Abstract
Donor-derived cell-free DNA (dd-cfDNA) is reportedly a valuable tool for graft surveillance following kidney transplantation (KTx). Possible changes in dd-cfDNA(%) reference values over time have not been evaluated. For long-term monitoring after KTx, changes in host cfDNA might represent a biasing factor in dd-cfDNA(%) determinations.
Plasma samples were obtained (n = 929) 12–60 months after engraftment in a cross-sectional cohort of 303 clinically stable KTx recipients. Total cfDNA(copies/mL), dd-cfDNA(%), and dd-cfDNA(copies/mL) were determined using droplet-digital PCR. Stability of threshold values in these stable KTx recipients over time was assessed by 80th, 85th, and 90th quantile regression.
Upper percentiles of total cfDNA showed a significant decline of −1902, −3589, and −4753 cp/mL/log(month) (P = 0.014, <0.001, and 0.017, respectively), resulting in increasing dd-cfDNA(%) percentiles by 0.25, 0.46, and 0.72%/log(month) (P = 0.04, 0.001, and 0.002, respectively), with doubling of the 85th percentile value by 5 years. In contrast, dd-cfDNA(cp/mL) was stable during the observation period (P = 0.52, 0.29, and 0.39). In parallel increasing white blood cell counts and decreasing tacrolimus concentrations over time were observed. After 5 years, the median total cfDNA was still 1.6-fold (P < 0.001) higher in KTx recipients than in healthy controls (n = 135) and 1.4-fold (P < 0.001) higher than patients with other medical conditions (n = 364).
The time-dependent decrease of host cfDNA resulted in an apparent increase of dd-cfDNA fraction in stable KTx patients. For long-term surveillance, measurement of absolute dd-cfDNA concentrations appears to be superior to percentages to minimize false positive results.
Traditional methods to assess transplant organ damage are imprecise or invasive, and therefore new biomarkers for noninvasive monitoring of graft integrity are needed. Donor-derived cell-free DNA (dd-cfDNA) is one of the best investigated biomarkers (1), with an increasing number of clinical validation studies published.
Measuring dd-cfDNA has a very clear and solid biological rationale, since any DNA released from an organ graft into the blood stream originates from a nonviable cell. Due to the very short half-life of cell-free DNA (cfDNA) in plasma (2), which is in the range of less than an hour, dd-cfDNA is particularly suitable for real-time noninvasive monitoring of graft health.
Most methods used in larger studies have monitored the amount of dd-cfDNA in percentage of the total cfDNA in the patients’ plasma [dd-cfDNA(%)]. This relative quantification has the advantage of being simple to measure and technically fairly robust and precise. Almost all published studies designed to establish clinically useful cut-off values to distinguish a healthy graft from one with injury have measured the dd-cfDNA(%) in the recipient’s plasma.
However, little is known about changes in host cfDNA, which represents the denominator in the percentage calculations, over time. The major source of cfDNA are the circulating white blood cells (WBCs). The dd-cfDNA(%) in healthy graft conditions has been found to be near 1% or less in kidney (3–6,), heart (7, 8,), and lung (9,) recipients, and in only low single digit percentage ranges in liver transplant recipients (10).
The concentration of any analyte in plasma follows the simple rules of homeostasis, where the distribution volume, production, and elimination ratios are the main factors to be considered. Reference values, which are used in the clinical laboratory, are established in reference populations under the assumption that those parameters represent intra- and interindividual constants maintained within narrow ranges under physiological conditions. The smaller the effects of any confounding factors are, the narrower the reference range will be (i.e., less biologic noise). Changes occurring in any influencing factors would lead to a widening of the reference range or threshold values, resulting in a lower diagnostic performance of the analyte to distinguish a diseased state.
For dd-cfDNA the same basic rules apply, but the use of a percentage value adds an additional source of possible uncertainty. If the denominator (concentration of total cfDNA in plasma) is not constant, or at least does not have a narrow window under all conceivable clinical conditions, dd-cfDNA(%) values can change without a change in the value of the analyte. There are countless publications describing changes of cfDNA concentrations under physiologic (11, 12,) and pathologic conditions (13–15,). Therefore, the assumption that host cfDNA is fairly constant seems unjustified. For example, even a short exercise leads to a doubling of cfDNA in plasma (11,) and longer exercises can result in more than 10-fold increases in total cfDNA plasma concentrations (12).
Even though soon after any surgery, higher amounts of cfDNA are conceivably present from tissue sources, generally the WBCs are considered the major (>95%) source of cfDNA physiologically present (16, 17,). The only solid organ making a significant single contribution seems to be the liver (10, 16,), most probably based on its mass and structure. Therefore, changes in the number of WBCs or their destruction could result in substantial variations in total cfDNA concentrations after organ transplantation. It has recently been shown in 2 large clinical studies that the absolute quantification of dd-cfDNA, expressed as copies/mL of plasma, was superior to percentage values in distinguishing biopsy proven antibody-mediated rejection (ABMR) (18,), as well as borderline T-cell mediated rejection (TCMR) and episodes of acute tubular necrosis (3,) from a healthy graft condition during graft surveillance after kidney transplantation (KTx). The 1% threshold value used for one dd-cfDNA(%) method was established over a narrow time span early (on average 32 days) after KTx and therefore is formally only valid for this early period after surgery (19).
The aim of the study presented herein was to investigate the dynamics of cfDNA during the 1 to 5 year period after kidney transplantation in clinically stable KTx recipients. The primary objective was to determine whether there were changes in the threshold values over time.
Materials and Methods
Study Design
The study cohort was part of a prospective clinical validation trial funded by the German Ministry of Education and Research (BMBF-# FKZ 031A584A+B). Samples were drawn at defined time-points between 12 and 60 months (Fig. 1) in a cross-sectional cohort of 303 patients. Certain patients from the earlier reported longitudinal cohort (3,) were included in this cross-sectional cohort. All blood collection procedures and laboratory procedures were as published elsewhere (3). The estimated glomerular filtration rate was calculated by the Chronic Kidney Disease Epidemiology Collaboration equation and normalized to 1.73 m2 body surface area.
Consort diagram of the study. N = number of subjects, n = number of samples.
A total of 929 samples were analyzed for total cfDNA as copies per mL plasma (cp/mL), dd-cfDNA(%) and dd-cfDNA concentration in cp/mL [dd-cfDNA(cp/mL)]. Patients were recruited under Institutional Review Board approval at routine visits for clinical evaluation, as well as biochemistry and immunosuppressive drug determinations performed at a single transplant center. Each patient provided written informed consent to participate in the study. Only apparently stable patients without clinically or biochemically suspected graft injuries were included in the study cohort. Patients were receiving standard immunosuppression and at the sampling time points 92% were receiving tacrolimus, in whom 85% this was in combination with mycophenolate, while 8% were receiving only mycophenolate. Figure 1 provides an overview of the samples included herein and the patients’ demographics are given in Supplemental Table 1.
Measurement of dd-cfDNA
All laboratory measurements were performed as described elsewhere (3,) and recorded de-identified in a central study database. All cfDNA determinations were performed using droplet digital PCR on a QX200 system (BioRad), where dd-cfDNA determination was based on a universal single point mutation panel (20, 21). Briefly, the method for cfDNA determination consisted of correction for cfDNA extraction efficiency by means of an artificial DNA, spiked into the plasma before DNA extraction, for which the Large volume DNA kit (Roche Diagnostics) was used. In a second step, the fragmentation rate of each individual sample was determined by a 2-length ddPCR and the results of the quantification of ddPCR were corrected for both DNA extraction efficacy and fragmentation-based PCR efficiency. The value for absolute dd-cfDNA(cp/mL) was calculated by multiplying the results of the fractional abundance of dd-cfDNA(%) by the measured total cfDNA(cp/mL) value.
Statistical Analysis
After unblinding, the data were grouped by sampling time point and evaluated.
As a comparison group for total cfDNA concentration 364 samples from patients with other medical conditions (OC), as well as 135 samples from apparently healthy controls (HC) was used. The OC group contained patients with a variety of different conditions, for example, patients prior to and after surgery for benign or nonbenign diseases, as well as patients receiving different chemotherapy regimens for malignancies (22, 23).
The ability to detect graft damage is mainly based on the correct definition of a threshold, which defines the upper limit of the dispersion range in KTx recipients with stable grafts. Since clinically silent (subclinical) rejections (mainly ABMR) were expected to occur in 10% to 20% of any apparently stable KTx recipients (24, 25), we focused on the range between the 80th and 90th percentile calculated for each time point in this clinically stable cohort.
Changes over time were inferred by means of quantile regression (26) on the 80th, 85th, and 90th percentile of each dependent variable with the number of log-months as the independent variable. Statistical significance of regression coefficients was calculated using 10,000 xy-pair bootstraps and 95% pointwise confidence intervals provided. Statistical significance was concluded at significance level α = 5%. All analyses were performed in R version 3.6.1. Continuous data were presented with median and interquartile range (IQR) or mean and standard deviation, with 95% confidence intervals where feasible, whereas frequencies are reported as proportions. Box plots show the 5th and 95th percentiles as whiskers, 25th and 75th percentiles as a box and a median line. Differences between KTx recipients and patients with other medical conditions (OC) or apparently healthy controls (HC) were calculated using Wilcoxon’s Rank Sum Test.
Results
Total cfDNA Dynamics Over Time
We observed a steady decrease of total cfDNA over time during the observation period (Fig. 2). The median values showed a steady decrease, reaching a 60 months value that was 30% lower than the year one value. From month 21 onwards the difference to the 12 month values were statistically significant (P < 0.001).
Time course of total cell free DNA (cf-DNA) during the 60 months study period. Boxes depict the 25th and 75th percentiles as a box and a median line; whiskers extend to minimum or maximum, but at most 1.5 × IQR; Left side with restricted Y-axis; right side with y-axis covering all outliers.
An important additional observation was that the range of total cfDNA concentrations at each time point was very high, illustrating the large variability of these values (Supplemental Table 2).
It should be noted that the 60 month total cfDNA concentrations (median = 4419cp/mL; IQR = 2857–7370) were still a 1.42 multiple of the median of the values in a group of patients with other medical conditions (OC: n = 364; median = 3313cp/mL; IQR = 2129–4647) and 1.56 multiple of the median above what was seen in healthy controls (HC: n = 135; median= 2835 cp/mL; IQR = 1831–3840), with a P-value <0.001 for both. Additionally, the OC group had no patients receiving any immunosuppressant drugs.
The median plasma creatinine as well as the estimated glomerular filtration rate in the study cohort was nearly constant over the entire observation period (Fig. 3).
Time course of plasma creatinine a) and estimated glomerular filtration rate (eGFR) b) during study period. Boxes depict the 25th and 75th percentiles as a box and a median line; whiskers extend to minimum or maximum, but at most 1.5 × IQR.
Decision Threshold Value Dynamics Over Time
When focusing on the putative threshold values, there was a statistically significant, time dependent increase in dd-cfDNA(%) upper quantiles (Fig. 4A) (cf. Fig. 5). This result is explained by the significantly decreasing denominator (total cfDNA) as shown in Figs. 2 and 5.
![a) Time course of donor-derived cell-free DNA percentage [dd-cfDNA(%)] during the study period. b) Time course of dd-cfDNA absolute concentration [dd-cfDNA(cp/mL)] during the study period. Percentiles for 50th to 90th percentiles for each time point are displayed.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/clinchem/66/10/10.1093_clinchem_hvaa175/1/m_hvaa175f4.jpeg?Expires=1712786294&Signature=adwKSH7qg5LEDKtK~mfmU4dGPaKYJngXzVKaf117flSMgBWlEYhpo7y3KBgRtGJ9hC5gk9gkgVJ-WxGhtlrN5m~8wy9I5QvMgTBjHvGkyWz1PI6J2lfReGMLe96gyiDW7vnLYuKueyOs~6iKHxpgVUP58m4C-pPjONgO2CbYJpEY10SR0sfo-ooR7WviS5mW8Cy5SLwW2phn25ws15Zo7lzDw8YPgm4kxNnhLYpl65yN9y~0y1eqOmWdxXp63f2-r-WC0a7z4~galYADZT7FUNQAj89R4G~2SD9jutRbA3~qJP4S~CWBqtH~dJ1cF8nbxH-ovggKT~Xd0Yz-45S0jw__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
a) Time course of donor-derived cell-free DNA percentage [dd-cfDNA(%)] during the study period. b) Time course of dd-cfDNA absolute concentration [dd-cfDNA(cp/mL)] during the study period. Percentiles for 50th to 90th percentiles for each time point are displayed.
Regression analyses of study variables vs. log-months after kidney transplant. Time dependencies are given as slope coefficients of quantile regression for given parameters (y-axes) vs. log-months for the different percentiles (x-axis). A deviation of the 95%-pointwise confidence intervals (gray area) from 0 indicates a significant change of values with time. cfDNA: cell-free DNA; dd-cfDNA: donor derived cell-free DNA; WBC: white blood cells.
In contrast to dd-cfDNA(%), absolute dd-cfDNA concentrations, in number of copies/mL in plasma remained stable during the observation period (Fig. 4B). The respective boxplots are presented in Supplemental Figs 1 and 2.
It is particularly noteworthy that the intercept of the 80th percentile (47.5 cp/mL) was nearly the same as the value shown to discriminate best between stable and biopsy-proven rejection patients during the first year after KTX in the first cohort of this study (52 cp/mL). This reference value established for the first year post-KTx (3) was around the 80th and 85th percentiles during the entire observation period.
Figure 5 presents an overview of the time dependencies of the parameters investigated. The slope values (y-axis) for each percentile (x-axis) are displayed together with the respective confidence intervals (grey areas). All parameters had a significant time dependency in the direction explained above, since the point of no difference (zero) did not fall within the confidence limits. The only exception was the absolute concentration of dd-cfDNA in cp/mL. The respective calculated numeric correlation values and significances for the upper percentiles are shown in Table 1.
Matrix of correlation values of cell-free DNA (cfDNA) parameters vs. log (months).
| Variable | 80th percentile | 85th percentile | 90th percentile |
|---|---|---|---|
| total cfDNA(cp/mL) | −1902 (P = 0.014) | −3589 (P < 0.001) | −4753 (P = 0.016) |
| dd-cfDNA(%) | 0.248 (P = 0.041) | 0.461 (P = 0.001) | 0.715 (P = 0.002) |
| dd-cfDNA(cp/mL) | 4.43 (P = 0.519) | 9.54 (P = 0.286) | 16.2 (P = 0.385) |
| Variable | 80th percentile | 85th percentile | 90th percentile |
|---|---|---|---|
| total cfDNA(cp/mL) | −1902 (P = 0.014) | −3589 (P < 0.001) | −4753 (P = 0.016) |
| dd-cfDNA(%) | 0.248 (P = 0.041) | 0.461 (P = 0.001) | 0.715 (P = 0.002) |
| dd-cfDNA(cp/mL) | 4.43 (P = 0.519) | 9.54 (P = 0.286) | 16.2 (P = 0.385) |
Slopes and P-values of quantile regression of cfDNA moieties vs. time (based on log-months) for the 80th, 85th, and 90th percentiles. cfDNA: cell-free DNA; dd-cfDNA: donor derived cell-free DNA.
Matrix of correlation values of cell-free DNA (cfDNA) parameters vs. log (months).
| Variable | 80th percentile | 85th percentile | 90th percentile |
|---|---|---|---|
| total cfDNA(cp/mL) | −1902 (P = 0.014) | −3589 (P < 0.001) | −4753 (P = 0.016) |
| dd-cfDNA(%) | 0.248 (P = 0.041) | 0.461 (P = 0.001) | 0.715 (P = 0.002) |
| dd-cfDNA(cp/mL) | 4.43 (P = 0.519) | 9.54 (P = 0.286) | 16.2 (P = 0.385) |
| Variable | 80th percentile | 85th percentile | 90th percentile |
|---|---|---|---|
| total cfDNA(cp/mL) | −1902 (P = 0.014) | −3589 (P < 0.001) | −4753 (P = 0.016) |
| dd-cfDNA(%) | 0.248 (P = 0.041) | 0.461 (P = 0.001) | 0.715 (P = 0.002) |
| dd-cfDNA(cp/mL) | 4.43 (P = 0.519) | 9.54 (P = 0.286) | 16.2 (P = 0.385) |
Slopes and P-values of quantile regression of cfDNA moieties vs. time (based on log-months) for the 80th, 85th, and 90th percentiles. cfDNA: cell-free DNA; dd-cfDNA: donor derived cell-free DNA.
Possible Causes of cfDNA Dynamics
Considering the large and convincing body of evidence that the vast majority of total cfDNA comes from circulating WBC, we examined possible interdependencies between WBC counts and total cfDNA including tacrolimus whole blood concentrations as possible contributing covariates. As shown in Supplemental Table 3, WBC medians correlated negatively with both total cfDNA and tacrolimus medians; in contrast to a positive association of tacrolimus concentrations with cfDNA. (Additional visualizations are shown in Supplementary Figs 3 and 4). An additional observation is that the cfDNA fragment length index, a measure of cfDNA fragmentation (3), increased over time, being inversely correlated to both tacrolimus and WBC medians. Although quantitatively not high enough to cause a quantification bias (2% difference at 12 vs. 60 months), it provides a hint into the pathophysiology of cfDNA in patients taking CNI inhibitors.
Discussion
Monitoring of dd-cfDNA has been validated in several clinical trials with somewhat different or even conflicting results in terms of diagnostic performance and threshold values used to distinguish KTx recipients with a healthy graft from those with graft damage (e.g., rejections). Most of the currently reported dd-cfDNA studies have measured only the relative fraction dd-cfDNA(%) of the donor’s cfDNA, which is the simplest, cheapest, and easiest method to measure dd-cfDNA. However, numerous publications have shown that it is naive to assume that the concentration of host cfDNA (denominator for %) remains constant under all conceivable clinical and physiological conditions (11, 12, 14, 15, 27,). The denominator of the percentage expression can and will vary over time. In addition, preanalytic variation, a well-known issue in laboratory medicine, has not been adequately evaluated by most dd-cfDNA test vendors. Lysis of WBC, for example, will lead to falsely low dd-DNA(%) values. Another more specific issue in transplantation is the fact that immunosuppressive drugs can lead to alterations of WBCs, in their composition as well as numbers. Calcineurin, for example, has been shown to play a role in the response to shear stress of eukaryotic cells, and some recent publications suggest that calcineurin and mTor (mammalian target of rapamycin) inhibitors have a negative effect on cell stability (28, 29,). This effect could lead to increased cfDNA release by an increased lysis of WBC, which are the main source (≥90%) of the total cfDNA in plasma. Tacrolimus has been found to result in a higher apoptosis rate in cells exposed to shear stress (28,). It is known that leukocytes are exposed to such shear stress constantly during capillary passage (30, 31,). One conceivable hypothesis therefore is that the link between tacrolimus and increased host cfDNA is the result of a higher apoptosis rate of WBC in the blood stream, resulting in a lower WBC count, as was observed in this cohort and consequently in an increase of total cfDNA. The immunosuppressive drug minimization strategies currently used in long-term stable solid organ recipients could lead to a systematic decrease of host and thereby total cfDNA due to a longer survival time of leucocytes and increased WBC counts, as a result of a decrease in the above-mentioned calcineurin inhibitor-related shear stress vulnerability. This would lead to increased dd-cfNDA(%) values. Noteworthy, a previous report (32) already showed a slow but significant increase in dd-cfDNA(%) with time after lung transplantation in absence of infection and rejection.
These considerations led us to investigate a possible time-dependent change of cfDNA in a large cohort of KTx patients in years 1 to 5 after engraftment. Whitlam et al. had already shown a decrease in total cfDNA in a small set of patients’ samples over a very long time period (up to 20 years), which resulted in apparently elevated dd-cfDNA percentage values (18). Our prospective study confirmed this initial report in a large prospectively collected cohort. A steady decrease of cfDNA over time was observed, and yet even after 5 years the cfDNA concentration was still higher than in patients with other medical conditions not receiving calcineurin inhibitors or in healthy individuals. This is consistent with our hypothesis concerning the influence of calcineurin inhibitors on host cfDNA. Mycophenolate, which is often combined with calcineurin inhibitors, is also tapered over time and could be in part another cause for the increasing WBC over time. Nevertheless, we could not find any mechanism in the literature concerning stability of leukocytes, which would explain the counter-intuitive effect reported herein of decreasing total cfDNA while increasing its major source.
The literature suggests that in a cohort of apparently stable KTx-patients, about 10%–20% will have subclinical rejection (25,), mostly ABMR (24,). Since the most important use of larger cohorts of apparently healthy individuals is to define the upper threshold of normal values to be used as reference, we focused on the upper percentiles while taking the mentioned unavoidable number of subclinical rejections into account. The threshold established during the first year after KTx (52 cp/mL) for dd-cfDNA (cp/mL) was around and between the 80th and 85th percentile in this long-term cohort with no trend over time. In contrast, the 85th percentile of the dd-cfDNA(%)at 15 months (0.61%) was already slightly higher compared to the first-year value (0.48%). More importantly, this value increased steadily over time, doubling near year 5. Because Whitlam et al. found a substantial increase of dd-cfDNA percentages in a small group of patients followed for over more than 10 years after KTx, we suspect that this increase continues over time (18). Both the Whitlam et al. data and the data presented herein indicate that neither total host nor dd-cfDNA(%) values will have reached stable concentrations at the end of a 5-year observation period. We chose to focus on the 80th to 85th percentile of this apparently stable KTx group since the putative limit of the reference range was based on the assumption that, above the 85th percentile, unrecognized (subclinical) rejections would unavoidably confound the results of the cohort. This is a rather conservative approach and it needs to be noted that the effect shown here would be even more pronounced if the 95th percentile were considered.
A practical solution for a reliable establishment of a valid threshold value is to eliminate the confounding effect of the variable total (host) cfDNA by the determination of the (absolute) concentration of the dd-cfDNA released by the graft (in copies/mL of patient plasma). Doing this adds some complexity, but the added costs are negligible compared to either the cost of current assays (especially in the US) or to the potential harm that an increasing number of false positive dd-cfDNA(%) values could cause over time. The added complexity and minimal increase in costs must be compared to the cost of unnecessary biopsies and additional diagnostic tests, which could result from the use of only dd-cfDNA(%) values during long-term post KTx monitoring.
It should also be noted that in a published direct comparison of dd-cfDNA(%) values with dd-cfDNA(cp/mL) (3), the use of dd-cfDNA(cp/mL) significantly improved the ability to distinguish TCMR from a healthy graft condition during the first year after KTx.
Given some published concerns about the value of dd-cfDNA after KTx (in particular, for detection of TCMR), it should be mentioned that effects of the technology used, such as percentage vs. concentration or shotgun vs. targeted assays, cannot be ruled out as confounding factors. We and others have shown that TCMR can be detected (3, 4,), while some others did not find sufficient detection rates (6, 33,). However, these apparently unfavorable clinical studies used only dd-cfDNA(%) rather than absolute concentrations and none considered the time after engraftment as a possible confounding factor. The time dynamics reported herein might, at least in part, explain the differences in these study results. Another recent study (which included only a very few cases of TCMR) reported on the attempt to distinguish (only a few cases) TCMR from other types of graft damage, including acute tubular necrosis (5,). Based on the fact that necrosis must lead to a release of graft DNA into the blood stream, such an evaluation seems per se futile. Therefore, the conclusion should not be that dd-cfDNA is generally a questionable biomarker in solid organ transplantation. The specific shortcomings of certain assays and study objectives also need to be considered (34). A better conclusion would be that, in contrast to absolute dd-cfDNA(cp/mL) quantification, percentage technologies can encounter specific problems and that the time after transplantation, immunosuppressive regimens, and perhaps other not yet investigated sources might add more noise to percentage dd-cfDNA(%) values than suspected so far.
The current study was limited to the analysis of cross-sectional samples from solely clinically stable patients. Studies that include samples from patients with, for example, late-rejections are needed to examine whether the effect seen of increased dd-cfDNA(%) over time in stable patients is present in other such cases as well. Nevertheless, the increased diagnostic performance of absolute versus relative dd-cfDNA determinations as already demonstrated for the first year after KTx (3) should be considered when choosing an assay method.
Our hypothesis driven investigations of time-related changes in host-derived cfDNA produced two major results: i) host cfDNA concentrations can be very variable (over two orders of magnitude) in stable KTx patients, and ii) the clear trend towards lower values persisted over the 1–5 years after transplantation. Both of these findings demonstrate that dd-cfDNA(%) percentage values, as opposed to absolute concentrations of dd-cfDNA(cp/mL), can vary over time in a graft damage independent manner, adding substantial noise to the test results.
We strongly recommended that laboratories use (as in almost all other laboratory testing) a quantitative value instead of a percentage to obtain the best possible diagnostic performance from dd-cfDNA, a biomarker with proven clinical value in patient care after transplantation. The use of only dd-cfDNA(%) for long term surveillance, for which reference values have not been established (19), does not seem to be justified in light of the dynamic changes in cfDNA demonstrated.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Author Contributions
All authors confirmed they have contributed to the intellectual content of this paper and have met the following 4 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; (c) final approval of the published article; and (d) agreement to be accountable for all aspects of the article thus ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved.
E. Wieland, M. Shipkova, V. Schauerte, and M. Kabakchiev were responsible for recruitment, sample collection, and documentation of clinical data. N. Mettenmeyer and J. Beck performed the laboratory work. T. Asendorf performed the statistical data analysis. M. Oellerich, P.D. Walson, E. Schütz, and J. Beck wrote the manuscript and edited the article, and provided scientific advice. V. Schauerte, E. Wieland, and M. Shipkova provided clinical and scientific advice.
Authors’ Disclosures or Potential Conflicts of Interest
Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest:
Employment or Leadership
E. Schütz, Chronix Biomedical; J. Beck, Chronix Biomedical GmbH.
Consultant or Advisory Role
P.D. Walson, Chronix Biomedical GmbH; M. Oellerich, Chronix Biomedical, Liquid Biopsy Center GmbH (LBC).
Stock Ownership
E. Schütz, Chronix Biomedical.
Honoraria
E. Schütz, Chronix Biomedical.
Research Funding
The German Federal Ministry of Education and Research (BMBF), Project Management Jülich to E. Schütz, M. Oellerich/Tim Friede and E. Wieland (KMU-innovativ-13: FKZ 031A584A+B).
Expert Testimony
None declared.
Patents
E. Schütz, US10,570,443; J. Beck, EP3004388, PCT/US2015/053304.
Other Remuneration
P.D. Walson, expenses for train travel reimbursed by Chronix Biomedical GmbH.
Role of Sponsor
The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, preparation of manuscript, or final approval of manuscript.
References
Nonstandard Abbreviations:
- ABMR
antibody-mediated rejection
- cfDNA
cell-free DNA
- dd-cfDNA
donor-derived cell-free DNA
- DNA
deoxyribonucleic acid
- HC
healthy controls
- IQR
interquartile range
- KTx
kidney transplantation
- OC
other medical conditions
- TCMR
T-cell mediated rejection
