compared with 94% with cyclophosphamide). However, in this study, the relapse rate and progression to a systemic form was higher in the methotrexate than in the cyclophosphamide arm (70% versus 47% at 18 months), which indicated the need for a longer methotrexate treatment duration (>12 months) if initially effective.

More recent studies with mycophenolate mofetil have also suggested that it could be used instead of cyclophosphamide in non-severe GPA disease, with mild renal involvement. However, and again, the subsequent relapse rates were high [154].

Finally, data on rituximab for limited GPA are promising but limited to small case series and are not approved for this indication (existing approvals for rituximab are only for severe ANCA-positive GPA; and limited GPA is often ANCA-negative).

Other Treatments in GPA

Intravenous Immunoglobulins

Intravenous immunoglobulins have been mainly used to treat refractory or recurrent GPA and/or GPA with ongoing and concomitant serious infection. The dose is 2 g/kg/month, over 2–5 days, with good immediate results, but is often only transient and/or not sustained (the disease often recurs after their discontinuation) [155]. However, the therapy has lower toxicity than with conventional immunosuppressants, ­particularly in terms of infection, and does not result in late neoplastic complications, which explains its transient use for GPA patients with ongoing infection, pregnant female patients, and/or children.

Plasma Exchange

Plasma exchange combined with induction chemotherapy has been proposed for patients with GPA and rapidly progressive necrotizing glomerulonephritis and/or alveolar hemorrhage, by analogy with anti-GBM antibody disease. However, the benefit of plasma exchange in such patients with ANCA vasculitis was mainly suggested by case series. Only one small prospective randomized study of patients with severe renal impairment, defined as a serum creatinine level >500 μmol/L, was conducted prior to 2010 and showed that plasma exchange improved the probability of good renal function at only 12 months but not later and had no impact on global survival [156]. The subsequent randomized international trial PEXIVAS included 704 patients with severe renal disease (20% required dialysis initially) and/or alveolar hemorrhage (18.5% had non-severe alveolar hemorrhage, 8.6% had severe alveolar hemorrhage) and showed no benefit of plasma exchange in terms of global survival or renal survival. Death from any cause or end-stage renal disease at the end of follow-up was 28.4% in the plasma exchange group versus 31% in the group not treated with plasma

exchange (p = 0.27), with similar findings in main subgroup analyses. There may still be a role for plasma exchange in patients with refractory disease and/or with kidney biopsy showing acute disease and minimal scarring, as a rescue adjunctive option, but with limited expectations, and a possible increased risk of infection (related to the procedure itself, although the reasons for this remains poorly understood).

Avacopan (and Other Complement Inhibitors/ Blockers): A Major Upcoming Change in the Treatment of GPA

Avacopan, an oral C5a receptor antagonist, was tested in an animal model of MPO-ANCA-induced vasculitis, and then in two open-label phase II studies, administered in addition to standard induction therapy with rituximab or cyclophosphamide with or without glucocorticoids, with promising results [46, 157, 158]. The safety profle of the drugs was extremely good. A larger randomized controlled study has thus been conducted and recently completed (ADVOCATE), which have confrmed these earlier results [159]. This double-­blind, randomized study included 330 patients with newly diagnosed or relapsing ANCA-positive GPA or microscopic polyangiitis. Sustained BVAS remission at Weeks 26 and 52 was achieved in 65.7% of the patients who received oral avacopan (30 mg, twice daily for 52 weeks) versus 54.9% in those treated instead with a conventional 5-month prednisone course (in addition, in both arms, to rituximabor cyclophosphamide-based induction, followed by azathioprine for the latter; p = 0.007). Of note, the renal parameters, including proteinuria, improved to a greater and faster level in avacopan recipients. Serious adverse events (excluding worsening vasculitis) occurred in 37.3% of the patients receiving avacopan versus 39% of those receiving prednisone. Patients with GPA could thus be treated with rituximab or cyclophosphamide and avacopan, instead of the conventional glucocorticoid treatment. Initial glucocorticoid pulses or a short initial course of glucocorticoids may still be considered (and were given to most of the ADVOCATE participants in fact) and might be useful [106].

Avacopan was approved by the FDA in late 2021 for the treatment of adult patients with severe GPA and microscopic polyangiitis. Once more largely accessible, this new drug may radically change the treatment of GPA (and microscopic polyangiitis), as it would lead to a (totally or at least partially) steroid-free approach. However, many subsequent questions have now been raised and will have to be studied, such as the optimal duration of avacopan use, its long-term safety, and cost-effciency [160]. Other complement inhibitors or blockers are also under investigation (ClinicalTrials.gov Identifer: NCT03712345 and NCT03895801).

CTLA4-Ig (Abatacept)

An open-label study reported promising results with abatacept for relapsing or refractory limited GPA [161], which will need to be further confrmed by the ongoing, prospective controlled study ABROGATE (ClinicalTrials.gov Identifer: NCT02108860).

Cotrimoxazole

Besides basic science evidence suggesting an association between GPA and Staphylococcus aureus carriage (in the nostrils; cf. Pathogenesis), some early studies in the 1980s reported that the prescription of antibiotics, especially cotrimoxazole (trimethoprim–sulfamethoxazole), combined with other treatments, could provide better disease control and reduce the relapse rate [29, 152]. When administered at a high dose (320 mg trimethoprim/day, sulfamethoxazole 1600 mg/day), combined with or following the usual maintenance treatments of GPA, cotrimoxazole further reduced the relapse rate by 40% at 1 year in one small controlled study, regardless of the presence of S. aureus on nasal swabs [29]. However, recent meta-analyses of all the studies on cotrimoxazole in GPA, including that only one, former randomized controlled study, have seriously questioned that fnding [162]. Cotrimoxazole (trimethoprim–sulfamethoxazole) cannot substitute for immunosuppressive maintenance, as it is clearly not as effective [153]. In addition, such a high dose can be hazardous for patients with end-stage renal disease and is not compatible with the use of methotrexate because of increased bone marrow toxicity.

However, and importantly, low-dose cotrimoxazole (160 mg trimethoprim with sulfamethoxazole 80 mg every day or, alternatively, 320 mg trimethoprim and sulfamethoxazole 1600 mg three times per week; dose to be adjusted in case of renal impairment) must be prescribed for prophylaxis against Pneumocystis jiroveci pneumonia in patients with GPA [120]. The risk of P. jiroveci pneumonia is relatively small but real in patients with GPA receiving cyclophosphamide or rituximab combined with high-dose glucocorticoids. Such a low dose likely has no beneft in terms of preventing relapse or controlling disease manifestations, but it was also shown to reduce the risk of infectious complications of other immunosuppressive treatments [162, 163], more common bacterial infections, especially upper respiratory tract infections. In case of allergy or intolerance to cotrimoxazole, prevention of Pneumocystis jiroveci pneumonia can be based on aerosolized pentamidine (300 mg every 4 weeks), oral dapsone (100 mg/day, which can cause hemolysis even without glucose-6-phosphate dehydrogenase defciency or more rarely methemoglobinemia), or atovaquone (1500 mg/day).

Other Agents

Smaller series or open-label studies have also been reported, most often with some encouraging results, but in highly spe-

cifc patient subsets, either with refractory GPA and/or at different points of their disease course, such as gusperimus (15-deoxyspergualin) or alemtuzumab (antiCD52), with interesting results, but their use has remained extremely limited, mainly because of their associated risks of adverse events [164, 165]. Other small series have also reported interesting results with ofatumumab, a fully humanized monoclonal antibody directed against CD20 on B cells, obinutuzumab, another anti-CD20 monoclonal, belimumab, a human monoclonal antibody that inhibits B-cell-activating factor (BAFF), tocilizumab, a humanized monoclonal antibody against the IL-6 receptor, or streptococcal enzymes IgG-degrading enzyme of Streptococcus pyogenes (IdeS) and endoglycosidase S (EndoS), which can to degrade and disarm IgG ANCAs.

Anti-TNF-α blockers such as in iximab and etanercept have been used in refractory ANCA vasculitis but are now seldom used, despite some benefcial results with in iximab (at 5 mg/kg on Days 0, 15, and 45 and then every 1–2 months) [166]. The results of the Wegener’s Granulomatosis Etanercept Trial (WGET) combined with conventional treatments were disappointing: etanercept added no beneft in preventing relapses as compared to placebo, and, more importantly, solid cancers, including unusual ones, developed in a concerning number of patients who had received etanercept (combined with glucocorticoids and cyclophosphamide), including several years after the end of the study [167]. Thus, the use of anti-TNF-α is not recommended for GPA.

Other potential therapeutic targets are constantly being identifed or studied, such as IL-17, IL-23, NETs, the Janus kinase/signal transducers and activators of transcription (JAK-STAT) signaling pathway, or dipeptidyl peptidase I (cathepsin C), but are all at extremely early stages of consideration. Intensive chemotherapy followed by autologous bone marrow or stem cell transplantation remain the anecdotal therapy for GPA and are under investigation in few selected centers.

Specifc Treatments for Certain Manifestations of GPA

GPA patients often need multiple other treatments, including local treatment of sinus in ammation with nasal rinses, myringotomy, dialysis for severe renal disease, local treatment of episcleritis, surgery for intestinal perforation, or pacing for severe conduction disorders. Kidney transplant for patients with end-stage renal disease or surgical reconstruction of nasal deformities can be considered in patients with sustained disease remission.

Treatment for subglottic and tracheobronchial stenoses is highly challenging and complex, especially because this complication often relapses and/or can worsen despite con-