Allograft solid-organ transplantation has become a standard of care in patients with end-stage organ disease. In some patients, these treatments, along with other advances in care, have transformed fatal disease into treatable and preventable disease.1-3 Transplant patients are placed on immunosuppressant drug therapy and are routinely monitored for indicators of rejection to prolong the survival of the donor allograft. Serologic and laboratory markers aid in monitoring graft function and in the early detection of AR which may occur as ACR or AMR.4 Transplant recipients may also undergo (a) surveillance biopsies to detect subclinical acute rejection (subAR) and other histologic changes that precede functional allograft decline and/or (b) for-cause biopsies when clinically indicated. With early intervention, these changes may still be amenable to treatment. Early detection of AR has led to significant improvement in allograft survival in the first 12 months post-transplantation.5,6
The importance of graft rejection and immunosuppression was discovered early on following the development of transplantation, a challenge that started to be overcome with the availability of immunosuppressants, though with the current standard-of-care for managing solid organ transplant patients, rejection remains a common problem with a high frequency of graft failure at 5 and 10 years.6-9 Though organ transplant recipients remain at risk for rejection indefinitely, the risk is highest in the first year post-transplant.10-12 Graft assessment is also used clinically to assist in the management of immunosuppression; the clinical value it brings is that it allows modification of immunosuppressive therapy to maximize graft longevity, which is a focus of post-transplant care. Histology has traditionally been used, often in conjunction with clinical and laboratory markers, as a common graft assessment tool.9,13-19 While histology is considered the gold standard of diagnosis at this point in time, it requires a biopsy, which is invasive and may be associated with significant risks, complications and access to care barriers.
The interpretation of relevant histopathologic findings for the purpose of guiding management also poses a challenge. For example, in kidney transplantation, there has been heterogeneity in the literature regarding the definition and treatment of borderline and subAR, making it challenging to compare results between studies.20-25 Borderline rejection is a heterogeneous diagnosis and includes the broad range of mild “inconsequential” inflammation to clinically significant ACR that can lead to inferior graft function.24 Further, subAR can be detected in up to 25% of surveillance biopsies performed in the first year following kidney transplantation (a number that may be decreasing as immunosuppressive regimens improve); however, the clinical relevance of finding borderline and subAR has been uncertain for many years.26-28 Nevertheless, it is now generally accepted that subAR is associated with worse graft outcomes 20,23,26,29-31 that may be improved with early diagnosis and treatment.32,33 In a study from the University of Wisconsin, where donor specific antibody (DSA)-based protocol biopsies were performed at months 3 and 12 for kidney transplant recipients, patients with treated subclinical acute AMR were found to have similar good outcomes as those who underwent protocol biopsy but who were without rejection.32 The treated subAR patients also fared significantly better than those who underwent a clinically indicated biopsy with or without AMR.32 Another study from Johns Hopkins University evaluated kidney transplant recipients though year 5 post-transplantation and found no statistically significant difference in graft loss for patients with treated subclinical AMR compared to their matched controls; however patients with subclinical AMR who did not receive treatment had a 3.34-fold (95% CI: 1.37–8.11; p = 0.008) higher risk of graft loss.33 Protocol biopsies in this study, which demonstrate value for renal surveillance testing, were performed at 1, 3, 6 and 12 months following ABO- and HLA-incompatible live donor transplants.33
Despite evidence supporting its value, up to 80% of transplant centers do not routinely perform surveillance biopsies after kidney transplantation because of the relatively low incidence of subAR, the risk–benefit ratio of protocol biopsies and the fact that protocol biopsies have not definitively been shown to improve outcomes.23,27,28,34 Further, the most current KDIGO guidelines on the management of kidney transplant recipients (published in 2009) also do not recommend surveillance biopsies.12 For those centers that do perform surveillance biopsies, the surveillance biopsy testing schedules are not standardized.23,27 The current care model for monitoring patients after kidney transplantation ranges from not using surveillance biopsies at all, to using them selectively in “high-immunologic risk” patients, to routine use in all patients.27,30,32,35 Though there is no one standard of practice regarding the performance of surveillance biopsies in kidney transplantation, some center protocols have been published.27,36
For other transplant types such as heart and lung, surveillance biopsies have traditionally been performed more frequently for the majority of patients, given the heightened risk of rejection early in the post-transplant period, though variations in surveillance biopsy practices and interpretive histopathologic challenges of biopsy apply to these transplant types as well.10,37-42 For example, International Society for Heart and Lung Transplantation (ISHLT) guidelines suggest that approximately 13 endomyocardial biopsies (EMBs) (or other non-invasive methods) should be performed in the first 12 months following heart transplantation.10 However, the multicenter prospective cohort study from the Genomic Research Alliance for Transplantation (GRAfT) reported that heart transplant recipients underwent approximately 8 EMBs over a median follow-up time of 17.7 months.41
Molecular diagnostic methods have emerged in an attempt to address limitations in current diagnostics including the measurement of donor-derived cell-free DNA (cfDNA) and gene expression profile (GEP) assays, which have been developed for a number of organ allografts including kidney, heart, liver, and lung.43-50 The principle underlying cfDNA assays to assess rejection is that the transplantation of a new organ involves transplantation of new genetic material, and genetic material is shed into the bloodstream as part of rejection.51 The fraction of cfDNA in the blood-stream may serve as a marker of rejection.46-49 While this is a straightforward principle, DNA concentration in the bloodstream is quite small, and therefore tests relying on cfDNA require advanced technological methods that can accurately capture and quantify the presence of cfDNA specific to the allograft.46,49,51 GEP tests tend to quantify expression of numerous genes in the allograft recipient and use these data in algorithms developed with sophisticated modeling or machine learning to determine whether rejection is occurring.50,52 Molecular tests can not only provide information about graft status in a minimally-invasive manner, but they can be sensitive enough to be able to detect AR before it is evident histologically or by other clinical and laboratory indicators.41,42,51 For example, in kidney transplant recipients, cfDNA values of 0.5% or more were associated with a nearly three-fold increase in risk of developing de novo DSAs; cfDNA was elevated a median of 91 days prior to DSA detection, suggesting a role for cfDNA as an early identifier of AR.36,53 Further, for various transplant types, an improvement in cfDNA has been reported in patients who receive treatment for rejection and have biopsy evidence of resolution.25,41,49
Circulating Donor-Derived Cell-Free DNA tests have also demonstrated the ability to identify subclinical recurrence in a surveillance setting. The Circulating Donor-Derived Cell-Free DNA in blood for diagnosing Acute Rejection in Kidney Transplant Recipients (DART) study by Bloom et. al demonstrated the performance of cfDNA in detecting allograft injury and AR; however, this study was not able to estimate the performance of cfDNA to discriminate AR in patients who may have had subclinical rejection because there were only 34 surveillance biopsies and only one finding of active rejection.47 The Assessing Donor-derived cell-free DNA Monitoring Insights of kidney Allografts with Longitudinal surveillance (ADMIRAL) study by Bu et. al subsequently compared 45 surveilled patients with findings of AR on protocol biopsy with cfDNA findings.36 This study found that elevation of cfDNA (≥0.5%) was significantly correlated with clinical and subclinical allograft rejection in kidney transplants and that persistently elevated cfDNA predicted over a 25% decline in the estimated glomerular filtration rate (eGFR) over three years.36 Mantios et. al, who tested cfDNA at months 1, 2, 3, and 5 post-transplantation also found that in a surveillance group, high levels of cfDNA from the second month post-kidney transplantation were correlated with non-increasing eGFR one year post-transplantation; moreover, in patients with AR, levels of cfDNA decreased substantially after initiation of treatment.54 In a recent study of 2,882 kidney allograft recipients from 14 transplantation centers in Europe and the United States, Aubert et. al found that the inclusion of cfDNA to a standard-of-care prediction model showed improved discrimination (area under the curve (AUC) 0.777 - 0.821; P = 0.0011); results were confirmed in external validation cohorts (n = 1,748).55 Importantly, cfDNA was able to detect any kind of rejection, including subclinical rejection, which the authors note “remains the most challenging diagnosis when using traditional kidney transplant markers alone.”55 There were no statistically significant differences in cfDNA levels between AMR and ACR and higher levels of cfDNA were associated with higher degrees of severity of both AMR and ACR.55 Importantly, the authors found that the independent association of cfDNA with AR remained significant when cfDNA was tested in the first 3 months post-transplantation (OR = 2.370; 95% CI 1.505–3.730, P < 0.001), between 3 months and 1 year post-transplantation (OR = 1.882; 95% CI 1.381–2.566, P < 0.001) and after 1 year post-transplantation (OR = 2.521; 95% CI 1.944–3.316, P < 0.001).55
Despite the fact that elevated cfDNA may be a predictor of AR and decreased eGFR, some kidney transplant recipients with elevated cfDNA remain clinically stable and it is not clear how to differentiate them from those who will develop adverse outcomes.56 Further, cfDNA has been reported to have a limited sensitivity for detecting borderline and early ACR, raising concern that reliance on this analyte could miss cases of rejection (though AMR is thought to be the more important cause of transplant failure).57 cfDNA also has a limited positive predictive value (PPV) (approximately 50-60%), suggesting that routine surveillance could potentially result in the performance of unnecessary biopsies, though this impact has not been observed in clinical utility studies to-date.56 Nevertheless, many transplant programs in the United States are routinely surveilling kidney transplant patients with cfDNA and treating both subclinical AMR and ACR, despite the fact that there continues to be significant heterogeneity in the classification and treatment of subclinical borderline and early ACR.22 cfDNA tests have been proposed to reduce variance in interpretive approaches in order to better identify those patients with subclinical borderline and early ACR who are at elevated risk of graft injury.58 In a retrospective single-center study of kidney transplant recipients, addition of cfDNA to Banff biopsy scores provided better prognostic assessment over certain biopsy findings alone.59 However, that study was limited by a number of factors including the lack of standardized treatment protocols and the potential for bias introduced by exclusion of cfDNA tests assessed without a biopsy.59
In heart transplantation, molecular biomarker testing has been used for over a decade and, unlike the outdated Banff guidelines for kidney transplantation (at the time of this work), the ISHLT guidelines do recommend the use of molecular biomarker testing in the evaluation of heart transplant recipients.10,60 Despite longstanding clinical experience with Allomap (CareDx, Brisbane, CA), ISHLT guidelines affirm that the impact of GEP-guided surveillance on long-term clinical outcomes in heart transplantation still needs further study.10 Additionally, the optimal monitoring schedule using molecular methods is unknown, though ISHLT-recommended schedules for heart transplantation have been published for the performance of DSA and biopsy “or other non-invasive methods as appropriate.”10 In recent years, cfDNA testing has become more widely utilized in heart and lung transplant recipients as studies have shown good performance for the detection of AR. Studies evaluating cfDNA have shown high sensitivity (78.5%-81%), specificity (76.9%-85%) and NPV (97.3%-99.2%) for AR in heart transplant recipients.41,61 Importantly, these studies also showed good performance for clinically significant ACR, previously thought to be better detected by GEP tests. Further, they have shown that cfDNA levels may rise weeks to months before AR is diagnosed on biopsy.41,42,62-64 For example, in the GRAfT study, 80% of cfDNA elevations in heart transplant recipients were associated with a negative EMB.41 Importantly, the majority of these (76%) were considered clinically significant because they were associated with allograft dysfunction or preceded histopathologic rejection by up to 3.2 months.41 In lung transplant recipients, higher cfDNA values were associated with a significantly higher hazard of chronic lung allograft dysfunction (CLAD) or death than those with lower levels.65 Moreover, during the COVID-19 pandemic, a home-based surveillance program using plasma cfDNA instead of surveillance bronchoscopy identified early episodes of AR and infection that would not have been identified using a clinically indicated biopsy strategy alone.40 As a result, some heart and lung transplant centers have adopted cfDNA for surveillance alongside reduction in the number of biopsies during the first years post-transplantation, when the risk of AR is highest and when levels of molecular markers such as cfDNA are predictable.10,11,39,40,42,64,66
Given the acceptance of GEP testing, studies subsequently attempted to evaluate the value of adding cfDNA to GEP in heart transplant patients, as both approaches were thought to have different strengths and weaknesses in ACR vs AMR detection.42 In a single-center retrospective review of combined GEP and cfDNA testing for AR surveillance, Gondi et. al found that combination testing at 6 months post-transplant resulted in the cancellation of 83.9% EMBs as opposed to 71.2% using a GEP-only surveillance approach.67 This decrease was driven primarily by the 87 GEP (+) and cfDNA (-) results that would have previously resulted in an EMB but were reassigned to a lower risk group with the addition of cfDNA. Notably, there were very few rejections in the cohort (two AMR and two grade 2 ACR), and cfDNA detected them all while GEP missed the cases with AMR.67 Another single center, retrospective analysis by Henricksen et. al found that at one year post transplantation, there were no differences between the paired testing and GEP-only cohorts in terms of survival or rejection-free survival, though paired testing was associated with fewer EMBs. However, their study only had 3 episodes of clinically significant AR and there was not a standard protocol for reacting to the results of the noninvasive surveillance.68 In a retrospective observational study of more than 400 samples from 112 unique heart transplant patients who underwent combination testing with both cfDNA and GEP, Rodgers et. al observed a significantly improved specificity without a significant change in sensitivity when using cfDNA instead of a combination testing approach (sensitivity: 36.8% cfDNA vs 50.5% combination testing, p=0.99; specificity: 86.3% cfDNA vs 46.8% combination testing, p < 0.0001); this study identified 13 samples associated with AR, 10 of which were ACR.69 Based on their findings, the authors conclude that cfDNA testing without GEP is sufficient, agreeing with other reports that cfDNA alone has the potential to replace or significantly reduce the need for EMB.41,42,61,62,69 A prospective cohort study of 565 plasma samples collected longitudinally from 65 adult and pediatric transplant recipients also reported that cfDNA showed superior test performance when compared to AlloMap [AUC cfdDNA = 0.83; AUC AlloMap = 0.72];62 notably, they compared the performance of cfDNA from their study with that of AlloMap as reported in another study by Deng et. al.70 Finally, the performance of combination testing for ACR surveillance was evaluated most recently in the Surveillance HeartCare Outcomes Registry (SHORE) study.71 The reported sensitivity, specificity, and likelihood ratio (LR) for combination testing was 32.3%, 91.7% and 3.90 (95% CI: 3.08-4.96); sensitivities, specificities and LRs for testing with GEP alone were 59%, 56.8% and 1.37 (95% CI: 1.20-1.56) and 44.7%, 84.6% and 2.91 (95% CI: 2.43-3.49) for cfDNA alone.71 Of the 156 grade 2 ACR events reported, only 31.4% were dual positive (49/156) while another 30% were dual negative (46/156) by molecular testing.71 The incidence of ACR in the GEP negative/ cfDNA negative group (1.5%; 95% CI: 1.1%-2.0%) was not significantly different from that of the GEP positive/cfDNA negative group (1.9%; 95% CI: 1.4%- 2.6%); however, in the GEP negative/ cfDNA positive and dual positive groups it was 4.3% (95% CI: 2.8%-6.6%) and 9.2% (95% CI: 7.1%-11.9%), respectively.71 Of the cfDNA positive tests (n=1026), 72 (7%) showed clinically significant ACR, similar to the dual positive tests. Notably, ACR was observed most frequently in the first 30 days post-transplant.
There is considerably limited evidence for combination testing using cfDNA and GEP in kidney transplantation.72,73 For this reason, the American Society of Transplant Surgeons (ASTS) and the European Society of Organ Transplantation (including expert members from the United States) also do not support GEP or combination testing for this indication.74,75
Just as there is no one standard of practice regarding the performance of surveillance biopsies, particularly in kidney transplantation, there is also not a standard practice for the performance of surveillance molecular tests, though some center protocols have been published in the literature.36,76 Questions remain regarding which patients would benefit most from routine serial testing, how often to test, and the appropriate steps to further evaluate an elevated cfDNA.25,77 Minimally invasive testing should “allow for a biomarker stratified approach for those at a higher risk of harboring subAR.”30
Importantly, these molecular tests are not meant to be used in isolation. Rather, they are part of a constellation of clinico-laboratory data used to evaluate the patient.57 Accordingly, the 2022 Banff-Canadian Society of Transplantation Joint Meeting and the Banff Minimally Invasive Diagnostics Working Group have concluded that these biomarkers should not be used alone to monitor patients in the setting of stable allograft function.57,78
While these technologies are newer, large and multicenter studies have supported their use in renal, heart, and lung transplantation as minimally and non-invasive methods to assess allograft status, modify immunosuppression regimens, and avoid unnecessary biopsies.11,40,46,47,50 Evidence continues to develop for other transplant allograft organs and other analytes.45,79-82
Contractor Advisory Committee (CAC) Summary
Kidney and Liver Allograft CAC
On November 16, 2022, Noridian and Palmetto jointly facilitated a CAC to discuss molecular testing for the diagnosis of acute rejection in kidney or liver transplant recipients where GEP, cfDNA, and multiomic tests were discussed. The subject matter experts (SMEs) expressed a high degree of certainty that there is sufficient evidence to support the use of molecular tests like cfDNA in kidney allograft patients.
Heart and Lung Allograft CAC
On November 17, 2022, Noridian and Palmetto jointly facilitated a CAC to discuss molecular testing for the diagnosis of acute rejection in heart or lung transplant recipients where GEP and cfDNA tests were discussed. The SMEs were unanimous that the evidence is sufficient to support the use of GEP and/or cfDNA for surveillance of acute organ injury to help replace an unnecessary biopsy. However, they also agreed that there is a lack of data to support the routine use of molecular combination testing. Moreover, they did not endorse the use of molecular diagnostic tests in lieu of biopsy in patients with a very high clinical and laboratory suspicion for acute rejection and who can also safely tolerate a biopsy. All SMEs agreed that these tests do not discern the etiology of organ injury, indicating that they had a high degree of confidence in the detection of allograft injury, but additional clinical testing was necessary to discern the cause.
Documents associated with the CAC meetings can be found here:
Contractor Advisory Committee (CAC) - JE Part B - Noridian (noridianmedicare.com)