Predicting Acute Cardiac Allograft Rejection Using Donor and Recipient Gene Expression
Section snippets
Background
The occurrence of acute allograft rejection in cardiac transplant patients is still a contributing factor to the limited survival of the implanted heart. Currently, there are no biomarkers in clinical use that can predict, at time of transplantation, the likelihood of post-transplantation acute rejection. Therefore, the Biomarkers in Transplantation (BiT) initiative set out to discover biomarkers that could predict, before transplantation, the risk of rejection in order to allow for a more
Methods
As a first step, seven clinical variables measured pre-transplantation were investigated for their power to predict post-transplantation rejection in a larger cohort of patients. A subset of these subjects had recipient whole blood collected pre-transplantation and donor heart tissue collected during transplantation. These samples were analyzed using Affymetrix Human Genome U133 plus 2.0 chips. In the next step, the biomarker discovery was focused on genes from whole blood and biopsy in this
Results
Biomarker panels containing clinical variables or genes from either blood or biopsy or genes from blood and biopsy were identified. The Area Under the Receiver Operating Characteristic Curve (AUC) of the genomic and/or clinical biomarkers ranged between 0.30 and 0.90.
Conclusions
The BiT team discovered novel biomarker panels for predicting, at time of transplantation, which patients will develop acute rejection post-transplantation. Based on this study, a biomarker combining 25 genes from the donor biopsy and 18 genes from the recipient blood provides the best, clinically relevant, prediction power (AUC=0.90). This biomarker, if validated in a separate patient cohort, may help in personalizing immunosuppressive treatment and frequency of acute rejection monitoring.