#2538 Remission and relapses in lupus nephritis among multi-racial Southeast Asians: a single center retrospective study

Abstract Background and Aims Lupus nephritis occurs in up to 60% of all patients with systemic lupus erythematosus (SLE) and has a relapsing/remitting course that can progress to kidney failure. This study aims to assess the treatment outcomes in adult patients treated for lupus nephritis and factors associated with remission among multi-racial Southeast Asians. Method This was a single-center, retrospective cohort study of adults with biopsy-proven lupus nephritis diagnosed between 2018 and 2020 at the Singapore General Hospital. Clinical, laboratory and pharmacotherapy data were collected from electronic health records every 6 months (“visits”) for three years from diagnosis. We excluded visits with incomplete follow-up data for immunosuppressive therapy and remission status. The outcomes were remission and relapses during follow-up. To account for variability across patients and repeated measures over time, mixed effects logistic regression analyses were used to examine factors associated with remission. The mixed effects model included random intercepts for patients and repeated measures for 6-monthly time points, and fixed effects for (1) follow-up time point alone; and (2) follow-up time point, age at diagnosis; gender; presence of diabetes and hypertension at diagnosis; laboratory results at diagnosis for estimated glomerular filtration rate (eGFR), urine protein-to-creatinine ratio, urine red blood cell number, serum complement C3 and C4 levels, presence of FSGS on kidney histology; and treatment with hydroxychloroquine, mycophenolate and multitarget therapy within 6 months before the follow-up visit. The base model included these factors that were determined a priori. The sensitivity analysis added factors that were significantly associated with remission after adjustment for follow-up duration. Results We included 73 patients with 301 visits. Figure 1 shows that remission and complete remission occurred in 63.0% and 39.7% respectively at Month 6. The remission rate increased to 82.9% at Month 24 but were lower during the third year of follow-up. The complete remission rate increased to 75.0% at Month 30 and was lower at Month 36. Nephritic relapses occurred in 1.4% at Month 6, increased to 5%–7% at Month 12–24, and increased to 11% in the third year of follow-up. Table 1 shows the clinical and histological characteristics of the cohort and in the groups with and without remission at follow-up. After adjusting for the follow-up duration, known SLE at kidney biopsy, hyperlipidemia, and acute kidney disease at presentation were associated with reduced likelihood of remission, while higher eGFR was associated with higher likelihood of remission. The mixed effects logistic regression found that follow-up at Month 18 and 24 (OR 3.478 [95% CI: 1.015–11.912] and 8.620 [95% CI: 2.008–36.995] respectively, compared to Month 6), higher eGFR (OR 1.032 [95% CI: 1.012–1.053]), and higher C4 (OR 1.662 [95% CI: 1.011–2.732]) levels were independently associated with remission during follow-up. Higher C3 (OR 0.043 [95% CI: 0.005–0.364]) levels at kidney biopsy and use of multitarget therapy within 6 months before follow-up (OR 0.172 [95% CI: 0.035–0.838]) were inversely associated with remission. The sensitivity analysis (addition of SLE and hyperlipidemia) found that SLE was inversely associated with remission (OR 0.127 [95% CI: 0.037–0.435]); this model had lower correct classification and poorer model fit than the base model. Conclusion This study highlighted that despite increased remission rates, especially during the first two years, approximately 5%–10% may relapse in the 2nd and 3rd year. Factors independently associated with remission were follow-up time point, higher eGFR, higher C4, lower C3 and absence of multi-target therapy. Future studies should aim to further elucidate the relationship between different induction regimens including multitarget therapy and clinical outcomes.