Long-term efficacy and safety of the Lupus-Cruces Nephritis protocol: a propensity score study of the Lupus-Cruces and Lupus-Bordeaux cohorts

Abstract

Objective To assess the efficacy and toxicity of the Lupus-Cruces Nephritis (LCN) protocol compared with standard of care (SOC) with cyclophosphamide (CYC) or mycophenolate in patients with lupus nephritis (LN) during an extended follow-up time up to 10 years.Methods Patients with biopsy-proven class III, IV or V LN treated with LCN were compared with SOC. Patients in the LCN were treated with a CYC plus repeated methylprednisolone pulse-based regimen. The achievement of complete renal response (CRR) and the progression to chronic kidney disease (CKD) were the two main outcomes. Glucocorticoid (GC)-related toxicity, major infections and damage accrual were also analysed. A propensity score (PS)-adjusted multivariate analysis was used to overcome the confounding-by-indication bias.Results 147 patients were included in this study (47 LCN and 100 SOC). CRR at 12 months was 85% vs 44%, respectively (p<0.001). Eventually, 96% patients in the LCN group achieved CRR vs 74% patients in the SOC (p=0.002). In the multivariate PS-adjusted Cox model, LCN patients were more likely to eventually achieve CRR (PS-adjusted HR 3.5, 95% CI 2.2 to 5.5, p<0.001). The risk of progression to CKD was lower in LCN patients (PS-adjusted HR 0.3, 95% CI 0.11 to 0.82, p=0.019). The risks of GC-induced toxicity, renal or GC-related damage accrual and major infections were also lower in the LCN group: adjusted HR 0.09, 95% CI 0.02 to 0.39; PS-adjusted HR 0.14, 95% CI 0.04 to 0.4; PS-adjusted HR 0.2, 95% CI 0.046 to 0.95; respectively.Conclusions This study confirms the LCN protocol as an effective and safe, in addition to widely available and affordable, regimen for the induction therapy of LN.