Persistently Elevated Donor-Derived Cell-Free DNA After Kidney Transplantation in Atypical Hemolytic Uremic Syndrome: A Preliminary Observational Study

1. Introduction

1.1. Donor-Derived Cell-Free DNA: A Promising Non-Invasive Biomarker

Kidney transplantation remains the preferred treatment for patients with end-stage renal disease, offering improved survival and quality of life compared with dialysis [1]. Despite advances in immunosuppression and post-transplant care, allograft rejection continues to threaten long-term graft survival and function. Acute rejection may lead to irreversible graft injury if not identified and treated promptly [2].

The current diagnostic reference standard for rejection is histopathological evaluation of kidney allograft biopsy specimens. Although biopsies provide essential diagnostic information, they are invasive, carry procedural risks such as bleeding or infection, and may be affected by sampling error and interobserver variability [3,4]. These limitations have stimulated the search for safer, repeatable, and more objective diagnostic adjuncts.

One of the most promising approaches is donor-derived cell-free DNA, a fragment of DNA released into the recipient circulation as a result of allograft cell injury and turnover. Cell-free DNA is generated predominantly during apoptosis and necrosis, and its presence in circulation may reflect tissue injury. In transplant recipients, the proportion of cfDNA originating from donor tissue provides a quantitative measure of allograft-derived injury [5].

Numerous studies have demonstrated an association between elevated dd-cfDNA and episodes of acute rejection, with increased levels indicating heightened cell turnover and injury in the transplanted organ [6,7]. Before detecting changes in conventional clinical markers like serum creatinine or proteinuria, dd-cfDNA levels may rise, potentially allowing for earlier clinical intervention [8]. This has positioned dd-cfDNA as a useful adjunct in post-transplant surveillance and in the assessment of immunological risk [9].

Beyond rejection, dd-cfDNA has also shown promise in identifying other forms of graft injury, including BK virus nephropathy and ischemia-reperfusion injury, although specificity remains a major concern. Overall, dd-cfDNA offers an appealing, non-invasive tool for post-transplant monitoring, yet important diagnostic caveats remain, particularly in recipients with complex comorbidities or rare conditions such as atypical hemolytic uremic syndrome.

1.2. Diagnostic Challenges of dd-cfDNA in Kidney Transplant Recipients with aHUS

Atypical hemolytic uremic syndrome is a rare, life-threatening disease characterized by complement dysregulation leading to thrombotic microangiopathy, endothelial damage, and multiorgan injury, with the kidneys being among the most commonly affected organs [10,11]. Patients with aHUS frequently progress to end-stage renal disease and may require kidney transplantation. However, transplant outcomes in this population remain complex because of the risk of disease recurrence and persistent microvascular injury after transplantation [12].

These pathophysiological mechanisms pose a unique challenge for the interpretation of dd-cfDNA. Chronic, recurrent, or subclinical endothelial injury associated with aHUS, even in the absence of overt alloimmune activation, may independently enhance graft-derived cell injury and the release of dd-cfDNA. This raises concerns about the specificity of dd-cfDNA as a marker of rejection in this context, because elevated values may not necessarily reflect immune-mediated allograft injury but may instead reflect complement-mediated endothelial injury and associated cell turnover [13].

1.3. Aim of the Study

This study sought to address this knowledge gap by examining the dynamics of plasma dd-cfDNA after kidney transplantation in recipients with and without aHUS. By evaluating the relationship between dd-cfDNA levels, DSA status, serum creatinine, and clinical course, this study aimed to assess the reliability and limitations of dd-cfDNA as a biomarker of graft injury in the specific context of aHUS.

2. Results

Recipient-level peri-transplant characteristics are presented in Table 1. Non-aHUS and aHUS recipients are shown as separate patient-level rows.

Overall, maintenance immunosuppression was based on corticosteroids, tacrolimus, and mycophenolate mofetil in all recipients. Delayed graft function was recorded in one non-aHUS recipient and in one aHUS recipient.

Table 2. Cohort characteristics and testing schedule in kidney transplant recipients with and without aHUS. Available cohort and testing characteristics are summarized in Table 2.

Variable	Non-aHUS recipients	aHUS recipients
Number of recipients	12	2
First kidney transplantation	12/12	2/2
Study period	April–December 2022	April–December 2022
Plasma collection time points	D7, D14, M1, M2	D7, D14, M1, M2
Clinical graft status during observation	Stable renal function during follow-up	Preserved graft function during follow-up
Clinically documented acute rejection	None documented	None documented
Protocol biopsies	Not performed systematically	Not performed systematically
DSA detected during follow-up	0/12	0/2
dd-cfDNA assay	AlloSeq cfDNA, NGS	AlloSeq cfDNA, NGS
DSA assay	LABScreen Single Antigen Beads	LABScreen Single Antigen Beads

Table 2. Cohort characteristics and testing schedule in kidney transplant recipients with and without aHUS. Available cohort and testing characteristics are summarized in Table 2.

Variable	Non-aHUS recipients	aHUS recipients
Number of recipients	12	2
First kidney transplantation	12/12	2/2
Study period	April–December 2022	April–December 2022
Plasma collection time points	D7, D14, M1, M2	D7, D14, M1, M2
Clinical graft status during observation	Stable renal function during follow-up	Preserved graft function during follow-up
Clinically documented acute rejection	None documented	None documented
Protocol biopsies	Not performed systematically	Not performed systematically
DSA detected during follow-up	0/12	0/2
dd-cfDNA assay	AlloSeq cfDNA, NGS	AlloSeq cfDNA, NGS
DSA assay	LABScreen Single Antigen Beads	LABScreen Single Antigen Beads

In non-aHUS kidney transplant recipients, median %dd-cfDNA values showed a gradual decline over time, ranging from 0.72% [0.28%, 2.33%] at day 7 to 0.12% [0.05%, 0.34%] at M1, consistent with stabilization of graft function. In contrast, recipients with aHUS demonstrated persistently elevated dd-cfDNA levels, with median values fluctuating around 1% [0.88%, 1.14%] during the first month of follow-up (Figure 1).

No donor-specific anti-HLA antibodies were detected in any patient throughout the observation period. No apparent correlation was observed between %dd-cfDNA levels and serum creatinine levels across the cohort. Given the small sample size and repeated measurements within individuals, this observation should be interpreted descriptively rather than as definitive evidence of independence between these markers.

3. Discussion

This study examined early dd-cfDNA kinetics in kidney transplant recipients with and without aHUS. In non-aHUS recipients, dd-cfDNA levels declined within the first month after transplantation, consistent with recovery from perioperative injury and stabilization of graft function. By contrast, recipients with aHUS showed persistently elevated dd-cfDNA around the commonly used 1% threshold despite the absence of detectable DSA, absence of clinically documented rejection, and preserved graft function during follow-up.

These findings raise important questions regarding the interpretation of dd-cfDNA in the context of aHUS. Previous studies in broader kidney transplant populations have consistently shown that dd-cfDNA levels around or above 1% are associated with acute rejection and poorer graft outcomes [6,7,8]. However, our data suggest that in aHUS recipients, elevated dd-cfDNA may not carry the same predictive meaning as in unselected kidney transplant populations. Instead, the observed elevations may reflect ongoing endothelial injury and cell turnover related to underlying complement-mediated pathology, rather than exclusively alloimmune-mediated graft damage [10,11,12,13]. This highlights a potential limitation of applying universal dd-cfDNA thresholds across biologically diverse transplant populations.

The role of dd-cfDNA as a biomarker of allograft rejection has been progressively refined since its initial clinical validation [5,14]. A 1% threshold discriminates active rejection with high negative but limited positive predictive value, reflecting that dd-cfDNA rises with any source of graft injury rather than with rejection specifically [5]. This signal is strongest for antibody-mediated rejection, in which microvascular inflammation produces greater endothelial DNA release than the predominantly tubulointerstitial injury of T-cell-mediated rejection [9,15]. A recent meta-analysis confirmed higher pooled diagnostic accuracy for antibody-mediated than for any-cause rejection [8]. Even low-grade or borderline T-cell-mediated rejection may, however, be accompanied by elevated dd-cfDNA and an increased risk of subsequent graft injury [16], underscoring that the biomarker reports tissue injury along a continuum rather than a binary rejection state.

Beyond a single cross-sectional measurement, the temporal behavior of dd-cfDNA carries additional prognostic information. In the longitudinal ADMIRAL cohort, persistently elevated values—rather than isolated peaks—predicted future de novo donor-specific antibodies and a clinically meaningful decline in glomerular filtration rate, and elevations may precede biopsy-proven rejection by several weeks [17,18]. Surveillance studies using the same next-generation sequencing assay applied here have similarly linked persistent early post-transplant elevation to non-improving graft function [19]. This kinetic perspective is directly relevant to the present cohort: a transient peri-operative rise that resolves is expected, whereas values that remain elevated beyond the first month are more concerning. In aHUS, ongoing complement-mediated endothelial injury can sustain dd-cfDNA release even without rejection, complicating this phenomenon. Distinguishing persistent elevation due to alloimmune injury from that due to underlying microvascular disease is, therefore, the central interpretive challenge in this population and one that a fractional threshold alone cannot resolve.

The lack of published information on dd-cfDNA changes in aHUS patients makes it harder to understand the clinical situation. It remains unclear whether the biomarker behaves similarly to other transplant populations or whether elevated dd-cfDNA in aHUS reflects baseline vascular injury, recurrent or subclinical thrombotic microangiopathy, subclinical rejection, or overlapping mechanisms. This knowledge gap calls for studies specifically addressing rare transplant populations, because universal dd-cfDNA thresholds may not be appropriate in all clinical contexts.

None of the patients in our cohort developed detectable DSA during follow-up, and no apparent correlation was observed between dd-cfDNA and serum creatinine. In a cohort without documented rejection and with stable graft function, the absence of such a correlation is expected and should be interpreted cautiously rather than as positive evidence that dd-cfDNA captures occult injury. This disconnect demonstrates the complexity of graft monitoring in aHUS recipients. Serum creatinine is an insensitive and delayed marker of injury, whereas dd-cfDNA may capture biological processes not reflected in routine biochemical testing. Importantly, recent evidence shows that antibody-mediated rejection can occur even in the absence of detectable DSA, with DSA-negative antibody-mediated rejection sharing molecular signatures and dd-cfDNA release patterns with DSA-positive cases [20]. These findings emphasize that a negative DSA result does not fully exclude alloimmune injury.

In the context of aHUS, persistently elevated dd-cfDNA without detectable DSA may reflect non-alloimmune endothelial injury, low-grade immune activation below the threshold of serological detection, or a combination of both. Accordingly, dd-cfDNA should not be interpreted as a stand-alone diagnostic test in this setting. A multimodal approach integrating DSA, graft function, proteinuria, infection status, hemolysis and thrombotic microangiopathy markers, complement activation markers, and biopsy findings when clinically indicated is likely to provide a more reliable assessment.

Our study adds to a limited body of literature addressing dd-cfDNA kinetics in rare transplant populations. The persistently elevated levels observed in aHUS recipients highlight the need for more nuanced interpretation of this biomarker in the setting of chronic microvascular injury. Future research should investigate disease-specific thresholds or integrate dd-cfDNA with complementary biomarkers, such as complement activation products, endothelial injury markers, and gene expression profiling. Such approaches may help disentangle alloimmune from non-alloimmune injury and improve risk stratification in complex patient groups.

This study has important limitations. First, only two aHUS recipients were included; therefore, the findings should be interpreted as hypothesis-generating. Second, the study was retrospective and single-centre. Third, protocol biopsies and molecular biopsy assessment were not systematically available, so subclinical rejection could not be definitively excluded. Fourth, detailed markers of complement activation and thrombotic microangiopathy were not available for integrated analysis. Fifth, dd-cfDNA was analyzed as a fractional value; absolute dd-cfDNA quantification may provide additional information in future studies. Finally, follow-up was limited to the early post-transplant period, and longer-term graft outcomes were not assessed.

4. Materials and Methods

This retrospective observational study included 14 kidney transplant recipients: 12 non-aHUS recipients with stable renal function and 2 recipients with aHUS. The aHUS subgroup consisted of two female kidney transplant recipients aged between 20 and 40 years. Both patients received a kidney transplant from a deceased donor. In one patient, aHUS was the primary cause of kidney failure, whereas in the other patient, aHUS was considered most likely secondary to primary focal segmental glomerulosclerosis.

In the first patient, genetic testing performed as part of the diagnostic work-up for aHUS revealed a heterozygous pathogenic variant in the CFH gene. In addition, the C3 GGTA haplotype, which has been associated with dense deposit disease, was identified. The patient also carried the CFH H3 risk haplotype (TGTGT), associated with a less favorable disease course and an increased risk of kidney allograft loss, as well as one copy of the MCP/CD46 aHUS risk haplotype (GGAAC). Genetic analysis also demonstrated heterozygous ADAMTS13 variants associated with familial thrombotic thrombocytopenic purpura.

In the second patient, genetic testing revealed heterozygous CD46 variants c.946+23G>T and c.*783T>C. The role of the c.946+23G>T variant in modulating disease risk remains uncertain; however, its presence has been associated with alternative splicing of the CD46 transcript, involving selective inclusion of exon 8 and resulting in impaired functionality of the gene product. The c.*783T>C variant has been described as a risk variant for hemolytic uremic syndrome. In addition, the patient carried the CFH H3 risk haplotype (TGTGT), associated with a less favorable disease course and an increased risk of kidney allograft loss.

In both patients, eculizumab therapy was continued after kidney transplantation and was subsequently converted to ravulizumab. No recurrence of aHUS was observed after transplantation in either patient. Both patients received basiliximab induction and standard triple maintenance immunosuppression consisting of corticosteroids, mycophenolate mofetil, and tacrolimus.

All patients underwent their first kidney transplantation in the Department and Clinic of Nephrology, Transplant Medicine and Internal Medicine at Wroclaw Medical University between April and December 2022. Protocol biopsies were not performed systematically as part of this study.

Blood samples were collected at four time points: one week after transplantation (D7), two weeks after transplantation (D14), one month after transplantation (M1), and two months after transplantation (M2). Samples were collected into tubes dedicated to cfDNA stabilization (Cell-Free DNA BCT RUO, Streck, La Vista, NE, USA). cfDNA was isolated from 4 mL of plasma using the MagMAX Cell-Free DNA Isolation Kit (Applied Biosystems, Thermo Fisher Scientific Inc., Waltham, MA, USA), according to the manufacturer's protocol.

cfDNA concentration was measured using fluorescence-based methods with the Quantus Fluorometer (Promega Corporation, Madison, WI, USA; RRID:SCR_026279) and QuantiFluor ONE dsDNA System (Promega Corporation, Madison, WI, USA). cfDNA quality was verified using automated electrophoresis with the 4200 TapeStation System (Agilent, Santa Clara, CA, USA; RRID:SCR_018435) and High Sensitivity D1000 ScreenTape (Agilent, Santa Clara, CA, USA).

dd-cfDNA in kidney transplant recipients was evaluated by next-generation sequencing using the AlloSeq cfDNA Kit (CareDx Inc., Brisbane, CA, USA). Sequencing was performed on a MiSeq instrument using the MiSeq Reagent v3 150-cycle Kit (Illumina Inc., San Diego, CA, USA). Sequencing results were analyzed with AlloSeq cfDNA Software (CareDx).

Donor-specific anti-HLA antibody profiling was performed using LABScreen Single Antigen Beads (One Lambda, Inc., Thermo Fisher Scientific, Canoga Park, CA, USA) according to the manufacturer's instructions. Samples were analyzed on a Luminex FlexMAP 3D platform (Luminex Corporation, Austin, TX, USA), and results were interpreted using HLA Fusion Software, version 4.7 (One Lambda, Inc.).

Donor-specific anti-HLA antibody positivity was defined as the presence of at least one anti-HLA class I and/or class II antibody directed against the corresponding donor HLA antigen. Anti-HLA antibody specificity testing was performed only in samples with a positive screening result for anti-HLA antibodies of the respective HLA class. The final classification of antibody specificity was based on the mean normalized fluorescence intensity value (MFI_norm) obtained for bead populations corresponding to a given serological specificity with a positive or borderline signal.

For anti-HLA class I antibodies, the laboratory-defined cut-off value was MFI_norm = 1100, with a reported measurement uncertainty of 55%. Results were classified as positive when MFI_norm was >=1700, borderline when MFI_norm was >=500 and <1700, and negative when MFI_norm was <500. For anti-HLA class II antibodies, the laboratory-defined cut-off value was MFI_norm = 1250, with a reported measurement uncertainty of 40%. Results were classified as positive when MFI_norm was >=1750, borderline when MFI_norm was >=750 and <1750, and negative when MFI_norm was <750. The cut-off values were established with reference to a non-immunized male population, defined as individuals without a history of transplantation or blood transfusion.

For the purposes of the present analysis, patients were considered DSA-positive when at least one donor-specific anti-HLA antibody met the laboratory criteria for a positive result.

Statistical analyses were conducted using STATISTICA version 13 (TIBCO Software Inc., 2017, Dell, OK, USA; RRID:SCR_014213). Given the small sample size and the exploratory nature of the study, analyses were primarily descriptive. dd-cfDNA values are presented as medians with interquartile ranges where appropriate. Associations between %dd-cfDNA and serum creatinine were assessed descriptively; formal inferential conclusions were avoided because of the limited cohort size.

The study protocol was reviewed and approved by the Bioethical Committee of Wroclaw Medical University (approval no. 37/2022). All participants provided written informed consent before inclusion in the study, in accordance with the Declaration of Helsinki.

An AI-based language model (ChatGPT, OpenAI, San Francisco, CA, USA) was used to assist in language refinement. No AI tool was used for data analysis or generation of scientific conclusions.

The authors made and verified all scientific content, interpretations, and conclusions.

5. Conclusions

In this preliminary observational study, kidney transplant recipients with aHUS demonstrated persistently elevated early post-transplant dd-cfDNA despite the absence of detectable DSA, absence of clinically documented rejection, and preserved graft function during follow-up, whereas non-aHUS recipients showed a decline in dd-cfDNA during early follow-up. These findings suggest that aHUS may represent a biologically distinct setting in which universal dd-cfDNA thresholds require cautious interpretation. Larger biopsy-anchored studies integrating dd-cfDNA with DSA, complement activation markers, endothelial injury biomarkers, and molecular graft assessment are needed to clarify the clinical meaning of dd-cfDNA elevation in this rare transplant population.