Fc‐FcγR interactions during infections: From neutralizing antibodies to antibody‐dependent enhancement

Summary

Advances in antibody technologies have resulted in the development of potent antibody‐based therapeutics with proven clinical efficacy against infectious diseases. Several monoclonal antibodies (mAbs), mainly against viruses such as SARS‐CoV‐2, HIV‐1, Ebola virus, influenza virus, and hepatitis B virus, are currently undergoing clinical testing or are already in use. Although these mAbs exhibit potent neutralizing activity that effectively blocks host cell infection, their antiviral activity results not only from Fab‐mediated virus neutralization, but also from the protective effector functions mediated through the interaction of their Fc domains with Fcγ receptors (FcγRs) on effector leukocytes. Fc‐FcγR interactions confer pleiotropic protective activities, including the clearance of opsonized virions and infected cells, as well as the induction of antiviral T‐cell responses. However, excessive or inappropriate activation of specific FcγR pathways can lead to disease enhancement and exacerbated pathology, as seen in the context of dengue virus infections. A comprehensive understanding of the diversity of Fc effector functions during infection has guided the development of engineered antiviral antibodies optimized for maximal effector activity, as well as the design of targeted therapeutic approaches to prevent antibody‐dependent enhancement of disease.

1. INTRODUCTION

Antibody responses induced upon infection or vaccination are a critical component of protection against infectious diseases. The binding of antibodies to epitopes on the surface of a pathogen or infected cell can lead to neutralization, physically inhibiting cell entry or fusion, as well as induction of downstream effector cell responses. These effector functions are largely driven by Fc domain engagement with Fcγ receptors (FcγRs), a family of receptors differentially expressed across leukocytes. The diverse heterogeneity of the Fc structure and its corresponding binding affinities for FcγRs has profound immunomodulatory consequences, resulting in comprehensive antiviral protection and induction of long‐term immunity. However, in certain viral contexts, namely dengue virus, antiviral antibodies can instead play a pathogenic role and contribute to enhanced disease; a phenomenon termed as antibody‐dependent enhancement (ADE). In recent years, following the success of monoclonal antibody (mAb) therapeutics for the treatment of cancer and autoimmunity, there has been increased interest in the development of mAbs to confer broad protection against multiple viral infections. Indeed, the COVID‐19 pandemic greatly stimulated efforts to understand the mechanisms of antibody‐mediated protection and to design potent neutralizing mAbs, resulting in multiple anti‐SARS‐CoV‐2 mAbs being approved for emergency use. Importantly, increasing evidence has demonstrated that optimal antiviral protection by mAbs goes beyond direct neutralization, and requires the activation of FcγR‐mediated immune functions. These findings have led to novel approaches to engineer mAbs that can preferentially engage FcγRs for maximal protection. Although evidence of ADE in viral contexts beyond dengue virus remains limited in biologically relevant in vivo models or clinical studies, concerns about ADE have hindered the development of optimized mAbs. In this review, we discuss the current understanding of the mechanisms of antibody‐mediated neutralization and coupled FcγR‐dependent antiviral functions, with a focus on Ebola virus, influenza virus, and SARS‐CoV‐2. Additionally, we consider the experimental and clinical evidence of ADE across multiple infectious agents, as well as the impact of ADE concerns on the development of antiviral therapeutics.

2. IgG ANTIBODY STRUCTURE

Most antiviral human mAbs investigated to date are immunoglobin G (IgG), the most abundant immunoglobin among the five isotypes (IgA, IgD, IgE, IgG, and IgM). ¹ , ² IgG has been the preferred isotype for therapeutic mAbs, due to its critical involvement in linking innate and adaptive immune responses, as well as its ease of production and favorable in vivo pharmacokinetics. Structurally, the IgG molecule consists of two functional components connected by a flexible hinge region: two identical Fabs (fragment antibody binding) domains and one Fc (fragment crystallizable) domain. Each Fab domain contains the variable regions of one heavy and one light chain, which mediate the highly specific recognition and binding to antigens. Somatic recombination and hypermutation of the variable regions allow for an extremely diverse repertoire of antibodies with extensive pathogen recognition. However, the induction of innate and adaptive immune responses upon antigen binding is dependent on the combined function with the Fc domain. The Fc mediates engagement with a family of receptors called Fc receptors (FcγRs), as well as with FcRn and the C1q subunit of the complement system. Antigen–antibody immune complexes (ICs) cross‐link membrane‐bound FcγRs to initiate downstream activating or inhibitory signals depending on the cell type and the FcγR engaged, resulting in pleiotropic immunomodulatory effector functions.

The diverse cellular responses induced by various Fc‐FcγR interactions are tightly regulated by the heterogeneity of the Fc domain, which arises from differences among the four subclasses of human IgG (IgG1‐4) and from the composition of the Fc‐associated glycan ³ (Figure 1A‐B). The Fc comprises of two homodimers of the heavy chain constant domains, CH2 and CH3, with the two CH3 domains being closely associated and the CH2 domains forming a horse‐shoe shape, separated by an internal gap. ⁴ Among the four IgG subclasses, IgG1 is the most abundant, followed by IgG2, IgG3, and IgG4 in descending order. ⁵ Although the subclasses are over 90% homologous, they differ in the primary amino acid sequence predominately within the hinge region and N‐terminal CH2 domain where the FcγR binding interface is mapped. ⁶ Such differences affect the affinity of the Fc domain for the various FcγRs, as well as the length and flexibility of the hinge region, partially explaining the unique binding profile that each IgG subclass has with the different FcγR subtypes. ⁶

FIGURE 1.

Overview of IgG Fc domain and FcγR determinants that regulate the induction of diverse cellular effector functions. (A). The structural differences among the four IgG subclasses (IgG1‐4) impact their affinity for the various FcγR subtypes and differ in their abundance in human serum. Type I FcγRs are categorized as either activating (red) or inhibitory (blue), based on the presence of a cytoplasmic ITAM or ITIM motif(s), respectively. The relative binding affinities of the IgG subclasses for the FcγR subtypes, and SNPs of FcγRIIa and FcγRIIIa, are summarized. (B) The core biantennary N‐linked glycan at the N297 site on the IgG Fc (shaded gray) can be modified by additional fucose, sialic acid, galactose, and bisecting N‐acetylglucosamine saccharides. These four posttranslational modifications, with or without the addition of other saccharides (light gray), alter the IgG affinity for specific Type I and Type II FcγRs. (C) FcγRI, the high‐affinity receptor, is unique in its ability to bind monomeric antigen‐bound IgG. Multimeric IgGs, when complexed with antigens or opsonized cells, exhibit increased affinity for all FcγRs, leading to FcγR cross‐linking and subsequent activation of downstream signaling pathways. (D) Engagement of immune complexes (ICs) with either activating FcγRs or the inhibitory FcγRIIb leads to opposing signaling cascades, having either pro‐inflammatory or anti‐inflammatory effects. FcγRIIb signaling directly inhibits activating signals, preventing inappropriate effector cell activation. (E) Multiple FcγR subtypes are often co‐expressed on leukocytes and their expression levels and patterns can change due to cytokine and chemokine signals released during infection or inflammation. Cytokines such as IFN‐ γ increase activating FcγR expression, while cytokines such as IL‐4 upregulate FcγRIIb expression and reduce activating FcγR levels. (F) It is proposed that Fab epitope specificity and the resulting quaternary conformation of the Fc domain may also contribute to the capacity of IgG antibodies to interact with FcγRs. Some evidence indicates that allosteric conformational changes in the Fc domain may be induced upon Fab binding. Created with Biorender.com.

All four IgG subclasses additionally possess a biantennary N‐linked glycan at the asparagine 297 site on the CH2 domain. The Fc glycan consists of a core seven saccharides, including four N‐acetylglucosamine and three mannose residues, which can be modified by additional fucose, N‐acetylaglucosamine, galactose, and sialic acid saccharides—of which 30 different glycoform combinations have been identified ⁷ , ⁸ (Figure 1B). However, it is estimated only 11 glycoforms make up about 90% of the human IgG profile in healthy individuals. ⁹ The Fc glycan sits within the hydrophobic pocket between the two CH2 domains and does not directly interact with FcγRs, but rather influences the three‐dimensional conformation and stability of the antibody, thereby modulating the binding affinity to the various FcγRs. ¹⁰ , ¹¹ Glycosylation of the Fc is essential for the functional activities of IgG, as it has been demonstrated by several studies that genetic substitution at the N297 site (N297A) to prevent any glycosylation leads to abrogation of FcγR binding, and consequently inhibits both pro‐ and anti‐inflammatory effector functions. ¹² , ¹³ Furthermore, studies have shown that the Fc glycoform profile of serum IgG is dynamic and can vary greatly between health and disease states, including during pregnancy and following vaccination. ¹⁴ , ¹⁵ , ¹⁶ The combination of the four IgG subclasses and Fc glycoforms together leads to hundreds of possible Fc compositions, each with sterically distinct conformations that modulate the affinity for the different FcγRs. ¹⁷ Therefore, the Fc structure plays a crucial role in determining downstream cellular responses and is highly regulated to prevent excessive or unwarranted inflammation.

3. FcγR STRUCTURE, EXPRESSION, AND FUNCTION

The FcγRs that interact with IgG antibodies are categorized into two main types: Type I and Type II FcγRs. They are grouped based on their distinct structures and the conformational state of the Fc in which they engage, with Type I and Type II receptors binding the open and closed states, respectively. ¹⁸ , ¹⁹ Type I FcγRs consist of six structurally related receptors that belong to the immunoglobulin superfamily (IgSF), all possessing 2–3 Ig‐like extracellular domains. Type I FcγRs bind in a 1:1 complex, where the loop region of the FcγR, positioned between the Ig‐like domains, engages with the Fc CH2 domain adjacent to the hinge region. ²⁰ , ²¹ Type II FcγRs include the C‐type lectin receptors CD23 and DC‐SIGN and bind in a 2:1 Fc:FcγR complex. ¹⁸ , ²² Type II FcγRs are low affinity for IgG Fc and bind preferentially to sialylated Fc glycoforms, due to the conformational change allowing for a “closed” Fc state. ¹⁸

Of the six Type I FcγRs, there are four activating receptors, namely FcγRI (CD64), FcγRIIa (CD32a), FcγRIIc (CD32c), and FcγRIIIa (CD16a), and one inhibitory receptor, FcγRIIb (CD32b), which have cytoplasmic immunoreceptor tyrosine activating (ITAMs) motif(s) or an immunoreceptor tyrosine inhibitory (ITIM) motif, respectively. One ITAM or one ITIM is located on the ligand‐binding receptor α‐chain (for FcγRIIa, FcγRIIc, and FcγRIIb); two ITAMs are located on the associated FcR γ‐subunit (for FcγRI and FcγRIIIa). FcγRIIIb is a GPI‐anchored protein and does not possess an intracellular α‐chain or ITAM domain; however, it can transduce activation signals through cross‐linking and associating with other receptors and accessory chains, although with less robust activity. ²³ , ²⁴ FcγRs also exhibit significant genetic heterogeneity, with multiple single‐nucleotide polymorphisms (SNPs) identified, as well as copy number variants described for all FcγR encoding genes. ²⁵ Among them, two SNPs, V158F for FcγRIIIa and H131R for FcγRIIa represent the most studied variants, as they influence the affinity for IgGs (IgG1 and IgG3 for FcγRIIIa and IgG2 for FcγRIIa). ²⁵ Importantly, several genetic association studies have demonstrated that these SNPs are associated with increased susceptibility to bacterial infections, chronic inflammatory, and autoimmune pathologies, as well as response to antibody therapeutics in cancer patients. ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² , ³³ , ³⁴ , ³⁵

Type I FcγRs can be further subdivided into high and low‐affinity receptors, based on their ability to interact with monomeric or multimeric IgG. FcγRI is the only receptor that can bind monomeric IgG with a relatively high affinity, due to an additional Ig‐like domain, whereas all other FcγRs bind only to multimeric IgG ICs or opsonized cells. ³⁶ The size of IgG immune complexes has also been reported to impact FcγR binding to low‐affinity FcγRs in vitro, with larger ICs increasing binding affinity, particularly to FcγRIIa and FcγRIIIa receptors ³⁷ (Figure 1C). This low affinity for monomeric IgG protects against erroneous effector cell activation and inflammation unless there is an immunogenic or pathogenic trigger present. ³⁸

As mentioned earlier, fine‐tuning of activating or inhibitory FcγR‐mediated immune responses is partly achieved by posttranslational modifications to the Fc glycan and differences among IgG subclasses. Despite the identification of many unique Fc glycoforms, research has primarily concentrated on the core fucose and terminal sialic acid due to their significant roles in Type I and Type II FcγR engagement. Afucosylated IgG1 demonstrate increased affinity for the activating FcγRIIIa, and thereby mediates enhanced activating effector functions, including ADCC by NK cells and macrophage activation. ²² , ³⁹ , ⁴⁰ Sialylated Fc glycoforms have preferential capacity to interact with the Type II FcγRs DC‐SIGN and CD23 due to increased flexibility of the Fc domain through sialic acid destabilizing the quaternary structure (Figure 1B). ¹⁹ DC‐SIGN engagement results in cytokine production and control of inflammatory myeloid cell activation, whereas CD23 is involved in B‐cell affinity selection. ¹⁵ , ⁴¹ , ⁴² Additionally, IgG subclasses display unique binding affinities for all FcγRs, including differences across FcγR polymorphisms and affinity for activating over inhibitory FcγRs (A/I ratio). Human IgG1 and IgG3 have the greatest A/I ratio, with IgG1 and IgG3 having the greatest affinity for FcγRI, followed by FcγRIIa or FcγRIIIa, respectively. In contrast, IgG2 and IgG4 subclasses have a relatively reduced affinity for all FcγRs, except for IgG4 affinity to FcγRI (Figure 1A). ⁶ , ⁴³ IgG1 additionally has a higher affinity for the FcγRIIIa‐V158 variant compared with FcγRIIIa‐F158, which is associated with increased cell activation. ⁴⁴ , ⁴⁵ Given the apparent differences in IgG Fc profiles between healthy and disease states, including those observed for COVID‐19, ⁴⁶ , ⁴⁷ human immunodeficiency virus (HIV), ⁴⁸ and dengue, ⁴⁰ , ⁴⁹ , ⁵⁰ further investigation is needed to understand the nuances of Fc structure's contribution to disease pathogenesis and to inform optimal therapeutic design.

In addition to the role of the Fc domain structure, FcγR expression patterns and the balance of activating to inhibitory receptor engagement are also important in regulating Fc‐FcγR interactions and downstream signaling effects. Each FcγR has a unique expression pattern among different leukocyte populations, with cell subsets often co‐expressing multiple FcγR types at one given time. ⁵¹ For example, monocytes often express FcγRI, FcγRIIa, and FcγRIIIa, whereas NK cells predominately only express FcγRIIIa. The inhibitory FcγRIIb is present on all cell types except for NK cells and T cells, and is the only Type I FcγR expressed on B cells. ⁵² In response to infection and inflammation, FcγR expression levels and patterns can change due to cytokine and chemokine signals released upon leukocyte activation or inhibition. For example, IL‐4 increases FcγRIIb and decreases activating FcγR expression, while IFN‐ γ induces the opposite effect ⁴¹ , ⁵³ (Figure 1E). The balance in activating and inhibitory FcγR expression is critical in regulating immune responses and protecting against autoimmunity, as it can alter the threshold for IgG‐mediated activation.

Upon Fc‐FcγR engagement, FcγRs cluster within the cell membrane and cross‐link through IgG ICs binding simultaneously to multiple FcγRs (Figure 1C). All activating FcγRs initiate similar signal transduction cascades, triggering multiple key pathways that can induce several cellular functions including antibody‐dependent cellular phagocytosis (ADCP), antibody‐dependent cytotoxicity (ADCC), degranulation, cytokine signaling, superoxide production, antigen‐presenting cell (APC) activation, and B‐cell selection. ¹⁸ FcγR receptor clustering leads to phosphorylation of the ITAM domain by the SRC family of membrane‐anchored kinases, and recruits Syk kinases which further propagate the signal. ⁵⁴ , ⁵⁵ This activation leads to engagement of the phosphoinositide 3‐kinase (P13K) and subsequent recruitment of BTK, GAB2, and phosphoinositide‐specific phospholipase Cγ (PLCγ). These proteins catalyze the production of inositol triphosphate (IP3) and DAG, resulting in an influx of intracellular calcium (Ca²⁺) levels and activation of protein kinase C (PKC). These pathways collectively lead to actin remodeling, facilitating phagocytosis of IgG‐coated ICs, ⁵⁶ , ⁵⁷ , ⁵⁸ , ⁵⁹ as well as ADCC, generating cytotoxic compounds and reactive oxygen species release. ⁶⁰ , ⁶¹ , ⁶² Additional MAPK pathways are also activated, inducing transcription factors to upregulate genes involved in pro‐inflammatory cytokine and chemokine release, cell survival, and cell differentiation. ⁶³ , ⁶⁴

In a finely regulated counterbalance, engagement of the inhibitory FcγRIIb acts in opposition of activating signaling cascades (Figure 1D). Similarly, FcγRIIb receptors aggregate and cross‐link upon IC binding, leading to phosphorylation of its ITIM domain by SRC family kinases. Phosphorylated ITIM recruits SRC homology 2 domain‐containing inositol polyphosphate 5‐phosphatase (SHIP‐1), causing hydrolysis of phosphatidylinositol 3,4,5‐triphosphate (PIP₃), and subsequently preventing activation of proteins such as PLCγ, GAB2, and BTK. In turn, Ca²⁺ influx and MAPK pathways are inhibited, resulting in direct abrogation of activating signals. ⁶⁵ , ⁶⁶ , ⁶⁷ These complex signaling mechanisms underscore the balance between activation and inhibition of immune responses and highlight the intricate regulatory network governing the various Fc‐FcγR functions.

4. ANTIBODY‐MEDIATED NEUTRALIZATION: DIRECT INHIBITION TO FcγR‐DEPENDENT PROTECTION

Antiviral antibodies mediate protection in several ways, yet in the development of mAbs and vaccines, much of the focus is often on an antibody's neutralizing capabilities. Indeed, neutralizing antibodies (nAbs), acquired either through natural immunity or vaccine‐induced responses, are key correlates of protection against many viral infections. The classical definition of neutralization refers to the ability of antibodies to prevent infection of host cells by binding to viral proteins and physically blocking viral entry or fusion. The identification of neutralizing antibodies has relied on in vitro neutralization assays that measure virus infectivity using only antibodies and target cells in isolation. These assays cannot replicate the complex in vivo microenvironment, and more importantly, the pleiotropic effector responses elicited upon FcγR engagement. Consequently, the classification of antibodies as neutralizing may sometimes be misleading, as studies from several research groups have shown that the maximal therapeutic potency of many mAbs against various viral infections is achieved only when neutralization is coupled with the ability to mediate FcγR‐dependent effector functions. ⁶⁸ , ⁶⁹ , ⁷⁰ , ⁷¹ , ⁷² , ⁷³ , ⁷⁴ , ⁷⁵ It is therefore important to distinguish between the in vitro neutralizing activity and the in vivo mechanisms of protection of an antibody. For this review, we discuss both direct virus neutralization and the coupled FcγR‐mediated effector functions of antiviral antibodies.

4.1. Fab‐mediated neutralization

For both enveloped and non‐enveloped viruses, infection requires attachment to cell surface receptors, resulting in virion internalization by endocytic pathways or viral‐host membrane fusion, and subsequent intracellular release of the viral genome. ⁷⁶ Various mechanisms of antibody‐mediated neutralization have been identified, differing based on the function of the viral epitope the Fab domain binds to, thereby inhibiting specific steps of viral entry. The mechanisms of neutralization have largely been determined utilizing evidence from epitope mapping, as well as structural cryo‐EM and X‐ray crystallography studies of antigen–antibody complexes.

One mechanism of neutralization involves the prevention of virion attachment to host cells entirely by antibodies binding to or near the receptor and co‐receptor‐binding sites, sterically inhibiting the interaction between the virion and target cell. Due to its importance for viral entry, many potent neutralizing mAbs target the primary receptor‐binding site on viral glycoproteins. ⁷⁷ For example, in the context of HIV‐1, several mAbs currently in clinical testing target the CD4 binding site on gp120, preventing attachment to host CD4 receptors (e.g., mAbs 3BNC117 and VRC01). ⁷⁸ Similarly, for influenza virus, anti‐hemagglutinin (HA) mAbs prevent attachment to α‐sialic acid, and for SARS‐CoV‐2, the receptor‐binding domain (RBD) on the spike S1 subunit is targeted by mAbs to prevent attachment to angiotensin‐converting enzyme 2 (ACE2). Additionally, some antibodies block viral epitopes which engage host cell co‐receptors necessary for enhanced viral entry. For instance, anti‐HIV‐1 antibodies prevent gp120 glycoprotein engagement with the C‐C chemokine receptor 5 (CCR5) co‐receptor. ⁷⁹ , ⁸⁰ There are also examples of mAbs against enteroviruses and HIV‐1 which may act by disrupting the structure of the virion, leading to loss of the viral genome or insufficient cell attachment. ⁸¹ , ⁸² , ⁸³

Following attachment to the host receptor, viral entry is achieved for many enveloped viruses by fusing the viral and host membranes together in a multistep mechanism. Receptor‐binding causes significant structural changes to occur in the viral glycoproteins, shifting the virion into a post‐fusion conformation and generating the necessary energy to merge the membranes. This fusion allows the viral genome to be released through pores into the host cytoplasm. ⁸⁴ Consequently, antibodies have been described that target the membrane fusion machinery or inhibit conformational changes required for fusion. Multiple neutralizing antibodies have been identified that target the hydrophobic fusion peptide sequence common to enveloped viruses, which becomes exposed following attachment to the host cell, including for influenza virus, ⁸⁵ , ⁸⁶ SARS‐CoV‐2, ⁸⁷ and HIV‐1. ⁸⁸

Antibody‐mediated neutralization can also occur after virion internalization. For both enveloped and non‐enveloped viruses, cell entry can involve hijacking of the host endocytic pathways and subsequent transport by clathrin‐coated vesicles, non‐clathrin‐coated pits, micropinocytosis, or caveolae. ⁷⁶ Similar to surface membrane fusion inhibition, antibodies internalized alongside the virion can potentially prevent infection by disrupting pH‐induced structural changes and uncoating of the viral particle, or by blocking fusion to the endosomal membrane. ⁸⁹ For Ebola viruses (EBOV), mAbs have been identified that can bind to the base of the surface glycoprotein (GP) and likely prevent conformational changes to fusion machinery; however, these are strain‐specific. ⁹⁰ Alternatively, cross‐reactive anti‐EBOV mAbs demonstrate an ability to bind to the partially occluded region of the receptor‐binding site (RBS) exposed following glycan‐cleavage within late endosomes, and thereby block binding to the NPC1 receptor on the endosomal membrane, including two FDA‐approved mAbs, ansuvimab (Ebanga, mAb114), and odesivimab (REGN‐3471). ⁹¹ , ⁹² , ⁹³ , ⁹⁴ Additionally, antibodies can inhibit viral egress of newly formed virions and reduce the number of virus particles released. ⁹⁵ , ⁹⁶ , ⁹⁷ , ⁹⁸ In the case of influenza virus, multiple antibodies targeting distinct antigenic sites on HA and neuraminidase (NA) restrict viral release by preventing diffusion of viral surface proteins or by inducing viral aggregation at the cell surface. ⁹⁹ , ¹⁰⁰ Although this mechanism occurs post‐viral entry and cell infection, and thus may not be considered a true form of direct neutralization, it still reduces viral infectivity by preventing spread to surrounding cells. By targeting various stages of the viral lifecycle, from attachment and entry to post‐internalization events, neutralizing mAbs function in multiple ways to directly prevent viral infections, making them a critical component of antiviral immunity.

4.2. Fc‐mediated effector functions

The diverse expression patterns and levels of FcγRs across various leukocyte populations enable a wide array of downstream innate and adaptive immune functions, promoting a multifaceted response tailored to specific pathogens or disease states. Importantly, the immunomodulatory effects of Fc‐FcγR engagement are not limited to the classical, most well‐studied effector functions of antibody‐dependent cellular cytotoxicity (ADCC) and antibody‐dependent cellular phagocytosis (ADCP). FcγR‐mediated pathways additionally contribute to antigen‐presenting cell activity, macrophage polarization, cytokine and chemokine signaling, and dendritic cell (DC) activation and maturation. FcγR interactions also have the capacity to modulate adaptive immune responses, regulating T‐ and B‐cell activation, as well as IgG production. In the context of viral infections, Fc‐mediated effector functions are essential for comprehensive immune responses, facilitating direct elimination of virus‐infected cells, as well as enhancing long‐term immunity.

One of the most well‐characterized FcγR‐mediated functions during an immune response is the direct killing of the pathogen, as well as the elimination of infected cells through cytotoxic and phagocytic processes. Cytotoxicity against IgG‐coated cellular targets is thought to be mediated by NK cells following engagement of FcγRIIIa, the sole FcγR type expressed by these cells. ³⁸ FcγRIIIa cross‐linking on NK cells results in ITAM phosphorylation of the associated γ‐ or ζ‐chain, which in turn leads to the rapid increase in intracellular Ca²⁺ levels, cellular activation, degranulation, and release of cytotoxic compounds (e.g., granzyme and perforin) near target cells, thereby inducing cell death. Although the downstream signaling events and biological consequences of FcγRIIIa‐mediated NK cell activation have been extensively studied and characterized in numerous studies, experimental evidence suggests that NK cells have a rather limited role in the in vivo IgG‐mediated cellular cytotoxicity. Instead, seminal in vivo studies reported that the clearance of IgG‐opsonized erythrocytes is predominantly mediated by macrophages. ¹⁰¹ In contrast to NK cells, which express only a single activating FcγR (FcγRIIIa in humans and FcγRIII in mice), macrophages typically express an array of activating FcγRs (FcγRI, FcγRIIa, and FcγRIIIa), as well as the inhibitory FcγRIIb, which through its ITIM domain inhibits activating FcγR signaling. ³⁸ Studies using FcγR knock‐out mice have previously revealed that the in vivo cytotoxic activity of mAbs requires the expression of activating FcγRs, suggesting a dominant role for FcγRIV, an FcγR that is not expressed on mouse NK cells. ³ , ¹⁰² , ¹⁰³ , ¹⁰⁴ Blockade or genetic deletion of FcγRIV resulted in complete loss of the cytotoxic activity of mAbs against tumor or virus antigens, highlighting the importance of this receptor in the mAb‐mediated in vivo cytotoxicity. ¹⁰³ , ¹⁰⁴ , ¹⁰⁵ , ¹⁰⁶ Follow‐up studies using conditional FcγRIV knock‐out strains identified FcγRIV‐expressing macrophages and monocytes, as the main leukocyte types that confer cytotoxicity in vivo ¹⁰⁵ ; an observation also confirmed following macrophage ablation by clodronate liposomes. ¹⁰³ , ¹⁰⁵ These findings have also been confirmed in recent studies using conditional knock‐out mouse strains, as well as in adoptive transfer experiments, demonstrating that the FcγR‐mediated cytotoxicity of IgG‐coated targets is mediated exclusively by liver‐resident macrophages, and not NK cells. ¹⁰⁷

FcγR‐mediated cytotoxicity of IgG‐coated targets by macrophages is primarily conferred through phagocytic mechanisms that are also shared among various phagocytes like neutrophils, dendritic cells, monocytes, and eosinophils. Following FcγR engagement by the IgG‐opsonized target, intracellular production of PI(4,5)P₂ and DAG leads to actin cytoskeleton remodeling, initiating internalization and formation of phagosomes. ⁵⁷ , ¹⁰⁸ Early phagosomes then go through a maturation process by fusing with endocytic vesicles and lysosomes, or preformed granules to mediate microbe degradation. The cell type, differentiation stage, and its unique FcγR expression pattern dictate the efficiency and end function following phagocytosis. For macrophages, monocytes, and granulocytes like neutrophils, phagocytosis generally leads to immediate elimination of opsonized targets upon exposure to multiple antimicrobial agents, including hydrolytic enzymes (proteases, lysozymes, lipases, and cathepsins) and reactive oxygen species (ROS). ¹⁰⁹ Alternatively in the case of DCs, phagocytic pathways differ to allow for effective processing and presentation of pathogenic antigens (discussed more below). ¹¹⁰ Engagement of FcγRs also induces activation of multiple transcriptional networks, consequently leading to the upregulation of the expression of several cytokines and chemokines, which further fine‐tune the immune response beyond the immediate effector functions. For example, for macrophages, FcγR and cytokine signaling leads to macrophage polarization into either a pro‐inflammatory (M1) phenotype, with high production of reactive nitrogen and oxygen intermediates for elevated microbicidal activity, or into an immunoregulatory (M2) phenotype, important for infections by parasites or allergy. In turn, FcγR expression is then regulated by the polarization of macrophages and continued cytokine environment. IL‐4 stimulated M2 macrophages demonstrate lower FcγRIIa/FcγRIIb expression ratio, whereas pro‐inflammatory IFN‐γ M1 macrophages show significant increase in FcγRI expression. ¹¹¹

Apart from the effects on innate immune responses, FcγR engagement has important impact on both the humoral and cellular arms of adaptive immunity. A key homeostatic mechanism that regulates DC maturation is the opposing signaling function of the activating and inhibitory FcγRs co‐expressed on the surface of DCs, namely, FcγRIIa and FcγRIIb. Whereas engagement of FcγRIIb typically fails to induce DC maturation, leading in turn to T‐cell anergy, engagement and signaling through the activating FcγRIIa is associated with enhanced endosomal maturation, efficient processing of endocytosed IgG immune complexes through lysosomes, as well as increased antigen processing and presentation on MHC molecules. The critical balance of activating and inhibitory FcγR signaling on the maturation of APCs, like DCs, and the induction of T‐cell responses has been demonstrated in numerous studies using engineered antibodies or genetically modified mouse strains. For example, genetic deletion of the Fcgr2b gene on DCs resulted in enhanced cellular maturation following the uptake of IgG‐opsonized targets, increased expression of co‐stimulatory molecules, enhanced antigen presentation, and the induction of potent T‐cell responses. ¹¹² , ¹¹³ , ¹¹⁴ Similar effects were observed in a model of CD20⁺ lymphoma, where selective engagement of FcγRIIa on DCs using Fc‐engineered anti‐CD20 mAbs led to the induction of protective T‐cell responses directed against CD20.

IgG immune complexes that are generated during an immune response also have the capacity to modulate B‐cell responses primarily through engagement of the Type I (FcγRIIb) and Type II FcγRs (CD23) expressed on B cells. FcγRIIb is expressed throughout B‐cell development and its primary function is to counterbalance any activating signals that are initiated following cross‐linking of the B‐cell receptor (BCR) by antigens. A series of studies have previously defined a critical role for FcγRIIb in the regulation of B‐cell activation and the selection of high‐affinity clones. ¹¹⁵ , ¹¹⁶ , ¹¹⁷ , ¹¹⁸ Engagement of FcγRIIb on B cells results in the induction of pro‐apoptotic signaling, which is attenuated by co‐engagement of the BCR. Therefore, the threshold for B‐cell activation is the outcome of the opposing signaling activity of FcγRIIb and BCR, analogous to the function of FcγRIIa:FcγRIIb signaling on DCs. Engagement of B‐cell FcγRIIb by IgG immune complexes leads to receptor clustering and ITIM phosphorylation, as well as the recruitment of SHIP phosphatases, which directly oppose BCR‐mediated signaling by hydrolyzing phosphoinositide intermediates (e.g., phosphatidylinositol‐3,4,5‐triphosphate). ⁶⁶ , ¹¹⁷ , ¹¹⁸ As a result, signaling downstream of the BCR is attenuated and B‐cell clones with insufficient BCR affinity are eliminated. Several lines of experimental evidence support a key role for the expression levels of FcγRIIb on B cells in setting the threshold for B‐cell activation. For example, mice deficient for B‐cell FcγRIIb expression or signaling are characterized by high titer, low‐affinity IgG responses due to inefficient B‐cell selection. ¹¹⁵ , ¹¹⁶ Similarly, human genetic association studies have previously demonstrated that FCGR2B variants that impact the expression levels (e.g., promoter SNPs of FCGR2B) or signaling of FcγRIIb (e.g., I232T SNP) are associated with increased risk for autoimmune pathologies, including systemic lupus erythematosus (SLE). ¹¹⁹ , ¹²⁰ , ¹²¹

B cells additionally express the Type II FcγR, CD23, which also contributes to the regulation of B‐cell selection. Although originally described as the low‐affinity IgE receptor, having a key role in the modulation of IgE responses, studies have also demonstrated that CD23 has the capacity to interact with sialylated IgG complexes. ¹⁵ Such interactions induce the upregulation of FcγRIIb on B cells, which in turn raise the threshold for B‐cell activation and the induction of high‐affinity IgG responses. ¹⁵ The interplay between CD23 and FcγRIIb in the modulation of IgG responses has been demonstrated in studies on influenza hemagglutinin (HA) immunization using HA‐anti‐HA IgG immune complex as immunogens. ¹⁵ , ¹²² Vaccination with sialylated IgG immune complexes resulted in the induction of high‐affinity IgG responses with the capacity to protect mice from heterologous influenza challenge, highlighting the importance of the sialylated IgG‐CD23‐FcγRIIb axis in the induction of protective IgG responses.

The contribution of FcγRIIb to the regulation of IgG responses is not limited to the modulation of BCR signaling on B cells during B‐cell activation, but extends to plasma cells, which are characterized by loss of surface BCR expression. In this absence of pro‐survival signals from the BCR, plasma cell FcγRIIb engagement by IgG immune complexes that are generated during an immune response results in cell apoptosis, thereby representing a homeostatic mechanism to regulate plasma cell survival and IgG production. ⁶⁶ , ¹¹⁸ , ¹²³ Given the diversity of effector functions that are mediated upon Fc‐FcγR interactions, FcγR pathways result in the regulation of the function and activation status of several leukocyte cell types, modulating several aspects of the innate and adaptive immune response. Using specific examples from several infection pathogens, the biological consequences of FcγR engagement by IgG immune complexes during immune responses are discussed in detail below.

5. BARRIERS TO DESIGNING OPTIMAL ANTIBODY THERAPEUTICS

Initial approaches for identifying protective antiviral mAbs for clinical use or characterizing effective antibody responses following vaccination have mainly relied on in vitro binding and neutralization assays. During the COVID‐19 pandemic, maximizing neutralizing antibody responses targeting the SARS‐CoV‐2 S protein was a primary goal in the development of the leading mAbs and vaccines. ¹²⁴ All mAbs approved for emergency use authorization (EUA) in the United States were selected for clinical testing due to high neutralizing activity, including the REGN‐COV mAbs, casirivimab (REGN10933), and imdevimab (REGN10987), Evusheld (tixagevimab (AZD8895) and cilgavimab (AZD1061)), bamlanivimab (LY‐CoV555), estesevimab (LY‐CoV016), bebtelovimab (LY‐CoV1404), and sotrovimab (S309). ¹²⁵ , ¹²⁶ , ¹²⁷ , ¹²⁸ , ¹²⁹ , ¹³⁰ This focus on potent neutralization for initial selection is with good reason, as direct neutralization is a key component of optimal antiviral protection. Several studies have highlighted that increased in vitro neutralizing levels in polyclonal antibody samples is associated with decreased risk of hospitalization and severe COVID‐19 disease. ¹³¹ , ¹³² However, in vitro neutralization potency does not necessarily correlate with in vivo protection, as in vitro assays cannot capture the important contributions of Fc‐FcγR interactions. ¹³³ Indeed, an increasing body of evidence using mechanistic in vivo models, across multiple types of viral infections, has revealed a clear distinction between in vitro neutralizing activity and in vivo antiviral protection of neutralizing mAbs. ³⁸ , ¹³⁴ , ¹³⁵ , ¹³⁶

Demonstrating the significance of optimal Fc‐FcγR interactions in antiviral protection during the development of therapeutic mAbs has encountered several challenges due to conventional selection approaches and the limitations of in vivo models. First, as mentioned, most mAbs have historically been selected based on their neutralization potency against viruses, potentially excluding candidates with robust effector functions. It is likely that a fine balance exists between epitope specificity and Fc‐mediated effector functions, as demonstrated by multiple antibodies showing varied ADCC, ADCP, and other Fc effector activities depending on their specific epitopes, even with identical IgG subclass and Fc glycosylation. This is well illustrated in a series of studies evaluating anti‐influenza antibodies targeting the HA glycoprotein. Strain‐specific mAbs against the HA head domain often exhibit reduced Fc effector function capacity, while those targeting the more conserved central stalk domain of HA typically demonstrate potent cytotoxic activity. ⁷¹ , ¹³⁶ , ¹³⁷ , ¹³⁸ , ¹³⁹ The influence of epitope specificity on Fc‐mediated activity has been observed for antibodies against other viral infections. Anti‐EBOV mAbs targeting the chalice bowl and fusion loop require FcγR engagement to mediate protective activity. In contrast, neutralizing anti‐EBOV antibodies against epitopes with close membrane proximity, such as the HR2 and MPER, exhibit minimal FcγR dependence. ¹⁴⁰ Heterogeneity of the specific Fc effector functions activated by anti‐HIV‐1 antibodies with different binding epitopes, for both mAbs and endogenous responses by elite controllers, has also been observed. ⁷⁴ , ¹⁴¹ , ¹⁴²

Although the specific mechanisms that impact the differential FcγR engagement between epitope‐specific antibodies remain to be determined, these findings indicate that the angle which antibodies bind and the resulting spatial configuration may determine their capacity to engage with FcγRs (Figure 1F). Existing experimental approaches using structural analysis or biophysical assays to investigate mAb Fc affinity for FcγRs commonly ignore the stoichiometry of the Fc domain, particularly when it is complexed with antigens. Without such consideration, a full understanding of the coordination between epitope binding, Fc orientation, and subsequent FcγR engagement is limited. Furthermore, whether antigen binding can induce additional steric conformational changes in the Fc domain still requires more investigation (reviewed by Ref. 143), but recent studies have reported that antigen binding to human IgG1 is accompanied by a change to open Fc conformations which favor FcγR binding. ¹⁴⁴ , ¹⁴⁵ In addition to the established role of IgG subclass and Fc glycosylation, epitope specificity may be another factor influencing FcγR engagement and thus antibody protective functions. Although examples of thorough screening methods exist, ¹⁴⁶ it will be important for mAb screening studies to include systematic characterization of the Fc domain, including subclass, glycosylation, and FcγR affinity. Understanding the relationship between the Fab and Fc features of antibodies and how they correlate with enhanced in vivo protection will likely accelerate the selection of improved therapeutics with optimal efficacy.

A second barrier to understanding the contribution of Fc‐FcγR interactions to antiviral protection is the lack of sufficient biological relevance in commonly used cellular assays and in vivo models. Many in vitro studies exist to investigate the FcγR effector functions of mAbs, yet they often are limited by their reproducibility, throughput efficiency, and in vivo congruity. Commonly used reporter‐based assays use cell lines that express only a single FcγR, such as FcγRIIIa or FcγRIIa, which are believed to be primarily responsible for ADCC and ADCP activities, respectively. ¹⁴⁷ However, leukocytes naturally co‐express multiple activating FcγRs alongside the inhibitory FcγRIIb and rely on the coordinated signaling between one another for a comprehensive response. Other assays, which use peripheral blood mononuclear cells (PBMC) from donated blood commonly only involve isolation and measurement of NK cell FcγRIIIa‐mediated ADCC or monocyte‐derived dendritic cell (moDC) activation. The read‐out from these assays are typically single activation markers and disregard the potential importance of engagement of other leukocyte populations and various FcγRs by specific mAbs. With a single‐cell type approach, the significant functional redundancy between leukocyte populations, cytokine signaling, and various downstream effects are not captured. Additionally, FcγR polymorphism differences among donor‐derived cells can cause high inter‐assay variability. Consequently, current in vitro systems do not accurately reflect the complex and diverse Fc effector functions by which IgGs mediate antiviral activities.

Conventional in vivo models, while better at capturing the diversity of effector leukocytes and the interactions between activating and inhibitory FcγRs, still fall short due to significant interspecies differences. These include variations in FcγR expression patterns, functions, and affinities for human IgG antibodies, which differ significantly from human FcγRs (reviewed in Refs. 38, 148). For example, humans express FcγRIIc and FcγRIIIb, whereas mice and rhesus macaques lack equivalent receptors. Additionally, FcγR expression patterns on specific leukocytes can vary widely. In humans, FcγRIIIa is expressed on NK cells and specific macrophage and monocyte subsets. In contrast, the homologous mouse receptor, FcγRIV, is constitutively expressed on monocytes and neutrophils. ¹⁴⁹ Therefore, the absence of a receptor homologous to human FcγRIIIa on mouse NK cells limits the ability to assess NK‐mediated ADCC activity in vivo. These interspecies differences also extend to nonhuman primate models. ¹⁴⁸ , ¹⁵⁰ The limited translational relevance from mechanistic studies using conventional animal models makes it challenging to fully understand the contributions of Fc‐FcγR interactions to antiviral protection. To address this issue, multiple groups have developed FcγR‐deficient, as well as partial and fully FcγR humanized mouse models to elucidate the function of human mAbs in vivo. ¹⁴⁹ , ¹⁵¹ FcγR humanized mouse strains are fully immunocompetent but lack all mouse FcγR genes and instead express all human FcγRs as transgenes, regulated by their endogenous human promoters to ensure a human‐specific FcγR expression pattern across leukocytes. ¹⁵¹ Thus, FcγR humanized mouse models can more accurately recapitulate the specific Fc‐FcγR interactions engaged by human IgG on certain cell types and reproduce subsequent cellular functions. Through the use of FcγR humanized mouse models, it was demonstrated that FcγR activity is essential for protection by many neutralizing monoclonal antibodies targeting influenza virus, HIV‐1, Ebola virus, SARS‐CoV‐2, and RSV. ⁶⁹ , ⁷¹ , ¹⁵² , ¹⁵³ These models have also been employed to study FcγR‐dependent mechanisms by which antibodies mediate anti‐tumor cytotoxicity and autoimmune diseases. ¹⁰⁵

6. Fc‐MEDIATED FUNCTIONS OF ANTIVIRAL IGG ANTIBODIES

Despite the challenges in identifying and characterizing the Fc functions of antiviral mAbs in appropriate in vitro and in vivo models, a clear role for Fc‐FcγR interactions in achieving optimal protection has emerged, particularly for broadly neutralizing antibodies (bNAbs). The emergence of viral escape variants, which existing neutralizing mAbs or vaccine‐induced responses can no longer recognize, is a significant concern. This was evident during the emergence of Omicron variants during the COVID‐19 pandemic. However, bNAbs demonstrate that while Fab‐mediated neutralization may be reduced, Fc effectors can remain largely unaffected. ¹⁵⁴ , ¹⁵⁵ Furthermore, evidence suggests that bNAbs can induce indirect mechanisms of protection upon FcγR engagement, leading to long‐lasting cellular immune responses comparable to vaccine‐like effects. Here, we discuss the evidence of FcγR‐dependent protection by neutralizing antibodies against influenza virus, SARS‐CoV‐2, and EBOV, with a specific focus on its importance for broadly cross‐reactive antibodies.

6.1. Influenza virus

The development of influenza vaccines has largely focused on designing vaccine immunogens that can induce broadly protective antibody responses against constantly evolving influenza strains. Significant evidence from studies on broadly neutralizing mAbs and endogenous antibody responses following vaccination has indicated the critical involvement of Fc effector functions for optimal in vivo protection. ¹⁵⁶ , ¹⁵⁷ , ¹⁵⁸ , ¹⁵⁹ It was first demonstrated that protection by passive transfer of serum from H1N1‐immunized mice was lost in FcR γ^−/− (Fcer1g ^−/−) mice. ¹⁶⁰ Subsequent foundational studies evaluating a panel of bNAbs, which target both the HA head and stem domains and recognize multiple influenza A strains and subtypes, revealed that activating FcγR engagement is required for protection in vivo. Furthermore, non‐neutralizing but broadly cross‐reactive antibodies also depend on Fc‐FcγR interactions for protection, unlike strain‐specific neutralizing mAbs, which retain protective activity when expressed as Fc variants with diminished FcγR binding or assessed in FcγR knock‐out mice. ⁷¹ , ¹³⁶ Follow‐up mechanistic studies determined that modifying human IgG1 bNAbs (FI6v3 and FY1) to selectively engage the FcγRIIa receptor enhances protection in FcγR humanized mice, compared with selective FcγRIIIa engagement or wild‐type antibodies. It was shown that FcγRIIa‐mediated engagement on DCs improves maturation and antigen presentation, thereby eliciting protective CD8+ T‐cell responses. ¹⁶¹ Consistent with mechanistic findings on the limited role of FcγRIIIa engagement with certain bNAbs, results from two maternal seasonal influenza vaccination trials observed that in vitro ADCC activity of anti‐H1 stalk polyclonal antibodies was not associated with reduced infection, whereas FcγRIIa‐mediated activation was. ¹⁶² Collectively, these results indicate that multiple Fc‐mediated effector functions can contribute to protection in vivo depending on the different epitope bound. A similar role of FcγR‐dependent protection was observed for a panel of cross‐reactive, non‐neutralizing antibodies against influenza virus B strains. ¹⁶³

The critical role of FcγR engagement has informed the design of influenza vaccine antigens to preferentially elicit bNAb responses with enhanced Fc‐mediated functions against conserved epitopes. ¹⁶⁴ , ¹⁶⁵ , ¹⁶⁶ , ¹⁶⁷ For instance, a phase I clinical trial administering sequential immunization with chimeric HAs (cHAs), which share the same H1 stalk domain but have varying HA head domains, induced cross‐reactive group 1 anti‐stalk antibodies with strong in vitro ADCC activity and increased T‐cell responses. ¹⁶⁵ , ¹⁶⁸ Further investigation showed that passive transfer of the human IgG antibodies elicited by vaccination with cHA immunogens protected FcγR humanized mice against lethal influenza virus infection. However, this protection was absent in FcγR‐deficient mouse strains, suggesting that the in vivo protection depends on Fc‐FcγR interactions. ¹⁶⁹ Additional studies have demonstrated that bNAbs induced by alternative vaccination platforms, including vectored, mRNA, and nanoparticle‐based vaccines, also activate Fc‐mediated effector functions. ¹⁶⁷ , ¹⁷⁰ , ¹⁷¹ Vaccination approaches targeting other conserved epitopes on the influenza virion, such as matrix‐2 ion channel (M2) and nucleoprotein (NP), similarly mediate broad protection and activate Fc effector functions. ¹⁷² , ¹⁷³ , ¹⁷⁴ Therefore, the design of optimal universal influenza virus immunogens and vaccination strategies must continue to prioritize eliciting durable IgG responses with robust Fc effector functions.

6.2. SARS‐CoV‐2

During the COVID‐19 pandemic, the rapid development and deployment of mAb therapeutics were important in mitigating the effects of SARS‐CoV‐2 infections early on. In the United States, five mAb therapies, both as single mAb formulations or as combination cocktails, received EUA for pre‐ and post‐exposure prophylaxis, with additional mAb therapies approved for use in other countries (Table 1). All approved mAbs are of the IgG1 subclass and specifically target the receptor‐binding domain (RBD) of the spike (S) protein on the SARS‐CoV‐2 virion. These mAbs primarily provide protection through potent Fab‐mediated neutralization, as reviewed elsewhere. ¹⁹⁷ , ¹⁹⁸ , ¹⁹⁹ However, following the emergence of new SARS‐CoV‐2 variants of concern, beginning with the Omicron lineage, a substantial reduction in the binding affinity and neutralization activity of mAbs therapies was observed. ¹⁷⁵ , ²⁰⁰ , ²⁰¹ , ²⁰² , ²⁰³ , ²⁰⁴ Only bebtelovimab retained in vitro neutralizing activity against initial Omicron subvariants, ²⁰⁰ yet it showed minimal effectiveness against later BQ.1 and XBB subvariants. ²⁰⁵ , ²⁰⁶ This marked decrease in neutralization potency against circulating SARS‐CoV‐2 variants ultimately led to the discontinuation of all mAbs as recommended treatments for COVID‐19 in the United States. This underscores the need for development of broadly reactive mAbs with greater resistance to neutralization evasion by viral mutations.

TABLE 1.

Fc domain modifications of approved therapeutic monoclonal antibodies against infectious diseases.

INN	Infectious disease target	Subclass	Binding epitope	Fc modifications	First approval (Country, year)	Ref.
Unmodified Fc domain—baseline FcγR and FcRn affinity
Casirivimab + Imdevimab	SARS‐CoV‐2	IgG1	RBD, RBM	—	Japan, 2021	69, 175
		IgG1	RBD, core	—
Bebtelovimab	SARS‐CoV‐2	IgG1	RBD, core	—	US, 2022	175
Bamlanivimab	SARS‐CoV‐2	IgG1	RBD, RBM	—	US, 2021	127, 130, 176
Regdanvimab	SARS‐CoV‐2	IgG1	RBD, RBM	—	South Korea, 2021	177, 178
Ansuvimab	Ebola virus	IgG1	GP	—	US, 2020	93
Atoltivimab + Maftivimab + Odesivimab	Ebola virus	IgG1	Glycan cap	—	US, 2020	179, 180
		IgG1	GP fusion loop	—
		IgG1	GP1 head / sGP	—
Ibalizumab	HIV‐1	IgG4	CD4 extracellular domain 2	—	US, 2018	181
RabiShield	Rabies virus	IgG1	G glycoprotein (site III)	—	India, 2016	182, 183
Docaravimab + Miromavimab	Rabies virus	mIgG2b	G glycoprotein (site III)	—	India, 2019	184
		mIgG1	G glycoprotein (site II)	—
Ormutivimab	Rabies virus	IgG1	G glycoprotein	—	China, 2022	185
Raxibacumab	B. anthracis	IgG1	Anthrax toxin, PA	—	US, 2012	186, 187
Obiltoxaximab	B. anthracis	IgG1	Anthrax toxin, PA	—	US, 2016	188
Bezlotoxumab	C. difficile	IgG1	Enterotoxin B	—	US, 2016	189, 190
Palivizumab	RSV	IgG1	F protein	—	US, 1998	191, 192
Diminished Fc effector function—minimal FcγR affinity; baseline affinity for FcRn
Etesevimab	SARS‐CoV‐2	IgG1	RBD, RBM	LALA	US, 2021	127, 130, 176
Extended half‐life—baseline FcγR affinity; enhanced affinity for FcRn
Sotrovimab	SARS‐CoV‐2	IgG1	RBD, core	LS	Aus, 2021	155, 175
Extended half‐life and reduced Fc effector function—reduced FcγR affinity; enhanced affinity for FcRn
Amubarvimab +Romlusevimab	SARS‐CoV‐2	IgG1	RBD, RBM	YTE	China, 2021	193, 194
		IgG1	RBD	YTE
Nirsevimab	RSV	IgG1	F protein	YTE	EU, 2022	191, 195
Extended half‐life and diminished Fc effector function—minimal FcγR affinity; enhanced affinity for FcRn
Tixagevimab +Cilgavimab	SARS‐CoV‐2	IgG1	RBD, RBM	YTE‐TM	EU, 2022	196
		IgG1	RBD, RBM	YTE‐TM

Note: Receptor‐binding domain (RBD); receptor‐binding motif (RBM); glycoprotein (GP); Bacillus anthracis (B. anthracis); protective antigen (PA); Clostridium difficile (C. difficile). Ref: The Antibody Society. Therapeutic monoclonal antibodies approved or in review in the EU or US. (August 9, 2024); www.antibodysociety.org/resources/approved‐antibodies. LALA: L234A/L235A; LS: M428L/N434S; YTE: M252Y/S254T/T256E; YTE‐TM: M252Y/S254T/T256E/L234F/L235E/P331S.

Several studies have highlighted the role of Fc‐mediated effector functions in the antiviral efficacy of mAbs and immune responses against SARS‐CoV‐2. Studies examining antibody responses of hospitalized COVID‐19 patients in the United States and Canada observed reduced IgG1 and IgG3 responses, FcγR binding, and ADCC and ADCP activity was associated with increased risk of COVID‐19 mortality. ²⁰⁷ , ²⁰⁸ in vitro and in vivo mechanistic studies across various animal models have independently confirmed that neutralizing anti‐SARS‐CoV‐2 mAbs rely on Fc‐FcγR interactions for their protective efficacy. For instance, SARS‐CoV‐2 challenge studies using a mouse‐adapted strain in BALB/c mice found that the in vitro neutralization potency did not always correlate with in vivo protection, and Fc‐null Fc variants (G236R/L328R; GRLR) of multiple mAbs exhibited significantly diminished protective efficacy. ¹³³ In K18‐hACE2 transgenic mouse strains, protection conferred by multiple mAbs against SARS‐CoV‐2 was lost when expressed as Fc variants (e.g., L234A/L235A (LALA) and L234A/L235A‐P329G (LALA‐PG)) that lack FcγR binding and the ability to induce Fc effector functions. ⁶⁸ , ⁷⁰ , ²⁰⁹ In contrast, GASDALIE (G236A/S239D/A330L/I332E) variants that enhance FcγR binding provided complete protection and reduced viral dissemination. ⁷⁰ Further cell depletion studies identified various leukocyte populations that are essential for the protective activity by different mAbs, including NK cells, Ly6C^hi monocytes, and neutrophils ⁷⁰ in one study, while Ly6C^hi monocytes and CD8⁺ T cells were essential for maximal viral control and protection against weight loss in another. ⁶⁸ A similar dependence on Fc‐FcγR interactions was observed in a Syrian hamster model, where the passive transfer of the COV2‐2050 mAb (expressed as wild‐type human IgG1) prevented weight loss, a protection that was lost with the FcγR‐null LALA‐PG variant. ⁶⁸ FcγR‐dependent protection and viral load reduction were also observed for neutralizing mAbs targeting regions of the spike protein outside the receptor‐binding domain (RBD). ²¹⁰

For clinically approved mAbs, Fc effector functions have been less extensively characterized, with the exceptions of sotrovimab (S309) and the REGN mAb cocktail (casirivimab and imdevimab). Studies using FcγR humanized mice demonstrated that the REGN cocktail conferred complete protection against lethal challenge with a mouse‐adapted SARS‐CoV‐2 strain, whereas Fc‐null variants (GRLR) or FcγR‐deficient mice did not exhibit such protection. Furthermore, a GAALIE (G236A/A330L/I332E) variant of the REGN cocktail, engineered to enhance binding to the activating FcγRs, FcγRIIa, and FcγRIII, and reduce affinity for the inhibitory FcγRIIb, provided superior protection at a dose five times lower than that required by the wild‐type variants. ⁶⁹ Similarly, sotrovimab exhibited FcγR dependence, as prophylactic treatment with an Fc‐null variant (S309‐GRLR) in K18‐hACE2 or FcγR humanized mice resulted in minimal reduction of lung viral titers. ¹⁵⁵ Notably, although sotrovimab's in vitro neutralization potency was diminished against Omicron variants BA.1, BA.1.1, and BA.2 compared with the ancestral WA1/2020 D614G strain, sotrovimab still reduced BA.1, BA.1.1, and BA.2 viral RNA titers in vivo. ¹⁵⁵ Despite the evidence supporting the importance of Fc‐FcγR engagement for optimal protection in vivo, multiple clinically approved mAbs were engineered with Fc modifications early on in their development due to the uncertainty surrounding the role of ADE for SARS‐CoV‐2. This included Fc modifications that either extend antibody half‐life (through improved FcRn binding) at the expense of reduced affinity to other FcγRs, or eliminate FcγR binding entirely, as seen with the YTE/TM modifications in tixagevimab/cilgavimab ¹⁹⁶ , ²¹¹ and LALA modification in etesevimab, respectively ²¹² (Table 1). While Fc‐mediated activity is not universally required for all anti‐SARS‐CoV‐2 mAbs, it can play an important role in enhancing the overall therapeutic effectiveness of mAbs against certain epitopes. Optimizing Fc‐FcγR interactions during the early stages of mAb development could offer a strategy to counteract the neutralization evasion caused by viral mutations, as observed throughout the COVID‐19 pandemic.

6.3. Ebola virus

Following the 2014 Ebola epidemic, the development of antibody‐based therapeutics with potent antiviral activity against Ebola viruses intensified and has led to the approval of two mAbs in the United States. One of the first mAbs identified with high neutralizing activity was KZ52; however, it demonstrated no protection in a nonhuman primate (NHP) model of EBOV infection. Since then, several neutralizing and non‐neutralizing mAbs have been tested in pre‐clinical models to identify broadly protective post‐exposure treatments. An early study of the MB‐003 cocktail, consisting of three IgG1 mAbs (13C6, 13F6, and 6D8), demonstrated that the afucosylated glycoforms, compared with an otherwise identical mAb cocktail, showed improved protection in an NHP model, indicating the potential contribution of FcγRIIIa‐mediated effector functions. ²¹³ Subsequent investigations have continued to explore the mechanisms of protection, highlighting the need for both potent direct neutralization and Fc effector functions. A screening of 171 mAbs by the Viral Hemorrhagic Fever Immunotherapeutic Consortium (VIC) revealed that, generally, non‐neutralizing mAbs provided less in vivo protection compared with neutralizing mAbs, although there was significant variability in protective activity within each group. Notably, some highly protective non‐neutralizing mAbs were identified, while some robust neutralizing mAbs offered minimal protection. ⁹⁰ This and other studies observed that in vivo protection by anti‐EBOV mAbs correlated with the induction of at least one Fc effector function, and polyfunctional antibody responses were associated with increased efficacy. ⁷⁵ , ⁹⁰ , ²¹⁴ Additionally, in vivo mouse and guinea pig models, ²¹⁵ , ²¹⁶ as well as machine learning applications identified afucosylated Fc as an important predictor of protection, whereas fucosylated (G2FB) and sialylated (G2S1) glycoforms were negatively associated with protection. ⁷⁵ These results are consistent with the understanding that increased affinity for activating Type I FcγRs receptors over FcγRIIb and Type II FcγRs promote protective pro‐inflammatory activity. in vitro mechanistic studies of endogenous antibody responses in survivors and isolated mAbs have shown NK cell‐dependent ADCC, phagocytosis, and antibody‐mediated complement deposition are associated with in vivo protection. ⁹³ , ²¹⁷ , ²¹⁸ There have been limited studies using biologically relevant in vivo models, however, one study has demonstrated that the protection provided by mAbs targeting certain epitopes with high neutralizing activity, such as the GP fusion loop, was significantly reduced in FcγR‐deficient mice. Furthermore, Fc modifications to the anti‐GP fusion loop mAbs that enhance binding affinity for activating FcγRs (GASDALIE) resulted in increased protection compared with the wild‐type hIgG1 mAb. In contrast, neutralizing mAbs targeting other epitopes, including HR2 and MPER, showed a limited reduction of protection when FcγR binding was abrogated.

The mixed dependency on both direct neutralization and various Fc effector functions for optimal in vivo protection by different mAbs is evident in the two currently approved antibody therapeutics. The first approved monotherapy, ansuvimab (Ebanga, mAb114), is a human IgG1 mAb isolated from a survivor of the 1995 Kikwit Ebola outbreak. It targets the glycoprotein on Zaire ebolavirus and has demonstrated strong neutralizing ability, as well as in vitro ADCC activity. ⁹³ , ²¹⁹ However, the contribution of ADCC or other Fc effector functions has not been identified in appropriate in vivo models. The second approved mAb cocktail, REGN‐EB3, consists of three IgG1 mAbs (atoltivimab, maftivimab, and odesivimab), which bind to non‐overlapping epitopes on the Zaire EBOV GP. Maftivimab (REGN3479) and atoltivimab (REN3470) demonstrate strong and mild neutralizing activity against live EBOV strains, respectively, whereas odesivmab (REGN3471) displayed weak neutralization even at high antibody concentrations. In contrast, odesivmab and atoltivimab displayed in vitro activation of FcγRIIIa signaling upon binding to EBOV‐GP expressing target cells, whereas maftivimab showed no FcγRIIIa‐mediated activation. ¹⁷⁹ The combination of both neutralizing and effector activities likely contributes to the effective protection of REGN‐EB3 against EBOV in NHPs and human trials. ¹⁷⁹ , ¹⁸⁰ Although the outlined in vitro studies highlight the role of FcγR‐mediated functions in facilitating protective responses against Ebola, the precise mechanisms of protection by anti‐Ebola antibodies need further investigation to optimize and develop broadly protective mAb therapies across all orthoebolaviruses, beyond just EBOV.

7. ANTIBODY‐DEPENDENT ENHANCEMENT

As highlighted above, effective antiviral immune responses that combine Fab‐mediated neutralization with diverse Fc effector functions result in efficient pathogen clearance and the induction of long‐term immunity. However, in some viral contexts, dysregulated humoral and cellular responses can contribute to disease pathogenesis, in addition to the effects of viral infection alone. This is evident for SARS‐CoV‐2 infection, where the spectrum of COVID‐19 disease presentation varies from asymptomatic cases to severe disease with life‐threatening complications. Antibody‐mediated responses overwhelmingly provide protection against infection, yet there are examples where antibodies can facilitate increased viral entry into host cells by harnessing FcγR pathways, enhancing the infection and disease manifestations. This phenomenon is known as antibody‐dependent enhancement (ADE). The primary evidence for ADE has predominantly come from studies analyzing the heterogeneity of antibody responses between people with varying disease severity following dengue virus (DENV) infection. ADE has been suggested to also play a role in enhanced disease for other viral infections, however, the lack of biologically appropriate models used to investigate the contribution of Fc‐FcγR interactions extends to ADE research, limiting the ability to make strong causal inferences.

7.1. Dengue

Dengue fever is a vector‐borne disease caused by a positive (+) stranded RNA virus (dengue virus (DENV)) from the Flavivirus family. There are four DENV serotypes identified worldwide, namely DENV‐1, DENV‐2, DENV‐3, and DENV‐4. For most DENV infections, patients are asymptomatic or experience dengue fever (DF), defined by mild symptoms such as fever, joint pain, and headaches. However, a minority of people experience dengue hemorrhagic fever (DHF), characterized by persistent acute fever, thrombocytopenia, and vascular leakage, which can further lead to dengue shock syndrome (DSS), causing tissue injury, multi‐organ failure, and mortality. Upon infection with a DENV serotype, immune responses include both serotype‐specific antibodies which confer long‐lasting immunity, and cross‐reactive antibodies capable of binding to other DENV serotypes. ²²⁰ , ²²¹ The potential role of ADE in severe dengue disease was first proposed following an observational study in Thailand, which found that both infants with a primary DENV infection and older children with a secondary infection had higher levels of cross‐reactive antibodies, and a greater prevalence of DHF/DSS. ²²² Further investigation revealed that secondary infection with a DENV serotype distinct from the primary serotype, transferred from DENV‐immune mothers to their neonates or elicited by a prior infection, is associated with a greater risk of DHF/DSS. ²²³ , ²²⁴ It was proposed that while at high IgG concentrations infection is blocked due to potent neutralization, at sub‐neutralizing concentrations, anti‐DENV IgGs promote infection. ²²⁵ This effect was demonstrated in a Nicaraguan pediatric cohort study which reported that children with high levels of preexisting anti‐DENV IgG had no difference in relative risk of DHF/DSS when compared to DENV‐naïve children, after adjustment for sex and age. In contrast, children with low levels of neutralizing anti‐DENV IgG had 7.64 times the risk compared with naïve children (95% CI: 3.19–18.28). ²²⁶ Despite the continued epidemiological evidence, the precise in vivo mechanisms by which ADE contributes to varying dengue disease outcomes have only been recently elucidated.

Building on early in vitro studies demonstrating that ADE occurs in FcγR‐expressing phagocytic cells, ²²⁷ , ²²⁸ , ²²⁹ , ²³⁰ it was hypothesized that preexisting anti‐DENV IgGs form antibody–virion immune complexes that engage FcγRs. In turn, FcγRs facilitate the internalization of the virion through phagocytic pathways, resulting in increased viral replication. ²³¹ Mechanistic in vitro evidence using FcγR‐expressing cell lines determined that sub‐neutralizing antibodies promote ADE through activating receptors FcγRIIa and FcγRIIIa, while cross‐linking of FcγRIIb inhibits ADE by reducing FcγR‐mediated phagocytosis. ²³⁰ , ²³² , ²³³ , ²³⁴ , ²³⁵ , ²³⁶ , ²³⁷ In corresponding in vivo ADE studies, passive transfer of cross‐reactive polyclonal and monoclonal anti‐DENV antibodies followed by DENV challenge in susceptible mouse models has shown that blocking the Fc binding to activating FcγR prevents enhanced viremia or disease. ²³⁸ , ²³⁹ , ²⁴⁰ , ²⁴¹ Due to limitations in the clinical manifestation of DENV in nonhuman primate models, there remains limited evidence of the specific mechanisms of FcγRs mediating ADE, but similar findings have been reported. ²⁴² , ²⁴³

Importantly, the magnitude of heterospecific antibody titers does not fully explain the differences in dengue disease severity, as only a minority of people with preexisting, non‐neutralizing IgGs progress to DHF or DSS. ²³⁸ , ²⁴⁴ Additionally, DHF/DSS has been reported following primary infection, not just in infants. ²⁰⁵ , ²⁴⁵ , ²⁴⁶ A recent prospective cohort study showed that baseline neutralizing antibody levels after two or more prior DENV infections were comparable between people who would go on to have inapparent or symptomatic infections. Collectively, these findings indicate that in vitro neutralizing titers do not always correlate with protection or enhanced disease. It is likely that antibody characteristics beyond direct neutralization may change with subsequent DENV infections and contribute to disease outcomes. Given the dependence of DENV ADE on Fc‐FcγR interactions, disease susceptibility is likely determined not only by anti‐DENV IgG titers, but also by the affinity of these antibodies for FcγRs due to differences in Fc domain structure. Indeed, cohort studies identified that elevated levels of afucosylated anti‐DENV IgG, and specifically IgG1, are associated with increasing dengue disease severity. ⁴⁹ , ⁵⁰ , ²⁴⁷ Maternal anti‐DENV IgGs enriched for afucosylated glycoforms were also found to be associated with enhanced clinical disease in their infants, indicating that increased affinity of FcγRIIIa by afucosylated IgG can contribute to ADE. ²⁴⁷ Importantly, analysis of the Fc glycan composition of IgG antibodies from dengue patients with differential disease severity revealed that elevated levels of afucosylated Fc glycoforms precedes the development of severe symptoms, suggesting that dysregulated Fc glycosylation is a key driver for disease pathogenesis, rather than a manifestation of severe disease. ⁴⁹

A recent follow‐up in vivo study ⁴⁰ using a fully FcγR humanized and DENV‐permissive mouse model (Ifnar1 ^−/−; FcγR humanized) aimed to dissect the mechanisms by which afucosylated anti‐DENV IgG antibodies modulate disease pathogenesis. Transfer of IgGs derived from patients with symptomatic dengue disease to these mouse strains resulted in symptomatic disease, which was characterized by the exact same clinical and histopathological features seen in patients with severe disease, including thrombocytopenia, hepatocellular necrosis, and thrombosis ⁴⁰ Using this mouse model, it was shown that engagement and activation of the various human FcγRs by anti‐DENV IgGs had differential impact on disease pathogenesis, and Fc variants with increased affinity for FcγRIIIa (ALIE variant [A330L/I332E or afucosylated]) were associated with increased morbidity and mortality. These findings confirmed prior studies that demonstrated that Fc domain afucosylation is associated with susceptibility to symptomatic dengue disease, as well as with disease severity. ⁴⁹ , ⁵⁰ This study also revealed that a critical step in the in vivo pathogenesis of dengue disease is the engagement of FcγRIIIa on splenic macrophages by afucosylated IgGs. FcγRIIIa‐afucosylated IgG interactions result in aberrant macrophage activation, inflammatory sequelae, significant morbidity, and mortality, highlighting the importance of the FcγRIIIa‐afucosylated Fc axis in dengue disease. These studies suggest that dengue pathogenesis is not the outcome of altered DENV tropism and increased infectivity of FcγR‐expressing cells, but it is primarily driven by aberrant activation of pro‐inflammatory pathways in response to FcγRIIIa engagement on macrophages by afucosylated IgGs. Importantly, these findings are in stark contrast to previous reports that suggested FcγRIIa as the predominant receptor involved in ADE. However, these studies typically use single FcγR‐expressing cell lines or in vivo models which do not accurately reflect human FcγR structure and function, ²⁴⁸ highlighting the importance of using biologically relevant models.

7.2. ADE in other viral infections

Under experimental conditions similar to those commonly used in in vitro assays for ADE of DENV, several studies have previously demonstrated that IgG antibodies against multiple, diverse viral pathogens can also mediate ADE in vitro. Indeed, analogous to ADE of DENV infection, IgG antibodies against HIV‐1, ²⁴⁹ , ²⁵⁰ Ebola virus, ²⁵¹ , ²⁵² , ²⁵³ influenza virus, ²⁵⁴ , ²⁵⁵ , ²⁵⁶ , ²⁵⁷ , ²⁵⁸ SARS‐CoV‐2, ²⁵⁹ and Zika virus ²³³ , ²⁶⁰ , ²⁶¹ , ²⁶² have been shown to promote infection of FcγR‐expressing cells at sub‐neutralizing concentrations through engagement of FcγRs. However, the biological significance of these findings is limited, given the lack of in vivo studies in animal models to support a pathogenic role for IgG antibodies and FcγR pathways in these infections. Likewise, with the exception of dengue disease, epidemiological studies in patient cohorts have not demonstrated an association between preexisting IgG antibodies with disease susceptibility or severity.

Apart from increasing the infection of FcγR‐expressing leukocytes, it has been proposed that one mechanism IgG antibodies might exacerbate disease pathogenesis is by inducing uncontrolled activation of pro‐inflammatory pathways through dysregulated IgG responses. This phenomenon, termed vaccine‐associated enhanced respiratory disease (VAERD), has been described primarily for respiratory viral pathogens, like RSV and influenza virus, and has been used to describe the increased susceptibility to symptomatic disease following vaccination of pediatric populations with certain immunogens. ²⁶³ , ²⁶⁴ In the context of RSV, immunologic analysis of vaccinees that received formalin‐inactivated RSV immunogens revealed that vaccine‐elicited IgG responses were characterized by poor neutralizing activity as a result of incorrect conformation of specific RSV antigens. ²⁶⁵ In in vivo mouse disease models, it has been demonstrated that RSV challenge following formalin‐inactivated RSV vaccination results in inappropriate airway inflammation and lung injury. These effects have been attributed to excessive complement‐mediated activation by immune complexes comprising non‐neutralizing IgG antibodies, as well as to dysregulated CD4 T‐cell responses and aberrant expression of Th2 cytokines. ²⁶⁶ , ²⁶⁷ , ²⁶⁸ Similar observations have been reported in immunogenicity studies of influenza virus vaccine candidates that elicited non‐neutralizing IgG antibody responses in pigs and ferrets, which contributed to IgG‐mediated inflammation and lung injury. ²⁶⁹ , ²⁷⁰ , ²⁷¹ , ²⁷² Apart from vaccine‐elicited IgG antibodies, a potential pathogenic role for FcγR‐mediated signaling has been suggested in the context of SARS‐CoV‐2 infection, as elevated levels of afucosylated Fc glycoforms have been reported in severe COVID‐19 patients. ⁴⁶ , ²⁷³ Given the heterogeneity of immune responses elicited upon infection or vaccination, it is likely that disease susceptibility and severity is the outcome of multiple complex immunologic determinants that modulate host antiviral immunity, rather than due to engagement and activation of specific FcγR pathways.

8. IMPACT OF ADE CONCERNS ON MAb DEVELOPMENT

Despite the evidence presented above, the mechanisms by which IgG antibodies contribute to pathogenesis, as well as the precise role (if any) of FcγR pathways in disease remain unknown. On the contrary, both clinical observations and experimental evidence from multiple in vivo animal disease models suggest that Fc‐FcγR interactions provide protective rather than pathogenic effects, and are critical for the antiviral protection mediated by IgG antibodies during infection. Indeed, numerous studies in small and large animal disease models have clearly demonstrated that both non‐neutralizing and neutralizing (even when tested at and sub‐neutralizing doses) antibodies against HIV‐1, RSV, SARS‐CoV‐2, Ebola virus, influenza virus, and other viral pathogens depend on Fc‐FcγR interactions for their antiviral activity and their administration is not associated with accelerated disease pathogenesis, increased viral replication, or IgG‐mediated inflammation. ⁶⁸ , ⁷¹ , ⁷² , ⁷³ , ⁷⁵ , ⁹³ , ¹⁰⁶ , ¹³³ , ¹³⁶ , ¹³⁷ , ¹⁴⁰ , ¹⁵³ , ²¹⁰ , ²⁵³ , ²⁵⁹ , ²⁷⁴ , ²⁷⁵ , ²⁷⁶ , ²⁷⁷ , ²⁷⁸ , ²⁷⁹ , ²⁸⁰ , ²⁸¹ , ²⁸²

Apart from in vivo studies, evidence from the clinical use of mAbs or polyclonal IgG against RSV, SARS‐CoV‐2, and Ebola virus supports a favorable safety profile with minimal pathological consequences. Indeed, prophylactic administration of either anti‐RSV mAbs (palivizumab) or polyclonal RSV immune globulin (RespiGam) in pediatric populations, as well as therapeutic administration of high doses of convalescent plasma or neutralizing anti‐SARS‐CoV‐2 mAbs in hospitalized COVID‐19 patients is not associated with worse disease outcomes. ²⁸³ , ²⁸⁴ , ²⁸⁵ , ²⁸⁶ , ²⁸⁷ Similarly, neutralizing anti‐SARS‐CoV‐2 mAbs with intact (e.g., sotrovimab) or diminished Fc effector functions (e.g., tixagevimab and etesevimab) exhibited comparable safety profiles in COVID‐19 patients. ²⁸⁵ , ²⁸⁸ , ²⁸⁹ Even when mAbs were Fc‐engineered for enhanced binding to activating FcγRs (either though glycoengineering or protein engineering), there was no evidence for increased disease pathogenesis or tissue damage in pre‐clinical disease models. ⁶⁹ , ⁷³ , ¹⁴⁰ , ¹⁶¹ Similarly, no adverse events suggestive of excessive IgG‐mediated inflammation have been reported in clinical studies evaluating glycoengineered (afucosylated) mAbs optimized for increased affinity for the activating FcγRIIIa (e.g., the anti‐Ebola mAb cocktail ZMApp ²⁹⁰ ) or the Fc‐optimized anti‐HIV (elipovimab) or anti‐HBV/HDV mAbs (tobevibart), which exhibit increased binding to activating FcγRs.

In summary, while in vitro assays have demonstrated that IgG antibodies can promote infection of FcγR‐expressing cells that are normally non‐permissive, neither in vivo infection models nor clinical evidence support a pathogenic role for antiviral IgG antibodies, even when assessed under conditions that favor in vitro ADE (e.g., sub‐neutralizing concentration, poor neutralizing activity, enhanced for activating FcγR engagement). These findings highlight the limited translational and biological relevance of established in vitro experimental systems that have been previously used to demonstrate ADE by antiviral antibodies. More importantly, the assumption that FcγR engagement might have pathologic consequences has impacted the development of anti‐SARS‐CoV‐2 mAbs. Despite evidence showing FcγR engagement is necessary for protective antiviral activities by these mAbs, several anti‐SARS‐CoV‐2 mAbs that are currently in clinical testing or use have been engineered to prevent FcγR interactions. As a result, these mAbs may not achieve maximal therapeutic activity and might lack the immunomodulatory effects needed to elicit protective adaptive immune responses, as their antiviral function relies entirely on Fab‐mediated neutralization, which is subject to viral escape.

9. CONCLUDING REMARKS OF FUTURE DIRECTIONS FOR MAb DEVELOPMENT

Antiviral protection by neutralizing antibodies was once considered to be entirely the outcome of Fab‐mediated antigenic recognition; however, recent advances have significantly improved our understanding of the important role played by Fc effector functions of IgG antibodies during infection. A substantial body of experimental data supports that in vitro neutralizing activity often does not predict the ability of antiviral antibodies to confer protection in vivo, prompting the systematic characterization of the Fc domain function in mAb‐mediated protection against infection. Technological and methodological advances in high‐throughput assays and in Fc domain engineering, as well as the development of animal models that recapitulate the unique structural and functional complexity of human FcγRs, facilitated mechanistic studies on the role of Fc‐FcγR interactions during infection. These studies revealed specific FcγR pathways that are associated with protection against infection, as well as defined the biological effects of FcγR engagement on innate and adaptive immune responses. It is now well‐established that engagement of FcγRs by IgG immune complexes, that are generated during an immune response to infection or vaccination, results in defined signaling cascades that mediate pleiotropic effector functions. These functions ensure the swift clearance of opsonized virions and infected cells through cytotoxic and phagocytic mechanisms, as well as the induction of protective T‐cell responses by modulation of APC functional activity. Understanding the role of the specific FcγR pathways that drive protective innate and adaptive antiviral immunity has led to the rational design and development of novel antibody‐based therapeutics currently in clinical testing (NCT06216470, NCT05484206, NCT04423393, NCT05461170, and NCT04856085). These therapeutics are engineered to selectively activate the precise FcγR pathways essential for conferring potent and durable protection against infection, ¹⁶¹ setting the paradigm for the future development of optimized mAbs with superior therapeutic efficacy against infectious pathogens.

CONFLICT OF INTEREST STATEMENT

The authors declare no competing interests.

Edgar JE, Bournazos S. Fc‐FcγR interactions during infections: From neutralizing antibodies to antibody‐dependent enhancement. Immunol Rev. 2024;328:221‐242. doi: 10.1111/imr.13393

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no new data were created or analyzed in this study.