Detection of Rejection in Kidney Transplant Patients Using an Algorithm That Combines Donor Fraction and Absolute Donor-derived Cell-free DNA

S. Bunnapradist¹, E. Ahmed², M. Maninder², Z. P. Demko², P. R. Billings², H. Tabriziani², P. Gauthier³

¹David Geffen Sch of Med @ UCLA, Los Angeles, CA, ²Natera, Inc., San Carlos, CA, ³Natera, Inc., Los Angeles, CA

Meeting: 2021 American Transplant Congress

Abstract number: LB 76

Keywords: Biopsy, Genomic markers, Kidney transplantation, Rejection

Topic: Clinical Science » Kidney » Kidney Acute Antibody Mediated Rejection

Session Information

Session Name: Kidney Acute Antibody Mediated Rejection

Session Type: Poster Abstract

Session Date & Time: None. Available on demand.

Location: Virtual

*Purpose: Donor-derived cell-free DNA (dd-cfDNA) in the plasma of renal allograft patients is a clinically validated biomarker for allograft injury and rejection. Several dd-cfDNA assays have shown that >1% dd-cfDNA is associated with a high risk for active rejection (AR). Additional studies have shown the advantage of measuring absolute dd-cfDNA concentration to avoid the variability that dd-cfDNA fraction encounters due to the host-derived cfDNA component. Here we present initial results from a new algorithm that combines both dd-cfDNA donor fraction and absolute dd-cfDNA concentration in the plasma and the results were compared with our previous algorithm.

*Methods: For this proof-of-concept study, we collected 241 plasma samples from 226 transplant patients. Our cohorts included both related and unrelated donors. Maintaining the 1% dd-cfDNA fraction cut-off, a set of plasma samples (n=200) were used for training the algorithm to set the optimal threshold for absolute dd-cfDNA cut-off (6.9 AU). Samples that exceeded either the 1% dd-cfDNA fraction or the new absolute dd-cfDNA threshold were considered high risk for rejection. The performance of the updated algorithm was further evaluated on the test cohort (n=41), and the results were compared to the previous algorithm that used the 1% dd-cfDNA fraction threshold alone. Biopsy samples were defined as: a) AR, with TCMR and/or ABMR rejection, and b) clinically stable; five patients had TCMR (2xIA, 2xIB, 1xIIB), one had ABMR and two had mixed rejection.

*Results: The updated algorithm demonstrated improved performance, with an observed sensitivity of 8/8 (100%), as compared to the previous algorithm with a 1% dd-cfDNA threshold, 6/8 (75%), without compromising the specificity (91%; 30/33).

*Conclusions: Host-derived cfDNA can be influenced by a number of physiological and pathological factors, which can therefore affect the reported dd-cfDNA fraction and potentially decrease test accuracy. Thus, an algorithm that incorporates absolute dd-cfDNA concentration to dd-cfDNA fraction is clinically meaningful as it increases sensitivity in detecting rejection in renal allograft patients without affecting the specificity. Future work to further optimize and validate this two-dimensional dd-cfDNA assay is currently ongoing.

To cite this abstract in AMA style:

Bunnapradist S, Ahmed E, Maninder M, Demko ZP, Billings PR, Tabriziani H, Gauthier P. Detection of Rejection in Kidney Transplant Patients Using an Algorithm That Combines Donor Fraction and Absolute Donor-derived Cell-free DNA [abstract]. Am J Transplant. 2021; 21 (suppl 3). https://atcmeetingabstracts.com/abstract/detection-of-rejection-in-kidney-transplant-patients-using-an-algorithm-that-combines-donor-fraction-and-absolute-donor-derived-cell-free-dna/. Accessed May 15, 2025.

« Back to 2021 American Transplant Congress