Abstract
Background:
Observational studies suggest that cell-free DNA (cfDNA) is a biomarker of tissue injury in a range of conditions including organ transplantation. However, the lack of model systems to study cfDNA and its relevance to tissue injury have limited the advancements in this field. We hypothesized that the predictable course of acute humoral xenograft rejection (AHXR) in organ transplants from genetically engineered donors, provides an ideal system for assessing circulating cfDNA as a marker tissue injury.
Methods:
Heterotopic heart xenotransplantation was performed on baboons (n=7) using genetically-modified pig donor hearts. Plasma cfDNA pre-transplant and at serial timepoints post-transplantation (n=75) were shot-gun sequenced. After alignment of sequence reads to pig and baboon reference sequences, we computed the percentage of xenograft-derived cfDNA – xdcfDNA - (relative to recipient) by counting uniquely aligned pig and baboon sequence reads.
Results:
The xdcfDNA (%) was high early post-transplantation and decayed exponentially to low stable levels (baseline); decay half-life was 3.0 days. Post-transplant baseline xdcfDNA levels were higher for transplant recipients that subsequently developed graft loss compared to the one animal that did not reject the graft (3.2% vs 0.5%). Elevations in xdcfDNA (%) coincided with increased troponin and clinical evidence of rejection. Importantly, elevations in xdcfDNA (%) preceded clinical signs of rejection or increases in troponin levels.
Conclusion:
Cross-species xdcfDNA kinetics in relation to acute rejection are similar to the patterns in human allografts. These observations in a xenotransplantation model support the body of evidence suggesting that circulating cell-free DNA is a marker of tissue injury.
INTRODUCTION
Research spanning over two decades [1] suggests that cell-free DNA (cfDNA) is a potential biomarker of tissue injury in a broad range of clinical settings including cancer [2, 3], trauma [4, 5], sepsis [6], and allograft rejection [1, 7–10]. However, the adoption of cfDNA as a clinical practice, or as a research tool to assess mechanisms of tissue injury have been limited. Of particular relevance to transplantation, is specificity, the ability to differentiate allograft cfDNA from recipient cfDNA. Resolution of this challenge has been made possible with the advent of array and next generation sequencing (NGS) technologies for analyzing donor/recipient genotype differences [7, 11, 12]. However, the high background of recipient cfDNA relative to donor-derived cfDNA still possess a high likelihood of erroneous assignment and may account for low assay sensitivities observed [11]. In allotransplantation, determining whether perturbations of cfDNA originate from donor or recipient is, in part, mitigated by measuring the ratio of donor cfDNA to total cfDNA, thus creating an internal standard [12]. The question of causality remains challenging especially in lung transplant recipients, where elevation in cfDNA may be triggered by co-existing infection and rejection [9], whether cellular (ACR) or antibody-mediated (AMR). Thus, while human transplantation seems an ideal model for studying the application of cfDNA as a biomarker of tissue injury, the ambiguities in differentiating AMR, ACR and infection underscores the need for pre-clinical models to study cfDNA in relation to clearly defined stages of rejection and transplant outcomes.
The time course and immunological features of xenograft rejection are well described [13, 14]. The 1,3-galactosyltransferase gene-knockout (GT-KO) pigs has modulated the time course and severity of rejection, significantly improving survival of transplanted organs from GT-KO pigs into non-human primate (NHP) recipients [13, 14]. Organs from GT-KO pigs are protected from hyperacute rejection and acute humoral xenograft rejection (AHXR) that progresses to graft failure. Even with immunosuppressive drug prophylaxis, these xenografts develop thrombotic microangiopathy that begins within the initial two to three weeks after transplantation, and progress to ischemic failure within two to three months. With the complexities of hyperacute rejection now largely resolved, the current primary goal is long-term xenograft survival. Achieving this goal requires diagnostic tools for early detection of tissue injury (due to AHXR) to prompt initiation of therapy before irreversible damage occurs. Previous studies have demonstrated the use of donor-derived cfDNA as a marker for allograft injury [7, 9, 11, 12], immunosuppressive drug efficacy [15], or long term outcomes [16]. Quantification of xenograft-derived cfDNA (xdcfDNA) may allow us to detect xenograft injury earlier and with greater accuracy than currently available methods – manual palpation, echocardiography and serum troponin assay. We hypothesized that the xenotransplantation model, with its predictable time course to AHXR, provides a unique model to validate cfDNA as a biomarker of tissue damage.
METHODS
Xenotransplantation and Non-Human Primates (NHPs) monitoring Specific-pathogen-free (SPF) Olive baboons (2–3 years of age, 7–15 kg) underwent heterotopic cardiac xenotransplantation procedure using weight-matched donor hearts obtained from 6–8-week-old genetically-modified Yorkshire swine. Genetically modified pigs were generated by Revivicor, Inc. (Blacksburg, VA) and were derived from both male and female porcine sources. Procedures and care were in accordance to ACUC guidelines. SPF baboon were obtained from the University of Oklahoma Primate Center [17]. Pig genetic modifications, xenotransplantation surgical procedures, and NHPs monitoring have been previously described [13] and are specified in Supplemental Materials. All recipient SPF baboons received immunosuppression, as listed in Supplemental Table 1. Post-transplant monitoring for AHXR (palpation, troponin, telemetry, echocardiogram) or infection [17, 18] is presented in Supplemental Materials. Serial plasma samples were collected for assay of xdcfDNA. A summary of the study design is shown (Figure 1).
Figure 1: Study design.
Genetically engineered pig’s heart were transplanted to weight-matched baboons. Recipient baboons were monitored prospectively for xenograft rejection (Supplementary Materials). Plasma samples were collected prospectively prior to clinical examination to assay for xenograft-derived cfDNA. Briefly, cfDNA ((containing pig (pink) and baboon (gray) cfDNA)) was isolated from plasma, used to make cfDNA libraries (12 PCR cycles plus addition of indexes = red bars at the end) prior to NGS. Reads were aligned to baboon or pig reference genomes; baboon and pig specific sequences were counted and used to compute %xdcfDNA. The serial xdcfDNA levels was assessed in relation to xenograft rejection.
Troponin analysis
The troponin assay employed in this study utilized the i-STAT system (Abbott Diagnostics; Lake Bluff, IL), a 2-site enzyme-linked immunosorbent assay (ELISA) using antibodies for human cardiac troponin I. To standardize the human assay for pig troponin, we i) obtained purified pig troponin, prepared several concentrations in triplicates to tested for troponin, ii) obtained peripheral blood from pigs before and during cardiac explantation and tested for troponin, and iii) homogenized pig’s heart tissue in PBS, prepared several dilutions of the extract and tested them for troponin.
XdcfDNA quantification from plasma
Cell-free DNA isolation, library preparation and short-gun sequencing
Cell-free DNA was isolated from plasma (500 uL) on a validated automated platform (QiaSymphony, Qiagen) [7], and quantified by a florescent DNA quantification method (Quant-iT PicoGreen dsDNA Assay Kit, Thrmo Fisher) prior to library construction (12 PCR cycles using Nugen Ultralow Library System, Nugen) on a second automated platform (Epimotion, Eppendorf). Library quality was checked using a high-sensitivity DNA assay (high-sensitivity DNA assay, Agilent 2100 Bioanalyzer), and then shotgun sequencing was performed (HiSeq 2500, 2 × 50 bp, Illumina).
Computational workflow for xdcfDNA quantification
Our computational workflow is built to identify and count specific xenograft- and recipient-derived cfDNA reads. After alignment of cfDNA reads to pig [19] and baboon [20] reference genomes, sequence reads were subjected to a series of analytical steps to remove non-pig and non-baboon reads (e.g. microbial [15]), poor quality reads, read duplications and erroneous reads introduced by library construction and NGS. All steps were achieved using previously described tools [21–23] with modifications detailed in Supplementary Materials. The number of sequence reads aligned to pig or baboon genomes were counted; reads that aligned to both genomes were discarded. The percentage of xenograft reads to the sum of xenograft plus recipient reads was computed as %xdcfDNA. The error rate of the assay was determined as the xdcfDNA in NHP pre-transplant plasma samples.
Assessing the physical characteristics of cell-free DNA
We took advantage of our NGS approach that sequences cfDNA libraries from both ends to deduce the length of the cfDNA fragments. To do this, we identified properly paired cfDNA fragments, identified their ends on the reference genome, and deduced the cfDNA length by counting the base pairs between the ends. To investigate the nucleotide pattern of cell-free DNA and of their ends, we used properly aligned pig or baboon reads and extracted the sequences from the reference fasta file for positions in the +/− 25 base pair interval around the fragment ends. Only first reads were utilized. We then computed the percentage of each base composition (A, T, C or G) at the end and at every position +/− 25 bp around the end. The frequency of each of the four bases at each position was plotted. To verify this analysis, we also extracted the sequences from position 0 to 25 of the properly aligned sequence reads, performed similar analysis and compared the nucleotide composition to the previous analysis. We performed similar length distribution and nucleotide composition analyses using previously analyzed cfDNA sequence data from allotransplant recipients [7, 9, 11, 16].
Data and statistical analysis
To compare technological characteristics of cfDNA analyses, we obtained sequencing and computation data from previously analyzed plasma cfDNA of cardiac allotransplantation recipients [7, 9, 11, 16] and compared the characteristics to the xenograft cfDNA assayed in current study. The xdcfDNA and troponin values showed a skewed distribution, thus non-parametric methods were used to compare rejection from non-rejection time-points. Rejection time-points were defined by a composite of clinical evaluation, echocardiographic assessment of left ventricular pressure, telemetry and troponin T (Tn) measurements [13] (Supplementary Material). The discriminatory effect of the xdcfDNA test for rejection and non-rejection time-points was calculated as a ratio of difference between the means and the standard deviation of the non-rejection controls time-points (Glass’ method [24]). The post-transplant kinetics of xdcfDNA was assessed assuming a one-step and two-step decay kinetics, retaining the kinetic model with the higher R-square value. Baseline xdcfDNA, defined as lowest xdcfDNA level at quiescent state, was computed as the average of the two lowest xdcfDNA levels or as the plateau of the decay kinetic curve attained; both measurements were similar (1.5% vs. 1.5%). To determine xenograft cell turnover rate (number of xenograft cells dying to produce the baseline xdcfDNA level), we used previously described methods [9, 25], assuming baboon total blood volume of 60 ml/Kg, hematocrit of 35%, a median post-transplant cell-free DNA concentration in recipient plasma of 173.5 ng/ml for time-points without rejection, cell-free DNA clearance half-life of 16.3 minutes [25], pig genome ploidy of 2 and size of 2.8 X 109 base pairs [19]. The relationship between xdcfDNA levels and Tn was assessed using a linear regression model assuming the null hypothesis as a line with a slope of 0. Statistical analysis was performed with GraphPad Prism [26].
RESULTS:
Non-human xenograft recipients clinical course
Median post-transplantation follow-up for the 7 NHPs studied was 72.5 days, range 3->200 days (Table 1); 5 NHPs developed xenograft rejection; AHXR, (median time to rejection was 42 days, range 3–62 days); 1 NHP showed no rejection during follow-up; and 1 NHP had only early (2 days) post-transplant follow-up.
Table 1:
xdcfDNA kinetics characteristics
| Non-Human Primate (NHP) | Day 0 Post-Operative xdcfDNA (%) | Baseline xdcfDNA (%) | Rejection event | Xenograft survival (days) | |
|---|---|---|---|---|---|
| Day post-transplant | xdcfDNA (%) | ||||
| NHP 1 | 20.8 | 0.5 | NA | NA | >200 |
| NHP 2 | 26.9 | 2.4 | 42 | 20.3 | 70 |
| NHP 3 | 40.4 | 4.5 | 29 | 20.7 | 98 |
| NHP 4 | -- | 2.3 | 62 | 12.7 | 75 |
| NHP 5 | 13.1 | -- | 3 | 14.2 | 3 |
| NHP 6 | 29.2 | 5.4 | 48 | 28.3 | 48 |
| NHP 7 | 30.8 | -- | -- | -- | -- |
Post-operative xdcfDNA levels were measured on Day 0 after transplant surgery, baseline xdcfDNA=average of xdcfDNA measures after the post-operative xdcfDNA decay, rejection event=first clinical indication of rejection, graft survival = interval in days from transplantation to explantation, NA=not applicable, -- not measured or not attained
Assessment of xenograft troponin
The human troponin assay used in this study detected purified pig’s troponin in a concentration-dependent manner (Supplementary Figure 1a). Pigs blood obtained before cardiac surgery (no cardiac injury) showed low troponin levels, whereas, pig blood samples obtained during cardiac surgery showed high levels of troponin (Supplementary Figure 1b). Further, the assay also detected pig troponin in extracts prepared from homogenized pig heart tissue (Supplementary Figure 1c). After xenotranplantation, elevations of troponin in recipient blood correlate with histopathology evidence of xenograft rejection; but not NHP heart damage (Unpublished results).
XdcfDNA assay characteristics
Average cfDNA sequencing depth was 10 million reads per sample, 68% of the reads were retained and were analyzable for xdcfDNA after eliminating low quality reads, PCR duplicates and non-aligned reads (Supplementary Figure 2a, Table 2). The error rate for assigning reads as xenograft-derived cfDNA was 0.06%. XdcfDNA measures showed a skewed distribution (Supplementary Figure 2b). For allotransplantation, after NGS, 78% of cfDNA sequence reads were retained after removing low quality reads, PCR duplicates and non-aligned reads, but only 2.7% of total reads were analyzable for donor-derived cfDNA (Table 2).
Table 2:
Plasma cfDNA assay characteristics for allo- and xenotransplantation
| Allotransplantation | Xenotransplantation | |||
|---|---|---|---|---|
| Median | IQR | Median | IQR | |
| Total reads (%) | 100 | - | 100 | - |
| High quality read (% of total reads) | 77.9 | 75.9 – 79.2 | 68.1 | 61.3 – 70.8 |
| Analyzable reads (% of total reads) | 2.7 | 2.5 – 3.0 | 67.8 | 61.1 – 70.7 |
| Identifiable graft-derived reads (% of total) | 0.005 | 0.002 – 0.018 | 2.8 | 0.5 – 7.5 |
Total reads = total cfDNA reads obtained from NGS; High quality reads = reads remaining after removing poor quality, non-aligned, and PCR duplication reads; Analyzable reads = informative reads used for computing graft-derived cfDNA. For allotransplantation, these overlap SNPs where the recipient genotype is homozygous and different from donor genotype [7, 11]. For xenotransplantation, these are reads uniquely aligned reads to pig or baboon reference genomes; Identifiable graft reads = percentage of graft-derived reads relative to total cfDNA reads. IQR = interquartile range.
Immediate post-transplant xdcfDNA kinetics
The concentration of plasma cfDNA was 197.5 + 91.8 ng/mL before transplantation, increased to 1892.1 ng/mL + 1184.9 ng/mL immediately after transplant surgery, and then decayed to a low level, corresponding to quiescent state with no rejection (baseline)-Figure 2a. We observed an immediate increase in %xdcfDNA after transplantation (median = 27.5%, range = 12.1–42.4%) followed by an exponential decay to a baseline of 1.5% (half-life 3.0, 95% CI = 1.6 – 6.1 days, Figure 2b). Cell turnover calculations (Supplementary Materials) indicate that the baseline xdcfDNA of 1.5% correspond to xenograft cell-turnover of 298 xenograft cells per second.
Figure 2: Kinetics of plasma cell-free DNA.
(A) Cell-free DNA was extracted from prospectively collected plasma samples and quantified using a florescent DNA quantification Assay. The x-axis intervals were selected based on observed kinetics of cfDNA amount in relation to transplant surgery within individual NHPs. Error bards represent standard deviation. (B) The percentage of xdcfDNA quantified by shotgun sequencing and plotted over days after transplant surgery; only non-rejection time-points were included. A first-order decay model (red solid line) is shown ((R2 =0.62, Y0=26.9, half-life=3.0 days, plateau (baseline)=1.5%))
Temporal course of %xdcfDNA and relationship to xenograft rejection
Table 1 summarizes %xdcfDNA measures and clinical follow-up for individual NHPs. In the four NHPs that developed xenograft rejection and loss beyond the first week of transplantation, mean baseline %xdcfDNA was seven times higher (median = 3.7%, range = 2.4 – 5.4%) than in the one NHP without rejection (0.5%). The NHP (NHP 5) that developed early xenograft rejection showed persistently elevated %xdcfDNA levels (Table 2, Figure 3).
Figure 3: Individual NHP xdcfDNA and troponin trends.
The xdcfDNA (solid black line, left y-axis) or troponin (dashed black line, right y-axis) against days post-transplantation; negative x-axis values represent pre-transplantation; positive x-axis represents post-transplantation. (A)= NHP 1 with no rejection event, (B-F) NHP 2–6 has rejection events, rejection time-points denoted “R”. * in (D) denotes cardiac arrest, (G) NHP 7 with one-post-transplant xdcfDNA assessment due to short follow-up during the study period.
For the 1 NHP that showed no rejection (Figure 3a, NHP 1), early post-operative elevated xdcfDNA was followed by rapid decayed and remained low over the period monitored. However, in the 5 NHPs that showed rejection, %xdcfDNA elevations were coincident with xenograft-rejection events (“R” in Figure 3b-f), and %xdcfDNA remained elevated until allograft loss occurred. In three of the NHPs with rejection (Figure 3b NHP 2, c NHP 3, f NH P 6), %xdcfDNA elevations preceded clinical evidence of rejection (telemetry, echocardiogram, palpation, and troponin levels) by 7, 10 and 27 days, respectively. %xdcfDNA elevations also occurred at an episode of unexplained xenograft cardiac arrest on day 43 in NHP 4 (* in Figure 3d), which then declined rapidly post-cardiac defibrillation, consistent with reversal of xenograft injury. Taken together, %xdcfDNA levels coincident with rejection events (n=11) were higher than non-rejection timepoints (n=27): median = 17.6%, interquartile range (IQR) = 9.7–20.7% vs. 2.0% IQR = 1.0–4.1%, p<0.001, effect size 7.76 (Figure 4a).
Figure 4: The xdcfDNA in relation to xenograft rejection.
(A) Comparing xdcfDNA at rejection time-points (n=11) to non-rejection time-points (n=27), p-value determined by Mann Whitney U test. (B) Linear regressing troponin on xdcfDNA. Regression equation is provided. P-value determined assuming the null hypothesis with a slope of 0.
Correlation of post-transplant %xdcfDNA with serum cardiac troponin (Tn) levels. Serum Tn levels were high immediately after transplant surgery followed by a decay reaching a nadir of 1.5 ng/ml. The Tn decay half-life was shorter than for xdcfDNA (0.8 vs 3.0 days). Tn levels were higher at times of rejection compared to no rejection (6.9 vs. 1.5 ng/ml, p<0.001). Both %xdcfDNA and Tn were elevated in all instances of acute rejection. However, %xdcfDNA rise preceded the rise in Tn and clinical measures of cardiac injury for three NHPs that progressed to graft failure (Figure 3b, c, f, NHPs 2, 3, 6). %xdcfDNA and Tn were linearly correlated (R2 = 0.80, p <0.001, Figure 4b). The x-intercept of the regression line, where Tn is theoretically 0, corresponded to a %xdcfDNA level of 1.9%.
Physical characteristics of xenograft- and recipient-derived cfDNA
The recipient (baboon) derived cfDNA was predominantly mononucleosomes with a peak length of 165 bp, and a periodicity of 10 base pairs, similar to human plasma cell-free DNA. Xenograft-derived cfDNA also showed a predominance of mononucleosomes with a periodicity of 10bp. However, peak fragment length was shorter than baboon cfDNA (144 bp vs. 165, p< 0.001) - Figure 5a. The GC content was higher for xenograft-derived cfDNA fragments compared to baboon cell-free DNA (45.1% vs. 42.0%, p<0.001); GC content of baboon cfDNA was similar to that human cfDNA. Further analysis indicates that the base composition at cell-free DNA fragment ends are not random, and are conserved between xenograft-derived, recipient-derived, plasma cell-free DNA fragments (Figure 5b).
Figure 5: Structure of xenograft-derived cell-free DNA.
(A) Length distribution: end coordinates of properly aligned paired reads were obtained from the baboon (blue), pig (red) or human (gray) reference sequences to deduce the length of the cell-free DNA fragments. The frequency of each length fragment was plotted. Peak length fragment noted. (b) Relative nucleotide composition (y-axis in percentage) at +/−25 bases relative to end (“0”) of properly aligned baboon (blue), pig (red), or human (gray) cell-free reads were computed for each of the 4 nucleotides (A, T, C or G) and plotted over the nucleotide position relative to the cell-free DNA cleavage site designated as 0 (x-axis); negative and positive coordinates denotes 5’ side or 3’ side in relation to the cleavage site
DISCUSSION
Two decades of research suggests that circulating cfDNA is a marker of tissue damage [1, 7, 8, 10, 15]. This technology has the ability to detect early signs of tissue injury before irreversible organ damage and failure ensues. Our analysis indicate that cross-species xenotransplantation is a model to study cfDNA as a marker of acute tissue injury: 1) XdcfDNA post-transplant kinetics are similar to those reported after human heart allotransplantation, all-be-it quantitatively larger. 2) The temporal course of %xdcfDNA is closely correlated with xenograft rejection, similar course is observed in allotransplant rejection. 3) %xdcfDNA is highly correlated with, and precedes rises in serum cardiac troponin (Tn) levels, a widely accepted marker of cardiac injury. 4) Greater precision for identifying cfDNA of graft origin improves the detection of xdcfDNA. The relevance and implications of these findings are discussed.
We observed significant technological assay differences in the xenotransplantation model compared to the allograft transplantation (Table 2). A significant limitation to identifying donor-derived cfDNA in allotransplantation is the high background of recipient cfDNA. Even with NGS or genotype arrays to distinguish donor and recipient reads, less than 0.01% of total sequence reads is assigned as donor-derived [7, 9, 11, 12]. High background relative to donor-derived cfDNA may account for the low sensitivity of cfDNA to detect acute rejection in some allotransplant models [11]. The xenograft model presented here overcomes this limitation owing to the large cross species genomic differences and high xenograft cell turnover rate. Thus, 68% of sequence reads are analyzable; xenograft derived reads constitute about 3% and higher of total sequence reads. The large cross species difference also contributes to a low error rate observed. Thus, the xenotransplantation model is technologically ideal to study changes in donor-derived cfDNA as a biomarker of graft injury.
The current study demonstrated elevations in %xdcfDNA immediately after transplantation, followed by a decline over the ensuing 2–3 days, as previously described in allotransplantation [9, 11] and another model [25]. The established baseline of %xdcfDNA is worth discussion. Compared to the NHP with sustained normal xenograft function, baseline %xdcfDNA was seven times higher in NHPs that subsequently developed clinical rejection and progressive xenograft failure. In lung transplant recipients higher donor-derived cfDNA levels after the initial transplant decay are correlated with increased rates of allograft loss [16]. These data from our xenograft model provide further evidence that cfDNA is a marker of persistent graft injury that partains poor outcomes.
The higher %xdcfDNA observed in our model compared to allotransplantation is an indicator of higher cell turnover. Our time course experiments demonstrate that serial monitoring showed rising %xdcfDNA preceding clinical diagnosis of xenograft rejection. These observations from serial measurements underscore the value of cfDNA for monitoring the onset and progression of graft injury compared to pre-determined threshold approaches. Changes in %xdcfDNA over time proved superior to current methods which rely on physical examination, imaging to assess graft function, and troponin; the human troponin assay used in this study reliably detects pig’s troponin. Xenograft examination by palpation or imaging are low sensitivities and high interobserver variabilities, similar to histopathology [27, 28]. As such, rejection is likely detected late when irreversible graft injury has occurred. Our observations suggest that serial monitoring of donor-derived cfDNA could provide major paradigm shifts in the management of organ transplant recipients.
Human cfDNA is predominantly mononucleosomal with a periodicity of 10 bp and greater GC content than genomic DNA, similar to our findings. Recent studies suggest that cfDNA physical characteristics may be tissue specific [29–31], and vary with disease states [32]. Consistent with these reports, xenograft (or tissue-derived) cfDNA was shorter than baboon cfDNA, which is predominantly derived from hematopoietic cells. Our future studies will address the mechanisms underlying the observed differences in recipient and donor-derived cfDNA, specifically exploring structural characteristics and their relevance to specific phenotypes such as AMR, ACR and infection. Such studies will likely uncover structural cfDNA features that are clinically relevant.
Study Limitations:
The small number of NHP studied is a potential limitation. However, despite the small sample size, xdcfDNA analyses accurately discriminated the presence or absence of acute rejection.
Supplementary Material
ACKNOWLEDGEMENT
Alison Davis, PhD and Kelly Byrne for helpful reviews and comments and NHLBI Sequencing and Computational Cores for shotgun sequencing and computation support.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ETHICS COMPLIANCE
This study was performed in strict compliance with the ethical standards set forth by the World Medical Association, as stated in the Declaration of Helsinki, as well as the International Society for Heart and Lung Transplantation’s Statement on Transplant Ethics.
CONFLICT OF INTEREST AND FUNDING SOURCES DISCLOSURE
GRAfT study () was supported by the NHLBI intramural research program. DLA is an employee of Revivicor, Inc. None of the remaining authors have any conflicts of interest or relevant financial relationships with an external or commercial entity to disclose. Xenotransplantation work at the NHLBI was funded through contract HHSN268201300001C and gift funds from Revivicor, Inc. No external or commercial entities were involved in this study’s design, collection, analysis, or interpretation of data. The authors had access to all of the data in this study and take complete responsibility for the integrity of the data and the accuracy of the data analysis. The decision to write this report and to submit it for publication was made independently by the authors.
REFERENCES
- 1.Lo YM, et al. , Presence of donor-specific DNA in plasma of kidney and liver-transplant recipients. Lancet, 1998. 351(9112): p. 1329–30. [DOI] [PubMed] [Google Scholar]
- 2.Fournie G, et al. , Plasma DNA as a marker of cancerous cell death. Investigations in patients suffering from lung cancer and in nude mice bearing human tumours. Cancer Lett, 1995. 91. [DOI] [PubMed] [Google Scholar]
- 3.Kim K, et al. , Circulating cell-free DNA as a promising biomarker in patients with gastric cancer: diagnostic validity and significant reduction of cfDNA after surgical resection. Ann Surg Treat Res, 2014. 86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Filho EMR, et al. , Elevated cell-free plasma DNA level as an independent predictor of mortality in patients with severe traumatic brain injury. J Neurotrauma, 2014. 31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Margraf S, et al. , Neutrophil-derived circulating free DNA (CF-DNA/NETS): a potential prognostic marker for posttraumatic development of inflammatory second hit and sepsis. Shock, 2008. 30. [DOI] [PubMed] [Google Scholar]
- 6.Avriel A, et al. , Admission cell free DNA levels predict 28-day mortality in patients with severe sepsis in intensive care. PLoS One, 2014. 9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Agbor-Enoh S, et al. , Applying rigor and reproducibility standards to assay donor-derived cell-free DNA as a non-invasive method for detection of acute rejection and graft injury after heart transplantation. J Heart Lung Transplant, 2017. 36(9): p. 1004–1012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bloom RD, et al. , Cell-Free DNA and Active Rejection in Kidney Allografts. J Am Soc Nephrol, 2017. 28(7): p. 2221–2232. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.De Vlaminck I, et al. , Noninvasive monitoring of infection and rejection after lung transplantation. Proc Natl Acad Sci U S A, 2015. 112(43): p. 13336–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Grskovic M, et al. , Validation of a Clinical-Grade Assay to Measure Donor-Derived Cell-Free DNA in Solid Organ Transplant Recipients. J Mol Diagn, 2016. 18(6): p. 890–902. [DOI] [PubMed] [Google Scholar]
- 11.De Vlaminck I, et al. , Circulating cell-free DNA enables noninvasive diagnosis of heart transplant rejection. Sci Transl Med, 2014. 6(241): p. 241ra77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Snyder TM, et al. , Universal noninvasive detection of solid organ transplant rejection. Proc Natl Acad Sci U S A, 2011. 108(15): p. 6229–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Mohiuddin MM, et al. , Chimeric 2C10R4 anti-CD40 antibody therapy is critical for long-term survival of GTKO.hCD46.hTBM pig-to-primate cardiac xenograft. Nat Commun, 2016. 7: p. 11138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Zhu X, Dor FJ, and Cooper DK, Pig-to-non-human primate heart transplantation: immunologic progress over 20 years. J Heart Lung Transplant, 2007. 26(3): p. 210–8. [DOI] [PubMed] [Google Scholar]
- 15.De Vlaminck I, et al. , Temporal response of the human virome to immunosuppression and antiviral therapy. Cell, 2013. 155(5): p. 1178–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Agbor-Enoh S, et al. , Early Graft Injury Measured by Donor-Derived Cell-Free DNA Predicts Early Mortality in Lung Transplant Recipients. The Journal of Heart and Lung Transplantation, 2017. 36(4): p. S65. [Google Scholar]
- 17.Mohiuddin MM, et al. , B-cell depletion extends the survival of GTKO.hCD46Tg pig heart xenografts in baboons for up to 8 months. Am J Transplant, 2012. 12(3): p. 763–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Horvath KA, et al. , Left ventricular pressure measurement by telemetry is an effective means to evaluate transplanted heart function in experimental heterotopic cardiac xenotransplantation. Transplant Proc, 2010. 42(6): p. 2152–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Sus_scrofa_10.2: Organism overview Available from: http://www.ensembl.org/Sus_scrofa/Info/Annotation; Accessed November 28, 2017.
- 20.Baylor College of Medicine, Human Genome Sequencing Center. Non-Human Primate Genome Sequencing: Panubis_2.0; Available from: https://www.hgsc.bcm.edu/non-human-primates/baboon-genome-project; Accessed November 28, 2017.
- 21.Bolger AM, Lohse M, and Usadel B, Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics, 2014. 30(15): p. 2114–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Langmead B and Salzberg SL, Fast gapped-read alignment with Bowtie 2. Nat Methods, 2012. 9(4): p. 357–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Li H, et al. , The Sequence Alignment/Map format and SAMtools. Bioinformatics, 2009. 25(16): p. 2078–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Hedges LV, Distribution Theory for Glass’s Estimator of Effect size and Related Estimators. Journal of Educational Statistics, 1981. 6(2): p. 107–128. [Google Scholar]
- 25.Lo YM, et al. , Rapid clearance of fetal DNA from maternal plasma. Am J Hum Genet, 1999. 64(1): p. 218–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Prism G, GraphPad Prism, in version 7 2008, GraphPad Software: La Jolla, CA. [Google Scholar]
- 27.Bhorade SM, et al. , Interobserver variability in grading transbronchial lung biopsy specimens after lung transplantation. Chest, 2013. 143(6): p. 1717–24. [DOI] [PubMed] [Google Scholar]
- 28.Marboe CC, et al. , Nodular endocardial infiltrates (Quilty lesions) cause significant variability in diagnosis of ISHLT Grade 2 and 3A rejection in cardiac allograft recipients. J Heart Lung Transplant, 2005. 24(7 Suppl): p. S219–26. [DOI] [PubMed] [Google Scholar]
- 29.Ivanov M, et al. , Non-random fragmentation patterns in circulating cell-free DNA reflect epigenetic regulation. BMC Genomics, 2015. 16 Suppl 13: p. S1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Snyder MW, et al. , Cell-free DNA Comprises an In Vivo Nucleosome Footprint that Informs Its Tissues-Of-Origin. Cell, 2016. 164(1–2): p. 57–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Sun K, et al. , Plasma DNA tissue mapping by genome-wide methylation sequencing for noninvasive prenatal, cancer, and transplantation assessments. Proc Natl Acad Sci U S A, 2015. 112(40): p. E5503–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Jiang P, et al. , Lengthening and shortening of plasma DNA in hepatocellular carcinoma patients. Proc Natl Acad Sci U S A, 2015. 112(11): p. E1317–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.