Molecular Classification System for Cardiac Allograft Rejection is positively associated with the pathology severity of rejection - 08/01/26
, A. Giarraputo 2, J. Patel 3, M. Fedrigo 4, S. Varnous 1, J.-P. Duong 5, J. Dagobert 6, P. Rouvier 7, P. Leprince 1, P. Achouh 8, X. Jouven 9, P. Bruneval 10, A. Angelini 11, J. Kobashigawa 12, A. Loupy 13Résumé |
Introduction |
Targeted molecular profiling combined with reproducible formalin-fixed paraffin-embedded (FFPE) EMB-based technology has the potential to support cardiac rejection diagnosis.
Objective |
We aimed to develop and validate targeted gene expression diagnostic models of cardiac rejection and show their association with ISHLT pathological grades.
Method |
We built a multicenter cohort of 591 FFPE-EMBs collected from four international centers between 2011 and 2021. Biopsies were graded according to the ISHLT working formulations, including 188 AMR cases, 289 ACR cases, and 114 non-rejection cases, randomly split in a derivation ( n = 475) and a validation cohort ( n = 116). Tissue gene expression was analyzed on FFPE-EMB using the Banff Human Organ Transplant gene set. Molecular classifiers for AMR and ACR were built using a supervised model. Association between molecular scores and pathology severity of rejection were analyzed in both derivation and validation set.
Results |
The derivation included a total of 151 AMR (pAMR1H+: n = 46, pAMR1I+: n = 36, pAMR2-3: n = 69), 232 ACR (ACR 1R: n = 143, ACR 2-3R: n = 89) and 92 non-rejection cases. The validation set included a total of 37 AMR (pAMR1H+: n = 12, pAMR1I+: n = 15, pAMR2-3: n = 10), 57 ACR (ACR 1R: n = 31, ACR 2-3R: n = 26) and 22 non-rejection cases. Median AMR scores in AMR, ACR and non-rejection cases were 0.655 (IQR = 0.316), 0.217 (IQR = 0.254) and 0.140 (IQR = 0.209), respectively. Median ACR scores in ACR, AMR and non-rejection cases were 0.678 (IQR = 0.347), 0.263 (IQR = 0.364) and 0.302 (IQR = 270), respectively. AMR and ACR molecular scores were strongly associated with the pathology assessment of severity of rejection according to AMR and ACR international working formulations, respectively (derivation set: ACR: P for trend = 1.017E-46, AMR: P for trend = 2.911E-52; validation set: ACR: P for trend = 1.827E-13, AMR: P for trend = 5.174E-10, Fig. 1 ). Molecular AMR score was not associated with ACR severity; neither was the ACR molecular score with AMR severity.
Conclusion |
Tissue-based molecular diagnostic system developed closely aligned with histological grading of cardiac allograft rejection, enhancing diagnostic precision, and offering a reliable companion tool for routine practice.
Le texte complet de cet article est disponible en PDF.Plan
Vol 119 - N° 1S
P. S29-S30 - janvier 2026 Retour au numéro
Article précédent
- Pronostic value of initial TIMI flow after ischemic Out-of-Hospital Cardiac Arrest
- E. Lasik, F. Picard, O. Varenne, P. Brami, V. Pham
- Interest in the use of the “MonCœur” Smartphone application for monitoring out patients with heart failure
- E. Berthelot, E. Audureau, D. Logeart, F. Roubille, J.-M. Tartière, T. Damy, F. Bauer
Bienvenue sur EM-consulte, la référence des professionnels de santé.
L’accès au texte intégral de cet article nécessite un abonnement.
Déjà abonné à cette revue ?