Advanced Morphologic Analysis for Diagnosing Allograft Rejection: The Case of Cardiac Transplant Rejection

Computational Pathology Assessments of Cardiac Stromal Remodeling: Clinical Correlates and Prognostic Implications in Heart Transplantation

Both overt and indolent inflammatory insults in heart transplantation can accelerate pathologic cardiac remodeling, but there are few tools for monitoring the speed and severity of remodeling over time. To address this need, we developed an automated computational pathology system to measure pathologic remodeling in transplant biopsy samples in a large, retrospective cohort of n=2167 digitized heart transplant biopsy slides. Biopsy images were analyzed to identify the pathologic stromal changes associated with future allograft loss or advanced allograft vasculopathy. Biopsy images were then analyzed to assess which historical allo-inflammatory events drive progression of these pathologic stromal changes over time in serial biopsy samples. The top-5 features of pathologic stromal remodeling most strongly associated with adverse outcomes were also strongly associated with histories of both overt and indolent inflammatory events. Our findings identify previously unappreciated subgroups of higher- and lower-risk transplant patients, and highlight the translational potential of digital pathology analysis.

Failing to Make the Grade: Conventional Cardiac Allograft Rejection Grading Criteria Are Inadequate for Predicting Rejection Severity

BACKGROUND

Cardiac allograft rejection is the leading cause of early graft failure and is a major focus of postheart transplant patient care. While histological grading of endomyocardial biopsy samples remains the diagnostic standard for acute rejection, this standard has limited diagnostic accuracy. Discordance between biopsy rejection grade and patient clinical trajectory frequently leads to both overtreatment of indolent processes and delayed treatment of aggressive ones, spurring the need to investigate the adequacy of the current histological criteria for assessing clinically important rejection outcomes.

METHODS

N=2900 endomyocardial biopsy images were assigned a rejection grade label (high versus low grade) and a clinical trajectory label (evident versus silent rejection). Using an image analysis approach, n=370 quantitative morphology features describing the lymphocytes and stroma were extracted from each slide. Two models were constructed to compare the subset of features associated with rejection grades versus those associated with clinical trajectories. A proof-of-principle machine learning pipeline-the cardiac allograft rejection evaluator-was then developed to test the feasibility of identifying the clinical severity of a rejection event.

RESULTS

The histopathologic findings associated with conventional rejection grades differ substantially from those associated with clinically evident allograft injury. Quantitative assessment of a small set of well-defined morphological features can be leveraged to more accurately reflect the severity of rejection compared with that achieved by the International Society of Heart and Lung Transplantation grades.

CONCLUSIONS

Conventional endomyocardial samples contain morphological information that enables accurate identification of clinically evident rejection events, and this information is incompletely captured by the current, guideline-endorsed, rejection grading criteria.

Digital pathology in cardiac transplant diagnostics: from biopsies to algorithms

In the field of heart transplantation, the ability to accurately and promptly diagnose cardiac allograft rejection is crucial. This comprehensive review explores the transformative role of digital pathology and computational pathology, especially through machine learning, in this critical domain. These methodologies harness large datasets to extract subtle patterns and valuable information that extend beyond human perceptual capabilities, potentially enhancing diagnostic outcomes. Current research indicates that these computer-based systems could offer accuracy and performance matching, or even exceeding, that of expert pathologists, thereby introducing more objectivity and reducing observer variability. Despite promising results, several challenges such as limited sample sizes, diverse data sources, and the absence of standardized protocols pose significant barriers to the widespread adoption of these techniques. The future of digital pathology in heart transplantation diagnostics depends on utilizing larger, more diverse patient cohorts, standardizing data collection, processing, and evaluation protocols, and fostering collaborative research efforts. The integration of various data types, including clinical, demographic, and imaging information, could further refine diagnostic precision. As researchers address these challenges and promote collaborative efforts, digital pathology has the potential to become an integral part of clinical practice, ultimately improving patient care in heart transplantation.

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

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.

Non-invasive detection of cardiac allograft rejection among heart transplant recipients using an electrocardiogram based deep learning model

Aims

Current non-invasive screening methods for cardiac allograft rejection have shown limited discrimination and are yet to be broadly integrated into heart transplant care. Given electrocardiogram (ECG) changes have been reported with severe cardiac allograft rejection, this study aimed to develop a deep-learning model, a form of artificial intelligence, to detect allograft rejection using the 12-lead ECG (AI-ECG).

Methods and results

Heart transplant recipients were identified across three Mayo Clinic sites between 1998 and 2021. Twelve-lead digital ECG data and endomyocardial biopsy results were extracted from medical records. Allograft rejection was defined as moderate or severe acute cellular rejection (ACR) based on International Society for Heart and Lung Transplantation guidelines. The extracted data (7590 unique ECG-biopsy pairs, belonging to 1427 patients) was partitioned into training (80%), validation (10%), and test sets (10%) such that each patient was included in only one partition. Model performance metrics were based on the test set (n = 140 patients; 758 ECG-biopsy pairs). The AI-ECG detected ACR with an area under the receiver operating curve (AUC) of 0.84 [95% confidence interval (CI): 0.78-0.90] and 95% (19/20; 95% CI: 75-100%) sensitivity. A prospective proof-of-concept screening study (n = 56; 97 ECG-biopsy pairs) showed the AI-ECG detected ACR with AUC = 0.78 (95% CI: 0.61-0.96) and 100% (2/2; 95% CI: 16-100%) sensitivity.

Conclusion

An AI-ECG model is effective for detection of moderate-to-severe ACR in heart transplant recipients. Our findings could improve transplant care by providing a rapid, non-invasive, and potentially remote screening option for cardiac allograft function.

The promise of machine learning applications in solid organ transplantation

Solid-organ transplantation is a life-saving treatment for end-stage organ disease in highly selected patients. Alongside the tremendous progress in the last several decades, new challenges have emerged. The growing disparity between organ demand and supply requires optimal patient/donor selection and matching. Improvements in long-term graft and patient survival require data-driven diagnosis and management of post-transplant complications. The growing abundance of clinical, genetic, radiologic, and metabolic data in transplantation has led to increasing interest in applying machine-learning (ML) tools that can uncover hidden patterns in large datasets. ML algorithms have been applied in predictive modeling of waitlist mortality, donor–recipient matching, survival prediction, post-transplant complications diagnosis, and prediction, aiming to optimize immunosuppression and management. In this review, we provide insight into the various applications of ML in transplant medicine, why these were used to evaluate a specific clinical question, and the potential of ML to transform the care of transplant recipients. 36 articles were selected after a comprehensive search of the following databases: Ovid MEDLINE; Ovid MEDLINE Epub Ahead of Print and In-Process & Other Non-Indexed Citations; Ovid Embase; Cochrane Database of Systematic Reviews (Ovid); and Cochrane Central Register of Controlled Trials (Ovid). In summary, these studies showed that ML techniques hold great potential to improve the outcome of transplant recipients. Future work is required to improve the interpretability of these algorithms, ensure generalizability through larger-scale external validation, and establishment of infrastructure to permit clinical integration.

Relationship between donor fraction cell-free DNA and clinical rejection in heart transplantation

BACKGROUND

Clinical rejection (CR) defined as decision to treat clinically suspected rejection with change in immunotherapy based on clinical presentation with or without diagnostic biopsy findings is an important part of care in heart transplantation. We sought to assess the utility of donor fraction cell-free DNA (DF cfDNA) in CR and the utility of serial DF cfDNA in CR patients in predicting outcomes of clinical interest.

METHODS

Patients with heart transplantation were enrolled in two sequential, multi-center, prospective observational studies. Blood samples were collected for surveillance or clinical events. Clinicians were blinded to the results of DF cfDNA.

RESULTS

A total of 835 samples from 269 subjects (57% pediatric) were included for this analysis, including 28 samples associated with CR were analyzed. Median DF cfDNA was 0.43 (IQR 0.15, 1.36)% for CR and 0.10 (IQR 0.07, 0.16)% for healthy controls (p < .0001). At cutoff value of 0.13%, the area under curve (AUC) was 0.82, sensitivity of 0.86, specificity of 0.67, and negative predictive value of 0.99. There was serial decline in DF cfDNA post-therapy, however, those with cardiovascular events (cardiac arrest, need for mechanical support or death) showed significantly higher levels of DF cfDNA on Day 0 (2.11 vs 0.31%) and Day 14 (0.51 vs 0.22%) compared to those who did not have such an event (p < .0001).

CONCLUSION

DF cfDNA has excellent agreement with clinical rejection and, importantly, serial measurement of DF cfDNA predict clinically significant outcomes post treatment for rejection in these patients.

Computational Analysis of Routine Biopsies Improves Diagnosis and Prediction of Cardiac Allograft Vasculopathy

Background: Cardiac allograft vasculopathy (CAV) is a leading cause of morbidity and mortality for heart transplant recipients. While clinical risk factors for CAV have been established, no personalized prognostic test exists to confidently identify patients at high vs. low risk of developing aggressive CAV. The aim of this investigation was to leverage computational methods for analyzing digital pathology images from routine endomyocardial biopsies (EMB) to develop a precision medicine tool for predicting CAV years before overt clinical presentation. Methods: Clinical data from 1-year post-transplant was collected on 302 transplant recipients from the University of Pennsylvania, including 53 'early CAV' patients and 249 'no-CAV' controls. This data was used to generate a 'clinical model' (ClinCAV-Pr) for predicting future CAV development. From this cohort, n=183 archived EMBs were collected for CD31 and modified trichrome staining and then digitally scanned. These included 1-year post-transplant EMBs from 50 'early CAV' patients and 82 no-CAV patients, as well as 51 EMBs from 'disease control' patients obtained at the time of definitive coronary angiography confirming CAV. Using biologically-inspired, hand-crafted features extracted from digitized EMBs, quantitative histologic models for differentiating no-CAV from disease controls (HistoCAV-Dx), and for predicting future CAV from 1-year post-transplant EMBs were developed (HistoCAV-Pr). The performance of histologic and clinical models for predicting future CAV (i.e. HistoCAV-Pr and ClinCAV-Pr, respectively) were compared in a held-out validation set, before being combined to assess the added predictive value of an integrated predictive model (iCAV-Pr). Results: ClinCAV-Pr achieved modest performance on the independent test set, with area under the receiver operating curve (AUROC) of 0.70. The HistoCAV-Dx model for diagnosing CAV achieved excellent discrimination, with an AUROC of 0.91, while HistoCAV-Pr model for predicting CAV achieved good performance with an AUROC of 0.80. The integrated iCAV-Pr model achieved excellent predictive performance, with an AUROC of 0.93 on the held-out test set. Conclusions: Prediction of future CAV development is greatly improved by incorporation of computationally extracted histologic features. These results suggest morphologic details contained within regularly obtained biopsy tissue have the potential to enhance precision and personalization of treatment plans for post-heart transplant patients.

Deep learning-enabled assessment of cardiac allograft rejection from endomyocardial biopsies

Endomyocardial biopsy (EMB) screening represents the standard of care for detecting allograft rejections after heart transplant. Manual interpretation of EMBs is affected by substantial interobserver and intraobserver variability, which often leads to inappropriate treatment with immunosuppressive drugs, unnecessary follow-up biopsies and poor transplant outcomes. Here we present a deep learning-based artificial intelligence (AI) system for automated assessment of gigapixel whole-slide images obtained from EMBs, which simultaneously addresses detection, subtyping and grading of allograft rejection. To assess model performance, we curated a large dataset from the United States, as well as independent test cohorts from Turkey and Switzerland, which includes large-scale variability across populations, sample preparations and slide scanning instrumentation. The model detects allograft rejection with an area under the receiver operating characteristic curve (AUC) of 0.962; assesses the cellular and antibody-mediated rejection type with AUCs of 0.958 and 0.874, respectively; detects Quilty B lesions, benign mimics of rejection, with an AUC of 0.939; and differentiates between low-grade and high-grade rejections with an AUC of 0.833. In a human reader study, the AI system showed non-inferior performance to conventional assessment and reduced interobserver variability and assessment time. This robust evaluation of cardiac allograft rejection paves the way for clinical trials to establish the efficacy of AI-assisted EMB assessment and its potential for improving heart transplant outcomes.

Noninvasive methods to reduce cardiac complications postheart transplant

PURPOSE OF REVIEW

Long-term success of heart transplantation is limited by allograft rejection and cardiac allograft vasculopathy (CAV). Classic management has relied on frequent invasive testing to screen for early features of rejection and CAV to allow for early treatment. In this review, we discuss new developments in the screening and prevention of allograft rejection and CAV.

RECENT FINDINGS

Newer noninvasive screening techniques show excellent sensitivity and specificity for the detection of clinically significant rejection. New biomarkers and treatment targets continue to be identified and await further studies regarding their utility in preventing allograft vasculopathy.

SUMMARY

Noninvasive imaging and biomarker testing continue to show promise as alternatives to invasive testing for allograft rejection. Continued validation of their effectiveness may lead to new surveillance protocols with reduced frequency of invasive testing. Furthermore, these noninvasive methods will allow for more personalized strategies to reduce the complications of long-term immunosuppression whereas continuing the decline in the overall rate of allograft rejection.

Deep Neural Network-Aided Histopathological Analysis of Myocardial Injury

Deep neural networks have become the mainstream approach for analyzing and interpreting histology images. In this study, we established and validated an interpretable DNN model to assess endomyocardial biopsy (EMB) data of patients with myocardial injury. Deep learning models were used to extract features and classify EMB histopathological images of heart failure cases diagnosed with either ischemic cardiomyopathy or idiopathic dilated cardiomyopathy and non-failing cases (organ donors without a history of heart failure). We utilized the gradient-weighted class activation mapping (Grad-CAM) technique to emphasize injured regions, providing an entry point to assess the dominant morphology in the process of a comprehensive evaluation. To visualize clustered regions of interest (ROI), we utilized uniform manifold approximation and projection (UMAP) embedding for dimension reduction. We further implemented a multi-model ensemble mechanism to improve the quantitative metric (area under the receiver operating characteristic curve, AUC) to 0.985 and 0.992 on ROI-level and case-level, respectively, outperforming the achievement of 0.971 ± 0.017 and 0.981 ± 0.020 based on the sub-models. Collectively, this new methodology provides a robust and interpretive framework to explore local histopathological patterns, facilitating the automatic and high-throughput quantification of cardiac EMB analysis.

The Role of Machine Learning in Cardiovascular Pathology

Machine learning has seen slow but steady uptake in diagnostic pathology over the past decade to assess digital whole-slide images. Machine learning tools have incredible potential to standardise, and likely even improve, histopathologic diagnoses, but they are not yet widely used in clinical practice. We describe the principles of these tools and technologies and some successful preclinical and pretranslational efforts in cardiovascular pathology, as well as a roadmap for moving forward. In nonhuman animal models, one proof-of-principle application is in rodent progressive cardiomyopathy, which is of particular significance to drug toxicity studies. Basic science successes include screening the quality of differentiated stem cells and characterising cardiomyocyte developmental stages, with potential applications for research and toxicology/drug safety screening using derived or native human pluripotent stem cells differentiated into cardiomyocytes. Translational studies of particular note include those with success in diagnosing the various forms of heart allograft rejection. For fully realising the value of these tools in clinical cardiovascular pathology, we identify 3 essential challenges. First is image quality standardisation to ensure that algorithms can be developed and implemented on robust, consistent data. The second is consensus diagnosis; experts don't always agree, and thus "truth" may be difficult to establish, but the algorithms themselves may provide a solution. The third is the need for large-enough data sets to facilitate robust algorithm development, necessitating large cross-institutional shared image databases. The power of histopathology-based machine learning technologies is tremendous, and we outline the next steps needed to capitalise on this power.

Molecular Diagnosis of Rejection in Heart Transplantation

Despite the overall success of heart transplantation as a definitive treatment for endstage heart failure, cardiac allograft rejection remains an important cause of morbidity and mortality. Endomyocardial biopsy has been the standard of care for rejection monitoring, but is associated with several diagnostic limitations and serious procedural complications. The use of molecular diagnostics has emerged over the past decade as a tool to potentially circumvent some of these limitations. We present an update on novel molecular approaches to detecting transplant rejection, focusing on 4 categories: microarray technology, gene expression profiling, cell-free DNA and microRNA.

Prediction of 1-year mortality after heart transplantation using machine learning approaches: A single-center study from China

BACKGROUND

Heart transplantation (HTx) remains the gold-standard treatment for end-stage heart failure. The aim of this study was to establish a risk-prediction model for assessing prognosis of HTx using machine-learning approach.

METHODS

Consecutive recipients of orthotopic HTx at our institute between January 1st, 2015 and December 31st, 2018 were included in this study. The primary outcome was 1-year mortality. Least absolute shrinkage and selection operator method was used to select variables and seven different machine-learning approaches were employed to develop the risk-prediction model. Bootstrap method was used for model validation. Shapley Additive exPlanations (SHAP) method was used for model interpretation.

RESULTS

381 recipients were included with average age of 43.783 years old. Albumin, recipient age and left atrium diameter ranked top three most important variables that affected the 1-year mortality of HTx. Other important variables included red blood cell, hemoglobin, lymphocyte%, smoking history, use of lyophilized rhBNP, use of Levosimendan, hypertension, cardiac surgery history, malignancy and endotracheal intubation history. Random Forest (RF) model achieved the best area under curves (AUC) of 0.801 and gradient boosting machine (GBM) showed the best sensitivity of 0.271. SHAP method was introduced to display the RF model's predicting processes of "survival" or "death" in individual level.

CONCLUSIONS

We established the risk-prediction model for postoperative prognosis of HTx patients by using machine learning method and demonstrated that the RF model performed the highest discrimination with the largest AUC when validated. This prediction model could help to recognize high-risk HTx recipients, provide personalized therapy plan and reduce organ wastage.

Markers of Immune Function in Heart Transplantation: Implications for Immunosuppression and Screening for Rejection

PURPOSE OF REVIEW

Recent developments in high-throughput DNA and RNA sequencing technologies have facilitated the development of noninvasive assays to monitor heart transplant rejection. In this review, we summarize existing assays employed for the surveillance of allograft rejection, as well as promising future directions for such tests in the molecular biology field.

RECENT FINDINGS

The AlloMap genome expression profiling assay remains the only noninvasive test for rejection surveillance and is incorporated into the International Society of Heart and Lung Transplantation guidelines. Other efforts have focused on messenger RNA (mRNA), microRNA (miRNA), and donor-derived cell-free DNA (dd-cfDNA) as potential viable biomarkers. Mitochondrial pathways in allograft necroptosis and inflammation signaling may represent a novel direction for future research endeavors. Although endomyocardial biopsy remains the gold standard, several converging areas of molecular biology could soon yield successful alternative methods of heart transplant rejection monitoring, with the distinct advantage of avoiding procedural complications.

The XVth Banff Conference on Allograft Pathology the Banff Workshop Heart Report: Improving the diagnostic yield from endomyocardial biopsies and Quilty effect revisited

The XVth Banff Conference on Allograft Pathology meeting was held on September 23-27, 2019, in Pittsburgh, Pennsylvania, USA. During this meeting, two main topics in cardiac transplant pathology were addressed: (a) Improvement of endomyocardial biopsy (EMB) accuracy for the diagnosis of rejection and other significant injury patterns, and (b) the orphaned lesion known as Quilty effect or nodular endocardial infiltrates. Molecular technologies have evolved in recent years, deciphering pathophysiology of cardiac rejection. Diagnostically, it is time to integrate the histopathology of EMBs and molecular data. The goal is to incorporate molecular pathology, performed on the same paraffin block as a companion test for histopathology, to yield more accurate and objective EMB interpretation. Application of digital image analysis from hematoxylin and eosin (H&E) stain to multiplex labeling is another means of extracting additional information from EMBs. New concepts have emerged exploring the multifaceted significance of myocardial injury, minimal rejection patterns supported by molecular profiles, and lesions of arteriolitis/vasculitis in the setting of T cell-mediated rejection (TCMR) and antibody-mediated rejection (AMR). The orphaned lesion known as Quilty effect or nodular endocardial infiltrates. A state-of-the-art session with historical aspects and current dilemmas was reviewed, and possible pathogenesis proposed, based on advances in immunology to explain conflicting data. The Quilty effect will be the subject of a multicenter project to explore whether it functions as a tertiary lymphoid organ.

In Situ Immune Profiling of Heart Transplant Biopsies Improves Diagnostic Accuracy and Rejection Risk Stratification

Recognizing that guideline-directed histologic grading of endomyocardial biopsy tissue samples for rejection surveillance has limited diagnostic accuracy, quantitative, in situ characterization was performed of several important immune cell types in a retrospective cohort of clinical endomyocardial tissue samples. Differences between cases were identified and were grouped by histologic grade versus clinical rejection trajectory, with significantly increased programmed death ligand 1+, forkhead box P3+, and cluster of differentiation 68+ cells suppressed in clinically evident rejections, especially cases with marked clinical-histologic discordance. Programmed death ligand 1+, forkhead box P3+, and cluster of differentiation 68+ cell proportions are also significantly higher in "never-rejection" when compared with "future-rejection." These findings suggest that in situ immune modulators regulate the severity of cardiac allograft rejection.

Cell-free DNA donor fraction analysis in pediatric and adult heart transplant patients by multiplexed allele-specific quantitative PCR: Validation of a rapid and highly sensitive clinical test for stratification of rejection probability

Lifelong noninvasive rejection monitoring in heart transplant patients is a critical clinical need historically poorly met in adults and unavailable for children and infants. Cell-free DNA (cfDNA) donor-specific fraction (DF), a direct marker of selective donor organ injury, is a promising analytical target. Methodological differences in sample processing and DF determination profoundly affect quality and sensitivity of cfDNA analyses, requiring specialized optimization for low cfDNA levels typical of transplant patients. Using next-generation sequencing, we previously correlated elevated DF with acute cellular and antibody-mediated rejection (ACR and AMR) in pediatric and adult heart transplant patients. However, next-generation sequencing is limited by cost, TAT, and sensitivity, leading us to clinically validate a rapid, highly sensitive, quantitative genotyping test, myTAI_HEART^®, addressing these limitations. To assure pre-analytical quality and consider interrelated cfDNA measures, plasma preparation was optimized and total cfDNA (TCF) concentration, DNA fragmentation, and DF quantification were validated in parallel for integration into myTAI_HEART reporting. Analytical validations employed individual and reconstructed mixtures of human blood-derived genomic DNA (gDNA), cfDNA, and gDNA sheared to apoptotic length. Precision, linearity, and limits of blank/detection/quantification were established for TCF concentration, DNA fragmentation ratio, and DF determinations. For DF, multiplexed high-fidelity amplification followed by quantitative genotyping of 94 SNP targets was applied to 1168 samples to evaluate donor options in staged simulations, demonstrating DF call equivalency with/without donor genotype. Clinical validation studies using 158 matched endomyocardial biopsy-plasma pairs from 76 pediatric and adult heart transplant recipients selected a DF cutoff (0.32%) producing 100% NPV for ≥2R ACR. This supports the assay’s conservative intended use of stratifying low versus increased probability of ≥2R ACR. myTAI_HEART is clinically validated for heart transplant recipients ≥2 months old and ≥8 days post-transplant, expanding opportunity for noninvasive transplant rejection assessment to infants and children and to all recipients >1 week post-transplant.