Blood-based immunological monitoring after heart transplant. Current status and future prospects

Abstract

The development of effective immunosuppressive strategies has been a key to successful outcomes in heart transplantation. Immunosuppression however is a double-edged sword with consequences of rejection when underutilized and risk for infection, malignancy, and drug toxicity when used in excess. The search for non-invasive blood-based monitoring not only to assess allograft rejection but also to monitor overall state of immunosuppression remains very attractive, potentially reducing cost and complications and allowing more frequent testing to assess response to anti-rejection therapy. This review outlines the current status of blood-based immunological monitoring after heart transplantation including serological monitoring, biomarkers, and evolution of molecular technologies.

Introduction

The endomyocardial biopsy is a flawed but durable gold standard.

Despite the development of durable mechanical circulatory support, heart transplantation remains the therapy of choice in the long term for selected patients with end-stage heart disease. In its infancy, heart transplantation was however limited by the lack of adequate immunosuppressive therapy, inability to monitor for allograft rejection, and unacceptable infection rates leading to high transplant mortality. The development of the percutaneous method for acquiring endomyocardial biopsies (EMB) for the assessment of rejection was a major step forward in post-transplant management [1], and this technique has survived the test of time over the last 45 years. Initial interpretation was limited to evaluation of acute cellular rejection (ACR), but in recent years, there has been increasing appreciation for the role of donor-specific antibody causing acute antibody-mediated rejection (AMR), a diagnosis associated with potentially worse outcomes than ACR [2]. While histologically its diagnosis has been more challenging, in recent years, a consensus has evolved regarding its interpretation on EMB [3]. However, despite its ubiquitous use, the EMB remains with significant limitations. Its invasive nature is associated with procedural discomfort and finite complication risk including tricuspid valve injury [4], myocardial perforation [5] and development of coronary fistula [6]. The technique also has limited sensitivity due to sampling error, inability to assess deeper myocardial tissue, and lack of agreement in histological interpretation even between experienced pathologists [7]. The biopsy however remains the gold standard for surveillance and diagnosis of acute rejection. Most programs continue to use this as the primary method for surveillance for the first 6–12 months, although the introduction of gene expression profiling (GEP) (see below) has reduced the number of biopsies needing to be performed in many patients. Patients at high risk for rejection, particularly AMR, should continue to be monitored by EMB in the first year. After the first postoperative year, EMB surveillance for an extended period of time (e.g., every 4–6 months) is recommended in heart transplant recipients at higher risk for late acute rejection to reduce the risk for rejection with hemodynamic compromise and the risk of death in black recipients who are at higher risk for rejection. A typical first-year protocol for EMBs is outlined in Fig. 1a (EMB Protocol) for patients being maintained indefinitely on long-term corticosteroids. In patients being considered to be weaned off corticosteroids, this may be performed as early as 8–12 weeks post-transplant in appropriately selected patients, but most programs will defer until the sixth month.

Fig. 1.

Typical surveillance protocols for allograft rejection for the first 12 months after heart transplantation (Tx)

More recently, molecular techniques have been evaluated to determine whether gene expression profiles within the myocardium on a biopsy can improve diagnostic yields [8]. A microarray-based system has been developed to assess EMB specimens (The Molecular Microscope®). Artificial intelligence techniques using principal component analysis and archetype analysis of rejection-associated transcripts determined in kidney transplants to be associated with AMR or ACR or both was assessed in EMBs and compared with both histological diagnoses and presence of donor-specific antibody. The initial experience is highly favorable with the test being able to estimate the probability and distinguish AMR from ACR. Loupy further determined distinct transcripts associated with AMR, including natural killer cell transcripts, macrophage transcripts, endothelial activation transcripts, and interferon gamma transcripts correlating closely with histological International Society of Heart and Lung Transplantation (ISHLT) pathology antibody mediated rejection (pAMR) grades [9]. This technology raises the potential to make a rejection diagnosis from a single biopsy core, improve histology systems, open the potential for automation, and provide quantitative results. It may also provide insights into pathologic processes and a personalized medicine approach to therapies.

While these advances show promise in increasing the diagnostic yield of the EMB, the search for a non-invasive blood-based test to assess allograft rejection remains very attractive, potentially reducing cost and complications and allowing more frequent testing to assess response to anti-rejection therapy.

Donor-specific antibodies

Both the presence of donor specific antibodies (DSA) at the time of transplant and the development of de novo DSA after transplant have been associated with adverse post-transplant outcomes, including rejection, cardiac allograft vasculopathy, and graft loss [10, 11]. However, even though DSA may be seen in up to a third of post-transplant recipients, the presence of DSA does not necessarily predict development of AMR, although the presence of class II antibodies do appear to correlate with increased risk of graft loss [12] even late after transplant [13]. Current guidelines from the ISHLT recommend routine monitoring for anti-human leucocyte antigen (anti-HLA) DSA post-transplant at 1, 3, 6, and 12 months [14].

While the data is less clear, non-HLA DSA have been implicated in acute and chronic allograft rejection. The antigens detected include major histocompatibility class I chain-related gene A (MICA), angiotensin II type 1 receptor (ATIR), endothelin type A receptor (ET_AR) endothelial cell antigens, vimentin, K-alpha-1-Tubulin (Kα1T), collagen-V, and non-HLA IgM antibodies [15, 16]. There is some evidence implicating anti-MICA and anti-endothelial antibodies in the development of AMR [17] and development of cardiac allograft vasculopathy [18]. The development of de novo HLA DSA in combination with AT1R antibodies may be particularly deleterious with regard to the development of rejection following heart transplantation [19].

However, at this time, the detection of these antibodies in the setting of acute and chronic rejection remains investigational. The clinical significance and the role of non-HLA antibodies in mediating cardiac allograft injury remain chiefly undetermined.

Markers of tissue injury and cardiac function

A number of early biomarkers associated with microvascular thrombosis (microvascular fibrin deposition, depletion of vascular tissue plasminogen activator, endothelial activation, and loss of vascular antithrombin), as well as detection of serum cardiac troponins and elevated serum soluble adhesion molecule levels, have been implicated in the later development of cardiac allograft vasculopathy [20]. In a recent review of studies assessing the utility of cardiac troponin to detect ACR, only high-sensitivity assays were determined to have a high sensitivity (82–100%) and negative predictive value (97–100%) [21]. The high negative predictive value of late generation assays may therefore be particularly useful clinically in ruling out rejection.

In a prospective biomarker study, assessment of biomarkers associated with myocyte injury (troponins), cardiac function (natriuretic peptides), fibrosis (galectin-3 and ST-2), and cardiorenal function (Cystatin-c) at 10 days after heart transplantation was correlated with 1-year outcomes [22]. Only cystatin C and high sensitivity troponin T (hsTnT) predicted 1-year mortality. Additionally, these two biomarkers in conjunction with galectin-3 were elevated in patients with decreased left ventricular ejection fraction (LVEF) measured 1 year after heart transplant. N-terminal pro B-type natriuretic peptide (NT-proBNP) did not show early prognostic power. In this study, none of the measured biomarkers predicted rejection, but hsTnT and NT-proBNP were increased significantly 12 months after heart transplant in patients with at least one rejection. A number of studies suggest that absolute and changes in serial natriuretic peptide measurements in adult and pediatric heart transplant recipients predict acute allograft rejection [23–25].

Gene expression profile

The first clinically available non-invasive blood test for assessment of heart transplant rejection utilizes GEP (AlloMap®, CareDx, Brisbane, CA). In an analysis of known alloimmune pathways and leukocyte microarrays, 252 candidate genes were identified for which real-time polymerase chain reaction (PCR) assays were developed [26]. From this, an 11-gene real-time PCR test was derived from a training set using linear discriminant analysis and converted into a score (0–40). This was subsequently validated prospectively in an independent set. The test distinguished biopsy-defined moderate/severe cellular rejection from quiescence (p = 0.0018) in the validation set and had agreement of 84% with grade ISHLT ≥ 3A rejection. Patients more than 1-year post-transplant who had scores below 30 had a 99.6% negative predictive value for ISHLT grade ≥ 3A rejection. A European-based (CARGO II) study subsequently validated the performance of the test [27], confirming the high negative predictive value but very low positive predictive value of GEP testing. A randomized study assigned more than 600 patients at low risk for rejection who had undergone cardiac transplantation 6 months to 5 years previously to be monitored for rejection with the use of GEP or with the use of routine EMB, in addition to clinical and echocardiographic assessment of graft function [28]. GEP was non-inferior with respect to the composite primary outcome of rejection with hemodynamic compromise, graft dysfunction due to other causes, death, or re-transplantation. In a subsequent smaller single-center study of 60 patients (EIMAGE), selected patients were randomized earlier after transplant (at 55 days) to GEP or surveillance biopsies [29] (Fig. 1a,b). GEP was again comparable with EMB for rejection surveillance and did not result in increased adverse outcomes. GEP was also shown to be useful to guide corticosteroid weaning and comparable to EMB for this purpose (Fig. 1c, d). The use of GEP (AlloMap®) has been included in the ISHLT guidelines for the management of patients after heart transplantation and has been useful in decreasing the number of protocol biopsies for rejection monitoring [30].

There are some important caveats regarding the use of GEP. The test has not been validated for patients transplanted less than 55 days. It should be avoided if patients are on > 20 mg of prednisone per day (as some of the tested genes are steroid responsive). The test should be avoided within 21 days of treated rejection and within 30 days of blood transfusion. Importantly, GEP has not been tested in patients with AMR. The EMB therefore remains the test of choice for surveillance in the first 3 months following transplantation, when the procedure is typically performed weekly for the first month and then every 2 weeks (Fig. 1a-d (all protocols)).

Cell-free DNA

An inevitable consequence of cellular turnover is the release of DNA into the circulation. Cell-free DNA (cfDNA) is released from healthy, inflamed, or diseased tissue from cells undergoing apoptosis or necrosis. Circulating cfDNA has become an attractive area of research as a non-invasive diagnostic biomarker since its discovery in human blood plasma 70 years ago. In transplantation, the detection of donor-derived cfDNA (dd-cfDNA) in the blood of transplant recipients has been proposed as a non-invasive marker of injury caused by both ACR and AMR. Single-center studies have demonstrated that elevated dd-cfDNA levels may be detected after acute rejection in heart transplantation, and increased levels precede histological rejection on EMB [31–33].

A number of techniques have been used to detect dd-cfDNA. The initial approach was to use shotgun whole-genome sequencing to detect and quantify cfDNA, but the complexity and cost of the analyses ultimately limit its application as a clinically useful tool. A more rapid and cost-effective surveillance strategy uses targeted quantification of dd-cfDNA, but this requires genotyping of the transplant donor, which is impractical. Targeted amplification of single-nucleotide polymorphisms using next generation sequencing has been recently validated to quantify dd-cfDNA in transplant recipients without the need for donor or recipient genotyping [34]. This technique has been shown to detect ACR in heart transplant recipients in a multicenter retrospective case control study [34]. In renal transplantation, a prospective multicenter trial demonstrated efficacy of this technique in detecting acute rejection [35].

The donor-derived cell-free DNA outcomes AlloMap registry (D-OAR) examined the characteristics and performance of dd-cfDNA in a routine, clinical surveillance setting in 760 heart transplant recipients at 26 centers across the USA [36]. The dd-cfDNA levels were paired to biopsy-based diagnosis of rejection. The median dd-cfDNA was 0.07% in reference heart transplant recipients and 0.17% in samples classified as acute rejection (p = .005). At a 0.2% threshold, dd-cfDNA had a 44% sensitivity to detect rejection and a 97% negative predictive value. In separate cohort at risk for AMR, defined as patients with pre-transplant panel reactive antibody (PRA) ≥ 10%, presence of donor-specific antibodies at the time of transplant, any post-transplant detection of DSA or biopsy-proven AMR, or for-cause biopsy due to reduced left ventricular ejection fraction, dd-cfDNA levels were elevated 3-fold in AMR compared with patients without AMR (p = .004). Current ongoing analysis in the D-OAR registry will determine the predictive value of combining dd-cfDNA testing with AlloMap to improve the performance of these tests.

MicroRNAs

Micro-ribonucleic acids (mi-RNA) are highly conserved small non-coding 21–25 base single-stranded RNAs that regulate gene expression by binding to complementary messenger RNA (mRNA) nucleotide sequences. Approximately, 50% of all protein-coding genes are under the control of miRNAs. A single mRNA may be regulated by multiple miRNAs and conversely, a single miRNA may regulate multiple mRNAs [37]. Detection of miRNAs has a potential in the diagnosis of allograft rejection in heart transplantation. In one study, the role of 14 pre-selected miRNAs in 113 human cardiac transplant recipients was evaluated. Four miRNAs were identified (miR-10a, miRs-31, miRs-92a, and miRs-155) that were differentially regulated in both the tissue and serum of patients with allograft rejection [38]. These individual miRNAs had an excellent ability to diagnose allograft rejection with an area under the curve (AUC) ranging between 0.87 and 0.98 and were also able to accurately differentiate ACR and AMR. It has yet to be determined whether the technique can distinguish clinically relevant rejection from mild rejection or whether concurrent processes such as infection may affect results. It remains unclear if serum miRNA (as opposed to tissue miRNA) expression alone is sensitive enough to reliably identify clinically significant rejection.

In another study, a group of heart transplant recipients with histologically verified ACR were compared with a control group without allograft rejection by assessing the levels of a selected set of miRNAs in serum specimens. Levels of seven miRNAs (miR-142-3p, miR-101-3p, miR-424-5p, miR-27a-3p, miR-144-3p, miR-339-3p, and miR-326) were significantly higher in ACR group compared with the control group [39].

Beyond allograft rejection, miRNAs are potential biomarkers of cardiac allograft vasculopathy (CAV). Endothelium-derived miRNAs have been identified in patients with CAV [40]. Sensitivity and specificity of these biomarkers compared with coronary angiography or intravascular ultrasound still need to be validated in larger cohorts, and it is unclear whether these miRNAs can prospectively detect asymptomatic post-transplant patients at risk for the development of CAV.

T cell immune function

The challenge in management of transplant recipients is to balance immunosuppression, avoiding either too much or too little, so as to prevent complications such as infections, malignancy, and drug toxicity on one hand and acute or chronic rejection of the graft on the other hand. Therapeutic drug monitoring alone is inadequate to measure the immune response, and their primary value is for preventing drug toxicity. The ImmuKnow® assay (Viracor-Eurofins, Lee’s Summit, Missouri) detects global CD4+ T cell-mediated immunity and helps identify patients at risk for infection and rejection. This assay is performed on peripheral blood and measures adenosine triphosphatase (ATP) release from activated lymphocytes and helps predict the immune state. It has been useful in assessing infection and rejection risk in solid-organ transplants, including kidney, pancreas, and small bowel. A meta-analysis of more than 500 solid organ transplant recipients (heart, kidney, kidney-pancreas, liver, and small bowel) from 10 US centers was performed using the ImmuKnow® assay [41]. Blood samples were taken from recipients at various times post-transplant and compared with clinical course. In this analysis, 39 biopsy-proven cellular rejections and 66 diagnosed infections occurred. A recipient with an immune response value of 25-ng/ml ATP was 12 times more likely to develop an infection than a recipient with a stronger immune response. Similarly, a recipient with an immune response of 700-ng/ml ATP was 30 times more likely to develop ACR than a recipient with a lower immune response value. The intersection of odds ratio curves for infection and rejection in the moderate immune response zone occurred at 280-ng/ml ATP providing an immunological target of immune function for solid organ recipients.

In a single center study of almost 300 heart transplant recipients, a total of 864 immune monitoring assays were performed between 2 weeks and 10 years post-transplant. There were 38 infectious episodes and 8 rejection episodes. The average immune monitoring score was significantly lower during infection than steady state (187- vs 280-ng ATP/ml, p < 0.001). The association between high immune monitoring scores and rejection risk was inconclusive in this study due to the small number of rejection episodes [42].

Clinically, the assay is most useful in determining extent of immunosuppression during active infection and can guide reduction of immunosuppression if the immune monitoring score is low (especially if < 100-ng ATP/ml). Patients in whom immunosuppression dosing is limited by drug intolerance, ImmuKnow® may provide determination of efficacy of immunosuppression beyond therapeutic drug monitoring. If the test is performed in asymptomatic patients, a score < 100-ng ATP/ml can prompt reduction in immunosuppression given the high intendant risk of infection, and elevated scores (> 700-ng ATP/ml) may require consideration for fortification of immunosuppression. In the early post-transplant period, routine use of ImmuKnow® is of limited value, as enhanced immunosuppression frequently leads to low scores.

The test does not however appear to predict risk of infection or rejection in pediatric heart transplant recipients [43, 44].

Conclusions

The advent of effective immunosuppression facilitated the development of cardiac transplant as a clinically useful strategy for the treatment of advanced heart disease. Since its inception however, the challenge has been to balance immunosuppression so as to minimize allograft rejection, drug toxicity, infection, and malignancy. The development of the endomyocardial biopsy was a significant advance in the monitoring and management of rejection. However, due to its invasive nature, there has been significant interest in the development of blood-based immunological monitoring after transplant. However, these tests have demonstrated limitations and no single test has demonstrated superiority, with most methodologies reaching the clinical arena demonstrating high negative predictive values.

Biomarkers of cardiac injury and function may provide incremental data to help determine post-transplant graft function but alone have limited discriminatory power. GEP is used clinically and has decreased the need to perform protocol biopsies for rejection monitoring in some heart transplant recipients. It is however not validated for AMR. Monitoring of DSA helps risk stratify patients at risk for AMR, and dd-cfDNA may help identify patients at risk of both ACR and AMR before histological changes are noted on biopsy. Assessment of circulating miRNA is a promising strategy which requires prospective validation and is being evaluated as promising biomarkers for both acute rejection and CAV. The ImmuKnow® test has been clinically useful in adult heart transplant recipients predominantly for assessing over-immunosuppression and risk of infection. The development of new technologies for post-transplant monitoring has been slow, chiefly due to limited numbers of patients participating in studies and low event rates, especially rejection, but the rapid development of molecular technologies, which have led to better understanding of transplant immunology, holds promise for a bright future in this arena.

Funding

Not applicable being a review article.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Informed consent

Not applicable being a review Article.

Ethics committee approval and human and animal rights statement

Not applicable being a review Article.