Assessment of Donor Derived Cell Free DNA (dd-cfDNA) at Surveillance and at Clinical Suspicion of Acute Rejection in Renal Transplantation

Abstract

In our prospective, unicenter cohort study, we collected blood samples from 30 newly kidney transplanted patients, at month 1, 2, 3, and 5 for dd-cfDNA analysis, along with creatinine/eGFR and DSA monitoring, and from 32 patients who underwent an indication biopsy and whose dd-cfDNA levels were measured at the time of biopsy and 1 month afterwards. Fourteen of 32 (43.8%) patients in the biopsy group were diagnosed with TCMR and 5 of 32 (15.6%) with ABMR. Dd-cfDNA proved to be better than creatinine in diagnosing rejection from non-rejection in patients who were biopsied. When a dd-cfDNA threshold of 0.5% was chosen, sensitivity was 73.7% and specificity was 92.3% (AUC: 0.804, 0.646-0.961). In rejection patients, levels of dd-cfDNA prior to biopsy (0.94%, 0.3-2.0) decreased substantially after initiation of treatment with median returning to baseline already at 1 month (0.33%, 0.21-0.51, p = 0.0036). In the surveillance group, high levels of dd-cfDNA (>0.5%) from second month post-transplantation were correlated with non-increasing eGFR 1 year post-transplantation. The study used AlloSeq kit for kidney transplant surveillance for first time and confirmed dd-cfDNA's ability to detect rejection and monitor treatment, as well as to predict worse long-term outcomes regarding eGFR.

Keywords: biomarker; dd-cfDNA; kidney allograft; rejection; transplantation.

FIGURE 1

Box and whisker plot and ROC analysis showing the median donor-derived cell-free DNA (dd-cfDNA) and creatinine levels observed in patients with and without allograft rejection. (A) Box and whisker plot for dd-cfDNA (left) showing a median of 0.24% seen in patients with no rejection and 0.94% in patients with allograft rejection; p = 0.004. Box and whisker plot for creatinine (right) with a median creatinine of 2.15 mg/dL in patients with no rejection versus 2.45 mg/dL in patients with allograft rejection; p = 0.3. (B) The ROC analysis for dd-cfDNA: area under the receiver-operating characteristic curve (AUROC) 0.804. The ROC analysis for creatinine: AUROC 0.609.

FIGURE 2

The dd-cfDNA kinetics with anti-rejection treatment. Total of 15 patients with biopsy and rejection [antibody-mediated rejection (ABMR) or T-cell mediated rejection (TCMR)]. Values shown are at month 0 (time of biopsy and diagnosis), and at 1 and 2 months (after rejection treatment was initiated). For the sake of clarity, four patients with high levels of dd-cfDNA (>2.0%) were excluded from the graph presented. Each diamond represents a biopsy specimen.

FIGURE 3

Two-way repeated-measures ANOVA performed to examine the effect of group (newly transplanted patients were grouped according to the percentage of dd-cfDNA and the number of measurements) and time on the eGFR revealed non-significant main effect of group (p = 0.2235), non-significant main effect of time (p = 0.2008) and non-significant interaction between factors (the effects of group and time on eGFR) (p = 0.0652). In more detail, the analysis determined that the mean value of eGFR has not been significantly different between the groups and the timepoints. A difference of the mean value of eGFR between the timepoints was observed only for those with dd-cfDNA < 0.5% (p = 0.004). The two-way repeated-measures ANOVA analysis with the Greenhouse-Geisser correction (performed to check if the data do not meet the compound symmetry assumption) confirmed the previous estimates.

FIGURE 4

The two-way repeated-measures ANOVA that was run to examine the effect of group (newly transplanted patients and newly transplanted patients who had experienced biopsy) and time on the dd-cfDNA revealed non-significant main effect of group (p = 0.5480), a significant main effect of time [F (1, 36) = 5.72, p = 0.0221] and non-significant interaction between factors (the effects of group and time on dd-cfDNA) (p = 0.3083). In more detail, the analysis determined that the mean value of dd-cfDNA has not been significantly different between the groups, but has been significantly different between the timepoints (month 1 and 2). The difference of the dd-cfDNA between the timepoints was observed mainly for the newly transplanted patients (p = 0.001). The two-way repeated-measures ANOVA analysis with the Greenhouse-Geisser correction confirmed the previous estimates.