Background/Purpose: Lupus nephritis (LN) is the leading cause of morbidity and mortality in systemic lupus erythematosus (SLE). The objective of this study is to validate urine platelet factor 4 (PF4) and E-selectin as novel biomarkers for disease activity and renal damage in LN. Both markers were selected based on recently completed proteomic screens of urine from LN patients.

Methods: 113 biopsy-proven LN patients (89 active LN and 24 inactive LN), 45 chronic kidney disease (CKD) patients and 41 healthy controls were recruited for enzyme-linked immunosorbent assay (ELISA) testing of urine PF4 and E-selectin levels. Urine biomarker levels were normalized by urine creatinine. Disease activity was assessed by the SELENA-SLEDAI. Renal disease activity was assessed by Systemic Lupus International Collaborating Clinics Renal Activity Score (SLICC-RAS) and R-SLEDAI (Total score of the four kidney-related parameters).

Results: Urinary PF4 and E-selectin levels were significantly increased in active LN patients (PF4 [median (Q1-Q3)]: 0.272 (0.0675-1.432) ng/mg; E-selectin: 30.84 (0-76.32) pg/mg) compared to inactive LN (PF4: 0.0225 (0-0.09325) ng/mg, P<0.001; E-selectin: 0 (0-0) pg/mg, P<0.001), CKD (PF4: 0.062 (0-0.18) ng/mg, P<0.001; E-selectin: 0 (0-36.16) pg/mg, P=0.046), and healthy controls (PF4: 0.005 (0-0.017) ng/mg, P<0.001; E-selectin: 0 (0-0) pg/mg, P<0.001), respectively. Urinary PF4 and E-selectin levels were able to discriminate LN patients from healthy controls (Area Under the Curve (AUC) for PF4: 0.84 (P<0.001); AUC for E-selectin: 0.77 (P<0.001)) as well as active LN from inactive LN (AUC for PF4: 0.79 (P<0.001); AUC for E-selectin: 0.80 (P<0.001)). The sensitivity and specificity of urinary PF4 (sensitivity: 0.75; specificity: 0.78) in detecting LN activity was higher than that of positive anti-dsDNA (sensitivity: 0.38; specificity: 0.67) or decreased C3/C4 (sensitivity: 0.56; specificity: 0.63). Urinary E-selectin (sensitivity: 0.92; specificity: 0.65) was more sensitive than anti-dsDNA or C3/C4, but less specific than anti-dsDNA in detecting LN activity. A significant positive correlation was noted between urine PF4 and E-selectin levels with SLEDAI (PF4: ρ=0.55, P<0.001; E-selectin: ρ=0.30, P=0.001), R-SLEDAI (PF4: ρ=0.55, P<0.001; E-selectin: ρ=0.41, P<0.001), SLICC (PF4: ρ=0.45, P<0.001; E-selectin: ρ=0.59, P<0.001), urine protein-creatinine ratio (PF4: ρ=0.39, P<0.001; E-selectin: ρ=0.58, P<0.001) respectively. PF4 also significantly correlated with serum creatinine levels (ρ=0.21, P=0.03). Importantly, in a urine/biopsy concurrent cohort, urine PF4 was good at discriminating type IV LN from type II/III LN with an AUC of 0.73 (P=0.02). Urinary E-selectin was able to discriminate type II LN from type III/IV LN (AUC: 0.84, P=0.03) and type II/III LN from type IV LN (AUC: 0.80, P=0.002).

Conclusion: Urinary PF4 and E-selectin are potential diagnostic biomarkers of disease activity in lupus nephritis and proliferative renal pathology. The performance of urinary PF4 and E-selectin in monitoring lupus nephritis should be further validated in larger longitudinal cohorts.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/urinary-pf4-and-e-selectin-as-novel-biomarkers-for-disease-activity-and-renal-damage-in-lupus-nephritis/