Article Text

Download PDFPDF

Short report
FLNC truncations cause arrhythmogenic right ventricular cardiomyopathy
Free
  1. Francesca Brun1,
  2. Marta Gigli1,
  3. Sharon L Graw2,
  4. Daniel P Judge3,
  5. Marco Merlo1,
  6. Brittney Murray4,
  7. Hugh Calkins4,
  8. Gianfranco Sinagra1,
  9. Matthew RG Taylor2,5,
  10. Luisa Mestroni2,
  11. Cynthia A James4
  1. 1 Cardiovascular Department and Department of Medical Surgical and Health Sciences, University of Trieste, Trieste, Friuli-Venezia Giulia, Italy
  2. 2 Cardiovascular Institute, University of Colorado Anschutz Medical Campus, Aurora, Colorado, USA
  3. 3 Division of Cardiology, Medical University of South Carolina, Charleston, South Carolina, USA
  4. 4 Division of Cardiology, Johns Hopkins, Baltimore, Maryland, USA
  5. 5 Adult Medical Genetics Program, University of Colorado Anschutz Medical Campus, Aurora, Colorado, USA
  1. Correspondence to Professor Luisa Mestroni, Cardiovascular Institute, University of Colorado Anschutz Medical Campus, Aurora 80045-2581, Colorado, USA; luisa.mestroni{at}CUAnschutz.edu

Abstract

Background Arrhythmogenic right ventricular cardiomyopathy (ARVC) is a heart muscle disease that affects predominantly the right ventricle and is part of the spectrum of arrythmogenic cardiomyopathies (ACMs). ARVC is a genetic condition; however, a pathogenic gene variant is found in only half of patients.

Objective Filamin C gene truncations (FLNCtv) have recently been identified in dilated cardiomyopathy with ventricular arrhythmia and sudden cardiac death, a phenotype partially overlapping with ARVC and part of the ACM spectrum. We hypothesised that FLNCtv could be a novel gene associated with ARVC.

Methods One hundred fifty-six patients meeting 2010 ARVC Task Force Criteria and lacking variants in known ARVC genes were evaluated for FLNC variants. Available family members were tested for cosegregation.

Results We identified two unique FLNCtv variants in two families (c.6565 G>T, p.Glu2189Ter and c.8107delG, p.Asp2703ThrfsTer69), with phenotypes of dominant RV disease fulfilling ‘definite’ diagnosis of ARVC according to the 2010 Task Force Criteria. Variants in other cardiomyopathy genes were excluded in both kindreds, and segregation analysis revealed that p.Asp2703ThrfsTer69 was a de novo variant. In both families, the disease phenotype was characterised by prominent ventricular arrhythmias and sudden cardiac arrest.

Conclusion The identification of FLNCtv as a novel cause of ARVC in two unrelated families expands the spectrum of ARVC non-desmosome disease genes for this disorder. Our findings should prompt inclusion of FLNC genetic testing in ARVC to improve diagnostic yield and testing of at-risk relatives in ARVC.

  • arrhythmogenic cardiomyopathy
  • filamin C
  • sudden cardiac death
  • arrhythmias
  • dilated cardiomyopathy
  • arrhythmogenic right ventricular cardiomyopathy

Statistics from Altmetric.com

Introduction

Arrhythmogenic right ventricular cardiomyopathy (ARVC) is a rare heart muscle disorder that predominantly affects the right ventricle (RV).1 ARVC is typically characterised by progressive fibrofatty replacement of the RV myocardium that predisposes to life-threatening ventricular arrhythmias and sudden cardiac death, and to slowly progressive RV dilatation and dysfunction.2 According to the recent 2019 Heart Rhythm Society (HRS) Expert Consensus Statement on Evaluation, Risk Stratification, and Management of Arrhythmogenic Cardiomyopathy (ACM),3 ARVC is considered part of the spectrum of ACM phenotypes, where there may be a predominant RV (ARVC), left ventricular (ALVC, overlapping with dilated cardiomyopathy (DCM)) or biventricular involvement. ARVC is a genetic disease, largely autosomal dominant, caused by variants in genes encoding the cardiac desmosome (PKP2, DSP, JUP, DSG2 and DSC2) and rarer non-desmosomal genes (TMEM43, LMNA, TTN, PLN, SCN5A, DES, TGFB3, CDH2, CTNNA3).4 The identification of ARVC gene variants in patients with DCM indicates that there is also genetic overlap between the two diseases.5 6

In ARVC, a causal gene variant is found in approximately 50% of cases, leaving a significant part of the patient population with an unknown cause.7 We and other investigators have reported filamin C gene truncation variants (FLNCtv) in patients with DCM and a severe arrhythmogenic phenotype.8–10 Variants in FLNC have also been reported in myofibrillar myopathy, hypertrophic and restrictive cardiomyopathies, all characterised by prominent ventricular arrhythmias,11–13 suggesting that filaminopathies can generate a spectrum of different cardiac disorders with high risk of arrhythmias and sudden cardiac death. Intriguingly, the FLNC protein has known actin crosslinking, Z-line and integrin binding functions and is a critical protein for mechanotransduction in cardiac myocytes,14 a major pathogenic mechanism in ARVC. In addition, we have found that the FLNC protein localises to the intercalated disk and is associated with reduced desmoplakin, making FLNC an excellent candidate for causing ARVC.8 10

Here, we report the identification of FLNCtv as a novel cause of ARVC in two unrelated families, expanding the spectrum of disease genes for this disorder.

Methods

Study subjects

We analysed patients with ARVC with available genetic information enrolled in the Johns Hopkins ARVC Registry and in the Familial Cardiomyopathy Registry, a multicentre registry including the Cardiovascular Department, University of Trieste (Italy) and the University of Colorado Cardiovascular Institute (CUCVI) (USA). The diagnosis of ARVC was performed according to the 2010 Task Force Criteria.1 Patients were enrolled after informed consent according to the respective Institutional Review/Ethics Committees. Family history was extensively investigated and ≥3 generation pedigrees were constructed.

Genetic testing

Next-generation sequencing was performed on Illumina platforms in the CUCVI research laboratory, as previously reported.8 The Johns Hopkins cohort included 107 patients in whom pathogenic/likely pathogenic variants in PKP2, DSG2, DSP, DSC2, JUP, TMEM43, SCN5A and PLN had previously been excluded. Twenty-one patients had whole-exome sequencing (Illumina HiSeq2000 platform) and 86 were sequenced using a custom Agilent 500 kb panel of 40 candidate genes. In the Familial Cardiomyopathy Registry, the patients’ samples were analysed using the TruSight One-Sequence panel (Illumina, Redwood City, California, USA), which queries 4813 genes associated with clinical phenotypes, as previously described.15 Variants were validated with bidirectional Sanger sequencing, and classified as ‘pathogenic’ (P) or ‘likely pathogenic’ (LP) according to the American College of Medial Genetics criteria.16 Variants of uncertain significance were excluded from the analysis. Whenever possible, additional affected family members were used for segregation analysis. For the FLNC gene, we used the reference transcript ID NM_001458.

Results

FLNCtv identification

We studied 156 ‘gene elusive’ patients with ARVC with no known ARVC disease variants (107 of the Johns Hopkins Hospital cohort and 49 of the Familial Cardiomyopathy Registry). We identified two unique FLNCtv variants in two families with phenotypes of dominant RV disease fulfilling ‘definite’ 2010 Task Force Criteria for ARVC17 (figure 1A). Variants in other cardiomyopathy genes were excluded in both kindreds. The first variant (family TSRVD029) predicted a loss of function (c.6565 G>T, p.Glu2189Ter) in exon 40. The second variant (family JHRVD001) caused a frameshift in exon 48 (FLNC c.8107delG, p.Asp2703ThrfsTer69). Both FLNCtv variants were absent in gnomAD (gnomad.broadinstitute.org)18 and ClinVar (www.ncbi.nlm.nih.gov/clinvar/)19 databases (accessed 6 Jan 2019) and classified as P, according to the American College of Medical Genetics criteria16 (table 1). In the Genome Aggregation Database (https://gnomad.broadinstitute.org), FLNC has a pLI (probability of being loss-of-function intolerant) score of 1 and an o/e (observed/expected score) of 0.17 (upper 90% CI=25), indicative that the transcript is extremely intolerant of loss-of-function variants. Furthermore, there is a statistically significant difference in the prevalence of all truncating variants in FLNC and the prevalence in our ARVC population (p<0.00001—one-sample proportion test), where the ARVC percentage is 1.28% (95% CI 0.16 to 4.55).

Figure 1

FLNCtv and arrhythmogenic right ventricular cardiomyopathy (ARVC). (A) Family pedigrees. Squares represent men; circles, women; diagonal lines, deceased individuals; black shading, ARVC phenotype; grey shading, affected not fulfilling ARVC criteria; arrow, proband; plus signs (+), confirmed carriers of FLNCtv; negative signs (−), wild-type genotype. (B) Cardiac MRI of subject TSRVD029-IV:3, showing a dilated LV (end-diastolic volume 110 mL/m2) with normal contractility, while the RV was globally hypokinetic (ejection fraction 44%) and dilated RV (end-diastolic volume 136 mL/m2). (C) ECG of the proband of family JHRVD001 showing V1–V5 T-wave inversion, a typical feature found also in desmosome ARVC, and frequent premature ventricular contractions with left bundle branch block morphology.

Table 1

Clinical features of the FLNC variant carriers

ARVC phenotype

In family TSRVD029, the proband, a 66-year-old woman (figure 1A, III:3; table 1) presented with a 20-year history of syncope and palpitations. Her ECG showed a complete right bundle branch block with T-wave inversion V1 to V4. The echocardiogram demonstrated mild RV dilatation, a RV fractional area change of 18% and RV apical inferolateral wall akinesia. The left ventricle (LV) showed normal structure and function. Non-sustained ventricular tachycardia with left bundle branch block (LBBB) morphology and frequent (>500/h) premature ventricular contractions (PVCs) were detected with 24-hour Holter ECG, resulting in a ‘definite’ diagnosis based on the 2010 ARVC Task Force Criteria (table 1).1 The proband’s sister had refractory supraventricular arrhythmias. The proband’s nephew (figure 1A, IV:3; table 1), a 45-year-old asymptomatic athlete, was found to have frequent PVCs (735/24 h), non-sustained ventricular tachycardia with LBBB morphology with inferior axis. Evaluation is ongoing via implantable loop recorder. Cardiac magnetic resonance (figure 1B) showed a dilated RV (end-diastolic volume 137 mL/m2) with impaired systolic function and diffuse hypokinesis (RV ejection fraction 44%), a dilated LV (end-diastolic volume 110 mL/m2) with LV ejection fraction of 52%. He had two minor Task Force Criteria (family history and arrhythmia) fulfilling a diagnosis of ‘possible’ ARVC since the RV dysfunction was global, not regional. Finally, a third-degree relative (III:1) died of progressive heart failure while awaiting heart transplant.

In family JHRVD001, the proband, a 37-year-old woman (figure 1A, II:1; table 1), presented with sudden cardiac arrest while playing tennis. The patient had T-wave inversion V1 to V5 and LBBB superior axis PVCs at admission (figure 1C). Cardiac magnetic resonance revealed major RV structural abnormalities with regional dyskinesia in the RV outflow tract, RV dysfunction (RVEF 32%) and normal LV function (LVEF 55%), fulfilling ‘definite’ diagnostic ARVC criteria (three major criteria—repolarisation, arrhythmia, structure; table 1). The patient received an implantable cardioverter defibrillator, and had recurrent sustained ventricular tachycardia and frequent PVCs (18%) during the follow-up. No additional family members exhibited symptoms and cascade genetic testing revealed the variant was de novo.

Discussion

FLNCtv have previously been associated with an arrhythmogenic phenotype in a variety of cardiomyopathies. Recently, the arrhythmogenic pattern of patients with DCM carrying pathogenic or likely pathogenic FLNCtv variants has emerged as a key issue in the management of those patients.10 Indeed, the current 2019 HRS guidelines include FLNC among genes that warrant early consideration for implantable cardioverter defibrillator for the prevention of sudden cardiac death, even in the presence of only mild left ventricular dysfunction (LV ejection fraction between 35% and 45%).3

Here, we report that FLNCtv can be associated with an ARVC phenotype, showing a predominant RV involvement and life-threatening ventricular arrhythmias. In population-based cohorts, FLNC is predicted to be extremely intolerant to protein-truncating variants. Although RNA and myocardial tissue were not available from the families here investigated, it should be noted that we previously found decreased FLNC protein expression in the explanted heart of one patient with a FLNCtv, supporting haploinsufficiency as the disease mechanism.9 Interestingly, we also found altered desmosomal protein localisation (desmoplakin, SAP97) in FLNCtv carriers8: these findings reflect the clinical continuum and the genetic overlap between ARVC and DCM, as recently suggested by Corrado et al 20 and defined by the 2019 HRS Consensus Statement on ACM.3 ARVC and DCM involving respectively RV and LV are two extremes of the same disease and demonstrate phenotypic overlap when the arrhythmic pattern is evident and predominant.21 Further confirming this principle is the increasing finding of pathogenic variants in desmosomal genes in DCM.15 22

Our findings highlight the complexity of ACM and the need for a better understanding of the pathogenic mechanisms leading to ARVC. In this regard, it should be noticed that advancements in sequencing technology have led to widespread availability of clinical genetic testing, and while emerging evidence supports the use of genetic testing in the clinical practice, current guidelines recommend the referral of patients with ACM to experienced centres,3 to improve risk stratification for clinically affected patients with cardiomyopathy and their families, especially with regard to the risk of sudden cardiac death and malignant ventricular arrhythmia. Finally, although FLNCtv are apparently rare (approximately 1%) in the ARVC population, their clinical impact should prompt FLNC gene testing both in the clinical and research settings. Furthermore, more investigations are needed to allow full understanding of the contributions of FLNC variants in cardiomyopathies, arrhythmias and sudden cardiac death.

Acknowledgments

The authors are grateful to the family members for their participation in these studies.

References

Footnotes

  • Contributors FB, MG, MM, BM: clinical characterisation of family members, data analysis and final approval of the manuscript submitted. SLG: design, analysis and interpretation of targeted sequencing data; revising critically the manuscript for important intellectual content; final approval of the manuscript submitted. MT, GS, DPJ, HC, LM, CJ: conception, design, analysis and interpretation of data; drafting of the manuscript and revising it critically for important intellectual content; final approval of the manuscript submitted.

  • Funding This study was supported by the NIH R01 HL69071, HL116906, NCATS/CTSA UL1 TR001082 and AHA17GRNT33670495 (LM); NIH 1K23HI067915 and R01HL109209 (MT); 2UM1HG006542 and NCATS UL1 TR001079 (HC). This work was also supported in part by the Trans-Atlantic Network of Excellence grants from the Foundation Leducq (14-CVD 03 and 16-CVD 02) and the EU FP7 SarcoSi IRSES. The Johns Hopkins ARVD/C Program is supported by the Leonie-Wild Foundation, the Francis P. Chiaramonte Private Foundation, the Leyla Erkan Family Fund for ARVD Research, the Dr. Satish, Rupal, and Robin Shah ARVD Fund at Johns Hopkins, the Bogle Foundation, the Healing Hearts Foundation, the Campanella family, the Patrick J. Harrison Family, the Peter French Memorial Foundation and the Wilmerding Endowments.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.