Background Our preliminary proteomic analysis revealed elevated levels of urinary angiostatin, CXCL4 and VCAM-1 in patients with active lupus nephritis
Objectives To study the performance of urinary angiostatin, CXCL4 and VCAM-1 in differentiating between active renal and non-renal SLE.
Methods Patients who fulfilled the ACR classification for SLE with active renal disease, active non-renal disease and inactive disease were randomly recruited, together with a group of healthy controls. Stored urine samples of the participants were assayed for angiostatin, CXCL4 and VCAM-1 and compared among the different groups. ROC analysis was performed to obtain the best cut-off values to calculate the performance of these markers in differentiating among the different groups of patients. SLE disease activity was assessed by the SELENA-SLEDAI and physician's global assessment (PGA). Specificity and sensitivity of these markers in differentiating between active renal and non-renal SLE was determined by 2x2 contingency tables. In patients with active renal disease, correlation between these urinary biomarkers with clinical renal parameters was also performed.
Results 227 SLE patients (80 inactive SLE; 67 active non-renal disease; 80 active renal disease; 94% women, age 39.2±13.8 years) and 53 controls were studied. Urinary angiostatin, CXCL4 and VCAM-1 levels normalized for creatinine were significantly higher in patients with active renal than non-renal disease (angiostatin 18.4±27.1 vs 1.6±2.91 pg/ng; p<0.0001; CXCL 9.11±15.7 vs 5.12±10.4 pg/ng; p=0.003; VCAM-1.41±1.31x103 vs 0.72±1.10x103 pg/ng; p<0.0001). The levels of these urinary protein markers were also significantly higher in active SLE patients than inactive SLE patients or healthy controls. Urinary angiostatin, CXCL4 and VCAM-1 correlated significantly with the renal SLEDAI (Rho 0.66, 0.45 and 0.51, respectively; p<0.001 in all), total SLEDAI score (Rho 0.60, 0.46 and 0.53, respectively; p<0.001 in all) and urine protein-to-creatinine (uPCr) ratio (Rho 0.73, 0.51, 0.59, respectively; p<0.001) in all the SLE patients studied. Urinary angiostatin was more specific (specificity 0.82) than elevated anti-dsDNA and low C3 (specificity 0.64 and 0.66) in differentiating active renal SLE from non-renal SLE. In a subset of patients with biopsy proven active lupus nephritis (N=68), these urinary protein markers could not differentiate between proliferative (III/IV) from non-proliferative (I/II/V) types of lupus nephritis. However, urinary CXCL4 (Rho 0.25; p=0.049) and VCAM-1 (Rho -0.28; p=0.02), but not angiostatin (Rho 0.11; p=0.39), correlated significantly with the histologic activity index. There was no significant association between these protein markers and the histologic chronicity index or renal SLEDAI score. On the other hand, urine angiostatin levels (Rho 0.36; p=0.003), but not CXCL4 (Rho 0.07; p=0.59) or VCAM-1 (Rho -0.11; p=0.36), correlated significantly with the uP/Cr ratio in this subgroup of patients.
Conclusions Urinary angiostatin, CXCL4 and VCAM-1 are potentially useful biomarkers for SLE, in particular lupus nephritis. Further longitudinal studies are necessary to delineate the sensitivity and specificity of these two urinary protein markers in predicting renal flares and prognosis in SLE patients.
Disclosure of Interest None declared