Background: The transplant of kidneys with high KDPI, and the incidence and duration of delayed graft function (DGF) are risk factors associated with incomplete recovery of kidney function (IRF). In both situations, kidney biopsies are essential to exclude the diagnosis of acute rejection (AR).

Purpose: This study evaluates the clinical utility of monitoring donor-derived cell-free DNA (dd-cfDNA), a biomarker of kidney injury, in recipients of high Kidney Donor Profile Index (KDPI) transplants during the period of kidney function recovery.

Methods: This is a single center, prospective, cohort study in recipients of high KDPI kidneys who developed delayed graft function (DGF), defined as the need for dialysis within the first week after transplantation. All patients received a single 3 mg/kg anti-thymocyte globulin dose, tacrolimus, mycophenolate, and prednisone. Blood samples were obtained at 14 (D14) and 30 (D30) days to measure the percentage (%) of dd-cfDNA. The dd-cfDNA > 1% is associated with increased probability for acute rejection (AR).

Results: This preliminary analysis includes data from 69 patients. The median KDPI was 73% [IQR 47-87] and the mean cold ischemia time was 27± 7 hours. The incidence of DGF was 89.8% with a median duration of 6 (IQR 2 – 9) days. At D14 the median GFR was 12 ml/min/1.73 m2 (IQR 6-21) and the median dd-cfDNA levels were 0.43% (IQR 0.02-0.78), with 19 patients (28%) showing dd-cfDNA >1%. Surveillance biopsies during DGF were performed in 42 (61%) patients showing one T-cell mediated rejection IA (dd-cfDNA of 0.41%) and 4 borderline changes (dd-cfDNA 0.82%, 0.85%, 0.98%, 1.90%). Using dd-cfDNA >1% to guide the indication of the surveillance biopsy, we would have avoided 22 (52%) biopsies but missed 1 (2%) patient with AR. At D30, the median GFR was 36 ml/min/1.73 m2 (IQR 28-48) and the median dd-cfDNA levels were 0.40% (IQR 0.01-0.66), with 12 patients (19%) showing dd-cfDNA >1%. Surveillance biopsies performed in 21 (30.43%) patients with incomplete recovery of kidney function showed 4 borderline changes (dd-cfDNA 0.17%, 0.66%, 0.67%, and 2.80%). Using dd-cfDNA >1% to guide the indication of the surveillance biopsy, we would have avoided 12 (57%) biopsies without missing any AR episodes. Overall, 5 (7%) patients showed dd-cfDNA >1% at D14 and D30, but none of them showed biopsy confirmed AR.

Conclusions: This preliminary analysis suggests that monitoring dd-cfDNA may assist clinical decision during the period of DGF and in kidney transplant recipients with incomplete recovery of graft function.