Premium
A78: Urine Biomarkers Role in Predicting the Future Development of Renal Functional Loss With Lupus Nephritis in Children and Adults
Author(s) -
Abulaban Khalid,
Brunner Hermine,
Nelson Shannen L.,
Bennett Michael,
Ying Jun,
Song Huijuan,
Kimmel Paul,
Kusek John,
Feldman Harold,
Ramachandran Vasan,
Rovin Brad H.
Publication year - 2014
Publication title -
arthritis and rheumatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 4.106
H-Index - 314
eISSN - 2326-5205
pISSN - 2326-5191
DOI - 10.1002/art.38494
Subject(s) - lupus nephritis , medicine , urine , nephritis , systemic lupus erythematosus , disease
Background/Purpose: Lupus nephritis (LN) is frequently associated with a poor long‐term prognosis. Current non‐invasive blood and urine tests do not reliably predict the course of LN. The objective of this study was to evaluate the performance of candidate urine biomarkers in predicting future kidney function in adults and children with LN. The biomarker candidates studies were liver‐type fatty acid binding protein (L‐FABP), albumin (Alb), monocyte chemoattractant protein 1 (MCP‐1), Uromodulin, Transferrin and Hepcidin. Methods: L‐FABP, Alb, MCP‐1, Uromodulin, Transferrin and Hepcidin were measured by ELISA in urine from 70 adults and 42 children collected at the time of enrollment into prospective observational LN cohorts. Urine analytes were normalized to urine creatinine and logarithmically transformed. The association of each analyte to renal function loss (RFL), defined as a sustained increase of ≥25% in serum creatinine (SCr; adults) or a decrease in eGFR of ≥20% (children), was determined using a fixed effect model after adjusting for the age group (adult vs. child). In addition, the results were confirmed using Wilcoxon Rank Sum tests. Logistical models were used to predict the presence of RFL using each biomarker or a combination of the biomarkers. Biomarker performance in predicting RFL was assessed as the area under a ROC curve (AUC) corresponding to the logistical model. Results: 13 children and 22 adults had RFL during the mean follow‐up period of 6.1 months and 60 months, respectively. Overall patients with RFL showed significantly higher levels ALB than those without RFL (p < 0.05, ). In addition, the levels of L‐FABP, MCP‐1, and Transferrin were also marginally higher in RFL (p‐values < 0.1). The AUC using the combination of urine L‐FABP, Alb, MCP‐1 and Utransferrin was 0.66, slightly higher than those using any single biomarker as the predictor (ranging from 0.52–0.63).Patient type Biomarker/Cr$ Renal function loss$ Preserved renal function$ p (Fixed effect model/Wilcoxon)N35 77 –LFABP 1.75 ± 1.58 1.17 ± 1.62 0.084 / 0.066Albumin 5.58 ± 2.31 4.46 ± 2.19 0.017 / 0.030MCP‐1 5.77 ± 1.63 4.89 ± 2.45 0.064 / 0.034UROMODULIN 2.04 ± 1.30 2.14 ± 1.14 0.683 / 0.644UTRANSFERRIN 2.27 ± 2.27 1.49 ± 1.98 0.073 / 0.070All patients with LNHEPCIDIN 3.85 ± 1.15 3.57 ± 1.67 0.378 / 0.656N 22 48 –LFABP 1.69 ± 1.26 1.28 ± 1.69 0.326 / 0.229Albumin 5.90 ± 2.14 5.01 ± 2.01 0.100 / 0.147MCP‐1 5.98 ± 1.16 4.80 ± 2.70 0.061 / 0.057UROMODULIN 1.33 ± 1.06 1.65 ± 1.08 0.261 / 0.274UTRANSFERRIN 1.98 ± 1.78 1.41 ± 1.96 0.255 / 0.177Adults with LNHEPCIDIN 3.80 ± 0.95 3.28 ± 1.71 0.197 / 0.368N 13 29 –LFABP 1.84 ± 2.05 0.98 ± 1.50 0.136 / 0.218Albumin 5.05 ± 2.57 3.55 ± 2.21 0.060 / 0.116MCP‐1 5.38 ± 2.30 5.02 ± 2.02 0.634 / 0.402UROMODULIN 3.19 ± 0.68 2.96 ± 0.69 0.314 / 0.589UTRANSFERRIN 2.73 ± 2.94 1.64 ± 2.04 0.170 / 0.199Children with LNHEPCIDIN 3.91 ± 1.45 4.02 ± 1.51 0.830 / 0.803Conclusion: Urine biomarkers L‐FABP, Alb and MCP‐1 are likely predictive to RFL. Other biomarkers such as Uromodulin, Transferrin and Hepcidin are markers of disease activity, but not predictive of RFL.