Precision Medicine

The SARS-CoV-2 virus has infected at least 60 million people in the United States and resulted in the deaths of over a million. SARS-CoV-2 is a respiratory virus that causes mild-to-severe flu-like symptoms in most patients but severe respiratory distress syndrome (acute respiratory distress syndrome) in others. Surprisingly, one of the earliest findings reported from Wuhan, China, was that a significant number of patients had signs and symptoms consistent with injury of the heart, as evidenced by the fact that 25% of patients had an increase in serum troponin (troponin T [TnT]), a robust marker of cell damage.¹ Similarly, a study of 2,736 U.S. patients found that even small amounts of TnT leakage were associated with a higher risk of death even after adjusting for clinical factors.² That the elevated TnT reflected activation of the cardiac inflammasome was supported by the finding that CD4+ and CD8+ T lymphocytes were markedly lower in nonsurvivors than in survivors, and that proinflammatory cytokines including tumor necrosis factor–α were higher in nonsurvivors.³ While science has led to effective vaccines and antiviral medications, the diagnosis and treatment of COVID-19 myocarditis has remained an enigma.

Much of what we know about the pathobiology of viral myocarditis comes from studies of infections in mice because, like humans, mice have strain-specific responses to viruses, making the time from exposure to disease variable.⁴ For example, in susceptible strains of mice, an initial infection with a coxsackie virus leads to death within 4 days due to myocyte necrosis without histologically apparent myocarditis. In a different strain, the same virus can cause an initial inflammatory phase with macrophage activation and infiltration of mononuclear cells, resulting in persistent subacute myocarditis. Similarly, pathologic evaluation in humans with acute myocarditis has mirrored the inflammation, multifocal lymphocytic myocarditis, presence of inflammatory cells, and activation of the cytokine cascade seen in animal models. Although there are some postmortem autopsy data to identify the pathology associated with COVID-19 infections, there has been an absence of molecular and cellular data from endomyocardial biopsies.

The disease that has provided valuable information about the biology of myocarditis in humans is postpartum cardiomyopathy (PPCM). In 1990, we reported that 14 of 18 women with PPCM had a pathologic diagnosis of myocarditis on an initial endomyocardial biopsy (EMB).⁵ Ten of the 14 women received an immunosuppressive agent: 9 of them had subjective and objective improvement of their myocarditis on a second EMB 6 days later. In 2016, Ware et al⁶ tested the hypothesis that women with PPCM had a genetic predisposition to cardiac inflammation. They found that 15% of the women had truncating variants in at least 1 heart failure gene—the most common being the titin gene (TTN).⁶ Recent studies of PPCM have confirmed these initial results, and studies in patients with dilated cardiomyopathy (DCM) and in children with myocarditis have also shown genetic variants in other so-called heart failure genes including BAG3, DES, DMD, DSC2, DSP, TPM1, TEM3, and VCL.

Juxtaposed against these preclinical studies, there is no consensus regarding the clinical evaluation of patients with SARS-CoV-2 and signs and symptoms of heart failure. For example, the Task Force for the Management of COVID-19 of the European Society of Cardiology (2022)⁷ posited that while virus has been found early in the disease, studies performed later “have been less convincing,” leading them to propose that a definitive diagnosis of myocarditis “should be based on EMB or autopsy.” By contrast, a 2022 American College of Cardiology Consensus Decision Pathway on Cardiovascular Sequelae of COVID-19 recommended obtaining TnT levels and an echocardiogram in patients with a recent infection and a moderate or high suspicion of cardiac involvement. However, in those with a rising troponin and electrocardiographic or echocardiographic abnormalities, cardiac magnetic resonance (CMR) is recommended because it is “the most sensitive method to exclude ischemia and pre-existing cardiomyopathies while also confirming cardiac changes due to SARS-CoV-2.”⁸

Reliance on a single imaging finding such as CMR or on a biomarker such as cardiac troponin as the single diagnostic tool of choice has significant limitations for the following reasons: 1) no appropriately powered studies exist in which both EMB data and CMR data were obtained from the same patients at a comparable time point; 2) comparisons of diagnostic performance of CMR at different settings and/or using different equipment make cross-group comparisons difficult; 3) studies of CMR have followed patients for only a short time period; 4) in a study that purported to evaluate patients by CMR and EMB, no patients had biopsy-proven myocarditis; and 5) in a study in which all 57 of the participants had EMB evidence of myocarditis, CMR sensitivity was high for an “infarct-like pattern” but low for a “cardiomyopathic pattern” and very low for an “arrhythmic clinical presentation.”⁹

It is now well recognized that SARS-CoV-2 is a human pathogen that is associated with mild-to-moderate disease in most individuals but can be associated with severe disease in a few. While authoritative committees have opined based on CMR findings that SARS-CoV-2 can result in myocardial edema, fibrosis, and inflammation, the information on tissue characterization, microscopic anatomy, immunohistochemistry, predominant cell lines infiltrating the heart muscle, characterization of presence of SARS-CoV-2 in cardiac myocytes, and mechanisms of injury to the heart are lacking. In fact, it is possible that CMR is too sensitive in the diagnosis of myocarditis, which would account for the large number of CMR diagnosed cases yet a paucity of positive cases by EMBs. It is also easy to see that a practitioner not trained in the intricacies of CMR would be hard pressed to have a level of confidence in their own interpretation to make an informed clinical decision. Despite the fact that there is a pressing need for a comprehensive analysis of the cellular and molecular pathophysiology in the setting of myocarditis, it cannot go unnoticed that the number of centers and physicians that perform EMBs has declined due in large part to concerns over the safety of the procedure. However, a recent consensus statement from experts representing the heart societies from Japan, Europe, and the United Kingdom as well as a European retrospective study of 1,368 patients both opined that the risk of a major complication during or after an EMB was low when performed by highly experienced physicians at high-volume centers.^{10 , 11}

The EMB should stand at the center of any efforts to better understand the biology of myocarditis because it provides a binary definition of the pathology as well as tissue that can be evaluated using technologies never before available to cardiovascular scientists. For example, in a paradigm-changing collaborative study, teams in Germany and the United States used single-cell RNA sequencing to compare samples from human hearts with DCM or with arrhythmogenic cardiomyopathy to identify the cellular landscape and the transcriptional phenotype of each.¹² They found distinct differences within the right heart and left heart genotype-associated pathways as well as variability in gene expression at the single-cell level, suggesting that these variants could result in significant differences in therapeutic responses. They also leveraged machine learning to illuminate genotype specific molecular responses.

Genetics also appears to play a role in an individual’s response to a viral infection. In a study of patients with COVID-19 who developed cardiac failure and elevations in serum troponin, Brugada et al¹³ reported finding genetic variants in “heart failure genes” in nearly half of the patients with severe disease but only a rare case in those with nonsevere disease, suggesting a strong relationship between genotype and phenotype. Similarly, the release of free DNA (cell-free DNA) from damaged or dead cells has emerged as a potential biomarker of disease that appears capable of predicting outcomes, but also may provide additional diagnostic information due to the ability to discriminate between different sizes and subtypes.¹⁴ Circulating microRNAs also provide prognostic information about heart failure–related morbidity and mortality and their profiles appear to reflect neurohormonal signaling.^{15 , 16} All of these exciting areas of research would benefit by the creation of myocarditis (or heart muscle disease) tissue repositories that would curate ventricular tissue, DNA, and sera from multiple studies in ethnically and geographically distinct populations and provide a clinical service with high-volume EMB to which patients could be referred.

Viral infections have long been presumed to be etiologic factors in some patients who present with a nonhereditary DCM, but we now know that the genetic milieu also plays a significant role in disease. Because the time between infection and hospitalization can be extremely short, COVID-19 also provides a unique opportunity to identify novel disease-causing genes at the start of the disease. In an era when single cells can be extracted for molecular diagnostics, the EMB and other circulating molecular and genetic markers become not only a diagnostic tool, but also a source for information about the etiology and pathobiology of the disease. Identification of the unique phenotype of the infected heart coupled with artificial intelligence–enhanced molecular analysis will one day enable therapy for each new patient to be predicated on their precise genotype and not just on their similar phenotypes, thereby bringing precision and enhanced rigor to the treatment of heart failure patients.

Funding Support and Author Disclosures

Dr Feldman was the founder of and held equity in Renovacor Inc, a biotechnology company focused on gene therapy for patients with rare inherited cardiovascular disease that licensed a BAG3 portfolio of patents from Temple University. Renovacor was recently purchased by Rocket Pharmaceuticals and Rocket now funds work ongoing in the Feldman laboratory. No funds from Renovacor or Rocket Inc were used in the preparation of this paper. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.