Alpha-protein kinase 3 (ALPK3) truncating variants cause an autosomal dominant form of hypertrophic cardiomyopathy

Background

More than half of hypertrophic cardiomyopathy (HCM) remains genetically unsolved. ALPK3 truncating variants (ALPK3tv) have been described as a cause of autosomal recessive cardiomyopathy in a small number of paediatric cases, but the pathogenicity in heterozygosity as a possible cause of autosomal dominant HCM is unknown.

Aims

To determine the frequency of heterozygous ALPK3tv in patients with HCM and to confirm their pathogenicity by means of burden testing in independent cohorts, family co-segregation studies, and functional analysis of an allelic series of ALPK3tv using human embryonic stem cell-cardiomyocytes (hESC-CM). Phenotype was compared with a cohort of 1679 genotyped HCM patients.

Methods and results

In a discovery cohort of 770 index patients with HCM, 12 (1.56%) were heterozygous for ALPK3tv, odds ratio (OR) 16.01 (95% confidence interval (95% CI): 7.89 to 29.74, p<8.36e-11), compared to the Genome Aggregation Database (gnomAD) population. In a validation cohort of 2047 HCM probands, 32 (1.56%) carried heterozygous ALPK3tv, OR 16.17 (95% CI: 10.31 to 24.87; p<2.2e-16, compared to gnomAD). Combined logarithm of odds score in 7 families with ALPK3tv was 2.99. In comparison with a large cohort of genotyped HCM patients, the phenotype of 51 HCM patients with ALPK3tv (probands and relatives) was characterised by a higher prevalence of apical/concentric patterns of hypertrophy (60%) compared to both sarcomere-positives or negatives (p<0.001 overall), with the age at diagnosis (56±16ys) and maximum wall thickness (18±4mm) similar to sarcomere-negatives and LV systolic impairment at baseline (6%) and non-sustained ventricular tachycardia (31%) similar to sarcomere-positives. Short PR (10%, p=0.009 overall) and extensive fibrosis>15% of LV segments (49%) were distinctive features. During follow-up (5.3±5.7 years), 4 (9%) patients died of heart failure or had cardiac transplantation (p=0.012 vs sarcomere-negatives and p=0.425 vs sarcomere-positives). Analysis of hESC-CM showed that ALPK3 heterozygotes had phenotypic characteristics of HCM, including increased contractile force and delayed membrane repolarization.

Conclusions

Heterozygous ALPK3tv are pathogenic and segregate with a characteristic HCM phenotype.

Funding Acknowledgement

Type of funding sources: Public Institution(s). Main funding source(s): MRC UK, UCLH BRC

Higher responsiveness to rosuvastatin in polygenic versus monogenic hypercholesterolaemia: a propensity score analysis

Background

The underlying monogenic defect in familial hypercholesterolemia (FH) can be detected in ∼40% of cases. The majority of mutation-negative patients have a polygenic cause of high LDL-cholesterol (LDL-C) due to having inherited a greater than average number of common LDL-C raising single nucleotide polymorphisms (SNPs).

Purpose

We sought to investigate, whether the monogenic or polygenic defect in FH is associated with the response to rosuvastatin.

Methods

Individuals with a clinical diagnosis of FH were tested for mutations in LDLR and APOB genes. A previously established LDL-C-specific polygenic risk score (PRS) was used to examine the possibility of polygenic hypercholesterolemia in mutation negative patients. The propensity score analysis was performed to evaluate the variables associated with the response to rosuvastatin. The type of hypercholesterolemia (polygenic or monogenic) and following variables: age, gender, LDL-baseline, statin intolerance, ezetimibe use, rosuvastatin dose, diabetes and cardiovascular disease (CVD), were examined to minimize the bias of this observational study.

Results

LDLR/APOB mutation was found in 47 (42%) patients, whereas polygenic hypercholesterolemia was diagnosed in 65 (58%) of patients. Mean age was comparable in both groups (54±13 vs 51±13, p=0.134). CVD was diagnosed in ≈26% of individuals in both cohorts (p=0.343). There was no difference in the distribution of CV risk factors, such as arterial hypertension, smoking, diabetes, body mass index and in rate of statin intolerance. Monogenic subjects had higher baseline LDL-C compared to polygenic (Table 1). Adjusted model showed a lower percentage of change in LDL-C after rosuvastatin treatment in monogenic vs. polygenic subjects (46% vs 55%, p<0.001) (Figure 1). The probability of achieving LDL-C targets in monogenic FH was lower than in polygenic subjects (0.075 vs. 0.245, p=0.004). Polygenic patients were more likely to achieve LDL-C goals, compared to mutation-positive patients (OR 3.28; 95% CI:1.23–8.72).

Conclusion

Our findings indicate an essentially higher responsiveness to rosuvastatin in patients with a polygenic cause, as compared to those carrying monogenic mutations.

Figure 1

Funding Acknowledgement

Type of funding source: Public grant(s) – EU funding. Main funding source(s): 1. Polish Ministry of Science and Higher Education European Regional Development Fund under the Programme Innovative Economy 2007–2013 (POIG.01.01.02-22-079/09). 2. British Heart Foundation (PG 08/008). 3. The National Institute for Health Research, University College London Hospitals, Biomedical Research Centre. 4. The Foundation Leducq Transatlantic Networks of Excellence Program grant (no. 14 CVD03)

P3686A novel desmin gene variant as an important cause of biventricular arrhythmogenic cardiomyopathy

Introduction

Arrhythmogenic Cardiomyopathy (AC) is typically caused by mutations in the desmosomal genes, however non-desmosomal genes have been increasingly implicated. Desmin gene (DES) mutations have been previously reported in AC, but in many cases there are insufficient data to support their pathogenicity.

Purpose

We assessed our AC cohort for DES gene mutations and describe the clinical phenotype associated with a recurring variant present in 3 unrelated families.

Methods

Genetic testing was performed using next-generation sequencing for 41 genes in a total of 138 AC probands with a definite diagnosis of AC based on the revised 2010 Task Force diagnostic criteria. All candidate variants were confirmed using Sanger sequencing. Clinical and genetic cascade screening were expanded to the first-degree relatives of the probands. Retained tissue from deceased individuals was used for genetic testing. All living mutation carriers underwent clinical assessment including physical examination, 12-lead ECG, signal-averaged ECG, echocardiography, cardiac magnetic resonance imaging (MRI) and 24h Holter-monitoring.

Results

Two DES gene variants, p.Ser298Leu (n=1) and p.Leu115Ile (n=3), were identified in 4 out of the 138 probands (3%). The former coexisted with a pathogenic DSP gene mutation and has not been further evaluated. The latter is a novel variant, absent in control databases (gnomAD) and was the only variant present in 3 unrelated families (see figure). One carrier required heart transplant (A-II-1), two died suddenly (A-III-1, B-II-1) and one died of non-cardiac causes (B-I-2). Detailed clinical information was present in 8 mutation carriers (2 male, age 45±19 years). Seven (88%) had a definite diagnosis and one had a borderline diagnosis of AC. All cases (100%) had right ventricular (RV) wall motion abnormalities, 6 (75%) had a dilated RV, 6 (75%) a dilated LV and 6 (75%) had LV dysfunction (mild in 5 and severe in 1). LV late gadolinium enhancement (LGE) was present in all 6 carriers that had a cardiac MRI with a circumferential sub-epicardial distribution (see figure, case A-III-2). Non-sustained ventricular tachycardia (VT) was present in 7 (88%) and sustained VT in 2 cases (25%). The ventricular ectopic burden per 24h ranged from 426 to 10583 with a median value of 820.

Figure 1

Conclusion

Variants of the DES gene are rare causes of AC. The novel p.Leu115Ile variant seems to be prevalent in a large UK-based cohort and it causes a biventricular form of AC, with a characteristic scar pattern on MRI and severe outcomes.

Acknowledgement/Funding

Alexandros Protonotarios is supported by a BHF Clinical Research Training Fellowship no. FS/18/82/34024

The Genetic Spectrum of Familial Hypercholesterolemia (FH) in the Iranian Population

Familial hypercholesterolemia (FH) is an autosomal dominant disorder associated with premature cardiovascular disease (CVD). Mutations in the LDLR, APOB, and PCSK9 genes are known to cause FH. In this study, we analysed the genetic spectrum of the disease in subjects from the Iranian population with a clinical diagnosis of FH. Samples were collected from 16 children and family members from five different cities of Iran. Probands were screened for mutations in the LDLR, APOB, and PCSK9 genes using next generation sequencing, with results confirmed by Sanger sequencing. The likely pathology of identified variants was examined using in silico tools. Of the probands, 14 had a clinical diagnosis of homozygous FH and two of heterozygous FH. No mutations were found in either APOB or PCSK9, but nine probands were homozygous for seven different LDLR mutations, with p.(Trp577Arg) occurring in three and p.Val806Glyfs*11 occurring in two patients. Two mutations were novel: p.(Leu479Gln) and p.(Glu668*). Seven probands with a clinical diagnosis of FH were mutation negative. This pilot study, integrating clinical and molecular-based techniques, begins to elucidate the FH heterogeneity and the mutation spectrum in the Iranian population. Such information is important for future disease management and cost savings.