Early prognosis of the development of renal chronic allograft rejection by gene expression profiling of human protocol biopsies

Background. Chronic allograft rejection (CR) is the major cause of failure of long-term graft survival and is so far irreversible. Early prognosis of CR by molecular markers before overt histologic manifestation would be a valuable aid for the optimization of treatment regimens and the design of clinical CR trials. Oligonucleotide microarray-based approaches have proven to be useful for the diagnosis and prognosis of a variety of diseases and were chosen for the unbiased identification of prognostic biomarkers. Methods. Renal allograft biopsies were taken at month 6 posttransplantation (PT) from two groups who were, at that time, healthy recipients: one group developed CR at month-12 PT, the other group remained healthy. Gene expression profiles from the two groups at month-6 PT biopsies were analyzed to identify differentially expressed genes with prognostic value for CR development at month 12. Results. A set of 10 genes was identified that showed differential expression profiles between the two patient groups and had a complete separation of the 15% to 85% quantile range for each individual gene. This set of genes was sufficient to allow the correct prediction of the occurrence or nonoccurrence of CR in 15 of 17 (88%) patients using cross-validation (occurrence for a patient was predicted on the basis of the other patients' data only). In addition, a correct prediction could be made that a recipient with a normal biopsy 12 months PT developed CR within the following 6 months. Conclusions. Identified expression patterns seem to be highly prognostic of the development of renal CR.