Should proteinase-3 and myeloperoxidase anti-neutrophil cytoplasmic antibody vasculitis be treated differently: part 2

More than 30 years ago, anti-neutrophil cytoplasmic antibodies (ANCAs) with specifity for either proteinase 3 (PR3-ANCA) or myeloperoxidase (MPO-ANCA) were first described in the context of glomerulonephritis and vasculitis [1]. Most patients are single positive, meaning that only one ANCA serotype can be detected [1]. MPO-ANCA and/or PR3-ANCA were found in granulomatosis with polyangiitis (GPA; formerly known as Wegener’s granulomatosis), microscopic polyangiitis (MPA) and eosinophilic GPA (EGPA; formerly known as Churg–Strauss syndrome). In patients with GPA, PR3-ANCA is more often found, whereas in patients with either MPA or EGPA, MPO-ANCA is more often found [1]. Importantly, however, a substantial portion of patients with either GPA, MPA or EGPA are ANCA negative (see Table 1). In contrast to the ANCA-positive patients with ANCA-associated vasculitis (AAV), renal involvement is nearly always lacking in AAV patients in whom ANCA testing is negative. Disease associations of PR3-ANCA and MPO-ANCA Neg, negative testing for both MPO-ANCA and PR3-ANCA. Disease associations of PR3-ANCA and MPO-ANCA Neg, negative testing for both MPO-ANCA and PR3-ANCA. The distinction between GPA and MPA in the context of the ANCA serotype is far from perfect [1] and underlines the discordance between disease categorization and ANCA serotype categorization. Many clinical and genetic differences exist between patients with PR3-AAV and MPO-AAV [1] (Table 2). Similarities and differences between PR3-AAV and MPO-AAV Similarities and differences between PR3-AAV and MPO-AAV The incidence of PR3-AAV and MPO-AAV varies worldwide, possibly due to a combination of genetic pools and certain environmental factors. In general, PR3-AAV is more common in northern parts of the world, whereas MPO-AAV is more common in southern Europe, Asia and the Pacific, with the exception of New Zealand and Australia [1]. At diagnosis, PR3-AAV patients are younger than MPO-AAV patients and the male:female ratio is higher in PR3-AAV when compared with MPO-AAV. It has been demonstrated that single-nucleotide polymorphisms (SNPs) in the Human Leucocyte Antigen (HLA)-DPB region on chromosome 6 are present in a large percentage of PR3-AAV patients as opposed to MPO-AAV patients [2]. Importantly, the association between this particular SNP and PR3-ANCA was stronger than with the clinical diagnosis of GPA [2]. Also, a higher prevalence of SERPINA1 and PRTN3 polymorphisms [2] is present in PR3-AAV patients when compared with MPO-AAV patients. These findings support a pathogenic role for ANCA, that is, genetic studies show differences in major histocompatability complex, the PR3 antigen and α1-antitrypsin. Also, the cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) and programmed death 1 (PD-1) immune checkpoints seem to be differentially involved: MPO-AAV patients are more often G-allele carriers of a polymorphism in CTLA-4 when compared with PR3-AAV patients [1], whereas PD-1 seems to be functionally deficient in PR3-AAV [3]. Environmental triggering factors also differ. Silica exposure and smoking [4] are associated with MPO-AAV, whereas vitamin D deficiency [5] and chronic sinonasal inflammation due to Staphylococcus aureus are associated with PR3-AAV. Clinical features differ between PR3-AAV and MPO-AAV, as was already observed directly after the introduction of ANCA in clinical practice [1]. At diagnosis, more organs are affected in PR3-AAV compared with MPO-AAV. The combination of granulomatous inflammation in the upper and/or lower respiratory tract with renal involvement is frequently seen in PR3-AAV, as opposed to vasculitis limited to the kidney in MPO-ANCA. Asthma and pulmonary fibrosis are more frequent in MPO-AAV, whereas cavitating lung lesions and/or alveolar lung hemorrhage are more often found in PR3-AAV. Most importantly, ear, nose and throat involvement clearly differs between PR3-AAV and MPO-AAV. Necrotizing lesions in the nose and/or sinus are found in PR3-AAV, whereas nasal polyposis is found in MPO-AAV. Renal disease is common in AAV, estimated at 80%. However, the overall prevalence does not differ between PR3-AAV and MPO-AAV. Outcome differs according to ANCA serotype. PR3-AAV patients relapse earlier and more frequently than patients with MPO-AAV [1]. Patient and/or renal survival may be worse in MPO-AAV when compared with PR3-AAV, but studies demonstrate conflicting results [1]. Patient survival in PR3-AAV is largely determined by diffuse alveolar hemorrhage at presentation and renal relapses during follow-up, whereas renal and/or patient survival in MPO-AAV is related to smoldering disease, causing pulmonary fibrosis, cardiovascular disease and end-stage renal disease. The pathophysiology of the initial phases of PR3-AAV and MPO-AAV differs. In PR3-AAV, ciliary motility is severely reduced, facilitating chronic nasal carriage of S. aureus [6]. In addition, ‘chronic inflammatory and/or infectious rhino sinusitis’ in PR3-AAV patients results in the release of high-mobility group box 1 (HMGB1), whereas in MPO-AAV patients, ‘chronic rhinosinusitis with nasal polyps’ results in the release of HMGB1 [7]. The release of HMGB1 results in the generation of neutrophil extracellular traps (NETs) [8], a process that might be pivotal for breaking tolerance to MPO and/or PR3 [9]. In addition, in MPO-AAV, silica exposure and smoking—both risk factors for MPO-AAV—are well-known causes of NET formation [10, 11], whereas in PR3-AAV, S. aureus causes NET formation [12]. Once ANCA is induced, there are only a few pathophysiological differences between MPO-ANCA and PR3-ANCA induction of glomerulonephritis and/or vasculitis [1]. While neutrophils express low levels of PR3 and MPO on their cell surfaces, PR3 and MPO are expressed extensively upon stimulation (e.g. with tumor necrosis factor α), making it possible for ANCA to bind. Both ANCA serotypes are able to activate neutrophils via Fab as well as Fcγ engagement, resulting in the release of inflammatory mediators. Importantly, ANCA-stimulated neutrophils induce the generation of NETs, a process that is potentiated by HMGB1 [13]. Recently Schreiber et al. [14] demonstrated that these NETs cause endothelial damage, a process that requires receptor-interacting protein kinase (RIPK)1/3/mixed-lineage kinase domain (MLKL)-dependent necroptosis. Moreover, ANCA activation of neutrophils and ANCA-induced NET formation results in complement activation via the alternative pathway, representing a pro-inflammatory amplification loop in AAV that is hypothesized to be essential for disease induction and/or progression [15]. In order to demonstrate the pathogenicity of ANCA in vivo, several animal models have been developed. Anti-MPO antibodies cause severe vasculitis and glomerulonephritis when initial immune complex formation occurs. Injecting mice with anti-MPO antibodies (raised in MPO−/− mice) results in a mild form of necrotizing and crescentic glomerulonephritis (NCGN), while an additional injection with lipopolysaccharide (LPS) causes more severe NCGN [15]. Similarly, humanized NOD-SCID-IL2Rγ−/− mice injected with human immunoglobulin G containing anti-PR3 antibodies showed the development of lung hemorrhage and mild kidney disease when LPS was injected as well. From these studies, we hypothesized that AAV is a two-hit disease model in which, along with ANCA, a pro-inflammatory signal such as caused by an infection is pivotal for disease development. In agreement with this hypothesis, we recently observed that ANCA increases were more often followed by a relapse when they occurred during the flu season [5]. Based on these pathophysiological differences and similarities, PR3-AAV and MPO-AAV are treated differently. Initial stage PR3-AAV (the so-called loco-regional GPA) may be treated with trimethoprim–sulfamethoxazole [16], whereas nonrenal or non-organ-threatening MPO-AAV may be treated with corticosteroids only [17]. Induction–remission therapy in MPO-AAV or PR3-AAV is not different when AAV patients have systemic vasculitis with organ involvement such as kidney involvement. In these cases, induction–remission therapy should consist of a combination of corticosteroids and either cyclophosphamide or rituximab. Rituximab may be the preferred choice, as has been demonstrated in the Rituximab versus cyclophposphamide for AAV (RAVE) study with PR3-AAV patients [18]. Although it is mentioned in several guidelines [19] that patients with nonsevere disease can be treated with either methotrexate or mycophenolate mofetil, there is not much published evidence that this therapy is equivalent to an induction regimen consisting of either rituximab or cyclophosphamide in these patients with nonsevere disease. Different approaches regarding treatment regimens should be used for maintenance therapy. As mentioned earlier, outcome in PR3-AAV patients differs from outcome in MPO-AAV patients and this affects treatment decisions. In trials regarding maintenance treatment of AAV, it has been demonstrated that rituximab 500 mg given every 6 months until 2 years after diagnosis is superior to azathioprine maintenance therapy [20]. It is unclear whether rituximab reinfusion should be given every 6 months or if rituximab should only be given when CD19+ B lymphocytes or ANCA has reappeared and/or increased markedly [21]. In this latter study, it was demonstrated that patients who were treated with a fixed schedule received significantly more rituximab. Furthermore, also in this study, PR3-ANCA was found to be a risk factor for a major relapse, suggesting that PR3-AAV should be treated with fixed schedules, whereas it is probably safe to treat MPO-AAV based on reappearance of ANCA and/or CD19+ B cells. Currently a rituximab maintenance regimen is approved by the US Food and Drug Administration but not yet by other regulators. So azathioprine is currently used as the best alternative for maintenance therapy in countries other than the USA. However, azathioprine has the disadvantage that it is associated with a high risk of a hypersensitivity reaction, which may occur in 9% of patients with AAV [22]. In these cases, other alternatives such as methotrexate and/or mycophenolate mofetil should be chosen. One important question is how long maintenance therapy should be continued. Recently a prospective randomized trial in which two different durations of maintenance immunosuppressive therapy were compared for the prevention of relapse in AAV was published [23]. In this study, it was demonstrated that prolonged azathioprine therapy (duration 4 years) was associated with a reduced relapse risk. An important risk factor for relapse was ANCA positivity at the start of azathioprine maintenance therapy (i.e. at the time of randomization). In line with this study, we performed a study in which patients who were PR3-ANCA positive at the start of maintenance therapy (i.e. after remission–induction with cyclophosphamide) were randomized to receive either standard (1 year after diagnosis and subsequent tapering) or extended (4 years after diagnosis and tapered thereafter) azathioprine maintenance therapy. In this study, 46% of patients relapsed in the standard therapy group versus 24% in the extended therapy group [24]. From these studies, it can be concluded that prolonged azathioprine might be an option for PR3-AAV but not for MPO-AAV, in which the risk for a relapse is substantially lower [1]. Also, other treatment options should be considered. In PR3-AAV it has been demonstrated that maintenance therapy with trimethoprim–sulfamethoxazole (2 × 960 mg/day) reduces the risk of relapse by 66% [16]. Otherwise, it can be postulated that patients with PR3-AAV infections should be treated as soon as possible, for instance, by instructing patients to start antibiotics once symptoms of bacterial infection are present. In addition, vitamin D deficiency in these PR3-AAV patients should be avoided, especially during fall and winter [5]. Otherwise, in MPO-AAV, smoking should be stopped and silica exposure should be avoided. In addition, since persistent proteinuria is an important risk factor for the development of end-stage renal disease, MPO-AAV patients should receive a combination of a salt-restricted diet and an angiotensin-converting enzyme inhibitor and/or an angiotensin II receptor blocker to reduce proteinuria. Cardiovascular risk factor management is important in all AAV patients but more so in patients with MPO-AAV [1]. New treatments in AAV are currently being studied [15]. Whether these or future (ANCA-specific and/or antigen-specific) treatments will be differentially used for PR3-AAV versus MPO-AAV remains to be investigated [25]. Finally, categorizing patients by ANCA serotype instead of clinical diagnosis may decrease selection bias for including patients in clinical trials. Evidence that PR3-AAV and MPO-AAV are two distinct varieties of one entity has accumulated over the last 30 years. Remarkably, in vitro and in vivo experiments have demonstrated great similarities between the modes of action of both serotypes. Therefore induction–remission therapy in generalized PR3-AAV and MPO-AAV is not different. After induction–remission therapy, however, a much higher relapse risk is found in PR3-AAV compared with MPO-AAV. Importantly, triggers for disease induction and relapse differ between the two serotypes. Therefore, breaking up AAV into PR3-AAV and MPO-AAV is currently mainly important for therapy of the initial (nonrenal) phases of the disease and for prevention of relapses. In addition, the discrimination between PR3-AAV and MPO-AAV might prove to be of great significance in the scope of future clinical trials. None declared. (See related articles by Watts. Should proteinase-3 and myeloperoxidase anti-neutrophil cytoplasmic antibody vasculitis be treated differently: part 1. Nephrol Dial Transplant 2019; 34: 381--383; Fouque and Fervenza. Editorial: a new era in anti-neutrophil cytoplasmic antibody vasculitis. Nephrol Dial Transplant 2019; 34: 379--381)

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.