Urinary osteoprotegerin: a potential biomarker of lupus nephritis disease activity

Abstract

Objectives: Urinary biomarkers may help in identification, treatment and assessment of response in patients with lupus nephritis (LN). Osteoprotegerin (OPG) is produced by the kidneys and lymphoid cells and may reflect renal disease activity better. The data on its utility are sparse. Methods: Fifty-eight patients with active LN (AN), 24 with active non-renal disease (ANR) and 39 with inactive disease (ID) were included. Median disease duration was 32 (1-204) months and median age was 27 (12-50) years. AN patients were followed up every three months for one year. Urine and serum samples were collected for OPG measurement by ELISA (pg/ml) and urinary values were normalised for creatinine excretion (pg/mg). Urine samples from 24 healthy individuals (HCs) and 20 patients each of rheumatoid arthritis (RA) and diabetic nephropathy (DM) served as controls. Variables were expressed as median (range). Results: At baseline, normalised urinary OPG (uOPG) was significantly higher (p < 0.001) in AN (1229 (0-8577)) than ANR (236 (0-14713)), ID (463 (7-4253)), HCs (366 (120-2849)) and DM (350 (127-1577)) but it was not different from RA (1511 (122-8849)). uOPG correlated modestly with rSLEDAI (r = 0.4, p < 0.001) and SLEDAI (r = 0.31, p < 0.001) but not with serum OPG (sOPG). uOPG but not sOPG could differentiate between AN and ANR groups. In the longitudinal study, uOPG and sOPG decreased significantly with treatment at all follow-up visits but the trend of fall in sOPG was erratic. uOPG values at baseline, 3, 6, 9 and 12 months were 1229 (0-8577), 466 (3-4874), 104 (0-1598), 325 (0-4025) and 555 (6-6771) pg/mg, respectively. uOPG but not sOPG rose before conventional markers in three patients who had a relapse of LN. In two patients who developed chronic kidney disease, uOPG remained persistently high. For differentiating AN from ANR patients, uOPG performed the best on receiver operator characteristics analysis (AUC = 0.72) when compared with anti-dsDNA antibodies, C3, C4 and sOPG. Conclusion: uOPG is derived from kidneys and helps differentiate active SLE patients with and without LN. It shows modest correlation with disease activity and has a potential to predict poor response to therapy and relapse of LN.