Eosinophilic granulomatosis with polyangiitis (EGPA), previously referred to as Churg-Strauss syndrome, is a condition associated with asthma and eosinophilia and characterized by eosinophil-rich and necrotizing granulomatous inflammation, frequently affecting the respiratory tract, and necrotizing vasculitis predominantly affecting small to medium vessels (1–4). This rare autoimmune disease has an incidence ranging from 0.5 to 4.2 cases per million person-years, and a global prevalence estimated to be between 10 and 18 cases per million inhabitants (1, 5). The mean age at diagnosis is 50 years and there are no differences in incidence between men and women (6).

EGPA is classified as a antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) even though ANCA are present in a minority of cases (2). Indeed, its clinical and biochemical presentation is different from granulomatosis with polyangiitis (GPA) and microscopic polyangiitis (MPA). EGPA is characterized by three phases: 1) a prodromal “allergic” phase with asthma, allergic rhinitis and sinusitis, 2) a subsequent variable period of up to 8–10 years with eosinophilia and pulmonary infiltrates, eosinophilic cardiomyopathy and gastrointestinal involvement, and 3) a third phase in which the features of vasculitis, as palpable purpura and glomerulonephritis, occur in association with ANCA positivity, usually with anti-myeloperoxidase (anti-MPO) specificity (1, 5, 7). However, the phases may not follow this order, there may be an overlap between different phases and some patients may not exhibit vasculitic complications (8).

The clinical presentation of EGPA is heterogeneous and current evidence suggests that two different phenotypes are associated with ANCA positivity or negativity. Approximately 40% of individuals with EGPA are found to be positive for ANCA, with predominant perinuclear staining and anti-MPO specificity in about 65% of cases (6, 7, 9). The prodromal phase of EGPA is commonly characterized by asthma and ear-nose-throat (ENT) disease. Asthma is observed in over 90% of patients, while ENT disease occurs in approximately 60–80% of individuals. Importantly, both asthma and ENT disease are prevalent in both ANCA-positive and ANCA-negative patients (6). However, the classical manifestations of vasculitis, such as glomerulonephritis, peripheral neuropathy, and purpura, tend to be more common in ANCA-positive patients, while cardiac involvement and gastroenteritis, addressed as eosinophilic features, are more frequent in ANCA-negative patients (6, 7, 9, 10).

Due to the heterogeneity of the clinical presentation, the diagnosis of EGPA can be challenging. None of the disease features alone can be considered pathognomonic. Moreover, EGPA commonly presents as a phasic disease, where both clinical and pathological findings vary based on the specific anatomical site affected and the phase of the illness. Churg and Strauss originally described allergic granulomatosis and angiitis in 1951 based on a study of 13 patients, of which 11 were autopsied. These patients exhibited severe asthma and shared common histological features, including blood and tissue eosinophilia, necrotizing vasculitis, and necrotizing granulomas centered around necrotic eosinophils. However, it is important to note that not all patients displayed all three of these pathological criteria, making them unreliable for clinical diagnosis in the majority of cases (11).

The diagnosis of EGPA should be considered in individuals who have a history of asthma, chronic rhinosinusitis, and eosinophilia, and subsequently develop end-organ involvement, particularly peripheral neuropathy, lung infiltrates, and cardiomyopathy, but also skin, gastrointestinal and kidney involvement (4, 12). Although many attempts to develop diagnostic criteria have been made (13–15), none of them has been validated for diagnosis (12). In the MIRRA trial eligibility criteria to define EGPA have been proposed, but they still need validation. These criteria encompassed asthma, eosinophilia, and the occurrence of at least two of the following manifestations: histological confirmation of eosinophilic vasculitis, perivascular eosinophilic infiltration, or eosinophil-rich granulomatous inflammation; neuropathy; pulmonary infiltrates; sino-nasal abnormalities; cardiomyopathy; glomerulonephritis; alveolar hemorrhage; palpable purpura; and ANCA positivity (16). Recently, the Diagnosis and Classification criteria in Vasculitis Study (DCVAS) defined the American College of Rheumatology/European Alliance of Associations for Rheumatology (ACR/EULAR)-endorsed criteria for the classification of small-and medium-sized vessel vasculitis, including EGPA (2). These criteria consist of positively scored parameters that increase the likelihood of EGPA diagnosis, such as a maximum eosinophil count ≥1 × 10⁹/L (+5 points), nasal polyps (+3), obstructive airway disease (+3), extravascular eosinophilic-predominant inflammation (+2), and motor neuropathy and/or mononeuritis multiplex not caused by radiculopathy (+1). In contrast, there are certain parameters that, when negatively scored, reduce the likelihood of EGPA. These include a cytoplasmic ANCA (C-ANCA) pattern on immunofluorescence or positive anti-proteinase 3 (PR3)-ANCA (−3), as well as the presence of hematuria (−1). If the cumulative score reaches 6 or more, a patient with a diagnosis of small-or medium-sized vessel vasculitis can be classified as having EGPA. This scoring system achieved a sensitivity of 85% and a specificity of 99% (2).

When investigating a patient suspected of having EGPA, it is crucial to both exclude known causes of eosinophilia and ascertain the presence of a vasculitic process (11, 17). EGPA can be differentially diagnosed from several other diseases, including:

In summary, the diagnosis of EGPA is primarily based on clinical evaluation. The presence of asthma, rhinitis or sinusitis, along with peripheral eosinophilia and symptoms suggestive of vasculitis, supports the diagnosis. However, when feasible, obtaining a tissue biopsy is recommended to further confirm the diagnosis.

Compared to GPA and MPA, where kidney involvement is prevalent (70 and 95% of cases, respectively), only a minority (25%) of patients with EGPA will develop renal disease (6, 34, 35), more often associated with the ANCA-positive phenotype. In a retrospective study on 116 patients with EGPA, ANCA positivity was more frequently observed in cases with renal involvement (75% vs. 25.7%) (34), reaching peaks of 84% in ANCA+ cases (33). The association of ANCA positivity with glomerulonephritis in EGPA is so well established that in the 2012 Revised International Chapel Hill Consensus Conference Nomenclature of Vasculitides the sentence “ANCA is more frequent in EGPA when glomerulonephritis is present” has been added to EGPA definition (14). In a retrospective work of 31 patients with EGPA, one half presented with urinary abnormalities and the other one with RPGN (34). However, RPGN is often observed in cases of renal involvement (6, 33, 34, 36). Thus, clinical presentation of EGPA cases with renal involvement can be protean, ranging from urinary abnormalities to acute kidney injury (AKI) and rapidly progressive glomerulonephritis (RPGN) (Table 1), stressing the need for renal biopsy to confirm and characterize the specific kidney modifications (32, 37). As for GPA and MPA, the most frequent histological manifestation of EGPA is necrotizing pauci-immune glomerulonephritis (Figure 2A) (33, 34, 36), with up to 78% of AKI in EGPA showing this pattern of injury (33). As in the other ANCA-associated glomerulonephritis, the progressive formation of crescents eventually leads to Bowman capsule rupture, which have been recently demonstrated to better stratify the prognosis of these patients if associated with more widely used Berden classification and Renal Risk Score (38, 39). Even if rare, cases of EGPA with positive ANCA can present with necrotizing and transmural arteritis of small and/or medium sized arteries at renal biopsy (Figure 2B), which recently showed to have a worse prognostic significance (40, 41). Most EGPA patients with RPGN display a focal or a crescentic histological class according to Berden et al. (33, 42), suggesting that renal involvement is usually detected early in this setting compared to MPA patients with anti-MPO antibodies, in which renal involvement is diagnosed at advanced stages. This is probably determined by the severity of extra-renal symptoms, which lead to a rapid diagnosis of systemic vasculitis. As compared to the ANCA+ forms, ANCA-negative patients may show alternative patterns of injury, ranging from membranous nephropathy (MN) (10%) to membranoproliferative glomerulonephritis (3%) or acute interstitial nephritis (10%), rarely with giant cell reaction and/or interstitial non-necrotizing granulomas (Figure 2C), or more frequently with an eosinophilic predominance (Figure 2D) (33). Prominent interstitial eosinophilic infiltration is frequently present together with pauci-immune RPGN in almost half of cases, showing pathogenetic differences as compared to the interstitial inflammation seen in GPA and MPA (33). In particular, the preferential activation of T helper 2 (Th2) phenotype orchestrates the adaptive immune response and enhances eosinophilic reactions through IL-5 secretion and GPA33 as a more frequent pathway in ANCA-negative EGPA. This mechanism may determine the eosinophilic feature, in which eosinophils are essential to mediate tissue damage (20). On the other hand, up to 10% of cases can show MN as the sole renal manifestation of EGPA (33), mainly in ANCA-negative cases, showing some common genetic background with HLA alleles found in other MN forms (43). Furthermore, the presence of an overlap syndrome between AAV and IgG4-related disease has been reported, being both MN and IgG4+ plasma cell rich interstitial nephritis manifestations of IgG4-related disease (Figure 3) (24, 44). Hence, MN and ANCA-negative EGPA association may be not casual, but further studies are required to explore this hypothesis. In addition to glomerular diseases, obstructive uropathy due to ureteral involvement has been occasionally reported.

The treatment of EGPA is based on remission-induction and remission-maintenance (45–47). Remission-induction treatment should be tailored to disease severity, defined according to the presence at the diagnosis of at least one adverse prognostic factor included in the Five-Factor Score (FFS) (12). The FFS is a prognostic tool used to assess the risk of mortality in patients diagnosed with EGPA and includes: renal insufficiency defined as serum creatinine >1.58 mg/dL, proteinuria exceeding 1 g per day, cardiomyopathy, gastrointestinal involvement and central nervous system involvement (44). FFS was revised in 2011 by Guillevin et al., which added age > 65 years to cardiomyopathy, gastrointestinal involvement, and renal insufficiency (creatinine ≥1.7 mg/dL), eliminating central nervous system involvement. Independently from which score is chosen, renal involvement is sufficient to define a severe disease and drive a more aggressive treatment. In Figure 4, the indications from the three main references for the treatment of EGPA are summarized (12, 48, 49). Patients with severe disease are recommended to receive pulsed intravenous glucocorticoids (GCs) (typically daily methylprednisolone pulses of 500–1,000 mg each over 3 days, for a maximum total dose of 3 g), followed by high-dose oral GCs (0.75–1 mg/kg per day) (12, 49). In severe disease cyclophosphamide (CYC) should be added to GCs for remission induction (49). In a randomized controlled trial (RCT) on 48 EGPA patients with FFS ≥ 1, of which 19 with renal involvement, a lower rate of minor relapses was observed after 12 CYC pulses than after 6 CYC pulses (administered every 2 weeks for 1 month, then every 4 weeks thereafter, at a dose of 0.6 g/m² per pulse) (50). However, 12 CYC pulses did not improved response rate or reduced severe relapses (50). Therefore, there is no consensus on the optimal duration of CYC therapy, which must be balanced between the efficacy and the potentially harmful dose-related side effects (51). The latest EULAR guidelines recommend switching to a less intensive remission maintenance therapy after 6 pulses of CYC if remission is achieved (49). In cases where patients show slow improvement but do not achieve complete remission within 6 months, longer induction periods with CYC may be considered, extending up to 9–12 months (12).

Data from observational studies suggested RTX as a potential alternative for remission induction (28, 52–54). The RCT REOVAS, presented as abstract at the 2021 American College of Rheumatology convergence, found that in EGPA patients with FFS ≥ 1 RTX (two 1-gram pulses on day 1 and 15) is comparable with CYC (nine intravenous pulses over 13 weeks) for induction of remission (defined as Birmingham Vasculitis Activity Score [BVAS] of zero and a prednisone dose ≤7.5 mg per day). The two groups showed comparable adverse events, cumulative prednisone doses, and quality of life (55). Differently from previous observational studies (28, 54), the response to RTX was similar in ANCA-positive and ANCA-negative patients (55). Hence, this study suggests that RTX may be an alternative to CYC and that EGPA treatment should not be influenced solely by ANCA status (49). Nevertheless, the REOVAS trial was structured as a superiority trial. Given that the primary endpoint of on-treatment remission was attained at comparable rates in both cohorts, there exists insufficient substantial evidence to definitively establish non-inferiority (55).

There are no data on GCs tapering strategies in EGPA. Therefore, evidence obtained from the PEXIVAS trial on GCs tapering in GPA and MPA should be used as orientation (56). The goal is to achieve through a stepwise reduction a dose of 5 mg prednisolone equivalent per day by 4–5 months (49). The future goal for treatment strategies is to spare GCs. In this perspective, mepolizumab may be an interesting add-on therapy. The MIRRA RCT, which investigated the efficacy and safety of mepolizumab compared to placebo in achieving remission in EGPA patients without organ-or life-threatening manifestations, demonstrated a higher percentage of remission and more weeks in remission in patients treated with mepolizumab, allowing a reduction in the GCs cumulative dose (16). Some case reports described the effectiveness of mepolizumab as an add-on therapy to induce remission, in particular in the case of pure interstitial nephritis (57–59). Mepolizumab was given in addition to GCs and CYC (59), or to GCs alone (57, 58). In another case report, the addition of mepolizumab to GCs and RTX achieved successful treatment of a case of EGPA with multiorgan involvement, notably including renal damage determined by necrotizing pauci-immune glomerulonephritis and severe interstitial nephritis (60). The hypothesis that a synergistic treatment with complementary mechanisms of action, such as the combination of RTX and MEPO, could enhance the remission rates of EGPA components, finds further validation in the outcomes of the European multicenter retrospective observational study conducted by Bettiol et al. (61) This study demonstrated the efficacy of sequential RTX and mepolizumab therapy in inducing and sustaining remission for both systemic and respiratory EGPA symptoms. Additionally, two other anti-IL-5 therapies, reslizumab and benralizumab, have undergone investigation, yielding promising results as GCs sparing agents in phase 2 open-label trials involving a limited number of patients (62, 63). In theory, these drugs should be more effective in ANCA-negative individuals who exhibit a profile with eosinophilic features. However, this hypothesis has not yet been substantiated by clinical data (64, 65). Currently, the efficacy and safety of benralizumab are under evaluation, comparing it to mepolizumab in EGPA patients who are receiving standard care therapy (NCT 04157348). Another noteworthy anti-IL-5 agent, notable for a long-acting properties, currently under investigation is depemokimab (NCT05263934) (66). However, further studies are needed to evaluate the actual role of mepolizumab and other anti-IL-5 agents in the induction therapy of EGPA with renal involvement.

After remission-induction, remission-maintenance is necessary to avoid relapses, but also to reduce the risk of drug-related toxicity. However, data regarding patients with severe EGPA are scant and not exhaustive. Observational studies failed to prove a better relapse-free survival with azathioprine (AZA), methotrexate (MTX) and leflunomide combined with GCs, compared to GCs alone (67, 68). However, these drugs are often used in clinical practice during the maintenance phase (12, 48, 69). RTX seems to be effective also for remission maintenance. In an observational study on 21 patients with EGPA, scheduled maintenance RTX (500 mg every 6 months) reduced the relapse rate compared to single gram infusion administered in case of relapse (29). The use of CYC for remission maintenance is not indicated because of its toxicity (49). Moreover, in a single-center prospective RCT that investigated the use of oral CYC versus MTX for 1 year following remission induction with CYC in different subtypes of AAV, no significant difference in relapse rates was observed between the two treatment arms in the subgroup of 30 patients with severe EGPA (70).

The treatment of severe systemic relapses is based on RTX or CYC. RTX is preferred when re-treatment with CYC is contraindicated, or in patients who previously achieved remission on RTX or failed to respond to CYC. CYC should be considered in cases of severe or life-threatening complications and/or in patients who have previously failed to respond to RTX (12, 28, 29, 52, 70).

It’s now recognized that ANCA status describes two different phenotypes of EGPA, characterized by different clinical presentation, and probably also pathogenesis. However, there is no evidence that different ANCA phenotypes necessitate different therapeutic approaches (12). Preliminary evidence suggested that ANCA-positive patients may be more susceptible to RTX (28, 54), but these results have been questioned by the REOVAS trial (55). In the context of ANCA-negative cases linked to Th2-related etiology, anti-IL5 therapies are emerging as a promising approach (71). Nevertheless, clinical data remains relatively limited, and another potentially effective biologic, omalizumab, an anti-IgE monoclonal antibody, has exhibited efficacy in asthma treatment but not in extrapulmonary manifestations (72, 73). Hence, the prospect of selectively targeting eosinophilic or vasculitic inflammation undeniably represents a groundbreaking approach. However, to successfully integrate these novel treatments into the landscape of this complex disease, additional clinical data is required, and significant progress must be made in understanding the etiopathogenesis of renal involvement in EGPA.

The presence of kidney involvement in AAV determines higher morbidity and mortality rates (7, 74). However, EGPA patients with renal involvement have favorable outcomes when treated with GCs and immunosuppressants. Compared to GPA and MPA, patients’ survival is good (34, 75–79). In the retrospective work by Sinico et al., the five-year mortality rate in EGPA patients with renal abnormalities was 11.7%. In the same study, favorable renal survival outcomes were reported. Only one patient reached end-stage renal disease (ESRD), and none of the patients experienced a doubling of serum creatinine levels after a mean follow-up of approximately 5 years (34). In the study by Durel et al., after a median follow-up of 48 months 92% of patients were alive and 17% (11 patients) reached ESRD, with nine patients (14%) on chronic dialysis and two (3%) who underwent kidney transplantation (33). In another retrospective study, out of twelve patients monitored over a median duration of 43.5 months, two patients ESRD, nine exhibited chronic kidney disease with an estimated glomerular filtration rate (eGFR) < 60 mL/min, and two patients maintained a normal eGFR (36). ANCA status may have prognostic implications since overall survival seems to be worse in ANCA-negative patients, while relapses do appear to occur more frequently in ANCA-positive patients, even if some controversies exist (6, 9). The worse prognosis of ANCA negative patients is probably caused by the higher frequency of cardiac involvement (12).

EGPA is a rare form of vasculitis, and renal involvement is present in just 25% of cases. However, kidney disease is an adverse prognostic factor and may prompt a more aggressive treatment based on GCs and immunosuppressants. At the moment, the management of EGPA and kidney disease in EGPA is challenging, as it remains a condition that is not easily diagnosed and without treatments validated by clinical trials. RCT are necessary to fill this gap and to test current and newer biological therapies. Moreover, other unresolved challenges remain for the future, as understanding better the pathogenesis and the role of genetics, and if these are truly associated with the two phenotypes based on ANCA status. This will open new perspectives on the treatment of EGPA, allowing the development of therapies tailored to the different EGPA subsets.

FR and RS: conceptualization. FR, MC, and RS: data creation. FR, VL’M, and RS: writing. FR, VL’M, and FP: visualization. FR, VL’M, MC, FP, and RS: review editing. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.