Spectrum and Prevalence of Mutations Involving BrS1-12-Susceptibility Genes in a Cohort of Unrelated Patients Referred for Brugada Syndrome Genetic Testing: Implications for Genetic Testing

Abstract

Objective

To provide the spectrum and prevalence of mutations in the 12 Brugada Syndrome (BrS)-susceptibility genes discovered to date, in a single large BrS cohort.

Background

BrS is a potentially lethal heritable arrhythmia syndrome diagnosed electrocardiographically by coved-type ST segment elevation in the right precordial leads (V1-V3; type-1 Brugada ECG pattern) and the presence of a personal/family history of cardiac events.

Methods

Using PCR, DHPLC, and DNA sequencing, comprehensive mutational analysis of BrS1-12-susceptibility genes was performed in 129 unrelated patients with possible/probable BrS [46 with clinically diagnosed BrS (ECG pattern plus personal/family history of a cardiac event) and 83 with type 1 ECG pattern only].

Results

Overall, 27 (21%) patients had a putative pathogenic mutation, absent in 1400 Caucasian reference alleles, including 21 patients with an SCN5A mutation, 2 CACNB2B, 1 KCNJ8, 1 KCND3, 1 SCN1Bb, and 1 HCN4. The overall mutation yield was 23% in type 1 ECG pattern only patients versus 17% in clinically diagnosed BrS patients, was significantly greater among young men < 20 years of age with clinically diagnosed BrS, and among patients who had a prolonged PQ interval.

Conclusions

We identified putative pathogenic mutations in ~20% of our BrS cohort, with BrS2-12 accounting for < 5%. Importantly, the yield was similar between patients with only a type 1 BrS ECG pattern and those with clinically established BrS. The yield approaches 40% for SCN5A-mediated BrS (BrS1) when the PQ interval exceeds 200ms. Calcium channel-mediated BrS is extremely unlikely in the absence of a short QT interval.

INTRODUCTION

Brugada Syndrome (BrS) is a rare heritable arrhythmia syndrome associated with an increased risk of sudden cardiac death (SCD) secondary to re-entrant polymorphic ventricular tachycardia and ventricular fibrillation(1). The diagnosis of BrS is based on the presence of coved-type ST segment elevation in the right precordial leads (V₁-V₃) on surface electrocardiogram (ECG), referred to as a type 1 Brugada ECG pattern, in the absence of structural heart disease or pharmacologic agents known to cause a Brugada-like ECG pattern (www.brugadadrugs.org). Additionally, to make a proper diagnosis, the diagnostic ECG pattern should be present together with either personal symptoms, family history of premature SCD, or at least 1 additional relative with a positive type 1 Brugada ECG pattern(2). Due to the transient and dynamic nature of the Brugada ECG pattern, administration of sodium channel blockers (e.g. ajmaline, flecainide, or procainamide) are used to unmask the type 1 ECG pattern(2). In general, BrS is understood as a disorder that affects young male adults with arrhythmogenic manifestation first occurring in the 4th decade of life with sudden death usually occurring during sleep(3,4). However, BrS may also manifest in the young and during infancy, and when familial, BrS is inherited as an autosomal dominant trait, however over half may be sporadic in nature(5).

Over the past 20 years, 12 BrS-susceptibility genes (BrS 1-12) have been identified. Loss-of-function mutations in the SCN5A-encoded α-subunit of the cardiac sodium channel represents the most common genetic substrate for BrS (annotated as type 1 BrS or BrS1), accounting for 15% to 30% of the disorder(5,6,7,). Besides sodium channel dysfunction, mutations involving the L-type calcium channel alpha1 (α1), beta2 (β2), and alpha2delta (α2δ) subunits encoded by CACNA1C(8), CACNB2B(8) and CACNA2D1(9), respectively, may cause 10-15% of BrS(9).

Over the last five years, 8 of the 12 BrS-susceptibility genes have been identified, where mutations result in either i) a reduction of the cardiac sodium channel (I_Na) current [GPD1L(10), SCN1B(11) (including the alternatively spliced exon 3A; SCN1Bb(12)), SCN3B(13) and MOG1(14)], ii) an increase in the transient outward potassium (Ito) current [KCNE3(15), and KCND3(16)], iii) an increase in I_KATP current (KCNJ8)(17), or iv) reduction in pacemaker (I_f) current (HCN4)(18).

Although the relative contribution of SCN5A to BrS has been well characterized by several investigating laboratories(19), the molecular genetic contribution of the BrS1-12 susceptibility genes has not been systematically analyzed in a single large cohort of unrelated BrS cases. In 2011, two expert consensus documents were published on the use of genetic testing in the clinical evaluation of cardiac channelopathies and cardiomyopathies(20,21). While the Heart Rhythm Society (HRS)/European Heart Rhythm Association (EHRA) Expert Consensus Statement(21) recommended genetic testing for patients with a clinical diagnosis of BrS, the Canadian Cardiovascular Society (CCS)/Canadian Heart Rhythm Society (CHRS) joint position paper(20) also advised genetic testing even in the setting of an isolated type 1 Brugada ECG pattern.

In the present study, we provide the spectrum and prevalence of BrS1-12 associated gene mutations discovered in a large BrS cohort. Furthermore, through the assessment of the genetic testing yield for 46 patients with clinically diagnosed BrS and for 83 unrelated patients with only a type 1 Brugada ECG pattern, we provide, for the first time, concrete data to further guide genetic testing recommendations for these two patient populations.

METHODS

Study Population

The study population consisted of 46 unrelated patients with clinically diagnosed BrS and 83 unrelated patients having either a spontaneous or drug induced type 1 Brugada ECG pattern as their sole finding, who were referred to either the Windland Smith Rice Sudden Death Genomics Laboratory at Mayo Clinic, Rochester, Minnesota, or to the Molecular Cardiology Laboratory, Fondazione IRCCS Policlinico San Matteo, Pavia Italy, for genetic testing. A clinical diagnosis of BrS was made using the strict criteria provided in the Consensus Conference Document(2). Briefly, a clinical diagnosis of BrS was assigned to an individual presenting with a diagnostic type 1 Brugada ECG pattern (coved-type ST segment elevation in the right precordial V1-V3 leads) present either spontaneously and/or after intravenous injection of a sodium channel blocking agent (ajmaline, flecainide or procainamide) and either a personal or family history of arrhythmic syncope, cardiac arrest, or sudden cardiac death. Patients with an acquired cause of a type 1 ECG pattern were excluded. Patients reporting palpitations, atypical chest pain, and/or a history of syncope with clinical characteristics strongly suggestive of vasovagal syncope were considered asymptomatic. This study was approved by both the Mayo Foundation Institutional Review Board and the Medical Ethical Committee of Fondazione IRCCS Policlinico San Matteo. Informed consent was obtained for all patients.

ECG Analysis

Twelve-lead ECGs were recorded at baseline at a paper speed of 25 mm/s. P-wave duration, PR-, QRS-, and QT- intervals were measured manually from basal ECGs and the QTc was calculated according to Bazett’s formula. The presence of a spontaneous type 1 Brugada ECG was evaluated both in basal ECGs and in 12-leads 24-hour Holter recordings.

Mutational Analysis

Following informed consent, a comprehensive open reading frame/splice site mutational analysis of all amino acid coding exons and intron borders of the 12 BrS-susceptibility genes (SCN5A, GPD1L, CACNA1C, CACNB2B, SCN1B (including the alternatively spliced exon 3A; SCN1Bb), SCN3B, KCNE3, KCNJ8, KCND3, CACNA2D1, MOG1, and HCN4) was performed using polymerase chain reaction (PCR), denaturing high performance liquid chromatography (DHPLC), and DNA sequencing as previously described(22). PCR primer sequences and PCR / DHPLC conditions are provided in Supplemental Tables 1-13.

In order to be considered a putative pathogenic mutation, the genetic variant had to be i) a non-synonymous variant and ii) absent in at least 700 ethnically-matched controls (≥ 1400 reference alleles plus all available online databases, including the 1000 Human Genome Project database (www.1000genomes.org)(23). Control genomic DNA was obtained from the European Collection of Cell Cultures (HPA Culture Collections, UK), the Human Genetic Cell Repository sponsored by the National Institute of General Medical Sciences and the Coriell Institute for Medical Research (Camden, New Jersey), and from the Blood Transfusional Centre in IRCCS Policlinico San Matteo of Pavia (Italy).

Mutations were annotated using the single letter nomenclature whereby F892I for example denotes a non-synonymous single nucleotide substitution producing a missense mutation whereby the wild-type amino acid phenylalanine (F) has been replaced by an isoleucine (I) at amino acid 892.

Statistical Analysis

Comparisons of groups identified on the basis of the clinical characteristics and genotype were performed in univariate analysis. Student’s t-test was used for continuous variables. Fisher test was used for categorical variables and odds ratios (OR) for unadjusted data and their 95% confidence intervals (95% CI) were calculated. Continuous variables are presented as mean ± standard deviation (SD). Two-sided P value <0.05 was considered statistically significant. Prism 3.02 was used for statistical analysis.

RESULTS

Study Population

Table 1 summarizes the clinical demographics of the 129 unrelated patients referred for laboratory-based BrS genetic testing. The majority were male (104, 81%), of Caucasian descent (122, 95%), with a mean age at diagnosis of 43 ± 14 years. Sixteen (12%) suffered cardiac events, including 6 (4.6%) with documented aborted cardiac arrest (ACA)/sudden cardiac death (SCD). A type 1 Brugada ECG pattern (Figure 1) was present spontaneously in 61 patients (47%), was induced only with a sodium channel blocker in 61 patients (47%) and appeared during fever in the remaining 7 patients. Among those with a spontaneous type 1 ECG pattern, the diagnostic ECG was not evident in 18% at the time of the diagnosis but only during follow-up visits (follow-up 24 ± 14 months). A strict clinical diagnosis of BrS was assigned to 46 patients (27 males, average age at diagnosis 43±15 years, 30 with a spontaneous type 1 ECG pattern, 14 with a drug induced type I ECG pattern, 2 with a spontaneous ECG pattern during fever; 16 symptomatic, 36 with a positive family history of cardiac events). Eighty-three asymptomatic patients (77 males, average age 42±14 years) with no family history were referred for genetic testing solely due to the patient’s idiopathic type 1 Brugada ECG pattern that occurred either spontaneously (31 cases) or following intravenous injection of a sodium channel blocker (47 cases) or during fever (5 cases).

Table 1.

Demographics of Unrelated Patients Referred for BrS Genetic Testing

	Overall	Patients with Clinically Diagnosed* BrS	Patients with Type 1 Brugada ECG Pattern only
Patient Demographics
No. of Probands	129	46	83
Age at Diagnosis (years)	43±14	43±15	42±14
Range (yrs)	8-81	10-81	8-74
Males (%)	104 (81)	27 (59)	77 (93)
Females (%)	25 (19)	18 (39)	7 (8)
Average QTc (ms)	409±27	410±27	409±27
Average PQ interval (ms)	172±29	177±28	170±30
Symptomatic Patients (%)	16 (12)	16 (35)	0 (0)
Family History of cardiac events / unexplained sudden death (%)	37 (29)	37 (80)	0 (0)
Type 1 ST-segment elevation at baseline (%)	61 (47)	30 (65)	31 (37)
Type 1 ST-segment elevation with sodium blockade (%)	68 (53)	16 (35)	52 (63)
Mutation Detection Yield
Total Yield (%)	27 (21)	8 (17)	19 (23)
Males (%)	20 (19)	8 (30)	11 (14)
Females (%)	7 (28)	0 (0)	7 (100)
SCN5A Positive (%)	21 (16)	6 (13)	15 (18)
PQ interval (ms)	191±31	199±32	188±31

^*Clinically Diagnosed BrS = patients with a Type 1 Brugada ECG pattern (coved-type ST segment elevation in the right precordial V1-V3 leads) present either spontaneously and/or after intravenous injection of a sodium channel blocking agent (ajmaline, flecainide or procainamide) with either a personal and/or family history of arrhythmic syncope, cardiac arrest, or sudden cardiac death.

Figure 1. Diagnostic Brugada Syndrome Electrocardiogram (ECG).

Representative ECG traces of a type I Brugada ECG pattern, not present in basal conditions (A), but appearing during Holter recording (B) in a 56-year old male.

Mutational Analysis

Putative Pathogenic Mutations

Overall, 27 putative pathogenic mutations (21 SCN5A, 2 CACNB2B, 1 KCNJ8, 1 KCND3, 1 SCN1Bb, and 1 HCN4) were identified in 27/129 (21%) unrelated BrS cases (20 males, 7 females) (Table 2). Each mutation was absent in not only ≥ 1400 ethnic-matched reference alleles but also all publicly available databases including the 1000 Human Genome Project (www.1000genomes.org)(23).

Table 2.

Summary of Brugada Syndrome-Associated Mutations

Case Number	Gene	Exon	Nucleotide Change	Mutation	Mutation Type	Location	Age (Years)	Sex	Symptomatic (Yes/No)	Family History (Yes/No)
Major BrS Genotype-SCN5A
1	SCN5A	2	80G>A	R27H	Missense	N-terminal	8	M	No	No
2	SCN5A	2	127C>T	R43X*	Nonsense	N-terminal	19	M	No	Yes
3	SCN5A	4	477 T>A	Y159X*	Nonsense	DI-S2	40	M	No	No
4	SCN5A	16	2466G>T	W822C*	Missense	DII-S4	24	M	Yes	No
5	SCN5A	16	2632C>T	R878C	Missense	DII-S5/S6	34	M	No	No
6	SCN5A	16	2674T>A	F892I*	Missense	DII-S5/S6	30	F	No	No
7	SCN5A	17	3175C>T	Q1059X*	Nonsense	DII-DIII	38	M	No	No
8	SCN5A	17	3175C>T	Q1059X*	Nonsense	DII-DIII	16	M	Yes	No
9	SCN5A	18	3352C>T	Q1118X	Nonsense	DII-DIII	51	M	No	No
10	SCN5A	21	3673G>A	E1225K	Missense	DIII-S1	46	F	No	No
11	SCN5A	21	3673G>A	E1225K	Missense	DIII-S1	43	M	No	No
12	SCN5A	21	3806A>G	N1269S*	Missense	DIII-S2/S3	53	M	No	unknown
13	SCN5A	23	4140C>G	N1380K*	Missense	DIII-S5/S6	31	M	Yes	No
14	SCN5A	Intron 24	4299+1 G>T	G1433sp	Splice	DIII-S5/S6	10	M	No	Yes
15	SCN5A	26	4501C>G	L1501V	Missense	DIII-DIV	55	M	No	No
16	SCN5A	28	4849-4851delTTC	F1617del	In-frame del	DIV-S3/S4	49	F	No	No
17	SCN5A	28	4952-4953insT	L1650+137X*	Frame shift	DIV-S4/S5	16	F	No	No
18	SCN5A	28	5150T>C	L1717P*	Missense	DIV-S5/S6	44	M	No	No
19	SCN5A	28	5227 G>A	G1743R	Missense	DIV-S5/S6	14	M	Yes	Yes
20	SCN5A	28	5324delT	N1774+11X*	Frame shift	DIV-S6	34	M	No	No
21	SCN5A	28	5494C>G	Q1832E	Missense	C-terminal	28	M	No	No
Minor BrS Genotypes
22	CACNB2B	10	1018G>A	V340I*	Missense	GK domain	58	F	No	No
23	CACNB2B	13	1497G>C	E499D*	Missense	C-terminal	60	M	No	Yes
24	HCN4	8	2522C>T	S841L	Missense	Cytoplasmic	64	F	No	No
25	KCND3	7	1798G>A	G600R	Missense	Cytoplasmic	22	M	No	Yes
26	KCNJ8	2	1265C>T	S422L*	Missense	Cytoplasmic	30	M	No	No
27	SCN1βb	3A	611A>G	Q204R*	Missense	Cytoplasmic	41	F	No	No

^*= novel mutation for this cohort; M=male; F=female

Of the 21 SCN5A mutations, 8 were “radical” mutations (5 nonsense and 3 insertion/deletion frameshift mutations), 12 were missense mutations localizing to either the N-terminus (1), transmembrane spanning region (1, IS1-IS4; 3, IIS1-S4; 3, IIIS1-S4; or 1, IVS1-S4), linker domain (2, DIII-DIV) or C-terminus (1), and 1 was an in-frame deletion, Figure 2. However, it must be recognized that nearly 2% of healthy Caucasians and 4% of seemingly healthy non-whites also host rare missense SCN5A variants, leading to a potential conundrum in the interpretation of the genetic test results(24).

Figure 2. Channel Topology of the Nav1.5 Pore-Forming Alpha Subunit Encoded by SCN5A and the Location of BrS-associated Mutations.

Putative pathogenic SCN5A BrS-associated mutations, absent in at least 700 Caucasian controls, identified in this study are indicated by yellow circles and non-synonymous “functional polymorphisms” are indicated by white circles. * = mutation novel to this cohort.

Importantly, none of the Nav1.5 missense mutations resided in low probability of pathogenicity regions of the channel (i.e. DI-DII or DII-DIII linker regions) where the vast majority of rare variants identified in health control populations reside(25). Instead, 5/12 (42%) SCN5A missense mutations resided in the critical pore-forming or S4 voltage sensing regions of Nav1.5. Putative pathogenic mutations in all other genes (CACNB2B, KCNJ8, KCND3, SCN1B, and HCN4) were missense mutations.

While 21 (16.3%) patients hosted SCN5A mutations overall, only 6 patients (4.6%) were identified with a mutation in one of the 11 other BrS-susceptibility genes (Table 2). Two patients (1.5%) were identified with a mutation in an auxiliary L-type calcium channel subunit. An asymptomatic 60- year-old Caucasian male, presenting with a spontaneous type 1 ECG pattern, a QTc of 428 ms, and a positive family history was identified as hosting E499D-CACNB2B and an asymptomatic 58-year-old Caucasian female with a spontaneous type 1 ECG pattern, a QTc of 447 ms, and a negative family history was identified with V340I-CACNB2B. The remaining four patients hosted one of the following mutations: KCNJ8-S422L, KCND3-G600R, SCN1Bb-Q204R and HCN4-S841L (Table 2). The two patients carrying KCNJ8-S422L and KCND3-G600R have been described previously and both mutations conferred a marked gain-of-function to their respective potassium channel(16,17). No mutations were identified in KCNE3, SCN3B, GPD1L, or MOG1.

Rare Genetic Variants

Besides these aforementioned putative BrS-associated mutations, 11 uncommon non-synonymous genetic variants (present in published and internal controls with a measurable frequency > 0.01% but < 1%) were identified in 13 additional patients (Table 3). Seven of the 11 rare variants (S216L-SCN5A(26), R1512W-SCN5A(27), R214Q-SCN1Bb(28), S160T-CACNB2B(9) [identified in 2 cases], S709N-CACNA2D1(9) [2 cases], S755T-CACNA2D1(29), and L450F-KCND3(16)) have been characterized functionally as electrophysiologically abnormal and/or associated with BrS or other genetically-transmitted arrhythmogenic diseases linked to BrS (i.e. Short QT Syndrome and Early Repolarization Syndrome), suggesting that these genetic variants may contribute to the development of BrS. However, despite their previous implication in these various disease states, we chose to be ultra-conservative and did not consider these variants as disease-causing mutations due to their presence in ostensibly healthy controls and therefore excluded them from our overall yield and genotype-phenotype correlations.

Table 3.

Summary of Additional Rare Genetic Variants Identified in BrS Probands

Gene	Exon	Nucleotide Change	Mutation	Number of BrS Cases Hosting the Genetic Variant (n=129)	Frequency in Controls (Caucasians)	Results Summary from Published Cellular Electrophysiology Functional Studies	Associated Disease
SCN5A	6	647C>T	S216L	1	4/1300 (18)	INA Loss of function(26)	BrS
SCN5A	26	4534C>T	R1512W	1	1/1300 (18)	Slow inactivation recovery of INA(27)	BrS
CACNA1C	17	2449C>T	P817S	1	4/476	-	-
CACNA1C	42	5150C>G	A1717G	1	0/796	-	-
CACNA1C	46	5918G>A	R1973Q	1	1/489	-	-
CACNB2B	6	479G>C	S160T	2	3/507	-	ERS
CACNA2D1	26	2126G>A	S709N	2	3/466	-	BrS
CACNA2D1	28	2264G>C	S755T	1	2/798	ICa Loss of function(29)	SQTS
KCNE3	1	248G>A	R83H	1	2/300	-	-
KCND3	3	1348C>T	L450F	1	1/800	Ito Gain of function(16)	BrS
SCNIBb	3A	641G>A	R214Q	1	4/807	INA Loss of function(28)	SIDS

BrS=Brugada Syndrome; ERS=Early Repolarization Syndrome; SQTS= Short QT Syndrome; SIDS=Sudden Infant Death Syndrome

Influence of Phenotype on the Mutation Detection Yield

In order to better understand potential phenotypic effects on the yield of mutational analysis, we further divided our 129 patient-cohort into 8 specific phenotypic categories: 1) asymptomatic, no family history, spontaneous ECG pattern only (31 cases), 2) asymptomatic, no family history, drug-induced ECG pattern only (52 cases), 3) asymptomatic, positive family history, spontaneous ECG pattern (17 cases), 4) asymptomatic, positive family history, drug-induced ECG pattern only (13 cases), 5) symptomatic, no family history, spontaneous ECG (6 cases), 6) symptomatic, no family history, drug-induced ECG only (3 cases), 7) symptomatic, positive family history, spontaneous ECG pattern (7 cases), and 8) symptomatic, positive family history, drug-induced ECG pattern only (0 cases). While the overall mutation discovery yield was 21%, this yield ranged from a low of 0% for those patients with symptoms, no family history, and a drug-induced ECG pattern (0/3) to as high as 50% for those patients with symptoms, no family history, and a spontaneous ECG pattern (3/6). However, none of these differences in yield achieved statistical significance due to the small sample sizes of each sub category (Figure 3). Notably, there was no difference in mutation detection yield between those patients with solely a spontaneous or drug-induced type 1 Brugada ECG pattern (19/83, 23%) compared to those who fully satisfied the current clinical definition of BrS (8/46, 17%, p = 0.51, Table 2)

Figure 3. Influence of Phenotype on the Mutation Detection Yield.

Depicted is a bar graph comparing the percent yield of the eight specific phenotypic categories showing the phenotypic effects on overall mutation yield of our cohort. The number in the bar represents the number of cases with a mutation and the percent yield is highlighted above each bar.

An interesting effect of age on the mutation detection yield was observed, especially among males. Overall, the 20 mutation positive males were younger (33 ± 15 years) compared to the 84 mutation negative males (43 ± 12 years, p=0.001). However, there was no real difference in average age between mutation positive (43 ± 16 years) and mutation negative females (51 ± 17 years, p=0.34). The overall yield was significantly greater among BrS patients younger than 20 years of age (6/8, 75%) compared to patients between 20 to 40 years (10/46, 22%) and those over the age of 40 years (11/75, 15%, p=0.0003, Figure 4). This significant difference was even more striking when comparing males only [< 20 years, (5/6), 83%; 20-40 years, (9/42), 21%; > 40 years, (6/56), 11%, p<0.0001]. The female group was too small to draw any meaningful conclusions [< 20 years, (1/2), 50%; 20-40 years, (1/4), 24%; >40 years, (5/19), 26%]. Interestingly however, this age effect on the mutation detection yield was only evident among the 46 clinically certain BrS patients [< 20 years, (4/4), 100%; 20-40 years, (3/13), 23%; > 40 years, (1/29), 3%, p < 0.0001] compared to the 83 referral patients with only a type 1 Brugada ECG pattern [< 20 years, (2/4), 50%; 20-40 years, (7/33), 21%; > 40 years, (10/46), 22%, p=0.42].

Figure 4. Effect of Age on the Mutation Detection Yield for the Overall, Clinically Strong, and Type I ECG Pattern Only Cohorts.

Depicted is a bar graph showing the percent yield of the 3 different age groups (<20 years of age, 20-40 years of age, >40 years) for the overall, clinically diagnosed, and type 1 Brugada ECG pattern only cohorts. The number in the bar represents the number of cases with a mutation and the percent yield is highlighted above each bar.

Genotype-Phenotype Correlations

When comparing SCN5A mutation positive individuals with those patients that were SCN5A negative, the PQ interval was significantly longer (191 ± 31 ms vs. 169 ± 28, p=0.0015). In fact, a SCN5A mutation was identified in 39% of those with a PQ ≥ 200 ms compared to only 8% of those with a PQ < 200 ms (OR 7; 95% CI 3-20; p<0.0001). For the subset with an isolated, idiopathic type 1 ECG pattern only, 38 % of the patients with a PQ ≥ 200 ms were SCN5A positive compared to 11% with a PQ < 200 ms (OR 8, 95% CI 1.5-16, p=0.006). Due to the rarity of mutations identified in BrS genes 2-12, we are unable to provide genotype-phenotype correlations for these specific BrS genotypes. However, none of the patients in this cohort had a QTc < 350 ms which might explain the total absence of CACNA1C-mediated BrS.

DISCUSSION

In 2011, two consensus documents were published on the diagnostic, prognostic, and therapeutic impact of genetic testing in the clinical evaluation of cardiac channelopathies and cardiomyopathies(20,21). For BrS, both documents recommended genetic testing for any patient for whom there is a clinical suspicion for BrS, and emphasized the importance of genetic testing of the index case in relation to overall family screening. While the Heart Rhythm Society (HRS)/European Heart Rhythm Association (EHRA) Expert Consensus Statement(21) indicated that either a comprehensive or (BrS1) SCN5A-targeted genetic screen “may be useful”, the Canadian Cardiovascular Society (CCS) /Canadian Heart Rhythm Society (CHRS) joint position paper(20) advised limiting BrS genetic testing to only SCN5A and to only consider the minor genes (BrS 2-12) under special circumstances. In addition, while the HRS/EHRA group advised genetic testing for those patients that fulfill the task force criteria for a clinical diagnosis of BrS which requires an expressed type 1 Brugada ECG pattern plus one or more clinical variables from the patient’s personal or family history (such as unexplained syncope or a family history of sudden cardiac death), the CCS/CHRS group recommended genetic testing in both clinically diagnosed BrS patients and for asymptomatic patients with only a type 1 Brugada ECG pattern.

Drawn from the largest cohort of unrelated patients referred for BrS genetic testing to be systematically analyzed for mutations in the 12 known BrS-susceptibility genes (as of June 1, 2011), there are several key observations that may further buttress and refine these expert opinion recommendations. First, our results show that there was no significant difference in mutation detection yield between those patients who fully satisfied the clinical definition of BrS (17%) and those patients with only a diagnostic type 1 Brugada ECG pattern (23%) suggesting that BrS genetic testing may be equally warranted for patients with solely an electrocardiographic manifestation of a type 1 Brugada ECG pattern for the main purpose of identifying probands and their family members that should take precautionary measures in certain conditions. This observation lends evidence to support the CCS/CHRS position on genetic testing of asymptomatic individuals, and suggests that perhaps the HRS/EHRA recommendation (requiring both an abnormal ECG and personal symptoms or family history) may be too strict. As implied in the Canadian guidelines, a positive SCN5A genetic test result plus a spontaneous/drug-induced type 1 Brugada ECG pattern may be sufficient for the clinical diagnosis of BrS. However, one must be mindful that the presence of a positive genetic test result is not predictive of clinical symptoms as there are pedigrees with probands that exhibit incomplete penetrance and a lifelong asymptomatic course. Nevertheless, if our observation is validated, then the genetic test might become part of the diagnostic criterion akin to the revised criteria for both arrhythmogenic right ventricular cardiomyopathy and Marfan syndrome. Furthermore, the identification of a mutation positive subject would at least enable the simple Brugada preventative measures of avoiding certain drugs (www.brugadadrugs.org) and reducing the degree of hyperthermia in the setting of febrile illnesses.

Second, whether dealing with patients with either BrS or just a type 1 Brugada ECG pattern, the minor BrS-susceptibility genes are indeed minor. In other words, the sensitivity of the BrS genetic test is affected minimally by their inclusion lending additional merit for the option of SCN5A (BrS1) only genetic testing rather than so-called comprehensive, multi-gene BrS genetic testing. This notion of targeted genetic testing for BrS was supported by both guidelines as well. In addition, this study provides useful pre-genetic test anticipatory guidance as to its pre-test probability of returning positive. Previously, analysis of SCN5A among over 2000 patients, derived from 9 different cohorts of BrS throughout the world, indicated an 11 – 28% yield (average = 21%) for possible/probable BrS1 status(19). Consistent with that range, the overall yield in this study was 16.3% for SCN5A-mediated BrS (i.e. BrS1). However, the yield was far greater among young men ≤ 20 years of age with clinically manifest BrS and among those with BrS and a prolonged PQ interval. Compared to a < 10% yield for a positive SCN5A test for those with a PQ interval < 200 ms, the yield was almost 40% when the PQ interval ≥ 200 ms. This is consistent with prior observations that SCN5A-positive BrS patients displayed prolonged HV and PQ intervals during electrophysiology study(30,31).

Finally, in the absence of a short QT interval, calcium channel-mediated BrS is extremely uncommon. Here, mutations in genes encoding the alpha 1 (CACNA1C), beta 2b (CACNB2B), and alpha2delta1 (CACNA2D1) were observed in less than 2% of this cohort. In contrast, perturbations involving the calcium channel macromolecular complex was implicated as the second most common genetic cause for BrS accounting for 12% of the disease, and up to 18% when including rare polymorphisms(9). A close examination of those seminal discoveries underscores the tight link between calcium channel mediated disease and the clinical phenotype of BrS with concomitant short QT interval, where 50% of patients with BrS/SQT hosted a mutation in an L-type calcium channel subunit(9). Once again, this illustrates the critical importance of the phenotypic classification and the opportunity for phenotype guided genetic testing within what is currently captured under the header of Brugada syndrome. Given the increased recognition of so-called “background genetic noise” or potential false positives with respect to genetic testing for heritable arrhythmias and/or cardiomyopathic syndromes(25,32), phenotype-targeted testing within the spectrum of J wave syndromes may minimally compromise the test’s sensitivity while significantly enhancing its specificity. Just like a SCN5A-centric genetic test for classical Brugada syndrome especially with concomitant PQ interval prolongation, this data would suggest similar consideration for primary genetic testing of the genes that encode the calcium channel’s pore-forming subunit and its auxiliary subunits, rather than SCN5A, for the ECG phenotype of type 1 Brugada ECG pattern with concomitant short QT intervals.

Limitations

There are two major limitations with this study, one dealing with the veracity of the phenotype and the other dealing with the certainty of the genotype. While the ECG phenotype was vetted for every one of the patients in this study by at least one of the authors, not every patient in this study was evaluated clinically by the authors, as our study population consisted of patients referred to our laboratories for genetic testing. Accordingly, for the 83 patients with only a type 1 Brugada ECG pattern, since we did not personally ask the questions to elicit the history for every patient in our cohort, we do not know that the personal and family history was indeed negative. However, this potential failure to elicit the clinical information is unlikely as one of the co-authors (LC) directly evaluated over 80% of the patients in this cohort, and the same referring physicians submitted patients with detailed characterization of positive symptomatology. With respect to the genotype, the major limitation of this study pertains to the issue of mutation calling(19,24,25). None of the 15 novel mutations reported in this study cohort have been characterized functionally. Instead, strict absence from our internal set of ethnic-matched reference alleles and all available online databases, including the 1000 Human Genome Project database(23), was required. Despite this stringent bar to establish rarity (i.e. not seen in over 3,000 reference alleles), it is possible that some of these variants listed as putative pathogenic mutations may be innocuous functionally. Conversely, this strict “absence from all controls” definition may have resulted in the exclusion of other variants that are nevertheless disease-contributing. S216L-SCN5A and R1512W-SCN5A illustrate this possibility. In vitro expression of S216L-SCN5A channels was associated with a 60% reduction in peak sodium current and moderate slowing of inactivation(26). Similarly, in vitro studies of R1512W-SCN5A demonstrated slower inactivation and recovery from inactivation consistent with a “loss of function” phenotype anticipated for BrS1 mutations(27). However, both S216L and R1512W have been previously reported in ostensibly healthy controls. Assignment of a given variant’s pathogenicity is a vexing problem and has been declared the Achilles’ heel of genetic testing(21,25,33). This issue underscores the critical need to align carefully the phenotype with the appropriate genetic test panel that best balances the sensitivity/specificity issue rather than simply continuing to add the next novel disease-susceptibility gene.

Conclusion

To our knowledge, this study is the first comprehensive mutational analysis of all 12 BrS-susceptibility genes discovered to date for a single large BrS cohort. SCN5A-mediated BrS (BrS1) is still the only common genetic substrate for BrS in general, particularly for young BrS males (<20 years) and those BrS patients with a PQ interval ≥ 200ms who may have a 40 – 50% pre-test probability for a positive SCN5A genetic test result. In addition, our data suggest that both clinically diagnosed BrS patients and patients presenting with only a spontaneous/drug-induced type 1 Brugada ECG pattern may equally warrant SCN5A genetic testing. Finally, the other 11 genes account for < 5% of our cases and in the absence of BrS plus a short QT interval, calcium channel-mediated BrS is very uncommon, far below the initial 10-15% estimates.

Abbreviations and Acronyms list

BrS

Brugada Syndrome

SCD

sudden cardiac death

ECG

electrocardiogram

HRS

Heart Rhythm Society

EHRA

European Heart Rhythm Association

CCS

Canadian Cardiovascular Society

CHRS

Canadian Heart Rhythm Society

CI

confidence interval

OR

odds ratio

SQT

Short QT