Reevaluating the Mutation Classification in Genetic Studies of Bradycardia Using ACMG/AMP Variant Classification Framework

Purpose Next-generation sequencing (NGS) has become more accessible, leading to an increasing number of genetic studies of familial bradycardia being reported. However, most of the variants lack full evaluation. The relationship between genetic factors and bradycardia should be summarized and reevaluated. Methods We summarized genetic studies published in the PubMed database from 2008/1/1 to 2019/9/1 and used the ACMG/AMP classification framework to analyze related sequence variants. Results We identified 88 articles, 99 sequence variants, and 34 genes after searching the PubMed database and classified ABCC9, ACTN2, CACNA1C, DES, HCN4, KCNQ1, KCNH2, LMNA, MECP2, LAMP2, NPPA, SCN5A, and TRPM4 as high-priority genes causing familial bradycardia. Most mutated genes have been reported as having multiple clinical manifestations. Conclusions For patients with familial CCD, 13 high-priority genes are recommended for evaluation. For genetic studies, variants should be carefully evaluated using the ACMG/AMP variant classification framework before publication.


Introduction
One of the inherited bradycardias that is currently being reported is inherited progressive cardiac conduction disease (IPCCD). Progressive cardiac conduction disease (PCCD) is an unidentified, heterogeneous, life-threatening disease that manifests as progressing fibrosis of the cardiac conduction system [1]. It is characterized by a decreased conduction rate, prolonged PR interval, and widened QRS wave, and it ultimately leads to complete atrioventricular block, syncope, and even sudden cardiac death [1]. Initially, patients present with only a widened QRS wave without a bundle branch block, and later, they develop complete atrioventricular block. Abnormalities in the conduction system may be related to changes in cardiac structure and function [2]. It is currently believed that the etiology of PCCD may be related to genetic factors, valvular disease, cardiomyopathy, and autoimmune disease [3]. PCCD caused by genetic factors was originally called progressive familial heart block (PFHB) [3], and some studies directly used PCCD or IPCCD to refer to progressive conduction system diseases related to genetic factors. It is believed that PCCD is caused by the SCN5A mutation [4], and it may also be correlated with TRPM4 [5], DSP [6], and others. Genetic studies about other kinds of familial bradycardia have been published over the past decade, such as sick sinus syndrome and heart block. However, those studies have still not been summarized, and the clinical significance of the related variants is still unknown.
In 1977, Sanger et al. developed Sanger's "chaintermination" or dideoxy technique for nucleic acid sequence testing [7]. The improvement of Sanger sequencing makes DNA sequence testing for complex species available [8]. In the course of the development of next-generation sequencing (NGS), genetic testing becomes quicker, cheaper, and easier [9]. For patients who suffer from inherited cardiac disease, NGS has become a potential choice for the diagnosis,

PS2
De novo (confirmed maternity and paternity) in a patient with no family history and diseases.

PS3
Functional studies supported the effect of related pathogenic variants.

PS4
Variants' prevalence significantly increased in affected individuals than controls.

PM1
Mutation happened in hot spot and known function domain.

PM2
Absent (or extremely low) in large population studies.

PM3
With recessive disease, detected in trans with pathogenic variants.

PM4
Variants (in-frame deletions/insertions in a nonrepeat region or stop-loss variants) lead to changes in protein length.

PM5
Different missense changes at known pathogenic amino acid residue.

PM6
De novo (without confirmation of maternity and paternity).

PP1
Variants known to be the causes affected multiple family members.

PP2
Missense variants in a gene that have a low rate of benign missense variation are common mechanism of disease.

PP3
Multiple lines of computational evidence support a deleterious effect on the gene or gene products.

PP4
Phenotype specific for disease with single genetic etiology.

PP5
Reputable source reports variants as pathogenic.

Evidence of benign
Stand-alone

BA1
Allele frequency is >0.5% base on population database.

BS1
Allele frequency is greater than expected for disorder.

BS2
Recessive heredity being observed in healthy adult.

BS3
Functional studies show no pathogenic effect.

BP1
Missense variant in gene where only loss of function is pathogenic.

BP2
Observed in genes with overlapping function without increased disease severity or observed in cis with a pathogenic variant.

BP3
Variants (in-frame deletions/insertions in a nonrepeat region or stop-loss variants) lead to changes in a repetitive region without known function.

BP4
Multiple lines of computational evidence suggest no impact on gene or gene product.

BP5
Variant found in a case with alternate molecular basis for disease.

BP6
Report as benign.

BP7
Splicing variant predict an algorithm which predict no impact to the splice consensus sequence.
prevention, and treatment of certain diseases [9]. The relationships between inherited ion channel disease, such as long QT syndrome (LQTs) [10] and Brugada syndrome (BrS) [11], inherited cardiomyopathy, such as dilated cardiomyopathy (DCM) [12], hypertrophic cardiomyopathy (HCM) [13], and arrhythmogenic right ventricular cardiomyopathy/dysplasia (ARVC/D) [14], and variant sequencing have been well studied. However, the role of genetic sequence variants in bradycardia is still under debate. Evaluation of sequence variants is a complex process. The integrity of both the genome and the protein being translated should be studied. In 2015, the American College of Medical Genetics and Genomics (ACMG) and the Association for Molecular Pathology (AMP) recommended an interpretative category of sequence variants and an algorithm for interpretation [15]. The ACMG/AMP classification framework is prominent in the evaluation of the Mendelian system. By evaluating the allele frequency, segregation, de novo, and protein expression, functional studies and other factors, sequencing variants can be scored as pathogenic or benign. The two parallel scoring systems divided mutations into 7 categories (Table 1). Sequence variants were then classified into a fivetier system: "pathogenic," "likely pathogenic," "uncertain significant," "likely benign," and "benign" ( Table 2). By using this method, evaluated genomic variants can be quantified. With the development of evaluation methods for sequence variants, a growing number of databases have been developed. InterVar [16] is a tool implementing ACMG/AMP criteria that can automatically analyze sequence variants. LitVar [17] links genomic variants in PubMed and PMC, making functional studies achievable. With those databases, sequence variants can be evaluated properly.
At present, most of the related mutant genes reported in the literature are not analyzed according to the ACMG guidelines. In this article, we summarized and reevaluated pedigree studies of bradycardia published in PubMed from 2008/1/1 to 2019/9/1 using the ACMG/AMP variant classification framework.

Database Search.
We searched the PubMed database by using the term "heart block" or "sick sinus syndrome" associated with "pedigree"

Study
Selection. The aim of this study was to evaluate genetic studies of bradycardia, in addition to the inclusion criteria and exclusion criteria, as follows: Inclusion criterion:  (Tables 1 and 2), we evaluated related sequence variants and proposed a clinical judgement.

Results and Discussion
We summarized genetic studies published in the PubMed database over 11 years (Figure 1). A total 1015 articles were enrolled after searching the database. 927 articles were excluded. Finally, 88 articles fit the profile; 99 variants and 34 genes were studied in the current article. Information in InterVar was gathered to evaluate all the sequence variants, and the relevant evidence for pathogenic and benign criteria was summarized (Table 3). For mutation cannot be defined in InterVar, we used gnomAD and Clin-Var to analyze frameshift mutations (Table 4) and large fragment deletions (Table 5). We also gathered information about splicing mutations (Table 6).
For the majority of related genes, the clinical manifestations were not unique. These mutations may lead to bradycardia, arrhythmia, myopathy, and nerve system disease. LMNA mutations may present as AVB and arrhythmia; DES, GJA5, TTN, LAMP2, and MECP2 mutations may present as AVB and myopathy; GNB5 mutation may present as CCD and nerve system disease; HCN4, KCNQ1, PRKAG2, and SCN5A mutations may present as CCD, myopathy, and arrhythmia.
Genetic diagnosis has become an inalienable part of the diagnosis, treatment, and prevention of SCD. Cardiac ion channel disease, closely related to sudden cardiac death (SCD), has been discussed for decades. In contrast, the relationship between bradycardia and genetic factors is still unclear. Syncope and SCD caused by bradycardia are lifethreatening diseases. If the relationship between genetic factors and bradycardia is eliminated, SCD could be prevented.
Pedigrees of bradycardia families have been reported for decades. However, those studies are lacking. Some of the studies do not include full information about related sequence variants, and some of the studies do not list the whole family tree. In addition, the methods used to evaluate sequence variants are complex, and different centers have their own experience. It is still doubtful whether those variants are pathogenic. Therefore, ACMG/AMP promotes a guideline for thorough evaluation. By analyzing the allele frequency, segregation, de novo, protein expression, functional studies, and other factors, sequencing variants can be scored into a five-tier system: pathogenic, likely pathogenic, uncertain significant, likely benign, and benign. As accurate as the guideline may be, pathogenicity has been defined as being greater than 90% of pathogenicity [15]. According to the precise classification of pathogenicity, pedigrees of familial bradycardia can be reevaluated. InterVar [16] is a tool implementing ACMG/AMP criteria that can automatically analyze sequence variants. In this article, we used InterVar to summarize 13 high-priority genes, as follows: ABCC9, ACTN2, CACNA1C, DES, HCN4, KCNQ1, KCNH2, LMNA, MECP2, LAMP2, NPPA, SCN5A, and TRPM4 (Table 3). High-throughput sequencing (next-generation sequencing) is quite expensive. In contrast, the gene panel is cheaper and easier to analyze. We recommend that patients with a family history of bradycardia have their clinical manifestations gathered and that related pathogenic genes be highly regarded.
For future reference, multicenter studies on the epidemiology of familial bradycardia should be organized. In Table 2: Sequence variant classification.       addition, detailed information about sequence variants should be addressed in related articles and should be evaluated under the ACMG/AMP classification framework. The relationship between bradycardia and genomic variants remains unknown, and epigenetics and modifier genes should be used to investigate the relationship between genes and diseases.

Limitation
We summarized sequence variants published in only the PubMed database. There should be more pathogenic genes studied related to bradycardia.

Conclusion and Future Direction
Only 13 pathogenic genes (99 sequence variants and 34 genes being studied) were identified after using the ACMG/AMP variant classification framework to reevaluate. For future reference, pedigree studies should be fully evaluated before being published. For patients with familial CCD, 13 high-priority genes are recommended for evaluation. Compared to whole genome sequencing, this will increase the clinical utility of genetic testing.

Data Availability
There are no restrictions on data access of this paper. All works have been provided in this paper

Conflicts of Interest
The authors declare that they have no conflicts of interest.

Authors' Contributions
Liting Cheng and Xiaoyan Li contribute the same to this article.