Subclinical inflammation in protocol biopsies relates to tacrolimus exposure and human leukocyte antigen (HLA) matching. We aimed to characterize transcripts associated with rejection and tacrolimus exposure and the latter's association with transplant outcomes. We tested whether gene expression is associated with rejection using strictly normal protocol biopsies (n = 17) and biopsies with T cell-mediated rejection (TCMR) or antibody-mediated rejection (ABMR) according to Banff criteria (n = 12). Subsequently, we analyzed these transcripts in a set of 4-month protocol biopsies (n = 137) to assess their association with donor and recipient characteristics, the intensity of immunosuppression, and the graft outcome. Differential expression (false discovery rate (FDR) < 0.01, fold (change (FC) > 3) between normal and rejection biopsies yielded a set of 111 genes. In the protocol biopsy cohort (n = 137), 19 out of these 111 genes correlated with tacrolimus trough levels at the time of biopsy (TAC-C0), and unsupervised analysis split this cohort into two clusters. The two clusters differed in donor age and tacrolimus trough levels. Subclinical rejection, including borderline lesions, tended to occur in the same cluster. Logistic regression analysis indicated that TAC-C0 at the time of biopsy (OR: 0.83, 95%CI:0.72-0.06, p = 0.0117) was associated with cluster 2. In a follow-up averaging 70 ± 30 months, this patient group displayed a significant decline in renal function (p = 0.0135). The expression of rejection-associated transcripts in early protocol biopsies is associated with tacrolimus exposure and a faster decline in renal function.
Keywords: biopsies; gene expression; rejection; renal transplantation; tacrolimus.