Journal of the American College of Cardiology Vol. 60, No. 8, 2012© 2012 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00

STATE-OF-THE-ART PAPER

Genetics of Hypertrophic CardiomyopathyAfter 20 YearsClinical Perspectives

Barry J. Maron, MD,* Martin S. Maron, MD,† Christopher Semsarian, MB, BS, PHD‡

Minneapolis, Minnesota; Boston, Massachusetts; and Sydney, New South Wales, Australia

Hypertrophic cardiomyopathy (HCM) is the most common familial heart disease with vast genetic heterogeneity,demonstrated over the past 20 years. Mutations in 11 or more genes encoding proteins of the cardiac sarco-mere (�1,400 variants) are responsible for (or associated with) HCM. Explosive progress achieved in under-standing the rapidly evolving science underlying HCM genomics has resulted in fee-for-service testing, makinggenetic information widely available. The power of HCM mutational analysis, albeit a more limited role than ini-tially envisioned, lies most prominently in screening family members at risk for developing disease and exclud-ing unaffected relatives, which is information not achievable otherwise. Genetic testing also allows expansion ofthe broad HCM disease spectrum and diagnosis of HCM phenocopies with different natural history and treat-ment options, but is not a reliable strategy for predicting prognosis. Interfacing a heterogeneous disease such asHCM with the vast genetic variability of the human genome, and high frequency of novel mutations, has createdunforeseen difficulties in translating complex science (and language) into the clinical arena. Indeed, probanddiagnostic testing is often expressed on a probabilistic scale, which is frequently incompatible with clinical deci-sion making. Major challenges rest with making reliable distinctions between pathogenic mutations and benignvariants, and those judged to be of uncertain significance. Genotyping in HCM can be a powerful tool for familyscreening and diagnosis. However, wider adoption and future success of genetic testing in the practicing cardio-vascular community depends on a standardized approach to mutation interpretation, and bridging the communi-cation gap between basic scientists and clinicians. (J Am Coll Cardiol 2012;60:705–15) © 2012 by theAmerican College of Cardiology Foundation

Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jacc.2012.02.068

For over 50 years, hypertrophic cardiomyopathy (HCM)has been recognized as a familial cardiac disease with highlyvisible risk for sudden death and disease progression, char-acterized by heterogeneous phenotypic expression, naturalhistory, and genetic profile (1–6). HCM is the mostcommon monogenic cardiovascular disease with a preva-lence of 1:500 (7).

For HCM, the molecular era emerged more than 20 yearsago with identification of disease-causing mutations ingenes coding for proteins of the cardiac sarcomere (5,8,9).These breakthrough observations defining the basic geneticsubstrate were accompanied by considerable optimism andexpectation that mutational analysis would revolutionize

From the *Hypertrophic Cardiomyopathy Center, Minneapolis Heart InstituteFoundation, Minneapolis, Minnesota; †Hypertrophic Cardiomyopathy Center, TuftsMedical Center, Boston, Massachusetts; and the ‡Agnes Ginges Centre for Molec-ular Cardiology; Centenary Institute; Sydney Medical School, University of Sydney;and Department of Cardiology, Royal Prince Alfred Hospital, Camperdown, Sydney,New South Wales, Australia. Dr. B. J. Maron is an advisor for GeneDx and a granteeof Medtronic. Dr. M. S. Maron has reported that he has no relationships relevant to thecontents of this paper to disclose. Dr. Semsarian is the recipient of a PractitionerFellowship from the National Health & Medical Research Council, Canberra, Australia.

Manuscript received October 5, 2011; revised manuscript received January 20,2012, accepted February 3, 2012.

HCM with regard to diagnosis and prediction of clinicalcourse, as well as guiding management (10–12).

More recently, striking scientific advances in moleculargenetics have resulted in availability of comprehensivecommercial genetic testing to the practicing cardiovascularcommunity, while paradoxically creating many unansweredquestions, and communication gaps surrounding translationof genetic information to clinical decision making. Therefore,it is timely to place the benefits and challenges of genotyping inHCM into perspective, for a disease that epitomizes applica-tion of genetic science to cardiovascular medicine.

Historical Perspectives

HCM is inherited in an autosomal dominant Mendelianpattern with variable expressivity and age-related penetrance(3–5,9,11,13). Offspring of an affected individual have a50% probability of inheriting a mutation and risk fordisease; alternatively, sporadic cases may be due to de novomutations in the proband but absent from the parents.

For its first 25 years, diagnosis of HCM could only bemade through integration of examination, electrocardio-

gram, and invasive angiographic/hemodynamic studies, dis-

bt

706 Maron et al. JACC Vol. 60, No. 8, 2012Genetics in Hypertrophic Cardiomyopathy August 21, 2012:705–15

proportionately identifying pa-tients with left ventricular (LV)outflow obstruction (1) (Fig. 1).

In the early 1970s, echocardi-ography afforded noninvasive vi-sualization of left ventricular hy-pertrophy (LVH), and reliableidentification of family memberswith and without the HCM phe-notype (14). Introducing basicscience to HCM, by interfacingdeoxyribonucleic acid (DNA)–based methodologies and classi-cal segregation linkage analysis

with echocardiography, allowed mapping HCM to a caus-ative locus on chromosome 14 in 1989 (8). In 1990,sequence analysis of a candidate gene revealed a pathogenicmissense mutation in the beta-myosin heavy chain gene(MYH7Arg 403 Gln) to be responsible for HCM (9).

Molecular Basis of HCM

The substrate, 2011. Two decades of intensive investiga-tion have defined the vast and daunting heterogeneity of theHCM substrate. The early report of 7 mutations in 1 gene(MYH7) (10) has now expanded to 11 or more causativegenes (5,13,15–21) with �1,400 mutations (Dr. H. Rehm,personal communication, August 2011), expressed primarilyor exclusively in the heart. These genes encode thick and thinmyofilament proteins of the sarcomere or contiguous Z-disc(Table 1). Mutations in several additional sarcomere (orcalcium-handling) genes have been proposed, but with lessevidence supporting pathogenicity (Table 1). Of thosepatients with positive genetic tests, about 70% are found tohave mutations (of either definite or uncertain pathogenic-ity) in the 2 most common genes, MYH7 and myosin-inding protein C (MBPC3), while other genes includingroponin T, troponin I, �-tropomyosin, and �-actin each

account for a small proportion of patients (1% to 5%).Types of mutations. The vast majority (about 90%) ofpathogenic mutations altering physical and functional prop-erties of proteins are missense, in which a single normalamino acid is exchanged for another (e.g., replacement ofarginine for glutamine). Alternatively, more radical muta-

Abbreviationsand Acronyms

GINA � GeneticInformation Non-Discrimination Act

HCM � hypertrophiccardiomyopathy

LV � left ventricle/left ventricular

LVH � left ventricularhypertrophy

VUS � variant of uncertainsignificance

Figure 1 HCM Landmarks

Genetic and clinical advances over the �50-year history of hypertrophic cardiomyo

tions affect many amino acids in the protein, resulting in avery different product (i.e., frameshift), and are generallypredicted to trigger more substantial clinical consequences.Frameshift mutations are caused by insertion or deletion of�1 nucleic acids in the coding region often resulting inshortened truncated proteins (frequently found in theMYBPC gene), or abnormal splicing of messenger ribonu-cleic acid (mRNA).

Genetic Testing for HCM

Background. HCM genetic testing was initially confinedto the realm of a few research laboratories focused onenhancing a basic understanding of this disease. Testingresults were unpredictable, as laboratories lacked sufficient

(HCM). ICD � implantable cardioverter-defibrillator; SD � sudden death.

Molecular Substrate of HCMTable 1 Molecular Substrate of HCM

Strongest evidence for pathogenicity

Thick filament

1. �-myosin heavy chain MYH7

2. Regulatory myosin light chain MYL2

3. Essential myosin light chain MYL3

Thin filament

4. Cardiac troponin T TNNT2

5. Cardiac troponin I TNNI3

6. Cardiac troponin C TNNC1

7. �-tropomyosin TPM1

8. �-cardiac actin ACTC

Intermediate filament

9. Cardiac myosin-binding protein C MYBPC3

Z-disc

10. �-actinin 2 ACTN2

11. Myozenin 2 MYOZ2

Lesser evidence for pathogenicity

Thick filament

12. �-myosin heavy chain MYH6

13. Titin TTN

Z-disc

14. Muscle LIM protein CSRP3

15. Telethonin TCAP

16. Vinculin/metavinculin VCL

Calcium handling

17. Calsequestrin CASQ2

18. Junctophilin 2 JPH2

HCM � hypertrophic cardiomyopathy.

pathy

mm

erci

alG

enet

icTe

stin

gS

ervi

ces

inth

eU

nite

dS

tate

sfo

rH

CM

*ab

le2

Com

mer

cial

Gen

etic

Test

ing

Ser

vice

sin

the

Uni

ted

Sta

tes

for

HC

M*

Com

pany

Beg

anH

CM

Test

ing

(Yea

r)W

ebsi

teTe

leph

one

No.

ofH

CM

Gen

esin

Pan

el†

Turn

arou

ndTi

me

(Wee

ks)§

§

No.

ofP

roba

bilit

yC

ateg

orie

s‡

Gen

etic

Con

sult

atio

nA

vaila

ble§

Var

iant

Rec

lass

ifica

tion

�

Cos

t(P

roba

ndTe

stin

g)

Cos

t(O

ther

Fam

ilyM

embe

rs)

No.

ofO

ther

Gen

etic

Dis

ease

sTe

sted

eneD

x(G

aith

ersb

urg,

Mar

ylan

d)2

00

8w

ww

.gen

edx.

com

30

1-5

19

-21

00

18

85

��

$3

.37

5#

¶$

35

03

50

rans

geno

mic

-FA

MIL

ION

(New

Hav

en,C

onne

ctic

ut)

20

08

ww

w.fa

mili

on.c

om8

77

-27

4-9

43

21

24

–63

��

$5

,40

0#

$9

00

10

orre

lage

nD

iagn

ostic

s(W

alth

am,M

assa

chus

etts

)2

00

7w

ww

.cor

rela

gen.

com

86

6-6

47

-07

35

16

6–8

7�

�$

3,9

75

$2

50

40

artn

ers*

*(C

ambr

idge

,Mas

sach

uset

ts)

20

03

ww

w.p

cpgm

.par

tner

s.or

g6

17

-76

8-8

50

01

8†

†5

5�

�$

3,2

00

‡$

40

01

00

ting

also

avai

labl

eab

road

,inc

ludi

ngN

orw

ay(R

iksh

ospi

tletM

edic

alG

enet

icLa

bora

tory

),D

enm

ark

(Sta

tens

Ser

umIn

stitu

te),

and

the

Uni

ted

Kin

gdom

(Nat

iona

lHea

lthSer

vice

).†

Usu

ally

incl

udes

met

abol

icph

enoc

opie

s(e

.g.,

LAM

P2

,PR

KA

G2

,Fab

rydi

seas

e).‡

Toas

sess

atio

nalp

atho

geni

city

.§Cou

rtes

yin

terp

reta

tion

and/

orcl

arifi

catio

nof

test

ing

repo

rtby

tele

phon

e/em

ail.

�Sys

tem

atic

notifi

catio

nto

phys

icia

n/pa

tient

(“lo

okba

ckte

stin

g”)a

vaila

ble.

¶In

sura

nce

prog

ram

limits

dire

ctpa

tient

cost

to$

10

0,w

here

this

polic

yis

perm

issi

ble

erlo

cala

ndst

ate

law

s.#

Test

ing

com

pany

take

spr

imar

yre

spon

sibi

lity

for

esta

blis

hing

the

insu

ranc

eco

mpo

nent

ofpa

ymen

tin

conj

unct

ion

with

the

com

mer

cial

carr

ier.

**P

artn

ers

Hea

lthCar

eCen

ter

for

Per

sona

lized

Gen

omic

Med

icin

e(H

arva

rdU

nive

rsity

,Cam

brid

ge,

sach

uset

ts).

††

Als

oav

aila

ble

asa

pan-

card

iom

yopa

thy

pane

lof

46

gene

s,in

clud

ing

HCM

(cos

t:$

3,9

00

).‡

Test

spe

rfor

med

grat

isfo

rco

segr

egat

ion

tore

solv

eVU

S.§

§W

hen

mut

atio

nis

know

nin

prob

and,

and

with

gene

rally

shor

ter

turn

arou

ndtim

esof

2to

4w

eeks

.CM

�hy

pert

roph

icca

rdio

myo

path

y;VU

S�

varian

tof

unce

rtai

nsi

gnifi

canc

e.

707JACC Vol. 60, No. 8, 2012 Maron et al.August 21, 2012:705–15 Genetics in Hypertrophic Cardiomyopathy

resources to accommodate all clinical requests. In 2003,genetic testing entered the mainstream of the healthcaresystem with automated DNA sequencing providing rapid,reliable, and comprehensive molecular diagnosis on a fee-for-service basis. DNA testing typically requires transport-ing 5 to 10 ml of whole blood to institutional or commerciallaboratories (Table 2), all offering similar panels, includingestablished HCM causative genes, phenocopies, and possi-bly other genes with lesser evidence for pathogenicity. U.S.proband testing may be associated with significant cost andeconomic burden, but also with a variety of insurance andbilling strategies available.Pathogenic versus nonpathogenic mutations. A mutationcan be considered pathogenic (or likely pathogenic) on thebasis of the preponderance of evidence from the followingcriteria (15,22) (Table 3): 1) cosegregates with the HCMphenotype (i.e., LV hypertrophy) in family members, i.e., theonly criterion that relies on data obtained directly from thepatients; 2) previously reported or identified as a cause ofHCM; 3) absent from unrelated and ethnic-matched normalcontrols; 4) protein structure and function is importantlyaltered (e.g., frameshift with truncation); and 5) amino acidsequence change in a region of the protein otherwise highlyconserved through evolution with virtually no variation ob-served among species, suggesting its importance to basiccellular function.

On the other hand, many amino acid substitutions in DNAsequence do not cause disease and are regarded as benignpolymorphisms (i.e., variants not generally expected to bedeleterious), by convention occurring in �0.5% to 1.0% ofethnic-specific normal control populations. Nevertheless, therelevance for causing disease attached to a significant minorityof such identified variants remains unclear, even after applyingall criteria for pathogenicity. As a result, these mutations areassigned by genetic test reports into an ambiguous category(i.e., variants of uncertain significance [VUS or VOUS])(15,21,22) with virtually no clinical utility for family screening.

Indeed, distinguishing pathogenic mutations from VUS,or rare nonpathogenic variants, has increasingly emerged asa dilemma for interpreting testing results in HCM, and hasbeen regarded as the “Achilles heel” of this diagnosticstrategy (15). This issue has become particularly challengingas the reduced cost of technology now allows comprehensiveDNA sequencing of the exome and even the whole genome(23,24,25). Although affording scientific insights, this de-velopment also substantially increases the recognition ofVUS, potentially creating only further ambiguity in testreports. That VUS occur more commonly in human controlpopulations than previously appreciated (genetic variationreferred to collectively as “background noise”) underscoresthe need for more definitive criteria to differentiate patho-genic mutations from benign genetic “noise.”

Notably, generally accepted guidelines for interpreting VUSare currently lacking, with estimated frequencies varying dra-matically from 5% to 50%, largely dependent on the number of

pathogenicity classes used to categorize mutations and the Co T G T C P

*Tes

mut

und

Mas H

708 Maron et al. JACC Vol. 60, No. 8, 2012Genetics in Hypertrophic Cardiomyopathy August 21, 2012:705–15

number of genes comprising the testing panel (Fig. 2). Labo-ratory commercial testing strategies vary widely, employingfrom 3 to 7 descriptive pathogenicity classes to constructformal reports, thereby creating the distinct possibility thatdifferent interpretations of pathogenicity may emanate fromdifferent laboratories for the same mutation (Table 2).

Indeed, it is not universally appreciated in the clinicalcardiovascular community that molecular diagnosis and assign-

Current Criteria Used to Determine Probability for Pathogenicity ofTable 3 Current Criteria Used to Determine Probability for Path

Pathogenicity Criterion Description

Cosegregation Determine whether mutation is presentwith LVH and absent in those without

Prior evidence of pathogenicity Documentation that mutation is HCM diin �1 patient in published literature,individual experience of a testing labo

Control population Confidence for pathogenicity increased wabsent from large, ethnicity-matchedhealthy population

Major disruption protein structure,and function

Mutant proteins are judged to have subsaltered physical properties�

*Presented in approximate order of the power for establishing disease-causing status for a given mscreening the threshold burden of evidence for a gene already known to cause HCM is lower thanodds) score �3. ‡Databases proposed to systematically assemble genetic data from all clinicaassembled its own control group. §�100 subjects (200 alleles) considered the minimum; controsequence, associated with insertions/deletions, or resulting in truncated (shortened) protein; 2) olocation in a functionally important site in genetic sequence leading to dysfunction. ¶Includes: 1) inusing extracted protein of various species; 3) predictive software program models.

HCM � hypertrophic cardiomyopathy; LVH � left ventricular hypertrophy.

Figure 2 Pathogenicity in Probands

Genetic testing interpretation categories are depicted with respect to the relative sthese test results occur. These variant classifications are those commonly used bBenign and likely benign variants are not responsible for disease (polymorphisms)of which are currently unresolved in terms of disease pathogenesis or causation.nificance. Such category assignments are made largely on the basis of expert pro

ment of mutations to pathogenic status is often made on aprobabilistic basis, and not necessarily as a definitive binary(“yes” [positive] or “no” [negative]) test result. Certainly, theexpectation threshold may have been set too high for genotyp-ing initially, given the relatively few definitive scenarios thatemerge in HCM, with the pathogenicity of sequence variantsoften difficult to establish with certainty. This principle is oftendifficult to merge with clinical practice decision making, and

CM Mutation*icity of an HCM Mutation*

Potential Limitations for Interpretation

tives Often impracticalFamily size may be small/relatives unavailableFamily compliance unpredictableRequires resources for imaging/DNA studies in �3 relatives

(other than proband) including �1 with HCM phenotype†

causinge

Absence of established comprehensive, curated, and cooperativedatabase tabulating mutations‡

High rate of novel (de novo; “private”) mutations in 65% ofprobands

Interpretation of pathogenicity can be inconsistent amongtesting laboratories

utationibly

Often insufficient size§Control subjects should be unrelated, ethnicity-specific and free

of the disease in questionPotentially pathogenic variants can occur in subjects judged

clinically normalMany rare benign (missense) variants in normals, termed

“background noise”

lly Inferred from evidence obtained from in nonhuman sources¶

, although meeting any 1 of the criteria can be sufficient to assign pathogenicity. †In clinical familyw gene in a research setting requiring linkage analysis over 3 generations with LOD (logarithm ofsearch laboratories include MutaDATABASE and Clin-Var; currently each clinical laboratory hass of up to 400 subjects are in use. �Includes: 1) Frameshift with profoundly altered amino acid

g in amino acids otherwise highly conserved through evolution in humans and across species; 3)utated mouse or rabbit models of HCM; 2) in vitro basic metabolic cell culture assay experiments

h of each for clinical practice, and independent of the frequency with whichng laboratories in formatting reports, to express the probability of pathogenicity.nts of uncertain significance (VUS) are single nucleotide variants the significancee of a mutation (in proband), is considered indeterminate, and of no clinical sig-al judgment by clinical laboratory geneticists using currently available data.

an Hogen

in relaLVH

sease–or in thratory

hen mostens

tantia

utationfor a nel and rel groupccurrinvivo m

trengty testi. VariaAbsencfession

iup

foswcttlai(n�iab

bpmnuei

ioflc

a3miGnlwmLmdi

709JACC Vol. 60, No. 8, 2012 Maron et al.August 21, 2012:705–15 Genetics in Hypertrophic Cardiomyopathy

translating complex molecular science to patient care hasproved more challenging than initially anticipated.

In addition, not generally recognized by clinicians is thepossibility that pathogenic classifications of variants canchange over time as new relevant information becomesaccessible (e.g., a VUS can be reassigned to pathogenic [oralternatively, mutations regarded as disease-causing may bedowngraded], scenarios that could be considered a second“Achilles heel” of genetic testing). Responsibility for noti-fication of such reassignments to physicians and patients hasunavoidably been assumed by genetic testing laboratories.However, the optimal strategies for this process are unclearand standard procedures have not been established toresolve these unique circumstances.

Clinical Applications of Genetic Testing

Mutation-specific screening of family members for HCM.A common question asked by cardiologists is “Should myHCM patient have genetic testing?” and it reflects thepersistent confusion regarding molecular strategies in thisdisease (Figs. 3 and 4). Some form of family screening for HCMs universally recommended (26), with the preferred first optionsually clinical testing with cardiac imaging and electrocardiogra-hy to identify phenotype-positive relatives (27) (Table 4).

Screening of family members is also the predominant roleor genetic testing (i.e., to identify those at risk for devel-ping disease who do not have LV hypertrophy) (26). Thistrategy is initiated by successfully genotyping the probandith clinically expressed HCM. Failure to identify the

ausative mutation in the proband is an indeterminate resulthat provides no useful information and precludes predictiveesting in family members. It should be underscored that theikelihood of obtaining a positive test in the proband is onlybout 50%, as all genes causing HCM have not yet beendentified, and are absent from testing panels (19,20,26)Fig. 5). Furthermore, many of the detected mutations willot be judged pathogenic, thereby eliminating substantially50% of families from the option of genotyping for

dentification of relatives at risk for HCM. Therefore, in onlyminority of HCM probands will the result of the genetic test

e actionable for family screening.If a pathogenic mutation is identified in a proband, it

ecomes a tool for screening at-risk relatives, with severalossible diagnostic scenarios (26) (Figs. 3 and 4). In theost robust application of genetic testing, relatives testing

egative for the known family mutation are considerednaffected. This result largely alleviates the psychological andconomic burden of further cardiovascular surveillance, imag-ng tests or lifestyle and competitive sports restrictions (28).

However, extraordinarily rare (but possible) scenarios canmpact the use of genetic testing to definitively exclude the riskf developing HCM and make it imprudent to permanentlyorgo imaging tests: 1) reassignment of a mutation fromikely pathogenic to VUS; 2) proband with 1 disease-

ausing mutation in fact has a second mutation (unknown l

nd undetected), which is transmitted to an offspring;) laboratory or human error in processing/reporting autation; and 4) de novo mutation, not accounted for in

nitial proband testing, occurs in subsequent generations.enotype positive-phenotype negative. Penetration of ge-

etic testing into clinical practice has created an importantegacy, evident as a new and rapidly expanding patient subsetithin the HCM spectrum, i.e., genetically affected familyembers (usually MYH7 or MYBPC3 mutations) withoutVH (29), although often with other markers such asyocardial crypts or scarring, elongated mitral leaflets, or

iastolic dysfunction (11,30 –32) (Figs. 3 and 4). Suchndividuals have generated clinical decision-making di-

Figure 3 Predictive Cascade (Generational) Testing

(A) A 45-year-old proband with a pathogenic MYH7 mutation. Genetic testingprovided 3 diagnostic scenarios in third-generation offspring: age 10 years,unaffected status is virtually certain by absence of the mutation (reassuring tothe family given multiple HCM-SCDs); age 14 years, gene positive but pheno-type negative is at risk for developing disease, requiring continued clinical sur-veillance and imaging; age 22 years, diagnosis in patient with phenotypicallyexpressed HCM (but only mild LVH) is confirmed. (B) A 33-year-old mother withclinical HCM diagnosis (probable pathogenic MYBPC mutation); 3 offspringtested negative for that mutation, with risk to develop HCM virtually excluded.Circles � females, squares � males; plus sign � present, zero � absent;solid symbols � clinical HCM diagnosis; and slash � deceased. CMR � car-diovascular magnetic resonance; HCM � hypertrophic cardiomyopathy; LVH �

left ventricular hypertrophy; SCD � sudden cardiac death.

emmas: which (if any) should be disqualified from

tctataasd

tcoitl

710 Maron et al. JACC Vol. 60, No. 8, 2012Genetics in Hypertrophic Cardiomyopathy August 21, 2012:705–15

intense competitive sports to reduce risk, or be consid-ered for prophylactic implantable cardioverter-defibrillator therapy (28,29)?

While Bethesda Conference #36 consensus recommen-dations do not exclude genotype positive-phenotype nega-tive (G� P�) from sports (28), the paucity of evidence thathe nonhypertrophied LV is in fact electrically unstable (33)an make clinical decisions difficult (29). Given the rela-ively short period of time that genetic testing has beenvailable, the precise proportion of the G� P� populationhat will develop overt disease expression remains uncertain,lthough believed to be low. Until rigorous penetrance datare available, it is prudent to extend standard HCMurveillance with cardiac imaging at least through midlife toetect development of the phenotype.On the basis of mutated HCM mouse and rabbit models,

here is emerging interest in drug strategies (antioxidants,alcium-channel blockers, statins) to prevent developmentf the phenotype and disease progression in G� P�ndividuals (34,35). While there are no data that supporthis hypothesis, studies are in progress assessing pharmaco-

Figure 4 Genetic Family Screening Strategies in HCM

*First option for assessment of family members would be a clinical screening evathose relatives with negative or indeterminate clinical testing (for the HCM phenotwith coexisting conditions (e.g., systemic hypertension or physiologic athlete’s heatical obstacles including limited family size or patient compliance. ECG � electrocgenetic diagnosis; VUS � variant of uncertain significance; other abbreviations as

ogic prevention strategies in patients.

Suspected HCM phenocopies. An application of genetictesting is a disease-specific diagnosis in patients with LVH,due to mutations in genes distinct from those that encodesarcomere proteins. Metabolic myocardial storage cardio-myopathies, often misdiagnosed clinically, comprise animportant but small fraction of patients genotyped forsuspicion of HCM (�1%): regulatory subunit of adenosinemonophosphate-activated protein kinase (PRKAG2) glyco-gen storage disease; lysosme-associated membrane protein(LAMP2; Danon disease), an X-linked dominant lyosomalstorage disorder; Fabry, an X-linked recessive disease due tomutations in the galactosidase alpha (GLA) gene, and�-galactosidase A deficiency leading to multiorgan intracel-lular glycosphingolipid deposition (36–40).

Diagnostic distinction between sarcomeric HCM and itsphenocopies is crucial, given differences in natural historyand management strategies. For example, mutations inLAMP2 are usually associated with rapid and potentiallylethal clinical course within the first 3 decades, requiringearly consideration for heart transplant (37,38). In Fabrydisease, clinical benefits have been attributed to enzyme

with imaging tests and ECG; the option of genetic testing is triggered largely forGenetic testing may potentially lead to definitive diagnosis of HCM in patientsenetic screening often not productive; cosegregation, while an option, has prac-

raphy; ICD � implantable cardioverter-defibrillator; PGD � pre-implantationre 3.

luationype). †rt). ‡G

ardiogin Figu

replacement therapy with recombinant �-galactosidase A,

diGAoarast

tmagducgpch

P

Scuwi

mieovcspedcshardMdcshrmap

711JACC Vol. 60, No. 8, 2012 Maron et al.August 21, 2012:705–15 Genetics in Hypertrophic Cardiomyopathy

including regression of LVH and improved myocardialfunction and exercise capacity (39,40).

Clinical suspicion sufficient to trigger genetic testing can beraised by: Wolff-Parkinson-White pattern in PRKAG2 andLAMP2, or greatly increased precordial voltages and massiveLVH in patients with LAMP2 mutations (36–38). Fabryisease can be suspected by symmetric LVH and late gadolin-

um enhancement in the posterobasal LV (39,40) (Fig. 6).enotype–phenotype relationships/differential diagnosis.positive genetic test in phenotypically expressed HCM is

nly confirmatory, but often useful when morphologic featuresre mild, equivocal, or atypical. Genetic testing can potentiallyesolve ambiguous clinical diagnoses such as HCM versusthlete’s heart or systemic hypertension; however, in thisetting anticipated mutational yield is very low, and a negativeest is indeterminate and does not exclude HCM (26).

Availability of genetic testing has defined and expandedhe HCM clinical spectrum showing a variety of sarcomereutations to be responsible for diverse disease expressions

nd new patient subgroups (Fig. 7). Patterns of LVH varyreatly in closely related individuals (41) with phenotypes asissimilar as apical aneurysms, end-stage, and massive ornusual patterns of hypertrophy, including wall thickeningonfined to the apex, are all part of this heterogeneousenetic disease (3,4,6,31,32,42,43). These observations dis-el the notion that HCM is many similar but unrelatedonditions, and create the unifying principle of a single buteterogeneous disease of the sarcomere.

redicting Prognosis With Mutations

ingle mutations. The early period of HCM genetics washaracterized by substantial optimism and, in retrospect, thenrealistic expectation that the discipline of molecular geneticsould lead to a new paradigm in predicting the outcome of

Proposed Clinical Family Screening Strategies WithEchocardiography or Cardiovascular MagneticResonance (and 12-Lead Electrocardiography) forDetection of HCM With Left Ventricular Hypertrophy*†

Table 4

Proposed Clinical Family Screening Strategies WithEchocardiography or Cardiovascular MagneticResonance (and 12-Lead Electrocardiography) forDetection of HCM With Left Ventricular Hypertrophy*†

Age �12 yrs

Optional unless:

Malignant family history of premature death from HCM, or other adversecomplications

Competitive athlete in an intense training program

Onset of symptoms

Other clinical suspicion of early left ventricular hypertrophy

Age 12–21 yrs‡

Every 12–18 months

Age �21 yrs

Imaging at onset of symptoms, or possibly at 5-year intervals at least throughmidlife; more frequent intervals for imaging are appropriate in familieswith malignant clinical course, or history of late-onset HCM

*In family members who had not undergone genetic testing, or in whom testing was unresolved orindeterminate. †Modified from Maron et al. (27). ‡Age range takes into consideration individualvariability in achieving physical maturity, and in some patients may justify screening at an earlierage; initial evaluation should occur no later than early pubescence.

HCM � hypertrophic cardiomyopathy.

ndividual patients (5,12). Early notions that specific missense

utations are linked to prognosis (5,10,11) have been unreal-zed, in part by initial underappreciation of HCM as anxceedingly complex biological entity with vast genetic heter-geneity (11,15,21). While in the past some HCM genes andariants have been judged to convey more severe diseaseonsequences than others (5,10–12,44,45), there is now con-ensus on the basis of studies in large cohorts of unrelatedatients that routine screening for specific mutations (eachxceedingly uncommon in a genotyped population) cannot beesignated either “benign” or “malignant” or reliably predictlinical outcome (17,18,21,26,46,47) (Fig. 4). Indeed, risktratification with conventional clinical markers (e.g.,istory-taking or echocardiography) has served HCM wells a more rigorous prediction model for identifying high-isk patients who benefit from implantable cardioverter-efibrillator therapy (48).

ultiple mutations. Through large-scale gene sequencing,ata have emerged in HCM suggesting that double (or triple) orompound pathogenic mutations can be associated with moreevere disease expression and adverse prognosis (e.g., advancedeart failure or sudden death, even in the absence of conventionalisk markers [49–51]) (Fig. 8). While it is possible that multipleutations will prove to be prognostic markers or arbitrators of

mbiguous risk profiles, current evidence is preliminary androspective long-term studies in large populations are required.

Figure 5 Prevalence of HCM Genes in Probands

Distribution of genes encoding proteins of cardiac sarcomere in hypertrophiccardiomyopathy (HCM) probands who undergo clinical genetic testing. A varietyof laboratories report a wide range in mutational yield (24% to 63%), leaving asignificant proportion of the HCM population genotype negative. Highest yieldsare in HCM probands with multiple affected family members, while lower ratesare expected in probands without other affected family members (19,20).

712 Maron et al. JACC Vol. 60, No. 8, 2012Genetics in Hypertrophic Cardiomyopathy August 21, 2012:705–15

Consideration for Other Initiatives

Assisted reproduction. Pre-implantation genetic diagnosisallows couples with an increased risk for transmitting a severeor potentially lethal genetic disease the opportunity to conceivea child who will not inherit the pathogenic mutation (52) (Fig. 4).After in vitro fertilization of harvested ova, a single cell isremoved from each embryo (8 cells; 3-day stage) and testedfor the specific mutation; only embryos free of the mutationare transferred into the uterus.Gene therapy. There is no established role for widespreadexperimental genetic engineering in HCM, a disease oftenassociated with normal longevity. Gene therapy is also mostapplicable to recessively inherited defects caused by muta-tions associated with decreased or absent enzyme function.

Genetic Counseling

All HCM patients and relatives should be fully informed byvirtue of some form of genetic counseling. Those whoundergo genetic testing should receive specific pre- andpost-test counseling, including discussion of risks, benefits,and options available (26,53). Certified genetic counselorsplay an important role by collecting detailed family historydata to facilitate cosegregation studies and clarifying thepathogenicity of mutations, as well as discussing family

Figure 6 Genetic Diagnosis of Sarcomeric HCM Phenocopies

(A) LAMP2 (Tyr109Ter mutation, resulting in truncated protein) in 12-year-old boy1,425 g). (B) Histopathology showing marked scarring (stained blue) containing clsions are granular degraded lysosome material (Maron et al. [37], with permissionwith LAMP2 (IVS6-2A-G splice site mutation), showing marked LV cavity dilatation(Maron et al. [38], with permission of Elsevier). (D and E) CMR from 57-year-old wwall (of left ventricle); HCM � hypertrophic cardiomyopathy; LA � left atrium; LV �

planning, and mitigating the psychosocial impact of inher-

iting a potentially deleterious disorder (53). While themultidisciplinary approach has considerable merit in bring-ing together diverse expertise into 1 program (e.g., cardiol-ogist, counselor, nurse), in the United States many HCMpatients are evaluated outside of academic hospital settings,often making this model difficult to follow. As a conse-quence, cardiologists and nursing staffs assume a greater rolein genetic counseling. In addition, all testing vendors haveclinical geneticists available for post-test consultation.

Legal Implications

Some patients are hesitant to contemplate genetic analysisfor a variety of reasons, most often confidentiality concerns.However, in the United States, there is legislative protectionfrom genetic discriminatory practices. Since 2008, GeneticInformation Non-Discrimination Act (GINA) (54) hasbeen a federal law for which most employers and healthinsurance providers are prohibited from denying or termi-nating insurance, employment, or promotion based solelyon a mutation or family history of genetic disease. GINAdoes not protect against discrimination in the military, forlife, disability, or long-term care insurance, or when there isa documented medical condition. GINA is a major advan-

ed suddenly with massive hypertrophy (septal thickness 65 mm, heart weightof distorted myocytes with distinctive vacuolated sarcoplasm (arrows); inclu-American Medical Association). (C) Explanted heart from 24-year-old woman

ll thinning due to diffuse transmural and subepicardial scarring (white areas)with Fabry disease and typical modest/symmetric LVH. Ao � aorta; FW � freeentricle; RV � right ventricle; VS � ventricular septum.

who diustersof the

and waomanleft v

tage in promoting the dissemination of genetic testing.

713JACC Vol. 60, No. 8, 2012 Maron et al.August 21, 2012:705–15 Genetics in Hypertrophic Cardiomyopathy

Promise and Cautionary Notes

The science underlying HCM genotyping has become exceed-

Figure 7 Utility of Genotyping for Defining DiverseHCM Phenotypes Within the Clinical Spectrum

(A) LV apical aneurysms (arrowheads) with mid cavity muscular obstruction(Maron et al. [32], with permission of the American Heart Association); (B) “end-stage” remodeling with enlargement of LV (and atria) and wall thinning, associatedwith systolic dysfunction; (C) massive hypertrophy (wall thickness 34 mm) inanterolateral LV free wall (ALFW) (Maron et al. [42], with permission of Elsevier).(D to F) Morphologic abnormalities in absence of LVH: (D) primary elongation ofanterior mitral leaflet (arrows) (Maron et al. [31], with permission of the AmericanHeart Association); (E) multiple LV myocardial crypts (arrows); (F) late gadoliniumenhancement indicative of replacement myocardial fibrosis (arrows). (G and G1)De novo phenotypic conversion at advanced age: (G) LVH absent at age 46 years;(G1) apical HCM (*) evident at age 51 years (Maron et al. [43], with permission ofElsevier). D � distal cavity; P � proximal cavity; other abbreviations as in Figure 6.

ingly complex, given the interfacing of a highly heterogeneous

genetic disease with an increasing appreciation for the funda-mentally complex nature of normal human genetic variability(i.e., with about 4 million largely benign nucleotide variantsunique to each person). Nevertheless, novel and rapidly evolv-ing strategies for generating genetic data, using next-generation technology, have emerged that will allow exomeand whole-genome sequencing, leading to even greateramounts of genetic information. This opens up the possibilityof defining new genes responsible for HCM (55), but also amultitude of novel variants and VUS for which clinical rele-vance is uncertain.

Therefore, the application of genomics to HCM stands at acrossroads. If genetic testing is to evolve and have a moresubstantial role in the management of patients with HCM,future efforts should focus on clarifying pathogenicity moreprecisely for the substantial number of novel variants presentlyrecognized and those that will inevitably be identified by newmolecular techniques.

Consequently, it is timely to create clinically advantageouscollaborations among the commercial and academic testinglaboratories to promote the exchange of genetic informationand development of standardized mutational classificationschemes that can be more easily translated into patient care.Databases that have been proposed for systematic assembly ofgenetic data from clinical laboratories include MutaDATA-BASE and Clin-Var. Such an initiative would almost certainlyimprove the quality of genetic test reports by reducing ambi-guity in terminology and interpretation.

Current genetic testing reports, which are both the windowto this science and a potential guide to patient care, often seemtoo nuanced for the expectations of clinical cardiovascularmedicine. As a result, clinical decisions are sometimes made inthe context of rapidly evolving but often imprecise science,underscoring the role of clinical correlation and judgment forintegrating diagnostic genetic testing into patient care.

Another transforming question is, given the complex andprobabilistic nature of genetic testing and the imperfect state ofknowledge, how (or by whom) should the results and impli-cations of genetic testing be adjudicated? Should this respon-sibility be confined to those elite (“master”) clinical geneticistsharboring a higher level of expertise? (15). Perhaps, morerealistically, genetic test reports could be modeled with moreclinically relevant language so that patient-specific interpreta-tions do not require special knowledge or teaching.

Conclusions

Closing this communication gap currently separating clini-cians from basic scientists is paramount. To this purpose itwould be important to work toward greater standardizationand less ambiguity in reporting mutations, hopefully en-abling such testing to achieve the confidence of the clinicalcommunity and ultimately its full potential for translatingbasic genetic information into HCM decision making and

patient management.

714 Maron et al. JACC Vol. 60, No. 8, 2012Genetics in Hypertrophic Cardiomyopathy August 21, 2012:705–15

Reprint requests and correspondence: Dr. Barry J. Maron,Hypertrophic Cardiomyopathy Center, Minneapolis Heart Insti-tute Foundation, 920 East 28th Street, Suite 620, Minneapolis,Minnesota 55407. E-mail: hcm.maron@mhif.org.

REFERENCES

1. Braunwald E. Lambrew CT, Rockoff SD, Ross J Jr., Morrow AG. Idiopathichypertrophic subaortic stenosis. I. A description of the disease based on ananalysis of 64 patients. Circulation 1964;30 Suppl 4:3–119.

2. Teare D. Asymmetrical hypertrophy of the heart in young adults. BrHeart J 1958:20:1–8.

3. Maron BJ. Hypertrophic cardiomyopathy: a systematic review. JAMA2002;287:1308–20.

4. Maron BJ, McKenna WJ, Danielson GK, et al. ACC/ESC clinicalexpert consensus document on hypertrophic cardiomyopathy. J AmColl Cardiol 2003;42:1687–713.

5. Seidman CE, Seidman JG. Identifying sarcomere gene mutations in hyper-

C

A 73y/F

Family T

Figure 8 Double Mutations as a Potential Risk Factor in HCM

Asymptomatic proband (arrow) with 2 disease-causing mutations, MYBPC3 and TNventional risk factors. Both parents have HCM, and each contributed 1 mutation twith permission of the Heart Rhythm Society. Solid symbols are those clinically afmutation; � � without mutation; other symbols and abbreviations as in Figure 3.

trophic cardiomyopathy: a personal history. Circ Res 2011;108:743–50.

6. Maron MS, Maron BJ, Harrigan C, et al. Hypertrophic cardiomyop-athy phenotype revisited after 50 years with cardiovascular magneticresonance. J Am Coll Cardiol 2009;54:220–8.

7. Maron BJ, Gardin JM, Flack JM, Gidding SS, Kurosaki TT, Bild DE.Prevalence of hypertrophic cardiomyopathy in a general population ofyoung adults. Circulation 1995;92:785–9.

8. Jarcho JA, McKenna WJ, Pare JA, et al. Mapping a gene for familialhypertrophic cardiomyopathy to chromosome 14q1. N Engl J Med1989;321:1372–8.

9. Geisterfer-Lowrance AA, Kass S, Tanigawa G, et al. A molecular basisfor familial hypertrophic cardiomyopathy: a beta cardiac myosin heavychain gene missense mutation. Cell 1990;62:999–1006.

10. Watkins H, Rosenzweig A, Hwang DS, et al. Characteristicsand prognostic implications of myosin missense mutations in familialhypertrophic cardiomyopathy. N Engl J Med 1992;326:1108–14.

11. Ho C. Genetics and clinical destiny: improving care in hypertrophiccardiomyopathy. Circulation 2010;122:2430–40.

12. Watkins H. Sudden death in hypertrophic cardiomyopathy (editorial).N Engl J Med 2000;342:422–4.

13. Richard P, Charron P, Carrier L, et al. Hypertrophic cardiomyopathy:distribution of disease genes, spectrum of mutations, and implications

37y/M

B 77y/M

VS

MYBPC3 Gln998GluTNNI3 Arg145Trp

experienced resuscitated cardiac arrest (RCA) at age 37 years, but without con-offspring. Electropherograms show identified mutations. From Maron et al. (51),with HCM. N � normal on clinical screening with imaging; � � heterozygote for

N13,o theirfected

for a molecular diagnosis strategy. Circulation 2003;107:2227–32.

715JACC Vol. 60, No. 8, 2012 Maron et al.August 21, 2012:705–15 Genetics in Hypertrophic Cardiomyopathy

14. Henry WL, Clark CE, Epstein SE. Asymmetric septal hypertrophy.Echocardiographic identification of the pathognomic anatomic abnor-mality of IHSS. Circulation 1973:47:225–33.

15. Tester DJ, Ackerman MJ. Genetic testing for potentially lethal, highlytreatable inherited cardiomyopathies/channelopathies in clinical prac-tice. Circulation 2011;123:1021–37.

16. Niimura H, Bachinski LL, Sangwatanaroj S, et al. Mutations in thegene for human cardiac myosin-binding protein C and late-onset familialhypertrophic cardiomyopathy. N Engl J Med 1998;338:1248–57.

17. Ackerman MJ, VanDriest SL, Ommen SR, et al. Prevalence andage-dependence of malignant mutations in the beta-myosin heavychain and troponin T gene in hypertrophic cardiomyopathy. J Am CollCardiol 2002;39:2042–48.

18. Van Driest SL, Ackerman MJ, Ommen SR, et al. Prevalence andseverity of benign mutations in the beta-myosin heavy chain, cardiactroponin T, and alpha-tropomyosin genes in hypertrophic cardiomy-opathy. Circulation 2002;106:3085–90.

19. Erdmann J, Daehmlow S, Wischke S, et al. Mutation spectrum in alarge cohort of unrelated consecutive patients with hypertrophiccardiomyopathy. Clin Genet 2003;64:339–49.

20. Andersen PS, Havndrup O, Hougs L, et al. Diagnostic yield, inter-pretation, and clinical utility of mutation screening of sarcomereencoding genes in Danish hypertrophic cardiomyopathy patients andrelatives. Hum Mutat 2009;30:363–70.

21. Landstrom AP, Ackerman MJ. Mutation type is not clinically useful inpredicting prognosis in hypertrophic cardiomyopathy. Circulation2010;122:2441–9.

22. Richards CS, Bale S, Bellissimo DB, et al. ACMG recommendationsfor standards for interpretation and reporting of sequence variations:Revisions 2007. Genet Med 2008:10:294–300.

23. Mestroni L, Taylor MR. Hearing the noise the challenges of humangenome variation in genetic testing. J Am Coll Cardiol 2011;57:2328–9.

24. Meder B, Haas J, Keller A, et al. Targeted next-generation sequencingfor the molecular genetic diagnostics of cardiomyopathies. Circ Car-diovasc Genet 2011;4:110–22.

25. The 1000 Genomes Project Consortium. A map of human genomevariation from populations—scale sequencing. Nature 2010;467:1061–73.

26. Ackerman MJ, Priori SG, Willems S, et al. HRS/EHRA expertconsensus statement on the state of genetic testing for the channelo-pathies and cardiomyopathies. Heart Rhythm 2011;8:1308–39.

27. Maron BJ, Seidman JG, Seidman CE. Proposal for contemporaryscreening strategies in families with hypertrophic cardiomyopathy.J Am Coll Cardiol 2004;44:2125–32.

28. Maron BJ, Zipes DP. 36th Bethesda Conference: eligibility recom-mendations for competitive athletes with cardiovascular abnormalities.J Am Coll Cardiol 2005;45:1312–75.

29. Maron BJ, Yeates L, Semsarian C. Clinical challenges of genotypepositive (�) – phenotype negative (�) family members in hypertrophiccardiomyopathy. Am J Cardiol 2011;107:604–8.

30. Germans T, Wilde AA, Dijkmans PA, et al. Structural abnormalitiesof the inferoseptal left ventricular wall detected by cardiac magneticresonance imaging in carriers of hypertrophic cardiomyopathies mu-tations. J Am Coll Cardiol 2006;48:2518–23.

31. Maron MS, Olivotto I, Harrigan C, et al. Mitral valve abnormalitiesidentified by cardiovascular magnetic resonance represent a primaryphenotypic expression of hypertrophic cardiomyopathy. Circulation.2011;124:40–7.

32. Maron MS, Finley JJ, Bos JM, et al. Prevalence, clinical significance,and natural history of left ventricular apical aneurysms in hypertrophiccardiomyopathy. Circulation 2008;118:1541–9.

33. Christiaans I, Lekanne dit Deprez RH, van Langen IM, Wilde AA.Ventricular fibrillation in MYH7-related hypertrophic cardiomyopa-thy before onset of ventricular hypertrophy. Heart Rhythm 2009;6:1366–9.

34. Semsarian C, Ahmad I, Giewat M, et al. The L-type calcium channel

inhibitor diltiazem prevents cardiomyopathy in a mouse model. J ClinInvest 2002;109:1013–20.

35. Patel R, Nagueh SF, Tsybouleva N, et al. Simvastatin inducesregression of cardiac hypertrophy and fibrosis and improves cardiacfunction in a transgenic rabbit model of human hypertrophic cardio-myopathy. Circulation 2001;104:317–24.

36. Arad M, Maron BJ, Gorham JM, et al. Glycogen storage diseasespresenting as hypertrophic cardiomyopathy. N Engl J Med 2005;352:362–72.

37. Maron BJ, Roberts WC, Arad M, et al. Clinical outcome andphenotypic expression in LAMP2 cardiomyopathy. JAMA 2009;301:1253–9.

38. Maron BJ, Ho CY, Kitner C, et al. Profound left ventricularremodeling associated with LAMP2 cardiomyopathy. Am J Cardiol2010;106:1194–6.

39. Weidemann F, Niemann M, Breunig F, et al. Long-term effects ofenzyme replacement therapy on Fabry cardiomyopathy. Evidence for abetter outcome with early treatment. Circulation 2009;119:524–29.

40. Patel MR, Cecchi F, Cizmarik M, et al. Cardiovascular events inpatients with Fabry disease. Natural history data from the Fabryregistry. J Am Coll Cardiol 2011;57:1093–9.

41. Ciro E, Nichols PF, Maron BJ. Heterogeneous morphologic expres-sion of genetically transmitted hypertrophic cardiomyopathy: two-dimensional echocardiographic analysis. Circulation 1983;67:1227–33.

42. Maron MS, Lesser RJ, Maron BJ. Management implications ofmassive left ventricular hypertrophy in hypertrophic cardiomyopathysignificantly underestimated by echocardiography but identified bycardiovascular magnetic resonance: implications for management strat-egies. Am J Cardiol 2010;105:1842–3.

43. Maron BJ, Haas TS, Kitner C, Lesser JR. Delayed onset of apicalhypertrophic cardiomyopathy in adulthood. Am J Cardiol 2011;108:1783–7.

44. Saltzman AJ, Mancini-DiNardo D, Li C, et al. Short communication:the cardiac myosin binding protein C Arg502Trp mutation: a commoncause of hypertrophic cardiomyopathy. Circ Res 2010;106:1549–52.

45. Olivotto I, Girolami F, Ackerman MJ, et al. Myofilament protein genemutation screening and outcome of patients with hypertrophic cardio-myopathy. Mayo Clin Proc 2008;83:630–8.

46. Gersh BJ, Maron BJ, Bonow RO, et al. 2011 ACCF/AHA guidelinesfor the diagnosis and treatment of hypertrophic cardiomyopathy. J AmColl Cardiol 2011;58:2703–38.

47. Semsarian C, Yu B, Ryce C, Lawrence C, Washington H, Trent RJ.Sudden cardiac death in familial hypertrophic cardiomyopathy: are”benign” mutations really benign? Pathology 1997;29:305–8.

48. Maron BJ, Spirito P, Shen W-K, et al. Implantable cardioverter-defibrillators and prevention of sudden cardiac death in hypertrophiccardiomyopathy. JAMA 2007;298:405–12.

49. Ingles J, Doolan A, Chiu C. Seidman JG, Seidman CE, Semsarian C.Compound and double mutations in hypertrophic cardiomyopathypatients: implications for genetic testing and counseling. J Med Genet2005;42:e59.

50. Kelly M, Semsarian C. Multiple mutations in genetic cardiovasculardisease: a marker of disease severity? Circ Cardiovasc Genet 2009;2:182–90.

51. Maron BJ, Maron MS, Semsarian C. Double or compound sarcomeremutations associated with sudden death in hypertrophic cardiomyop-athy in the absence of conventional risk factors. Heart Rhythm2012;9:57–63.

52. Basille C, Frydman R, El Aly A, et al. Preimplantation geneticdiagnosis: state of the art. Eur J Obstet Gynecol Reprod Biol2009;145:9–13.

53. Ingles J, Yeates L, Semsarian C. The emerging role of the cardiacgenetic counselor. Heart Rhythm. 2011;8:1958–62.

54. Abiola S. Recent developments in health law. The Genetic Informa-tion Nondiscrimination Act of 2008. J Law Med Ethics 2008;36:856–60.

55. Bagnall RD, Ingles J, Semsarian C. Molecular diagnostics of cardio-myopathies: the future is here. Circ Cardiovasc Genet. 2011;4:103–4.

Key Words: cardiomyopathy y genes y hypertrophy.