Background/Purpose: Lupus nephritis (LN) complicates 60% of systemic lupus erythematosus (SLE) patients within 5 years of diagnosis. Glucocorticoids and disease modifying immunosuppressive drugs (IS) are associated with adverse effects, treatment failure and relapses and treatment duration is uncertain. Our aims were to study long-term renal outcomes of LN patients who discontinued treatment after achieving disease remission and to identify predictors of renal relapse off IS therapy.

Methods: This was a retrospective cohort study of LN in our Lupus Clinic who discontinued IS treatment. Patient data was retrieved from October 1991 and through to 1^stDecember 2015, or until their last follow-up date. Renal relapse was defined as raised urine protein-creatinine ratio (UPCR) of ≥ 50mg/mmol or ≥50mg/mmol from baseline (equivalent to proteinuria 0.5g/L), or with renal biopsy proven recurrence of nephritis. Predictive factors for renal relapse were identified using logistic regression model. Data analysis was performed using SPSS Version 23.0.

Results: 43 patients were identified, 25 (58%) had sufficient data for analysis. 60% were Caucasians, 28% of Afro-Caribbean origin and 12% were Asian. 23 patients (92%) were female. Mean age at LN diagnosis was 33.5 ± 11.4 years. 7 had class IV and 7 had class V LN (28% for each group). The remaining patients had class III LN (16%), class III/IV LN (12%), and class II LN (8%). An association with anti phospholipid syndrome (APS) was found in 28% and 7 with positive aPL antibodies that remained asymptomatic. Mean duration of LN treatment was 5.6±4.2years. 17 patients (68%) received IV cyclophosphamide (CYC) as induction therapy. Azathioprine was the commonest maintenance agent (76%). Most received hydroxychloroquine throughout maintenance phase (76%). At the time of treatment discontinuation, mean eGFR was 87.7± 21.87 ml/1.73m²/min. Mean serum albumin was 41.3± 3.88 mmol/L and mean UPCR was 59.0± 85.11 mg/mmol. Anti- dsDNA antibodies persisted in 24% of the cohort. Despite discontinuation, 80% remained under active surveillance. Only 4 patients lost to follow-up. There was 1 death with cause unknown. Renal relapses were recorded in 7 patients (28%) and 1 patient developed non-renal flare. Mean time to event estimated at 47.2months (95% CI: 36.7,57.7). None developed end-stage renal disease after treatment discontinuation. Young age at LN diagnosis, low C3/C4, high UPCR, low eGFR upon discontinuation of treatment and ethnicity were six predictors of possible renal relapse. Our fitted model was statistically significant, χ²=18.22, p=0.011 and correctly classified 88% of cases.

Conclusion: Our long-term LN relapse rates after discontinuation of treatment are similar to previous studies. Nevertheless, reliable comparisons are limited by the retrospective nature of our study. It is possible to discontinue treatment for LN but the optimal duration of IS therapy remains uncertain and randomized clinical trials of treatment withdrawal are warranted.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/outcomes-of-lupus-nephritis-patients-following-discontinuation-of-treatment/