PO:16:234 | Differential efficacy of induction treatments for lupus nephritis at onset and relapses: an observational study

Giuseppe Alvise Ramirez^1|2, Gabriele D Gallina¹, Chiara Calabrese^1|2, Serena Nannipieri^1|2, Marta Calatroni^3|4, Francesco Reggiani^3|4, Giovanni Benanti¹, Enrica Paola Bozzolo¹, Raffaele Ceniccola², Luca Moroni^1|2, Marco Matucci-Cerinic^1|2, Gabriella Moroni^3|4, Lorenzo Dagna^1|2 | ¹IRCCS San Raffaele Hospital, Unit of Immunology, Rheumatology, Allergy and Rare Diseases Milan, Italy; ²Vita- Salute San Raffaele University Milan, Italy; ³IRCCS Humanitas Research Hospital, Nephrology and Dialysis Division Rozzano, Milan, Italy; ⁴Humanitas University, Department of Biomedical Sciences Rozzano, Milan, Italy

Background. Lupus nephritis (LN) is a frequent manifestation of SLE. Relapses are associated with progressively lower treatment response and increased risk of chronic kidney damage. Mycophenolate mofetil (MMF) and cyclophosphamide (CYC) are first-line options for the induction treatment of new-onset LN, with alternatives including Rituximab (RTX) or calcineurin inhibitors (CNI). Limited data are available on the comparative efficacy of these treatments at disease onset versus relapse. This study aimed to evaluate whether induction regimens for LN differ in efficacy between onset and FLARES, and whether switching treatment after relapse impacts renal outcomes.

Methods. This retrospective, double-center observational study included LN patients from 1964 to 2024. Renal responses were categorized according to EULAR criteria as Complete Response (CR), Partial Response (PR), or Failure (FL). The best renal response for each flare was considered. Patients receiving a different induction regimen at relapse compared to onset were classified as “switchers.” Statistical analysis was performed using SPSS; categorical variables were analyzed with Chi-square test, with significance set at p: 0.05.

Results. Data from 286 patients and 474 LN episodes were analyzed, including 188 relapses. Median age at onset was 32 years (IQR 24–41). 111 patients (39%) experienced one or more relapse(s), occurring a median of 37 months (IQR 3–88) after onset. Median follow-up after relapse was 155 months (IQR 69–261). At onset, induction therapy included CYC (43%), MMF (20%), CNI (4%), and RTX (3%). For relapses, treatments included CYC (29%), MMF (25%), CNI (12%), and RTX (5%). Additional patients received azathioprine or corticosteroids only. Renal outcomes at onset vs relapse were: CR in 76% vs 79%, PR in 12% vs 10%, and FL in 11% vs 11%, respectively. CR rates were high with MMF (83%), CNI (78%), and CYC (71%), though CYC had a higher FL rate (18%) compared to others (4%). Among relapsing patients, 59% were switchers. Switchers had numerically higher CR rates than non-switchers (94% vs 76%, p=0.08). Among those who switched to MMF, patients previously treated with CYC had significantly higher CR rates compared to those previously treated with RTX (86% vs 0%, p=0.03), and also compared to MMF non-switchers (86% vs 79%, p=0.07). Switchers to CYC from MMF or CNI had no FL. CNI switchers and non-switchers had similar outcomes, with only 4% FL. No CR were observed in relapsing patients initially treated with RTX.

Conclusions. MMF and CYC are the most effective agents for inducing LN remission at both onset and relapse, with MMF associated with fewer failures. CNI are also effective at relapse, with low FL rates. RTX at onset may be associated with more severe relapses. Switching therapy after relapse may improve CR rates, without increasing PR or FL.