Molecular mismatch as a marker for immunologic risk in heart transplant recipients - PROJECT SUMMARY The overall success of heart transplantation (HTx) as a lifesaving therapy for patients with end-stage heart failure is attributable to the development and refinement of the current immunosuppressive drug regimens. However, these agents carry well-known risks for malignancies, new onset diabetes, chronic kidney disease, and other conditions that limit quality of life and long-term survival. The ability to titrate immunosuppression (IS) to precisely match an individual recipient’s need would prevent both over-IS with its associated toxicities, and under-IS with its risks of graft rejection. We propose to investigate the role of HLA eplet mismatches (EMM) between donors and recipients as a biomarker for immunologic risk in HTx patients. In kidney transplantation (KT), EMM analysis has been validated as such a biomarker, and established EMM thresholds defining high and low risk categories are being used to guide IS. Since high-resolution HLA genotyping, required to calculate EMM, has not historically been performed, the ideal prospective study of EMM in HTx would require long-term follow-up and thus not yield meaningful results for several decades. To answer the question now, as to whether EMM is a valuable biomarker for immunologic risk in HTx, we propose the use of three patient cohorts which will provide complementary outcomes data with which to model associations between EMM and immunoreactivity. In Aim 1 we will use a national dataset to impute high-resolution HLA genotyping based on the low-resolution HLA typing that is captured in the registry. By leveraging population-specific high-resolution haplotype frequency data and reported ethnicity, we will calculate imputed EMM for a large cohort (N=51,530) of HTx recipients. Test imputations demonstrate good calibration when verified against known high-resolution typing. We will evaluate the association of EMM with rejection, re-transplantation, and mortality captured in the database. Using machine learning we will evaluate the EMM risk thresholds defined in KT patients and explore ways to improve them for HTx recipients. In Aim 2 we will assess the association of EMM with cardiac allograft vasculopathy (CAV), and with biopsy-proven rejection using a protocol HTx biopsy dataset. In Aim 3 we will study a single-center observational cohort to examine the association of EMM with early indicators of immunoreactivity in HTx recipients, including the development of de novo donor specific antibody and the presence of donor-derived cell free DNA. There is an urgent need for a valid biomarker to gauge immunologic risk in HTx recipients. Our use of existing and prospectively collected data from complementary patient cohorts will enable us to evaluate EMM as a biomarker for HLA mismatch driven immunoreactivity. We aim to demonstrate that EMM can be utilized to accurately and risk-stratify patients at the time of HTx. This work will establish a foundation for future clinical trials that can fine-tune IS management, reducing the preventable morbidity and mortality that accompanies both over- and under-IS.
