• COVID-19
• Systemic vasculitis
• Vaccination

The complement C5a-C5a receptor (C5aR) signalling axis plays a crucial role in the pathogenesis of both antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) and COVID-19. Complement components such as the membrane attack complex, C5a, C3a and factor B are higher in patients with active AAV compared with healthy controls.1 Similarly, experimental studies have shown abundant expression of C5aR1 and increased levels of soluble C5a in severe COVID-19.2 Following the ADVOCATE trial, avacopan, a C5aR antagonist, has been approved by the US Food and Drug Administration (FDA) as an adjunct therapy for remission induction in AAV.3 Similarly, vilobelimab, an anti-C5a monoclonal antibody, was granted Emergency Use Authorization by the FDA for severe COVID-19 based on the results of the PANAMO study.4 The risk of severe COVID-19 is increased in patients receiving rituximab and high-dose glucocorticoids.5

Here, we report the outcomes of eight patients with AAV treated with avacopan who developed COVID-19, which to our knowledge has not been reported to date. We retrospectively identified patients ≥18 years old with AAV at the Johns Hopkins Vasculitis Center in Baltimore, Maryland and the Vasculitis and Glomerulonephritis Center at the Massachusetts General Hospital in Boston, Massachusetts who contracted COVID-19 while on avacopan. Data on ANCA serotype, concomitant immunosuppressive drug use, COVID-19 severity, treatment and vaccination status were retrieved.

Patient characteristics are outlined in table 1. All patients were previously vaccinated against COVID-19. The timing of vaccination in relation to B cell depletion is outlined in table 1. Spike antibody levels prior to this episode of COVID-19 were positive (n=3), negative (n=1) or not tested (n=4). The median (IQR) time from the start of induction therapy to avacopan initiation was 15 (10–30) days. The median (IQR) time from the start of induction therapy to COVID-19 infection was 116 (75–161) days. The median (IQR) time from avacopan initiation to COVID-19 was 78 (59–149) days. In addition to avacopan, induction treatment included rituximab+cyclophosphamide+glucocorticoids (n=6) and rituximab+glucocorticoids (n=2). Concomitant immunosuppression at the time of COVID-19 infection included rituximab (n=8), prednisone (n=2) and oral cyclophosphamide (n=1). All patients remained on their treatment, but cyclophosphamide was held. Two patients did not receive COVID-19 treatment, while the remainder received nirmatrelvir/ritonavir (n=3), bebtelovimab (n=1), molnupiravir (n=1), and remdesivir and dexamethasone (n=1). The severity of COVID-19 was mild in all patients except for one who was hospitalised and died from acute on chronic respiratory failure secondary to severe COVID-19 on a background of ANCA-related interstitial lung disease. In this patient, the diffusing capacity of the lungs for carbon monoxide was 31% predicted preceding COVID-19. Of the seven surviving patients, all recovered completely from acute COVID-19, and no patient experienced long COVID.

Despite significant immunosuppression, COVID-19 was mild in nearly all patients with AAV receiving avacopan. The one patient with severe COVID-19 had advanced fibrotic lung disease. Larger studies with a comparator arm are needed to confirm the effects of C5aR blockade with avacopan in COVID-19.