Validation of Urinary CXCL10 As a Marker of Borderline, Subclinical, and Clinical Tubulitis

Background. Renal allograft injury secondary to subclinical and clinical tubulitis remains an important cause of allograft fibrosis and loss despite modern immunosuppression. The goal of this study was to validate the previously reported use of urinary CXCL10 (interferon-gamma-induced protein of 10 kDa) as a noninvasive marker of tubulitis in an independent clinical cohort. Methods. Urine samples (n = 102) from 91 patients with protocol or indication biopsies were assayed for urinary CXCL10 using ELISA. The groups analyzed were as follows: normal histology (n = 22); interstitial fibrosis and tubular atrophy (IFTA) (n = 20); IFTA and borderline tubulitis (n = 13); borderline (n = 13), subclinical (n = 17); and clinical tubulitis (n = 17) without IFTA. Results. The ratio of urinary CXCL10 to creatinine (CXCL10:Cr) was found to distinguish borderline, subclinical and clinical tubulitis from normal histology, and IFTA. The area under the curve receiver operating characteristic curve to distinguish normal versus borderline and subclinical tubulitis was 0.845 (OR 1.407, P = 0.0184); normal versus borderline, subclinical and clinical tubulitis was 0.835 (OR 1.400, P = 0.0127). CXCL10: Cr demonstrated a sensitivity of 73.3% and specificity of 72.7% for normal versus borderline and subclinical tubulitis at a cut-off of 1.97 ng CXCL10/mmol Cr. Conclusion. This study validates urinary CXCL10 as a noninvasive, sensitive, and specific marker for tubulitis in an independent cohort. The straightforward urine processing is accessible to clinical laboratories. We propose that CXCL10 may be useful as a supplementary noninvasive screening test for tubulitis in renal transplant patients, with a level more than 1.97 ng CXCL10/mmol Cr being a threshold to consider biopsy.