Abstract 4371583: Donor-Derived Cell-Free DNA Stratifies Risk of Mortality and Graft Dysfunction in Severe Acute Cardiac Allograft Rejection

Z Zaid Safiullah (National Heart Lung and Blood Institute, Bethesda, Maryland, United States) H Han Su H Hyesik Kong (NHLBI NIH, Bethesda, Maryland, United States) M Moon Jang (NIH, Bethesda, Maryland, United States) P Palak Shah (Inova Schar Heart and Vascular, Falls Church, Virginia, United States) G Gerald Berry (Stanford University, Palo Alto, California, United States) H Hannah Valantine (Stanford University, Stanford, California, United States) X Xin Tian (Wuya College of Innovation) S Sean Agbor-Enoh (National Heart, Lung, and Blood Ins, Bethesda, Maryland, United States)

Abstract

Introduction: Acute rejection (AR) is a risk factor for mortality following heart transplant. Prior studies relied on endomyocardial biopsy (EMB), an invasive gold standard with poor sensitivity, and excluded EMB-negative antibody mediated rejection. They also did not evaluate risk in relation to sustained LV dysfunction, a reversible precursor to death. Plasma donor-derived cell-free DNA (%dd-cfDNA) is sensitive and can detect AR earlier than EMB. Its prognostic utility is unknown. This study assesses the association of AR and %ddcfDNA at diagnosis of AR with risk of adverse outcomes. Methods: In the prospective multicenter GRAfT study (NCT02423070), heart transplant recipients were enrolled and serial plasma samples were collected to quantify %dd-cfDNA via shotgun sequencing. Acute cellular rejection (ACR) was defined as grade ≥2 and AMR as grade ≥1 on EMB. EMB-negative AMR was defined as donor specifici antibody (DSA) positivity with LV dysfunction. AR was classified as mild (ACR 2 or AMR 1) or severe (ACR ≥3, AMR ≥2, or DSA+/LV dysfunction). The primary outcome was a composite of sustained LVEF <50% for ≥3 months and/or death. Cox regression models assessed the association between AR, %dd-cfDNA at diagnosis, and clinical outcomes. Patients were further stratified by a validated %dd-cfDNA threshold of 0.25%. Results: Among 277 patients, AR occurred in 26.9%: 16.2% with ACR, 9.4% with pathologic AMR, and 6.6% with DSA+LV dysfunction. Severe AR occurred in 4.4%, and mild AR in 18.8%. At 5 years post-transplant, the primary outcome occurred in 25.2%, and overall survival was 80.4%. In multivariable analysis, severe—but not mild—AR was associated with increased risk. Among those with severe AR, %dd-cfDNA > 0.25% at diagnosis was associated with higher risk of the primary outcome (adjusted HR 6.06 [95% CI,1.78–20.6]; p < 0.005) and death (HR 10.3 [95% CI, 2.85–37.3]). %dd-cfDNA levels remained persistently elevated after AR treatment in patients who experienced adverse outcomes. Conclusion: We demonstrate the severe AR is associated with risk of poor outcomes, using contemporary definitions. The %dd-cfDNA levels at diagnosis of severe AR offers novel prognostic utility, identifying patients at high risk for death or graft dysfunction. The persistently elevated %dd-cfDNA levels following treatment could indicate poor response to therapy and need for intensified or prolonged therapy.

Article Details

Journal Circulation
Volume / Issue Vol. 152, Issue Suppl_3
Published November 04, 2025
ISSN 0009-7322
Publisher Lippincott Williams & Wilkins

Journal Info

Circulation

Lippincott Williams & Wilkins

ISSN: 0009-7322 Health Sciences

Authors (9)

Z

Zaid Safiullah

National Heart Lung and Blood Institute, Bethesda, Maryland, United States

H

Han Su

H

Hyesik Kong

NHLBI NIH, Bethesda, Maryland, United States

M

Moon Jang

NIH, Bethesda, Maryland, United States

P

Palak Shah

Inova Schar Heart and Vascular, Falls Church, Virginia, United States

G

Gerald Berry

Stanford University, Palo Alto, California, United States

H

Hannah Valantine

Stanford University, Stanford, California, United States

X

Xin Tian

Wuya College of Innovation

S

Sean Agbor-Enoh

National Heart, Lung, and Blood Ins, Bethesda, Maryland, United States