Considering complexity in the genetic evaluation of dilated cardiomyopathy

Abstract

DCM is a cardiovascular disease of genetic etiology that causes substantial morbidity and mortality, but also presents considerable opportunity for disease mitigation and prevention in those at risk. Foundational to the process of caring for patients diagnosed with DCM is a clinical genetic evaluation, which always begins with a comprehensive family history and clinical evaluation. Genetic testing of the proband, the first patient identified in a family with DCM, within the context of genetic counseling is always indicated, regardless of whether the DCM is familial or nonfamilial. Clinical screening of at-risk family members is also indicated, as is cascade genetic testing for actionable variants found at genetic testing in the proband. Clinicians now have expansive panels with many genes available for DCM genetic testing, and the approaches used to evaluate rare variants to decide which are disease-causing continues to rapidly evolve. Despite these recent advances, only a minority of cases yield actionable variants, even in familial DCM where a genetic etiology is highly likely. This underscores that our knowledge of DCM clinical genetics remains incomplete, including variant interpretation and DCM genetic architecture. Emerging data suggest that the single-variant Mendelian disease model is insufficient to explain some DCM cases, and rather that multiple variants, both common and rare, and at times key environmental factors, interact to cause DCM. A simple model illustrating the intersection of DCM genetic architecture with environmental impact is provided.

INTRODUCTION

Cardiovascular disease, caused by an intersection of environmental and genetic factors, is the leading cause of mortality of men and women in the United States¹. Advances in knowledge about the genetic background of cardiovascular disease have provided approaches to manage genetic risk and environmental exposures and thereby reduce morbidity and mortality. Through use of genetic evaluation and testing, individuals can learn of their genetic predisposition and use this information to mitigate or prevent cardiovascular disease arising from a primary genetic cause.

Among genetic cardiovascular conditions is dilated cardiomyopathy (DCM). DCM, estimated to impact 1 in 250 individuals², is a disease of the myocardium characterized by left ventricular (LV) enlargement and systolic dysfunction. Of several causes of DCM, the most common diagnosis is ischemic. The next most common diagnosis assigned is idiopathic, and is applied when other usual clinically detectable causes (except genetic) have been excluded. Herein in this review, the term DCM is used in place of the more precise term “idiopathic DCM.” Some DCM patients can be shown to have familial DCM. With clinical screening of first-degree relatives following a DCM diagnosis in a patient, single-site studies have shown that 15-30% of families will have familial DCM.³ A multisite US study, the Dilated Cardiomyopathy Precision Medicine Study, is now evaluating the proportion of familial DCM in an anticipated 1300 DCM patients and their family members.⁴ Nevertheless, despite the fact that familial transmission strongly supports the conclusion that the family’s DCM has a genetic basis, actionable genetic cause is only identified in a minority of cases, suggesting that a comprehensive understanding of familial DCM genetics remains elusive, and that additional yet unknown genes or genetic mechanisms may be at play.

Nevertheless, the majority of patients with DCM, even with clinical screening of family members, do not appear to have clinically detectable familial disease. Hence, the persistent question of whether genetics also underlies non-familial DCM has remained without a definitive answer. Preliminary data has shown that the frequency of genetic cause identified in those with non-familial (“sporadic”) DCM is similar to those reported with familial DCM.^5,6 The hypothesis that most of DCM, whether familial or non-familial, has an underlying genetic basis is now being tested in the Dilated Cardiomyopathy Precision Medicine Study.^4,7

More broadly, a unifying concept of how all of genetic variation interacts to cause DCM, termed genetic architecture, is not yet complete. For example, preliminary data from the Dilated Cardiomyopathy Precision Medicine Study indicates that 20% of DCM probands have two or more rare variants that may plausibly be biologically relevant for the DCM phenotype.⁴ Furthermore, the impact of common variant genetic background on the DCM phenotype when combined with biologically relevant rare variants remains unexplored. Environmental impact from pregnancy, drugs or toxins has also been suggested to be relevant to the development of DCM. More recent literature supports the case that peripartum-, alcohol-, or chemotherapy-related DCM has a genetic background similar to that of idiopathic DCM.^8–11 Nevertheless, specific mechanisms of how and in whom these environmental risk factors interact with genetics to cause DCM remains minimally understood. Collectively, defining the genetic architecture of DCM while also considering environment impact remains an ongoing challenge and an emerging area of DCM research.

THE GENETIC EVALUATION OF DCM

Identifying underlying genetic causes of DCM helps manage disease more thoughtfully in the patient, termed a “proband” in a genetics pedigree, and offers the opportunity to define genetic risk in the proband’s family members in order to prevent disease development (Figure 1). Usually DCM is not identified until symptoms present, at which time the disease has progressed to late stage and the opportunity to reverse myocardial damage is reduced¹². If genetic cause is able to be identified in the proband, genetic risk of DCM can then be assessed in asymptomatic family members with genetic testing. Family members then have the opportunity to learn if they share a genetic predisposition so clinical screening can be implemented, as currently recommended^13,14. However, at present, clinical genetic testing still leaves many probands with DCM without a clear genetic diagnosis. These cases of uncertainty and complexity introduce challenges and need for precision medicine care.

Figure 1. The Genetic Evaluation of DCM.

The genetic evaluation of DCM begins with identification of the proband and initiating the guideline-based^13,14 family-based process of comprehensive phenotypic evaluation, family history assessment, genetic counseling, and genetic testing process. Surveillance and management of family members is largely dependent on the genetic testing outcome and risk assessment of the proband. For those who are phenotype positive, medical and device therapy should proceed according to guidelines.

Multi-disciplinary centers specializing in cardiovascular genetics have been described as the optimal approach to evaluating genetic cardiovascular disease^15–17, including a cardiologist with genetic expertise in collaboration with a cardiovascular genetic counselor and/or medical geneticist. Further, the family should be treated as the unit of care¹³. While a family-based model is generally more common in the care of genetic disease, cardiology historically tends to center on the proband. However, given the family-level implications in the genetic diagnosis of DCM, treating the family as the unit of care maximizes the impact of the genetic evaluation.

Phenotypic Considerations

Family History and Pedigree Construction.

The first step and a key component of a complete genetic evaluation for DCM is a comprehensive family history that includes at least three generations placed into a pedigree format. Other aspects, particularly family-based specifics as described below, should also be included.^13,18 The pedigree serves as a critical tool to define the inheritance pattern, phenotype(s) of the family, and to identify at-risk family members. It is also a helpful family counseling tool to visually explain the biologically related family members at-risk for the proband’s presumptive genetic disease.

Phenotype Clarification.

Clearly defining the phenotype is also a critical first step in genetic analysis. Dilated cardiomyopathy is a major subtype of cardiomyopathy among others, including hypertrophic, restrictive, and arrhythmogenic right ventricular cardiomyopathy, all of which have a genetic background that has also been well described¹⁹. While the presentations of each of these cardiomyopathies are usually distinct, mixed-phenotype cases have also been described beyond those meeting standard clinical criteria for DCM^20–22. Examples of features that may be unusual or confound a DCM phenotype are provided, as well as those that may have features suggestive of specific genetic cause (Table 1). In addition, a DCM genetic evaluation should also consider disease beyond the cardiovascular system, such as symptoms of underlying neuromuscular or metabolic disease. In the setting of unusual or mixed DCM phenotypes, a broader-based panel testing approach may be considered to capture genetic complexity beyond the primary DCM gene list.

Table 1.

Variable DCM Clinical Presentations with Implications for Genetic Cause

Phenotypic feature	Usual DCM phenotype	Variable DCM Phenotype	Implications for Genetic Evaluation and Risk Assessment
Systolic dysfunction	LV or bi-ventricular dysfunction.	More prominent RV enlargement or systolic dysfunction than is usually observed with DCM.	*Consider ARVC with predominant LV involvement; this phenotype may be observed in DSP-related DCM.
Conduction system disease	Heart block not a usual early feature except with specific genes.	With advanced disease conduction system abnormalities not uncommon with DCM	LMNA, DES well established to cause prominent conduction system disease (1st, 2nd, 3rd degree heart block) before or with early systolic dysfunction
Supraventricular arrhythmias, including Atrial Flutter, Atrial Fibrillation	Atrial fibrillation is not uncommon with progressive DCM.	Prominent early arrhythmia	Especially common with LMNA cardiomyopathy
Ventricular Arrhythmias	Ventricular arrhythmias common with progressive, advanced stage DCM.	Prominent ventricular arrhythmias with minimal LVE or systolic dysfunction; less common as the presenting feature.	Characteristic of FLNC; also common with late phase LMNA, DES; common with ARVC with predominant LV features.
Skeletal myopathy, muscular dystrophy	Unusual to present primarily to cardiology; more commonly presents to neuromuscular clinics.	Subtle features may be identifiable with focused medical history and exam.	Consider LMNA, DES, EMD; and DMD in female (heterozygous) carriers; rarely TTN, SGCD.
DCM with LV hypertrophy	Non-hypertrophied, even a thinned LV myocardium	LVH of 13-15 mm; rarely LVH>15 mm	Implies sarcomeric genetic etiology, especially MYH7.
Non-or minimally-dilated, hypocontractile LV, but with LVH	Never a usual DCM phenotype	Consistent with a “burned out” or a hypocontractile LV that earlier may have met criteria for HCM.	Almost always a gene encoding a sarcomeric protein involved in HCM (MYH7, TNNT2, TPM1, TNNC1, TNNI3)
Selected Syndromic Presentations that may present to cardiologists
DCM with hearing loss	Rare DCM presentation	Rare DCM presentation	EYA4
Early DCM with mild LVH suggestive of restrictive cardiomyopathy	Uncommon DCM presentation.	Uncommon DCM presentation.	Rule out iron overload, including hemochromatosis, an autosomal recessive disease due to variants in HFE.
			Rule out amyloidosis, including TTR rare variant cause, which can be familial. The common variant in TTR (V122I) is carried in 3-4% of African Americans and is associated with amyloid risk.

ARVC, arrhythmogenic right ventricular cardiomyopathy; HCM, hypertrophic cardiomyopathy; LV, left ventricle; LVE, left ventricular enlargement; LVEF, left ventricular ejection fraction; LVH, left ventricular hypertrophy.

^*While almost all DCM genetic testing panels have the genes implicated for ARVC so panel selection would not be affected, this insight may alert the clinician to consider other features of the phenotype that may be relevant for diagnosis or care (e.g., use of CMR to assess ARVC task force diagnostic criteria; greater emphasis on an arrhythmia evaluation).

Genetic Testing and Clinical Implications

DCM Clinical Genetics.

Clinical genetic testing is the next key step and is now indicated with a DCM diagnosis, whether familial or non-familial.^13,14 Within a Mendelian framework, DCM in most cases is inherited in an autosomal dominant pattern¹⁹. Genetic heterogeneity, reduced penetrance, and variable expression are commonly observed. Genes causing a primary DCM phenotype come from a diverse gene ontology and demonstrate substantial locus and allelic heterogeneity². Considering this complex architecture is important to understanding the variability frequently observed in DCM pedigrees.

Genetic Testing and Gene Selection.

Identifying which genes are most appropriate to include in a genetic testing panel is a first critical step in facilitating genetic testing. Clinical genetic testing laboratories offer curated panels including anywhere from 50 to over 100 genes²³. While each of these genes have some degree of evidence suggesting a relationship with a cardiomyopathy phenotype, only about a dozen of these genes account for a majority of monogenic DCM, with only a portion of these having robust evidence supporting a monogenic relationship (Figure 2)^9,24. A recent case-control study demonstrated variants in TTN, DSP, MYH7, LMNA, BAG3, TNNT2, TNNC1, PLN, ACTC1, NEXN, TPM1, and VCL are significantly enriched in DCM cases⁹. Further, loss of function variants in TTN ²⁵, SCN5A²⁶, FLNC²⁷, BAG3²⁸, DSP²⁵, PLN^26,29, VCL³⁰, and LMNA³¹ have been specifically shown to have a DCM-causing effect. In addition, RBM20 is emerging as a key DCM gene, with pathogenic variation particularly well described in the hot spot region located in exon 9, affecting amino acids 634, 636, 637, and 638³². The Clinical Genome Resource (ClinGen) Cardiovascular Working Group is completing expert review and curation of DCM-associated genes in the literature³³ and these curations are available via the ClinGen website (https://search.clinicalgenome.org/kb/gene-validity).

Figure 2. Genes with evidence supporting an association with DCM.

Genes in which variants have been published in humans with DCM arranged by gene ontology. Genes with the most substantial evidence in their disease association with DCM are noted in bold text. Those with loss of function as an established DCM-causing mechanism are noted with an asterisk (*).

Variant Interpretation.

Rigorous and reliable variant interpretation, a cornerstone of genetic testing, is the next critical aspect. Despite recent intensive efforts, the interpretation of genetic variants continues to present challenges for the clinician. A summary of the clinical implications of variant classification categories defined by the American College of Medical Genetics is shown (Figure 3). Variants classified as “pathogenic” or “likely pathogenic” (P, LP) are considered clinically actionable and can be used for decision-making and risk prediction for relatives³⁴. However, actionable variants are only identified in a minority of DCM cases with current clinical genetic testing when applying published variant adjudication standards^7,34,35. The current estimates of identified actionable P or LP variants ranges from 15-40% of DCM cases, although this testing sensitivity is likely an overestimate, as observed in recently published preliminary dataset from the DCM Precision Medicine Study where actionable P/LP variants were detected in less than 20% of DCM probands. Up to date estimates of the sensitivity of genetic testing require a family-based approach with rigorous clinical and variant adjudication methods of a large, diverse patient cohort, such as the data that will be generated in the full Precision Medicine Study dataset. Nonetheless, at present, clinical genetic testing leaves many families with uninformative results, including variants of uncertain significance (VUS) or no variants identified⁷.

Figure 3. Definitions and clinical implications of DCM genetic results.

a) First-degree relatives (FDRs) with negative cascade genetic testing (GT) only discharged from follow up in the setting of normal clinical screening and no other indications for cardiovascular follow up. b) If VUS(s) are observed along with a P/LP variant are suspected to further modify the disease, clinical judgment should be carefully exercised when determining if family members should be discharged from screening. Expert CV genetics centers are best to manage the risk assessment and make care recommendations in the setting of unsolved complexity in these cases. c) Cardiologists, genetic counselors, and/or geneticists with expertise in cardiovascular genetics. d) In trying to resolve VUSs, expert evaluation of the pedigree to identify other informative affected family members where testing could inform if a variant is segregating with disease in the family may aid in classification resolution. (B, benign, LB, likely benign).

Variants of Uncertain Significance.

A VUS classification is assigned when available evidence is insufficient to assign a P, LP, or likely benign (LB) or benign (B) classification³⁴. Variants classified as VUSs are very commonly reported in DCM patients. Further, patients may have multiple VUSs reported, or a combination of LP or P variants along with one or more VUSs. Preliminary observations from The DCM Precision Medicine Study⁴ found at least one VUS in 46% of DCM probands, and 21% of these probands had multiple variants, including multiple VUSs or a combination of VUS(s) with a P or LP variant⁷.

Those ordering genetic testing for DCM should be informed and comfortable with VUS results, as VUSs can have a spectrum of biological relevance with ranging probability of clinical effect – from little to no predicted impact to disease modifying or high likelihood to be disease causing^36,37. Most cardiovascular genetics centers have experience managing the genetic uncertainty raised by VUSs³⁸. Current widely accepted US guidelines recommend that VUSs not be used for clinical decision-making. Thus, all possible efforts to resolve the classification of a VUS into a benign or pathogenic category should be made, such as tracking the segregation of a VUS(s) in an extended pedigree and correlate its presence or absence with DCM-related phenotypes ^34,39.

Clinical Management of at-risk family members

Guidelines recommend that at-risk family members who have not yet shown a DCM phenotype be clinically evaluated at regular intervals based on their age (Table 2), including an electrocardiogram and an echocardiogram, or cardiac magnetic resonance (CMR) imaging when an echocardiogram is not sufficient. Other testing and/or clinical evaluation is also indicated for concern of disease beyond the cardiovascular system.¹³ The use of CMR imaging is emerging as a critical clinical tool for detailed characterization of DCM, including more accurate quantitative measures, as well as unique approach to characterize the myocardium. Similarly, advanced echocardiography techniques, such as speckle-tracking (STE), also provides more quantitative measures of myocardial function. The use of CMR and/or STE imaging techniques has particular promise for identifying DCM at early, pre-clinical stages in at-risk family members.

Table 2. Suggested screening intervals for relatives at-risk for DCM who have no clinical phenotype.

This table is adapted for DCM screening from the HFSA practice guideline for the management of genetic cardiomyopathies¹³. Relatives at-risk for DCM include those who are unaffected but carry a genetic abnormality, or those with a first degree relative (parent, sibling, or child) with DCM that is genetically unsolved. Screening intervals can be informed in part by the age of onset and characteristics of the disease manifest in the family’s proband, as well as the expected phenotype based on the genetic information, when known.

Age	0-5 Years	6-12 Years	13-19 Years	20-50 Years	Over 50 years
Suggested screening interval	Annual	Every 1-2 years	Every 1-3 years	Every 2-3 years	Every 5 years

Family Counseling and Communication

Genetic counseling, the process of helping patients understand and adapt to the medical, psychological, and familial implications of genetic disease⁴⁰, is another key component of the genetic evaluation of DCM.^13,14 Accordingly, genetic counselors with expertise in cardiovascular disease can explain complex topics such as genetic risk, discuss implications of test results, provide psychosocial support as well as facilitate the family communication process, and are a recommended part of the care team.^13,14,38,41 However, while genetic counselors have the training and expertise to enable family communication, improved or novel approaches are still needed to assist with various communication needs and factors of all families affecting the successful sharing of DCM risk information.

The traditional model of genetic counseling and care begins with the proband presenting for evaluation. The proband is then charged with the task of communicating family risk with their relatives in a way that they understand the information and that motivates family members to seek follow up care, which can often be challenging⁴². Many factors create barriers to this critical step to family communication in genetic disease^43–45. Provider assistance to support the proband in their communication of genetic risk to their family is an essential part of family-based care. This reality also indicates the need for novel methods of information dissemination to family units.

EMERGING CONCEPTS IN DCM GENETICS

The genetics of DCM-related phenotypes

Recent publications have highlighted the genetics of DCM-related phenotypes, a term used to refer to a DCM phenotype beyond idiopathic where other usually clinically detectable causes, such as coronary artery disease causing ischemic cardiomyopathy, or cardiotoxicity, have been excluded. Many authorities have also included pregnancy-associated (PACM) and peripartum cardiomyopathy (PPCM) as different from idiopathic DCM. However, over time PACM and PPCM have been shown to have a similar genetic background to that of idiopathic DCM.^8,9,46 Finding specific molecular mechanisms of the pre- and post-partum condition that link to specific genetic mechanisms in humans to cause PPCM has remained elusive.⁴⁷

DCM primarily attributed to cardiotoxicity, including alcohol- and chemotherapy-related cardiomyopathy, have also been shown to have a genetic background, specifically with enrichment of truncating variants in TTN^10,11. These DCM-related phenotypes, previously phenotypes of uncertain genetic significance, are now known to have a genetic background, although whether all of DCM in these circumstances results from a rare variant genetic background remains to be fully established. This is important not only to the interpretation and management of disease in these patients but also to the risk stratification of their family. Further, this leads to a consideration of the many mechanisms in which one may meet a DCM disease threshold (Figure 4), with a combination of gene or gene(s) with or without environmental impact(s) in order to express the phenotype. As this interaction of disease-contributing factors continues to unfold, the utility and public health reach of genetic information to prevent DCM will also expand.

Figure 4. Simplified models of gene-environment interactions causing DCM.

Each colored bar represents genetic and environmental factors in individuals with DCM. The height of each bar represents its phenotypic impact to cause DCM. Blue bars represent the genetic contribution of one or more variants in DCM genes to the phenotype. Orange bars represent the environmental contributions to the phenotype. The upper horizontal line shows a threshold for clinically detectable DCM, and when blue, orange, or a combination of both, cross the line, DCM, can be clinically detected. Box A shows classic autosomal dominant DCM, occurring due to a single, highly penetrant, pathogenic variant, whereas Box B shows an individual who developed DCM only from environmental cause. In Box C, neither individual C1 nor individual C2 develop DCM from chemotherapy (CTX) or a moderate impact genetic variant (a), respectively. However, individual C3 develops DCM from two moderate impact variants (a, b), and individual C4 develops DCM when CTX is combined with a moderate impact variant (a). Box D represents cases of near-DCM, not reaching the DCM threshold from genetic, environmental, or gene and environment interaction (D1, D2, D3, respectively). With an added genetic predisposition due to common variants (yellow box), the same single variant, environmental impact, or both (D4, D5, D6, respectively) now reach the DCM threshold. Environmental impacts other than CTX may also facilitate DCM, e.g., excess alcohol ingestion, hypertension.

Uncertainty and genetic complexity

Despite recent advances in knowledge, many DCM patients, even those in families with multiple affected members, remain without a confirmed (P or LP) genetic diagnosis with currently available testing following established best-practice ACMG guidelines³⁴ or ACMG guidelines adapted for DCM.⁷ This high proportion of uninformative results may be explained in many ways: lack of detection with current technology; private family variants lacking case and/or experimental data required to meet strict variant classification standards; or complex non-Mendelian and non-traditional inheritance patterns that have yet to be clearly defined.

The Mendelian paradigm has been the primary model considered for DCM. However, to date, this framework has only been useful to identify a genetic explanation in a minority of families. While gene discovery may explain some unsolved pedigrees, it is unlikely to define the etiology for all unsolved DCM. Multiple variant etiologies, including oligogenic and polygenic models, need to be considered. Exemplary pedigrees in the literature have demonstrated the complex characteristics of such models^2,48,49. Further, as the genetics of DCM-related conditions beyond idiopathic disease continues to unfold, the clinical implications to family members of individuals with these DCM-related phenotypes will also introduce challenges to providers involved in the management of these patients and families in clinical practice.

Clinical recommendations remain conservative in the setting of uncertainty, advising that at-risk relatives, including parents, siblings, and children of individuals with DCM, be treated as if they are at increased risk (Figure 1). This applies not only to pedigrees where the proband has inconclusive or negative results, but also to probands who have P or LP variants but their pedigree does not appear to be fully explained by the single identified variant. Until the genetic background of the entire family, not only the proband, is solved and predictive testing can conclusively identify and definitively rule-out DCM risk, at-risk relatives should obtain regular clinical surveillance at their age-defined intervals, as described in current management guidance^13,14.

CONCLUSION

DCM is a cardiovascular disease of genetic etiology with a substantial opportunity for mitigation and prevention. A clinical genetic evaluation, with family history, phenotype clarification, genetic testing and counseling of the proband and at-risk family members, is foundational to the process. The understanding of the genetics of DCM continues to rapidly expand, including identification of new gene-disease relationships, approaches to variant adjudication and interpretation, and consideration of new disease models beyond traditional frameworks. Ongoing research to clarify and confirm the genetic architecture of DCM is key to further progress in the field.