Abstract
Congenital heart disease (CHD) is the most common birth defect. Despite considerable advances in care, CHD remains a major contributor to newborn mortality and is associated with substantial morbidities and premature death. Genetic abnormalities are a significant cause of CHD; however, identifying precise defects has proven challenging due to genetic heterogeneity, incomplete penetrance, polygenes, and other nongenetic contributions. With advances in genomic technology and the decreased cost of sequencing, copy number variants and de novo mutations have been increasingly identified as a cause of CHD. The patients with identifiable causes of their CHD are much more likely to have associated extra cardiac manifestations, including neurodevelopmental disorders. Clinical genetic testing is increasingly available and able to diagnose these disorders and will likely allow for increasingly accurate neonatal and prenatal diagnosis to guide care.
Keywords
Congenital heart disease, genetics, birth defect
Chapter Outline
Introduction 221
Monogenic Causes of Specific Cardiac Malformations 223
Atrial Septal Defect 223
Ventricular Septal Defect 223
Bicuspid Aortic Valve 225
Supravalvar Aortic Stenosis 225
Additional Lesions 225
Copy Number Variation 225
De Novo Mutations 227
Common Variants 228
Noncoding Mutations 228
Clinical Genetic Evaluation and Testing 229
Conclusions 229
Acknowledgment 230
References
Further Reading 234
Introduction
Congenital heart disease (CHD) is the most common birth defect, occurring in approximately 1% of newborns . Multiple epidemiological studies suggest there is a genetic contribution to CHD. Twin studies support a genetic contribution; monozygotic twins have a concordance rate of 25% for CHD (although the exact lesion may differ between the twins) as compared to 4.9% in dizygotic twins . The relatively low concordance rate of CHD among identical twins suggests either that mutations occur after splitting of the zygote or that other stochastic developmental processes and/or environmental exposures also play a significant role in many individuals. Moreover, having a first-degree relative with CHD is associated with a relative risk of 3.2 to have CHD . After exclusion of chromosomal defects, the population-associated risk given a positive family history of CHD is 4.2%. Parental consanguinity is associated with a 2–3-fold increased risk of CHD, likely due to the shared recessive variants . Overall, the estimated heritability for CHD is 35% as compared to 85% for schizophrenia . The recurrence risk differs by lesion, and some lesions like atrial septal defect (ASD), ventricular septal defect (VSD), tetralogy of Fallot (TOF), and hypoplastic left heart syndrome (HLHS) have a 3% recurrence risk when one child is affected and the parents are unaffected, while other lesion like Ebstein anomaly have a lower 1% recurrence risk . Left-sided CHD lesions ranging from bicuspid aortic valve (BAV) to HLHS have higher heritability . Known environmental factors, such as maternal diabetes, rubella, phenylketonuria, and certain medications such as thalidomide and indomethacin are thought to account for 2% of isolated CHD cases .
The genetic architecture of CHD is incompletely understood. It has been long known that chromosomal defects and single-gene disorders can cause CHD, often in the context of a multisystem disease. Common examples of aneuploidy with CHD include trisomy 21, associated with atrioventricular canal defects and TOF, and Turner syndrome (45, X) associated with aortic coarctation. Families with Mendelian, nonsyndromic CHD have provided insight into the genes important for cardiac development, including NKX2.5 and GATA4 . These monogenic conditions are highly penetrant and act in an autosomal dominant manner due to haploinsufficiency with sensitivity to gene dosage. Mendelian monogenic disorders associated with CHD and extracardiac manifestations include Noonan and Holt–Oram syndromes. Fewer X-linked and autosomal recessive genes have also been identified and include ciliopathies such as Bardet–Biedl syndrome. For the most part, familial multigenerational CHD is often associated with milder lesions that do not greatly impact survival or reproductive fitness, since previous generations of individuals with severe CHD lesions often did not live long enough in order to transmit these genes. There is significant heterogeneity in CHD with many different genes causing the same type of cardiac lesion and with a single gene causing many different types of CHD. For example, mutations in NOTCH1 can cause a relatively benign lesion like BAV or a severe lesion like HLHS . Some genes like JAG1 are commonly associated with a genetic syndrome that includes CHD, like Alagille syndrome ( Chapter 16 ), but can occasionally also cause nonsyndromic CHD . The most common segmental aneuploidy is the 22q11 microdeletion associated with DiGeorge syndrome, usually arising de novo, and accounting for 34% of truncus arteriosus and 16% of TOF .
The recurrence risk for children of individuals with CHD is substantially increased over the general population risk of 1%. Women with CHD have a 16% risk of recurrence, often with the same heart lesion while the recurrence risk for fathers with CHD is substantially lower . CHD is slightly more common in males than females and explains why the recurrence risk is higher for the offspring of females since recurrence risk for birth defects is generally higher in the less commonly affected gender in whom there is a higher probability of a monogenic or oligogenic basis.
All known genetic causes of CHD are currently estimated to account for 20%–25% of all CHD cases. Some proportion of the remaining 75% of CHD is likely to be multifactorial with polygenic inheritance, in many cases with contributing environmental factors .
Monogenic Causes of Specific Cardiac Malformations
Atrial Septal Defect
ASD is a common cardiac lesion with an incidence of 1:1500 live births . Schott et al. identified the first gene, NKX2.5 , associated with isolated CHD by studying four large families with autosomal dominant secundum ASD ( Table 13.1 ) . These four families also had atrioventricular conduction defects, suggesting that NKX2.5 mutations may alter conduction. Subsequently, additional groups have also reported other cases of ASD with and without conduction defects caused by NKX2.5 mutations . Many mutations cluster in the homeodomain and significantly reduce transcriptional activation, but mutations also occur outside this region . NKX2.5 mutations explain about 1%–4% of sporadic, isolated ASD and up to 12.5% of familial cases .
Gene | CHD Lesion(s) | Associated Problems | Treatment Implications | Frequency | Inheritance |
---|---|---|---|---|---|
CRELD1 | AVSD | None | N/A | 6% of isolated, nonsyndromic AVSD | AD |
ELN | Supravalvar aortic stenosis | None | N/A | AD | |
GATA4 | ASD, VSD, pulmonic stenosis | Can be associated with other structural birth defects | Need to screen for other associated anomalies | Up to 12% of VSD | AD |
GATA6 | VSD | Can be associated with other structural birth defects | Need to screen for other associated anomalies | Rare | AD |
JAG1 | Peripheral pulmonary artery stenosis | Alagille syndrome, cholestasis, posterior embryotoxon | Rare | AD | |
NKX2.5 | ASD, VSD, conotruncal anomalies, left-sided lesions, Ebstein anomaly, and pulmonary atresia | Atrioventricular conduction defects | Control arrhythmias with pacemaker and/or AICD | 1%–4% of ASDs | AD |
NOTCH1 | BAV, HLHS | 4% of BAV | AD | ||
TBX1 | Interrupted aortic arch and VSD | None | N/A | Rare | AD |
TBX20 | ASD, VSD | Rare | AD | ||
TBX5 | ASD | Holt–Oram syndrome, limb anomalies | <1% | AD |
Additional studies have also identified NKX2.5 mutations in other cardiac lesions such as conotruncal anomalies, left-sided lesions, Ebstein anomaly, and pulmonary atresia . Furthermore, even in the absence of CHD, NKX2.5 can cause atrioventricular conduction defects, and these conduction defects can progress to complete heart block and even cause sudden cardiac death .
GATA4 encodes a zinc finger transcription factor with activity modulated by interaction with other proteins such as those in the NKX2 family. Missense and frameshift mutations throughout GATA4 have been identified in secundum ASD without conduction defects . GATA4 mutations are also frequently associated with VSD and pulmonary stenosis among other outflow tract lesions .
T-box transcription factors are expressed in the developing heart and contribute to CHD . TBX5 interacts with NKX2.5 and GATA4 and causes syndromic CHD associated with Holt–Oram syndrome. Missense and nonsense mutations in a related gene, TBX20 , are associated with ASD and are emerging as another cause of CHD . One large family with secundum ASD was found to have a gain of function mutation in TBX20 that enhanced transcriptional activity, opposite to the loss of function mechanism of the majority of TBX20 mutations, demonstrating sensitivity to gene dosage in either direction .
Ventricular Septal Defect
VSD is the most frequent type of CHD, comprising over 20% of all CHD, with an incidence of 6:1000 live births . NKX2.5, GATA4 , and TBX20 mutations are all associated with VSD . Multiple studies have identified missense mutations throughout GATA4 in VSD , and in one study, this gene accounted for approximately 12.5% of isolated VSD cases . Another zinc-finger transcription factor, GATA6 , has also been shown to segregate with autosomal dominantly inherited VSD .
Bicuspid Aortic Valve
In addition to the association with Turner syndrome, BAV is a common anomaly that is often familial . Almost 2% of the population has BAV . NOTCH1 mutations have been associated with BAV in 4% of sporadic cases . NOTCH1 mutations have also been associated with aortic calcification even with an anatomically normal tricuspid aortic valve . Many other studies have also implicated NOTCH1 variants in other left-sided obstructive lesions including HLHS .
Supravalvar Aortic Stenosis
Supravalvar aortic stenosis can be associated with Williams syndrome or can occur in isolation in a familial form that is autosomal dominantly inherited. Missense, nonsense, frameshift, translation initiation, and splice site mutations have been identified in the elastin gene, ELN , in individuals who lack the other phenotypic characteristics associated with William syndrome (a contiguous gene deletion syndrome that encompasses ELN ).
Additional Lesions
The genes previously discussed are also implicated in multiple other lesions such as atrioventricular septal defects (AVSD), conotruncal lesions, and d-transposition of the great arteries (TGA). Although most AVSDs are observed in patients with Down syndrome, 6% of isolated, nonsyndromic AVSD can be caused by missense mutations in CRELD1 , a cell adhesion molecule . Conotruncal lesions are commonly associated with deletion 22q11, and isolated conotruncal lesions can be caused by mutations in NKX2.5 , GATA4 , and NOTCH1 . TBX1 is located within the 22q11 microdeletion interval, and rare variants in TBX1 are associated with nonsyndromic interrupted aortic arch and VSD .
Copy Number Variation
Recent advances in genomic technologies have identified many additional causes of CHD in the last 5 years. Copy Number Variants (CNVs) are segmental aneuploidies, usually of several contiguous genes. They range in size from 1 kb to several Mbs and affect roughly 10% of the human genome . CNVs may be associated with more than one clinical problem, because CNVs often include more than one gene with different genes in the CNVs responsible for different clinical manifestations and/or because single genes within the CNV have pleiotropic effects. For many CNVs associated with CHD, CHD is observed in all individuals with the CNV, and it is still unclear what the additional factors are that determine the presence and type of CHD.
CNVs are typically detected using genome-wide methods of detection including single nucleotide polymorphism microarrays or oligonucleotide arrays. Next generation sequencing is also being used to define CNVs by using sequencing depth of coverage and/or long stretches of homozygosity. The resolution in CNV size has increased over time. There are still some challenges in differentiating pathologic CNVs from benign polymorphic CNVs as we continue to increase the resolution with higher density arrays and/or sequencing data.
Some CNVs are recurrent and at observed at higher frequency in individuals with specific types of CHD, often associated with extracardiac abnormalities. With TOF series, de novo CNVs 5%–10% of patients . Patients with CNVs were more likely to have extracardiac abnormalities. In two studies of left ventricular outflow tract lesions, CNVs were more common among subjects than controls . In heterotaxy, a developmental disorder of the left–right body axis , there were twice as many rare CNVs among the heterotaxy cases compared to the controls (14.5% vs 7.4%, respectively). A study of 2000 patients with various types of CHD including ~800 individuals with TOF found a significant burden of rare deletions >100 kb containing genes. In 538 left-sided lesions, TOF, or heterotaxy a significant increase in CNV burden was observed when comparing CHD trios with healthy trios (OR=3.5–4.6) . Restricting the CNVs to genes previously implicated in CHD, another study of more than 800 CHD cases without known genetic diagnoses, large, rare CNVs (>100 kb for losses and >200 kb for gains) were observed in 4.3% of cases compared to 1.8% of controls .
There is significant phenotypic variability in the type of CHD associated with most CNVs. Several CNVs have been recurrently implicated across studies ( Table 13.2 ). CNVs explain ~10% of CHD cases with ~5% of CNVs being de novo . On average, CNVs associated with isolated CHD have fewer genes and are smaller than CNVs associated with syndromic CHD . Identifying the minimal critical interval within a series of overlapping CNVs allows for the identification of candidate genes for isolated CHD. CNV and sequence data suggest that ETS1 is the pathogenic gene altered by 11q24.2–q25 deletions in Jacobsen syndrome and that CTBP2 is the pathogenic gene in 10q subtelomeric deletions. The 1q21.1 CNV has been observed recurrently in mixed CHD population and in TOF and smaller 100–200 kb duplications including GJA5 are increased in frequency among the TOF cases.