Contribution of clinically relevant copy number variations in long QT syndrome: a large cohort study

Abstract

Abstract Background Copy number variations (CNVs) have increasingly been recognized in the cardiovascular setting. Their role in hypertrophic cardiomyopathy or sudden cardiac death has been described, but their involvement in Long QT syndrome (LQTS) has been scarcely studied. Most of these descriptions are confined to case reports or small series. Purpose The aim of this study is to investigate the proportion of clinically relevant CNVs among a large cohort of index cases referred for genetic testing due to LQTS. Methods Retrospective study in which the presence of likely pathogenic (LP) or pathogenic (P) CNVs was evaluated in a cohort of index cases referred for genetic study with LQTS phenotype. Index cases were genotyped with customized libraries through next generation sequencing (NGS). CNVs analysis was conducted by a proprietary method by means of comparative analysis of normalized read depth data. Clinically relevant (LP or P) CNVs were then confirmed by other techniques. Results 1183 index cases were referred for genetic testing to our laboratory with long QT syndrome phenotype. Some of these index cases were referred for suspicion of Jervell-Lange-Nielsen syndrome (JNLS) or for LQTS associated with congenital deafness (n=10) or Anderson-Tawil syndrome (n=1). Around 27% (n=322) of the index cases had a positive result. CNVs analysis was run on 1071 (91%) of the index cases. Most of the CNVs were considered negative (n=994, 93%), 59 could not be analysed (6%) and in 11 patients (1%), a LP/P CNV was identified. All CNVs but three (72.72%), were confirmed by orthogonal molecular techniques (Sanger sequencing or multiplex ligation-dependent probe amplification (MLPA)). In all, but two, the genetic study was considered positive due to the CNV findings. One of these patients carried a pathogenic lost-of-function (LOF) CNV in the LDLR gene (incidental finding) but the report was negative because he was referred for suspicion of JLNS. In another case a CNV in the DSP gene was identified in mosaicism (30-40%). However, in this case, the positive report was attributed to the presence of an additional pathogenic variant in CACNA1C. Regarding the other index cases with LP/P CNVs: 4 carried a deletion in the KCNH2 gene, 3 carried a deletion in KCNQ1 gene (one of them in homozygosis and was studied for suspicion of JLNS) and 1 carried a duplication in the KCQN1 (LOF variant). In addition, one patient carried a deletion of exon 3 in the RYR2 gene. Even if the patient was referred for suspicion of LQTS, the report was considered positive. These clinically relevant CNVs represented around 3% of the positive genetic results. Conclusions In our large cohort of patients referred for genetic testing for LQTS, around 1% of clinically relevant CNVs were identified. As such large genomic rearrangements underlie a non-neglectable portion of cases, we consider that their analysis should be performed as part of the routine genetic testing.Deletion exons 6-14 in KCNH2 gene (CNV)MLPA with deletion exons 6-14 in KCNH2