Cell-Free DNA in the Early Course After Heart Transplantation

Abstract

<h3>Purpose</h3> Cell-free DNA (cfDNA) has been used as a marker of rejection after heart transplantation (HTx). The commonly reported outcome is donor fraction (DF). However, cfDNA from donor (dd-cfDNA) and recipient (rd-cfDNA) can be quantified separately. We aimed to investigate the influence of donor and recipient-derived factors as well as intraoperative measures on levels of cfDNA. <h3>Methods</h3> Blood samples were collected in parallel to the first endomyocardial biopsy in 45 patients (8 children, 37 adults) after HTx. Levels of rd- and dd-cfDNA and DF were determined using digital PCR. Regression analyses were performed with respect to the influence of donor- (age, body surface area (BSA), gender) and recipient-derived factors (age, BSA, gender, kidney function pre-HTx, ventricular assist device (VAD) pre-HTx, previous heart operation) as well as perioperative measures (operation time, ischemic graft time, early rejection, presence of acute kidney injury). <h3>Results</h3> The first biopsy was taken on a mean of 11 days after HTx (median 10, range 5-14, IQR 9-12). Mean ischemic time was 180 min (median 180, range 62-394, IQR 118-215), mean operation time was 376 min (median 333, range 174-745, IQR 239-510). Mean rd-cf-DNA levels were 190,000 copies/μl (median 140,000, range 33,000-1,400,000, IQR 83,000-200,000), mean dd-cfDNA-levels 700 copies/μl (median 400, range 150-7000, IQR 270-620). Mean DF was 0.48 (median 0.33, range 0.09-5.3, IQR 0.2-0.5). In a multiple regression model, ischemic graft time was the only factor with a statistically significant impact on levels of dd-cfDNA, but not on DF. The biggest influence on rd-cfDNA-levels was the use of a VAD pre-HTx, but this was not statistically significant. Strong correlations were seen between age and BSA of donor and recipient. <h3>Conclusion</h3> The separate quantification of both rd- and dd-cfDNA, in contrast to solely reporting DF, allows for the distinction of factors mainly influencing the recipient (such as the pre-HTx use of VAD) and the donor (such as ischemic graft time). In this proof-of-concept study, most of the investigated measures did not reach statistical significance. The correlation of ischemic graft time and dd-cfDNA (but not DF), however, shows the possibility of precisely detecting graft injury