Abstract: PO2399
Donor-Derived Cell-Free DNA Identifies Patients with Antibody-Mediated Rejection and Strongly Correlates Histologically with Microvascular Inflammation
Session Information
- Clinical and Immunologic Predictors of Post-Transplant Outcomes
October 22, 2020 | Location: On-Demand
Abstract Time: 10:00 AM - 12:00 PM
Category: Transplantation
- 1902 Transplantation: Clinical
Authors
- Obrisca, Bogdan, Fundeni Clinical Institute, Bucharest, Romania
- Butiu, Maria, University of Washington, Seattle, Washington, United States
- Bakthavatsalam, Ramasamy, University of Washington, Seattle, Washington, United States
- Smith, Kelly D., University of Washington, Seattle, Washington, United States
- De Castro, Iris C., University of Washington, Seattle, Washington, United States
- Blosser, Christopher D., University of Washington, Seattle, Washington, United States
- Sibulesky, Lena, University of Washington, Seattle, Washington, United States
- Kling, Catherine, University of Washington, Seattle, Washington, United States
- Ismail, Gener, Fundeni Clinical Institute, Bucharest, Romania
- Sorohan, Bogdan Marian, Fundeni Clinical Institute, Bucharest, Romania
- Leca, Nicolae, University of Washington, Seattle, Washington, United States
Background
Accurate and timely detection of rejection is central to improving long-term kidney transplant outcomes. We sought to characterize the association of dd-cfDNA level with rejection status and histological lesions.
Methods
We included all patients (n=54) that underwent a kidney transplant biopsy for suspicion of rejection at our center between 9/17-12/19. Concurrent dd-cfDNA and DSA testing was obtained. Rejection type (Banff 2017) and histological lesions were tested for association with dd-cfDNA level.
Results
18 patients had ABMR (6 mixed ABMR/TCMR), while 12 patients had TCMR alone. Of those with ABMR, 94.4% had a dd-cfDNA level >1%, compared to 16.6% of those with TCMR alone (p<0.001). Of those with a high dd-cfDNA (>1%), 76.2% had both DSAs and ABMR. In multivariate logistic regression analysis, a high dd-cfDNA was a more important predictor of ABMR than a DSA MFI level over 2500 (Fig1). A high dd-cfDNA accurately discriminates ABMR (AUC=0.96; 95%CI, 0.92 to 1.00; p<0.001), with a positive and negative predictive value of 80.9% and 96.9%, respectively. Dd-cfDNA level showed a strong correlation with microvascular inflammation (Fig2), but not with tubulitis or interstitial inflammation.
Conclusion
Our study confirms the high diagnostic accuracy of dd-cfDNA for ABMR, while the strong association with elementary lesions of microvascular inflammation supports the specificity of this test for antibody-mediated allograft injury.
Binary logistic regression analysis regarding variables associated with antibody-mediated rejection
The absolute level of dd-cfDNA and its correlation with rejection type and individual pathological lesions
Funding
- Commercial Support – CareDx