Send to

Choose Destination
Transplantation. 2017 Oct;101(10):2553-2561. doi: 10.1097/TP.0000000000001662.

Urinary Metabolomics for Noninvasive Detection of Antibody-Mediated Rejection in Children After Kidney Transplantation.

Author information

1 Department of Pediatrics, University of British Columbia, BC Children's Hospital Vancouver, BC, Canada. 2 Department of Pediatrics and Child Health, University of Manitoba, Children's Hospital at Health Sciences Center, Winnipeg, MB, Canada. 3 Department of Pathology, University of Manitoba, Health Sciences Center, Winnipeg, MB, Canada. 4 The Metabolomics Innovation Center, University of Alberta, Edmonton, AB, Canada. 5 Section of Nephrology, Department of Internal Medicine, University of Manitoba, Health Sciences Center, Winnipeg, MB, Canada. 6 Department of Immunology, University of Manitoba, Winnipeg, Manitoba, Canada. 7 Manitoba Center for Proteomics and Systems Biology, Faculty of Medicine, University of Manitoba, Winnipeg, Manitoba, Canada.



Biomarkers are needed that identify patients with antibody-mediated rejection (AMR). The goal of this study was to evaluate the utility of urinary metabolomics for early noninvasive detection of AMR in pediatric kidney transplant recipients.


Urine samples (n = 396) from a prospective, observational cohort of 59 renal transplant patients with surveillance or indication biopsies were assayed for 133 unique metabolites by quantitative mass spectrometry. Samples were classified according to Banff criteria for AMR and partial least squares discriminant analysis was used to identify associated changes in metabolite patterns by creating a composite index based on all 133 metabolites.


Urine samples of patients with (n = 40) and without AMR (n = 278) were analyzed and a classifier for AMR was identified (area under receiver operating characteristic curve = 0.84; 95% confidence interval, 0.77-0.91; P = 0.006). Application of the classifier to "indeterminate" samples (samples that partially fulfilled Banff criteria for AMR; n = 65) yielded an AMR score of 0.19 ± 0.15, intermediate between scores for AMR and No AMR (0.28 ± 0.14 and 0.10 ± 0.13 respectively, P ≤ 0.001). The AMR score was associated with the presence of donor-specific antibodies, biopsy indication, Banff ct, t, ah and cg scores, and retained accuracy when applied to subclinical cases (creatinine, <25% increase from baseline) or had minimal or no transplant glomerulopathy (Banff cg0-1). Exploratory classifiers that segregated samples based on concurrent T cell-mediated rejection (TCMR) identified overlapping metabolite signatures between AMR and TCMR, suggesting similar pathophysiology of tissue injury.


These preliminary findings identify a urine metabolic classifier for AMR. Independent validation is needed to verify its utility for accurate, noninvasive AMR detection.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Wolters Kluwer
Loading ...
Support Center