Cardiovascular disease and sudden cardiac death: between genetics and genomics.

In 1980, polymorphic variance in the human genomewasused to generate a first, physical map allowing monogenic traits to be traced to certain chromosomal regions. Due to the recognition of fragment length polymorphisms (microsatellite markers) with several different alleles at the same chromosomal locus, single nucleotide polymorphisms (SNP) became neglected in this instance for a long time. The obvious interindividual variability and the recognition that many cardiovascular parameters (e.g. blood pressure or ECG parameters) are continuous traits, raised the question of whether evident phenotypic differences in cardiovascular diseases and clinical outcome (e.g. in heart failure, myocardial infarction, or atrial fibrillation) have a common genetic component either due to small size effects or by multiple pathway interaction. To date, many investigations addressing differences in the distribution of SNP alleles in cardiovascular diseases and controls have been triggered by the increasing, but still incomplete knowledge of genomic variance after whole-genome sequencing (3 × 10 nucleotide pairs, among them 1% protein coding) (Figures 1 and 2). Per individual there are (i) around 3.3 million SNP, but within the nearly 20 000 genes only a minority can be found (.10 000 non-synonymous and .10 000 synonymous SNP); and (ii) 250–300 loss-of-function gene variants (including premature stops, splice site disruptions, and frameshifts); some might have been implicated in inherited disorders before and due to a germline mutation rate of 1 × 10 8 up to 30 de novo mutations might be expected in the next generation. Thus, the majority of SNP are located in intergenic regions of the human genome. For a long time, these non-coding sequences have been thought to be a ‘gene desert’ or a ‘no man’s land’, since functional elements, e.g. those affecting gene expression or those involved in gene regulation, were largely unknown before the ENCODE (ENCyclopedia Of DNA Elements) Project was reported. Of note, many structural changes [including short insertionsor deletions (indels) as well as larger copy number variants up to several megabases] in non-coding sequences of the human genome are also known and differ extensively between individuals in a similar polymorphic manner, e.g. as shown for several hundreds of thousands indels. Due to improvedRNA sequencing technologies, nucleotide variation at these functional genomic elements will be further linked to changes in transcript sequence and tissue-specific transcriptional profiles. How can this genome information be used in cardiovascular medicine, in particular for explaining sudden cardiac death (SCD) susceptibility? In industrial countries, SCD is a substantial health issue, and affects worldwide 4–5 million cases per year, thereby 20% of the total mortality. Often, and especially in youths and young adults, there is a clear monogenic component due to the presence of an inherited arrhythmia syndrome or cardiomyopathy. Since these entities are genetically quite heterogeneous (5–25 causative genes per disease) and mutation detection rates (genetic testing sensitivity) range from10–80%,modern technologies suchasnext-generationsequencing (NGS) are being applied more and more. With the exception of hypertrophic cardiomyopathy (HCM), these conditions are considered as rare diseases (prevalence ,1 : 2000) and patient care and genetic testing are recommended to be facilitated in specialized centres. But what about the common and mainly polygenic conditions causing SCD, e.g. coronary artery disease (CAD; linked to 80% of SCD), and where first-degree relatives of an SCD victim still have a higher rate for SCD (after adjustment for risk factors) than individuals with a negative family history? In this issue of the journal, Hernesniemi and co-workers present a comprehensive study in which they developed a weighted genetic (better: genomic) risk score (GRS) for CAD and occurrence of SCD. A key hurdle for the associating SNP alleles with a certain phenotype is that the majority of ‘harmful’ SNP alleles will only have a small phenotypic effect