Background

Subclinical and clinical T cell–mediated rejection (TCMR) has significant prognostic implications in pediatric renal transplantation. The goal of this study was to independently validate urinary CXCL10 as a noninvasive biomarker for detecting acute rejection in children and to extend these findings to subclinical rejection.

Methods

Urines (n= 140) from 51 patients with surveillance or indication biopsies were assayed for urinary CXCL10 using enzyme-linked immunosorbent assay and corrected with urinary creatinine.

Results

Median urinary CXCL10-to-creatinine (Cr) ratio (ng/mmol) was significantly elevated in subclinical TCMR (4.4 [2.6, 25.4], P< 0.001, n= 17); clinical TCMR (24.3 [11.2, 44.8], P< 0.001, n= 9); and antibody-mediated rejection (6.0 [3.3, 13.7], P= 0.002, n= 9) compared to noninflamed histology (1.4 [0.4, 4.2], normal and interstitial fibrosis and tubular atrophy, n= 52), and borderline tubulitis (3.3,[1.3, 4.9], n= 36). Elevated urinary CXCL10: Cr was independently associated with t scores (P< 0.001) and g scores (P= 0.006) on multivariate analysis. The area under receiver operating curve for subclinical and clinical TCMR was 0.81 (P= 0.045) and 0.88 (P= 0.019), respectively. This corresponded to a sensitivity-specificity of 0.59-0.67 and 0.77-0.60 for subclinical and clinical TCMR at cutoffs of 4.82 and 4.72 ng/mmol, respectively.

Conclusion

This study demonstrates that urinary CXCL10: Cr corresponds with microvascular inflammation and is a sensitive and specific biomarker for subclinical and clinical TCMR in children. This may provide a noninvasive monitoring tool for posttransplant immune surveillance for pediatric renal transplant recipients.