David Jayne, Alice Guerini, Andreas Kronbichler, Federico Alberici, Matija Crnogorac, Annette Bruchfeld, Federica Mescia, Anna Ricchiuto, Iva Gunnarsson, Rosanna Lacetera, Anna Juto, Mark A. Little, Renato Alberto Sinico, Martina Uzzo, Vladimir Stoyanov, Francesco Scolari, Stefania Affatato, Giorgio Trivioli, Gaetano La Manna, Mario Cozzolino, Jennifer Scott, Krešimir Galešić, Stephen P. McAdoo, Jennifer O'Brien, Federico Pieruzzi, and Marco Allinovi
Background and Aims Renal involvement in ANCA-associated vasculitis (AAV) impacts significantly on patients’ prognosis. The role of different induction regimens on remission rates and long-term renal outcomes according to renal histological characteristics has not been explored yet. Method AAV patients with biopsy-proven renal involvement were collected retrospectively from eleven centers and stratified according to the induction regimen employed: Rituximab (RTX), Cyclophosphamide (CYC) or both (RTX-CYC). Kidney biopsies were classified according to the Berden and Brix classifications. Renal remission rate was assessed 6 months after the induction regimen and defined as a renal Birmingham Vasculitis Activity Score (BVAS) of 0. Among patients who achieved remission at 6 months, renal relapse was defined as a renal-BVAS>0 associated with an increase in immunosuppressive treatment. ESRD was defined as an eGFR Results 323 patients were identified and followed-up for a median time of 36 months (IQR 18-72). The cohort included 38% patients with GPA and 62% with MPA, 53% patients were MPO-ANCA and 41% PR3-ANCA positive. The median baseline eGFR in the overall cohort was 19 ml/min/1,73m2 (IQR 12- 34). 58% of patients were treated with CYC, 24% with RTX-CYC and 18% with RTX. According to the Berden classification, 24% biopsies were classified as Focal, 31% as Crescentic, 33% as Mixed and 12% as Sclerotic. The Brix score was assessable in 270/323 (84%) patients: 17%, 52% and 31% were respectively in the Low, Medium and High-risk class. The overall renal remission at 6 months was 90%; according to the Berden classification, 94% patients achieved remission in the Focal, 88% in the Crescentic, 91% in the Mixed and 86% in the Sclerotic class. According to the Brix risk score, 88% patients achieved remission in the High risk, 91% in the Medium and 96% in the Low-risk class. According to induction regimen employed, 91%, 90% and 90% patients achieved remission in the RTX, CYC and RTX plus CYC group respectively. In a logistic regression model adjusted for sex, age, ANCA type, AAV diagnosis, creatinine and proteinuria at onset, the induction regimen employed was not predictive of renal remission at 6 months, neither in Berden Focal plus Crescentic and Mixed plus Sclerotic classes, nor in Brix High and Low plus Medium risk classes. Of the 185 patients with at least 6 months of follow-up available after remission, 25% experienced a renal relapse. In a Cox regression model adjusted for sex, age, ANCA type, AAV diagnosis, creatinine and proteinuria at onset, the induction regimen or histological score were not predictive of renal relapse. In the unadjusted survival analysis with the Kaplan-Maier curve, patients in the Crescentic group treated with RTX had a shorter ESRD-free survival compared to the CYC group (p=0.033) and the RTX-CYC group (p=0.044); figure 1: This was confirmed also with a Cox regression analysis adjusted for sex, age, ANCA type, AAV diagnosis, creatinine and proteinuria when comparing the RTX group with the CYC one (HR 8.30 [95% CI 1.64 to 42.01], p=0.011); figure 2: While the eGFR changes over time in the Focal plus Crescentic and Mixed plus Sclerotic classes showed a similar trend between treatment groups, in the Crescentic class the median eGFR values in the RTX group tended to be lower compared to the CYC and the RTX-CYC ones; figure 3: The rate of severe infections in the RTX, CYC and RTX-CYC group was respectively 6.3, 8.5 and 8.8 per 100 patient-years during the first 12 months. Conclusion in a retrospective multicenter survey, response rates and relapse risk after different induction regimens in AAV patients with renal involvement were comparable in the overall cohort and in the different histopathological subgroups. Although in a small subset of patients, the ESRD-free survival in the Crescentic class was shorter in the RTX group compared to the CYC one.