Donor-derived cell-free DNA and its utility in kidney transplantation: A myth or a reality

Abstract

Renal allograft rejection and its detection are challenging problems for transplant clinicians. Transplant physicians rely on serum creatinine, estimated glomerular filtration rate, proteinuria, donor-specific antibodies, and graft biopsy to detect rejection. The sensitivity and specificity in these blood and urine tests are low, and the invasiveness of graft biopsy has led transplant clinicians to seek alternative diagnostic tools. Cell-free DNA (cfDNA) is a fragment of DNA released from cell death due to necrosis and apoptosis. Donor-derived cfDNA (dd-cfDNA) has been proposed as a potential non-invasive biomarker for detecting rejection. However, one must interpret it cautiously in conditions such as ischemia-reperfusion injury, delayed graft function, BK virus nephropathy, post-kidney biopsy, and dual kidney transplantation, which may cause dd-cfDNA elevation. There is a lack of standardized cutoff values for diagnosing various types of rejections. Low specificity, higher cost, and lack of universal availability are the multiple obstacles to using this tool. There is a need to establish clinical guidelines for its future utility in early rejection detection, graft surveillance, and tailoring of immunosuppression.

Key Words: Donor-derived cell-free DNA; Rejection; Graft surveillance; Tailoring immunosuppression; Kidney transplantation

Core Tip: Donor-derived cell-free DNA is a biomarker for detecting antibody-mediated rejection. It has a potential role for renal allograft surveillance, assessing the immune system reactivity, and monitoring response to rejection therapy. This minireview will focus on its clinical utilization, method of estimation, quantification, and will provide a way for its integration with clinical parameters and gene profiling for its best utility.

INTRODUCTION

Optimal surveillance of the kidney transplant recipients is of extreme importance. Serum creatinine, drug levels, proteinuria, hematuria, donor-specific antibodies (DSA), and Doppler ultrasound are crude methods for monitoring renal allografts, each with its limitations. Among these, serum creatinine is the most commonly used laboratory test to identify graft dysfunction. Serum creatinine is a late indicator, occurring only when 50% of kidney function is lost[1]. The gold standard is a kidney biopsy, but this has its limitations. Graft biopsy may be complicated by hematuria or other major complications in 1%-3.5% of cases[2], and around one quarter may not have sufficient tissue to facilitate an accurate diagnosis[3]. Over time, interest has grown in non-invasive markers, such as cell-free DNA (cfDNA), a fragment of DNA released into the circulation from dying cells through apoptosis or necrosis. It has been studied to screen for cancers and trisomy 21[4,5]. Donor-derived cfDNA (dd-cfDNA) in the blood of kidney transplant patients has been studied as a noninvasive marker of graft rejection[6,7].

WHAT IS DD-CFDNA AND HOW IS IT PRODUCED?

dd-cfDNA is the DNA fragment released from injured donor cells[7]. DNA resides in the nucleus and, together with histone, forms a nucleosome[8]. The necrosis and apoptosis of cells cause the release of nucleosomes into the bloodstream[9,10]. In circulation, the nucleosome is then cleaved by nucleases into double-stranded DNA. Free DNA is cleared by the liver, spleen, and kidneys in a short period, from 30 minutes to two hours[11,12]. Elevated dd-cfDNA levels indicate injury to the donor cells and may serve as an early marker of rejection.

WHAT IS THE AVAILABLE EVIDENCE AND HOW CAN IT BE UTILIZED IN CLINICAL PRACTICE?

The dd-cfDNA has been utilized in the early diagnosis of rejection and has shown promising results in the early diagnosis of antibody-mediated rejection (ABMR). The dd-cfDNA assay has also helped to guide early biopsy decisions, monitor therapy response to antirejection therapy, and design surveillance strategies in high-risk recipients. Figure 1 shows various clinical utilities of dd-cfDNA.

Figure 1 shows various clinical utilities of donor-derived cell-free DNA. dd-cfDNA: Donor-derived cell-free DNA; ABMR: Antibody-mediated rejection.

Utility in the diagnosis of rejection reactions

dd-cfDNA is a valuable marker for rejection. The DART study (Diagnosing Active Rejection in Kidney Transplant Recipients) was the first breakthrough study that shed light on the role of dd-cfDNA in rejection[13]. A threshold dd-cfDNA level of 1% or above was predictive of T cell-mediated rejection (TCMR) greater than Banff grade 1B or ABMR[13]. Subsequently, single-nucleotide polymorphisms (SNPs) were found to be promising in detecting dd-cfDNA and identifying graft injury and rejection[14]. Most studies and a meta-analysis have shown that dd-cfDNA is more effective at detecting ABMR than TCMR, particularly in borderline or mild rejection[15-17]. Dauber et al[18] used quantitative polymerase chain reaction of insertions and deletions to measure dd-cfDNA and found a threshold of 2.7% could detect TCMR[18]. Another essential utilization of dd-cfDNA is for a specific group of patients with DSA but no histological lesions. The utilization of dd-cfDNA in these patients can predict ABMR[19]. A literature review has identified that dd-cfDNA has a high negative predictive value[13-16], which means that low levels or a negative result reliably exclude rejection. On the other hand, the positive predictive value is low and variable, ranging from 48% to 68.6%[13-16], and these patients may require additional parameters, such as other chemokines or a kidney biopsy. Interestingly, dd-cfDNA has been shown to reduce after treatment of rejection[20]. A meta-analysis conducted by Xing et al[21] yielded interesting findings. They reinforced the previous finding of better diagnostic performance in ABMR, but had limited ability to diagnose TCMR. They found an area under the curve (AUC) of 0.87 with a good sensitivity of 81% and specificity of 80% for ABMR. On the other hand, for TCMR, the AUC was moderate (0.80), but with a low sensitivity of 59%, meaning that many cases of TCMR will not be detected[21]. Various thresholds of dd-cfDNA were recognized in a multicentric prospective study, where a threshold of 2.85% could predict mixed rejection. Similarly, 2.03% could predict acute TCMR. Acute ABMR was predicted by 1.15%, and chronic active ABMR was predicted by 1.09%[22,23]. A literature review of various studies reveals that the threshold for dd-cfDNA varies across studies, with a cutoff of around 1% predictive of ABMR[13,19]. In contrast, a threshold of greater than 2% is required for TCMR[18,21]. Multiple studies have shown correlations between elevated dd-cfDNA and various histological features of ABMR, such as microvascular inflammation and capillaritis[13,15]. Still, it is not yet a substitute for biopsy. It is hoped that its role may be explored further to identify better phenotype-specific thresholds and integration with histological and molecular diagnostics.

Although we have sufficient evidence so far, the transplant community needs to identify the AUC, the mean threshold level with reasonable sensitivity, specificity, and predictive values to utilize cc-dfDNA. A universal validation is also required to utilize dd-cfDNA in clinical practice. Table 1 presents summaries of various studies on kidney transplantation that utilize dd-cfDNA for predicting rejection. TCMR and ABMR can elevate dd-cfDNA levels, though the correlation with ABMR is more promising. One should understand that dd-cfDNA alone does not reliably distinguish between rejection phenotypes. Integration with DSA testing, histology, and clinical parameters is required for correct diagnosis and tailored immunosuppression to identify the specific rejection type.

Table 1 Summaries of studies utilizing donor-derived cell-free DNA assay in the diagnosis of rejections.

Ref.	Journal/year	Objective	Area under the curve	Threshold dd-cfDNA	Sensitivity and specificity/predictive value	Findings
Bloom et al[13]	J Am Soc Nephrol/2017	To investigate dd-cfDNA is a marker of rejection	0.74 for TCMR and ABMR	1%	PPV 61%; NPV 84%	Levels above 1% indicate a probability of active rejection of TCMR (> BANFF 1 B) or ABMR
Sigdel et al[14]	Clin Med/2018	To study single nucleotide polymorphism (SNP) multiplexed PCR to measure dd-cfDNA in various types for the detection of allograft rejection/injury	0.87	1%	Sensitivity 89%; specificity 73%; PPV 52%; NPV 95%	SNP-based dd-cfDNA assay detected allograft rejection with good performance
Huang et al[15]	Am J Transplant/2019	To investigate dd-cfDNA is a marker of rejection	0.42 for TCMR, 0.82 for ABMR	0.74%	Sensitivity 100% for ABMR; Specificity 71.8% for ABMR; PPV 68.6% for ABMR; NPV 100% ABMR	dd-cfDNA could not discriminate TCMR from no rejection -dd-cfDNA strongly discriminates TCMR from no rejection
Whitlam et al[16]	Am J Transplant/2019	To evaluate the validity of dd-cfDNA, total cell-free DNA, and graft fraction to correlate with individual Banff scores	0.89%	0.75%	Sensitivity 85%; PPV 48%; NPV 95%	dd-cfDNA was more associated with AMR -Failed to diagnose borderline or type 1A TCMR
Wijtvliet et al[17]	Transpl Int/2020	Metanalysis on dd-cfDNA to predict rejection				dd-cfDNA is a useful marker for ABMR but not for TCMR
Dauber et al[18]	Transpl Int./2020	To use INDEL PCR for dd-cfDNA for the detection of TCMR	0.84 to discriminate TCMR from no rejection	2.7%	Sensitivity 88%; Specificity 81%	INDEL PCR for dd-cfDNA could discriminate between TCMR and no rejection at a median level of 5.24%
Zhang et al[19]	Front Immunol /2020	To evaluate the diagnostic performance of dd-cfDNA in discriminating antibody-mediated rejection (ABMR) or de novo donor-specific antibodies (DSA) without histological lesions in kidney allograft recipients	0.90	1%	Sensitivity 89%; sensitivity 74%; PPV 76%; NPV 88%	dd-cfDNA correlates with AMR and may pick up cases of ABMR with stable graft function, and can help in early diagnosis
Benning et al[20]	Transpl Int/2023	To investigate dd-cf DNA in predicting rejections in patients with graft biopsy and see response to treatment	AUC 0.72 for any rejection including borderline vs no rejection		Median 2% for ABMR; median 0.92% for TCMR; median 0.42% for borderline rejection	dd-cfDNA distinguishes between active rejection and no rejection. dd-cfDNA level decreased due to antirejection therapy
Xing et al[21]	Biomol Biomed/2024	Metanalysis on dd-cfDNA to predict rejection	AUC for TCMR 0.80, AUC for ABMR 0.87		Sensitivity for TCMR 59% and specificity 83%; sensitivity for ABMR 81% and specificity 80%	Diagnostic ability for TCMR is limited, but it is a useful marker to detect ABMR
Aubert et al[22]	Nat Med/2024	Observational multicentric study of 1134 patients	AUC from 0.77-0.82	2.85% ± 0.68% for mixed rejection; 2.03% ± 1.13% for acute TCMR; 1.15% ± 0.15% for active AMR; 1.09% ± 0.15% for chronic active AMR		dd-cfDNA strongly correlated with ABMR, TCMR, and mixed rejection
Akifova et al[23]	Nephrol Dial Transplant/2024	A randomized prospective study in recipients with donor-specific antibody (DSA) and high dd-cfDNA and DSA with clinical indication of biopsy was compared		> 50 cp/mL	PPV 77%; NPV 85%	Biopsy in dd-cfDNA was performed earlier, leading to early management

dd-cfDNA: Donor-derived cell-free DNA; TCMR: T cell–mediated rejection; ABMR/AMR: Antibody-mediated rejection; DSA: Donor-specific antibody; SNP: Single nucleotide polymorphism; INDEL: Insertion–deletion; PCR: Polymerase chain reaction; AUC: Area under the curve; PPV: Positive predictive value; NPV: Negative predictive value; Banff: Standardized histopathological classification for kidney allograft pathology; IFTA: Interstitial fibrosis and tubular atrophy; DGF: Delayed graft function.

Potential role in early ABMR screening

Diagnosing ABMR in its early stages allows clinicians to intervene promptly; this may change the outcome, as traditional markers such as creatinine, urine protein estimation, and DSA monitoring may lead to late diagnosis. Graft biopsy in clinically indicated cases may have a chronic lesion leading to poor response to therapy. In high-risk allograft recipients with DSA, interval monitoring of dd-cfDSA may prompt clinicians to consider a biopsy and detect early ABMR in cases of high dd-cfDNA. Such an approach has been beneficial in an early biopsy to detect ABMR in a randomized, prospective study[23]. Further evaluation of this concept is warranted globally in populations with diverse characteristics to assess the utility of dd-cfDNA in high-risk allograft recipients. Figure 1 shows various clinical utilities of dd-cfDNA.

Monitoring the response to anti-rejection therapy

Like their role in diagnosis, dd-cfDNA assays can be used to see the response to treatment. Based on this theme, a study from Germany monitored dd-cfDNA at days 7, 30, and 90 and found a significant reduction in dd-cfDNA following treatment for rejection[20]. Shen et al[24] also reported on dd-cfDNA following rejection therapy, and the change in the reduction of dd-cfDNA was found to have a positive correlation with GFR at one, three, and six months. Gupta et al[25] also reported a decrease in dd-cfDNA after rejection therapy. This suggests that dd-cfDNA may be a potential marker for treatment response, and serial monitoring could help clinicians track the resolution of rejection and inform therapeutic decision-making.

Role in monitoring and planning immunosuppression

The potential role of dd-cfDNA is in the surveillance of graft function and the planning of immunosuppression. The Assessing Donor-derived cell-free DNA Monitoring Insights of kidney Allografts with Longitudinal surveillance study monitored dd-cfDNA over a three-year period. Elevation of dd-cfDNA to 0.5% or more has been shown to predict both clinical and subclinical rejection and a threefold increased risk of de novo DSA[26]. Interestingly, dd-cfDNA levels were elevated 91 days before the development of DSA. Persistent elevation of dd-cfDNA on more than one occasion above 0.5% predicted showed a 25% decline in estimated glomerular filtration over three years.

Stable dd-cfDNA may identify patients who are eligible for a reduction in immunosuppression. The reduction of immunosuppression has been successfully employed without an increased risk of rejection in patients with stable dd-cfDNA[27]. Similarly, another study using dd-cfDNA successfully stratified rejection risk, and the mycophenolic acid dose was adjusted without adverse outcomes[28]. There is a need for further, well-designed prospective studies to utilize dd-cfDNA for surveillance and adjusting immunosuppression doses.

WHAT ARE POTENTIAL CONFOUNDERS THAT MAKE INTERPRETATION OF THE ASSAY DIFFICULT?

Although the dd-cfDNA is a valuable diagnostic test for detecting allograft rejection, there are various confounders that clinicians should keep in mind while interpreting the results. These include such as infections, ischemia-reperfusion injury, or other forms of graft stress. Figure 2 shows various confounding conditions that may lead to elevated dd-cfDNA levels or interfere with the assay, independent of rejection.

Figure 2 Shows various confounding conditions that may lead to elevated donor-derived cell-free DNA levels or interfere with the assay, independent of rejection. dd-cfDNA: Donor-derived cell-free DNA; IFTA: Interstitial fibrosis and tubular atrophy; DGF: Delayed graft function.

BK virus nephropathy

BK virus can cause viral-induced cell necrosis and indirect damage to tubular cells. Therefore, the elevation of dd-cfDNA in BK virus infection is an accepted phenomenon. The DART study was analyzed to evaluate the association between dd-cfDNA and BK viral load and biopsy findings. Interestingly, patients diagnosed with BKV and BKVAN had median dd-cfDNA levels of 0.58% and 3.38%, respectively[29]. Yet, in another study, BK virus nephropathy led to modest dd-cfDNA elevation, but less than that seen in ABMR. However, this elevation is comparable to TCMR[30]. Another study found high urine dd-cfDNA in biopsy-proven BK virus nephropathy[31]. Elevation of dd-cfDNA in BK virus infection can confound rejection diagnosis. Therefore, one should carefully interpret the result in the presence of the BK virus infection.

Ischemia-reperfusion injury and delayed graft function

Ischemia-reperfusion injury and delayed graft function (DGF) cause cellular damage through acute tubular injury and the activation of innate immune responses. It has been demonstrated that dd-cfDNA levels rise immediately after surgery (to more than 5% of total cfDNA) and then decrease to less than 5% within a week[32].

It is essential to understand the kinetics of dd-cfDNA while interpreting its value post-transplant. The decline in dd-cfDNA is rapid post-transplant and follows an L-shaped curve[33]. The results showed that the mean dd-cfDNA concentration was 20.69% at three hours, 5.22% by about 16 hours, and 0.85% by Day 7. The decrease was slower in recipients of deceased donors, where the donor had cardiac death, as compared to recipients of the kidney from live donors. Deceased donation was associated with higher initial dd-cfDNA than live donation, and they had more DGF[33]. Another study showed that higher 24-hour dd-cfDNA levels are linked to functional DGF and lower 6-month eGFR. Persistently high dd-cfDNA at Day 7 points to worse long-term graft survival. On the other hand, early dd-cfDNA normalization (< 0.5% at 1 week) predicts better graft function in the future[34]. These facts may complicate the diagnosis of rejection in the immediate post-transplant period, and clinicians should consider these kinetic factors when investigating for rejection.

Other confounding factors

Many other causes can lead to an elevation in dd-cfDNA. A study examining pre-graft biopsy dd-cfDNA and monitoring post-biopsy dd-cfDNA found that dd-cfDNA levels rise after a renal allograft biopsy. This rise in dd-cfDNA is transient and returns to baseline values in 24-48 hours[35]. Interstitial fibrosis and tubular atrophy (IFTA) have been associated with high dd-cfDNA levels, although this finding was not statistically significant[19]. Further studies are needed to see the relation between IFTA and dd-cfDNA levels. Other confounders that could increase dd-cfDNA levels include blood transfusion[36], dual kidney transplantation[37], and pregnancy[38]. Morbid obesity has been shown to have an inverse relationship with obesity and should be considered when interpreting the results for dd-cfDSA[39]. Other potential confounders, such as calcineurin toxicity and recurrent glomerular disease, may theoretically cause dd-cfDNA elevation and warrant further evaluation.

Therefore, elevated dd-cfDNA is not specific to rejection. Its level may rise due to ischemia-reperfusion injury, infections, drugs, and other forms of graft injury. There is a need in the future to develop multivariable predictive models combining dd-cfDNA with clinical, serological (C-reactive proteins, white blood cells, drugs, and DSA), and radiology markers to improve specificity.

WHAT METHOD OF QUANTIFICATION SHOULD BE USED?

dd-cfDNA is quantified as both a ratio and an absolute value. In ratio quantification, dd-cfDNA is expressed as a ratio to total cfDNA where dd-cfDNA is used as the numerator, with total free cell DNA serving as the denominator. Absolute values are expressed as copies per milliliter of plasma[40]. Both these methods have pros and cons. Total plasma cfDNA is dependent on the physiological status of the recipients. In addition to rejection conditions, such as malignancy[41], stroke[42], and infarction[43], which can increase cfDNA levels, another important point to remember is that more than 90% of cfDNA in a plasma sample originates from white blood cells such as neutrophils and lymphocytes. These cells undergo natural apoptosis in the blood, possibly leading to more cfDNA[44]. Although much of the literature has reported dd-cfDNA in ratio[45], critics have pointed out the variability of cfDNA, making the ratio at times erroneous. Few studies have shown that the absolute value of dd-cfDNA may be more accurate than the ratio in predicting rejections[16,46,47]. Further studies are warranted to gain a deeper understanding of the most effective methods for quantification.

WHAT ARE THE ASSAYS AVAILABLE FOR DD-CFDNA AND WHAT ARE THEIR DIAGNOSTIC ABILITIES?

Allosure (CareDx), Prospera (Natera), and Viracor Transplant Rejection Assessment in Cell-free DNA (TRAC) (Eurofins) are various assays used to detect dd-cfDNA. These tests help in early detection of allograft injury and rejection. Most assays are for detecting dd-cfDNA SNPs[13]. These SNPs are homozygous in the recipient but differ from the donor's[13]. Allosure targets are 266 SNPs[48,49], and Prospera targets 13392[14,50].

Allosure has a limit of quantification as low as 0.2%[48]. It has high analytical precision with a coefficient of variation ranging from 4.5% to 7.7%[48]. Various studies have reported sensitivities ranging from 0.59 to 0.83[13,48] and AUC values up to 0.86[49]. This indicates that Allosure has moderate to strong discrimination of rejection. The Prospera assay utilizes a broader panel of 13962 SNPs[50] and has a low detection threshold of 0.15%-0.23%[50]. Sensitivities have been reported from 0.55 to 0.887[14,51] while specificity ranges from 0.69 to 0.726[14,51]. The AUC has been reported to be between 0.75 and 0.87, indicating moderate to high diagnostic accuracy[14,51]. These studies suggest that Allosure and Prospera have shown variable diagnostic performances. This could be due to differences in study design, patient populations, types of rejection analyzed, and timing of sample collection. There is a need for careful clinical interpretation of these tests. There is also a need to integrate these tests with other diagnostic tools and clinical judgment.

Melancon et al[51] did a head-to-head comparison of Allosure and Prospera[51]. Both assays reported an AUC of 0.73 for Allosure and 0.75 for Prospera, indicating similar overall diagnostic accuracy. Prospera showed a slightly higher sensitivity of 0.55 as compared to 0.45. This may give an edge in the detection of rejection, but this came at the cost of lower specificity (0.69 vs 0.85). From Melancon et al's data[51], we can infer that Prospera may be more sensitive in detecting rejection. Conversely, Allosure offers greater specificity, potentially reducing false positives. A head-to-head comparison showed that both tests demonstrated similar diagnostic performance.

The Viracor TRAC assay employs a quantitative PCR method to measure dd-cfDNA. This assay demonstrated a sensitivity of 0.579 and a specificity of 0.853, with an AUC of 0.85[52]. The Viracor TRAC measures dd-cfDNA levels from 0.5% to 60%. These findings suggest that Viracor TRAC offers good specificity and overall diagnostic ability in detecting allograft injury in transplant recipients. However, further studies are warranted to analyse the Viracor TRAC utility in detecting dd-cfDNA.

Future prospective multicenter trials comparing dd-cfDNA assays across diverse patient populations and types of rejection are needed. Furthermore, dd-cfDNA must be integrated with biomarkers and histology to enhance diagnostic precision. Table 2 summarizes Allosure, Prospera, and Viracor TRAC's studies and their diagnostic abilities[52].

Table 2 Summaries of studies of Allosure, Prospera, Viracor TRAC, and their diagnostic abilities.

Ref.	Journal/year	Assay used	SNPs	Limit of quantification	Limit of detection	Sensitivity	Specificity	Area under curve	Coefficient of variation
Grskovic et al[48]	J Mol Diagn/2016	Allosure	266	0.2%	0.16%	0.83	0.84		7.7% (dd-cfDNA < 2%). 4.5% (dd-cfDNA ≥ 2%)
Bloom et al[13]	J Am Soc Nephrol/2017	Allosure	266			0.59	0.85	0.74
Jordan et al[49]	Transplant Direct/2018	Allosure	266	0.2%-16%				0.86
Sigdel et al[14]	J Clin Med/2018	Prospera	13392			0.887	0.726	0.87 (87%)
Altuğ et al[50]	Transplantation/2019	Prospera	13962	0.15% (related donors); 0.23% (unrelated donor)	0.15% (related donors); 0.23% (unrelated donor)				4.29%
Melancon et al[51]	Kidney360/2020	Allosure vs Prospera	266 vs 13392			0.45 vs 0.55	0.85 vs 0.69	0.73 (73%) vs 75 0.75 (75%)
	Viracor TRAC	Quantitative PCR	0.5%-60%		0.579		0.853

dd-cfDNA: Donor-derived cell-free DNA; SNP: Single nucleotide polymorphism.

WHAT ARE THE OTHER BIOLOGICAL AND TECHNICAL FACTORS WHICH COULD AFFECT DD-CFDNA?

In addition to the common confounders discussed earlier, biological and technical factors can also contribute to variability in measured levels. In the early transplant period, dd-cfDNA levels are high due to a variety of reasons, including ischemia-reperfusion injury, surgical trauma, and early immune activation[13]. It is essential to remember this fact while interpreting the dd-cfDNA result. One should wait at least two weeks to start surveillance with dd-cfDNA to avoid false elevation in dd-cfDNA levels[48]. Another essential factor to consider is the genomic disparity between donor and recipient. A greater genetic disparity, such as that between unrelated donors, may lead to better accuracy. In contrast, low disparity may limit accuracy due to reduced ability to distinguish donor-specific variants[6]. As explained, pregnancy[38] or blood transfusion[36] are other confounders due to persistent microchimerism, which can contribute low levels of non-self-DNA and potentially affect assay specificity. It is essential to consider technical variations from different assays. Various methodologies, such as SNP-based targeted sequencing and shotgun approaches, had distinct sensitivity and quantification thresholds[32,48]. Multiple factors can interfere with the results before analysis for dd-cfDNA. These factors include sample handling, hemolysis, and time to plasma separation. Furthermore, differences in normalization methods and bioinformatics pipelines may also lead to variability in dd-cfDNA measurement across laboratories[14,48].

Because of this heterogeneity, it is essential to establish global standards for assay methodology and reporting. The debate over using a universal vs context-specific dd-cfDNA threshold is still disputed. Different studies reported variable thresholds for ABMR vs TCMR. Generally, lower thresholds may be appropriate in ABMR detection, while higher values may better correlate with TCMR. In the DART study[13], a dd-cfDNA threshold of 1.0% was found to discriminate between rejection and no rejection, particularly for ABMR. However, lower dd-cfDNA levels were seen in TCMR, highlighting the limitations of a single universal cutoff. Conversely, Aubert et al[22] found a threshold of 1% for ABMR and 2.85% for TCMR. These differences likely result from variations in patient populations, types of rejection, and assay platforms[52]. Since these biological and technical factors can significantly influence dd-cfDNA measurements, care must be taken to avoid misinterpretation. Establishing uniform practices and assay standardization worldwide is necessary to interpret these assays accurately. Choosing between a universal dd-cfDNA threshold and context-specific thresholds is also clinically significant and a topic of ongoing discussion. For example, post-transplant timing or donor type, such as live vs. deceased, remains an area of research. There is a need to identify thresholds relevant to the clinical context that improve the diagnostic capabilities of this assay.

WHAT ARE FUTURE STUDIES AIMING FOR?

Transplant clinicians and researchers are optimistically looking for the utility of dd-cfDNA in future studies. Various studies combine dd-cfDNA with gene expression profiling, such as Allomap and tools based on clinical parameters such as iBox. AlloMap is a gene profiling assay that helps to detect peripheral blood immune reactivity. It has been used successfully to differentiate between quiescence and rejection states in kidney transplant patients[39]. iBox is another validated tool. It analyzes multiple clinical parameters such as estimated GFR, proteinuria, presence of DSA, and renal allograft histology, and estimates the long-term risk of graft failure. It has been successfully used in kidney transplantation[53]. Another novel idea is to combine dd-cfDNA with Human Leukocyte Antigen–DR isotype-positive, Tumor Necrosis Factor Receptor 2-positive Regulatory T cells (HLA-DR+TNFR2+ Tregs). Tregs play a crucial role in transplant tolerance by suppressing effector immune responses. The reduction of HLA-DR+ Tregs has been demonstrated in cases of acute rejection[54]. Measuring dd-cfDNA together with HLA-DR+TNFR2+ Tregs can potentially determine the future risk of rejection. BK viremia is one of the important confounders making interpretation of dd-cfDNA difficult. Researchers are actively searching for gene expression profiling along with dd-cfDNA to diagnose rejection confidently in the presence of BK viremia.

dd-cfDNA and Allomap may identify low-risk individuals who can benefit from minimal immunosuppression. Although dd-cfDNA has been shown to monitor response to antirejection therapy[20,24,25] effectively, researchers are further consolidating the evidence of this biomarker in the future. The Molecular Microscope^® Diagnostic System (MMDx) is a diagnostic tool based on gene expression profiling to interpret kidney transplant biopsy tissue at the molecular level[55]. It analyzes thousands of genes within a biopsy sample and histopathological assessments to improve the ability to diagnose rejection with precision. By analyzing the expression of over a thousand genes within biopsy samples, MMDx provides a molecular perspective that complements traditional histopathological assessments, aiming to improve the accuracy and objectivity of transplant rejection diagnoses. Future analysis of dd-cfDNA and MMDx may provide exciting information to the transplant community. Efforts are underway to study further the prognostic and diagnostic abilities of dd-cfDNA to avoid kidney biopsies in the future. The myth of dd-cfDNA being an effective fluid (liquid) biopsy in the future will be explored. Figure 3 shows potential future aims that future studies are focusing on.

Figure 3 Potential future aims for ongoing studies. ABMR: Antibody-mediated rejection; TCMR: T cell-mediated rejection; AKI: Acute kidney injury; CKI: Chronic kidney injury; cfDNA: Cell-free DNA.

WHAT ARE THE PERTINENT ONGOING STUDIES?

Several ongoing pertinent studies combine dd-cfDNA with other molecular and clinical parameters. The KOAR study (NCT03326076) studies dd-cfDNA, AlloMap, and iBox to predict long-term outcomes. The result is expected to be announced by 2031[56]. Another ongoing study (NCT05084768) will measure the plasma level of dd-cfDNA and HLA-DR+TNFR2+ Tregs at various intervals during the first six months to assess the risk of acute rejection[57]. The study is estimated to be completed by 2026. The SPKCareDx Study (NCT04855422) is another ongoing study that will analyze dd-cfDNA and gene expression (AlloMap Kidney) to diagnose TCMR and ABMR in simultaneous pancreas-kidney transplant recipients. The study will investigate the reliability of these markers in the presence of BK virus viremia[58]; however, BK viremia makes the diagnosis of rejection tricky and may confound traditional rejection markers. Yet another ongoing trial (NCT04786067) will explore the role of dd-cfDNA and AlloMap in guiding the minimization of immunosuppressive medications. The study will identify individuals with the least reactive immune systems who could be transitioned to belatacept monotherapy[59]. An ongoing study focuses on calibrating dd-cfDNA (Prospera) against MMDx for ABMR, TCMR, acute kidney injury, and chronic kidney injury (CKI). It will determine cutoff values for rejection, acute, and CKI[60]. Another trial in the pediatric population (NCT05477082) will examine whether gene expression profiling and dd-cfDNA will be useful to see the response of biopsy-proven TCMR and ABMR[61]. The DEFILE study (NCT06476717) is a prospective, single-center, observational trial that is still ongoing and will assess the role of dd-cfDNA for detecting kidney allograft injuries. It will analyze two cohorts. The first cohort will analyze 100 renal allograft recipients from deceased donors in whom dd-cfDNA will be determined at Day 14 and then at months 1, 2, 3, 4, 6, 9, and 12 post-transplantation. The second cohort will consist of 40 patients who underwent allograft biopsy with histological or molecular diagnosis of rejection, and dd-cfDNA will be monitored to assess the response of antirejection therapies. The study's ultimate objective is whether dd-cfDNA can reduce the need for invasive biopsies and improve post-transplant care[62]. Like the DEFILE study, the QIDNEY (NCT03765203) is another prospective, multicenter study. Its primary objective is to assess whether dd-cfDNA can correctly identify rejection or injury as compared to traditional clinical markers. By comparing dd-cfDNA levels with biopsy-confirmed outcomes, this study aims to improve the monitoring of the renal allograft and reduce the need for invasive procedures[63]. Transplant communities, by looking into more details of the utility of dd-cfDNA in the detection of rejection, are aiming to avoid needle biopsies to reach a diagnosis of rejection.

COULD THE CLINICAL UTILITY OF DD-CFDNA BE A COST-EFFECTIVE OPTION IN THE FUTURE?

Although dd-cfDNA could be a promising future biomarker, its utilization in clinical practice needs evaluation in the context of health care resource utilization. Renal biopsy remains the gold standard for diagnosing rejection. However, it carries a complication rate of 1%–3.5%[2]. Furthermore, renal biopsy requires admission, imaging support, trained personnel and a histopathologist to interpret. Therefore, the cumulative cost of repeated surveillance biopsies and patient burden may be high, especially in high immunological risk kidney transplant recipients. dd-cfDNA could pick ABMR earlier than kidney biopsy and may allow the clinician to initiate therapy before chronicity sets in. Furthermore, dd-cfDNA testing may reduce the need for protocol or for-cause biopsies. Monitoring dd-cfDNA can guide clinicians to tailor off immunosuppression for individual patients. The reduction of immunosuppression has been successfully employed without an increased risk of rejection in patients with stable dd-cfDNA[27]. Similarly, another study using dd-cfDNA successfully stratified rejection risk, and the mycophenolic acid dose was adjusted without adverse outcomes[28]. In heart transplants, the utility of dd-cfDNA has led to reduced endomyocardial biopsies, which are projected to cost 8545 United States dollars less than endomyocardial biopsies[64].

The results of two ongoing studies may shed more light on the cost-effectiveness of using dd-cfDNA. The DEFILE trial[62] is a prospective, multicenter study assessing the use of dd-cfDNA in planning immunosuppression and reducing unnecessary biopsies in kidney transplant recipients. A similar study, the QIDNEY study[63], is still going on, evaluating the role of dd-cfDNA to monitor graft function. It also focuses on reducing invasive procedures. Thus, dd-cfDNA will improve diagnostic accuracy, lead to improved outcomes, and reduce costs. Future studies are needed to address reimbursement decisions and guide widespread adoption.

WHAT SHOULD BE THE WAY FORWARD?

Although the evidence for the utility of picking ABMR is high[13,19], its results for selecting low-grade TCMR remain equivocal, and a higher threshold of greater than 2% is required for TCMR[18,21]. There are various potential confounders such as ischemia reperfusion and DGF[33], IFTA[19], blood transfusion[36], dual kidney transplantation[37], pregnancy[38], and morbid obesity[39]. One should be careful to interpret the results of dd-cfDNA in the presence of these conditions. Other confounders include patients who had allogenic bone marrow transplantation, kidney transplantation among monozygotic twins, and recipients of multiorgan[65]. There still exist myths about the utility of dd-cfDNA to assess various aspects of kidney transplant recipients, which warrants further studies. Currently, it is not universally used in the diagnosis of rejection. Most organizations and transplant societies still have to make recommendations for the utility of dd-cfDNA and the recommendations available from a few vary. The American Society of Transplant Surgeons suggests using serial dd-cfDNA levels in kidney transplant recipients with stable graft function to exclude the presence of subclinical ABMR. It further recommends measuring dd-cfDNA levels in kidney transplant recipients with acute allograft dysfunction to exclude the presence of rejection, particularly ABMR[66]. The European Society of Organ Transplantation made a similar recommendation as AST[67]. Banff has no official position statement; however, in 2022, the Banff-Canadian Society of Transplantation published updates in kidney transplantation. In these updates, it was stated that elevated dd-cfDNA was a predictor of rejection. But since dd-cfDNA remained clinically stable, the update mentioned that monitoring dd-cfDNA alone in stable allograft function remains unclear[68]. The British Transplantation Society (BTS) has no comment about the utility of dd-cfDNA in kidney transplantation. However, in the context of cardiothoracic transplantation regarding the use of dd-cfDNA in diagnosing acute rejection, the BTS stated that the evidence is in infancy and that no clinical recommendation can be made[69]. Figure 4 shows an integrated approach to using dd-cfDNA to diagnose rejection and renal allograft surveillance.

Figure 4 An integrated approach to using donor-derived cell-free DNA for the diagnosis of rejection and renal allograft surveillance. dd-cfDNA: Donor-derived cell-free DNA.

There is a need for consensus among transplantation societies for the utility of dd-cfDNA in the context of kidney transplantation. At present, the clinicians should integrate dd-cfDNA along with clinical assessment to manage kidney transplant recipients.

FUTURE DIRECTIONS AND RECOMMENDATIONS

The use of dd-cfDNA alone in the clinical management of kidney transplant may be challenging. There is a need for an integrated pathway consisting of dd-cfDNA, gene expression profiling, and clinical parameters to diagnose cause of allograft dysfunction at the earliest, optimize individual immunosuppression medications and asses long term risk of graft failure. We recommend the following: There is a need to improve assay sensitivity and specificity, establish a standardized threshold, exclude confounding factors, and diagnose clinical and subclinical rejection confidently. There is a need for standardization for cutoff, reporting unit (ratio vs. absolute value), and interpretation of methodology used (Allosure vs Prospera) across the globe to ensure uniform practice among clinicians.

CONCLUSION

dd-cfDNA is a valuable biomarker for picking up the earliest ABMR. It may have a potential role in predicting graft surveillance, tailoring immunosuppression for individuals, and monitoring response to rejection therapy. There is a need for standardization for cutoff, reporting unit (ratio vs absolute value), and interpretation of the methodology used. Prospective observational multicentric studies are needed to improve their diagnostic abilities through integration with clinical parameters, gene profiling, and the use of other biomarkers. Furthermore, there is a need for uniform consensus among transplant societies and organizations to establish guidelines for its clinical use in kidney transplantation.