FP187TWENTY YEARS OF LUPUS NEPHRITIS MANAGEMENT - EVALUATION OF A MULTI ETHNIC PATIENT COHORT TO IDENTIFY FACTORS AFFECTING OUTCOME

Abstract

Introduction and Aims: Lupus Nephritis (LN) is a common & severe manifestation of systemic lupus erythematosus (SLE) that can lead to ESRF over time. Long-term outcome data from the Euro-Lupus Nephritis & MAINTAIN cohorts suggest that a less stringent cut-off of proteinuria at one year (<0.8g/d & <0.7g/d respectively) better predicted good renal outcomes at 7 years than the usual remission criteria of <0.5g/d. This cut-off has yet to be validated with urinary protein creatinine ratio (uPCR) or in a larger cohort. We reviewed our cohort of LN patients to identify factors affecting both short & long-term outcomes. Method(s): Demographic, clinical & outcome data were reviewed for all renal biopsies showing LN from 1/1/1996 to 1/1/2016. Multiple logistic regression analysis was applied to identify factors affecting response to treatment at 1 year. Complete remission (CR): uPCR (mg/mmol) <50 & estimated glomerular filtration rate (eGFR mls/min/1.73m2) >60, or if <60 at screening, not fallen by >20%. Partial remission (PR): uPCR <300 with a >50% improvement from baseline &eGFR criteria as for CR. Non-remission (NR): failing to achieve PR by 1 year. Data were analysed by biopsy or by patient as appropriate and displayed as totals or median (range). Receiver operating characteristic (ROC) curves were generated to identify thresholds of eGFR & proteinuria at 1 year predictive of good long-term outcome. Result(s): 824 biopsies were performed on 484 patients over the 20 years. Median age at diagnosis: SLE 29 yrs; LN 32 yrs. Female: 81%. Ethnic distribution: 26% white, 27% black, 31% South Asian &3% East Asian. 95% fulfilled SLICC criteria for SLE. Median biopsy number/patient: 1 (1-8) with 40% of patients having >1 biopsy. Median length of follow-up from diagnosis LN was 8 yrs (0-46). At latest follow-up: started renal replacement therapy (RRT): 61 (13%), median time 5yrs (0-43); died: 25 (5%), median time4yrs (0-16);eGFR<60 (CKD3a or worse): 100 (21%), median time 4yrs (0-29). ISN/RPS Class of biopsy: III or IV+/-V61%, pure V 25%. 43% of biopsies were performed at the first presentation ofLN, 39% at flare, 12% due to persistent activity & 6% to assess activity prior to treatment withdrawal or for pregnancy planning. Of the 487 biopsies with active disease & treatment data so far assessed: 46% of patients achieved CR (median time 5 mths) with 26% achieving PR (median time 7 mths); 63% of patients were treated with Rituximab alongside oral treatment, most frequently myco-phenolate mofetil, 11% IV cyclophosphamide, 9% Rituximab & cyclophosphamide & 17% oral treatments alone. Multiple logistic regression identified baseline factors predictive of response at 1 year: proteinuria at biopsy (odds ratio (OR) CR 0.64 (95% CI 0.49-0.82) for increase between each non-linear, clinically relevant range (0-50,50-100,100-300,300-500,500-1000,> 1000). For the first presentation ofLN, OR of CR was elevated at 3.22 (95% CI 1.15-8.99). Factors at 1 yr predictive of 2-46 yr survival with eGFR >60mls/min/1.73m2: ROC curves identified proteinuria of <76mg/mmol (AUC 0.713, p<0.0001, sensitivity 71%, specificity 65%) & eGFR of >74.5ml/min (AUC 0.8242, p<0.0001, sensitivity 74%, specificity 81%). Conclusion(s): In a large ethnically/racially diverse cohort of predominantly young women with LN, unique for the high proportion receiving rituximab & steroid sparing regimens, we demonstrate excellent long term outcomes. Importantly, this is the largest cohort to date that validates the use of a less stringent proteinuria cut off (using uPCR) of 76mg/mmol at 1 yr after treatment to define patients likely to have preserved renal function in the long term.