MicroRNAs as non-invasive biomarkers of heart transplant rejection

JP Duong Van Huyen, M Tible, A Gay… - European heart …, 2014 - academic.oup.com
JP Duong Van Huyen, M Tible, A Gay, R Guillemain, O Aubert, S Varnous, F Iserin…
European heart journal, 2014academic.oup.com
Aim Rejection is one of the major causes of late cardiac allograft failure and at present can
only be diagnosed by invasive endomyocardial biopsies. We sought to determine whether
microRNA profiling could serve as a non-invasive biomarker of cardiac allograft rejection.
Methods We included 113 heart transplant recipients from four referral French institutions
(test cohort, n= 60, validation cohort, n= 53). In the test cohort, we compared patients with
acute biopsy-proven allograft rejection (n= 30) to matched control patients without rejection …
Aim
Rejection is one of the major causes of late cardiac allograft failure and at present can only be diagnosed by invasive endomyocardial biopsies. We sought to determine whether microRNA profiling could serve as a non-invasive biomarker of cardiac allograft rejection.
Methods
We included 113 heart transplant recipients from four referral French institutions (test cohort, n = 60, validation cohort, n = 53). In the test cohort, we compared patients with acute biopsy-proven allograft rejection (n = 30) to matched control patients without rejection (n = 30), by assessing microRNAs expression in the heart allograft tissue and patients concomitant serum using RNA extraction and qPCR analysis. Fourteen miRNAs were selected on the basis of their implication in allograft rejection, endothelial activation, and inflammation and tissue specificity.
Results
We identified seven miRNAs that were differentially expressed between normal and rejecting heart allografts: miR-10a, miR-21, miR-31, miR-92a, miR-142-3p miR-155, and miR-451 (P < 0.0001 for all comparisons). Four out of seven miRNAs also showed differential serological expression (miR-10a, miR-31, miR-92a, and miR-155) with strong correlation with their tissular expression. The receiver-operating characteristic analysis showed that these four circulating miRNAs strongly discriminated patients with allograft rejection from patients without rejection: miR-10a (AUC = 0.975), miR-31 (AUC = 0.932), miR-92a (AUC = 0.989), and miR-155 (AUC = 0.998, P < 0.0001 for all comparisons). We confirmed in the external validation set that these four miRNAs highly discriminated patients with rejection from those without. The discrimination capability of the four miRNAs remained significant when stratified by rejection diagnosis (T-cell-mediated rejection or antibody-mediated rejection) and time post-transplant.
Conclusion
This study demonstrates that a differential expression of miRNA occurs in rejecting allograft patients, not only at the tissue level but also in the serum, suggesting their potential relevance as non-invasive biomarkers in heart transplant rejection.
Oxford University Press