Perfusate Cell-Free DNA (cfDNA) Predicts Lung Utilization during Clinical Ex Vivo Lung Perfusion

Abstract

Ex vivo lung perfusion (EVLP) has led to an increase in the number of donor lungs transplanted. However, current methods of assessment are largely subjective and subject to variability by clinician interpretation. We therefore sought a more accurate, robust and quantitative method for EVLP lung assessment. Given that cell-free DNA (cfDNA) is released by injured and dying cells, we hypothesized that the quantity of cfDNA in EVLP perfusate could be used as such a method for lung assessment. EVLP perfusate samples from 36 clinical EVLP cases lungs (19 transplanted, 17 rejected) were collected at 1h and 4h of perfusion, centrifuged to remove cell debris, and then snap frozen and stored at -80C. Samples were subsequently thawed and cfDNA isolated using the Qiagen QIAmp MinElute ccfDNA kit. The fragment size distribution of the samples and their concentration was measured using the Agilent Fragment Analyzer. We successfully isolated 200-1000 ng of cfDNA per ml of perfusate from each sample. The cfDNA had the peak DNA fragment sizes we would expect from serum cell free DNA, i.e. multiples of ∼167bp. Peaks of up to 1,000 bp in size were identified, suggesting that EVLP perfusate medium is excellent for the preservation of cfDNA. Perfusate cfDNA concentration in utilized lungs vs. rejected lungs were 627+287 vs. 1079+573 ng/ml, respectively (p=0.01). ROC curve analysis suggests a potential cutoff value for transplant utilization of 663 ng/ml (Fig 1. Scatter plot of DNA amounts isolated from perfusate samples separated by utilized/declined). We have demonstrated that it is possible to isolate perfusate cfDNA from human donor lungs undergoing EVLP and that the quantity of cfDNA is predictive of donor lung utilization. Further validation and refinement of this assay with separate cohorts are underway.