New population-based exome data are questioning the pathogenicity of previously cardiomyopathy-associated genetic variants

10 Feb 2013

Next generation sequencing (NGS) has emerged as a powerful technique for the study of genetic disorders. Whole exome studies are currently feasible at a reasonable time and costs. The authors of this very interesting paper analyzed the genetic variants found in all 84 genes which had been associated with cardiomyopathies to date in 6500 individuals from a general population included in the NHLBI-Go Exome Sequencing Project.

A comprehensive search and classification of the previously reported variants thought to be disease causing of Hypertrophic (HCM), Dilated (DCM) or Arrhythmogenic (ARVC) Cardiomyopathy were performed. The prevalence of these variants in the 6500 general population group was calculated. The aim of the study was to identify possible false-positive variants.

Ninety-four (14%) out of 687 variants previously associated with HCM were identified in the 6500 general population cohort, 58 (17%) out of 337 variants associated with DCM, and 38 (18%) out of 209 variants associated with ARVC. These variants were very frequent in the 6500 individuals cohort and corresponded to a genotype prevalence of 1:4 for HCM, 1:6 for DCM, and 1:5 for ARVC, which was several hundreds of times higher than the expected frequency of the disease in a general population (HCM 1:500, DCM 1:2500, and ARVC 1:5000). The authors validated the results by sequencing selected variants found in the cohort (the most frequent ones) in another cohort of 534 patients with similar results.

The pathogenicity of the variants found in the 6500 individuals was tested by means of a commercially available prediction software (PolyPhen-2), and then was compared with the variants not identified in the study cohort. 42% of the variants found in the 6500 exomes were classified as probably damaging versus 52% of the variants not present in the study cohort (p<0.01). On the contrary, a significant proportion of variants in the cohort (42%) were classified as benign versus 18% of non-found variants (p<0.01). Of note, 24% of the variants found in the cohort had some effect demonstrated in functional studies and 13% had co-segregation with the disease in families reported in the literature.

Topic(s):

Myocardial Disease

This is a very important study that should make us to consider that a significant proportion ( 15%) of the assumed causative variants of cardiomyopathies are actually frequent variants in the general population. It could be argued that they might have a role in the disease, although they are not the only cause, as a proportion of them have demonstrated disease co-segregation and there are functional studies to support it.

All identified “cardiomyopathy variants” in the 6500 exomes are listed in the manuscript, together with details of their prevalence and the presence/absence of prior information available on the family co-segregation, functional studies and the results of the evaluation by the prediction software (Polyphen-2). There are some well known mutations in that list that can be recognized. The TNNT2 R278C, for example, had an estimated frequency of 1 in 1000 exomes. R278C was first reported in one of the earliest papers in 1995, and since then labeled as a malignant mutation.

Interestingly, despite the “p values”, most variants either identified or not found in the general population gave similar results on a predictive software, with a remarkable close percentage of probably damaging effect in both groups (42% vs 52%). The authors limit the study to one of the available tests (Polyphen-2) and do not include other softwares such as Pmut, Mutation Taster, etc. Despite the information from these softwares might help in some cases, by themselves does not seem to give us the answer about the pathogenicity of a particular variant. It is not uncommon to have opposite results with different softwares on the same variant. The value of the combination of different prediction tests might be helpful and remains unexplored.

Conclusion:

In conclusion, the data published by the Danish group suggest that an small but significant proportion (15%) of variants currently included in international databases as probably damaging, are likely not monogenic causes of Cardiomyopathy. These variants are well over- represented in the general population.

References

Alexander G. Bick, Jason Flannink, Kaoru Ito, Susan Cheng, Ramachandran S. Vasan et al. Burden of Rare Sarcomere Gene Variants in the Framingham and Jackson Heart Study Cohorts. Am J Hum Genet. 2012 Sep 7;91(3):513-9. (see comments on Sept-2012 Newsletter of the WG, by Tiina Helio: http://www.escardio.org/communities/Working-Groups/cmp/education/papers/Pages/burden-rare-sarcomere.aspx)

Watkins H, McKenna WJ, Thierfelder L, Suk HJ, Anan R, O’Donoghue A et al. Mutations in the genes for cardiac troponin T and alpha-tropomyosin in hypertrophic cardiomyopathy. N Engl J Med. 1995;332:1058-64.

Theopistou A, Anastasakis A, Miliou A, Rigopoulos A, Toutouzas P, Stefanadis C. Clinical features of hypertrophic cardiomyopathy caused by an Arg278Cys missense mutation in the cardiac troponin T gene. Am J Cardiol. 2004;94: 246-9.

Monserrat L, Mazzanti A, Ortiz-Genga M, Barriales-Villa R, García-Giustiniani, Gimeno JR. The interpretation of genetic tests in inherited cardiovascular diseases. Cardiogenetics 2011; Vol 1:e8.

Notes to editor

Presented by : Francisco Ruiz-Espejo, Inmaculada Pérez-Sánchez and Juan Gimeno, Inherited Cardiac Unit, University Hospital Virgen Arrixaca, Murcia, Spain.
Eur J Hum Genet. 2013 Jan 9. doi: 10.1038/ejhg.2012.283. [Epub ahead of print]

The content of this article reflects the personal opinion of the author/s and is not necessarily the official position of the European Society of Cardiology.

Show navigation Hide navigation

New population-based exome data are questioning the pathogenicity of previously cardiomyopathy-associated genetic variants

References

Notes to editor