Graft-derived Cell-free DNA as a Noninvasive Biomarker of Cardiac Allograft Rejection: A Cohort Study on Clinical Validity and Confounding Factors

Background. Circulating graft-derived cell-free DNA (dd-cfDNA) is a new marker of cardiac allograft damage that is used for noninvasive rejection diagnostics. We performed dd-cfDNA (%) in heart transplant recipients during the first posttransplant year. Methods. In 87 patients, serial dd-cfDNA determination at predefined time-points was performed in 770 single samples. dd-cfDNA fraction (%) was measured using an established universal droplet digital polymerase chain reaction method, providing same-day turn-around. Rejection was diagnosed according to clinical parameters and biopsies. Results. Median dd-cfDNA (%) was high (5.36%) immediately after reperfusion and decreased to a median (interquartile range) of 0.10% (0.05%-0.24%) in clinically stable patients by postoperative day 10. Compared to dd-cfDNA (%) samples in clinically stable patients, values were higher (P < 0.001) in biopsy-proven rejection ISHLT 1R (0.42% [0.15%-0.53%]) and 2R rejection (0.84% [0.39%-0.97%]). Moreover, dd-cfDNA (%) was already significantly increased 9-30 days before biopsy-proven rejection (0.36% [0.20%-0.61%]). An as yet unknown finding was a slightly, but significantly (P < 0.0001) higher dd-cfDNA (%) value in samples of stable patients with pericardial effusions (PEs) (n = 94; 0.18% [0.07%-0.30%]) compared to samples of non-PE patients (n = 132; 0.07% [0.04%-0.17%]). Using a cutoff of 0.35%, sensitivity and specificity of dd-cfDNA for cardiac rejection were 0.76 and 0.83 (area under the curve [AUC] ROC-curve: 0.81 [95% confidence interval, 0.73-0.89]). Omitting PE samples from the control group yielded an AUC of 0.86 [95% confidence interval, 0.76-0.95]. Samples drawn <12 hours after endomyocardial biopsy showed high (0.40% [0.15%-1.21%]) dd-cfDNA values, also in ISHLT0R (0.36% [0.10%-0.60%]). Conclusions. dd-cfDNA plasma values were significantly associated with cardiac rejection. However, PE or improper sampling (eg, shortly after biopsy) should be considered as confounders for rejection diagnoses using dd-cfDNA.