Many genes reportedly tied to long QT syndrome have weak evidence
Click Here to Manage Email Alerts
For more than half of the genes thought to be associated with long QT syndrome, there is insufficient evidence to support disease causation, researchers wrote in Circulation.
Genes with little supporting evidence
A clinical domain channelopathy working group found that, of the 17 previously reported genes, nine were classified as having limited or disputed evidence as long QT syndrome-causative genes:
- AKAP9, KCNE2, SCN4B and SNTA1 “lacked evidence of significant segregation of the suspected variant in multiple affected cases beyond chance alone,” according to the researchers.
- For ANK2, the linked region on chromosome 4 encompassed approximately 16 million base pairs, including dozens of genes that were not thoroughly evaluated, and the genetic variant, p.Glu1458Gly, has a population frequency too high to be an autosomal-dominant cause.
- Similarly, KCNJ5 was within a 16 million base pair-linked region, and the reported variant, Gly387Arg, is present in 1 in 200 individuals of East Asian descent.
- CAV3, KCNE1 and KCNJ2 had limited evidence to support an etiologic role, and the genetic evidence of disease causality was based on the candidate gene approach and limited in scope.
Genes with strong evidence
Moreover, genes with conclusive associations with long QT syndrome included:
- KCNQ1, KCNH2 and SCN5A were found to have definitive evidence for causality.
- TRDN, CALM1, CALM2 and CALM3 had strong or definitive evidence supporting an etiology for long QT syndrome with atypical features.
- CACNA1C was found to be associated with Timothy syndrome but had only moderate evidence supporting association with a cardiac-only phenotype related to long QT syndrome.
“Taken as a whole, this contemporary, evidence-based evaluation of reported long QT syndrome disease-genes challenges the classic concept of the genetic landscape in long QT syndrome according to which there are three ‘major’ genes responsible for 75% to 95% of cases and a myriad of ‘minor’ genes that are each responsible for a small fraction of the long QT syndrome patient population,” Arnon Adler, MD, of the department of cardiology at the Peter Munk Cardiac Centre at University Health Network and assistant professor at the University of Toronto, and colleagues wrote. “The current reappraisal portrays a vastly different landscape with only three genes (KCNQ1, KCNH2 and SCN5A) causing typical long QT syndrome and another four genes (CALM1-3 and TRDN) responsible for rare cases of infantile/pediatric long QT syndrome with atypical features.”
After a search using PubMed, three gene curation teams scored the level of evidence for 17 genes reported to cause long QT syndrome by various reports. Then, a clinical domain channelopathy working group conducted the final gene classifications for causation after assessment of the score provided by the gene curation teams. The three teams were masked to each other’s scoring during the assessment.
“A method fraught with significant limitations is the candidate gene approach. In these studies, investigators would sequence the candidate genes in a series of patients with the disease of interest, identify rare genetic changes, and conclude that these genetic changes must be relevant to the disease cause,” Michael H. Gollob, MD, of the division of cardiology and department of medicine at Toronto General Hospital and the department of physiology at the University of Toronto, told Healio. “These studies commonly never asked the same question in a healthy cohort, ie, do rare genetic changes occur in healthy individuals for the same gene? Had this control experiment been done in most previous reports, it would be evident that most genes investigated through a candidate gene approach would not show an excess of rare variants in cases as compared to controls. If healthy controls show the same proportion of rare genetic variants, how could one conclude that the gene is relevant to disease? Unfortunately, this flawed approach has led to many genes being reported as disease causes in the medical literature, including the long QT syndrome genes AKAP9 and SCN4B, for example.”
Takeaway for genetic test providers
“One of the goals of our work is to encourage test providers to be responsible in the genes provided on their panels, and to only include genes with evidence for disease. Should providers choose to provide data on genes lacking evidence for disease causation, they should clearly label such genes as within the realm of research,” Gollob said in an interview. “Test providers that do not make such adjustments in years to come should expect a sustained pressure to do so from all stakeholders, including patient groups, genetic counsellors, geneticists and disease specialists.” – by Scott Buzby
Disclosures: One author reports he is a consultant for Audentes Therapeutics, Boston Scientific, Gilead Sciences, Invitae, Medtronic, MyoKardia and St. Jude Medical and has a potential equity/royalty relationship with AliveCor. The other authors report no relevant financial disclosures.
Perspective
Back to Top
Dan M. Roden, MD
Variants in 17 genes have been implicated as causes of the congenital long QT syndrome. Using criteria and methods developed by the ClinGen consortium, these investigators found strong evidence for causation for only three genes, reasonable evidence in specific clinical situations for five others, evidence for a role in acquired but not congenital long QT syndrome for two and little evidence for seven of 17 genes. The immediate clinical impact of this review should be to confine routine genetic testing for congenital long QT syndrome to the three major genes, KCNQ1, KCNH2 and SCN5A.
The results are not terribly surprising. The way in which we assign pathogenicity has evolved over the last decade, and we now understand that we need stronger evidence than a single case with a variant in a logical gene. One interesting outstanding question is whether the effect size of variants in the two genes associated with acquired long QT syndrome is large enough to merit their use in routine drug prescribing.
Read more about
Click Here to Manage Email Alerts