Potential Role of Donor-Derived Cell Free DNA as a New Biomarker in Cardiac Allograft Vasculopathy (FreeDNA-CAV)

Purpose There is an unmet need for a noninvasive biomarker for cardiac allograft vasculopathy (CAV), that could obviate the need to perform surveillance coronary angiograms in HT recipients. Donor-derived cell free DNA (ddcf-DNA) has shown a good ability to rule out cellular rejection, but its performance as a biomarker for CAV has not been tested. We sought to evaluate the performance of ddcf-DNA as a biomarker of CAV in a cohort of patients undergoing routine coronariography. Methods We prospectively obtained ddcf-DNA levels in all patients who underwent routine coronary angiography >1 year after HT in our center between Jan and Sept 2019. Patients with concomitant acute rejection (n=5) or multi organ transplant (n=8) were excluded. Ddcf-DNA levels were correlated with CAV degree (ISHLT 2010 classification). Results We included 54 HT recipients (mean age 57 years, 70% men), with a mean of 10.2 years (IQR 4.3-16.9) after transplant. Coronary angiogram showed CAV0 in 59%, CAV1 in 22%, CAV2 in 6% and CAV3 in 13% (41% of patients with any degree of CAV). Median ddcf-DNA values for each CAV group are shown in figure 1A. Median levels did not differ significantly between patients with and without CAV: 0.64% in CAV0 (IQR 0.19-1.26) as compared to 0.525% (IQR 0.12-1.8) in CAV 1, 2 or 3, p=0.94. Area under the ROC curve for CAV diagnosis was 0.51 (Figure 1B). No significant differences were found between patients with "stable CAV" (no new significant coronary lesions since previous angiogram, n=39) and "progressive CAV" (presence of new coronary stenosis, n=15): median values were 0.5% (IQR 0.17-1.1) and 0.87% (IQR 0.12-4.2) respectively, p=0.51. No correlation was found between ddcf-DNA and Troponin I (r=0.03, p=0.86) or NTproBNP levels (r=-0.09, p=0.59). Conclusion In this preliminary experience, ddcf-DNA level did not seem to correlate with CAV, although sample size might be too small to reach definite conclusions. We hope that the ongoing recruiting (n=100) will shed light in this matter.