Abstract: TH-PO1224
Dd-cfDNA Predicts Long-Term Graft Outcomes: Kidney Allograft Outcomes AlloSure Registry (KOAR) Results
Session Information
- Late-Breaking Research Posters
November 06, 2025 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Transplantation
- 2102 Transplantation: Clinical
Authors
- Wojciechowski, David, The University of Texas Southwestern Medical Center, Dallas, Texas, United States
- Alhamad, Tarek, Washington University in St Louis, St. Louis, Missouri, United States
- Kumar, Dhiren, Virginia Commonwealth University, Richmond, Virginia, United States
- Klein, Jeffrey A., The University of Kansas Medical Center, Kansas City, Kansas, United States
- Chuang, Peale, CareDx Inc, Brisbane, California, United States
- Huang, Edmund, Cedars-Sinai, Los Angeles, California, United States
- Sureshkumar, Kalathil K., Allegheny Health Network, Pittsburgh, Pennsylvania, United States
- Lawrence, Christopher, CareDx Inc, Brisbane, California, United States
- Ponsirenas, Renata, CareDx Inc, Brisbane, California, United States
- Shen, Ling, CareDx Inc, Brisbane, California, United States
- Woodward, Robert, CareDx Inc, Brisbane, California, United States
- Maw, Thin Thin, University of Southern California, Los Angeles, California, United States
- Wiseman, Alexander C., Porter Adventist Hospital Harley E Rice Medical Library, Denver, Colorado, United States
- Brennan, Daniel C., Johns Hopkins Medicine, Baltimore, Maryland, United States
Group or Team Name
- KOAR Principal Investigators.
Background
We evaluated the prognostic value of dd-cfDNA for graft dysfunction and graft loss in the Kidney Allograft Outcomes AlloSure Registry (KOAR NCT03326076).
Methods
1,743 kidney transplant recipients (KTR) from 56 U.S. centers were prospectively enrolled from 2018 to 2020. At the first dd-cfDNA elevation, patients were stratified into the High (dd-cfDNA≥1.0%) or Intermediate groups (0.5%≤dd-cfDNA<1% with ≥61%RCV from prior test). Patients without longitudinal elevations remained in the Low group. We used χ2 test to compare graft dysfunction (eGFR ≤30 mL/min/1.73m2), and applied extended Cox models to estimate graft loss risk, using dd-cfDNA state changes as a time-dependent covariate and mortality as a competing risk.
Results
1,258 evaluable KTR were divided into the Low [809; 64%], Intermediate [216; 17%], and High [233; 19%] groups. At 3 years, 7.5%, 11.4%, and 20% (p<0.001) had graft dysfunction in the Low, Intermediate, and High groups, respectively. The 3-year cumulative incidence of graft loss was 1.4%, 3.9%, and 7.6% in the Low, Intermediate, and High groups (Figure 1). On multivariate analysis, KTR in the Intermediate and High groups had a significantly higher risk of graft loss (Table 1).
Conclusion
Dd-cfDNA elevations were strongly associated with lower graft function over time and independently predicted graft loss. Our findings support dd-cfDNA as a prognostic biomarker and highlight the importance of monitoring both thresholds and relative changes to guide care.
Table 1. Estimated HR for 3-Year Graft Loss*
| Characteristic | HR [95% CI] | P value |
| Low (reference) | ||
| Intermediate | 3.7 [1.4 - 9.9] | 0.009 |
| High | 6.4 [2.8 - 14.4] | <0.001 |
| cPRA>80% (reference) | ||
| cPRA≤80% | 1.5 [1.2 - 1.9] | <0.001 |
| de novo transplant (reference) | ||
| repeat transplant | 1.9 [1.5 - 2.4] | <0.001 |
*Mortality modeled as a competing risk.
Funding
- Commercial Support – CareDx, Inc.