1. DISEASE CHARACTERISTICS

1.1 Name of the disease (synonyms)

Marfan syndrome type 1 and related phenotypes.

1.2 OMIM# of the disease

154700.

1.3 Name of the analysed genes or DNA/chromosome segments

FBN1.

1.4 OMIM# of the gene(s)

134797.

1.5 Mutational spectrum

Over 1700 different disease-causing mutations have been described (UMD database1; Collod-Béroud et al2; Collod-Beroud G, personal communication).

All types of mutations have been reported. From a study of 1013 probands with a pathogenic FBN1 mutation, the distribution was as follows: 56% missense mutations; 17% frameshift mutations; 14% nonsense mutations; 11% splice mutations; 2% in-frame deletions.3

1.6 Analytical methods

Two different strategies for FBN1 mutation screening procedures are currently applied:

  1. 1)

    direct sequencing of genomic exonic DNA with flanking intronic sequences;

  2. 2)

    or DHPLC or high-resolution melting with confirmation by direct sequencing

When no mutation is identified, a search for FBN1 genomic rearrangements by MLPA or related techniques could be proposed in clinically convincing cases. Indeed, this search seems to increase the FBN1 mutation uptake number by a significant percentage. From a study of 101 patients with Marfan syndrome or related phenotypes but with an absence of FBN1 mutation after direct sequencing, two FBN1 genomic deletions (2%) were found using MLPA.4 Similarly, Liu et al5 identified two FBN1 genomic deletions using RT-PCR out of a series of 60 patients (3.3%), 55 of whom met diagnostic criteria for MFS.

SSCP analysis does not seem to be a satisfying technique for FBN1 mutation screening, as it has been shown that it was less efficient than direct sequencing. Indeed, Loeys et al6 detected 73 sequence variants in 95 patients after screening by SSCP. They identified 13 additional mutations by performing direct sequencing in patients with normal SSCP.

1.7 Analytical validation

Sequencing of both strands was carried out. When a mutation is identified, validation of the results using a second primer set is recommended, +/− using a second technique (PCR with restriction enzyme digestion, high-resolution melting or DHPLC) when possible.

1.8 The estimated frequency of the disease (incidence at birth (‘birth prevalence’) or population prevalence)

There was a population prevalence of approximately 3 out of 10 000.7

1.9 If applicable, prevalence in the ethnic group of the investigated person

Not applicable.

1.10 Diagnostic setting

Comment:

FBN1 mutation screening does not seem to be useful for the positive diagnosis of Marfan syndrome in patients fulfilling international Ghent criteria.8 However, it seems useful in the following situations, to determine whether follow-up and preventive treatment for aortic dilatation is indicated:9, 10

  1. 1)

    patients not fulfilling international Ghent criteria, in particular patients with isolated ectopia lentis and patients with suggestive cardiovascular features combined with skeletal findings, or in sporadic cases of young age11

  2. 2)

    predictive testing in young children (offspring of an affected parents) or relatives (large clinical heterogeneity)12

The decision with regard to searching for an FBN1 gene mutation in such cases will vary and depend on specific family and individual circumstances. The indications of genotyping could be extended to all cases/families in which the proven genetic diagnosis could influence the lifestyle (athletes), the initiation of treatment, or the rate of clinical controls/monitoring.

FBN1 mutation screening can also be indicated in an affected patient with reproductive issues. A prenatal test for Marfan syndrome is rarely requested, but it is expected that the greater availability of mutation testing of the FBN1 gene will increase requests for prenatal diagnosis. Prenatal diagnosis is technically possible by analysis of DNA extracted from foetal cells obtained by chorionic villus sampling at about 10–12 weeks of gestation.13 Prenatal diagnosis is possible when the disease-causing mutation has been identified in the family with careful exclusion of maternal DNA contamination when the mother is the affected parent. In a few cases, when a family can be sampled at large and the disease-causing mutation has not been identified, linkage analysis can be performed. Prenatal diagnosis can be then offered only if conclusive linkage has been obtained and an unambiguous disease-associated haplotype has been identified. A careful analysis of intra- and extragenic FBN1 markers is required.

Prenatal diagnosis can be discussed case by case, with couples requesting it in the framework of a genetic clinic, especially in families with severe cardiac manifestations. Practical use of prenatal diagnosis remains difficult because of the extremely broad variability of clinical expression, even within families, and our inability, at present, to predict the severity of the disease in a given individual. However, it is unlikely that a neonatal MFS will occur in newborns of an adult affected parent. Neonatal MFS cases are always caused by de novo FBN1 mutations.

Alternatively, preimplantation genetic diagnosis (PGD) can be offered for families in which the disease-causing mutation has been identified in an affected family member. However, rules, laws and regulations vary in different European countries, and PGD is illegal in some countries.

2. TEST CHARACTERISTICS

2.1 Analytical sensitivity (proportion of positive tests if the genotype is present)

The proportion is practically 100%.

The possibility of preferential amplification of one allele if primers are localized on an SNP or because of deletion exists, although these events are exceptional.

Classical criteria for determining the pathogenicity of an FBN1 mutation are the following:

  1. 1)

    Nonsense mutation

  2. 2)

    Splice-site mutations affecting canonical splice sequence or shown to alter splicing at mRNA/cDNA level

  3. 3)

    Out-of-frame and in-frame deletion or insertion

  4. 4)

    De novo missense mutation (with proven paternity and absence of disease in parents)

  5. 5)

    Missense mutation previously shown to segregate in a Marfan family

  6. 6)

    Missense replacing/creating cysteine (42% of missense mutations)

  7. 7)

    Missense mutation affecting cbEGF consensus sequence (22% of missense mutations)

  8. 8)

    Missense mutation involving a highly conserved amino acid (6% of missense mutations)

For other missense mutations, the search for segregation in the family should be performed if possible, as well as in the absence of the variant in 400 ethnically matched control chromosomes.

2.2 Analytical specificity (proportion of negative tests if the genotype is not present)

The proportion was practically 100%.

2.3 Clinical sensitivity (proportion of positive tests if the disease is present)

Clinical sensitivity can be dependent on variable factors such as age or family history. In such cases, a general statement should be given, even if a quantification can only be made case by case.

A few studies have addressed the question of clinical sensitivity for FBN1 mutation recognition. Results are variable depending on the method used for mutation screening, but also depending on the set of clinical criteria required for molecular diagnosis. Indeed, a high variable expressivity has been reported in FBN1 mutations and the clinical sensibility is higher when patients fulfilled the Ghent criteria. Results of the more recent studies, including a reasonable number of patients, are as follows:

  1. 1)

    Identification of FBN1 mutations in 86 out of 93 individuals presenting with classic Marfan syndrome, all fulfilling the Ghent criteria (93%), using SSCP and direct sequencing in negative cases6

  2. 2)

    Identification of FBN1 mutations in 74 out of 81 individuals presenting with MFS or Marfan-like phenotypes (91.35%), using DHPLC14

  3. 3)

    Identification of FBN1 mutations in 69 out of 105 individuals with suspected MFS, all fulfilling the Ghent criteria (76%), using direct sequencing15

  4. 4)

    Identification of FBN1 mutations in 90 out of 110 individuals fulfilling the Ghent criteria (82%), in 84 out of 315 individuals with incomplete MFS (27%), in 19 out of 38 individuals with EL (50%) and in none of the 45 individuals with isolated ascending aortic aneurysm using SSCP or DHPLC. The mutation rate was higher with DHPLC. For example, in individuals with classical MFS, the mutation detection rate was 91% using DHPLC vs 75% using SCCP16

  5. 5)

    Identification of FBN1 mutations in 80 out of 85 individuals fulfilling the Ghent criteria (88%) and in 36% of patients with other fibrillinopathies type I using DHPLC17

  6. 6)

    Identification of FBN1 mutations in 193 out of 266 individuals fulfilling the Ghent criteria (72.5%), in 61 out of 105 with incomplete Ghent criteria (58%) and in 3 out of 21 (14.3%) patients referred as possible MFS but with no major diagnostic criterion in any organ system.18

Some explanations can be given accounting for the imperfect clinical sensitivity for FBN1 mutation screening in MFS:

  1. 1)

    Genetic heterogeneity: mutations within the TGFBR1 and TGFBR2 genes have been reported in patients with MFS or suspected MFS.19 Sakai et al found one patient with a TGFBR1 mutation out of a series of 49 patients (2%) and two TGFBR2 mutations (4%);20 Mátyás et al21 reported 10 TGFBR1 or TGFBR2 mutations in 70 unrelated individuals with MFS-like phenotypes who previously tested negative for mutations in FBN1; Singh et al.22 found two TGFBR1 and five TGFBR2 mutations in 41 unrelated patients fulfilling or not fulfilling the diagnostic criteria of Ghent nosology, in whom mutations in the FBN1 coding region were not identified; Stheneur et al23 found six mutations in the TGFBR2 gene and one in the TGFBR1 gene in 105 MFS patients and nine mutations in the TGFBR2 gene and two mutations in the TGFBR1 gene in 247 patients with incomplete or probable MFS who were negative for an FBN1 gene mutation. Screening for TGFBR1/2 should be indicated in the first step when one of the following clinical or imaging features is encountered: hypertelorism, bifid uvula, cleft palate, craniosynostosis, clinical features of vascular Ehlers–Danlos syndrome, arterial tortuosity and aneurysms.

  2. 2)

    Incomplete detection of mutations with the method used: mutations in the 5′ upstream regions24 or intronic mutations.25

2.4 Clinical specificity (proportion of negative tests if the disease is not present)

Clinical specificity can be dependent on variable factors such as age or family history. In such cases, a general statement should be given, even if quantification can only be made case by case.

The proportion is probably 100%, but no data were available for this measure.

2.5 Positive clinical predictive value (lifetime risk of developing the disease if the test is positive)

The possibility is nearly 100%.

Exceptional cases of incomplete penetrance have been reported.26

It is noteworthy that a large number of MFS manifestations are age dependent. A child with an FBN1 mutation can be identified as at-risk but only presents MFS features at a later age.

Although all patients with an FBN1 pathogenic mutation will present a clinical feature at some time during life, it is possible that some patients will not fulfil international criteria for MFS throughout life.

2.6 Negative clinical predictive value (probability not to develop the disease if the test is negative)

Assume an increased risk based on family history for a nonaffected person. Allelic and locus heterogeneity may need to be considered.

When the index case in that family had been tested, the proportion was found to be nearly 100%

When the index case in that family had not been tested, it was concluded that predictive testing for family members should only be proposed when a pathogenic mutation has been identified in an index case.

3. CLINICAL UTILITY

3.1 (Differential) diagnosis: the tested person is clinically affected

(To be answered if in 1.10, ‘A’ was marked).

3.1.1 Can a diagnosis be made other than through a genetic test?

3.1.2 Describe the burden of alternative diagnostic methods to the patient

Cardiological (including echocardiography), orthopaedic (including X-rays) and ophthalmological investigations can altogether establish a diagnosis (but not always).

MRI to diagnose or exclude dural ectasia is occasionally necessary to establish the diagnosis, in patients not fulfilling the international criteria with the previously cited investigations. Dural ectasia is present in many other connective tissue disorders, such as Ehlers Danlos or Loeys–Dietz syndrome, hence this will not, on its own, allow the establishment of a diagnosis.

3.1.3 How is the cost effectiveness of alternative diagnostic methods to be judged?

Unknown.

3.1.4 Will disease management be influenced by the result of a genetic test?