Performance of Donor-Derived Cell-Free DNA (dd-cfDNA) for Discriminating Acute Lung Allograft Injury in a Multicenter Cohort

Abstract

Purpose Histological patterns of acute allograft injury including perivascular acute rejection (AR), lymphocytic bronchiolitis (LB), organizing pneumonia (OP) and acute lung injury (ALI) are frequent after lung transplant and have been associated with worse outcomes. Circulating dd-cfDNA has been proposed as a biomarker to discriminate acute allograft injury in lung recipients. We sought to determine the performance characteristics of three plasma dd-cfDNA thresholds for distinguishing acute lung allograft injury in a prospective, multicenter cohort. Methods The cohort was drawn from the Clinical Trials in Organ Transplantation-20 study and comprised 126 first adult lung recipients with 320 plasma samples that paired with a lung biopsy performed early within the first posttransplant year. Total cfDNA was isolated from ∼1mL of plasma using automated nucleic acid purification and the dd-cfDNA fraction determined at CareDx using next generation sequencing and validated bioinformatics. Biopsies were binned into typical sampling timeframes of 1, 3, and 6 months posttransplant. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for discriminating between a biopsy with vs. without acute allograft injury (defined as any of AR, LB, OP or ALI) was estimated at each timeframe for each of the following dd-cfDNA thresholds; 0.5%, 0.85%, and 1.0%. Results The Table reports the performance characteristics of each dd-cfDNA threshold at each early posttransplant timeframe. At 1month posttransplant the NPV, indicating the probability of a biopsy without acute allograft injury having a concurrent plasma dd-cfDNA value below the threshold, was poor for all evaluated thresholds. However, by 6months posttransplant the NPV improved, in particular for the 0.5% threshold. Conclusion In this prospective real-world cohort, a dd-cfDNA threshold of 0.5% achieved a high NPV for acute histological allograft injury at 6 months posttransplant. Histological patterns of acute allograft injury including perivascular acute rejection (AR), lymphocytic bronchiolitis (LB), organizing pneumonia (OP) and acute lung injury (ALI) are frequent after lung transplant and have been associated with worse outcomes. Circulating dd-cfDNA has been proposed as a biomarker to discriminate acute allograft injury in lung recipients. We sought to determine the performance characteristics of three plasma dd-cfDNA thresholds for distinguishing acute lung allograft injury in a prospective, multicenter cohort. The cohort was drawn from the Clinical Trials in Organ Transplantation-20 study and comprised 126 first adult lung recipients with 320 plasma samples that paired with a lung biopsy performed early within the first posttransplant year. Total cfDNA was isolated from ∼1mL of plasma using automated nucleic acid purification and the dd-cfDNA fraction determined at CareDx using next generation sequencing and validated bioinformatics. Biopsies were binned into typical sampling timeframes of 1, 3, and 6 months posttransplant. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for discriminating between a biopsy with vs. without acute allograft injury (defined as any of AR, LB, OP or ALI) was estimated at each timeframe for each of the following dd-cfDNA thresholds; 0.5%, 0.85%, and 1.0%. The Table reports the performance characteristics of each dd-cfDNA threshold at each early posttransplant timeframe. At 1month posttransplant the NPV, indicating the probability of a biopsy without acute allograft injury having a concurrent plasma dd-cfDNA value below the threshold, was poor for all evaluated thresholds. However, by 6months posttransplant the NPV improved, in particular for the 0.5% threshold. In this prospective real-world cohort, a dd-cfDNA threshold of 0.5% achieved a high NPV for acute histological allograft injury at 6 months posttransplant.