Thoracic Aortic Dilatation Syndromes

Background Aortic dilatation syndromes are comprised by a group of different syndromes, of which Marfan syndrome is the best described. Marfan syndrome is a heritable connective tissue disorder associated with mutations in the fibrillin-1 gene (FBN1). The genetic diversity is wide with more than 1000 different mutations in FBN1. Until now it has in many instances been both expensive, difficult and time consuming to get the correct genetic diagnosis; but the use of next generation sequencing (NGS)1;2 has the potential to make diagnosis faster, safer and much cheaper. Mutations in the TGFβR1 and TGFβR2 gene can also give rise to a Marfan syndrome phenotype, but also Loeys-Dietz syndrome and Familial Thoracic Aortic Aneurysm syndrome. Mutations in other genes can lead to a clinical suspicion of Marfan syndrome or related syndromes like Ehler-Danlos syndrome, Weill-Marchesani syndrome as well as others 3. Sporadic occurrence due to de novo mutations are seen in 20-30 % of diagnosed cases of Marfan Syndrome 4;5. New diagnostic revised criteria have been presented in 2010 3.

The frequency of Marfan syndrome has only been studied in a few studies and with very divergent results. In one Chinese study from 1990 a prevalence of 17.2 per 100,000 was found, compared with a study from 1997 with a prevalence of 4.6 per 100,000 6-9. Preliminary runs of the available registries suggest that 1000-2000 individuals with Marfan syndrome exist in Denmark, which is a substantially higher prevalence than in the above mentioned studies.

The pattern of symptoms and diseases in all these syndromes are quite varied, but with typical involvement of the cardiovascular, ocular and skeletal systems. New studies has also shown an increased frequency of migraine, sleep apnea and cholelithiasis in Marfan syndrome and probably with a range of symptoms and traits still not described 10-15.

There is a well known increased mortality among Marfan patients due to cardiovascular causes, because of aortic dilatation and dissection. However, this has only been studied in few and small studies; one American/Scottish study from 1993 found an increasing lifespan from 48 years in 1972 to 72 years in 1993 16-19. Data on lifespan in other thoracic aortic dilatation syndromes is scant. And likewise not much is known of the quality of life and socio-economic conditions of Marfan syndrome and other related syndromes. It is possible that the increased morbidity and mortality related to these syndromes affect socio-economic conditions adversely, as can be speculated in Marfan syndrome 20-23.

The skeletal manifestations were first described by AB Marfan in original description of the syndrome in 1896 24, but now a century later, many facets of the syndrome are still not well characterized, and many studies have been conducted on very selected groups of Marfan syndrome patients, which makes it problematic to extrapolate data.

Denmark has excellent registries; they represent a unique possibility for the study of rare diseases like Marfan and related syndromes concerning morbidity, mortality and socio-economy, as long as the diagnosis is certain. We have previously studied a range of rare syndromes in this manner 25-31.

However, diagnosis of Marfan syndrome and a range of Thoracic Aorta Dilatation syndromes are still problematic and a goal of the present study is to improve not only diagnosis but also the characterization of these syndromes. Likewise, we aim to identify new mutations, especially in frequent conditions like bicuspid aortic valve (0.46-1.37% of the population), which often leads to aortic dilatation and the need for surgical intervention 32. Bicuspid aortic valve syndrome follows an autosomal dominant inheritance with incomplete penetrance and male predominance (3:1), which has pointed towards genes on the X chromosome being involved 33.

We will identify all patients with Marfan syndrome and all other patients with thoracic aortic dilatation syndromes using a multi-facetted approach. Patients will be identified through a combination of all available registries in Denmark (The National Registry of Discharge Diagnoses, The Registry of Causes of Death, The Cancer Registry) and department takings care of these patients (the 2 Centres of Rare Diseases in Copenhagen and Aarhus, and the 3 largest Departments of Cardiology).

In this project we will provide a unique combination of front line molecular technologies and well defined patient cohorts. The aims of the study are the following:

1. To establish a rapid and cheap genetic test for diagnosis of all these syndromes in a combined fashion.
2. To establish the incidence of all these syndromes in Denmark.
3. To establish morbidity, mortality and socio-economic patterns for these syndromes.
4. To examine genotype-phenotype relations. Study population

From preliminary runs in the National Registry of Discharge Diagnoses, we expect to include 500-1000 persons with Marfan syndrome. Some of these have had aortic surgery and tissue specimens are available.

The remaining syndromes are fewer in number and we expect altogether to include about 100-500 persons. Likewise, some of these have had aortic surgery and tissue specimens are available.

For every case (Marfan or any other syndrome) we will identify 100 sex and age-matched controls from Statistics Denmark enabling us to calculate precise estimates of morbidity, mortality and socio-economic measures.

Methods \& Endpoints The NGS technology enables the procurement of large amounts of sequencing data from either the entire or targeted parts of the genome. We will sequence the entire exom and thereafter target specific genes like FBN1, TGFβR1 and TGFβR2 and others (we found more than 40 genes of interest and this number is still rising). The capacity of NGS is so large that we aim to sequence a number of patients in the same run. This will speed up the diagnostic process and reduce costs substantially. Changes found in any one patient will then be compared with available databases of variations in the human genome (SNPdb and HapMap).

The DNA sequence of all examined genes will allow for genotype characterization, which will be correlated with the phenotype of a given individual. Both known and new mutations will expectedly be found, and we expect to find single-base mutations as well as genomic rearrangements (indels).

We will use tissue samples from the aorta or other relevant tissue for correlations between mutational status, the level of messenger ribonucleic acid (mRNA) expression and protein amount, thus, obtaining functional information. In a later project we will study the functional aspects of the genomic aberrations in cell models.

Digital gene expression values will be found using Cufflinks and mutations, SNPs and indels, will be called using the BWA SNP-caller, and copy number variation will be called using CNV-seq. We will perform deep coverage of the DNA and RNA from the same samples for validation of purportedly genomic events such as mutations, SNPs or breakpoints. Bioinformatics will be carried out under supervision by the experts at Department of Molecular Medicine bioinformatics group. We aim at devising new computerized techniques for analysis of sequence data.

Epidemiology The population of patients will be described as mentioned above, and after genetic characterization we will study morbidity, mortality and incidence for the different syndromes. We will link different registries and estimate the use of prescription medicine and we will use the registries at Statistics Denmark and study the socio-economic outcome for the different syndromes.

We will develop tables of expected clinical manifestations in each syndrome group and eventually link these to certain mutations (gene - function relationship), thus being able to predict the outcome of each individual. The tables will include age of onset, likelihood of symptom, etc.

Statistical methods Incidence rates will be calculated as cases per 100.000 per year and analysed with the use of Poisson regression. Morbidity and mortality will be described with the use of Kaplan-Meier curves, log-rank testing and Cox regression. Socio-economic data will be analysed using conditional logistic regression.

Permissions and collaborative partners Permission from the Science Ethical Committee is accepted. There are no specific ethical considerations in relation to the study, which will be conducted in accordance with the Declaration of Helsinki. Permission from the different Registries employed will be sought and Statistics Denmark has accepted a permission. Patients will be included from the outpatient clinics of the 2 centers of Rare Disease in Denmark and the 4 largest Departments of Cardiology. Data will be stored and analysed at the Department of Molecular Medicine, Aarhus University Hospital, Skejby.

Clinical importance The described syndromes impact a high cost on society (>3000 individuals with various clinical manifestation), and if successful we will be able to identify families with the syndromes and prevent diseases, as well as diagnose much earlier leading to better life quality and reduced treatment cost for society. Furthermore, a precise description of the genetic background is likely to lead to a much better treatment of for instance hypertension (angiotensin blocker instead of beta blocker treatment).

Publications The described studies will lead to a range of publications. All involved clinical departments will be mentioned in Acknowledgements in future publications.