The evolving use of biomarkers in heart transplantation: Consensus of an expert panel

Abstract

In heart transplantation, the use of biomarkers to detect the risk of rejection has been evolving. In this setting, it is becoming less clear as to what is the most reliable test or combination of tests to detect rejection and assess the state of the alloimmune response. Therefore, a virtual expert panel was organized in heart and kidney transplantation to evaluate emerging diagnostics and how they may be best utilized to monitor and manage transplant patients. This manuscript covers the heart content of the conference and is a work product of the American Society of Transplantation’s Thoracic and Critical Care Community of Practice. This paper reviews currently available and emerging diagnostic assays and defines the unmet needs for biomarkers in heart transplantation. Highlights of the in-depth discussions among conference participants that led to development of consensus statements are included. This conference should serve as a platform to further build consensus within the heart transplant community regarding the optimal framework to implement biomarkers into management protocols and to improve biomarker development, validation and clinical utility. Ultimately, these biomarkers and novel diagnostics should improve outcomes and optimize quality of life for our transplant patients.

1. Introduction

An expert panel conference on the evolving use of biomarkers in heart and kidney transplantation was held virtually on January 27, 2022 to January 28, 2022. This manuscript covers the heart content of the conference and is a work product of the American Society of Transplantation’s Thoracic and Critical Care Community of Practice. The objectives included the following:

• Review and appraise available analytical and clinical data for advanced diagnostic tests for heart transplant rejection

• Identify assay advantages and limitations

• Provide guidance for use of these tests in transplant recipients in specific clinical settings

• Determine unmet needs and areas for future investigation

In heart transplant, there are 2 areas where biomarkers have been clinically useful. These include acute rejection monitoring and defining the immune-responsive state to customize immunosuppression.

Currently, we have several biomarkers/diagnostic tests for use after heart transplantation. These include gene expression profiling (GEP), donor-derived cell-free deoxyribonucleic acid (ddcfDNA), donor-specific antibodies (DSAs), T cell immune monitoring (TCIM) assay, and intragraft messenger ribonucleic acid transcripts (IRTs). Assays in development include micro-RNAs, digital pathology, and exosomes, to name a few. All have been directly or indirectly compared to traditional histopathologic evaluation of the endomyocardial biopsy (EMB). The challenge is that the EMB may not be a true “gold standard” to detect acute rejection owing to artifacts that confound biopsy interpretation, sampling error, or shortcomings with the grading rubric itself. Moreover, it has been demonstrated that, among expert pathologists, there is only a two-thirds concordance rate to diagnose acute cellular rejection via biopsy interpretation.¹ Nevertheless, the biopsy has been used for over 50 years to manage heart transplant patients and should be considered the reference standard until a new modality proves sufficient clinical utility to become the new reference standard.

This paper discusses many aspects of these newer diagnostic tests and defines current unmet needs. Additionally, we hope to build consensus within the community as to the optimal framework to improve biomarker development, with the goal of improving clinical outcomes and optimizing quality of life for our transplant patients.

2. Review of the community survey of heart transplant programs

Prior to the conference, a survey was sent to transplant cardiologists in the United States. The aim of this survey was to assess the opinions of a broad group of stakeholders about the current use of biomarkers/diagnostic tests in heart transplantation, their optimal use in the future, the data necessary to inform their optimal use, and the appropriate pathways to gain endorsement from regulatory agencies. The survey was sent to the medical directors (transplant cardiologists) of 134 heart transplant programs with 34 responses (25%). The demographics of the 34 respondents to the survey included 28 (82%) heart transplant cardiologists, 3 transplant infectious diseases physicians, 1 (3%) pharmacist, and 2 (6%) other. Of the heart transplant physicians, 2 were pediatricians. Below are responses (including the highest percentage response) to select questions (complete survey results are included in Supplementary Material):

• 91% of programs currently employ ≥1 biomarkers/diagnostic tests in the care of transplant recipients with 78% using these in routine clinical care.

• 39% of programs believe that the minimum level of data necessary to support regulatory endorsement includes multicenter prospective evidence associating the biomarker/diagnostic test with the event of interest.

• 34% of programs believe that diagnosing ongoing clinical or subclinical rejection is the most impactful use of biomarkers/diagnostic tests in transplantation.

• 45% of programs are interested in a biomarker/diagnostic test that has a high negative predictive value, thereby facilitating the safe reduction in the patient’s net state of immunosuppression.

• 56% of programs felt that <$500 is a reasonable cost for a biomarker/diagnostic test that is to be used routinely.

• 94% of programs feel that it is acceptable/clinically feasible to run a battery of biomarkers/diagnostic tests either in parallel or series.

3. Systematic overview of current and emerging diagnostic tools and biomarkers

3.1. Human leukocyte antigen antibodies (donor-specific antibodies)

De novo DSAs have been reported in 10% to 30% of post-transplant patients with a mean follow-up of 7 years.² Patients with preformed DSAs and/or positive flow cytometry crossmatch at transplant displayed significantly lower freedom from first-year antibody-mediated rejection (AMR).^3,4 Both preformed and de novo DSAs are associated with increased incidence of AMR, coronary artery vasculopathy (CAV) and heart allograft loss.^5,6 De novo DSAs against anti-human leukocyte antigen (HLA) class II, particularly DQ, are at high risk for CAV development.⁷ As a result of this associated risk of DSAs, posttransplant monitoring for DSAs has been recommended at 1, 3, 6, and 12 months postoperatively in accordance with International Society for Heart and Lung Transplantation (ISHLT) guidelines.⁸ Patients who are low risk should be monitored annually for DSAs after the first year. Sensitized patients should be monitored more frequently. It has also been suggested that DSA testing should be performed for any patient presenting with symptoms or signs of graft dysfunction. DSAs with graft dysfunction and restrictive physiology should be considered for treatment.⁸ DSAs that remain at higher dilutions, complement binding DSAs, DSAs that persist, and DSAs arising late after transplantation have been associated with adverse outcomes, but the benefit of intervention has not been established.^9–13 Additional DSA attributes such as immunoglobulin G subclass and the impact of Fc receptor genotypes may improve diagnostic and prognostic information but need further study.^14,15

More research is required to determine whether treating antibodies in these situations (with and without graft dysfunction) would improve outcomes. Randomized, controlled trials are needed to assess the benefit of treatment in posttransplant sensitized patients. Other future studies will include whether sensitization should trigger a heart biopsy, assessing various treatment approaches to DSAs and validating the suspected causal link between antibodies and CAV.

Solid-phase assays such as the Luminex platform have been the most common method to detect and identify DSAs. Although solid-phase assays are more sensitive than cell-based assays, their adoption in clinical trials has been impeded by intra and interlaboratory variability. Differences can result from the test conditions, laboratory technician performance, and kit manufacturing processes (ie, concentration of antigens on latex beads). Standardization of these operating procedures is needed to reduce these differences.

3.2. Non–human leukocyte antigen antibodies

Many heart transplant patients present with AMR in the absence of DSAs, suggesting a role for non-HLA antibodies in allograft injury. Antibodies specific to vimentin, MHC class I polypeptide sequence chain A, angiotensin II type 1 receptor, perlecan, and endothelial cells have been detected concurrent with AMR and allograft vasculopathy.^15–20 Furthermore, non-HLA antibodies detected in combination with DSAs are associated with greater rejection severity. However, a causal role for non-HLA antibodies has not yet been definitively determined. Few clinical studies have used large unbiased cohorts with sufficient rejection events, adequate DSA monitoring, and complete donor HLA typing to show an independent role for non-HLA antibodies in rejection lesions and transplant outcomes. Non-HLA antibodies detected concurrent with or following rejection or de novo DSAs may be a product of allograft injury and severity rather than the causal agent. Therefore, currently we can only conclude that these non-HLA antibodies serve as biomarkers associated with current or past cardiac allograft injury. New non-HLA bead panels allow detection of antibody specific for multiple non-HLA antigens and are leading the way for larger, better controlled cohort studies. Recent studies by Butler et al²⁰ and See et al¹⁹ show elevated non-HLA antibody reactivity at the time of cardiac rejection (acute cell-mediated rejection [ACR] and AMR respectively). Sensitization to 4 non-HLA targets were found to associate with cardiac AMR whereas 18 non-HLA antibody targets correlated with ACR; 2 targets, vimentin and tubulin alpha 1B, were found to be overlapping between these studies. Thus, we need a larger body of evidence to use non-HLA antibodies as an independent diagnostic biomarker for diagnosis, treatment, prognosis, and outcome after heart transplantation.

3.3. Gene expression profiling in peripheral blood cells

Gene expression profiling of peripheral blood mononuclear cells can provide information regarding the recipient’s alloimmune response to the donor heart. Commercially available GEP testing monitors the expression of 11 genes to identify cardiac allograft recipients who are at low risk for ACR. A multicenter randomized clinical trial that enrolled patients between 6 months and 5 years after transplant with a median follow-up of 19 months showed that patients who were monitored with GEP and those who underwent routine biopsies had similar 2-year cumulative rates of the composite primary outcome (rejection with hemodynamic compromise and graft dysfunction because of other causes, death, or retransplantation). Patients who were monitored with the use of GEP underwent fewer biopsies per person-year of follow-up.²¹ As a result of this study and others, GEP has been included in the ISHLT 2010 guidelines for the care of heart transplant patients.²² A limitation of GEP is that it was developed and validated only for ACR but not AMR.

3.4. Donor-derived cell-free deoxyribonucleic acid

Donor-derived cell-free DNA can be isolated from a variety of human body fluids such as blood, urine, saliva, lymph, breast-milk, bile, spinal fluid, and amniotic fluid. It serves as a noninvasive biomarker for disease, infection, and tissue injury/rejection. In a study by Deshpande et al,²³ ddcfDNA showed excellent agreement with clinical rejection and, importantly, serial measurement of ddcfDNA predicted clinically significant outcomes after treatment for rejection in these patients.²³ In a multicenter, prospective cohort study by Agbor-Enoh et al,²⁴ the investigators demonstrated that immediately posttransplant, there was a marked elevation of ddcfDNA, with an exponential decay that occurs within 2 to 3 weeks. This level of ddcfDNA became the basal level for stable patients without rejection. Importantly, there was a significant increase in ddcfDNA in patients with AMR and ACR. The performance of ddcfDNA assessed against the EMB had an area under the curve of 0.9. The positive predictive value (PPV) of ddcfDNA was ~20%, but many of the false positives occurred in the setting of graft dysfunction in the absence of biopsy-proven rejection or preceded rejection by ~3 months. The use of ddcfDNA to screen for rejection would safely avoid 81% of all surveillance endomyocardial biopsies.²⁴ A limitation of using ddcfDNA is that it does not differentiate between AMR and ACR.

3.5. T cell immune monitoring assay

T cell immune monitoring is a measure of the strength of a patient’s immune system via measured T cell activity. The clinical application of TCIM in heart transplantation has been contentious because of the small sample size of published cohort studies as well as discordant results.^25,26 However, in a larger patient study, TCIM testing predicted the risk of infection in heart transplant patients, but the association between high immune monitoring (IM) scores and risk of rejection was inconclusive.²⁷ TCIM appears to help to avoid overimmunosuppression, may be most effective when monitoring serial levels, and may assist in the assurance of adherence to immunosuppression therapy.

3.6. Microribonucleic acids

MicroRNAs are noncoding ribonucleic acids (RNAs) that typically lead to messenger RNA (mRNA) transcript destabilization, suppression of translation, and eventual mRNA transcript degradation. Approximately 50% of protein-coding genes are regulated by microRNAs, and as biomarkers, they are remarkably stable in the circulation with 90% contained within exosomes or bound to proteins or lipid particles.²⁸ They are promising biomarkers in solid organ transplantation with the potential to diagnose and distinguish the 2 major subtypes of rejection with an area under the curve ranging from 0.87 to 0.98.²⁹ Aside from their potential use as biomarkers, they have been shown to be therapeutic targets, and prior studies have suggested that pharmacologic inhibition of specific microRNAs may attenuate rejection.^29,30 However, this biomarker is early in development and is not ready for commercial use.

3.7. Intragraft messenger ribonucleic acid transcripts

When biopsies are performed, the measurement of intragraft gene expression mRNA transcripts has significant potential to improve biopsy interpretation. IRTs can be assessed using a variety of platforms, including RT-PCR, microarrays, RNA sequencing, Nano string, and others. A system for assessing EMBs has been developed in the INTERHEART study (ClinicalTrials.gov NCT02670408) to define TCMR, ABMR, and parenchymal injury.^31–36 The molecular microscope (MMDx) is a central biopsy diagnostic system that uses genome-wide microarrays to measure gene expression in intact RNA (stabilized in RNAlater) with high precision (>99%) and compares the biopsy to a reference set, using ensembles of predefined machine learning-derived algorithms.³² MMDx correlates with histology, but often with discrepancies, at least part of which is expected given the interobserver disagreement in histologic diagnosis of rejection.^1,35 Because it gives automated assessments, MMDx emerges as a platform with which to assess body fluid biomarker performance—DSAs, autoantibodies, ddcfDNA, and others—against the biopsy findings already in progress for kidney transplants.³⁷

3.8. Digital pathology in heart transplantation

Although histologic grading of EMBs has been the mainstay of rejection diagnosis for decades, the subjective and qualitative nature of this approach presents clear limitations. In practice, traditional histologic grading suffers from poor reliability and modest accuracy while providing minimal information about long-term allograft outcomes.^38–44 An emerging paradigm for addressing these shortcomings leverages quantitative image analysis software to provide more objective and rigorous assessments of EMB histology. Computer-aided analysis of digital pathology slides can provide reproducible, easily disseminated histology assessments. Peyster et al³⁹ demonstrated that automated analysis of digitized EMBs can be used to generate highly standardized, clinical-quality, acute rejection grading across centers. The “morphologic biomarkers” extracted through digital pathology workflows can also be combined with more conventional forms of patient data to generate nuanced and accurate predictions of important long-term outcomes. A recent publication combining novel morphologic biomarkers with traditional clinical risk factors showed excellent capacity for predicting cardiac allograft vasculopathy years before overt disease onset.⁴⁵ Lastly, spatial immune profiling using digital pathology methods has shown potential to improve the diagnostic and prognostic value of EMB histology samples, uncovering novel biology associated with allograft injury that may have valuable precision medicine implications.⁴⁶ These promising translational studies justify continued investment to further define the role of digital pathology as a tool for diagnosis and discovery in transplant medicine.

3.9. Extracellular vesicles

Extracellular vesicles (EVs) are found in nearly all bodily fluids and secreted in both normal and pathologic states. EVs circulating in body fluids represent an attractive candidate for rejection detection, as their cargo mirrors the originating cell (donor-derived) and its pathophysiological status. Once isolated from body fluids, multiomics can be applied to EVs as downstream analyses, including lipidomics, miRNomics, and proteomics. A recent clinical study collected plasma from 90 heart transplant patients (53 training cohort, 37 validation cohort) before the EMB. The concentration of EVs was significantly increased, and their diameter decreased in patients undergoing rejection with the trend being highly significant for both AMR and ACR (P <.001). Among EV surface markers, CD3, CD2, ROR1, SSEA-4, HLA-I, and CD41b were identified as discriminants between controls and acute cellular rejection, whereas HLA-II, CD326, CD19, CD25, CD20, ROR1, SSEA-4, HLA-I, and CD41b discriminated controls from patients with AMR.⁴⁷ Further studies demonstrating the utility of this biomarker after transplant will be required for this biomarker to have commercial potential.

4. Biomarker evaluation

At the conclusion of these discussions, the participants of the conference were asked to evaluate each diagnostic in the form of a table (Table 1). The lead discussant had previously reviewed the available literature and had indicated whether the following had been performed:

• Randomized controlled trials (Y/N)

• Cohort studies (registry data) (Y/N)

• Case–control studies (Y/N)

• Case series/reports (Y/N)

Table 1.

Conference participants’ evaluation of current diagnostics. A Likert scale was used (1 highest through 5 least) for each query.

Diagnostic/Biomarker	Randomized controlled trials (Y/N)	Cohort studies (ie, registry data) (Y/N)	Case-control studies (Y/N)	Case series/reports (Y/N)	Validity to detect rejection (1 highest through 5 least)	Reproducibility (1 highest through 5 least)	Utility to affect outcome (1 highest through 5 least)	Practicality for use (1 highest through 5 least)	Is negative predictive value important? (Y/N)	Is negative predictive value important? (Y/N)
Endomyocardial biopsy	N	Y	N	Y	2–3	3	1	3–4	Y	Y
HLA antibodies (donor-specific antibodies)	N	Y	Y	Y	2	2–3	2–3	2–3	–	–
Non-HLA antibodies	N	Y	Y	Y	5	5	5	5	–	–
Donor-derived cell-free DNA	N	Y	Y	Y	2	2	2–3	2–3	Y	Y
Gene expression profiling	Y	Y	Y	Y	3–4	2	2–3	2–3	Y	Y
microRNA	N	N	Y	Y	3–4	5	5	5	–	–
T cell immune monitoring	N	N	Y	Y	4–5	4–5	3	3	–	–
Intragraft messenger RNA transcripts	N	Y	Y	Y	1–2	1	2	3–4	Y	Y
Digital pathology	N	Y	Y	Y	2	2	2	–	Y	Y
Exosomes	N	N	Y	Y	5	5	5	–	–	–

HLA, human leukocyte antigen; N, no; RNA, ribonucleic acid; Y, yes.

The group then came to verbal consensus on the next 6 questions. A Likert scale was used (1 -highest to 5 - least) for each query below:

• Validity to detect rejection

• Reproducibility for detected rejection

• Utility to affect outcome

• Practicality for use

• Is negative predictive value important (Y/N)

• Is PPV important (Y/N)

In addition, conference participants discussed the advantages and limitations of each of the biomarkers, which is summarized in Table 2. They felt that the EMB is not completely reliable or reproducible, as noted in the Cardiac Allograft Rejection Gene Expression Observational II study,¹ where pathology-read biopsies of ISHLT 3A rejection had a concordance of only 28%. This is an issue as the current biomarkers/diagnostics are being compared to a flawed gold standard to detect heart transplant rejection.

Table 2.

Advantages and limitations of each biomarker/diagnostic.

Diagnostic/biomarker	Advantages	Limitations
Endomyocardial biopsy	The endomyocardial biopsy has been the gold standard to detect rejection for over 50 y and relies on tissue diagnosis.	Concordance among expert pathologists is ~67%. Artifacts are confounding. Additionally, in 2% of cases, patients have acute clinical decompensation with no corresponding pathology on biopsy (biopsy-negative rejection). The test is also invasive and less practical compared to a peripheral blood test.
HLA antibodies (DSAs)	DSAs are useful for risk stratification as both preformed and de novo DSAs are associated with increased incidence of AMR, CAV, and heart allograft loss. De novo DSAs against anti-HLA class II, particularly HLA-DQ, are at high risk for CAV development.	There is noteworthy intra- and interlaboratory variability to identify DSAs owing to differences in test conditions, laboratory technician performance, and kit manufacturing processes (ie, concentration of antigens on latex beads). Standardization of these operating procedures is needed to reduce these differences. In addition, the clinical utility of treating DSAs to improve outcomes has not been established.
Non-HLA antibodies	These tests may explain the mechanism for biopsy-negative rejection episodes.	Non-HLA antibodies are nonspecific during heart transplant, and the existing strength of evidence for use of this biomarker for rejection surveillance is low.
Donor-derived cell-free DNA	ddcfDNA is practical (noninvasive) and can detect acute rejection (ACR and AMR). High ddcfDNA test results signify abnormal pathology or allograft injury. Multiple technologies to measure donor fraction and applications beyond rejection such as predicting the development of DSA and CAV.	Inability to distinguish ACR from AMR is a limitation. Furthermore, no clinical utility trials have been performed, so it is unknown whether ddcfDNA monitoring can modify clinical outcomes.
GEP	The clinical utility of GEP has been shown in the IMAGE Trial.21 GEP is now included in the ISHLT guidelines to indicate absence of cellular rejection.	Only detects ACR and unable to detect AMR. The test requires core lab services for sample processing; however, mobile phlebotomy has helped with ease of acquiring samples.
microRNA	Highly stable and easy to measure in plasma or serum. Ability to detect both ACR and AMR.29 Aside from their potential use as biomarkers, they have been shown to be therapeutic targets, and prior studies have suggested that pharmacologic inhibition of specific microRNAs may attenuate rejection.30	Limited case series, lack of reproducibility for individual microRNAs. Methods of data collection for preliminary studies are not standardized, and therefore, results are not consistent across all studies. Further work needs to be done to establish validity. Still early in development.
T cell immune monitoring	T cell immune monitoring may have value in terms of demonstrating overimmunosuppression or trends. This assay may be helpful in patients with severe infections and malignancies to guide minimization of immunosuppression and may assist in the assurance of adherence to immunosuppression therapy.	T cell immune monitoring assay has inconsistencies. The clinical application of TCIM in heart transplantation has been contentious because of the small sample size of published cohort studies as well as discordant results.25,26
Intragraft mRNA transcripts	IRTs have the ability to detect rejection and distinguish ACR from AMR and can assess acute and chronic parenchymal injury. They provide mechanisms of the rejection process. May impact outcome by influencing rejection treatment if they provide a clinically relevant result compared to the pathology-read report.	Reliant on the heart biopsy. It recognizes a heterogeneity of pathways, which can confound the clinician. Needs RCT to confirm true utility in clinical decision making and impact in changing clinical outcomes. This diagnostic has been used as an adjunct to histology scoring in kidney transplant biopsies to improve rejection recognition.
Digital pathology	Digital pathology appears reliable and reproducible and has high potential. Strong evidence for cellular grading quality (multicenter cohort). Similar to IRTs, it may impact outcome by influencing rejection treatment if they provide a clinically relevant result compared to the pathology-read report.	Reliant on the heart biopsy. Lower quality evidence for rejection severity and future risk prediction.
Exosomes	Detection of cellular and antibody-mediated rejection was retrospectively demonstrated by proteomic analysis of circulating serum exosomes.	Lack of standardization for exosome purification and analysis precludes adoption. Evidence overall is minimal and still very early in exploratory stages.

ACR, acute cell-mediated rejection; AMR, antibody-mediated rejection; CAV, coronary artery vasculopathy; ddcfDNA, donor-derived cell-free deoxyribonucleic acid; DNA, deoxyribonucleic acid; DSA, donor-specific antibodies; HLA-DQ; GEP, gene expression profiling; HLA, human leukocyte antigen; IMAGE, Invasive Monitoring Attenuation through Gene Expression; IRT, intragraft messenger ribonucleic acid transcript; ISHLT, International Society for Heart and Lung Transplantation; RCT, randomized controlled trial; RNA, ribonucleic acid; TCIM, T cell immune monitoring.

5. Small group concomitant breakout and discussion

The participants of the biomarker conference were divided into 2 smaller groups to discuss 5 specific questions that arose from 3 premeeting conference calls held in the preceding 4 months. The premeeting discussions involved the current diagnostics as well as unmet needs. The breakout groups were evenly divided into those persons with expertise in each of these specific diagnostics as well as representing large and small heart transplant programs.

The participants of the biomarker heart transplant conference reconvened following the small group breakout discussions. Group leaders discussed their responses to the 5 predetermined questions and engaged the entire group in the reconvened session for discussion, as summarized below.

Question #1: Based on everything we have discussed, what do you consider to be the most impactful use of biomarkers/diagnostic tests in heart transplantation? In contrast to the pre-meeting survey, participants felt that the most impactful use for biomarkers/diagnostics may be risk stratification, based on time before and after heart transplantation. Testing for the purpose of risk stratification should be performed at specific time points as part of monitoring, bearing in mind that pretransplant comorbidities, demographics, and posttransplant complications such as infection and rejection will change the trajectory of risk. Consideration for donor comorbidities such as age, sex, and smoking history are also warranted. There were comments that a biomarker that could accurately monitor the strength of the alloimmune response can help with risk stratification. An accessible, affordable, and reproducible biomarker that can screen patients for a positive outcome such as rejection is desirable.

Question #2: What is the most important aspect of a biomarker to detect heart transplant rejection? It was noted that the EMB is a poor surrogate for clinical status. Intragraft biomarkers may be useful in the future, but their current use is not clear. The PPV of new biomarkers continues to be measured against the EMB, which is an unreliable gold standard, so the calculated PPV of new assays is probably much higher than what is reported. Ideally, biomarkers should indicate rejection prior to biopsy-determined rejection. It was commented that high PPV should be a top priority. For screening purposes, negative predictive value demonstrates minimal risk, but for a diagnostic test, high PPV will guide clinical practice.

Question #3: Which biomarkers should be used in combination with others for optimal clinical outcome? It was not clear what specific combination(s) would be most clinically helpful. However, DSAs and ddcfDNA may be helpful in specifying the type of rejection (ie, AMR) and clinical outcome. DSAs as a bimodality are insufficient; we must consider other assays such as dilution or complement fixation. Rising DSAs may suggest more active alloimmunity, indicating a need for more effective treatment. Other combinations were discussed but were of less interest.

Question #4: What is the minimum level of data necessary to support a regulatory endorsement and the routine use of a biomarker/diagnostic test? Large multicenter prospective cohort studies and randomized clinical trials (at least 3 to 5 sites) are certainly needed because registries do not track patients longitudinally to assess the relationship between the biomarker and clinical outcomes. In designing clinical trials, long-term complications such chronic kidney disease, cardiac allograft vasculopathy, and cancer are important endpoints. Dynamic risk stratification is needed, as risk will change over time. Ideally a biomarker that can identify patients who are at low risk for posttransplant complications would be desirable.

Question #5: Discuss unmet needs and areas of future investigations, comment on the outlook of diagnostics/biomarkers on the horizon. The biomarkers on the horizon such as micro-RNAs, exosomes, and digital pathology are of great interest but need to have further testing and validation. For current biomarkers, there is a need for tissue specificity in ddcfDNA assays (ie, examination of tissue-specific methylation). Technology is needed to advance the field to develop more organ-specific biomarkers.

5.1. Summary

There was broad agreement that larger and more sophisticated studies are needed to investigate, compare, and contrast the growing number of potential biomarkers in heart transplant medicine. In addition, it is clear that the EMB is not the gold standard to detect rejection and that we need a better definition of rejection to validate all current and future biomarkers. There was also recognition that there is unlikely to exist a single, preeminent assay or biomarker that serves all purposes and that combinations of biomarkers, employed either together or at different times after transplant, are the most probable future outcome. As such, biomarker studies should be well-powered through inclusion of multiple centers and diverse patient populations but also comprehensive in their inclusion of data, bio-samples, and outcomes to facilitate multibiomarker investigations under a common set of circumstances to determine clinical utility. Some molecular tests can also be used to discover molecular mechanisms to understand and reclassify the disease and injury states that determine function and outcomes. Pooling of existing resources as well as careful planning for future observational cohorts and randomized controlled trials will be needed to achieve these goals. The reward of such efforts will be enhanced diagnostic precision and advances in personalized patient care.

6. Expert panel consensus statements

After these final discussions, a poll of conference participants (n = 17) was taken. Consensus was viewed as majority support for any 1 answer.

1. The most impactful use of biomarkers/diagnostic tests in heart transplantation is risk stratification (76%).

2. The most important aspect of a biomarker is a high PPV to detect heart transplant rejection without a biopsy (65%).

3. Currently, DSAs and ddcfDNA used in combination may be the optimal biomarkers to assess clinical outcomes (65%).

4. Multicenter clinical trials are necessary to support regulatory endorsement and the routine clinical use of a biomarker/diagnostic test (71%).

5. A reliable immune-responsive test help to manage immunosuppression dosing is needed (94%).

6. The detection of DSA alters therapy (71%).

7. EMB is not the true “gold standard” to detect rejection (82%).

8. EMBs continue to have a role in heart transplant patient management (88%).

Abbreviations:

ACR

acute cell-mediated rejection

AMR

antibody-mediated rejection

ddcfDNA

donor-derived cell-free deoxyribonucleic acid

DSA

donor-specific antibody

EMB

endomyocardial biopsy

EV

extracellular vesicle

GEP

gene expression profiling

HLA

human leukocyte antigen

IRT

intragraft messenger ribonucleic acid transcripts

IMAGE

Invasive Monitoring Attenuation through Gene Expression

ISHLT

International Society for Heart and Lung Transplantation

MMDx

molecular microscope

PPV

positive predictive value

TCIM

T cell immune monitoring

Data availability

No data is available for this paper.