The rise of complement in ANCA‐associated vasculitis: from marginal player to target of modern therapy

The complement system is critical for the pathogenesis of ANCA‐associated vasculitis (AAV). Complement activation via the alternative pathway is triggered by ANCA‐induced neutrophil degranulation and leads to the generation of C5a, a potent anaphylatoxin that enhances recruitment and priming of further neutrophils. Avacopan, an oral inhibitor of the C5a receptor, disrupts this pathogenic loop and may allow glucocorticoid‐free disease control, marking a new era in the treatment of AAV.

Summary

The complement system plays a central role in autoimmune diseases, including anti‐neutrophil cytoplasmic antibody (ANCA)‐associated vasculitis (AAV). Although complement deposition is scarce in AAV pathological samples, complement activation is required for the development of necrotizing crescentic glomerulonephritis (NCGN) in mouse models of AAV and occurs via the alternative pathway. The anaphylatoxin C5a, produced by the final complement pathway, is determinant to drive the disease in animal models. C5a primes human neutrophils and enhances their activation induced by ANCA; activated neutrophils, in turn, release factors that lead to C5a generation, establishing a self‐amplifying loop. C5a is also significantly increased in the serum of AAV patients with active disease compared to those in remission or healthy controls. Inhibition of the C5a receptor with avacopan is an emerging therapy that will probably allow AAV treatment with glucocorticoid‐free regimens.

The complement system, which comprises dozens of circulating and membrane‐bound proteins involved in the immune response, appears dysregulated in several autoimmune diseases. In anti‐neutrophil cytoplasmic antibody (ANCA)‐associated vasculitides (AAVs), its role has long been considered of minor importance. Nevertheless, recent advances have contributed to show that the complement is critical in these conditions, and have paved the way for therapies that inhibit its activation.

Compared to other small‐vessel vasculitis, biopsies of patients with microscopic polyangiitis (MPA) and granulomatosis with polyangiitis (GPA), the two main AAV variants, show scarce immunoglobulin and complement deposits, a pattern called ‘pauci‐immune’ [1]. This was considered to result from negligible complement activation. Moreover, hypocomplementaemia is uncommon in AAV patients [2].

However, these observations did not rule out the possibility that the complement participates in AAV injury. In fact, seminal studies performed during the past two decades demonstrated that the complement is crucial for ANCA to mediate tissue injury. ANCA bind to proteinase 3 (PR3 ANCA) and myeloperoxidase (MPO ANCA) that are expressed by neutrophils primed by inflammatory stimuli, such as tumour necrosis factor (TNF)‐α, and activate them causing respiratory burst, degranulation and consequent vascular inflammation [3]. MPO ANCA injected in mice can induce necrotizing crescentic glomerulonephritis (NCGN), which closely resembles human AAV‐related glomerulonephritis [4]. However, complement depletion by cobra venom factor and blockade of the final complement pathway by knock‐out (KO) of C5 or by C5 inhibiting monoclonal antibody protect mice from NCGN [5, 6]. These findings first demonstrated that complement is required in ANCA‐induced lesions.

Furthermore, the manner of complement activation was explored by selectively inhibiting the different pathways. Mice with KO of factor B, a clue marker of the alternative pathway, failed to develop NCGN, while mice with KO of C4, a shared component of the classical and the lectin pathway, experienced NCGN comparable to controls [5]. These results showed that complement activation in ANCA‐related NCGN occurs via the alternative pathway, while the remaining pathways are not essential. This is consistent with the concept that AAV is not driven by immune complexes, which preferentially activate the classical pathway.

All three complement pathways lead to the generation of C5, which is further cleaved into C5a, a cytokine with anaphylatoxic and chemotactic properties, and C5b, a larger protein that prompts the assembly of the lytic membrane attack complex (MAC, also known as C5b‐9). Both C5a and C5b downstream signals mediate complement effects and may cause endothelium and tissue damage.

C5a and not MAC was shown to be pivotal in ANCA‐mediated lesions. Schreiber and colleagues showed that mice with C5a receptor (C5aR)‐deficient leucocytes are protected against ANCA‐induced NCGN, while mice KO for C6, a MAC component, develop NCGN comparable to controls [7]. Results of human in‐vitro experiments also revealed that C5a acts as a primer of neutrophils, increasing their expression of ANCA targets and enhancing their response to ANCA. More intriguingly, ANCA‐activated neutrophils release factors that activate complement and generate further C5a, thus establishing an amplification loop that sustains ANCA‐induced vascular inflammation. The recognition of this positive feedback mechanism based on the C5a axis suggested the possibility of its selective inhibition as a therapeutic option for AAV [8]. Figure 1 depicts the main steps of ANCA‐induced neutrophil activation and tissue damage, with a focus on the C5a pathway and its blockade.

Fig. 1.

Neutrophil activation and the C5a pathway in anti‐neutrophil cytoplasmic antibody (ANCA)‐associated vasculitis (AAV) pathogenesis. Neutrophils primed by inflammatory stimuli express myeloperoxidase (MPO) and proteinase 3 (PR3), the main ANCA antigens, on their membrane surface. The binding of ANCA leads to neutrophil activation, with release of reactive oxygen species (ROS) and toxic enzymes, and consequent vascular inflammation. Neutrophil degranulation also results in complement activation via the alternative pathway. C5a, a complement product with anaphylatoxic and chemotactic properties, recruits further neutrophils into the affected tissues and primes them, perpetrating the ANCA‐induced pathogenic process. Avacopan, an orally administered molecule, and the monoclonal antibody IFX‐1 inhibit the amplification loop mediated by the C5a pathway, targeting, respectively, the C5a receptor (C5aR) and C5a itself.

Results of pathological studies were also consistent with complement activation, at least in a fraction of patients with AAV. Deposition of C3c was found at low intensity in approximately one‐third of AAV patients and correlated with proteinuria and initial renal dysfunction [9]. Moreover, biopsies positive for factor B and properdin (alternative complement pathway) showed a significantly higher amount of crescentic glomeruli with a lower proportion of normal glomeruli [10, 11]. Importantly, while experimental evidence was derived from murine models of MPO ANCA‐vasculitis, complement deposition was observed in biopsies of both MPO ANCA‐positive and PR3 ANCA‐positive patients, possibly indicating that complement activation occurs in both types of vasculitis.

Of note, the participation of immune complexes and classical complement pathway is not to be fully excluded, as a substantial proportion of biopsies stained for the classical pathway factor C4d, while some showed electrondense deposits compatible with immune complexes [10]. The alternative complement pathway may also cause renal damage by inducing thrombotic microangiopathy, which overlaps with NCGN in a subset of patients with extremely poor prognosis [12]. These findings do not argue against the central role played by the C5a axis in AAV pathogenesis, but underscore the multiple and complex relations established by the complement system. Notably, it was shown that leucocytes infiltrating the glomeruli of AAV patients express low concentrations of C5aR, probably because of desensitization due to hyperactivation of this pathway [13].

Another approach to investigate the role of complement is to measure the levels of its activation products in patient serum. Gou et al. observed that circulating C3a and C5a of the common pathway and factor B of the alternative pathway are significantly increased in AAV patients in their active stage compared to those in remission, supporting systemic complement activation via the alternative cascade [14]. Factor B also correlated with the proportion of cellular crescents, disease activity assessed by Birmingham Vasculitis Activity Score (BVAS) and acute‐phase reactants, such as erythrocyte sedimentation rate. Interestingly, C3a and C5a levels were increased in AAV patients compared to patients with lupus nephritis, while MAC concentrations were higher in the latter.

In this issue of Clinical and Experimental Immunology, Moiseev et al. report the results of the first meta‐analysis evaluating circulating complement components in AAV patients with active disease, disease in remission and healthy controls [15].

The authors first evaluated 59 patients with active GPA or MPA and 36 healthy volunteers and entered results with those of four previous studies, achieving a sample of more than 220 patients and 120 controls [14, 16, 17, 18]. The meta‐analysis revealed significantly higher levels of C5a, MAC and factor B among AAV patients with active disease compared to those in remission and healthy controls. These results confirm complement activation during active disease phases and highlight that remission is associated with lower complement activity. Levels of Bb, C4d and C3a, belonging to the alternative, classical and common pathway, respectively, did not differ between the groups or were moderately increased in AAV patients.

Analysis of the initial cohort also showed that there were no differences in complement product levels between MPO ANCA‐positive and PR3 ANCA‐positive patients, MPA and GPA, severe and non‐severe vasculitis and forms with predominant granulomatous or vasculitic manifestations. These findings are of major importance, as experimental studies that demonstrated the requirement of complement for ANCA‐mediated damage were conducted on models of MPO ANCA‐induced NCGN, while little is known about other manifestations such as granulomatous lesions in PR3 ANCA‐positive patients. Although not proving that complement activation is required for such involvement, these results are in favour of complement participation in all AAV phenotypes.

Furthermore, concentrations of C5a and factor B exceeded the upper bound of the reference range only in fewer than one‐quarter of the patients and did not correlate with BVAS, glomerular filtration rate and acute‐phase reactants; for example, C‐reactive protein and erythrocyte sedimentation rate. This indicates that absolute levels of complement components per se cannot be used as disease biomarkers. However, tests were repeated when patients achieved remission and C5a levels were decreased. Future studies might explore whether the trend of C5a reflects disease status in AAV patients.

Thanks to experimental and clinical evidence, the complement system has been recognized as a major determinant of AAV lesions. More deeply, it was shown that C5a binding to C5aR enhances ANCA‐induced neutrophil activation that mediates tissue damage. We are about to understand the therapeutic implications of these pivotal advances. The C5a axis is, in fact, the target of new promising therapies that aim to disrupt its pathogenic effect.

Avacopan, an orally administered inhibitor of C5aR, prevents C5a from activating the C5aR but not from binding to the C5a‐like receptor 2, which acts as scavenger and is associated with milder renal disease [8]. Phase II trials CLEAR and CLASSIC proved that treatment with avacopan 30 mg twice a day is safe, and may replace glucocorticoids for induction of AAV remission [19, 20]. Results of ADVOCATE, a large randomized trial that compared avacopan plus placebo versus standard glucocorticoid therapy, both in combination with cyclophosphamide or rituximab, are awaited for approval and will probably lead to a major breakthrough in the treatment of AAV.

Another C5a pathway blocker in clinical development for AAV management is IFX‐1, a monoclonal antibody directed against C5a that has entered a Phase II trial [21]. Importantly, both avacopan and IFX‐ 1 do not impair the formation of MAC, which is necessary for defence against encapsulated bacteria, such as Neisseria meningitidis. Other complement‐targeting approaches, such as that with eculizumab, are limited to single case reports [22].

Fifteen years of strenuous research have changed the view of the complement system in AAV, from that of a marginal player to that of a key target for modern therapies. Results of therapeutic interventions will definitively confirm its role and will mark a new era in the treatment of AAV.

Disclosures

None.

Data availability statement

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