Rejection continues to be the greatest threat to long-term survival for heart transplant recipients. The gold standard for rejection surveillance remains endomyocardial biopsy (EMB), an invasive procedure fraught with a modicum of risk, discomfort, and usually a short hospital stay.1 Further, given the patchy nature of rejection combined with significant interobserver variability in pathologic assessment, EMB remains an imperfect standard. We have previously reported the development of a rapid and sensitive quantitative genotyping assay to identify and quantify fragments of cell-free DNA (cfDNA) in patient plasma.2,3 In the setting of solid organ transplantation, donor-specific cfDNA from the transplanted organ can be selectively analyzed from recipient cfDNA present in the circulation as the proportion of donor-specific to total cfDNA, measured as percent donor fraction (DF). In organ transplantation, the DF can be used as a biomarker for events causing injury specific to the allograft, such as rejection, whereas the total cfDNA reflects more generalized events associated with cell turn-over, such as infection or inflammation. In preliminary studies, we showed the utility of percent DF cfDNA levels as a highly sensitive marker for rejection in pediatric heart transplant recipients.3 Our protocol called for correlation of DF levels with the results of EMB. As a practical matter, study samples were routinely drawn at the time of EMB. We hypothesized that the injury caused by the biopsy itself may result in important increases in donor-specific cfDNA in plasma. We, therefore, devised a substudy including paired samples drawn before and after EMB to understand this relationship. Patients consented in the primary study were eligible for substudy enrollment. Patients with graft vasculopathy, cancer, mechanical circulatory support, cellular rejection grade >1, or any antibody-mediated rejection at the time of the sample draw were excluded, leaving 20 paired samples for analysis. Quantities of total and DF (%) cfDNA levels were determined using the myTAI-HEART test, a proprietary quantitative genotyping assay (TAI Diagnostics, Wauwatosa, WI).
Three to 10 ml of anti-coagulated blood were collected in Cell-Free DNA Blood Collection Tubes (Streck, Omaha, NE) to assess circulating levels of cfDNA. Samples were immediately coded, de-identified, and delivered to the laboratory. Clinical, laboratory, cardiac catheterization, and echocardiographic data were recorded at the time of collection, and data were managed using REDCap.4 Demographic variables included weight, age, height, and time from transplant. Procedural data included time from biopsy completion to the post-biopsy sample collection, bioptome size, and number of biopsies taken, along with any documented procedural or physiologic complications.
Separation of plasma from whole blood by centrifugation was carried out as previously described.5 Plasma was stored at −80°C until DNA extraction. cfDNA extractions were performed using ReliaPrep HT Circulating Nucleic Acid Kit (Promega, Madison, WI). Recipient genomic DNA was extracted by using ReliaPrep Large Volume gDNA Isolation System (Promega). All purified genomic DNA was resuspended in 0.1 × buffer containing ethylene-diaminetetraacetic acid. Total cfDNA (ng/ml plasma) content from circulating plasma was evaluated by quantitative real-time PCR as previously described.5 PCR analysis was carried out on an Applied Biosystems QuantStudio 7 (Thermo Fisher Scientific, Waltham, MA). The myTAI-HEART assay was used to calculate DF (%) cfDNA (TAI Diagnostics).3 The assay quantitatively genotypes a panel of high frequency single nucleotide polymorphisms selected for their ability to reliably discriminate between alleles. DF was determined using an algorithm that was not reliant on donor genotyping.3 Genomic equivalents (GE) are calculated by multiplying the total cfDNA (ng/ml plasma) by the DF (%) and then by 302 genomes per ng.
Data processing was performed in RStudio. Statistical analyses included linear modeling and the paired Student t test for difference in means. Data are reported as median and interquartile range.
Quantities of donor-specific cfDNA were reported as donor GE and increased from a median (interquartile range) of 14.8 (6.3–21.6) to 113.1 (44.1–247.2) after biopsy (p < 0.01; Figure 1). The GE of donor-specific cfDNA increased after biopsy in all patients with a 7.5-fold median increase. Patient characteristics and summary data are shown in Table 1. Age and weight are correlated (Pearson + 68%), and both are associated with GE change (p < 0.05), such that smaller and younger patients saw relatively more organ injury (average 14.5 GE less per year of age and 5.5 GE less per kg). Pediatric patients (aged <17 years) had an average GE increase of 35.1-fold (Median pre = 9.2, post = 148.0, n = 12) compared with adult patients who had an average GE increase of 4.4-fold (Median pre = 17.3, post = 56.1, n = 8; p < 0.01, after log transform). GE change did not correlate with bioptome size (p = 0.4). A longer time interval between samples was associated with a greater increase in total cfDNA (Short = <10 minutes, n = 15, mean increase = 1.15 ± 0.29 times; Long = >10 minutes, n = 5, mean increase = 2.50 ± 1.8 times; p < 0.008). There were no reported procedural complications.
Figure 1.
Boxplots of donor-specific cell-free DNA (cfDNA) expressed in genomic equivalents (GE) per ml. Boxes represent interquartile range (IQR) with median. Each sample pair (N = 20) is shown as a set of dots connected with gray lines from the pre-biopsy measure to the post-biopsy one.
Table 1.
Patient Characteristics and Summary Data
| Measure | Median (IQR), N = 20 |
|---|---|
| Age, years | 13.3 (7.5–22.0) |
| Weight, kg | 51.9 (21.8–59.5) |
| Sex | |
| Male | 12 |
| Female | 8 |
| Time since heart transplant, years | 0.4 (0.1–1.4) |
| Time Bx to sample 2, minutes | 6.5 (3.8–11.2) |
| DF pre-bx, % | 0.2 (0.1–0.3) |
| DF post-bx, % | 1.4 (0.5–2.8) |
| LV ejection fraction, % | 65.6 (60.2–68.5) |
| Systemic O2 saturation, % | 97.5 (97–99.2) |
| Cardiac Index, liter/min/m2 | 2.8 (2.5–3.8) |
| Bioptome | 1.5 mm (n = 16) / 3 Fr (n = 4) |
| Total cfDNA pre-bx, ng/ml | 22.5 (13.1–33.4) |
| Total cfDNA post-bx, ng/ml | 25.4 (20.8–41.2) |
| GE/ml pre-bx, genomes | 14.8 (6.3–21.6) |
| GE/ml post-bx, genomes | 113.1 (44.1–247.2) |
bx, biopsy; cfDNA, cell-free DNA; DF, donor fraction; GE, genomic equivalents; IQR, interquartile range; LV, left ventricle.
Standard EMB induces a significant and measurable injury to the transplanted heart. The magnitude of this effect is influenced by patient cohort (pediatric vs adult). In addition to understanding the utility of donor-specific cfDNA as a marker for rejection, early experience reveals that significant increases in total cfDNA can be seen as a sign of inflammation or global injury, such as may be seen in the setting of acute infection. In this series, longer time between EMB and time to acquisition of the second blood sample correlated with increased total cfDNA, suggesting that the procedure itself may be proinflammatory. The relationship between EMB and donor-specific cfDNA levels is evidence of the potential of this technology to serve as a marker of cardiac injury. As commercial quantitative assessment of donor-specific cfDNA becomes increasingly available as a biomarker to identify patients at low risk for rejection, it will be essential that practitioners are aware of the effect of sample timing as it relates to endomyocardial biopsy.
Disclosure statement
This work was supported by the National Institutes of Health award (5R01HL119747; National Heart, Lung, and Blood Institute) and by Advancing a Healthier Wisconsin award (UL1TR001436) as funded through the Clinical and Translational Science Institute at the Medical College of Wisconsin.
A.T-M. and M.M. own stock and are cofounders of TAI Diagnostics. K.S. owns stock and is an employee of TAI Diagnostics. S.Z. and M.H. own stock and are consultants to TAI Diagnostics.
The authors acknowledge Mary Goetsch, Huan ling Liang, and Donna Mahnke for their dedication and support to the lab and this project.
References
- 1.From AM, Maleszewski JJ, Rihal CS. Current status of endomyocardial biopsy. Mayo Clin Proc 2011;86:1095–102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ragalie WS, Stamm K, Mahnke D, et al. Noninvasive assay for donor Fraction of cell-free DNA in pediatric heart transplant recipients. J Am Coll Cardiol 2018;71:2982–3. [DOI] [PubMed] [Google Scholar]
- 3.Hidestrand M, Tomita-Mitchell A, Hidestrand PM, et al. Highly sensitive noninvasive cardiac transplant rejection monitoring using targeted quantification of donor-specific cell-free deoxyribonucleic acid. J Am Coll Cardiol 2014;63:1224–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Harris PA, Taylor R, Thielke R, et al. Research Electronic Data Capture (REDCap)—A metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009;42:377–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hidestrand M, Stokowski R, Song K, et al. Influence of temperature during transportation on cell-free DNA analysis. Fetal Diagn Ther 2012;31:122–8. [DOI] [PubMed] [Google Scholar]