RESUMEN
BACKGROUND: Urinary CXCL10 (uCXCL10) is associated with graft inflammation and graft survival, but the factors related to its excretion are not well known. HLA molecular matching at epitope level allow estimating the "dissimilarity" between donor and recipient HLA more precisely, being better related to further transplant outcomes. The relationship between uCXCL10 and HLA molecular mismatch has not been previously explored. METHODS: HLA class I and class II typing of some 65 recipients and their donors was retrospectively performed by high resolution sequence-specific-primer (Life Technologies, Brown Deer, WI). The HLA-Matchmaker 3.1 software was used to assess eplet matching. Urine samples collected on the day of the 1-year surveillance biopsy were available of these 65 patients. uCXCL10 was measured using a commercial enzyme-linked immunoassay kit. RESULTS: 1-year uCXCL10 was independently associated with HLA-DQB1 eplet mismatch load (ß 0.300, 95%CI 0.010-0.058, p = 0.006). Kidney transplant recipients with a HLA-DQB1 eplet mismatch load >3 showed higher values of uCXCL10 at 1-year (p = 0.018) than those with ≤3. Patients with a HLA-DQB1 eplet mismatch load >3 with subclinical AbMR had significantly higher levels of the logarithm of 1-year uCXCL10 (No AbMR 0.88, IQR 0.37; AbMR 1.38, IQR 0.34, p = 0.002) than those without AbMR. CONCLUSIONS: uCXCL10 specifically relates to HLA-DQ eplet mismatch load. This relationship can partly explain the previously reported association between uCXCL10 excretion and graft inflammation. An adequate evaluation of any potential non-invasive biomarker, such as uCXCL10, must take into account the HLA molecular mismatch.