Donor-derived Cell-free DNA Evaluation in Pediatric Heart Transplant Recipients: A Single-center 12-mo Experience

Abstract

Background: Endomyocardial biopsy (EMB) is considered the gold-standard method to diagnose rejection after heart transplantation. However, the many disadvantages and potential complications of this test restrict its routine application, particularly in pediatric patients. Donor-derived cell-free DNA (dd-cfDNA), released by the transplanted heart as result of cellular injury, is emerging as a biomarker of tissue damage involved in ischemia/reperfusion injury and posttransplant rejection. In the present study, we systematically evaluated dd-cfDNA levels in pediatric heart transplant patients coming for follow-up visits to our clinic for 12 mo, with the aim of determining whether dd-cfDNA monitoring could be efficiently applied and integrated into the posttransplant management of rejection in pediatric recipients.

Methods: Twenty-nine patients were enrolled, and cfDNA was obtained from 158 blood samples collected during posttransplant follow-up. dd-cfDNA% was determined with a droplet-digital polymerase chain reaction assay. EMB scores, donor-specific antibody measurements, and distress marker quantification were correlated with dd-cfDNA, together with echocardiogram information.

Results: The percentage of dd-cfDNA increased when EMBs scored positive for rejection (P = 0.0002) and donor-specific antibodies were present (P = 0.0010). N-terminal pro-B-type natriuretic peptide and high-sensitive troponin I elevation were significantly associated with dd-cfDNA release (P = 0.02 and P < 0.0001, respectively), as were reduced isovolumetric relaxation time (P = 0.0031), signs of heart failure (P = 0.0018), and treatment for rejection (P = 0.0017). By determining a positive threshold for rejection at 0.55%, the test had a negative predictive value maximized at 100%.

Conclusions: Collectively, results indicate that dd-cfDNA monitoring has a high negative prognostic value, suggesting that in heart transplanted children with dd-cfDNA levels of <0.55% threshold, protocol EMBs may be postponed.

Graphical abstract

FIGURE 1.

Distribution of dd-cfDNA levels based on patient (A), donor weight (B), and transplant survival (C). The correlation was performed using the nonparametric Spearman test. P values are 0.54, 0.39, and 0.34, respectively. dd-cfDNA, donor-derived cell-free DNA.

FIGURE 2.

Correlation between dd-cfDNA values and clinical parameters of graft status. Dd-cfDNA percentages are plotted in each graph based on (A) EMB rejection grade, (B) presence of serum DSAs, (C) NT-proBNP levels, (D) hs-troponin I blood levels, (E) IVRT, (F) ECG voltage, (G) clinical signs of heart failure, and (H) for-cause treatment. The number of samples in each category is indicated below the graphs. EMBs that scored positive for AMR are highlighted in red in graph A. EMB, endomyocardial biopsy, dd-cfDNA, donor-derived cell-free DNA; DSA, donor-specific antibody; HF, heart failure; hs-troponin I, high-sensitive troponin I; IVRT, isovolumic relaxation time; NT-proBNP, N-terminal pro-brain natriuretic peptide.

FIGURE 3.

ROC curves of dd-cfDNA related to clinical and blood parameters of allograft injury. ROC analysis of rejection markers and clinical parameters of cardiac injury is calculated and compared. AUC and P values are reported in the legend in brackets. AUC, area under the curve; dd-cfDNA, donor-derived cell-free DNA; DSA, donor-specific antibody; EMB, endomyocardial biopsy; hs-troponin I, high-sensitive troponin I; IVRT, isovolumic relaxation time; NT-proBNP, N-terminal pro-brain natriuretic peptide; ROC, receiver operating characteristic.

FIGURE 4.

Variable importance plot of the random forest model showing the most significant variables (clinical features) contributing to rejection prediction. The most significant ones are ordered in descending order (from top to bottom) by a mean decrease in the Gini coefficient. The top variables contribute more to the model than the bottom ones. dd-cfDNA, donor-derived cell-free DNA; DSA, donor-specific antibody; hs-troponin I, high-sensitive troponin I; IVRT, isovolumic relaxation time; NT-proBNP, N-terminal pro-brain natriuretic peptide.

FIGURE 5.

Dd-cfDNA trend in patients with clinical signs of rejection or with healthy allograft. The percentage of dd-cfDNA is reported for each measurement performed. Samples were collected monthly during routine posttransplant follow-up. Clinically and biopsy-proven heart failure and rejection are highlighted in orange and red, respectively. EMBs that scored negative are indicated by a black arrow. A, Patient 1 (ID 13) experienced heart failure with high NT-proBNP levels and coronary lesions as a consequence of chronic allograft rejection, then stabilized after treatment with a decrease in dd-cfDNA%. B, Patient 2 (ID 19) had a mixed ACR/AMR rejection in September 2022 and an AMR rejection in January 2023, heavily treated. DSAs were present during all follow-up time. NT-proBNP and hs-troponin I levels were above the reference values. Cardiac catheterization revealed high filling pressure and, at the last follow-up, coronary allograft rejection, and heart failure. C, In contrast, dd-cfDNA values of patients (N = 3; ID 1, 14, 24) with no signs of rejection were all <0.55% cutoff. ACR, acute cellular rejection; AMR, antibody-mediated rejection; dd-cfDNA%, donor-derived cell-free DNA percentage; DSA, donor-specific antibody; EMB, endomyocardial biopsy; FU, follow-up; hs-troponin I, high-sensitive troponin I; NT-proBNP, N-terminal pro-brain natriuretic peptide.