials in nanocomposites, such as the magnetic poly(d,l-lactide-co-glycolide) (PLGA) nanoparticles loaded with the anticancer drug doxorubicin (DOX) and conjugated to the breast cancer-targeting antibody Herceptin^®.[176] This composite demonstrated high affinity for breast cancer cells, sustained release of DOX, and potential as an MRI probe. Beyond oncology, magnetic iron oxide nanoparticles were functionalized with antibodies against nervous necrosis virus (NNV) for detection of this important pathogen in aquaculture.[177] Similarly, ferromagnetic nanoparticles functionalized with polyclonal antibodies against Campylobacter jejuni can selectively capture and isolate the bacterium from poultry samples, illustrating the applicability of these nanomaterials for infectious diseases as well.[178]

Apart from iron oxide, silica nanoparticles can also be functionalized for diagnostic or therapeutic aims. For example, luminescent silica nanoparticles covalently attached to goat anti-human immunoglobulin G (IgG) can aid cell-based diagnosis of systemic lupus erythematosus.[179] For theranostics, mesoporous silica (mSiO₂) nanoparticles were conjugated with antibodies specific for the tumor antigen CD105, labeled with ⁶⁴Cu for positron emission tomography (PET) imaging, and loaded with DOX.[180] The nanoparticles demonstrated enhanced tumor-vasculature targeting via CD105-mediated binding and improved delivery of DOX in a murine tumor model, illustrating modular design of these systems. In one complex design [181], a novel nanoparticle ‘vesicle shuttle’ system was composed of an Fe₃O₄ core for magnetic guidance and a silica shell for attachment of a pH-sensitive PEG corona, which was conjugated to two antibodies – one targeting antigens on circulating exosomes and the other targeting injured cardiac cells [Figure 5]. In this design, nanoparticles attached to circulating exosomes and were then brought to the site of injured cardiac cells by the combination of magnetic targeting and antibody-targeting. There the reduced pH of the infarction caused release of exosomes, improving cardiac function. This study demonstrates the versatility of metal oxide ANCs for multi-specific targeting as well as noncovalent collection and stimuli-responsive delivery. Currently, however, clinical use of metal oxide NPs is often restricted due to low solubility and potential toxicity, underscoring the need for additional study.

Figure 5. Schematic of the dually targeted vesicle-shuttle approach for treating myocardial infarction.

Anti-CD63 antibodies on the vesicle shuttle capture endogenous circulating exosomes. Dual targeting occurs via an external magnet, which directs the magnetic vesicle shuttle to the infarct site, and a conjugated anti-MLC antibody that targets damaged cardiomyocytes. Exosome release is triggered by the infarct environment, characterized by pH < 6.8, leading to acidosis-induced cleavage of hydrazone bonds and shedding of the corona for selective exosome delivery.

4.6. Protein Nanoparticles

Protein assemblies have excellent potential in terms of biocompatibility, biodegradability, versatility, and responsiveness to stimuli.[182–184] Nanostructures can be engineered from assembly of natural proteins, often those involved in structural functions (e.g., collagen).[185–188] Antibody conjugation confers targeting specificity to the protein NPs [189, 190]. For example, an albumin nanoparticle formulation of the chemotherapeutic paclitaxel was noncovalently coated with the monoclonal antibody bevacizumab for targeted delivery to cells expressing VEGF.[191] Similarly, gelatin nanoparticles modified with sulfhydryl groups were used to link NeutrAvidin (NAv), enabling the noncovalent attachment of biotinylated anti-CD3 antibodies.[192] These nanoparticles demonstrated specific targeting and efficient uptake into CD3-positive T-cells through receptor-mediated endocytosis. For stimuli-responsive nanoparticles, one approach based on ROS-sensitive linkers responds to the ROS-rich environment of the tumor microenvironment by releasing therapeutic antibodies.[193]

Compared to inorganic or polymer nanoparticles, the use of protein nanoparticles opens the possibility of precise architectural control. Antibody ‘nanocages’ were designed from self- assembly to mimic multivalent antibody structures, allowing tuning of avidity effects and valency-dependent signaling.[194] Similar self-assembled antibody nanostructures were designed to be pH-responsive for disassembly in the endosomal environment.[195] The goal of such approaches is to incorporate almost any antibody into tunable, stimuli-responsive nanocarriers. However, compared to other options, protein-based nanoparticles can be limited by lack of scalable production, intensive purification processes, and the limited stability of many proteins. In addition, since proteins vary widely in physicochemical properties, nanoparticle formulations generally need to be optimized for each protein, contributing to high development costs.[196] Thus, clinical translation of protein-based ANCs will require interdisciplinary attention to improve their feasibility as therapeutics.

4.7. Hybrid Nanoplatforms

When additional functionality, such as precise intracellular targeting, is required, more complicated hybrid systems may be designed. Antibody-directing antibody conjugates incorporate one mAb for targeting and a second mAb for intracellular binding.[197] In this system, a polymeric nanogel has a targeting mAb attached to its surface for uptake into a specific cell type, and the cargo antibody is then released upon nanogel disassembly. Similarly, self-assembly of cargo proteins with bivalent Fab domains can generate a nano-complex that mimics an opsonized pathogen, effectively delivering the cargo to immune cells such as macrophages by hijacking endogenous uptake pathways.[198] Hybrid systems are also useful for cargo protection, as illustrated by the use of polymer brushes covering a SiO₂ NP core,[199] in which the polymer brushes protected the encapsulated proteins from proteolysis. Such systems demonstrate potential versatility and extensibility of ANCs.

5. Special Delivery: Crossing the Blood-Brain Barrier (BBB)

The blood-brain barrier (BBB) is a significant challenge for delivery of therapeutics to the central nervous system (CNS), affecting treatment of brain tumors, neurodegenerative disorders, stroke, and other diseases. The BBB is a specialized multicellular interface that strictly regulates the exchange of substances between the bloodstream and the CNS, thereby maintaining the brain’s homeostasis. The BBB structure includes micro-vessel endothelial cells, astrocytes, microglial cells, pericytes, and a basal lamina composed of extracellular matrix proteins like laminin, heparan sulfate, and collagen [Figure 6a]. Its integrity is primarily maintained by tight junctions between adjacent endothelial cells, formed by proteins such as claudins, ZO-1, and occludin, which impart high electrical resistance reflecting the solute barrier. An additional barrier layer includes ATP-binding cassette transporters that actively expel various compounds back into the bloodstream, thereby limiting the availability of therapeutic agents (including therapeutic antibodies, recombinant proteins, and nucleic acids) in the brain. As a result, after parenteral administration, therapeutic mAbs typically achieve brain concentrations that are only 0.01–0.1% of their plasma levels, which is inadequate for producing a pharmacological effect in most cases.[200, 201]

Figure 6. Structure of blood-brain-barrier (BBB) and modes of nanomaterial transport across it.

(a) The BBB is formed by capillary endothelial cells, surrounded by basal lamina and astrocytic perivascular end feet. Astrocytes provide a cellular link to the neurons. Magnified view of the contact between two endothelial cells forming adherens junctions and tight junctions that inhibit paracellular permeability. (b) Various endogenous (top three) and exogenous (bottom two) mechanisms of nanoparticle transport across the BBB. Note that paracellular transport is typically uncommon for nanoparticles (NPs). However, ultrasmall NPs can exploit this pathway in areas with compromised tight junctions, such as in a diseased BBB or BBB subjected to focused ultrasound beam (FUS). Similarly, passive transcellular diffusion can take place when assisted by electroporation.

Notably, 145 therapeutic antibodies have been approved by the US FDA [202], but none of them target CNS disorders, reflecting the need for innovative technologies to enable the delivery of mAbs to the CNS. In this regard, ANCs represent a promising strategy for overcoming the BBB and delivering therapeutic agents to the CNS. [201] Several NP systems including polymeric implants [203], polymersomes [204], liposomes [205], micelles [206], magnetic NPs [207, 208], and silica NPs [209, 210] are being studied to deliver therapeutic agents across the BBB. These NP systems can be engineered to exploit endogenous and/or exogenous transport mechanisms to cross the BBB [Figure 6b].[211]

Endogenous strategies involve functionalization (e.g., positive charge application, and addition of targeting ligands, or antibodies) of the NPs to take advantage of existing transport mechanisms for naturally occurring substances, such as glucose or transferrin, allowing the NPs to be taken up by endothelial cells and transported into the brain. In one example, anti-Her2-targeted selenium NPs were capable of crossing the BBB and delivering both anticancer agents and MRI contrast agents directly to Her2-expressing brain tumors.[212] The anticancer agents were triggered for release upon internalization and endosome formation and were able to inhibit the growth and migration, as well as induce apoptosis, of glioblastoma cells. Several polymer nanomaterials have also been studied for BBB-crossing properties. For example, poly(β-L-malic acid) was targeted using an anti-transferrin receptor antibody and covalently loaded with antibodies that function as immune checkpoint inhibitors. These nanoparticles crossed the BBB using transferrin receptor-mediated transcytosis and stimulated a localized anti-tumor immune response in the brain.[213] A similar approach can be applied for medical imaging in the brain. Poly(β-L-malic acid) nanoparticles attached to gadolinium DOTA (an MRI contrast agent), and monoclonal antibodies targeting tumor antigens (cetuximab and trastuzumab) are proposed as a diagnostic tool for metastatic brain tumors.[214] Similar receptor-mediated endocytosis enabling brain uptake was observed for PEG-poly(lactide) nanoparticles.[145] While transferrin receptor is a common target, other receptors may provide additional specificity. For example, transcytosis of liposomes displaying both the anti-transferrin receptor and anti-Aβ antibodies bound to Aβ was increased by the membrane receptor for Aβ (receptor for advanced glycation end products or RAGE) [118], illustrating that multiple uptake mechanisms might be combined to improve delivery across the BBB.

In contrast to endogenous delivery, which uses biological pathways for transport, exogenous approaches for delivery usually involve chemical or physical mechanisms to temporarily disrupt the BBB. Techniques include osmotic agents, focused ultrasound, or laser or magnetic guidance, which increase permeability and thereby facilitate the passage of therapeutic agents. For example, iron oxide nanoparticles (IONPs) were targeted using antibodies against an epidermal growth factor receptor mutant found in glioblastoma cells (anti-EGFRvIII).[215] Pressurized infusion was used to enable the nanoparticles to cross the BBB, a technique known as convection-enhanced delivery (CED).[216] Exogenous techniques, while effective and potentially applicable to a variety of nanoparticle types, may be less specific than endogenous techniques since they rely on physical forces rather than using existing biological machinery.

In summary, ANCs represent a promising strategy for delivering therapeutics and contrast agents across the BBB. Most ANCs use endogenous strategies (a “Trojan horse” approach) to cross the barrier using endocytic or transcytosis pathways. Surface charge, functionalization moiety and size of the NPs are the critical determinants of the translocation mechanism.[145, 217, 218] For example, positively charged nanoparticles are more readily able to utilize the adsorptive transcytosis pathway compared to neutral or negatively charged nanoparticles, due to attractive electrostatic interactions with the negatively charged endothelial cell surface.[219] ANCs smaller than 200 nm, the approximate size threshold for clathrin-mediated endocytosis, are more likely to cross the BBB.[220] Specificity is also affected by size. For example, silica ANCs accumulated with higher specificity and amount in brain microvascular endothelial cells at smaller size (50 nm) compared to larger size (160 nm), as the larger size resulted in greater nonspecific binding.[221] Surface arrangement of the antibodies also affects uptake efficiency of ANCs, as NP surfaces with fewer high affinity ligands results in enhanced receptor-mediated transcytosis compared to surfaces with high avidity.[222] Further optimization of these strategies is required to provide better permeation and prevent adverse effects, such as damage to brain vasculature and parenchyma or loss of BBB function.

6. ANCs as Next Generation Tools for Nucleic Acid Delivery

RNA-based therapies can manipulate gene expression or complement inactive proteins to treat pathologies with recognized genetic targets.[223] These therapies require delivery of RNA into target cells while avoiding immune responses. A major conventional approach is viral delivery vectors, but viral vectors suffer from recognized limitations, such as immunogenicity [224] and pre-existing immunity [225, 226], as well as the related difficulty of re-dosing due to an adaptive immune response. Some vectors (e.g., lentiviral) also pose a hazard of genomic integration [227, 228], and viral vectors based on icosahedral geometries have a hard physical constraint on the amount of genetic material that can be delivered [229]. In practice, production of viral vectors is typically complicated and expensive at scale.[230] Payload size can be a hard constraint for viral capsids (~100 nm), which may accommodate gene cassettes up to 8–9 kb, posing challenges to deliver two or more transgenes.[231] Formulation of nucleic acids with ANCs, particularly LNPs, is a promising alternative to viral vectors.[232, 233] LNPs represent a flexible platform avoiding several of the challenges associated with viral vectors, as underscored by early examples of the mRNA vaccines for COVID-19 as well as the first RNAi therapy to be approved by the FDA (patisiran, treating hereditary transthyretin amyloidosis).[234] However, without targeting, systemically administered LNPs accumulate in the liver and are rapidly cleared [133], and thus antibody targeting is particularly important.

6.1. CRISPR/Cas9

Genome editing by the CRISPR/Cas9 system leverages a synthetic single-guide RNA (sgRNA) that directs the CRISPR-associated protein 9 (Cas9) endonuclease to specific DNA sequences. [235, 236] Downstream recombination with an exogenously supplied DNA template, or endogenous repair of the dsDNA break, leads to genomic changes. Clinical translation of CRISPR/Cas9 requires efficient and specific delivery of the gene-editing components (sgRNA, Cas9, and possibly the template DNA) to the correct target cells, particularly for in vivo applications. Furthermore, the possibility of off-target editing, which could inadvertently lead to genomic instability or malignancy, underscores the importance of precision delivery systems. Antibody-conjugated nanomaterials, particularly Ab-LNPs, have been explored as a promising solution to these challenges. For example, cationic LNPs were engineered to deliver Cas9 mRNA and sgRNAs for editing PLK1, a kinase required for mitosis, to glioblastoma cells via intracerebral injection.[237] For systemic treatments, the LNPs were targeted using anti-EGFR antibodies, which resulted in selective uptake into ovarian tumors. This approach achieved 70–80% gene editing, accompanied by increased survival, for in vivo models. Thus, ANC delivery approaches have the potential to impart high editing efficiency and specificity and possibly improve the safety profile of CRISPR/Cas9-based therapies. An important application is CAR T-cell therapy.

6.2. CAR T-Cell Therapy

Chimeric antigen receptor (CAR) T cell therapy generally involves genetic engineering, including genome editing, or mRNA delivery to T cells to express an altered receptor that redirects them to kill a targeted cell population, usually cancer cells [Figure 7a]. This approach is achieving remarkable clinical success in the treatment of liquid tumors.[238] Currently, CAR T cell therapy involves isolation of T cells from the patient, which are then genetically modified by ex vivo viral transduction to introduce the CAR, followed by selection and stimulation before reintroducing of the transformed cells into the patient. This individualized process contributes to very high procedure and production costs as well as long-term side effects due to genetic transformation and lack of targeting specificity [239], in addition to other limitations of viral vectors. Alternatively, LNP ANCs can be administered parenterally to achieve good efficacy in vivo with lower immunogenicity due to absence of viral proteins and reduced off target toxicity. In one case, conjugating anti-CD4 antibodies to LNPs enabled in situ mRNA delivery to CD4+ T cells after intravenous administration.[135] Versatility in targeting broad T cell populations was also demonstrated by incorporating anti-CD3, anti-CD5, or anti-CD7 antibodies with LNPs [240] to deliver a CD19 CAR (B-cell-specific) mRNA. In vivo studies showed up to 90% B cell depletion, depending on antibody combination and effective dose. Similarly, intravenous injection of anti-CD3-targeted LNPs delivered mCherry mRNA to Jurkat T cells in situ [241], causing T cell activation and migration. Likewise, another CD3-targeting LNP was designed to deliver a plasmid containing genes for interleukin 6 short hairpin RNA (IL-6 shRNA) and CD19-CAR (AntiCD3-LNP/CAR19 + shIL6).[242] The LNPs transformed CD3+ T cells into CAR T-cells targeting CD19, a leukemia marker, and simultaneously conferred IL-6 knockdown. Thus, the resulting transformants killed CD19-expressing leukemia cells while reducing cytokine release syndrome caused by IL-6. In vivo experiments demonstrated that this system was effective, showing antitumor effects comparable to traditional ex vivo CAR T-cell therapy. Multifunctional delivery was enabled by the flexibility of the LNP platform. Thus, LNP ANCs can potentially simplify CAR T-cell therapy by allowing direct in vivo rather than ex vivo treatment.

Figure 7. CAR T cell therapy and nanoparticle bispecific T-cell engager (NBiTE) T cell therapy.

(a) Production of CAR T cells begins with the collection of immune cells from the patient or donor through leukapheresis. T cells are isolated from other blood components and then activated and expanded. Gene transfer (e.g., by viral vector) introduces the chimeric antigen receptor genes (CARs) to the T cells. The modified T cells are further expanded, harvested and infused into the patient. (b) An NBiTE contains two distinct scFvs attached to the nanoparticle surface: one that binds to a T cell-specific antigen (i.e., T cell receptor) and another that targets a tumor-specific antigen, enabling cytotoxicity. Multivalent interactions at the nanoparticle-cell interface can increase efficiency in bridging T cells to tumor cells compared to molecular BiTEs.

Aside from nucleic acid delivery, an important step in CAR T-cell therapy is activation of the transformed T cells. During ex vivo therapy, activation is usually performed by magnetic beads coated with anti-CD3 and anti-CD28 antibodies, which together co-stimulate the T cells in a separate step from transfection. To simplify this process, LNP ANCs containing mRNA were designed with surface-conjugated human anti-CD3 and anti-CD28 antibody fragments, permitting simultaneous transfection and activation.[243] These combination LNPs facilitated rapid, one-step activation and transfection with mRNA, and the resulting CAR T cells decreased tumor burden in a murine xenograft model. Thus, LNP ANCs may advance CAR T-cell therapy through multiple mechanisms.

6.3. Modular Design for Targeting Nucleic Acid LNPs

To accelerate the development of LNP ANCs, a modular platform has been proposed in which a membrane-anchored lipoprotein embedded in the RNA-containing LNP binds to the antibody Fc domain (Figure 8). Then, various antibodies with different variable domains can be attached to the LNP via the affinity for the Fc domain. This approach (Anchored Secondary scFv Enabling Targeting (ASSET)) [244] was demonstrated using an anti-Ly6c mAb to deliver IL10 mRNA to inflammatory leukocytes, for treating inflammatory bowel disease.[245] Depending on the antibody used, ASSET can be targeted to specific receptor conformations.[246] In one case, the high-affinity conformation of α₄β₇ integrin was targeted by fusing domains D1 and D2 of the natural ligand of α₄β₇ integrin, MAdCAM-1, to an IgG–Fc conjugated LNP. The LNPs were loaded with siRNA for interferon-γ, and treatment resulted in an improved therapeutic outcome in experimental colitis.

Figure 8. Modular platform for anchoring antibodies to LNPs for gene editing.

An LNP containing nucleic acids, such as Cas9 mRNA and sgRNA, is nano-assembled using a microfluidic mixer. Then micelles functionalized with an Fc-binding protein (green) are fused to the LNPs. LNP targeting is determined by the IgG added (pink) for a particular application.

7. Emerging therapeutic applications for ANCs

7.1. Nanomaterial Bispecific T-cell Engagers

Targeted therapies rely on the presence of a cell surface receptor [247], so tumor cells may evade T-cell-mediated killing by altering receptor interactions, including reducing the expression of major histocompatibility complex (MHC) class I antigens or other critical surface markers.[248] One approach to overcome antigen escape uses bispecific T-cell engagers (BiTEs), proteins that essentially bridge T cells and tumor cells.[249–251] Notably, BiTE-mediated tumor cell killing is independent of MHC I, as it does not require antigen presentation. Also, BiTE strategies take place in situ and avoid the need for ex vivo engineering of T-cell receptors.[252, 253] Despite their potential, current challenges with BiTEs include short circulation times [251, 254] and inadequate binding between the BiTE and the T-cell surface [255], requiring multiple BiTEs to stabilize a T-cell / tumor cell pair.[256] In this context, nanomaterial-based BiTEs (NBiTEs) [Figure 7b] offer a platform for modulating structure and morphology as well as surface patterning, e.g., to induce multivalent coupling from antibody targeting.[257] [256], modulating the binding among T cells, tumor cells, and BiTEs. Furthermore, the nanomaterial of an NBiTE could be designed to deliver molecules, such as interleukin-2 (IL-2), to promote T-cell activation and enhance the antitumor effect.[258]

Successful NBiTE applications have been reported in preclinical models. For instance, PEGylated liposomes conjugated with both trastuzumab (anti-Her2 mAb) and rituximab (anti-CD20, B-cell-targeting mAb) exhibited a 25-fold increase in antibody potency in cell viability assays [259], with in vivo studies showing enhanced tumor inhibition compared to free antibodies. Likewise, a 30 nm polystyrene-based NBiTE with surface-bound antibodies targeting Her2 and calreticulin displayed significantly stronger binding compared to free anti-Her2 antibodies, attributable to their multivalent interactions.[260] This NBiTE stimulated phagocytosis of cancer cells, as well as enhanced antigen processing and presentation due to calreticulin interaction promoting engagement with antigen-presenting cells.

Modulation of the nanomaterial in NBiTEs can be used to prolong circulation time by surface functionalization.[261, 262] One approach involved engineering a HEK293 cell line to overexpress scFvs for CD3 and epidermal growth factor receptor (EGFR), for binding T cells and EGFR-expressing breast cancer cells, respectively.[263] This cell line was used to produce exosomes, which then showed the ability to promote interaction between T cells and breast cancer cells, resulting in antitumor activity. Overall, these studies highlight the potential of NBiTEs for in situ cancer immunotherapy.

7.2. Rheumatoid Arthritis

Rheumatoid arthritis (RA) is a chronic autoimmune disorder whose treatment includes small molecules as well as mAbs. [264–267] The non-targeted drugs are distributed systemically, resulting in undesirable side effects. ANCs may address this problem. For example, PLGA nanoparticles conjugated to anti-CD64 mAb (targeting macrophages at sites of inflammation) were loaded with the anti-rheumatic drug methotrexate as well as smaller superparamagnetic iron oxide nanoparticles.[268, 269] In another approach, AuNPs were used to deliver hyaluronate and Tocilizumab, two drugs used to treat RA.[270] This nanoparticle treatment significantly reduced signs of disease in a collagen-induced arthritis mouse model. Similarly, anionic carbosilane dendrimers (consisting of C-C and C-Si bonds) conjugated to infliximab were designed to target TNF-α, a pro-inflammatory cytokine mediator of RA.[271] Another TNF-α-targeted ANC was designed using poly(amidoamine) dendrimers functionalized with chondroitin sulphate (CS) and anti-TNFα mAbs.[272] While these approaches are still in the research stage, ANCs are poised to advance therapeutic options of RA.

7.3. Inflammatory Bowel Disease

Inflammatory bowel diseases represent another family of chronic autoimmune disorders, including Crohn’s disease and ulcerative colitis. Treatment strategies include anti-TNF-α mAbs like infliximab (IFX), cetrolizumab and adalimumab (ADA).[273] Despite their efficacy, repeated systemic administration of these biologics can lead to significant adverse effects, including systemic immune suppression and the development of anti-drug antibodies, which diminish therapeutic effectiveness over time.[274, 275] To address these challenges, recent research has focused on developing targeted nanomaterial-based delivery systems that can deliver anti-TNF-α antibodies directly to the inflamed tissues of the GI tract, reducing systemic exposure and enhancing therapeutic efficacy. In these cases, oral delivery provides anatomical targeting to the gut mucosa, and thus the main challenge for NP design is intact delivery of a therapeutic mAb, rather than mAb targeting of the NP. In one design, chitosan nanoparticles complexed with IFX were embedded into alginate microparticles for oral delivery of IFX.[276] The alginate microparticles shielded the antibody from the harsh conditions of the GI tract, enabling it to reach the lower GI tract. This delivery system significantly alleviated colitis in a mouse model of acute ulcerative colitis, making it a promising candidate for orally bioavailable IBD therapy. Alternatively, enteric-coated liposomes have been developed to deliver IFX specifically to the colon [277], showing significant anti-inflammatory effect and reduced TNF-α levels in a mouse model of induced colitis. This system could be a promising strategy for transporting biologics to the GI tract. A polymer carrier, polyesterurethane (PU), for IFX has also been explored for treating colitis [278], showing improvement in an animal model, aided in part by enhanced permeation across the inflamed mucosa. Another possible formulation is based on tannic acid and PEG polymers to create supramolecular NPs for oral delivery of IFX.[279]

In addition to IFX delivery, polymer nanoparticles delivering ADA are also under active study. PLGA NPs conjugated with ADA [280] demonstrate enhanced stability against proteolytic degradation and improved therapeutic efficacy compared to ADA in solution, particularly in reducing the severity of experimental colitis in mice. Dense packing of the antibody on the nanoparticle surface appeared to be protective against degradation while maintaining bioactivity. Similarly, PLGA-PEG nanoparticles with two different PEG lengths (2 kDa, and 5 kDa) were tested as oral anti-TNF-α mAb delivery systems.[281] While both of them significantly reduced TNF-α secretion in vitro, only the 2 kDa version was able to alleviate experimental acute colitis in mice, indicating the importance of chain length for in vivo efficacy of the polymeric NPs. Overall, the ability of nano- and micro-materials to protect biologics for delivery to the lower GI tract after oral administration indicates potential for ANCs to impact this class of diseases.

7.4. Light-activated Therapies

Photodynamic therapy (PDT) and photothermal therapy (PTT) are emerging minimally invasive treatments that employ light-activated photosensitizers to generate reactive oxygen species (ROS) or heat, respectively, which selectively destroy targeted cells, such as cancerous tissues. The integration of ANCs with PDT and PTT offers significant advancements in directing these therapies with higher precision and efficiency while minimizing damage to surrounding healthy tissues. For example, AuNPs conjugated with Rituximab, a mAb that targets CD20-positive B-cells, were developed for the targeted treatment of lymphomas and leukemias.[282] These ANCs demonstrated a dual therapeutic effect: the AuNPs served as drug carriers and mediators of phototherapy, while Rituximab functioned as targeting agent and an activator of complement-dependent cytotoxicity (CDC). Upon exposure to resonant femtosecond laser pulses, Rituximab was released from the nanospheres at a controlled rate. Cell necrosis resulted from two independent pathways – direct rupture of the plasma membrane by photothermal and photochemical effect and CDC. Release was precisely controlled by laser exposure. In addition to noble metals, metal oxide nanoparticles are also used for PTT. For example, molybdenum oxide-based NPs were conjugated to a mAb that targets carbonic anhydrase IX, a hypoxia marker prevalent in many solid tumors.[283] These ANCs were formulated to be effectively two-dimensional, adding greater surface area compared to standard nanoparticles, and were effective in photothermal yield in cell culture. Beyond cancer, PTT using ANCs is also being explored for the treatment of bacterial infections. For example, AuNPs conjugated with appropriately targeted antibodies can selectively target and kill bacterial pathogens such as Pseudomonas aeruginosa upon laser irradiation.[284, 285] Specific binding of the ANCs brought the AuNPs to the bacteria, and near-IR laser irradiation caused bacterial cell death in a dose-dependent manner, due to the photothermal effect. These approaches offer an important alternative to antibiotics as the problem of antimicrobial resistance grows.

While PDT and PTT are promising approaches, the limited penetration depth of the SPR wavelength (e.g., AuNPs absorbing in the visible range) constrains their use for deep tissue. A possible solution is to significantly lengthen the lifetime of light emission to allow particles to be excited before administration.[286] In one case, nanoparticles were synthesized containing Purpurin 18 (Pu18) as a photosensitizer for PDT and source of persistent luminescence, lanthanide-doped upconversion nanoparticles for bioimaging and enhanced photostability, and the Her2-targeting antibody trastuzumab. The nanoparticles are activated by near-IR irradiation (980 nm) before intravenous administration, resulting in ROS generation and luminescence by Pu18. These nanoparticles thus enabled bioimaging and PDT of deep tumors. Another strategy to improve the effectiveness of PDT and PTT against deep tumors is photodynamic priming (PDP). In PDP, photo-activation of ANCs temporarily alters the permeability of the tumor vasculature, cell membranes, and extracellular matrix, thereby increasing sensitivity to subsequent therapeutic agents (often co-delivered with ANCs), such as chemotherapeutic drugs, even at sub-curative doses. An example is photoactivatable liposomes targeting the PD-L1 immune checkpoint in pancreatic ductal adenocarcinoma (PDAC).[287] Upon photoactivation (at 690 nm), these liposomes disrupt tumor stromal collagen, enhancing tumor penetration and facilitating self-delivery into deeper tumor regions.

For neurological diseases, much attention has focused on the potential of optogenetics to convert light exposure to calcium gating. In general, optogenetics requires expression of engineered opsin proteins in the target neurons, which represents a technical barrier for application of the technique. The photothermal effect of certain nanoparticles offers an alternative to optogenetics that do not require genetic modification. For example, semiconducting polymer nanoparticles can be functionalized with anti-transient receptor potential cation channel subfamily V member 1 (anti-TRPV1) antibodies.[288] TRPV1 is a temperature-sensitive ion channel. When the antibody-targeted nanoparticles are bound to TRPV1 and heated by near-IR exposure, the TRPV1 channels to gate the intracellular calcium influx of the neural cells. In neuronal cell culture, this approach enabled the photo-activation of neurons within milliseconds. Combined with BBB-crossing properties, ANCs are a promising approach for controlling neuronal stimulation.

8. Pharmacokinetics and Biodistribution of ANCs

After administration, ANCs can exist in three primary forms in systemic circulation: intact ANCs, free antibodies, and dissociated nanomaterials. These forms dynamically change based on target binding, elimination, and deconjugation, posing a challenge for developing unified pharmacokinetic (PK) profiles due to their distinct internalization, recycling, and clearance mechanisms. Nevertheless, a meta-analysis of 161 studies (2009–2021) revealed that ANC pharmacokinetics are largely dictated by nanoparticle properties such as size, charge, and material composition.[289] For example, biodistribution appears to be driven by NP properties, as analysis revealed no significant differences between antibody-conjugated and naked NPs regarding distribution to the liver, spleen, or kidneys. Size was important for biodistribution, as larger complexes accumulated in the liver and smaller ones in the kidneys, while spleen distribution was relatively unaffected by size. Smaller NP conjugates consistently exhibited improved tumor penetration across material types. NP charge was also found to be significant, and negatively charged NPs were preferred to reduce passive internalization and enhance antigen specificity.

Although NP physicochemical properties predominantly determine PK properties of ANCs, the antibodies also exert an effect. Although antibody moieties (complete mAbs or fragments) minimally affect circulation half-life and biodistribution to different organs, they do enhance tumor uptake significantly compared to non-targeted nanoparticles (7.9 ± 1.9% ID g−1 vs. 3.2 ± 0.6% ID g−1). Tumor uptake of different nanoparticle types is promoted by different antibody types. In particular, lipidic NPs demonstrated superior tumor uptake when conjugated to antibody fragments, while polymeric or inorganic NPs achieved higher uptake with full mAbs. Full mAbs introduce greater PK variability compared to antibody fragments. For example, half-life variability was more pronounced for full mAb-NPs compared to fragment-NPs, especially for antibodies of chimeric species origin (–ximab, e.g., human and mouse) compared to humanized antibodies (–zumab). This variability may arise from individual antibody pharmacokinetics, improper FcRn binding in murine models, or inconsistencies in conjugation methods. Conjugation type also influences variability, as amine-based linker strategies yielded more consistent half-life profiles compared to thiol-based linkers.

9. Biosafety

An important biosafety concern is the cytotoxicity of NPs, which is influenced by factors such as size, shape, surface charge, and hydrophobicity [see Figure 2 of this review; Ref. [58]]. To mitigate toxicity, biocompatible materials and appropriate surface modifications can be employed [290, 291]. Stability under physiological conditions is another critical factor, to prevent premature release of therapeutic agents in addition to preserving functionality [93]. Toward this end, ANCs are often coated with relatively inert polymers like polyethylene glycol (PEG), to increase stability and reduce opsonization.

An emerging aspect significantly affecting the biosafety of ANCs is the conjugation chemistry used to attach antibodies to nanoparticles. The choice of conjugation chemistry has been shown to affect activation of the complement cascade of plasma proteins, which can alter nanoparticle biodistribution and induce toxicity [292]. For example, conjugation using DBCO caused aggregation of antibodies on the nanoparticle surface, leading to complement activation. In contrast, thiol-maleimide chemistry activated the complement system through a different mechanism involving maleimide-reacted albumin aggregation in plasma. These processes resulted in enhanced nanoparticle uptake by lung phagocytes—up to 140-fold—and adverse effects, such as a 50% reduction in platelet count. Thus, the conjugation strategy must consider immunological factors such as complement activation to mitigate unwanted effects, paving the way for safer and more effective ANC applications.

10. Challenges in Clinical Translation of ANCs

Significant advancements have been made in the preclinical development of ANCs for therapeutic applications, although there is currently a translational gap between preclinical successes and clinical trials. Although there are currently no FDA-approved drugs, a small number of agents are in clinical trials (Table 1). Challenges in clinical trials can stem from difficulties with NP penetration of tumor vasculature to reach tumor cells, short circulation half-life, and/or immunogenic reactions. Other challenges include formulation stability, biocompatibility, and regulatory issues. Given the novelty of ANCs, both preclinical and clinical research is needed to address these challenges.

Table 1.

Clinical trials of ANCs^a)

NP-based Drug	Alternative Drugs	Condition	Stage	State	Identifier
Epidermal Antigen Specific AuNP	Proinsulin Derived Peptide	Type 1 Diabetes	Phase I	Completed	NCT02837094
Anti-EGFR immunoliposome	Doxorubicin	Solid Tumor	Phase I	Completed	NCT01702129
Magnetic NPs coated with anti-EpCAM or anti-CD52 Abs	-	Leukemia	NA	Completed	NCT04290923
EGFR-targeting nanocells	Doxorubicin	Glioblastoma, Astrocytoma	Phase I	Unknown	NCT02766699

^a)From clinicaltrials.gov, search terms: ‘antibody nanoparticle conjugate’, ‘antibody conjugated nanomaterials’

10.1. Reproducibility and Controls in Synthesis.

Traditional bulk synthesis techniques often result in NPs with high polydispersity, and batch-to-batch variability can be introduced by subtle environmental factors due to the sensitivity of the seeding and growth processes. In addition to standardization of methods, microfluidic technologies may enable high-speed self-assembly with narrower size distribution and improved batch-to-batch reproducibility.[293–295] In addition, an alternative to traditional ‘bottom-up’ self-assembly and seeded syntheses are ‘top-down’ approaches analogous to lithography used in hard materials. For example, Particle Replication in Non-wetting Templates (PRINT) technology [296, 297] allows precise control over NP size, shape, and surface properties. Technologies such as these could improve reproducibility and reduce polydispersity to help advance ANC therapeutics into development.[298, 299]

10.2. Evaluation of In Vitro Efficacy and Screening.

In vitro evaluation is essential to assess biocompatibility of ANCs and determine ANC-cell interactions before animal testing. While significant progress has been made in understanding factors like immune evasion, tumor extravasation, and drug release, systematic screening of NP properties remains challenging.[300, 301] Conventional in vitro models typically use cell culture in multiwell plates, but these may not capture the complexity of physiological barriers and systems. Recent developments such as biomimetic “organ-on-a-chip” (or, “tumor-on-a-chip”) systems, [302, 303] can offer more realistic simulations of ANC behavior in physiological conditions. Integrating tumor-like spheroids into microfluidic channels can provide insights into NP accumulation and diffusion, potentially bridging the gap between in vitro and in vivo studies to maximize the probability of successful in vivo studies.

10.3. Assessment of In Vivo Performance.

In vivo studies are necessary for evaluating pharmacokinetics (PK), biodistribution, efficacy, and safety.[304–306] While some studies have shown PK scaling across species, discrepancies often arise between preclinical and clinical outcomes, due to, for example, the limitations of current tumor models in replicating human cancers.[307, 308] Available models, including cell line-based xenografts, patient-derived xenografts (PDXs, i.e., direct implantation of patient cancer tissue specimens into immunodeficient mice), and genetically engineered mouse models (GEMMs), each have limitations. For instance, PDXs rely on mice lacking a functional immune system and a proper tumor microenvironment, which diminishes their ability to fully replicate human tumor-stroma interactions.[309] Differences between murine and human fibroblasts also reduce their accuracy.[310] Furthermore, PDX development is time-consuming, costly, and challenging.[311] On the other hand, usefulness of GEMMs depends on unpredictable factors like the accuracy and fidelity of genetic mutations, and disease progression kinetics.[312] Moreover, some drugs, particularly human-specific monoclonal antibodies, cannot be assessed in GEMMs.[313] Models that more accurately mimic biologic and molecular heterogeneity of human diseases, (e.g., PDX model recapitulating the dynamics, histopathology and genome architecture of human prostate cancer [314], humanized mouse models featuring functional human myelomonocytic cells with subset diversity and numbers similar to humans [315], and prostate adenocarcinoma GEMMs produced by MYC activation and PTEN loss, phenocopying key features of human disease, including widespread metastasis to lymph nodes, liver, and lung [316]) could enhance the predictive value of preclinical studies.

10.4. Scalability and Manufacturing.

Transitioning from preclinical to clinical development presents significant challenges in chemistry, manufacturing, and controls (CMC). While scale-up of simple NPs is feasible with existing pharmaceutical manufacturing techniques, more complex nanomedicines, including ANCs, face additional challenges. For instance, integrating biological targeting ligands or combining multiple therapeutics may require novel manufacturing processes. Technologies like PRINT may offer reproducible NP fabrication, although scale-up is still under development. Alternative emerging technologies, like coaxial turbulent jet mixers, may promise high-throughput production of polymeric NPs.[317, 318] Despite current reliance on bulk synthesis, these advanced methods could accelerate the clinical translation of NPs. Aside from NP synthesis, another major challenge in the clinical translation of ANCs is the large-scale synthesis of targeting antibodies. While therapeutic mAbs have been in clinical use for decades, the intensive purification processes and the limited stability of many antibodies still represent a limitation and high cost for these biomolecules. For ANCs, the nanoparticle formulation and linkers would also be optimized for each antibody, increasing the customization needed for manufacturing.

11. Potential advantages of ANCs

ANCs may be logically positioned as next-generation successors of ADCs, given their shared mechanism to target therapeutic action. ADCs rapidly gained success as a drug class, with 14 FDA-approved therapies and > 100 under clinical investigation. Nevertheless, there is room for improvement in overcoming the inherent limitation of low drug-to-antibody ratios (DAR), as well as limited linker chemistries and off-target toxicities of ADCs. By leveraging the benefits of nanocarriers, ANCs address many of these challenges [319]. Although more research is required, in principle, ANCs possess multiple fundamental advantages compared to ADCs.

First, ANCs have the potential for orders-of-magnitude greater drug-loading capacity. ADCs typically have a DAR ranging between 4–8, limiting the amount of drug that can be delivered per biomolecular interaction. The low DARs necessitate the use of highly potent drugs that can lead to toxic side effects. In contrast, ANCs can achieve exceptionally high DARs (10² – 10⁵) depending on nanoparticle size and drug solubility, which should correspondingly enhance therapeutic efficacy and reduce toxicity through enhanced potency.

Second, ANCs are a more flexible platform for conjugation chemistries, allowing broader drug compatibility. The design of ADCs is significantly constrained by the chemical compatibility, stability and degradability requirements of the antibody-drug linker, limiting the types of drugs that can be effectively conjugated. In contrast, ANCs introduce a wide variety of chemical moieties that could be incorporated into the nanomaterial, and can also use non-covalent encapsulation or attachment. This chemical flexibility broadens the potential spectrum of drug cargo.

Third, ANCs can benefit from intrinsic nanomaterial properties. The nanomaterial in ANCs can act as a therapeutic agent itself (e.g., gold nanoparticles). In addition, the nanomaterial core of ANCs provides mechanical and chemical stability under physiological conditions. At the same time, stimuli-responsive properties of the nanomaterial can provide additional options for triggered drug release, such as in response to specific tumor microenvironment changes in pH or enzyme activity, delivering drugs more effectively to target sites due to the high DARs. Finally, the EPR effect of nanoparticles may improve accumulation at tumor sites, helping to address rapid clearance issues often encountered with ADCs.

12. Conclusion

Despite the current challenges, ANCs represent an expanding forefront for therapeutics, with particular strengths in combining precise targeting of antibodies with multifunctionality, versatility, and stability advantages offered by nanomaterials. Starting in the 1980s, therapeutic antibodies, and biologics more broadly, have revolutionized medicine on an accelerating trajectory. Indeed, drug expenditures in the U.S. on biologics increased by 55% between 2017–2022, while expenditures on small molecule drugs remained essentially constant.[320] Of the top ten best-selling drugs of 2023, eight are biologics.[321] Of these, six are mAbs, one is a peptide, and one is a nanoparticle (mRNA LNP vaccine). For ANCs, we are currently witnessing advancing research and efficacy reminiscent of the early stages of biologics development, positioning ANCs as a growth area for next-generation therapeutic approaches. While the greater structural and functional intricacy of ANCs compared to other established drug classes adds complexity to translational research and the development pipeline, the approval of liposomal and protein nanoparticle therapies, in addition to mAbs and antibody-drug conjugates, has paved the way for ANCs. ANCs are potentially a next-generation targeted drug delivery platform that overcomes some primary limitations of ADCs while retaining key advantages. With high drug-loading capacity, expanded conjugation potential, and advantageous nanomaterial properties, ANCs may be a logical successor technology to ADCs, which themselves represented a revolutionary approach.

In order to fulfill the potential of ANCs, several areas require further research. A major knowledge gap exists currently in understanding interactions between nanoparticles and biological systems, particularly biocompatibility, including reducing immunogenicity, avoiding accumulation of inorganic components, and crossing into hard-to-reach compartments such as the brain. These challenges may be met by approaches such as integrating biocompatible coatings or enhancing delivery efficiency to reduce overall required dosages, and developing preclinical models that more closely mimic human PK and PD. Related to these challenges, the effects on human physiology, as well as the downstream environmental effects of nanoparticles, should be studied to understand the long-term ramifications of ANC therapies for public health. On the design side, ANCs, like many areas of biomedicine, are ripe for meta-analysis by machine learning, to gain a systematic and predictive understanding of how variables such as size, charge, surface chemistry, and other factors influence ANC behavior, particularly in the body. Accurate modeling could enable improved and more complicated designs, especially for multifunctional systems (e.g., theranostic applications). Further in development, advances are needed for scalable manufacturing of the conjugates, particularly for decreasing heterogeneity. The interdisciplinary collaboration between materials science, biomolecular engineering, and clinical medicine will be pivotal in advancing the field. While ANCs are still in relatively early stages of development, now bolstered by a significant body of research, we anticipate that ANCs will become a new therapeutic modality in the not-distant future.

Biographies

Aniruddha Adhikari is a postdoctoral researcher in the Chen lab at the University of California, Los Angeles. He obtained his PhD in Biochemistry from S. N. Bose National Centre for Basic Sciences, India, after completing his Master’s degree from the Dept. of Biochemistry, University of Calcutta, India. His research focuses on nanohybrid drug delivery systems and nanomedicines for treating chronic diseases where redox modulation plays a crucial role. He is exploring bacteriophage-mediated nano-therapy and stimuli-responsive antibody-conjugated nanomaterials.

Irene A. Chen is a Professor in the Department of Chemical and Biomolecular Engineering and Department of Chemistry and Biochemistry at the University of California, Los Angeles. She obtained a PhD in Biophysics from Harvard and an MD in Health Sciences and Technology from Harvard/MIT. Her laboratory studies nanoscale biomolecular systems to understand their fundamental properties and address emerging challenges in biotechnology and infectious disease. The Chen lab focuses on the design and evolution of simple synthetic cells and bacteriophage-based technologies. https://research.seas.ucla.edu/irene-chen/