We analyzed the urine samples of patients with type 2 diabetes at various stages of diabetic nephropathy by lectin microarray to identify a biomarker to predict the progression of diabetic nephropathy. Japanese patients with type 2 diabetes at various stages of nephropathy were enrolled and we performed lectin microarray analyses (n = 17) and measured urinary excretion of fetuin-A (n = 85). The increased signals of urine samples were observed in Siaα2-6Gal/GalNAc-binding lectins (SNA, SSA, TJA-I) during the progression of diabetic nephropathy. We next isolated sialylated glycoproteins by using SSA-lectin affinity chromatography and identified fetuin-A by liquid chromatography-tandem mass spectrometer. Urinary excretion of fetuin-A significantly increased during the progression of albuminuria (A1, 0.40 ± 0.43; A2, 0.60 ± 0.53; A3 1.57 ± 1.13 ng/gCr; p = 7.29 × 10(-8)) and of GFR stages (G1, 0.39 ± 0.39; G2, 0.49 ± 0.45; G3, 1.25 ± 1.18; G4, 1.34 ± 0.80 ng/gCr; p = 3.89 × 10(-4)). Multivariate logistic regression analysis was employed to assess fetuin-A as a risk for diabetic nephropathy with microalbuminuria or GFR<60 mL/min. Fetuin-A is demonstrated as a risk factor for both microalbuminuria and reduction of GFR in diabetic nephropathy with the odds ratio of 4.721 (1.881-11.844) and 3.739 (1.785-7.841), respectively. Collectively, the glycan profiling analysis is useful method to identify the urine biomarkers and fetuin-A is a candidate to predict the progression of diabetic nephropathy.