Bronchoalveolar Lavage Fluid (BALF) as a Compartment for Donor-Derived Cell-Free DNA (dd-cfDNA) Assessment in Lung Recipients

Abstract

<h3>Purpose</h3> Circulating dd-cfDNA has been proposed as a biomarker to discriminate acute lung rejection. It is unknown whether BALF may represent a more sensitive compartment for dd-cfDNA analysis in lung recipients. Additionally, the relative contribution of the recipient or donor to the BALF cfDNA content remains uncertain. We sought to understand the feasibility of distinguishing dd-cfDNA in the BALF and determine the greatest contributor of BALF cfDNA. <h3>Methods</h3> The cohort was drawn from the Clinical Trials in Organ Transplantation-20 study and comprised 126 first adult lung recipients with paired BALF and plasma samples (n=320 each). Total cfDNA was isolated from ∼1mL of sample using automated nucleic acid purification. The % cfDNA contributed by the donor vs. recipient was distinguished in plasma and BALF samples using a next generation sequencing (NGS) assay that measures single nucleotide polymorphisms (SNPs) followed by bioinformatics algorithms. Because these algorithms were developed on plasma, for each BALF sample the SNP typing from the paired plasma sample was used to verify the accuracy of the donor vs. recipient cfDNA contribution in the BALF. <h3>Results</h3> Applying standard QC filters, 168 BALF samples "passed" and generated a dd-cfDNA result while 96 "failed" and did not generate a dd-cfDNA result. 45 samples yielded insufficient total cfDNA to support NGS while 11 generated a potentially invalid dd-cfDNA result. To understand reasons underlying inability to estimate BALF dd-cfDNA, we compared the volume of BALF used for total cfDNA isolation and the total cfDNA yield from the BALF in passed vs. failed samples. The BALF volume was similar between these groups, however the total cfDNA yield was notably less among samples that failed (median total cfDNA 0.38ng/uL in failed vs. 1.42ng/uL in passed samples). Considering only passed BALF samples, the median (25^th, 75^th percentile) fraction of cfDNA contributed by the donor was 6.95% (2.40%, 14.00%), with the remainder contributed by the recipient. <h3>Conclusion</h3> Our data point to challenges in consistently generating dd-cfDNA data from the BALF and suggest the dilute nature of BALF may be a contributing factor. Additionally, these results indicate that the recipient, rather than the donor, is the largest contributor to the BALF cfDNA content potentially reflecting intragraft recipient immune cells.