Predicting, preventing and individualize treatment of heart transplanted patients

In this project, we aim to address an important clinical problem in heart transplantation, allograft rejection, by develop novel applications to be used in clinical practice to optimize and improve the treatment and monitoring of the heart transplanted patient.

To carry this out, we will use machine-learning techniques together with patient material from a national transplant biobank (including omic and morphologic data) and 8 national quality registries (including clinical treatment and outcome data).

This biobank is unique in the world and includes the whole genome sequences from both donors and recipients.Our main research questions are:Which prognostic gene expression profiles in the recipient and donor genomes, contribute to early transplant damage and progressive transplant failure?What influence have patient characteristics in combination with patients’ genetic profile on post-transplant complications, allograft rejection, late graft failure, and mortality?Can machine learning as a precision tool identify previously unrecognized histological pattern in heart biopsy that correlate with patients’ prognoses and improve morphologic analysis of myocyte damage and allograft rejection?We will use a translational approach to transfer molecular and genetic findings into clinical practice and treatment.

We expect to increase the understanding of allograft rejection and find novel methods for diagnostics and therapy, individualizing the care of heart transplanted patients.