Jaap Van Waning

Noncompaction cardiomyopathy (NCCM) is a rare disease. NCCM is characterized by hypertrabeculation of myocardium the left ventricle. Recent improvements in cardiac imaging techniques led to an increasing in detection and showed that NCCM occurs less frequently than hypertrophic (HCM) and dilated cardiomyopathy (DCM). The noncompaction phenotype is hypothesized to be a morphologic expression of different pathophysiologic mechanisms in patients with a cardiomyopathy. Interestingly signs of hypertrabeculation may be observed in healthy population without signs of cardiomyopathy, for instance in athletes or in women during pregnancy. In these cases cardiac overload might be related to localized hypertrabeculation of the left ventricle.

In patients diagnosed with NCCM the clinical symptoms range from mild to severe outcomes. Heritability plays an important role in NCCM. In around one third of NCCM patients a genetic defect can be identified. In an additional 15% there is familial disease without a known genetic cause. In around 50% of the NCCM patients there is no evidence for a genetic disease; no mutation in a cardiomyopathy gene or familial disease. These cases may have yet unknown (complex) genetic cause(s) carrying small risk for relatives. Alternatively, NCCM may be attributed to non-genetic, secondary causes for hypertrabeculation.

Finding a genetic cause for NCCM is of importance for patients and their families, enabling to predict genetic risk and perform family cascade screening. This allows accurate identification of relatives which have an increased risk of developing a cardiomyopathy and it also offers the possibility to reassure relatives who do not carry the genetic risk variant. It is important to know if specific causes are related to outcome. In this thesis we investigated if the genotype is of additional value for clinical management. More precisely whether specific genetic defects could predict the clinical features and the outcome (i.e. the risk of severe cardiac complications) in NCCM.

Chapter 1. This thesis starts with a chapter that was published in a monography on NCCM. It presents practical guidelines for the genetic diagnosis of NCCM for clinical geneticists and cardiologists with an overview of the genetic causes for NCCM, a description of the routine of genetic diagnostics, genetic counseling, DNA testing and the current view on family screening for NCCM. The chapter illustrates the importance of integrating genetic diagnostics to clinical management of NCCM patients by conveying appropriate information to patients and their families, in order to make early diagnosis and timely treatment accessible for the families of all NCCM patients.

Chapter 2. This chapter presents the largest genetic NCCM study performed worldwide to date. The retrospective multicenter study was performed in collaboration with the four major cardiogenetic centers in the Netherlands (EMC, UMCG, UMCU and AMC) to establish 1) spectrum of genetic causes for NCCM and 2) correlations between genetics, clinical features, and outcomes, in 327 unrelated children and adults with NCCM. In this study we identified differences in genetic causes between children and adults. Children had more often a genetic cause, had more complex genetic causes with multi-systemic features and severe outcome. In contrast NCCM in adults was mostly linked to missense mutations in sarcomere genes. In about half the cases, mostly diagnosed in adulthood, there was no evidence for a genetic cause. These cases did not have familial disease and DNA testing did not result in the identification of a disease causing variant. The main findings of this study were that NCCM is a heterogeneous condition and genetic stratification has an important role in clinical care. NCCM was due to an inherited cause in half of the cases; in 32% a mutation could be identified and in 15% there was familial cardiomyopathy without a known mutation. Mutations in MYH7, MYBPCP and TTN were the most common (71%). Adults were more likely to have sporadic NCCM (no disease causing variants nor affected family members). The risk of having a reduced left ventricular systolic dysfunction was higher for genetic patients, with highest risk for patients with multiple mutations and TTN mutations. In children, high risk for major cardiac events was related to having a mutation. In adults risk for major cardiac events was related to left ventricular systolic dysfunction in mutation carriers. This was not observed in sporadic cases. Patients with MYH7 mutations had low risk for major adverse cardiovascular events (MACE). Our results showed that DNA diagnostics and family history carry information that may help to distinguish genetic from non-genetic NCCM. Moreover, it complement cardiologic management and prediction of outcome and subsequently guide management of patients with follow-up tailored to genetic status and associated risk. Hence, it showed the importance of genetic testing in NCCM.

Chapter 3. The third chapter of the thesis focused on family screening in NCCM. In this chapter we looked for a way of predicting the severity and cardiac morphology in relatives of NCCM patients. In this chapter we investigated the results of family screening for NCCM in the families that are followed at the cardiogenetic clinic at the Erasmus Medical Center. We focused on familial NCCM and present the outcome of – cardiac examinations and (when appropriate) genetic family testing of the families of 114 NCCM index patients. We were particularly interested if the specific NCCM phenotypes, NCCM with DCM, NCCM with HCM, and isolated NCCM (without HCM or DCM) segregated within families, and if these specific NCCM phenotypes were linked to distinct genetic defects and were related to outcome. Cardiac family screening revealed that 109 (23%) of 472-screened relatives from 58 (51%) families had a cardiomyopathy; 81 relatives from 39 families (of 54 index patients) with a mutation and 28 relatives from 19 families (of 60 index patients) without a mutation. A third of the relatives with an inherited mutation did not have a cardiomyopathy (non-penetrance), which is important information for the counseling of relatives. The relatives who were diagnosed with NCCM were less likely to have left ventricular systolic dysfunction or major adverse cardiac events than the index cases. NCCM with DCM occurred in 56% of all NCCM patients and was associated LV systolic dysfunction, mutations in the tail of MYH7 and 38% was asymptomatic. Isolated NCCM (43%) was associated with milder course, mutations in the head of MYH7, and 42% were asymptomatic. NCCM with HCM was associated with MYBPCP mutations. Familial segregation of distinct NCCM phenotypes was observed in 47% of the affected relatives; 55% isolated NCCM, 34% NCCM with DCM and 33% NCCM and HCM. Thirty-six relatives from 24 families had HCM or DCM without hypertrabeculation. The conclusion was that the distinct NCCM phenotype and related risk for relatives might be predicted by the phenotype and genotype of the index cases. These insights can be applied to personalize clinical management and family screening of NCCM.

Chapter 4. In this chapter we investigated if cardiac magnetic resonance imaging (CMR) could help to distinguish which NCCM patients have a mutation. As the previous studies have indicated, it is important to distinguish genetic NCCM with high risk for relatives, from patients with sporadic NCCM and low risk for relatives. Since NCCM is frequently diagnosed on CMR, we were interested if CMR could help to identify which patients were most likely to have a genetic defect. For this purpose we evaluated 62 NCCM patients referred for CMR who had genetic testing at the Erasmus Medical Center between 2005 and 2017. Thirty-three (53%) patients had a mutation. In all patients previously proposed CMR diagnostic modalities were ascertained (i.e. criteria according to Petersen (short- and long-axis), Stacey, Jacquier, Captur and Choi). Correlation between different CMR criteria varied from moderate to very strong. On average the NCCM patients met four of the different diagnostic modalities. Genetic patients met more diagnostic modalities than patients without a mutation. In multivariate binary logistic regression analysis with CMR and non-CMR parameters, independent positive predictors for a genetic defect were familial cardiomyopathy, trabecular mass, and meeting Petersen criteria in ≥2 out of 3 long axis views, while left bundle branch block and hypertension were negative predictors. The ROC-curve of this multivariate model had a good area under the curve of 0.89. Concluding that CMR criteria and patient characteristics may help to predict if a patient has a genetic defect. In this way CMR interpretations may contribute in guiding cardiologists to refer patients for genetic testing.

Chapter 5. Since NCCM is a relatively rare, genetically heterogeneous cardiomyopathy, there are few studies that are large enough to have sufficient statistical power to establish a correlation between the genetic defects and the clinical features. By performing a systematic review of the NCCM literature for reports of NCCM patients with documented genotype and phenotypes, we could combine and analyze the results of smaller studies to determine the genetic spectrum of NCCM and establish if genetic causes could help predict high-risk profiles for major adverse cardiac events (MACE) in NCCM. For this purpose we have conducted a systematic search of the literature regarding genetics and clinical features of NCCM. The literature search for patients diagnosed with noncompaction cardiomyopathy yielded 1978 papers. After removing duplicates from the different searches 990 reports remained of which 172 fulfilled our inclusion criteria. In this way, clinical and genetic data of 561 patients were included. In total 244 (43%) children were included showing a high prevalence of congenital heart disease and increased risk for MACE compared to adults. Children had more frequently an X-linked or mitochondrial inherited defect or chromosomal anomalies. In adult NCCM patients the main causes were single missense mutations in sarcomere genes. Overall, MYH7 was involved in 48% of the sarcomere gene mutations. MYH7 and ACTCN mutations had lower risk for MACE, than MYBPCP and TTN. In this analysis, the combination of NCCM and DCM was the most frequent observed among the different NCCM phenotypes – as described in the previous chapter, with more than half of the cases showing these overlapping cardiac features. The NCCM-DCM phenotype was associated with an increased risk for MACE and high risk for left ventricular systolic dysfunction. The literature review showed that genetic defects were related to age at presentation, concomitant congenital heart defects and cardiac outcome. These observations endorsed the conclusions from our study presented in Chapter 2 of this thesis, that DNA diagnostics for NCCM are important for clinical management and counseling.

Chapter 6. In nearly half of the families with familial NCCM sequencing of cardiomyopathy gene panels consisting of approximately 50 genes cannot identify a mutation. In the last chapter we present a novel NCCM gene in two families. To identify the genetic cause for familial NCCM whole exome sequencing was performed in two families, one from the University Medical Center Groningen and one from our center and. In these two families a variant in the filamin C gene (FLNC) segregated with the NCCM phenotype. Also in both families there was a patient with coarctation of the aorta and mitral valve abnormalities. Based on these results we present FLNC as a novel NCCM gene. Pathogenic variants in FLNC have been linked previously to HCM and DCM. We expand the spectrum of FLNC related cardiomyopathies with familial segregation of NCCM and concurrent coarctation of the aorta, mitral valve abnormalities.

Conclusions

Distinguishing which NCCM patients and their relatives have a high genetic risk, has important implications for both the patients and their relatives in health management and cardiac surveillance. Genetic patients can be distinguished by using DNA testing, family history and clinical markers. We showed that for prediction of adverse events in patients, information on genotype and phenotyping the left ventricle is important. The distinct cardiac phenotypes: isolated NCCM, NCCM with DCM and NCCM with HCM were related to genotypes, outcome and segregated within families. These studies contributed to our understanding of the etiology of NCCM and in the relation between the cause of the disease and the clinical features. We hope that patients and their families may benefit from these insights and that our study results may contribute to the development of new and improved diagnostic criteria, and guidelines for NCCM.