Targeted 46-gene and clinical exome sequencing for mutations causing cardiomyopathies

Abstract

With the implementation of high-throughput sequencing protocols, the exhaustive scanning of known and candidate disease genes has become a feasible approach to genetic testing of patients with cardiomyopathy. A primary objective of the present study was to assess the performance characteristics of a 46-gene next-generation sequencing (NGS) assay that targets well-established cardiomyopathy genes. A total of 25 samples were analyzed. Twelve of those had previously been sequenced using resequencing arrays and served as reference samples for the assessment of the assay's performance characteristics. The remaining 13 samples were derived from consecutive patients. Both the analytical sensitivity and the specificity of the assay were 100% and the percentage of low-coverage bases was 0.4%, at an average read depth of 210×. In order to assess the diagnostic yield of the test, 13 consecutive samples representing cases of Dilated (n = 7), Hypertrophic (n = 4) and Left Ventricular Non-Compaction Cardiomyopathy (n = 2), were subjected to the 46-gene NGS assay. Including predicted pathogenic variants in the gene TTN, a total of 22 variants (11 novel) were detected in 10 patients, with a clear preponderance of variants of unknown pathogenicity (class 3 variants, 21/22, 95%). Of the seven DCM cases, two were digenic, involving variants in the genes MYH7 and RBM20 in one case and in DSP and TTN in the other case. Three other patients carried single TTN variants predicted to be pathogenic. Of the four HCM patients, one was trigenic (LAMA4, PKP2 and TTN) and three were digenic (DSP and TTN, MYH7 and NEXN, NEXN and TTN, respectively). As to LVNC, one of the two patients had one variant in the gene ABCC9 and two predicted pathogenic variants in the gene TTN. Strikingly, out of the thirteen investigated cases, only a single case exhibited a likely pathogenic or pathogenic variant justifying a positive test report. The percentage of inconclusive cases thus amounted to 69%. Three cases were devoid of any relevant variant. Two of these “negative” cases were subsequently taken to initially evaluate the use of an alternative NGS assay addressing 4813 genes previously implicated in genetic diseases (the so-called clinical exome). Although showing similar sensitivity and specificity values, the coverage of the 46 established cardiomyopathy genes was less efficient (low-coverage bases: 5%). In a case of DCM, the assay revealed a disruptive variant in the gene encoding the adrenoreceptor beta 2 (ADRB2), a protein implicated in signal transduction and energy metabolism in the heart. In conclusion, the 46 gene assay is applicable to routine genetic diagnostics of cardiomyopathy. The test detects many variants of unknown pathogenicity which need to be followed-up in order to gain benefit for the patients and their families. Samples devoid of any relevant variant may be subjected to a clinical exome assay, in order to identify interesting novel candidate genes.