The pathogenesis of IgA nephropathy and implications for treatment

Abstract

IgA nephropathy (IgAN) is a common form of primary glomerulonephritis and represents an important cause of chronic kidney disease globally, with observational studies indicating that most patients are at risk of developing kidney failure within their lifetime. Several research advances have provided insights into the underlying disease pathogenesis, framed by a multi-hit model whereby an increase in circulating IgA1 that lacks galactose from its hinge region — probably derived from the mucosal immune system — is followed by binding of specific IgG and IgA antibodies, generating immune complexes that deposit within the glomeruli, which triggers inflammation, complement activation and kidney damage. Although treatment options are currently limited, new therapies are rapidly emerging that target different pathways, cells and mediators involved in the disease pathogenesis, including B cell priming in the gut mucosa, the cytokines APRIL and BAFF, plasma cells, complement activation and endothelin pathway activation. As more treatments become available, there is a realistic possibility of transforming the long-term outlook for many individuals with IgAN.

Introduction

First described in 1968 by Jean Berger and Nicole Hinglais, IgA nephropathy (IgAN) is defined by the presence of dominant or co-dominant IgA deposition within the glomerular mesangium¹. IgAN is now recognized as the most common primary glomerular disease globally, and an important cause of kidney failure^2,3. It is most often diagnosed in adults in their second or third decades and can affect any age group, although it is rarely observed in children under 5 years old⁴. IgAN is associated with a wide spectrum of clinical manifestations, from asymptomatic haematuria and/or proteinuria to intermittent episodes of visible haematuria (often triggered by an upper respiratory tract infection), to progressive kidney function decline that can lead to kidney failure (typically over several years to decades). In rare circumstances (<5% of cases), patients with IgAN present with nephrotic syndrome or rapidly progressive glomerulonephritis.

IgA vasculitis with nephritis shares many features with IgAN, including indistinguishable kidney histology, but, in contrast to IgAN, IgA vasculitis with nephritis typically manifests in additional tissues, including skin, joints and the gastrointestinal tract. Differences in the underlying pathogenic pathways between these conditions are becoming better understood (reviewed in Pillebout⁵).

The estimated incidence of IgAN is at least 2.5 per 100,000 adults per year^2,6. Its true incidence and prevalence are unknown owing to several factors, including differences in access to health care (for example, accessibility and affordability of a kidney biopsy), screening practices for urinary abnormalities, thresholds for performing a kidney biopsy and accuracy of data records. The prevalence of IgAN increases in a gradient from West to East, and is especially common in East Asia, where it is diagnosed in approximately 40−50% of all kidney biopsies performed, compared with 10−20% in North America, up to 20% in Europe and <5% in Africa³ (Fig. 1). Gender distribution also varies — equal in Asian populations, but more prevalent in males than in females (~2:1 ratio) in North America and Europe. Histological lesions at presentation and rates of kidney disease progression are often more severe in Asian-Pacific populations and, given that these differences persist in migrant populations, genetic factors are likely to contribute to this effect^3,7.

Fig. 1 |. Geographical variation in the prevalence of IgA nephropathy.

Percentages represent the estimated proportion of cases of IgA nephropathy compared with all native kidney biopsies performed. Reprinted from ref. 190, Springer Nature Limited.

Although some patients have a benign disease course, particularly those with preserved kidney function and isolated haematuria, 50−75% of adults with IgAN develop kidney failure within 20 years of their diagnosis^8,9. An analysis of the UK National Registry of Rare Kidney Diseases (RaDaR) of 2,299 adults and 140 children with biopsy-proven IgAN, and estimated glomerular filtration rate (eGFR) <60 ml/min/1.73 m² or proteinuria >0.5 g/day, demonstrated that nearly all were at risk of developing kidney failure within their lifetime unless their annual rate of kidney function decline could be slowed to <1 ml/min/1.73 m², a rate that is often not achievable with the current standard of care⁹ (Fig. 2). In addition, mesangial IgA deposition recurs in ~50% of patients within 5 years of kidney transplantation and is associated with reduced graft survival¹⁰. A reduction in life expectancy of 6 and 10 years was reported in IgAN cohorts from Sweden and the USA respectively, mainly owing to complications of kidney failure^11,12.

Fig. 2 |. Long-term outcomes in IgA nephropathy.

Data from the UK National Registry of Rare Kidney Diseases⁹. Patients with biopsy-proven IgA nephropathy and proteinuria >0.5 g/day or estimated glomerular filtration rate (eGFR) <60 ml/min/1.73 m²) were enrolled; 2,299 adults and 140 children were included in the analysis. This study demonstrates that most patients are at risk of kidney failure within their lifetime and shows a strong relationship between time-averaged proteinuria and the risk of kidney failure or death. a, Kaplan−Meier survival curves of time to kidney failure or death event in adults and paediatric patients. b, Kaplan−Meier survival curves of time to kidney failure or death event, according to total follow-up time-averaged proteinuria (g/g). Reprinted with permission from ref. 9, Wolters Kluwer.

The UK RaDaR study also highlighted that a high proportion of children with IgAN eventually develop kidney failure⁹ (Fig. 2). Important differences exist between children and adults with IgAN regarding clinical presentation, kidney biopsy findings, treatment responses and long-term outcomes. A study of a multi-ethnic international cohort of 1,060 children with IgAN followed into adulthood demonstrated that their overall trajectory of eGFR differed from adults by being highly non-linear, increasing until ~18 years of age, followed by a linear decline that was similar to that observed in adults¹³. The earlier increase was postulated to be due to several factors, including a higher frequency of renal inflammatory lesions in children than in adults that might be more responsive to treatment, and a higher capacity for tissue repair¹³.

Overall, these data highlight IgAN as an important global health issue that places a considerable burden, not only on affected individuals and their carers but also on health care systems in both high- and low-resource regions. The unmet need for therapies that are safe and effective in the long-term is therefore substantial.

In this Review, we examine the latest insights into the pathogenesis of IgAN that have informed the development of several new therapies, many of which are likely to reach clinical practice over the next few years.

IgAN pathogenesis: the ‘multi-hit’ hypothesis

Current understanding of the pathogenesis of IgAN supports a ‘multi-hit’ hypothesis^14,15 (Fig. 3), whereby each hit is required for clinically apparent IgAN to occur. Elevated circulating levels of IgA1 lacking galactose residues from its hinge region O-glycan side chains — commonly termed galactose deficient-IgA1 (Gd-IgA1) — is a central finding in IgAN that has been replicated in several independent cohorts globally (Hit 1)^16–20. This alone is not sufficient to cause IgAN, as first-degree relatives of patients with IgAN can also have high levels of serum Gd-IgA1 without developing kidney disease²¹. In susceptible individuals, increased levels of Gd-IgA1 are associated with the presence of IgA and IgG antibodies that recognize the O-glycans in the hinge region (Hit 2)^22,23. Whether these are true autoantibodies that have developed owing to loss of tolerance or whether they represent cross-reactive antimicrobial antibodies is unknown. Self-aggregation of Gd-IgA1, binding of Gd-IgA1 to a range of serum proteins, and to Gd-IgA1-reactive IgA and IgG antibodies, results in the formation of circulating Gd-IgA1-containing immune complexes (Hit 3)²⁴. These complexes reach the kidney, where they have a propensity for mesangial deposition. Accumulation of Gd-IgA1 immune complexes leads to mesangial cell activation, proliferation and the release of pro-inflammatory and pro-fibrotic cytokines (Hit 4). These processes are accompanied and accelerated by activation of the complement system via the alternative and/or lectin pathways²⁵. Overall, these events result in podocyte injury, disruption of the glomerular filtration barrier, and filtration of IgA immune complexes and mesangial-derived cytokines, which lead to haematuria, proteinuria, tubulointerstitial inflammation and fibrosis, and ultimately progressive loss of kidney function.

Fig. 3 |. Pathogenesis of IgAN.

a, Priming of B cells takes place in mucosa-associated lymphoid tissue (1); the gut is illustrated here as an example. In IgA nephropathy (IgAN), a combination of genetic and environmental factors — including infections, food antigens, alterations in the microbiota and aberrant mucosal immune responses — prime B cells to express poorly galactosylated IgA1 (termed galactose-deficient IgA1 (Gd-IgA1)); the cytokines BAFF and APRIL promote IgA class switching. (2) Patients with IgAN have an increased circulating level of Gd-IgA1, either due to mis-trafficking of mucosa-derived Gd-IgA1⁺ B cells to the bone marrow or overspill of Gd-IgA1 from mucosal sites into the circulation. b, In the circulation, Gd-IgA1 is recognized by IgG and IgA1 antibodies specific to the Gd-IgA1 hinge region, which could be autoantibodies or cross-reactive antimicrobial antibodies. c, This recognition results in the formation of Gd-IgA1-containing immune complexes (3). d, In the kidney, Gd-IgA1-containing immune complexes deposit within the glomerular mesangium (4), leading to mesangial cell activation, proliferation, production of inflammatory mediators and extracellular matrix components, and complement activation, with resultant glomerular and tubulo-interstitial injury. Several therapeutics have been developed for treatment in IgAN that target different steps in the disease pathogenesis, depicted in the blue boxes.

Role of molecular forms of IgA

In humans and higher primates, IgA exists as two isoforms — IgA1 and IgA2 — that differ by the presence of an extended hinge region that is found only in IgA1 (ref. 26). Both IgA1 and IgA2 can form monomers (mIgA) or polymers (pIgA)²⁶. Most IgA is produced within mucosa-associated lymphoid tissue (MALT), which comprises immune cells responsible for antigen sampling and induction of IgA production, and is found in mucosal tissues, including the gastrointestinal, respiratory and urinary tracts. Out of all MALT sites, the gut-associated lymphoid tissue (GALT), which includes Peyer’s patches of the small intestine and isolated lymphoid follicles, produces the most IgA²⁷.

In combination with cellular and chemical barriers, IgA acts as a first line of defence against pathogens at mucosal surfaces while maintaining homeostasis with commensal microbiota. Antigens from the mucosal lumen are constantly sampled by antigen-presenting cells, including dendritic cells, or resident cells such as intestinal microfold (M) cells, in a process termed immune surveillance²⁷. Microbe-derived pathogen-associated molecular patterns (PAMPs) are recognized by pattern-recognition receptors, including Toll-like receptors, expressed on the cell surface of immune cells within the MALT. Here, antigens are taken up by subepithelial dendritic cells that — either directly or by activating T cells — promote the activation and differentiation of naive B cells into antigen-specific, class-switched B cells. A T cell-dependent (TCD) or independent (TCI) co-stimulatory signal is required for B cell priming, and the cytokines B cell activating factor (BAFF) and A proliferation-inducing ligand (APRIL) have a key role in promoting TCI IgA class switching.

IgA⁺ B cells leave the MALT and migrate via the lymphatic system and circulation to central lymph nodes, where they differentiate further before returning to mucosal effector sites, where they mature to become antibody-secreting plasma cells within the lamina propria. IgA is then produced, mainly as pIgA; this molecular form comprises at least two IgA monomers that are connected by a joining chain (J chain), which is a 17-kDa protein that forms disulfide bridges between cysteine residues of the α-heavy chain²⁶. pIgA binds to the polymeric Ig receptor (pIgR) on the basolateral surface of mucosal epithelial cells and undergoes transcytosis to the luminal surface. pIgR is then cleaved at the mucosal surface, and a portion of the receptor remains bound to pIgA as the secretory component, forming secretory IgA (sIgA) that protects the IgA molecule from enzymatic degradation by bacterial proteases²⁸. Within the mucosal lumen, sIgA prevents entry of pathogens and pathogen-derived antigens and toxins into the bloodstream through receptor engagement, steric hindrance and immune exclusion^29,30. An exception to this process might occur in coeliac disease, where increased expression of the transferrin receptor (CD71) in the intestinal epithelium has been reported; CD71 can interact with sIgA (with or without peptides from the wheat gluten protein gliadin) and promote apical−basal retrotranscytosis^31,32.

By contrast, most circulating IgA is monomeric (~90%) and of the IgA1 isotype, and is mainly produced by B cells within the bone marrow³³. Elevated levels of antigen-specific IgA can be detected in the serum following systemic immunization, indicating that it has a protective role in systemic immune responses³⁴.

IgA1 hinge region O-glycosylation

The IgA1 hinge region is located between the first and second constant domains of the α-heavy chain and contains 18 amino acids³⁵ (Fig. 4). As with other O-glycosylated proteins, the hinge region undergoes sequential O-glycosylation as a post-translational modification through the enzymatic addition of O-glycan side chains to threonine and/or serine residues. O-glycosylation is thought to block access of microbial IgA1 proteases to the proteolytically susceptible IgA1 hinge region. The susceptibility of the hinge region probably explains the increasing proportion of plasma cells secreting IgA2, which lacks a hinge region, from the small intestine through to the microorganism-rich large intestine³⁶.

Fig. 4 |. IgA1 structure and glycosylation.

a, The IgA1 molecule includes a hinge region between the CH1 and CH2 domains of the α1 heavy chain that contains serine (Ser) and threonine (Thr) residues, and these can be variably glycosylated. Initially, N-acetylgalactosamine (GalNAc) is added to serine or threonine, which can be further extended with galactose (Gal) or sialic acid (N-acetylneuraminic acid). GalNAc−Gal can be further extended with sialic acid in α2,3-linkage with Gal. b, IgA nephropathy (IgAN) is characterized by increased serum levels of IgA1 that lacks terminal galactose (that is, poorly galactosylated IgA1 glycoforms), which are termed galactose-deficient IgA1 (Gd-IgA1). IgG antibodies can recognize the exposed GalNAc and bind to form IgA1 immune complexes.

Of nine potential O-glycosylation sites in IgA1, usually three to six are O-glycosylated³⁷. This process is initiated by the addition of N-acetylgalactosamine (GalNAc) mediated by N-acetylgalactosaminyltransferase 2 (GalNAcT2). The O-glycan chain can then be extended through ß-1,3 linkage of galactose to GalNAc by core 1 ß-1,3-galactosyltransferase (C1GalT1), which requires the core 1,3-galactosyltransferase-specific molecular chaperone (COSMC) protein²⁴. Sialic acid can be attached to either galactose, or directly to GalNAc, which prevents the subsequent addition of galactose.

All individuals generate a variety of IgA1 O-glycoforms and variability in the glycosylation of the hinge region gives the IgA1 molecule conformational flexibility, thus enhancing its pathogen-binding ability. Microbe-derived PAMPs can promote a reduction in galactosylation of IgA1, which increases susceptibility of the IgA1 hinge region to microbial IgA proteases^38,39.

In IgAN, circulating levels of Gd-IgA1 are increased compared with those observed in healthy individuals and Gd-IgA1 is detected within glomerular deposits; total IgA levels are increased in ~50% of patients^{16–20,40–43}. Reduced activity of C1GalT1, lower expression of COSMC or increased sialylation of GalNAc can each contribute to increased Gd-IgA1 production, and these phenotypes are likely to be influenced by both genetic factors and local factors within the mucosal microenvironment, such as the cytokine milieu^{24,35,38,44–47}. For example, addition of IL-6 and IL-4 to the media of cultured IgA1-producing cells derived from patients with IgAN decreased the expression and activity of C1GalT1 and increased the expression of the sialyltransferase ST6GalNAc2, leading to increased production of Gd-IgA1 (ref. 45).

Gd-IgA1 has characteristics that are pertinent to the development of IgAN — it is prone to self-aggregation, binds to a plethora of serum proteins and is recognized by specific IgA and IgG antibodies, generating immune complexes that deposit in the kidneys⁴⁸. Gd-IgA1 immune complexes can also activate complement directly as they bind C3 and mannose-binding lectin (MBL), and complement breakdown products can be detected within circulating immune complexes and co-localize with IgA deposits^49,50. Reduced IgA1 O-galactosylation also influences its interactions with various IgA receptors, resulting in varying degrees of inflammation and induction of both innate and adaptive immune responses⁵¹.

High-resolution mass spectrometry studies showed that decreased IgA N-glycan sialylation and galactosylation, and increased bisection, were associated with reduced eGFR in IgAN⁵². Interestingly, this study demonstrated a correlation between lower IgA1 sialylation and higher Gd-IgA1 levels⁵². Decreased O-linked sialylation was also reported from pooled IgA1 samples from patients with IgAN compared with healthy individuals in a previous mass spectrometry study¹⁸. Further studies are needed to clarify the relationship between O-linked sialylation and galactosylation of the IgA1-hinge region in IgAN.

Of note, mouse IgA exists as a single isoform that lacks the extended O-glycosylated hinge region of human IgA1, and mice do not express several human IgA receptors. Consequently, findings from experimental mouse models and their relevance to human IgAN should be considered with some caution. To address this, humanized mouse models have been developed that express human IgA1 with and without the IgA receptor CD89 (ref. 53).

Sources of pathogenic Gd-IgA1

A longstanding debate exists regarding the origin of the pathogenic Gd-IgA1 implicated in IgAN. One hypothesis suggests that it is mucosal in origin, whereas a second, not necessarily mutually exclusive hypothesis, suggests that it is produced by mucosa-derived cells located in central sites such as the bone marrow.

Several lines of evidence support a mucosal origin for pathogenic Gd-IgA1 in IgAN⁵⁴. Flares of disease activity with visible haematuria and proteinuria are commonly associated with mucosal infections such as pharyngitis or gastroenteritis, suggesting that perturbations in commensal flora at mucosal surfaces (for example, due to viral or bacterial infections, or other environmental causes) are a trigger for IgA production by cells within the MALT. Alterations of the tonsillar and gut microbiome have been described in patients with IgAN compared with those with tonsillar hyperplasia or healthy individuals, respectively^55,56. In a transgenic mouse model of spontaneous IgAN that expresses both human IgA1 and human CD89, depletion of gut microbiota by broad-spectrum antibiotics prevented mesangial IgA1 deposition, glomerular inflammation and development of proteinuria without altering human IgA1 levels in serum, supporting a gut mucosa origin for pathogenic IgA^57,58. Furthermore, studies of BAFF-overexpressing transgenic mice, which develop a spontaneous IgAN-like disease, demonstrated that raising these mice in a germ-free environment (thereby preventing colonization by commensal flora) prevented glomerular IgA deposition; glomerular IgA deposition was re-established following the introduction of gut commensal flora⁵⁹.

In both healthy individuals and patients with IgAN, IgA1 from mucosal sites is typically polymeric and poorly O-galactosylated, which is similar to the Gd-IgA1 observed within circulating immune complexes and in mesangial deposits in IgAN. By contrast, the majority of circulating IgA1 is monomeric, heavily O-galactosylated and is produced by bone marrow-resident plasma cells. sIgA, which is produced exclusively at mucosal sites, can be detected at higher levels in serum of patients with IgAN compared with healthy individuals, and can also be detectable in mesangial IgA deposits, where its presence is associated with a worse prognosis^60–63. Compared with healthy individuals, numbers of circulating IgA⁺ B cells increase to a greater degree in patients with IgAN during episodes of upper respiratory tract infection, which is indicative of an exaggerated mucosal IgA response⁶⁴. Patients with IgAN had higher numbers of gut homing (C-C chemokine receptor 9 (CCR9)⁺ β7 integrin⁺) regulatory B cells, memory B cells and plasmablasts⁶⁵, and circulating Gd-IgA1⁺ λ⁺ B cells with homing receptors for the upper respiratory tract and gut⁶⁶. These findings suggest that production of Gd-IgA1 might be stimulated during respiratory or gut infections, supporting a mucosal origin⁶⁷. Further characterization of circulating Gd-IgA1⁺ B cell populations should help to clarify their source.

The process by which mucosal Gd-IgA1 enters the circulation remains unclear. One proposed mechanism is reverse trafficking (retrotranscytosis) of secreted IgA1 from mucosal surfaces, possibly mediated by dysregulation of the lymphotoxin β-receptor ligand LIGHT (also known as TNFSF14)⁶⁸. Compared with healthy individuals and patients with other forms of chronic kidney disease (CKD), those with IgAN had a higher relative abundance of mucin-degrading bacteria (including Akkermansia muciniphila), which could deglycosylate secreted IgA1 in the gut lumen⁴⁷. In a mouse model, deglycosylation of IgA1 by A. muciniphila was followed by retrotranscytosis, whereby IgA crossed from the intestinal epithelium into the circulation and underwent glomerular deposition⁴⁷.

An alternative hypothesis suggests that pathogenic IgA is produced by mucosa-derived B cells that have ‘mis-homed’ to the bone marrow⁶⁹. This concept was supported by the observation that, compared with healthy individuals, patients with IgAN had a higher proportion of IgA1-producing plasma cells expressing J chain mRNA in the bone marrow^70,71, and lower numbers of these cells in duodenal biopsy samples⁷². These observations support a unifying hypothesis that pathogenic IgA-producing cells in the bone marrow are mucosal in origin. Both mucosal and bone marrow IgA production might be important sources of pathogenic IgA⁷¹.

Emerging reports in other diseases suggest that the source of pathogenic IgA production should be revisited. Accumulating data show that IgA-producing plasma cells can be localized to non-mucosal tissue and might even be observed in typically immune-privileged sites such as the central nervous system. IgA-secreting plasma cells originating from the gut were detected in the central nervous system of mice with experimental autoimmune encephalomyelitis, where they might exert a protective role in neurological injury⁷³. This observation prompted experiments in the BAFF-overexpressing mouse model of IgAN, in which mucosa-derived IgA-producing cells targeting Neisseria meningitidis could be found in the kidneys following nasal infection with that bacteria⁷⁴. Moreover, in studies of the grouped ddY spontaneous IgAN mouse model, IgA⁺ plasmablasts accumulated in the kidneys, where they produced IgA that was directed against mesangial antigens, including βII-spectrin⁷⁵. Further work is required to determine if local production of pathogenic IgA might also be occurring in the kidneys of patients with IgAN.

Immune complex formation in IgAN

Circulating Gd-IgA1 is found in the serum, mainly within high-molecular-weight immune complexes²². Studies of these complexes demonstrated that patients with IgAN have increased levels of IgG and IgA antibodies that recognize and bind to GalNAc residues in the hinge region that are exposed by the absence of galactose^14,23. The J chain can be detected in circulating Gd-IgA1 immune complexes, demonstrating that the IgA is polymeric.

The causes of the development of anti-Gd-IgA1 antibodies are unclear but probably include a combination of genetic and environmental factors. Certain HLA polymorphisms are risk factors for IgAN and might predispose individuals to antibody responses to specific environmental pathogens or loss of tolerance^76,77. The related B cell survival mediators BAFF and APRIL, which can be found at higher concentrations in patients with IgAN, promote the generation and survival of B cells that produce both IgA and IgG^78,79. Several environmental microbes express the GalNAc motif within polysaccharides on their cell surface and B cells might be primed to express IgA and IgG directed towards these polysaccharides; such antibodies could cross-react with the hinge region of Gd-IgA1 (molecular mimicry hypothesis)⁴. Formation of these antibodies might be induced after infection with Epstein−Barr virus, respiratory syncytial virus, herpes simplex virus or streptococci (summarized in Knoppova et al.⁴). A somatic mutation within the complementarity-determining region 3 (CDR3) of the heavy-chain variable region, which resulted in an alanine to serine substitution, influenced IgG binding to Gd-IgA1 and was only observed in samples from patients with IgAN and not in those from healthy individuals⁸⁰. Whether this mutation was associated with loss of tolerance or was triggered in response to a microbial pathogen is not known.

IgA−IgG immune complexes might contribute to IgAN disease activity directly. In an early study of IgA vasculitis in children, all had elevated levels of IgA immune complexes, but only those with nephritis had increases in both IgA−IgA and IgA−IgG complexes⁸¹. In patients with IgAN, levels of circulating anti-Gd-IgA1 IgG, of the IgG1 and IgG3 subclasses, are high, increase during disease activity, can be found within mesangial deposits and in urine, and correlate positively with disease severity and worse kidney outcomes^64,82–85. Mesangial IgG co-deposition with IgA is associated with a worse prognosis⁸⁶. In vitro, higher molecular weight Gd-IgA1-containing immune complexes stimulated mesangial cell proliferation, whereas uncomplexed Gd-IgA1 did not⁸⁷. Although glomerular IgG deposition is not always observed on routine immunofluorescence microscopy, in a detailed confocal IgAN microscopy study that used an anti-IgG nanobody, the presence of small amounts of IgG were detected in all cases and this IgG was specific for Gd-IgA1 (ref. 88).

IgA antibodies specific for the IgA1-hinge region are also detected in IgAN but their contribution towards disease progression is unclear, and serum levels do not correlate with Gd-IgA1 concentration, unlike IgG antibodies⁸⁹. Gd-IgA1−IgM immune complexes might also have a role in disease pathogenesis⁹⁰. In vitro, these complexes induced mesangial proliferation that could be inhibited with a small glycomimetic compound that disrupted IgM binding to Gd-IgA1 (ref. 90). Mass spectrometry studies have revealed other components with potential biological activity within Gd-IgA1 immune complexes, including complement and extracellular matrix components, such as fibronectin and type IV collagen^91,92. Delivery of complement components to the glomerular mesangium might be an important mechanism by which Gd-IgA1-immune complexes contribute towards complement activation.

Circulating IgA1 immune complexes have been proposed to induce shedding of membrane-bound CD89 from myeloid cells, leading to the formation of IgA1−CD89 complexes that are prone to mesangial deposition and inducing mesangial cell activation. CD89 is an Fcα receptor expressed by myeloid cells that exists in membrane-bound and soluble (sCD89) forms. Two isoforms of sCD89 have been described in vivo — a shorter isoform present in both healthy individuals and patients with IgAN, and a longer isoform that was only detected in patients with IgAN⁹³. Increased levels of IgA−sCD89 complexes were reported in patients with recurrent IgAN after kidney transplantation, compared with those with IgAN without recurrence and healthy individuals⁹⁴. In children with IgAN, levels of IgA1−sCD89 complexes and free sCD89 correlated with proteinuria and the presence of inflammatory lesions on kidney biopsy⁹⁵. However, the overall significance of IgA−sCD89 complexes in IgAN disease pathogenesis remains unclear⁹⁶.

Hepatic catabolism of circulating Gd-IgA1 immune complexes is thought to be impeded both by the size of the complexes, which reduces access to Kupffer cells, and by reduced binding to the hepatic asialoglycoprotein receptor that recognizes terminal galactose molecules in the IgA1 hinge region⁹⁷. This impaired systemic clearance results in increased delivery of Gd-IgA1 immune complexes to the glomeruli, where the properties of Gd-IgA1 promote their deposition and accumulation within the mesangium.

Mesangial IgA deposits in IgAN

The glomerular deposition of IgA and its subsequent clearance are dynamic processes. For example, when kidneys containing IgA deposits were inadvertently transplanted into recipients without IgAN, subsequent clearance of the deposits could be demonstrated on repeat biopsy^98,99.

Early work showed that the deposited immune complexes comprised predominantly J chain-containing IgA1, with little IgA2 (ref. 100). Studies of IgA eluted from mesangial deposits, initially using lectin-binding assays and then mass spectrometry, confirmed that these immune complex deposits are enriched for polymeric Gd-IgA1 (refs. 100–102). The presence of sIgA within mesangial deposits has also been demonstrated⁶⁰.

Interactions between deposited IgA and mesangial cells lead to variable activation of inflammatory and fibrotic pathways, likely influenced by both the IgA-immune complex composition and individual susceptibility; ‘latent’ mesangial IgA deposition is well-recognized in kidneys with otherwise normal appearance, which suggests that additional factors determine the mesangial response to deposited IgA¹⁰³. Which mesangial cell receptors are responsible for binding IgA remains unclear, and they might include the transferrin receptor, sCD89, transglutaminase-2 and/or ß-1,4-galactosyltransferase-1 (refs. 53,93,104,105). Murine studies suggest that activation of mesangial cells by IgA1-immune complexes requires both sCD89 and tissue transglutaminase 2 (ref. 53).

Following deposition of IgA, an inflammatory reaction can ensue, resulting in mesangial cell proliferation, release of pro-inflammatory and pro-fibrotic cytokines and growth factors, and extracellular matrix remodelling (Fig. 5). pIgA is believed to be specifically involved in driving the glomerular inflammatory response. In vitro, pIgA bound to mesangial cells with much greater affinity than mIgA, and led to the release of IL-8, observed to the greatest extent with pIgA from patients with IgAN⁴⁹. pIgA N-glycans differ from those of mIgA and might promote complement activation via the lectin pathway⁴⁹. In a separate study, pIgA enhanced mesangial cell production of tumour necrosis factor, IL-6 and macrophage migration inhibitory factor¹⁰⁶. In rats, intravenous injection of engineered pIgA, but not mIgA, led to its glomerular deposition in association with mesangial cell proliferation, matrix expansion, albuminuria and haematuria¹⁰⁷. Following release of cytokines and chemokines by activated mesangial cells, inflammatory cells, including macrophages, are recruited, which promotes glomerular injury. Mesangial cell-derived cytokines can also cause podocyte injury. Reductions in glomerular podocyte nuclear density were associated with increased proteinuria, tubular atrophy and interstitial fibrosis, and lower eGFR in patients with IgAN¹⁰⁸. The ensuing disruption of the glomerular filtration barrier leads to haematuria and increasing levels of proteinuria. Filtration of mesangial cell-derived cytokines, and Gd-IgA1 immune complexes themselves, drives progressive tubulointerstitial inflammation and fibrosis, leading to progressive loss of kidney function.

Fig. 5 |. Pathological consequences of IgA immune complex deposition in IgAN.

Following the deposition of IgA1-containing immune complexes in the glomerular mesangium, an inflammatory reaction ensues with the release of pro-inflammatory and pro-fibrotic mediators. These mediators induce several histological changes that are characteristic of IgA nephropathy (IgAN) and inform the Oxford classification of IgAN — mesangial hypercellularity (M), inflammatory cell recruitment into glomeruli (endocapillary hypercellularity, or E), and an uncontrolled inflammatory response that can lead to crescent formation (C). Release of mesangial cell-derived cytokines leads to podocyte injury (glomerular−podocyte crosstalk), podocyte loss and segmental sclerosis (S). Moreover, disruption of the glomerular filtration barrier leads to haematuria, proteinuria and increased filtration of IgA1-containing immune complexes. Filtration of these complexes, coupled with mesangial cell-derived cytokines (glomerular−tubular crosstalk), subsequently lead to tubular atrophy and interstitial fibrosis (T), and ultimately nephron loss. Activation of the complement system has an important role in accelerating each of these processes.

Several downstream pathways are activated following the glomerular deposition of IgA. Cytoplasmic spleen tyrosine kinase (Syk) couples immune cell receptors with activation of intracellular pathways such as the c-Jun N-terminal kinase (JNK) and p38-MAP kinase signalling pathways. Glomerular phospho-Syk expression was increased in biopsy samples from patients with IgAN, compared with those from patients with minimal change disease¹⁰⁹. Endothelin pathway activation has also been observed in kidney biopsy samples from patients with IgAN and is associated with poor clinical outcomes¹¹⁰. Endothelin-1 (ET-1) acts via the endothelin-A (ETA) and ETB receptors, and can promote deleterious effects in the kidney, including vasoconstriction, mesangial cell proliferation, podocyte disruption, production of extracellular matrix, inflammation and fibrosis^110,111. Endothelin-1 and ETA receptor staining in the glomerular and tubulointerstitial compartments was increased in biopsy samples from patients with proteinuric IgAN compared with samples from patients with IgAN without significant proteinuria or from healthy individuals^{110,112–114}. Other pathways commonly engaged in proteinuric CKD are probably also activated in IgAN, including those involving the mineralocorticoid receptor, transforming growth factor-β and Wnt−β-catenin signalling.

Complement activation in IgAN

Similar to other glomerular diseases, complement activation is thought to have a major role in driving glomerular inflammation and injury in IgAN²⁵. Co-deposition of complement component C3 with IgA is observed in >90% of cases of IgAN, and its absence is associated with a more benign course of disease²⁵. Deposition of the glomerular membrane attack complex is often observed, which is indicative of intrarenal complement activation; its presence is associated with progressive IgAN¹¹⁵. Circulating C3 levels can be reduced in IgAN, with a corresponding increase in circulating levels of C3 degradation fragments, indicating that systemic complement activation may occur, in addition to that observed within the kidney⁴².

Given that C1q deposition is rarely observed, the classical pathway is not thought to have a major role in IgAN. Instead, abundant evidence has shown that the alternative and/or lectin pathways are activated²⁵. Alternative pathway components, such as properdin and complement factor H-related (FHR) proteins, are frequently observed within glomerular deposits in IgAN and their presence is associated with progressive disease¹¹⁶. A common genetic deletion of CFHR3 and CFHR1 is associated with protection against IgAN, probably because their encoded proteins antagonize factor H-mediated inhibition of the alternative pathway¹¹⁷. The reduced frequency of this deletion in the Chinese population might explain, in part, the increased frequency of inflammatory glomerular lesions seen in Chinese patients with IgAN¹¹⁸. The presence of lectin pathway components within the glomerulus, such as C4d, MBL and MBL-associated serine protease 1 (MASP1) and MASP2, is reported in up to 40% of patients with IgAN, and is associated with worse prognosis over long-term follow-up^119,120.

Genetic and epigenetic factors in the development of IgAN

Similar to other immune disorders, IgAN has a complex genetic architecture. Overc recent years, genome-wide association studies (GWAS) have emerged as a powerful ‘hypothesis-free’ tool to understand the genetic basis of complex traits. Several GWAS in predominantly European and East Asian cohorts identified 30 independent genome-wide significant risk loci for IgAN^121–124. Regarding disease susceptibility, these IgAN risk loci can account for ~11% of disease risk, and partly explain the geographical variation in disease prevalence¹²¹. Risk loci for IgAN were identified in genes responsible for IgA production, regulation of the immune and haematopoietic systems, and integrity of the intestinal mucosal barrier. Using phenome-wide associations, a positive correlation was observed between risk of IgAN and serum IgA levels, and risk loci for IgAN were also common to multiple other immune-mediated diseases, including rheumatoid arthritis, hypothyroidism and asthma, suggesting a shared polygenic architecture for these traits^121,125. Overall, findings from these GWAS identified the network of intestinal IgA production as a key pathogenic disease pathway, and highlight the role of both adaptive and innate immunity in the pathogenesis of IgAN¹²⁵. These data have helped to refine and expand the multi-hit model of IgAN¹²⁶.

These GWAS findings were incorporated into a genome-wide polygenic risk score (GPS), which was inversely associated with age at diagnosis, and was significantly associated with a higher lifetime risk of kidney failure¹²¹. Individuals with IgAN in the top 10% tail of the GPS distribution were at a 48% increased risk of developing kidney failure compared with the rest of the cohort, although the GPS was not a significant predictor for kidney failure at the time of kidney biopsy after adjusting for age, sex and other clinical factors¹²¹. A separate GPS, developed from SNPs associated with IgAN from a European cohort, was examined for its utility to assess the prevalence of IgAN. A higher GPS was found in patients from the UK Biobank with haematuria, hypertension and microalbuminuria, and it was estimated that IgAN could account for 19% of individuals with non-cancer haematuria and 28% of those with haematuria, hypertension and microalbuminuria in this cohort¹²⁷.

Separately, quantitative trait GWAS for serum levels of Gd-IgA1 revealed three distinct loci encoding molecular partners that are essential for the enzymatic O-glycosylation of IgA1 (refs. 128–130). Interestingly, none of these loci have been identified in GWAS to be associated with the risk of developing IgAN.

IgAN is a highly variable disease, and a wide spectrum of clinical and pathological features are observed, implying that IgAN might not be the same disease across the world. Indeed, IgAN might represent a histological end point towards which different pathogenic pathways converge. The relative importance of each ‘hit’ of the multi-hit hypothesis is likely to vary between populations of different ancestries, resulting in differences in the presentation and clinical course of IgAN^118,131,132. For example, an SNP in the non-coding region of C1GALT1 contributes to increased Gd-IgA1 expression in healthy and IgAN populations. This SNP was found at a lower frequency in people of Chinese ancestry compared with those of European ancestry, and this difference was associated with lower serum levels of Gd-IgA1 in a Chinese cohort, despite known higher rates of disease progression within this population^128,129. Elucidating the contribution of genetic factors to the diversity of clinical and pathological presentations in IgAN requires further work. In the future, successful management of IgAN is likely to rely on the use of targeted therapies that are based on the knowledge of an individual’s risk of disease progression and on their underlying pathophysiology and responsiveness to a specific treatment. Genetic factors are likely to contribute substantially to each of these factors, and a key advantage is that the genome can be analysed long before clinical disease becomes apparent.

Epigenetics acts as a bridge between genotype and phenotype, helping to explain why some genetic alterations do not necessarily result in an altered phenotype. Covalent modifications of DNA and histone proteins, such as DNA methylation, and the action of non-coding RNAs, such as microRNAs (miRNAs), can modify gene expression without altering the genome. Several studies have identified altered expression of miRNAs in IgAN, including miRs 148b, 374b and let-7b, that are involved in regulating gene expression of enzymes with key roles in O-glycosylation of the IgA1 molecule^{126,133–135}, and miR-150-5p and miR-204 that regulate mediators of kidney fibrosis¹³⁶. If validated, these findings could in future be exploited for the development of novel biomarkers or targeted therapies.

Biomarkers in IgAN

There remains a lack of validated non-invasive biomarkers to help determine the diagnosis of IgAN, risk of disease progression, optimal treatment selection and probable response to treatment. Proteinuria, eGFR and blood pressure are used in clinical practice to stratify the risk of progression, but changes in these clinical parameters are often only detected after considerable, and often irreversible, kidney damage has occurred¹³⁷. Remission of haematuria has been proposed as a marker of favourable prognosis, but this requires further validation in prospective interventional studies¹³⁸. The Oxford classification describes histological features in IgAN (Fig. 5) that are each independently associated with worse prognosis (mesangial hypercellularity, endocapillary hypercellularity, segmental glomerulosclerosis, tubular atrophy and interstitial fibrosis, as well as glomerular crescents)^139,140. The subsequently developed International IgAN risk prediction tool (IIgANRPT) combines clinical parameters and kidney histology findings¹⁴¹. This tool has been widely validated and accurately predicts an individual’s risk of a 50% decline in eGFR or of kidney failure for up to 7 years after kidney biopsy. However, both the Oxford classification and IIgANRPT have not been prospectively validated for treatment decisions. Mechanistic insights into IgAN have increased the pool of candidate biomarkers but, to date, none has been validated for clinical use^137,142. This might reflect the complex pathogenesis and variability of IgAN, and a single diagnostic or prognostic biomarker is unlikely to emerge, in contrast, for example, to anti-phospholipase A2 receptor antibodies in primary membranous nephropathy.

Raised serum levels of Gd-IgA1 and IgG antibodies specific to Gd-IgA1 are observed in patients with IgAN compared with healthy individuals, and were associated with worse kidney survival in retrospective cohort studies^41,83,84. However, levels of these antibodies in patients with IgAN and healthy individuals can overlap significantly, and these tests lack the sensitivity and specificity required for clinical use^43,143–145. Technical issues also exist, complicating the replicability of these assays. As discussed, miRNAs are increasingly recognized to have a role in IgAN pathogenesis, and serum let-7b and miR-148b levels are reported to be raised in IgAN¹⁴⁶. Adding a panel of miRs (135a-5p, 146b-5p, 150-5p, 155-5p and 204) to the IIgANRPT improved its risk prediction performance in single-cohort pilot studies^136,147. These results require further external validation, but demonstrate how novel biomarkers could be used to refine existing risk prediction tools.

The need for biomarkers in IgAN has become more pressing with the development of therapies that target specific components of the pathogenic cascade. Numerous phase II and phase III clinical trials are now being conducted for therapies that target the mucosal immune system, B cells and plasma cells, the complement system and downstream pathways that are activated following IgA deposition. Validated biomarkers are urgently needed to help determine the optimal treatment choice for a given individual. Sample biorepositories are being collated within these clinical trials and, together with other biobanking efforts, including through CureGN, the UK National Institute for Health and Care Research BioResource and the IgAN kidney atlas, will lead to planned multi-omics investigations with the potential to reveal new insights into the disease pathogenesis and biomarkers that can be used to determine the likelihood of treatment response at an individual level.

Key therapeutic targets in IgAN

According to the 2021 Kidney Disease

Improving Global Outcomes guidelines, first-line treatment of IgAN is based on optimized supportive care, including maximized renin−angiotensin−aldosterone system blockade¹⁴⁸. Acknowledging the paucity of safe and effective therapies at the time, it was recommended that any patient who remained at a high risk of disease progression (that is, those with persistently high levels of proteinuria >0.5−1 g/day), despite optimized supportive care, should be offered the opportunity to participate in a clinical trial. Where this is not possible, consideration can be given to a course of systemic glucocorticoids in selected individuals after a thorough toxicity risk assessment has been performed. Systemic glucocorticoids address one aspect of IgAN (glomerular inflammation) but are associated with substantial adverse effects that preclude their long-term use and might only provide temporary beneficial effects on proteinuria and kidney function, with the original disease trajectory resuming once treatment is stopped^149,150. Therefore, there remains a crucial unmet need for effective and safe therapies that can alter the long-term disease course in IgAN.

Over 20 new therapies are in clinical development for the treatment of IgAN, and two treatments — targeted-release formulation (TRF)-budesonide and sparsentan — have been approved^15,151,152 (Table 1). This surge in interest has been driven by both an increased understanding of the underlying disease pathogenesis, and the recognition of a strong association between early reductions in proteinuria and kidney outcomes, which led regulators (including the FDA, the EMA, the UK Medicines and Healthcare products Regulatory Agency and the Chinese National Medical Products Administration) to accept proteinuria reduction as a reasonably likely surrogate for a treatment’s effect on progression to kidney failure in IgAN, and as an end point in clinical trials¹⁵³.

Table 1. Latest treatments to be approved or in clinical development for IgAN.

Agent	Sponsor	Target	Type	Phase (trial name)	Status	Ref.
Gut mucosa B cell priming
Nefecon	Calliditas	GALT	TRF-budesonide	Approved	Completed	160
B cell-directed therapies
Sibeprenlimab (VIS649)	Otsuka Pharmaceutical	APRIL	mAb	III (VISIONARY)	Active, not recruiting	164
Zigakibart (BION-1301)	Chinook/Novartis	APRIL	mAb	III (BEYOND)	Recruiting	165
Atacicept	Vera Therapeutics	APRIL + BAFF	mAb	III (ORIGIN 3)	Recruiting	168
Telitacicept	RemeGene	APRIL + BAFF	mAb	III	Recruiting	182
Povetacicept	Alpine Immune Sciences	APRIL + BAFF	mAb	II (RUBY-3)	Recruiting	183
Mezagitamab	Takeda	CD38	mAb	II	Active, not recruiting	156
Felzartamab	HI-Bio	CD38	mAb	II (IGNAZ)	Active, not recruiting	155
Complement inhibitors
Iptacopan (LNP023)	Novartis	Factor B	Small molecule	III (APPLAUSE-IgAN)	Active, not recruiting	170
RO7434656	Hoffman-La Roche	Factor B	Anti-sense oligonucleotide	III (IMAGINATION)	Recruiting	184
Vemircopan (ALXN2050)	Alexion	Factor D	Small molecule	II (RESTORE-D)	Recruiting	185
ARO-C3	Arrowhead	C3	RNA interference	I/II	Recruiting	186
Ravulizumab	Alexion	C5	mAb	II (SANCTUARY)	Recruiting	187
Cemdisiran	Alnylam	C5	Small interfering RNA	II	Completed	188
Avacopan (CCX168)	Chemocentryx	C5aR1	Small molecule	II	Completed	189
Endothelin receptor antagonists
Sparsentan	Travere Therapeutics	DEARA	Small molecule	Approved	Completed	176
Atrasentan	Chinook/Novartis	ERA	Small molecule	III (ALIGN)	Active, not recruiting	177

DEARA, dual endothelin angiotensin receptor antagonist; mAb, monoclonal antibody; TRF, targeted-release formulation. Data from www.clinicaltrials.gov (accessed on 9 July 2024).

New approaches to targeting the production of pathogenic IgA

B cell or plasma cell depletion

Although CD20⁺ B cell depletion with rituximab failed to show benefit in a small open-label study in IgAN, other B cell depletion strategies are currently being explored¹⁵⁴. Felzartamab and mezagitamab are CD38-directed therapies that target and deplete plasma cells, and are being studied in phase II studies in IgAN^155,156.

B cell or plasma cell modulation

The mucosal immune system, and especially the GALT, is believed to be a major source of circulating Gd-IgA1 in IgAN and therefore represents a therapeutic target. TRF-budesonide (also known as Nefecon) was developed to release its highest concentration of active drug in the terminal ileum, where mucosal IgA synthesis within Peyer’s patches is concentrated, after which the drug undergoes extensive first-pass metabolism by the liver, minimizing its systemic effects. In a phase II trial, treatment with Nefecon in patients with IgAN reduced levels of circulating Gd-IgA1, IgA−IgG immune complexes, sIgA and BAFF, and led to significant proteinuria reduction in treated individuals compared with those receiving placebo^157,158. Results from the phase III NefIgArd trial demonstrated that a 9-month course of 16 mg Nefecon significantly reduced proteinuria and preserved kidney function compared with placebo at 12 months, with sustained improvements observed at 2-year follow up^159,160. Based on these results, Nefecon became the first drug to be approved by the FDA and the EMA for the treatment of IgAN.

The production of IgA, and therefore Gd-IgA1, depends on the B cell survival factors APRIL and BAFF, which can be found at higher levels in patients with IgAN than in healthy individuals^78,79. Anti-APRIL therapies are currently being evaluated in IgAN and include sibeprenlimab and zigakibart. Treatment with sibeprenlimab was tested in a dose-finding phase II trial of 155 patients with IgAN, and significantly reduced Gd-IgA1 and IgA levels, as well as proteinuria, and stabilized eGFR at 12 months, in contrast to the placebo group¹⁶¹. Effects on IgG were less pronounced, which is important given the known association between IgG hypogammaglobulinaemia and infection¹⁶². Vaccine responses, including to SARS-CoV-2 mRNA vaccines, were preserved and no concerning safety issues were noted. A small open-label phase II study of zigakibart has been performed, with reported beneficial effects on Gd-IgA1 levels and proteinuria¹⁶³. Both sibeprenlimab and zigakibart are now being studied in large global phase III trials^164,165. Targeting both BAFF and APRIL is an alternative approach that might also be beneficial, and a number of agents are now being evaluated in IgAN (including atacicept, telitacicept and povetacicept) that contain parts of the extracellular portion of the receptor TACI (also known as TNFRSF13B), which binds both cytokines, thereby inhibiting their actions. The phase IIb ORIGIN trial included 116 patients and demonstrated the efficacy of atacicept in reducing Gd-IgA1 and proteinuria levels and in stabilizing kidney function compared with placebo, as well as its safety; atacicept is now being tested in a phase III trial^166–168.

New approaches to targeting immune complex-mediated glomerular inflammation

Once glomerular deposition of Gd-IgA1 occurs, activation of complement has a key role in driving kidney inflammation and damage²⁵. Several complement-directed therapies are being developed that target the alternative, lectin and/or common terminal pathways. Treatment with iptacopan, which is an oral Factor B inhibitor, resulted in a dose-dependent reduction in proteinuria in a phase II study in IgAN, with evidence of rapid blockade of the alternative pathway; a phase III study is now in follow-up^169,170. Narsoplimab, which is a MASP-2 inhibitor that blocks the initiation of the lectin pathway, lowered proteinuria in a small open-label study of patients with IgAN¹⁷¹. However, a global phase III trial was terminated early, as a pre-planned interim analysis did not demonstrate significant proteinuria reduction with narsoplimab treatment compared with placebo. Further analysis of this study is awaited. Other complement inhibitors being actively studied in IgAN include ravulizumab (C5 inhibitor), ARO-C3 (C3 inhibitor), RO7434656 (Factor B inhibitor) and vemircopan (Factor D inhibitor)²⁵.

New approaches to managing the consequences of IgA immune complex-mediated nephron loss

SGLT2 inhibitors

Given that IgAN is a common cause of glomerular disease and CKD, large numbers of patients with IgAN were included in the DAPA-CKD and EMPA-KIDNEY trials of sodium−glucose co-transporter 2 (SGLT2) inhibitors in non-diabetic CKD. Treatment with SGLT2 inhibitors led to a significant benefit in the IgAN cohorts in both progression of kidney disease and survival^172,173. As well as their known effects on tubuloglomerular feedback, several beneficial pleiotropic effects of SGLT2 inhibition have now been described, including a postulated anti-inflammatory effect in the kidneys¹⁷⁴. The use of SGLT2 inhibitors is now becoming more established in IgAN, especially as part of CKD management.

Endothelin receptor antagonists

The endothelin system is upregulated in IgAN and promotes a number of deleterious haemodynamic and intraglomerular effects¹¹¹. Sparsentan is a dual endothelin A receptor and angiotensin receptor antagonist. A phase III trial in IgAN demonstrated that treatment with sparsentan resulted in a marked reduction in proteinuria compared with ARB alone, with improvements in chronic eGFR slope and absolute eGFR change at 2 years^175,176. Sparsentan was the second drug to be approved by the FDA and EMA for the treatment of IgAN. Atrasentan is an endothelin A receptor antagonist that has been tested extensively in diabetic nephropathy, and a phase III trial in IgAN is in follow-up¹⁷⁷.

Mineralocorticoid receptor antagonists

Finally, FIND-CKD is a trial of the non-steroidal mineralocorticoid receptor antagonist finerenone in non-diabetic CKD, which has also enrolled large numbers of patients with IgAN and is currently in follow-up¹⁷⁸. Finerenone may become another addition to supportive care.

Future directions

Multiple agents are currently in the late stages of development for IgAN, and several of these are likely to receive regulatory approval soon. Determining which drug will be suitable for an individual at a given time point in their disease will require careful analysis of the emerging data, and the development of reliable non-invasive biomarkers to assess disease activity and likelihood of treatment response. Suppression of pathogenic IgA production by targeting the mucosal immune system, the cytokines APRIL and BAFF, or CD38⁺ plasma cells, in combination with therapies to rapidly suppress glomerular inflammation (for example, through complement inhibition) are emerging as exciting treatment approaches¹⁷⁹. A combination or sequence of therapies to rapidly induce and then maintain remission, akin to the management of other glomerular diseases such as antineutrophilic cytoplasmic antibody (ANCA)-associated vasculitis or lupus nephritis, might become a new treatment paradigm for IgAN. However, the safety and any added efficacy to a combined approach will need to be fully evaluated. With increasing treatment options, the clinical trial landscape in IgAN will become more challenging and inclusion criteria for these, including proteinuria thresholds, might need to be revisited.

Additional approaches for potential future novel therapies include modulation of the gut microbiome, degradation of Gd-IgA1 with IgA1 proteases, disruption of Gd-IgA1 immune complex formation and inhibition of IgA1 binding to mesangial cells. For example, non-absorbable oral antibiotics that alter the gut microbiome (e.g. rifaximin), IgA1 proteases and anti-CD89 antagonists that block sCD89 release, and potentially mesangial IgA binding, are all in early stages of clinical development in IgAN^58,180,181.

Conclusion

Several advances in our knowledge of the underlying pathogenesis of IgAN have been made over the past few decades. Developments in molecular techniques should help to answer outstanding questions, including the source of Gd-IgA1 in IgAN and the underlying pathogenic mechanisms that determine the heterogeneity in disease course that is observed across individuals. Harnessing collaborative efforts between academic centres and industry in biobanking samples and their future analyses will be crucial to these efforts, and may enable future refinement of diagnostic criteria within IgAN, as well as advancing the discovery of biomarkers and effective disease-specific therapeutics.

Key points.

• IgA nephropathy (IgAN) is an important cause of progressive kidney disease and kidney failure globally, with most patients being at risk of developing kidney failure within their lifetime.

• Advances in the understanding of the pathogenesis of IgAN have highlighted an (auto)immune basis for the disease, with increased circulating levels of galactose-deficient IgA1 (Gd-IgA1) being associated with the presence of IgA and IgG antibodies specific to these IgA1 O-glycoforms.

• The circulating Gd-IgA1 that forms immune complexes and is deposited within the glomeruli in IgAN is probably mucosal in origin.

• The presence of elevated levels of Gd-IgA1 alone is insufficient to trigger IgAN; genetic and epigenetic factors contribute to the susceptibility of developing IgAN and the risk of progressive disease.

• Several therapies that target mucosal B cell priming, B cell production of Gd-IgA1, complement activity and the endothelin system are in development for the treatment of IgAN.