Abstract
Recent work demonstrated that kidney allograft rejection unfolds along a biological continuum that can be quantified using histopathology-derived continuous indices. To investigate whether donor-derived cell-free DNA (dd-cfDNA) reflects this rejection continuum and complements these histopathology-derived indices, we analyzed the association between dd-cfDNA measures and the newly developed rejection indices in 249 indication biopsies from two independent cohorts. dd-cfDNA was analyzed as percentage, absolute copies/mL, and as a previously developed combined continuous model (CM) score integrating both measures to mitigate limitations of relative measurements. dd-cfDNA increased with histopathological activity and was highest in biopsies with microvascular inflammation (MVI), including antibody-mediated (AMR) and mixed rejection, paralleling high AMR/MVI and activity indices. T-cell-mediated rejection (TCMR) showed elevated TCMR/tubulointerstitial inflammation (TI) indices but lower and more variable dd-cfDNA, accompanied by increased total cfDNA, providing a plausible explanation for the reduced detectability of low-grade TCMR when dd-cfDNA is expressed as a percentage alone. Interclass correlation analyses revealed the strongest associations between dd-cfDNA and the AMR/MVI and activity indices. The combined CM score achieved the highest overall associations (sum R2 = 3.4), outperforming absolute and relative dd-cfDNA measures. Thus, dd-cfDNA may serve as a non-invasive readout of graft inflammation and extends the rejection-continuum concept into the non-invasive space.
Graphical Abstract
Introduction
The recently published study by Vaulet et al. compellingly demonstrates that kidney allograft rejection unfolds along a biological continuum rather than within rigid diagnostic subcategories [1]. By deriving continuous activity and chronicity indices from assessed histological lesion scores in a cohort study of 19,500 biopsies from 8,873 patients across 10 centers worldwide, the authors elegantly quantified the spectrum of rejection and distinguished microvascular inflammation from tubulointerstitial inflammation patterns. This is a huge step forward in the interpretation of Banff lesions in a more quantitative manner and shall provide a better and standardized patient care if implemented as histopathological standard.
An important next step is determining whether non-invasive biomarkers map onto this continuum. Donor-derived cell-free DNA (dd-cfDNA) as an indicator of graft damage is now widely used in clinical kidney transplant care, in both for-cause and surveillance settings [2–4]. There is ongoing discussion about whether such a “liquid biopsy” may complement the histopathological and clinical assessment [3–7]. Usually, dd-cfDNA is measured as percentage of total cfDNA, which, however, may cause misinterpretation in patients with high or low total cfDNA [8]. Absolute dd-cfDNA copy numbers may address some of these limitations but require standardized measurement approaches [8, 9]. To address this, we recently developed a combined continuous model (CM) score that integrates relative and absolute dd-cfDNA measurements into a single composite metric, with the aim of reducing false classifications and improving the robustness of dd-cfDNA interpretation [10].
The aim of this study was therefore to determine whether, and how, dd-cfDNA measures including the combined CM score correlate with the newly developed histopathology-based continuous rejection indices [1] and to identify which Banff lesion patterns are the principal drivers of dd-cfDNA release.
Methods
To this end, we analyzed a cohort of 249 indication biopsies from two previously published independent cohorts [11, 12]. Detailed descriptions of the patient populations have been reported previously. For each biopsy, histopathological diagnoses were assigned according to Banff criteria [13], and continuous rejection indices for antibody-mediated rejection/microvascular inflammation (AMR/MVI), T-cell-mediated rejection/tubulointerstitial inflammation (TCMR/TI), global activity, and chronicity were calculated as described by Vaulet et al. [1].
dd-cfDNA was analyzed as relative percentage of total cfDNA, as absolute copies per milliliter (cp/mL), and using the previously developed combined CM score integrating relative and absolute dd-cfDNA measurements [10]. Descriptive analyses were performed by histopathological diagnostic category, with indices and dd-cfDNA measures summarized as means ± standard deviations.
To assess associations between individual Banff lesion scores and cfDNA measures, Spearman rank correlation coefficients (ρ) were calculated between each lesion score and dd-cfDNA metrics (percentage, cp/mL, CM score) as well as total cfDNA. Lesions were grouped according to AMR-related, TCMR-related, and chronic injury categories [13].
To evaluate how well dd-cfDNA quantitatively tracks the continuous rejection indices across their dynamic range, interclass correlation analyses were performed. For each dd-cfDNA metric, biopsies were stratified into five equal-sized quintiles. Mean values and standard errors of dd-cfDNA measures and histopathology-derived indices were calculated within each quintile, and linear regression between quintile means was used to derive coefficients of determination (R2) and corresponding P-values. Overall explanatory association was compared across dd-cfDNA metrics by summing R2 values across indices.
Because the endpoints were continuous indices and ordinal lesion grades, Spearman’s ρ and R2 were used as descriptive measures of association. Our analyses were not intended to assess diagnostic discrimination or calibration for binary outcomes.
Results
Across all dd-cfDNA measures, the highest dd-cfDNA levels were observed in biopsies with high histopathological activity indices (Table 1). Biopsies characterized by MVI, including AMR, mixed rejection, and DSA-negative/C4d-negative MVI, showed the highest AMR/MVI and global activity index values and correspondingly elevated dd-cfDNA levels. Biopsies with TCMR displayed elevated TCMR/TI and activity indices but lower and more variable dd-cfDNA levels. Borderline changes were associated with moderately increased TCMR/TI and activity indices, falling between overt TCMR and non-rejection. Categories characterized by chronic or non-immune-mediated injury, such as interstitial fibrosis/tubular atrophy (IFTA), glomerulonephritis (GN), or calcineurin inhibitor (CNI) toxicity, exhibited low dd-cfDNA and low activity despite variable chronicity indices.
TABLE 1
| Histopathological Diagnosis | N | AMR/MVI [index] | TCMR/TI [index] | Activity [index] | Chronicity [index] | dd-cfDNA [%] | dd-cfDNA [cp/mL] | dd-cfDNA [CM-Score] |
|---|---|---|---|---|---|---|---|---|
| AMR | 46 | 5.04 ± 2.28 | 0.95 ± 0.85 | 4.85 ± 2 | 7.07 ± 3.52 | 2.01 ± 1.48 | 96.4 ± 99.8 | 0.46 ± 0.63 |
| TCMR | 16 | 0.43 ± 0.79 | 3.57 ± 1.25 | 4.56 ± 2.1 | 3.25 ± 2.65 | 0.56 ± 0.39 | 53.4 ± 84.4 | −0.14 ± 0.52 |
| Mixed Rej | 8 | 5.29 ± 0.96 | 4.47 ± 1.27 | 9.13 ± 1.64 | 6.13 ± 3 | 1.82 ± 1.11 | 153.6 ± 111.6 | 0.73 ± 0.47 |
| DSA- MVI | 9 | 5.34 ± 2.17 | 1.31 ± 1.04 | 5.78 ± 2.22 | 6.22 ± 3.9 | 1.81 ± 1.41 | 113.2 ± 101.9 | 0.5 ± 0.63 |
| Borderline | 20 | 0.6 ± 0.86 | 1.87 ± 0.23 | 2.7 ± 0.73 | 2.85 ± 2.54 | 0.65 ± 0.54 | 46.1 ± 66.1 | −0.16 ± 0.51 |
| IFTA | 46 | 0.34 ± 0.59 | 0.64 ± 0.63 | 0.8 ± 1.09 | 3.61 ± 1.95 | 0.41 ± 0.52 | 21.4 ± 28.4 | −0.4 ± 0.33 |
| GN | 20 | 0.57 ± 1.18 | 1.23 ± 1.2 | 1.5 ± 1.57 | 5.1 ± 2.83 | 0.33 ± 0.2 | 14.5 ± 9.6 | −0.46 ± 0.15 |
| CNI Tox | 20 | 0.47 ± 0.9 | 0.33 ± 0.37 | 0.35 ± 0.67 | 3.85 ± 2.46 | 0.58 ± 0.59 | 14.8 ± 11.2 | −0.38 ± 0.27 |
| UTI | 6 | 1.16 ± 1.34 | 2.66 ± 1.85 | 3.67 ± 3.27 | 5.83 ± 3.54 | 0.51 ± 0.32 | 14.2 ± 6.3 | −0.38 ± 0.16 |
| BKVAN | 19 | 0.89 ± 1.2 | 2.72 ± 1.6 | 4.11 ± 2.54 | 2.84 ± 2.06 | 0.56 ± 1.04 | 56.4 ± 88.7 | −0.2 ± 0.68 |
| Normal | 14 | 0.16 ± 0.58 | 0.26 ± 0.47 | 0.21 ± 0.43 | 1.29 ± 2.37 | 0.6 ± 1.02 | 29.8 ± 47.7 | −0.31 ± 0.5 |
| ATI | 8 | 0.1 ± 0.27 | 0.36 ± 0.38 | 0.38 ± 0.52 | 1.25 ± 1.04 | 0.87 ± 1.04 | 92.3 ± 101.7 | 0.12 ± 0.77 |
| Other | 17 | 0.45 ± 1.0 | 1.07 ± 1.61 | 1.35 ± 1.84 | 2.71 ± 2.47 | 0.32 ± 0.11 | 37.1 ± 49.5 | −0.29 ± 0.31 |
Indices and dd-cfDNA by histopathology.
Averages and standard deviations are given. AMR, antibody-mediated rejection; ATI, acute tubular injury; BKVAN, BK polyomavirus nephropathy; CM-score, combined model score; CNI Tox, calcineurin inhibitor toxicity; dd-cfDNA, donor-derived cell-free DNA; DSA, donor-specific antibodies; GN, glomerulonephritis; IFTA, interstitial fibrosis and tubular atrophy; MVI, microvascular inflammation; Rej, rejection; TCMR, T-cell-mediated rejection; TI, tubulointerstitial inflammation; UTI, urinary tract infection.
To further assess the contribution of individual Banff lesions to cfDNA release, Spearman correlation coefficients were calculated between each lesion score and dd-cfDNA measured as percentage, copies/mL, and combined CM score, as well as total cfDNA (Figure 1). Lesions associated with AMR showed the highest correlation coefficients with dd-cfDNA measures, with glomerulitis (g) and peritubular capillaritis (ptc) demonstrating the largest coefficients. For TCMR lesions, correlations with dd-cfDNA percentage were lower, whereas higher coefficients were observed with absolute dd-cfDNA copies/mL and total cfDNA. Chronic lesions showed lower correlation coefficients overall with dd-cfDNA: among chronic Banff lesions, only transplant glomerulopathy (cg) and, to a lesser extent, vascular fibrous intimal thickening (cv) were associated with increased dd-cfDNA levels, whereas interstitial fibrosis (ci) and tubular atrophy (ct) showed no positive association or an inverse relationship.
FIGURE 1
Spearman rank correlations between individual Banff lesion scores and circulating cfDNA measures in 249 kidney transplant biopsies. Correlations are shown for donor-derived cell-free DNA (dd-cfDNA) expressed as combined continuous model (CM) score, percentage, and copies/mL, as well as total cfDNA. dd-cfDNA shows strongest associations with microvascular inflammation lesions (g, ptc), whereas correlations with tubulointerstitial and chronic lesions are weaker. Dashed lines indicate thresholds for moderate and strong correlations; statistical significance is indicated as P < 0.05 (*), P < 0.01 (**), and P < 0.001 (***).
Building on this descriptive alignment, we next assessed how well dd-cfDNA quantitatively tracks the newly developed indices across their dynamic range. We therefore performed interclass (IC) correlations between the histopathology indices and the dd-cfDNA measures, with classes defined by quintiles (20% intervals) of the dd-cfDNA levels in the cohort. Across all dd-cfDNA measurements, the highest R2 in the IC correlations were observed for the AMR/MVI and activity indices (Figure 2). In contrast, R2 values for the TCMR/TI and chronicity indices were lower (Figure 2). Among the dd-cfDNA metrics, the combination score, which simultaneously considers percent and absolute dd-cfDNA, showed the highest overall association with the indices, with a sum R2 of 3.4 across all indices, followed by absolute dd-cfDNA levels in cp/mL (2.9) and relative dd-cfDNA percentages (2.6) (Figure 2). Supplementary Figure S1 shows the individual-biopsy relationships between each cfDNA measure and the histopathology-derived continuous rejection indices.
FIGURE 2
Interclass correlations between continuous indices of kidney transplant rejection and circulating cfDNA measures in 249 kidney transplant biopsies. Biopsies were stratified into quintiles based on the four cell-free DNA (cfDNA) measures, including the combined continuous model (CM) score, donor-derived cell-free DNA (dd-cfDNA) percentage, dd-cfDNA copies/mL, and total cfDNA (rows, top to bottom). Mean values of (dd-)cfDNA and histopathology-derived indices (AMR/MVI, TCMR/TI, activity, chronicity) are shown across quintiles, with linear regression used to derive R2 and P-values. The CM score showed the highest overall explanatory power compared with absolute and relative (dd-)cfDNA measures. Background colors indicate statistical significance.
Discussion
In this study, we evaluate the newly developed histopathology-derived continuous indices [1] in an independent cohort of clinically indicated biopsies and demonstrate how dd-cfDNA aligns with the inflammatory component of the rejection continuum. Together, our findings suggest that dd-cfDNA may serve as a non-invasive readout of the acuity of graft injury within the continuous rejection spectrum.
As expected, biopsies with MVI as seen in AMR, mixed rejection, and DSAnegC4dneg MVI showed the highest AMR/MVI and activity index values, along with the highest dd-cfDNA levels, highlighting once again the utility of dd-cfDNA, particularly in AMR contexts [3–5, 10, 12, 14–16]. Consistently, correlation analyses demonstrated strong associations between AMR lesions, notably g and ptc, and dd-cfDNA, as demonstrated previously [2, 12].
In contrast, dd-cfDNA levels were lower and more heterogenous in TCMR biopsies, a pattern that has been reported previously across multiple studies [2, 3, 5, 10–12, 14]. For TCMR-associated lesions, total cfDNA showed a strong influence on dd-cfDNA measurements (Figure 1). As total cfDNA constitutes the denominator of the percentage-based metric, this provides a plausible explanation for the limited increase in relative dd-cfDNA often observed in TCMR. In contrast, absolute dd-cfDNA expressed as copies per milliliter is not affected by recipient-derived cfDNA and therefore showed a stronger association with TCMR-related injury (Figure 2). These observations are strikingly consistent with the limited association between the TCMR/TI index and dd-cfDNA when expressed as a percentage (Figure 2). A potential biological explanation is that infiltrating recipient immune cells undergo cell death within the graft during TCMR, contributing substantially to circulating total cfDNA levels and thereby diluting the relative dd-cfDNA fraction when expressed as a percentage. Independently of this effect on total cfDNA, variability in dd-cfDNA measures among TCMR biopsies may also reflect biological differences in injury mechanisms, whereby some cases exhibit interstitial inflammation accompanied by subtle microvascular injury, leading to increased dd-cfDNA, whereas others are dominated by tubular injury that may preferentially drain into the urine and therefore not be reflected in circulating cfDNA. Notably, in our cohort, seven TCMR cases were grade Ia, six were grade Ib, and none were grade III, resulting in a 25% lower mean of the TCMR/TI index compared to the cohort described by Vaulet et al. [1]. This limited dynamic range likely contributes to the weaker alignment of dd-cfDNA with the TCMR/TI index and to mean CM scores near (and slightly below) zero, rather than clearly positive values. Nonetheless, our findings for the first time provide a plausible mechanistic explanation for why low-grade TCMR is difficult to detect when dd-cfDNA is expressed as a percentage alone.
As expected, patients with Borderline changes had only moderately elevated TCMR/TI and activity indices, which were lower than those with overt TCMR. This is consistent with previous findings indicating that approximately 2/3 of Borderline cases have a molecular phenotype compatible with no rejection [17]. In such ambiguous cases, histological assessment and biomarkers may complement one another, with dd-cfDNA providing additional support in distinguishing true rejection with ongoing graft injury from innocuous infiltrates, indicating which patients might profit from intensified surveillance [11]. Whether dd-cfDNA can distinguish Borderline lesions that require treatment from those that are clinically benign, however, remains speculative and will require confirmation in larger cohorts of borderline cases together with molecular phenotyping.
The consistently low dd-cfDNA levels observed in chronic and non-immune-mediated injury support the concept that dd-cfDNA primarily reflects acute inflammatory injury rather than chronic structural damage. Among chronic Banff lesions, only transplant glomerulopathy (cg) and, to a lesser extent, vascular fibrous intimal thickening (cv), both reflecting chronic vascular and endothelial injury, were associated with increased dd-cfDNA levels, whereas interstitial fibrosis (ci) and tubular atrophy (ct), representing largely scarring, showed no positive association or an inverse relationship. Taken together, chronic graft injury represents a relatively weak driver of dd-cfDNA release, consistent with the findings observed for the chronicity index (Figure 2).
Thus, our dd-cfDNA data, and in particular the novel combined CM score (integrating the percentage and absolute dd-cfDNA) in relation to the rejection-continuum indices help integrate tissue-based and blood-based measures: whereas the continuous histology-based indices quantify the tissue phenotype of inflammation and injury, dd-cfDNA captures the real-time acuity of graft damage in the blood and may further support longitudinal monitoring of clinical response, as indicated previously [11, 18]. Despite the limited dataset, the overall patterns seem consistent with the biological behavior of dd-cfDNA, which reflects acute inflammation that causes cellular injury and death with subsequent release of dd-cfDNA but remains low in chronic structural damage. This supports the concept of a continuum of rejection activity, with microvascular inflammation representing the high-acuity end of the spectrum.
The results presented herein add to the scientific understanding of dd-cfDNA in relation to the specific lesions described by light microscopy and complement the biopsy-derived framework introduced by the publication of Vaulet et al. [1]. Particularly the increase of total cfDNA in TCMR is a novel observation, which offers an explanation, why low grade TCMR is notoriously hard to detect with dd-cfDNA when expressed as percentage only.
Our study has several limitations: (i) the study population is restricted to indication biopsies and does not include protocol biopsies, which may introduce selection bias and limits generalizability to low-pretest-probability surveillance settings and subclinical phenotypes; (ii) both cohorts are European, which constrains external validity across different populations and clinical practices; (iii) TCMR cases were predominantly low grade, which limits inference regarding cfDNA-index relationships in advanced/high-grade (“full-blown”) TCMR; (iv) the combined CM score, while applied here as a pre-specified composite integrating dd-cfDNA percentage and copies/mL, still lacks independent external validation in prospective clinical cohorts; and (v) because our endpoints were the newly proposed continuous rejection indices and Banff ordinal lesion grades, the reported Spearman and R2 metrics should be interpreted as descriptive measures of association and not as stand-alone diagnostic performance assessments.
Taken together, these invasive and non-invasive measures may contribute to a unified framework in which histology and circulating biomarkers (dd-cfDNA) converge to quantify active injury, track its trajectory, and potentially aid clinicians to more effectively treat graft inflammation with active injury patterns.
Statements
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Ethics statement
The studies involving humans were approved by the Ethics Committee of the Medical Faculty of Heidelberg and by the Ethics Committee of Charité – Universitätsmedizin Berlin, respectively. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.
Author contributions
LB, ES, and KB designed the study. LB, JB, ES, and KB analyzed and interpreted the data and drafted the manuscript. Patients were recruited and results generated by LB, AA, BO, CM, and KB. JB and ES established and performed the quantification of dd-cfDNA in their laboratory. CM and KB supervised the project and revised the manuscript. All authors contributed to the article and approved the submitted version.
Funding
The author(s) declared that financial support was received for this work and/or its publication. The authors declare that this study received in-kind support from Insight Molecular Diagnostics Inc. (iMDx; provision of dd-cfDNA kits/reagents and access to testing infrastructure).
Acknowledgments
The graphical abstract has been designed using resources from Flaticon.com.
Conflict of interest
Insight Molecular Diagnostics Inc. (iMDx; GraftAssure) provided kits and reagents for dd-cfDNA assessment and supplied the testing infrastructure used to generate study results. AA received travel support from Insight Molecular Diagnostics Inc. KB acts as a consultant to Insight Molecular Diagnostics Inc. and has received research funds and/or honoraria from: Aicuris, Alexion, Astellas, AstraZeneca, Biogen, Biohope, Carealytics, CareDx, Chiesi, CSL Behring, DTB GmbH, Eledon, Hansa, HiBio, iMDx, MSD, Natera, Neovii, Oncocyte, Oska, Otsuka, Paladin, Pfizer, Pirche, Sanofi, smart care solutions, Stada, Takeda, Veloxis, Vifor and Xenothera. JB was employed by Chronix Biomedical GmbH (an Insight Molecular Diagnostics Inc. company). ES was employed by Insight Molecular Diagnostics Inc., which holds intellectual property rights (EP 3004388B, EP3201361B1, US11155872B2, US10570443B2). The Department of Nephrology, University Hospital Heidelberg (LB and CM) has received kits and supplies for dd-cfDNA testing from CareDx and from Insight Molecular Diagnostics Inc. CM is co-founder and CSO of TolerogenixX GmbH (Heidelberg, Germany), a biotechnology company that holds licenses and patents for MIC treatment, and has received honoraria from AstraZeneca, Boehringer, Hansa and Novartis.
The remaining author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The authors declare that this study received in-kind support from Insight Molecular Diagnostics Inc. (iMDx; provision of dd-cfDNA kits/reagents and access to testing infrastructure). The study was investigator-initiated and investigator-led. iMDx/Chronix personnel contributed to aspects of data analysis, interpretation, and manuscript drafting. However, the academic investigators retained responsibility for the study design, overall data interpretation, and the decision to submit the manuscript for publication.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontierspartnerships.org/articles/10.3389/ti.2026.16099/full#supplementary-material
References
1.
VauletTKoshyPWellekensKAubertOBottomleyCCallemeynJet alContinuous Indices to Assess the Phenotypic Spectrum of Kidney Transplant Rejection. Nat Commun (2025) 16(1):10417. 10.1038/s41467-025-65153-9
2.
AubertOUrsule-DufaitCBrousseRGueguenJRacapéMRaynaudMet alCell-Free DNA for the Detection of Kidney Allograft Rejection. Nat Med (2024) 30(8):2320–7. 10.1038/s41591-024-03087-3
3.
BuddeKOsmanodjaBSchrezenmeierEHalleckFMayerKABöhmigGAet alClinical Utility of Donor-Derived Cell-Free DNA Testing for Kidney Transplant Monitoring in Selected Patients. Kidney Int (2025) 107(5):792–6. 10.1016/j.kint.2024.09.023
4.
NaesensMWongG. Simply Measuring what Matters—Toward Clinical Utility of donor-derived Cell-Free DNA Testing?Kidney Int (2025) 107(5):799–805. 10.1016/j.kint.2025.01.039
5.
SeifNBloomRD. Using donor-Derived Cell-Free DNA in Kidney Transplant Patients: This Jury Needs More Evidence. Kidney Int (2025) 107(5):796–9. 10.1016/j.kint.2024.10.029
6.
PagliazziABestardONaesensM. Donor-Derived Cell-Free DNA: Attractive Biomarker Seeks a Context of Use. Transpl Int (2023) 36:12406. 10.3389/ti.2023.12406
7.
OellerichMSherwoodKKeownPSchützEBeckJStegbauerJet alLiquid Biopsies: Donor-Derived Cell-Free DNA for the Detection of Kidney Allograft Injury. Nat Rev Nephrol (2021) 17(9):591–603. 10.1038/s41581-021-00428-0
8.
OsmanodjaBAkifovaABuddeKChoiMOellerichMSchützEet alAbsolute or Relative Quantification of Donor-Derived Cell-Free DNA in Kidney Transplant Recipients: Case Series. Transpl Direct (2021) 7(11):e778. 10.1097/txd.0000000000001237
9.
OellerichMBuddeKOsmanodjaBBornemann-KolatzkiKBeckJSchützEet alDonor-Derived Cell-Free DNA as a Diagnostic Tool in Transplantation. Front Genet (2022) 13:1031894. 10.3389/fgene.2022.1031894
10.
BenningLAkifovaAOellerichMOsmanodjaBMorathCBeckJet alImproving diagnostic performance of kidney allograft rejection with a model combining relative fraction and absolute copies of donor-derived cell-free DNA: results from five independent cohorts [Preprint]. medRxiv. (2025). 10.64898/2025.12.01.25341374
11.
BenningLMorathCFinkARudekMSpeerCKälbleFet alDonor-Derived Cell-Free DNA (dd-cfDNA) in Kidney Transplant Recipients with Indication Biopsy—Results of a Prospective Single-Center Trial. Transpl Int (2023) 36:11899. 10.3389/ti.2023.11899
12.
AkifovaABuddeKChoiMAmannKBuettner-HeroldMOellerichMet alAssociation of Blood Donor-Derived Cell-Free DNA Levels with Banff Scores and Histopathological Lesions in Kidney Allograft Biopsies: Results from an Observational Study. Transpl Direct (2025) 11(5):e1794. 10.1097/txd.0000000000001794
13.
NaesensMRoufosseCHaasMLefaucheurCMannonRBAdamBAet alThe Banff 2022 Kidney Meeting Report: Reappraisal of Microvascular Inflammation and the Role of Biopsy-Based Transcript Diagnostics. Am J Transpl (2024) 24(3):338–49. 10.1016/j.ajt.2023.10.016
14.
AkifovaABuddeKOellerichMBeckJBornemann-KolatzkiKSchützEet alPerspective for Donor-Derived Cell-Free DNA in Antibody-Mediated Rejection After Kidney Transplantation: Defining Context of Use and Clinical Implications. Transpl Int (2024) 37:13239. 10.3389/ti.2024.13239
15.
BenningLBestardO. Shedding Light on Microvascular Inflammation: Understanding Outcomes, but what Sparks the Flame?Transpl Int (2024) 37:14032. 10.3389/ti.2024.14032
16.
AkifovaABuddeKAmannKBuettner-HeroldMChoiMOellerichMet alDonor-Derived Cell-Free DNA Monitoring for Early Diagnosis of Antibody-Mediated Rejection After Kidney Transplantation: A Randomized Trial. Nephrol Dial Transpl (2024) 40(7):1384–95. 10.1093/ndt/gfae282
17.
FreitasDGdeSellarésJMengelMChangJHidalgoLGFamulskiKSet alThe Nature of Biopsies with “Borderline Rejection” and Prospects for Eliminating This Category. Am J Transpl (2012) 12(1):191–201. 10.1111/j.1600-6143.2011.03784.x
18.
MayerKASchrezenmeierEDieboldMHalloranPFSchatzlMSchranzSet alA Randomized Phase 2 Trial of Felzartamab in Antibody-Mediated Rejection. N Engl J Med (2024) 391(2):122–32. 10.1056/nejmoa2400763
Summary
Keywords
dd-cfDNA, donor-derived cell-free DNA, kidney transplantation, rejection, rejection continuum
Citation
Benning L, Akifova A, Osmanodja B, Morath C, Beck J, Schütz E and Budde K (2026) Donor-Derived Cell-Free DNA as a Non-Invasive Readout of Activity Across the Rejection Continuum. Transpl. Int. 39:16099. doi: 10.3389/ti.2026.16099
Received
19 December 2025
Revised
05 February 2026
Accepted
12 February 2026
Published
24 February 2026
Volume
39 - 2026
Updates
Copyright
© 2026 Benning, Akifova, Osmanodja, Morath, Beck, Schütz and Budde.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Klemens Budde, klemens.budde@charite.de
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.