C3 Glomerulopathy: A Current Perspective in an Evolving Landscape

Complement 3 (C3) glomerulopathy (C3G) is understood as a disease caused by dysregulation in the alternative pathway (AP) of the complement system. The current disease classification of C3G was adopted in response to the recognition that the preexisting morphology-based classification did not reflect the underlying pathogenetic drivers of disease. The previous system focused on the histologic pattern of membranoproliferative glomerulonephritis (MPGN), but biochemical and genetic analysis has supported the current understanding that C3 dominance on immunofluorescence (IF), rather than a specific histologic pattern, is most indicative of AP dysregulation [1‒3]. IF with intense C3 dominant staining is observed across a wide spectrum of histologic patterns in addition to the MPGN pattern, including mesangioproliferative, diffuse proliferative, and crescentic glomerulonephritis [4, 5]. Due to these findings, the previous classification system of MPGN was deemed fundamentally outdated. The 2013 C3G Consensus Report established the definition of C3G with diagnosis requiring a kidney biopsy showing a C3-dominant glomerulonephritis in which the IF shows C3 staining ≥2 orders higher than any other immune reactant [6]. The IF parameters were selected with the goal of identifying patients with glomerulonephritis secondary to AP dysregulation, while avoiding the misidentification of disease secondary to immune complex deposition [6].

C3G can be subdivided into C3 glomerulonephritis (C3GN) and dense deposit disease (DDD). These are differentiated solely by differences in electron microscopy (EM) morphology. The diagnosis of DDD requires the pathognomonic EM findings of dense, osmiophilic, “ribbon-like” deposits that are typically intramembranous, but may also be found within the mesangium, Bowman’s capsule, or tubular basement membrane [7, 8]. DDD encompasses all cases previously defined as MPGN type II, as well as any other case that fits the definition regardless of the histologic pattern. In practical terms, C3G cases without the characteristic EM findings of DDD are classified as C3GN. Making such a distinction based solely on EM interpretation is not always clearcut despite the seemingly distinct morphological appearance seen with DDD [8]. C3GN includes cases previously classified as type I and type III MPGN, and any other glomerulonephritis that meets the C3-dominant IF criteria. Electron deposits in C3GN are variable in terms of size, shape, and distribution and can localize in the mesangial, subendothelial, subepithelial, and intramembranous space [8].

DDD is often used as the “gold standard” for diagnostic validation studies because there is strong agreement that the distinct EM findings are pathognomonic for AP dysregulation. Mass spectrometry following laser microdissection of the deposits in DDD glomeruli identified components of both the AP and terminal pathway (TP) [9]. When using DDD as a “gold standard” comparator, it is evident that there are nuances and potential limitations to the use of IF as the main diagnostic arbiter. In a cohort of 42 patients with DDD, roughly 10% of patients did not meet the IF-based C3G criteria due to the presence of IgG and/or IgM [10]. This suggests that IF has limitations as a diagnostic tool in differentiating glomerular disease driven by immune complex deposition versus a defect in the complement system. This limitation is best exemplified by the identification of underlying AP dysregulation in patients with primary/idiopathic immune complex MPGN (IC-MPGN). In MPGN patients with repeat kidney biopsies, up to 17% may switch IF patterns between diagnoses of C3G and IC-MPGN or vice versa [10]. Idiopathic IC-MPGN are, for the most part, cases with an MPGN pattern on kidney biopsy that do not meet C3G criteria due to the degree of immunoglobulin staining on IF. Importantly, these patients have no identifiable secondary cause for IC-MPGN. In a cohort of 140 patients with either C3G or idiopathic IC-MPGN, genetic or acquired abnormalities in the AP were identified in 65% and 56% of patients, respectively [11]. In a separate cohort that included patients with idiopathic IC-MPGN, 53.6% of 41 patients tested positive for a serum C3 nephritic factor (C3NeF), an acquired autoantibody that mediates dysregulated complement activation [5]. These similarities between patients with C3G and idiopathic IC-MPGN highlight the need to identify the current classification system as outdated. Iatropoulos et al. [1] have proposed a new classification system that segregates patients into clusters according to common clinical phenotypes. In this classification system, C3GN, DDD, and idiopathic IC-MPGN patients were segregated into 4 groups by an unsupervised hierarchical cluster analysis based on 34 demographic, clinical, genetic, biochemical, and histologic variables. While the practical utility of this proposed system is unclear, it does provide insight into the future of AP dysregulation-mediated glomerulonephritis.

C3G is a heterogenous disease with different underlying pathogenetic etiologies. These include (1) genetic mutations/variants, (2) autoimmune disease/acquired autoantibodies, (3) monoclonal immunoglobulin-mediated disease, and (4) C3G without an identified cause. The broad range of clinical presentations in C3G reflects this heterogeneity. Onset of diseases occurs in both pediatric and adult populations with a mean age of diagnosis of 9 and 39 years, respectively [12]. In the subset of patients with monoclonal immunoglobulin-associated C3G (MIg-C3G) patients, the majority are older than 50 years [13]. The clinical presentation of C3G varies across a wide spectrum, from asymptomatic hematuria to a crescentic rapidly progressive glomerulonephritis. The majority of patients present with hematuria and proteinuria in the setting of preserved kidney function or chronic kidney disease [4, 7]. Roughly, 30–40% will present with nephrotic syndrome and crescentic rapidly progressive glomerulonephritis in less than 10% of cases [4, 5].

Genetic mutations and rearrangements in complement genes account for a minority of cases. Roughly, 10–20% of cases may be attributed to genetic mutations compared with an estimated 40–80% of cases due to acquired autoimmune causes [4, 5, 12, 14, 15]. In a cohort of 134 patients with either C3G or IC-MPGN, 24 had complement gene mutations, 13 of which tested positive for a circulating C3NeF, suggesting the presence of genetic mutations and acquired autoantibodies are not necessarily mutually exclusive [5]. Genetic mutations in C3G are either gain of function mutations in the AP complement cascade or loss of function mutations in AP regulatory proteins. These mutations typically exhibit non-mendelian inheritance, although there are rare cases of autosomal recessive and dominant inheritance [16]. Mutations have been identified in genes that code for various components and regulators of the complement system including C3, factor H (CFH), factor I (CFI), and factor B (CFB) (see Table 1) [5, 11, 17]. Over 60% of mutations occur in genes coding for components of the AP C3 convertase, an essential enzyme involved in the complement cascade [11]. Interestingly, mutations in CFH, a key regulator of the AP pathway, have been identified in both C3G and atypical hemolytic uremic syndrome patients, suggesting that monogenic mutations are unlikely to be the sole determinant for disease expression [16]. Polymorphisms and genetic variants across complement genes may have additive effects in determining the phenotypic complement profile of an individual. The collective effects and interactions among multiple polymorphisms compose what is sometimes referred to as a “complotype,” which in theory ultimately determines risk for disease [16, 18, 19].

C3G and idiopathic IC-MPGN cases attributed to autoimmunity are thought to arise in a permissive genetic background that allows for the formation of pathologic autoantibodies that dysregulate the AP. Whole-genome sequencing of 146 patients with C3G/idiopathic IC-MPGN found certain HLA types to be associated with disease when compared to controls [21]. Nephritic factors compose the class of autoantibodies most commonly considered responsible for the AP dysregulations seen in C3G/idiopathic IC-MPGN. Nephritic factors are a heterogenous group of autoantibodies that function to augment the complement system through stabilization of activating convertase enzymes which are involved in the amplification step of the complement pathway [22]. These are typically polyclonal IgG or IgM antibodies that bind to neoepitopes found only on assembled convertase protein complexes, thereby stabilizing and prolonging the half-life of these enzymes [22, 23]. The C3NeF stabilizes the AP C3 convertase and can be present in roughly 50% of C3G/idiopathic IC-MPGN patients and up to 80% of DDD patients, although the prevalence appears to be lower in adult DDD patients [5, 11, 15, 24]. Certain C3NeF that is properdin-dependent is capable of stabilizing both the C3 and C5 convertases of the AP, often leading to low serum C3 levels as well as increased TP activity [15, 22]. Autoantibodies that stabilize the C5 convertase, such as the properdin-dependent C3NeF, are referred to as C5 nephritic factors (C5NeF). These autoantibodies are responsible for TP activation and are commonly detected in C3G patients with a higher prevalence in C3GN (88%) compared to DDD (37%) [15, 22]. C4 nephritic factors are responsible for stabilizing the convertases of the classical pathway (CP) and lectin pathway (LP). These autoantibodies are detected in up to 10–15% of C3G/idiopathic IC-MPGN patients and can coexist with other nephritic factors [25]. Other less common autoantibodies target epitopes on complement components such as CFH, CFB, and C3b [13, 26].

MIg-C3G is a subtype of C3G that results from the presence of a monoclonal immunoglobulin capable of directly or indirectly activating the AP which in turn results in complement-mediated glomerular injury. It can occur in the setting of a hematologic malignancy such as chronic lymphocytic leukemia (CLL), B-cell lymphoma, and multiple myeloma [27, 28]. It also commonly occurs in the setting of a monoclonal gammopathy that does not meet criteria for a malignancy, and is one of several entities classified as a monoclonal gammopathy of renal significance. Unlike most forms of monoclonal gammopathy of renal significance, in MIg-C3G there is C3 deposition in the absence of monotypic immunoglobulin on the kidney biopsy as the underlying paraprotein causes AP dysregulation in the fluid phase, rather than on a cell surface. Monoclonal gammopathy is strongly associated with C3G with a disproportionately higher incidence compared with the general population. C3G patients aged 50 years and older had an associated monoclonal gammopathy in 65.1% of cases compared to 3.2% in the general population [13, 29]. The incidence of monoclonal gammopathy in a cohort of patients with DDD was 71.4% (10/14) in patients aged 49 years or older, an unexpected finding given the lower incidence of DDD compared to C3GN in adult patients [30]. Several case reports have demonstrated in vitro inhibition of CFH by purified monoclonal intact immunoglobulin and monoclonal lambda light chains derived from patients with MIg-C3G [31‒33]. Chauvet et al. [34] isolated purified IgG from 41 MIg-C3G patients and reported the presence of both monoclonal and polyclonal autoantibodies capable of AP activation suggesting MIg may be indirectly involved. MIg-C3G patients had a much higher prevalence of anticomplement receptor 1 antibodies, a lower prevalence of C3NeF and C5NeF, and significantly less genetic abnormalities compared to C3G patients without monoclonal gammopathy [34]. Zand et al. [35] performed laser dissection and mass spectrometry on a MIg-C3G patient with CLL and reported evidence of AP, TP, and surprisingly a significant amount of C4 suggesting a potential role of the CP/LP. The exact mechanism and degree to which monoclonal immunoglobulin contributes to AP dysregulation are still largely speculative. However, accumulating evidence continues to support the notion that MIg-C3G is a distinct subtype of C3G with clinically distinct features.

A significant proportion of patients with C3G lack an identifiable driver of disease. Significant methodological heterogeneity in identifying genetic or acquired causes of C3G exists across studies. Due to this, it is difficult to estimate the prevalence of cases of C3G without an identifiable cause. In a recent publication, Hauer et al. [20] reported no genetic or acquired cause for C3G in 43% of patients; however, the prevalence of monoclonal gammopathy was not reported in this cohort. Ultimately, this subset of patients may represent individuals that possess genetic mutations/variants or acquired autoantibodies that have not been discovered. Some of these individuals may carry autoantibodies that are transient or cannot be detected by the limited sensitivity of current assays. Servais et al. [5] reported fluctuations in C3NeF in 32% of patients that initially tested positive. Similarly, monoclonal paraproteins may exist at concentrations below the limit of detection by the assays used frequently in practice. Mass spectrometry-based assays have vastly improved the sensitivity of detecting circulating monoclonal paraprotein at very low concentrations [36].

The complement system is composed of over 50 different proteins, including both soluble and surface-bound components [37]. Complete evaluation of the complement system includes four parts: (1) measurement of complement components, (2) measurement of complement activation products, (3) functional analysis of individual complement pathways, and (4) detection of autoantibodies to complement proteins. Unfortunately, a complete evaluation can be prohibitively expensive which often necessitates careful selection of the highest yield tests in general practice. The interpretation of complement laboratory results requires knowledge of the complement system pathways along with an understanding of what is being measured with each assay. During complement activation, there is an increase in the ratio of complement activation products relative to the concentration of inactivated components, which are being actively consumed. During periods of increased complement activation, complement proteins are depleted and function assays counterintuitively will measure a reduced level of activity. Consumption of a particular complement protein due to in vivo complement activation will result in reduced activity as measured by a functional assay for any pathway that utilizes that protein [38]. Measurement of serum C3 and C4 has been widely standardized with relatively reliable test performance. However, despite the reliability of the assays, the finding of low serum C3 levels may only be seen in 40–75% of C3G patients and cannot reliably rule in or out the disease [5, 39]. Testing for other complement components and activation products do not have well-established standards [40]. Furthermore, cross-reactivity of assays and short half-lives of certain complement components can often limit the clinical utility of some tests [41]. Increased serum C3 and C4 levels can be seen in the setting of obesity and acute inflammation [42]. Some have proposed utilizing a ratio of activated products to inactivated products to better assess complement activity due to variability in baseline concentrations of complement proteins [38]. Functional assays can be difficult to interpret given the various assay types used across laboratories [41, 43]. Testing for the presence autoantibodies, such as nephritic factors, is complicated by variability in assays which include hemolytic assays, ELISA-based assays, and surface plasmon resonance-based techniques [44]. Regarding genetic testing, there has been increased availability of genetic testing due to technological advancements; however, determining pathogenicity of a genetic variant requires extensive databases to compare to the general population as well as functional studies that can be cumbersome with the capabilities to do so in only a few select laboratories.

Overall, significant progress has been made in the field of laboratory complement testing. Despite this, many obstacles exist that prevent widespread application and limit the clinical and diagnostic utility of complement testing. Among these obstacles are financial barriers, appropriate testing in the correct clinical setting, prolonged turnaround times, lack of test standardization, and the complexities of proper interpretation of results.

Classically, infection-related glomerulonephritis (IRGN) will have kidney biopsy findings of an acute exudative glomerulonephritis, C3-positive staining on IF with or without immunoglobulin, and large subepithelial electron-dense deposits often referred to as “humps” [45, 46]. The same kidney biopsy findings can also be seen in C3G and can pose a particular challenge clinically when faced with a C3-dominant GN [46]. The clinical course of C3G is often characterized as a progressive, gradually evolving disease. In many cases, C3G may occur in episodes or “flares,” and is preceded by an infectious trigger in nearly 30% of cases [13]. Conversely, Nasr et al. [45] reported 17% (19/109) of patients with post-infectious glomerulonephritis (PIGN) had no clinical evidence of an infection. Among those who underwent a kidney biopsy in this cohort, 100% showed IF with either C3-dominant or C3-codominant staining with 27% showing C3-only staining on IF [45]. The presence of C3 IF staining alone may represent a resolving IRGN disease process in which the inciting immunoglobulin deposits have disappeared. In contrast, it may represent persistent AP activation in the setting of an antecedent infectious trigger. Both disease states are driven by AP dysregulation and may represent a disease spectrum rather than separate entities. Sethi et al. [46] identified either genetic or acquired abnormalities in the AP in 10 out of 11 patients diagnosed with “atypical” PIGN. Individuals that develop IRGN or C3G/IC-MPGN may have an immunologic predisposition to the development of autoantibodies that mediate complement dysregulation. Several examples exist in which infectious pathogens are implicated in the formation of autoantibodies associated with glomerulonephritis. Staphylococcus aureus and Bartonella species are implicated in anti-neutrophil cytoplasmic antibody-associated vasculitis, while Treponema pallidum and Tropheryma whipplei are seen in cases of membranous nephropathy [47]. A recent retrospective study of pediatric acute PIGN patients found 91% to have transient anti-FB antibodies that correlated with biochemical evidence of AP and TP activation [48]. The presence of transient C3NeF has also been reported in pediatric patients with post-streptococcal GN [49]. At this time, laboratory testing cannot differentiate these two entities; however, the findings published by Chauvet et al. [48] suggest that identification of anti-FB antibodies may be indicative of a more transient disease course in the appropriate pediatric population. Current KDIGO guidelines suggest monitoring patients with suspected IRGN for 12 weeks and to consider a kidney biopsy to evaluate for C3G if the patient continues to have persistent signs of disease [50].

While some recommend testing for monoclonal gammopathy in patients of 50 years and older, testing in younger age groups may still be warranted. Alternative diagnoses such as cryoglobulinemic vasculitis or proliferative glomerulonephritis with monoclonal immunoglobulin deposits should be excluded through performing IF on pronase-digested paraffin-embedded tissue (IF-P) in order to “unmask” deposits that are not readily detected on frozen tissue. Larsen et al. [51] reported unmasking deposits in 36% of cases that had been diagnosed with MIg-C3G when IF-P was performed. If IgG is seen on IF with light chain restriction, IgG subclass staining should be used to potentially support a diagnosis of proliferative glomerulonephritis with monoclonal immunoglobulin deposit. Following exclusion of these alternative diagnoses, thorough evaluation for an underlying hematologic malignancy should be performed, as identification would warrant treatment. Unfortunately, in current clinical practice, there are no commercial laboratory tests available that can isolate and test the ability of a monoclonal paraprotein to dysregulate the AP. While the status of AP activity can be ascertained within an individual through various diagnostic tests, definitive proof that the monoclonal protein is the underlying driver cannot be established. Ultimately, the clinician must pursue a treatment strategy in the setting of relative diagnostic uncertainty. Figure 1 shows a proposed diagnostic approach to this particular clinical dilemma.

C3G portends a poor prognosis, with a reported 10-year risk of progression to end-stage kidney disease nearing 50%. Following kidney transplantation, disease recurs in 66.7% of cases with 50% of those leading to allograft failure at a median time of 77 months [52]. While current therapeutic options are limited in efficacy, the current available evidence suggests a benefit from use of immunosuppression over conservative therapy [12, 14]. Treatment approach should focus on underlying etiology, with those patients with MIg-C3G generally receiving clone-directed therapy and patients with genetic mutations receiving complement-directed therapy. Identifying the ideal mechanism of immune suppression for those with autoantibodies or unknown etiology remains a challenge. Mycophenolate-based treatments in combination with corticosteroids are often used as first-line agents. Across various retrospective cohorts, there are significant differences in response rates, ranging from 12.5 to 86% (Table 2) [12‒14, 53‒55]. The cohort analyzed by Caravaca-Fontán et al. [12] suggested improved outcomes with MMF and corticosteroids; however, only 36% achieved complete remission (resolution of proteinuria and normalization of estimated glomerular filtration rate), mostly seen in those with autoantibodies, while those with pathogenic variants in complement genes did not achieve complete remission. Ravindran et al. [13] reported a response rate (resolution or 50% reduction of proteinuria) of 12.5%, which was significantly lower than several other cohorts. This was partially attributed to the higher proportion of C3G patients in this cohort with a genetic form of disease. Of note, discontinuation of mycophenolate-based treatments after achieving clinical remissions is often followed by high rates of relapse, reported at rates as high as 33–50% [12, 54].

There are limited data regarding the efficacy of eculizumab in C3G; however, it appears that there is a subset of patients that may respond to eculizumab (Table 3). In a small series of 26 patients, a rapidly progressive disease course and significant features of extra capillary proliferation were identified in individuals more likely to respond to eculizumab [56]. Some have suggested eculizumab may be best suited for patients that have higher sC5b-9 levels and are nephritic factor negative [57]. In a small series of 10 C3G patients, all 3 patients achieving partial remission were C3NeF negative [58].

Until recently, complement inhibition has been limited to eculizumab, the first FDA-approved complement inhibitor. Eculizumab targets the TP as a C5 antagonist, which has shown variable efficacy in C3G. The emergence of novel immune-modulating therapies that target the complement system has led to a renaissance of therapeutic investigations in the C3G/IC-MPGN disease space. These novel agents may be more efficacious than eculizumab as they can specifically target the AP, directly address complement dysfunction due to pathogenic mutations, and possibly have less infectious complications with indirect terminal complement pathway inhibition. Figure 2 illustrates the components within the complement cascade that are targets of current complement inhibitors. Avacopan is an oral C5aR1 antagonist that is currently FDA approved for treatment of anti-neutrophil cytoplasmic antibody vasculitis. In a phase II trial, avacopan demonstrated less chronic histologic progression compared with placebo from baseline at 52 weeks [59]. The partial response seen with avacopan likely reflects the lack of AP inhibition upstream from C5aR1 antagonism. The factor D inhibitor, danicopan, targets the early steps of the AP through preventing factor D-mediated cleavage of FB within the C3 proconvertase to become the enzymatically active C3 convertase. Danicopan failed to meet the primary end point in phase II trials potentially from a lack of systemic bioavailability [60]. Iptacopan, a FB inhibitor, acts at multiple sites along the proximal limb of the AP through its capacity to bind both FB within the C3 proconvertase and the cleaved byproduct, Bb, a functional component of the C3 convertase. In a phase II trial, iptacopan demonstrated a 45% mean reduction in proteinuria at 12 weeks, suggesting promising results from an ongoing open-label extension study [61]. Pegcetacoplan, an oral C3 inhibitor, also functions at multiple sites within the complement cascade as it can bind both C3 and the C3b fragment that is essential to all 3 complement pathways. In a phase II trial, pegcetacoplan demonstrated a 50.9% mean reduction in proteinuria and a stable mean estimated glomerular filtration rate at 48 weeks [62]. Both iptacopan and pegcetacoplan are currently being evaluated in ongoing phase III trials.

Patients with MIg-C3G have worse outcomes compared to C3G patients without an associated monoclonal gammopathy. These differences in outcomes may reflect the need to direct therapy toward the underlying clonal cell disorder implicated in driving the AP dysregulation. Retrospective studies have found that achieving a hematological response with targeted therapy is associated with improved renal outcomes [27, 28]. Chauvet et al. [28] reported data on MIg-C3G patients from the French national database which found that 83% of those achieving a hematologic response to therapy also had a renal response. In those without a hematologic response, a renal response was seen in only 28% of patients [28]. Ravindran et al. [27] reported a complete or partial response or stable disease in 43.8% of patients receiving targeted therapy with a renal response in 70% of hematologic responders. The size of these cohorts is relatively small, and there continues to be a growing need for larger trials to support these findings.

C3G is a heterogenous disease state with distinct pathogenetic drivers of disease. Progress in this field continues to better define the different etiologies and improve disease classification. The current mainstays of treatment are largely limited to mycophenolate-based regimens with corticosteroids which have extremely variable response rates across cohorts. This variability in response is potentially due to differences in disease pathogenesis within C3G cohorts. While further studies are needed to determine the best approach in the treatment of C3G, a thorough investigation into the underlying driver of disease and the specific mechanisms of complement dysregulation may provide insight into which patients may benefit from targeted therapies. The emergence of novel complement-directed therapies has provided more reason to better define the genetic and autoantibody phenotype of patients enrolled in clinical trials to further refine the current therapeutic approach.

We want to thank Kendra Reilly, MD, for her assistance with graphical design for Figures 1 and 2.

The authors have no conflicts of interest to declare.

The authors received no financial support for the research, authorship, and/or publication of this article.

Eric Keoni Magliulo, MD: conceptualization; data curation; and writing – original draft, review, and editing. Prasanth Ravipati, MD: conceptualization; supervision; and writing – review and editing.