TO009THE POTENTIAL OF DONOR-DERIVED CELL-FREE DNA AS A BIOMARKER FOR REJECTION IN KIDNEY TRANSPLANTATION: A SYSTEMATIC REVIEW AND META-ANALYSIS
Abstract Background and Aims Donor-derived cell-free DNA (dd-cfDNA) is actively studied as a non-invasive biomarker for the diagnosis of kidney allograft rejection, with conflicting results being reported. Therefore, we conducted a systematic review and meta-analysis to provide a clear overview of the potential value of dd-cfDNA to diagnose kidney rejection, and specifically antibody-mediated rejection (ABMR). Method A systematic literature search was performed in Pubmed, Web of Science, and the Cochrane library using the search terms: “cell free DNA AND kidney transplantation”. We included studies that 1) described dd-cfDNA fractions in blood of kidney allograft recipients; and 2) were published as English original research papers between January 1, 1994 and November 26, 2019. The methodological quality was assessed by the Methodological Index for Non-Randomized Studies (MINORS). Out of 287 hits, 55 duplicates were removed. 232 remaining abstracts were screened, leaving 36 articles for full-text review, of which 11 were included in a systematic review and data from 7 articles were synthesized in a meta-analysis. A weighted median of medians (WMM) (median with 95% confidence interval (CI) in squared brackets) was computed as a pooled estimate for the median dd-cfDNA fractions in different groups of patients. A meta-analysis was prespecified to be performed among following patient groups: 1) ABMR patients; 2) T-cell-mediated rejection (TCMR) patients; 3) stable patients; 4) patients without rejection at indication biopsy. The pooled ABMR group consisted of patient groups with pure ABMR and/or mixed ABMR/TCMR. Weighted median differences of medians (WMDM) were estimated for pairwise comparisons among the aforementioned groups, according to the approach proposed by McGrath et al. (2019). Results A large heterogeneity was observed between the selected studies in 1) study design; 2) inclusion criteria (consecutively transplanted patients versus patients with an acute kidney event); 3) statistical reporting (mean ± SD versus median with IQR or min-max); and 4) the definition of study groups. This made the results of the selected studies only partially comparable or suitable for further meta-analysis. Three studies calculated the fraction of dd-cfDNA in stable patients, resulting in a WMM of 0.29% [0.26-0.45]. In the ABMR group, a remarkably higher WMM (2.5% [1.4-2.9]) was observed, when combining the data from five separate studies. Four of those five studies also described dd-cfDNA levels in TCMR patients and patients without rejection at indication biopsy, resulting in a WMM of 0.27% [0.26-2.69] and 0.57% [0.3-0.67], respectively. As presented in Figure 1, median dd-cfDNA fractions were significantly higher in patients with ABMR vs. patients without rejection at indication biopsy (Figure 1A). As compared to patients with TCMR, a tendency to higher median dd-cfDNA fractions was observed in patients with ABMR (Figure 1B). Patients with TCMR had no higher median dd-cfDNA fractions than patients without rejection at indication biopsy. Conclusion Despite the large heterogeneity between the selected studies, our results point towards higher median dd-cfDNA fractions in patients with ABMR vs. patients without rejection at indication biopsy. Furthermore, a tendency to lower median dd-cfDNA fractions was noted in patients with TCMR compared to ABMR. dd-cfDNA might be useful in the diagnosis of ABMR if confirmed in further studies.
