Prospective evaluation of donor-derived cell-free DNA as a potential biomarker for cardiac allograft vasculopathy

Abstract Background Cardiac allograft vasculopathy (CAV) remains a major cause of morbidity and mortality among long-term heart transplant (HT) recipients. There is clearly an unmet need for a noninvasive biomarker of CAV that could obviate the need to perform surveillance coronary angiograms in these patients. Purpose Our aim was to evaluate the performance of Donor-derived Cell Free DNA (dd-cfDNA) as a biomarker of CAV. Methods We prospectively measured dd-cfDNA levels in all consecutive asymptomatic patients undergoing surveillance coronary angiography >1 year after HT at a single center, between Jan 2019 and Jan 2021. Endpoints included the association between dd-cfDNA levels and the presence CAV, according to ISHLT 2010 classification. Patients with history of acute cellular rejection ≥1R or antibody mediated rejection in the previous 6 months were excluded. Results We included 94 HT recipients, median age 57 years (IQR 50–67), 67% men, a median of 10.9 years after transplant. Coronary angiogram revealed CAV0, CAV1, CAV2 and CAV3 in 61%, 19%, 14% and 6% of patients, respectively. Median dd-cfDNA values for each CAV group were: CAV0 0.92% (IQR 0.46–2.0), CAV1 1.4% (0.38–2.8), CAV2 0.17% (0.07–0.52) and CAV3 0.24% (0.057–0.87); p=0.0535. Figure 1 summarizes baseline characteristics of the cohort and results. Comparison of dd-cfDNA levels in patients with CAV0 and CAV1–2-3 did not show significant differences (0.92%, IQR 0.46–2.0 vs 0.46%, IQR 0.075–1.5, p=0.059) (Figure 2A), nor did the comparison between patients with stable CAV (no new coronary lesions since previous angiogram, n=77) and progressive CAV (patients with new coronary stenoses, n=17); median dd-cfDNA values were 0.735% (IQR 0.195–2.0) vs 0.9% (IQR 0.12–1.8), p=0.76 (Figure 2B). A subanalysis according to time after HT was also found non-significant: less than 5 years (p=0.95), 5 to 10 years (p=0.14) and more than 10 years after HT (p=0.16) (Figure 2C). The AUC ROC curve for the diagnosis of CAV revealed the lack of ability to predict the presence of any degree of CAV (AUC ROC = 0.38). Conclusion In our experience, dd-cfDNA did not perform as a useful biomarker to avoid surveillance coronary angiograms for CAV diagnosis. Funding Acknowledgement Type of funding sources: Other. Main funding source(s): Sociedad Madrileña de Trasplantes