New Therapeutic Targets in Antineutrophil Cytoplasm Antibody–Associated Vasculitis

Antineutrophil cytoplasm antibody (ANCA)–associated vasculitis (AAV) is a rare systemic autoimmune disease that is characterized by necrotizing inflammation of predominantly the small blood vessels and the presence of circulating ANCAs directed against myeloperoxidase or proteinase 3. Current treatment strategies for severe disease, supported by the findings of several well‐coordinated randomized controlled trials, aim to induce remission with high‐dose glucocorticoids and either rituximab or cyclophosphamide, followed by relapse prevention with a period of sustained low‐dose treatment. This approach has dramatically improved outcomes in AAV; however, a significant proportion of patients develop serious treatment‐related side effects or experience relapse. Recent advances in our understanding of the pathogenesis of AAV have led to the identification of novel therapeutic targets that may address these problems, including strategies directed at the aberrant adaptive autoimmune response (B and T cell–directed treatments) and those targeting innate immune elements (complement, monocytes, and neutrophils). It is anticipated that these novel treatments, used alone or in combination, will lead to more effective and less toxic treatment regimens for patients with AAV.


Introduction
The antineutrophil cytoplasm antibody (ANCA)-associated vasculitides (AAVs) are a group of rare systemic inflammatory diseases which include granulomatosis with polyangiitis (GPA), microscopic polyangiitis (MPA), and eosinophilic GPA (EGPA). They are associated with the presence of autoantibodies to myeloperoxidase (MPO) or proteinase 3 (PR3), which are thought to play a pathogenic role, although ~10% of patients with AAV are ANCA negative and have similar clinical features as those who are ANCA positive (1). The AAVs are multisystem diseases that share common characteristics such as necrotizing inflammation of predominantly the small blood vessels, though the clinical presentation can vary widely, both in disease severity and in the spectrum of organ involvement.
Historically, outcomes in patients with AAV were poor, with a mortality rate of up to 80% at 1 year prior to the use of immunosuppressive treatment. This potential for poor outcomes was transformed with the introduction of cyclophosphamide, used in combination with glucocorticoids, in the 1970s.
Since then, there has been decade upon decade of improvements in patient outcomes (2,3), reflecting a combination of improved general medical care, earlier diagnosis, and more refined immunosuppressive regimens that have reduced toxicity from long-term cyclophosphamide use. These therapeutic strategies have been informed by a number of well-designed and collaborative clinical trials, and based on their findings, current guidelines stratify treatment depending on the severity and phase of the disease, usually with an initial period of intense immunosuppression to induce remission (often with cyclophosphamide or rituximab), followed by prevention of relapse with a period of sustained low-dose maintenance treatment (with drugs such as azathioprine, mycophenolate mofetil, or rituximab) (4).
With this approach, disease remission is attained in most patients during the first 6 months of treatment, and survival is estimated to be >90% at 1 year. However, unmet treatment needs remain: most deaths in the first year are now attributed to side effects of treatment, particularly infection. During long-term follow-up, therapy-related adverse events,  including malignancy and cardiovascular disease, remain problematic, thus underscoring the need to refine treatment regimens further (5,6). In addition, approaches that can induce more sustained remission, thus avoiding the accrual of damage caused by disease relapse and its retreatment, are badly needed in this patient population. Advances in our understanding of disease pathogenesisrelated to the aberrant adaptive T and B cell responses that underlie this autoimmune disease, and to the role of innate immune components, including neutrophils, monocytes, and complement, as mediators of vascular damage-provide an opportunity to identify more specific, less toxic treatments for AAV. Many novel therapeutic agents are being investigated in preclinical and early-phase clinical studies (selected trials are listed in Table 1). In this review, we summarize the possible future therapeutic options for AAV (7-10) (Figure 1), although EGPA is not included herein because it has a distinct pathogenesis and different therapeutic approaches.

Targeting B cells
The emergence of rituximab as an effective induction and remission-maintenance treatment is arguably the most significant development in the management of AAV since the introduction of cytotoxic therapy almost a half century ago. B cells are clearly central to the pathogenesis of the disease, as they produce ANCAs. Moreover, studies have shown that the number of activated B cells correlates with disease activity (11), and B cell repopulation, and possibly the phenotype of B cells, after treatment with rituximab may be predictive of relapse (12).
Several second-generation anti-CD20 drugs are in development; these differ from rituximab in their epitope specificity, pharmacokinetics, and ability to induce either complement or antibody-dependent cytotoxicity and apoptosis, which in turn may impact the rapidity, depth, and duration of their depleting effect on the circulating and tissue B cell pools. Ofatumumab, a fully humanized anti-CD20 monoclonal antibody (mAb), has been tested in one small case series of patients with AAV, with the results showing its therapeutic benefit. However, none of these second-generation agents has been tested in randomized controlled trials (RCTs) in patients with AAV (13).
Obinutuzumab, which has increased antibody-dependent cytotoxicity and a greater capacity for direct B cell killing compared to rituximab, has shown promise in a phase II study in patients with lupus nephritis (ClinicalTrials.gov identifier: NCT02550652) (14). Conversely, an early study of another humanized anti-CD20 antibody, ocrelizumab, in patients with lupus was terminated early due to a higher-than-expected rate of infections (15,16). Whether these second-generation anti-CD20 drugs might provide incremental benefit over rituximab in AAV, without increasing the incidence and frequency of short-or long-term toxic effects (e.g., hypogammaglobulinemia, impaired vaccine responses), will require more detailed study.
The inhibition of B cell cytokines and survival factors is an alternative approach to direct targeting to B cells. B lymphocyte stimulator (BLyS), for example, plays an important role in B cell survival, and circulating BLyS levels are higher in patients with AAV than in healthy individuals. After treatment with rituximab (in AAV and other autoimmune diseases), serum BLyS levels rise (17,18), which may herald the occurrence of a relapse. In vitro, BlyS is released from neutrophils stimulated with ANCAs, suggesting that it has a specific role in AAV (19).
Belimumab is an anti-BlyS mAb licensed for the treatment of lupus and under investigation in AAV. The Belimumab in Remission of Vasculitis (BREVAS) trial examined the addition of belimumab to azathioprine and glucocorticoids for remission maintenance in patients who received either rituximab or cyclophosphamide for induction (20). Regrettably, the trial was terminated early due to under-recruitment, and no benefit for the primary end point (improvement in the relapse rate) was observed. However, in the subgroup of patients treated with rituximab for induction, fewer relapses were seen with belimumab (0 of 14 patients experiencing relapse in the belimumab group compared to 3 of 13 patients in the placebo group). Although both the number of patients and relapses were small, this might suggest a potential benefit of belimumab after treatment with rituximab (20).
The combination of rituximab and belimumab will be investigated further in the Rituximab and Belimumab Combination Therapy in PR3-AAV trial (COMBIVAS) (ClinicalTrials.gov identifier: NCT03967925), in which patients will be treated with rituximab and glucocorticoids for remission induction, and will be randomized to receive either belimumab or placebo for 1 year. It is hypothesized that the addition of belimumab will potentiate the effect of rituximab on B cell depletion and prevent the return of autoreactive cells, or suppress a broader repertoire of B cells (including those not expressing CD20) than can be achieved with rituximab alone, thus inducing more rapid and sustained remission (21).
Bortezomib, a proteasome inhibitor that drives plasma cells with high immunoglobulin synthesis to apoptosis, is an alternative approach for targeting the CD20-negative cell population. In a mouse model of MPO-AAV, treatment with bortezomib depleted MPO-specific plasma cells and decreased the severity of glomerulonephritis (22). There is a single report of its successful use in a patient with treatment-resistant PR3-AAV (23). The routine use of bortezomib is likely to be limited because of its side effect profile, as >30% of patients develop painful peripheral neuropathy. However, several novel and potentially less toxic proteasome inhibitors are currently in development (24,25).
Novel cell-based approaches may also be used to target autoreactive B cells. Chimeric antigen receptor (CAR) T cells are autologous cells that can be engineered to specifically target CD19+ B cells-an approach that has shown efficacy in some hematologic malignancies. It is also possible that an extension of this technology-chimeric autoantibody receptor (CAAR) T cells-may target autoreactive B cells through their antigen-specific B cell receptor. CAAR T cells have been tested in a model of pemphigus in humanized mice, in which it was observed that CAAR T cells induced lysis of pathogenic B cells (26). Although still in the early stages of development, CAAR T cells may provide a curative approach in AAV and other autoimmune diseases for which the target autoantigens have been defined.

Targeting T cells
The importance of aberrant T cell responses in AAV is increasingly recognized, and studies in experimental models of MPO-AAV have been particularly informative (27). In mice, disease can be attenuated by depletion of either CD4 or CD8 T cells, and adoptive transfer of T cells can initiate glomerular injury independently of ANCAs. In patients with AAV, circulating ANCAs are predominantly class-switched IgG1 and IgG4, implying that helper T cells play a role in this process. Immunostaining has identified T cells in the glomeruli and tubulointerstitium in renal biopsy tissue from patients with renal AAV, and several abnormalities in circulating T cell phenotype or function have been reported in patients with active disease (28,29). An exhausted T cell phenotype has been shown to correlate with a reduced risk of disease relapse (30). Activation of circulating T cells is reported to persist, despite treatment, during periods of remission, and therefore there may be a particular role for anti-T cell therapies in preventing relapse (31).
There are several small studies using T cell-directed therapies for remission induction, typically in patients with refractory disease. In an open-label cohort study of 15 patients with relapsing or refractory AAV, treatment with anti-thymocyte globulin led to a favorable response in 13 patients (32). Alemtuzumab is a humanized anti-CD52 mAb that depletes all lymphocytes, with a particularly long-lasting effect on T cells; CD4+ T cell counts take ~60 months to recover (33). When CD4+ T cells do eventually repopulate after treatment with alemtuzumab, some reports | 365 in patients with multiple sclerosis show a skew toward a Treg cell phenotype, which could contribute to long-lasting immunomodulatory effects (34). Long-term follow-up of 71 patients with refractory AAV treated with alemtuzumab showed that remission was achieved in 80% of patients, although relapse and severe adverse events were common (35). A subsequent phase II RCT, the Alemtuzumab for ANCA Associated Refractory Vasculitis (ALEVIATE) trial, compared high-and low-dose alemtuzumab in a mixed cohort of patients with refractory AAV or Behçet's disease. A preliminary report from the study indicated that 6 months after treatment, remission was achieved in 65% of patients and, although relapse was common, 35% had sustained remission at 1 year, by which point 26% of patients had experienced an adverse event (36). Thus, there may be a role for alemtuzumab in patients with difficult-to-treat disease, although the potential for adverse events is high compared to standard treatment strategies, likely reflecting both drug toxicity and susceptibility to infection due to disease-related factors. There are also reports of autoimmune phenomena occurring after alemtuzumab treatment in other diseases, such as multiple sclerosis, which is thought to be driven by expansion of T cells that have escaped deletion and become chronically activated; whether patients with AAV are at similar risk should also be considered (37).
In patients with nonsevere disease, abatacept, a fusion protein comprising the Fc region of IgG1 fused to CTLA-4, has been tested. Abatacept prevents the costimulatory signaling occurring via CD80 and CD86 that is needed for antigen-presenting activation of T cells (38). In an open-label study of 20 patients with relapsing, non-severe GPA who received abatacept in addition to methotrexate, mycophenolate mofetil, or azathioprine, remission rates of 80% were observed, and >70% of patients were able to wean glucocorticoid treatment (39). An RCT that is currently recruiting patients, the Abatacept for the Treatment of Relapsing Non-Severe GPA (ABROGATE) trial (ClinicalTrials.gov identifier: NCT02108860), is evaluating this approach using a glucocorticoid-free regimen.
The Th17/interleukin-17 (IL-17)/IL-23 axis is also known to play a role in the pathogenesis of AAV. Serum levels of IL-23 and IL-17 are raised in patients with active disease, and stimulation of neutrophils with ANCAs has been shown to induce production of IL-17 (40,41). In one study, IL-17-deficient mice were protected from developing MPO-AAV (42). Monoclonal antibodies against IL-17 (seikinumab) and IL-23 (ustekinumab) have been tested in patients with psoriasis and those with rheumatoid arthritis, but there are no reports of their use in patients with AAV to date.

Targeting cytokines
Levels of circulating cytokines such as IL-6 and tumor necrosis factor (TNF) are elevated in patients with active AAV (43,44). A number of open-label studies and case series have demonstrated the successful use of anti-TNF therapies, although these results were not confirmed when tested in RCTs, the largest of which-the Wegener's Granulomatosis Etanercept Trial (WGET)-recruited 174 patients with GPA. In this trial, the patients were randomized to receive either etanercept or placebo, in addition to standard treatment with glucocorticoids and methotrexate or cyclophosphamide, for remission maintenance (45). There was no benefit from etanercept on the rate of sustained remission, and this negative trial outcome means that anti-TNF agents have largely been discounted as a potential therapeutic option in AAV, though they may yet have a niche role in the treatment of specific disease manifestations such as ocular inflammation (46). IL-6 promotes B cell differentiation, activates macrophages, and induces production of other cytokines; serum IL-6 levels are elevated in patients with AAV, and IL-6 is expressed at sites of tissue inflammation (43). There are several case reports describing patients with AAV treated with tocilizumab, a humanized anti-IL-6 receptor mAb, showing that complete and sustained remission was achieved in many of the patients with otherwise refractory disease (43,47). Others have reported less favorable outcomes with tocilizumab, including treatment failure or infectious complications (47). Given the successful use of tocilizumab in other systemic autoimmune rheumatic diseases, controlled studies may be warranted to define the role of anti-IL-6 therapy in AAV.

Targeting complement
While historically regarded as a "pauci-immune" vasculitis, with few immunoglobulin or complement deposits in tissue, the past decade has seen the important role of complement in disease pathogenesis come to light (8). In patients with AAV, careful examination has identified complement deposition at sites of tissue inflammation, and altered levels of plasma and urinary complement components have been shown to correlate with disease severity (48,49). Convincing evidence of complement involvement came from experimental mouse models, in which a series of elegant studies dissected a role for alternative pathway activation, and for the receptor of C5a (C5aR), a potent anaphylatoxin and chemoattractant, in disease pathogenesis (50). In vitro, C5a can prime neutrophils to respond to ANCA stimulation, and both ANCA-stimulated neutrophils and neutrophil extracellular traps (NETs) can activate the alternative complement cascade, leading to a positive feedback loop (51,52). Ultimately, it was shown that in mice transgenic for the human C5aR, a small molecule antagonist of the C5aR1 (avacopan, CCX168) was an effective treatment for MPO-AAV in a passive transfer model (53).
This compound then showed promising results in an earlyphase clinical study of patients with AAV, in which it was found to be a noninferior substitute for prednisolone during remission induction (54). However, that study was small (n = 67 patients), was of short duration (12 weeks), and included only patients with nonsevere disease. A subsequent phase III trial, the Avacopan in Patients With ANCA-Associated Vasculitis (ADVOCATE) trial (Clin-icalTrials.gov identifier: NCT02994927), completed recruitment of 300 patients in 2018. The patients were randomized to receive either avacopan or glucocorticoids during remission induction with either cyclophosphamide or rituximab, and top-line data released in late 2019 suggested noninferiority of avacopan at 26 weeks and superiority over glucocorticoids at 52 weeks, with an acceptable safety profile. However, we still await full publication of the study results, and it should be highlighted that the most severe cases (those with an estimated glomerular filtration rate of <15 ml/ minute, requiring dialysis or plasma exchange) were still excluded.
An alternative anti-C5a treatment, IFX-1, a mAb that targets C5a rather than C5aR, which may therefore have differing biologic effects from those of avocapan, is also being evaluated in phase II studies. Patients will be randomized to receive standard glucocorticoids, a combination of IFX-1 and reduced-dose glucocorticoids, or IFX-1 and no glucocorticoids during the remission-induction phase (European study, ClinicalTrials.gov identifier: NCT03895801) or randomized to receive standard of care plus IFX-1 or placebo (North American study, ClinicalTrials.gov identifier: NCT03712345). Recruitment is ongoing and completion is estimated by July 2021. Blockade of C5 cleavage is another potential treatment, though descriptions on eculizumab use in patients with AAV are limited to individual case reports (55).

Targeting neutrophils and monocytes
There is extensive evidence demonstrating the pathogenic role of neutrophils in AAV. Studies first published nearly 30 years ago have shown that ANCAs bind to and activate neutrophils, leading to degranulation and production of reactive oxygen species (56). ANCA stimulation of neutrophils has also been shown to activate intracellular signaling cascades, leading to increased neutrophil adhesion and transmigration at the vascular endothelium (57). ANCA stimulation can induce NETosis, a specialized form of cell death with release of NETs (extracellular meshes of decondensed chromatin and granular proteins). NETs are pathogenic in AAV: they can activate dendritic cells, autoreactive B cells, and complement; they are directly injurious to endothelium; and they may play a role in loss of tolerance to ANCA antigens (52,58). While many studies have focused on the role of neutrophils, monocytes also express the ANCA autoantigens and respond similarly to ANCA stimulation in vitro (59), and thus may contribute to tissue injury. A number of agents that target these neutrophil-and monocyte-mediated functions in AAV are in preclinical development.
Inhibiting NETosis may attenuate both vascular damage and potentiation of the autoimmune response by limiting aberrant extracellular expression of ANCA autoantigens. Peptidylarginine deiminase 4 (PAD-4) is essential for NET formation, as it plays a role in citrullination of histones, and PAD-4 inhibition decreases NET formation in vitro (60). In a mouse model of MPO-AAV, PAD-4 deficiency or use of a selective inhibitor decreased NETosis, MPO deposition, glomerular injury, and cell infiltration (61). PAD-4 inhibition has also been tested in mouse models of lupus, but as yet there have been no studies in humans.
Cathepsin C is a lysosomal peptidase that acts in the bone marrow to cleave neutrophil serine proteinases (NSPs), including neutrophil elastase and PR3, to their mature, active forms (62). Activated neutrophils release large amounts of these NSPs into the extracellular space, where they may initiate tissue inflammation and injury as a constituent of NETs. In a mouse model of MPO-AAV, knockout of cathepsin C protected from disease and decreased MPO-ANCA-induced IL-1β production in vitro (63). Cleaved NSP may also remain bound to the neutrophil cell surface and, in PR3-AAV, this translocation of PR3 to the cell surface may perpetuate disease; there is evidence in GPA that patients with higher levels of membrane-bound PR3 have more severe disease features (64,65). Thus, reducing cell surface expression of PR3 by preventing its activation by cathepsin C has been suggested as a potential therapeutic strategy. A recently developed pharmacologic inhibitor of cathepsin C was found to decrease the levels of membrane-bound PR3 on neutrophils, with no effect on neutrophil differentiation. When used in vitro, PR3-ANCA-mediated neutrophil activation was diminished and the compound also showed pharmacologic activity in mice, although it was not tested in an in vivo model of AAV (66).
Directly targeting MPO, the other ANCA autoantigen, is another approach that has been assessed in animal models. Like PR3, MPO is released from neutrophils and monocytes following activation, and may cause injury and activate autoreactive B and T cells. Extracellular MPO can deposit in glomeruli in AAV, and the amount of deposition correlates with the severity of disease (67). In vivo, treatment of mice with an MPO inhibitor decreased the severity of crescentic glomerulonephritis (67).
Spleen tyrosine kinase (Syk) is a cytoplasmic protein tyrosine kinase that is highly expressed in neutrophils and monocytes, and it plays a role in signaling for activatory Fc receptors and some integrins (68). Syk is activated in neutrophils following ANCA stimulation (69), and can be detected in leukocytes within glomerular lesions in patients with ANCA-associated renal disease (70,71). Fostamatinib, a small molecule inhibitor with selectivity for Syk, inhibits ANCA-induced neutrophil responses in vitro, and is an effective treatment for MPO-AAV in a rat model (70,72). Syk is also critical for B cell receptor signaling, and treatment with fostamatinib reduced autoantibody responses in an experimental model of anti-glomerular basement membrane disease, suggesting a potential dual therapeutic effect in AAV (73).
There may be concerns that targeting these innate immune responses may leave patients vulnerable to infection, though it is reassuring that congenital deficiencies of cathespin C and MPO, for example, have relatively mild clinical phenotypes. In addition, clinical studies of Syk inhibition in other diseases, including | 367 rheumatoid arthritis, IgA nephropathy, and idiopathic thrombocytopenic purpura, did not show a risk of severe infections, and thus future trials testing these approaches in patients with AAV are warranted.

Combination drug therapy
Drugs targeting the innate immune response may be an effective substitute for glucocorticoids during acute disease, though they may be less effective for suppressing the underlying adaptive response, which is needed to secure long-term remission. Conversely, specific targeting of B and T lymphocytes may not provide sufficiently rapid responses during acute flares to prevent accrual of organ damage. As an increasing number of potential therapeutic agents are identified, future studies will need to address how they are best used in sequence or in combination, either with each other or with existing therapies, and during different phases of the disease, to improve outcomes and reduce toxicity. The negative results in some RCTs in patients with AAV, such as in those treated with etanercept in the WGET study, may relate to their use as add-on therapy to conventional treatment, such that potential signals of biologic activity were lost, whereas the recent enthusiasm for complement inhibition arose following a successful trial designed to demonstrate that avacopan could wholly replace glucocorticoids during remission induction.
Conversely, "multi-target" therapy has recently emerged as an effective approach in patients with lupus nephritis, and combination approaches that target multiple aspects of the immune and inflammatory response may likewise be an effective way to provide rapid and sustained disease control in AAV, while avoiding toxicities caused by excessive exposure to individual drugs. One such combination approach using low-dose intravenous cyclophosphamide and rituximab, along with a rapid oral glucocorticoid taper, has been described in an open-label cohort study. Rates of remission at 6 months, mortality, long-term relapse rates, and renal outcomes were favorable when compared to those in matched historical controls from European Vasculitis Study Group studies (74,75). A similar approach, using a short course of oral cyclophosphamide and rituximab, has been reported in a single-center, retrospective case series of 129 patients, again showing favorable rates of remission induction and relapse (76). It is suggested that the early combination of cyclophosphamide and rituximab may allow reduction in glucocorticoid exposure during acute disease, while inducing sustained remission, perhaps through potentiation of B cell depletion.
However, concern for increased toxicity remains. A clinical trial of this combination treatment regimen-Exploring Durable Remission With Rituximab in Antineutrophil Cytoplasmic Antibody-Associated Vasculitis (ENDURRANCE) (ClinicalTrials.gov identifier: NCT03942887)-is currently recruiting patients, and will assess both immunologic responses (as its primary outcome) and clinical outcomes (including adverse events) following combination induction treatment with rituximab, low-dose intravenous cyclophosphamide, and low-dose glucocorticoids.

Conclusions
Modern immunosuppressive regimens have transformed outcomes in AAV, and several evidence-based treatment guidelines are now available, informed by the findings of high-quality RCTs. There are, however, unmet needs related to the potential for drug side effects and the unpredictable nature of disease relapse. Advances have been made in understanding the pathogenesis of AAV, which have identified many potential new targets for therapy that may be directed to various aspects of the adaptive and innate immune responses underlying disease. Some of these investigations have already progressed to clinical studies, such as targeting of the alternative complement cascade, and others remain in the preclinical stages of development. Future clinical trials of these novel therapeutic agents will need to establish their efficacy and, as an increasing number of potential treatments becomes available, will need to indicate how they can be used to complement or replace existing approaches. Moreover, with more agents at our disposal, future studies will need to incorporate the use of biomarkers and predictors of flare and stratification for patient factors that might influence treatment response, including age, comorbidities, and patterns of organ involvement (e.g., presence of granulomatous lesions, ANCA serotype and potentially genotype), so that subgroups of patients likely to benefit from a given therapy can be identified. This should allow for the development of more tailored treatment protocols that maximize response while minimizing the side effects from unnecessary drug exposure, and thus would improve outcomes in patients with AAV.

AUTHOR CONTRIBUTIONS
Drs. Prendecki and McAdoo drafted the article, revised it critically for important intellectual content, and approved the final version to be published.