Abstract 18175: Predicting Low-grade Rejection After Heart Transplantation In Children Using Whole Genome Sequencing
Introduction: Endomyocardial biopsy (EMB) remains the gold standard for detecting rejection after heart transplantation. EMB is costly, invasive and, particularly, burdensome in children. We aimed to predict low-grade rejection by whole genome sequencing (WGS) in blood sample.
Hypothesis: WGS can predict low-grade rejection in children after heart transplantation.
Methods: A total of 106 blood samples with corresponding EMB from 18 children who underwent heart transplantation from 2011 to 2014 were analysed. There were 18 baseline preoperative samples and 88 samples taken at the time of each EMB. Grade of rejection corresponding to each blood sample was determined (no rejection, n=34, 1R, n=45, and 2R, n=9). An input of 200 ng of total RNA per sample was prepared and indexed separately for Illumina sequencing using the TruSeq RNA Library Prep Kit v2 (Illumina, San Diego, CA). Next generation WGS was performed. Differential expression (DE) analysis was performed using the limma package. A k-nearest neighbor prediction method was used.
Results: The mean age at transplantation was 10.3 years (median 12 years, range: 4 months -17.5 years). Preoperative interpersonal variances in gene expression were negated by post-transplantation immunosuppression. Using the top 3 DE genes (ATP1A1, CD8B, MKS1) in low-grade rejection (0 vs 1R), 68 correct predictions of 79 samples were determined. A scatterplot with the log2-counts-per-million expression values for these 3 genes is shown in Figure 1. Drop-one-out cross validation determined a correct prediction rate of 88% (30/34) for no rejection (Grade 0) and 84% (38/45) for the rejection (Grade 1R). Prediction of grade 2R was not feasible due to small number of samples.
Conclusions: Early grade of allograft rejection after heart transplantation in children can be predicted in blood using genomic approach despite wide age range.
- Transplantation/medical aspects
- Gene expression
- Heart failure, pediatric
- Pediatric cardiology
- Biomedical informatics
Author Disclosures: W.Y. Shi: None. A. Xin: None. G.K. Smyth: None. Y. Hu: None. S. Praporski: None. A. Shipp: None. R.G. Weintraub: None. I.E. Konstantinov: None.
- © 2016 by American Heart Association, Inc.