The genetic architecture of pediatric cardiomyopathy

Abstract

To understand the genetic contribution to primary pediatric cardiomyopathy, we performed exome sequencing in a large cohort of 528 children with cardiomyopathy. Using clinical interpretation guidelines and targeting genes implicated in cardiomyopathy, we identified a genetic cause in 32% of affected individuals. Cardiomyopathy sub-phenotypes differed by ancestry, age at diagnosis, and family history. Infants < 1 year were less likely to have a molecular diagnosis (p < 0.001). Using a discovery set of 1,703 candidate genes and informatic tools, we identified rare and damaging variants in 56% of affected individuals. We see an excess burden of damaging variants in affected individuals as compared to two independent control sets, 1000 Genomes Project (p < 0.001) and SPARK parental controls (p < 1 × 10^-16). Cardiomyopathy variant burden remained enriched when stratified by ancestry, variant type, and sub-phenotype, emphasizing the importance of understanding the contribution of these factors to genetic architecture. Enrichment in this discovery candidate gene set suggests multigenic mechanisms underlie sub-phenotype-specific causes and presentations of cardiomyopathy. These results identify important information about the genetic architecture of pediatric cardiomyopathy and support recommendations for clinical genetic testing in children while illustrating differences in genetic architecture by age, ancestry, and sub-phenotype and providing rationale for larger studies to investigate multigenic contributions.

Keywords: ancestry; bioinformatics; clinical interpretation; exome; heart; infant; molecular diagnosis; variant.

Graphical abstract

Figure 1

Diagnostic yield of exome sequencing (A–C) Exome data were filtered for 37 known cardiomyopathy-associated genes, and variants were classified per clinical guidelines to identify pathogenic or likely pathogenic variants, the presence of which was considered a positive result. (A) Overall diagnostic yield. (B) Diagnostic yield by ancestry. (C) Diagnostic yield in infants less than 1 year old (gray bars) and older children (black bars), by subtype of cardiomyopathy. The number of participants in each age group is shown below each bar. DCM, dilated cardiomyopathy; HCM, hypertrophic cardiomyopathy; LVNC, left ventricular noncompaction; PCM, pediatric cardiomyopathy; RCM, restrictive cardiomyopathy.

Figure 2

Damaging variants in the ClinVar genes clustered by phenotype, ancestry, and gene Each row represents an individual participant and each column represents a gene. Variants are color-coded as indicated based by their variant classification. DCM participants are clustered first, followed by those with HCM, LVNC/mixed, non-LVNC mixed, and RCM cases. Participants with damaging variant findings are shown (226/528 individuals). CH, compound heterozygous (cis/trans configuration unknown); DCM, dilated cardiomyopathy; HCM, hypertrophic cardiomyopathy; LoF, loss of function; LVNC, left ventricular noncompaction; RCM, restrictive cardiomyopathy.

Figure 3

Damaging odds ratios in PCM affected individuals and control individuals (A) Odds and 95% confidence interval of having a damaging variant in PCM individuals compared to 1000 Genomes control individuals by gene set and ancestry. (B) Odds and 95% confidence interval of having a damaging variant in PCM individuals compared to secondary control individuals SPARK by gene set and ancestry. (C) Odds and 95% confidence interval of having a damaging variant in PCM individuals of European ancestry compared to PCM individuals of African ancestry by gene set. The adjusted odds ratios take phenotypic differences into account.

Figure 4

Variant burden analysis in cardiac discovery genes shows enrichment in the pediatric cardiomyopathy cohort (A) Variant burden, shown as percentage of samples with damaging variant(s), in curated, ClinVar, intolerant, and combined gene sets by phenotype. The percentage is shown above each bar and the number of individuals is shown in parentheses on the x axis. (B) Variant burden in curated, ClinVar, intolerant, and combined gene sets by ancestry and phenotype. PCM participants are compared to 1000 Genomes (1000G) data. Combined, genes present in curated, ClinVar, missense-intolerant, or LoF-intolerant gene lists. DCM, dilated cardiomyopathy; HCM, hypertrophic cardiomyopathy; LoF, loss of function; LVNC, left ventricular noncompaction; PCM, pediatric cardiomyopathy cohort; RCM, restrictive cardiomyopathy. ^∗Significant difference between PCM and 1000 Genomes (1000G) burden at p < 0.001. p values are PCM versus 1000 Genomes as per Wilcoxon rank-sum analysis.

Figure 5

Proportion of individual-level predicted damaging variants by phenotype, ancestry, and gene list Percentages of one, two, three, or four hits across ancestries (EUR, AMR, and AFR). Combined, genes present in curated, ClinVar, missense-intolerant, or LoF-intolerant gene lists. DCM, dilated cardiomyopathy; HCM, hypertrophic cardiomyopathy; LoF, loss of function; LVNC, left ventricular noncompaction; PCM, pediatric cardiomyopathy cohort; RCM, restrictive cardiomyopathy.

Figure 6

Analysis of cases explained by variants in exomes (A) Percentage of cases explained with ACMG evaluation and informatic predicted damaging variants. (B) Venn diagram of cases explained with ACMG evaluation and informatic predicted damaging variants. Values are number of cases.