Urinary epidermal growth factor is a novel biomarker for early diagnosis of antibody mediated kidney allograft rejection: A urinary proteomics analysis.

ABSTRACT

Although antibody mediated rejection (AMR) accounts for 20-30% of all acute renal allograft rejections, introducing biomarkers for a timely detection of allograft rejection has been remained challenging. This study investigated novel diagnostic biomarkers of AMR by examining of urine proteome in renal transplant patients. Thirty-six patients with kidney transplantation including 22 AMR patients and 14 patients with stable renal function (control group) were enrolled in this study. Urinary samples were collected and Label free quantification (LFQ) proteomics technique was applied on urine samples and data was subjected to Random Forest (RF) algorithm to predict main candidate proteins contributing in AMR. Finally, applicability of candidate diagnostic biomarkers was evaluated in new sets of AMR subjects, stable patients and healthy volunteers. A total of 1020 proteins were detected in urine proteome. RF algorithm predicted 20 differentially expressed proteins with the highest sensitivity and specificity and combination of EGF, COL6A, and NID-1 was identified as possible panel for early diagnosis of AMR. Applicability of EGF as diagnostic biomarker was validated in urine samples of independent set of AMR subjects. This is the first urinary proteomics study in AMR patients demonstrating that urinary EGF might be used as early diagnostic biomarker for AMR. SIGNIFICANCE: Renal antibody mediated rejection (AMR) accounts for 20-30% of all acute rejections of allografted kidneys. Although several possible biomarkers have been proposed to predict AMR, ineffectiveness of current urinary biomarkers in early diagnosing of AMR patients and in distinguishing AMR subjects from patients with stable kidney function casts doubts on their applicability in clinic. Here for the first time and based on the analysis of urinary proteome we showed that uEGF and uEGF/Cr might be candidate biomarkers to predict AMR with high diagnostic power.