Detection of urinary monocyte chemoattr actant protein-1 and interferon-γ-inducible protein-10 to recognize the disease activity of lupus nephritis / 中华风湿病学杂志

ABSTRACT

Objective In this study,we measured the levels of urinary monocyte chemoattractant (MCP)-1 and interferon-γ-inducible protein (IP-10) and further analyzed their associations with clinical and pathological data in lupus nephritis patients in order to find the non-invasive biomarkers which canpredict disease activity.Methods MCP-1,IP-10,VEGF levels were measured in urine samples from 64 lupus nephritis patients and 20 healthy volunteers.Clinical disease activity was determined by SLEDAI and BILAG scores.The lupus nephritis patients were divided into two groupsactive disease group (SLEDAI scores ≥ 10points,n=36) and non-active group (SLEDAI score＜10 points,n=28).Of all patients enrolled,37 patients had a concomitant kidney biopsy performed at the time of urine collection.The predictive performance of uri-nary MCP-1 and IP-10 for renal flare,the Student's t test,Mann-Whitney U test,Chi-square test,and re-ceiver operating characteristic (ROC) curves were constructed for analysis.Results The urinary MCP-1 and urinary IP-10 levels of the active group was significantly higher than that of the non-active group [MCP-1672.39(318.05,2 554.23)pg/ml vs 152.52,(55.61,330.44)pg/ml,Z=-4.717,P＜0.01; IP-10 (38±19) pg/ml vs (22±16) pg/ml,t=3.576 P＜0.01].The level of urinary MCP-1 was positively correlated with the levels of hematuria and 24 hours protein quan-tification,as well as the scores of SLEDAI and BILAG (rbemahuria=0.570,P=0.000; r24hpro=0.569,P=0.000; rSLEDAI=0.600,P=0.000; rBILAG=0.606,P=0.000),and it was also positively correlated with the scores of cellular crescent,wire loop,and AI (rCC=0.405,P=0.015; rwire loop=0.430,P=0.014; rAI=0.352,P=0.003),while nega-tively correlated with the level of C3 and plasma albumin (rc3=-0.564,P=0.000; ralb=-0.587,P=0.000).It had no correlation with the scores of wire loop and CI (P＞ 0.05).The level of uIP-10 was positively correlated with the protein quantification in 24 hours and the scores of SLEDAI and BILAG (r24hpro=0.305,P=0.018; rSLEDAI=0.334,P=0.009; rSILAG=0.496,P=0.000),while negatively correlated with the level of C4 (rC4=-0.301,P=0.016).The R0C curve of uMCP-1 to predict the activity of SLE showed that its specificity was 75.0％,sensitivity was 83.3％,and the area under the ROC curve was 0.85±0.05.The ROC curve of urinary IP-10 to predict the activity of SLE showed that its specificity was 50.0％,sensitivity was 97.2％,its area under the ROC curve was 0.74±0.06.The ROC curve of urinary MCP-1 to predict renal flare shows that its specificity was 45.5％,its sensitivity was 100％,and the area under the ROC curve was 0.74±0.80.The ROC curve of urinary IP-10 to predict renal flare showed that its specificitywas 36.4％,its sensitivity was 73.3％,and its area under the ROC curve was 0.49 ±0.10.Conclusion Urinary MCP-1 and urinary IP-10 predict renal flare in patients with lupus nephritis.Furthermore,urinary MCP-1 is a more specific and sensitive forecaster of renal flare in patients with a history of lupus nephritis than urinary IP-10.