Introduction

Dilated cardiomyopathy (DCM) is a severe myocardial disorder characterized by left ventricular (LV) enlargement accompanied by depressed ejection fraction (EF) in the absence of abnormal loading conditions or significant coronary artery disease. Typically diagnosed in young adults, DCM frequently progresses to heart failure (HF), and is frequently associated with arrythmias.^1,2 The etiology of DCM is highly heterogenous and can be broadly categorized into genetic and nongenetic causes, the latter encompassing postinflammatory, toxic, tachyarrhythmic, infiltrative, and storage‑related mechanisms.^1,3 Depending on the population studied, geographic area, and inclusion / exclusion criteria, approximately 30%–40% of DCM cases are attributable to pathogenic or likely pathogenic variants in the genes encoding sarcomeric, Z‑disc, cytoskeleton, nuclear envelope, desmosome, and ion channel proteins.^1,3,4 Currently, genetic testing (GT) in cardiomyopathies is recommended for diagnosis confirmation, management, prognostication, and addressing reproductive issues. However, despite these recommendations, the real‑world implementation of GT remains suboptimal.^1,4 This gap in GT is particularly evident in Poland, where little is known about the genetic background of DCM. Thus, this study provides the first prospective report on genetic testing in a large DCM cohort from southeastern Poland.

Patients and methods

This study was designed as a prospective, single‑center observational analysis. A total of 102 patients with DCM (88 men, mean [SD] age of 45.2 [11.8] years) were enrolled based on published criteria.^1,5 The exclusion criteria encompassed HF due to non‑DCM causes, urgent cardiac or noncardiac admission within the preceding 4 weeks, and contraindications to cardiac magnetic resonance (CMR). Comprehensive assessments were conducted to investigate the etiology of DCM, incorporating anamnesis, family history, and a variety of diagnostic methods. These included laboratory and genetic tests, electrocardiography, 48‑hour Holter electrocardiographic (ECG) monitoring, echocardiography, and CMR. Based on this evaluation, the patients were divided into 4 subgroups of inflammatory, toxic, arrhythmic, and genetic DCM background. In the cases where a specific etiology could not be identified, a term “idiopathic DCM” was applied. All participants were prescribed guideline‑recommended HF therapy. Ethical approval was given by the Krakow Medical Chamber Ethics Committee (7/KBL/OIL/2019), and informed written consent was secured from all participants.

Results

Between May 2019 and September 2020, 24 patients (23.5%) with a genetic etiology of DCM were enrolled. Of these, 21 (87.5%) were men, at a mean (SD) age of 35.5 (7.8) years. A total of 15 novel variants (62.5%) were identified. In all cases, the status of the variants was heterozygous. Fourteen individuals (58.3%) carried mutations in the TTN gene, and to the best of our knowledge, 9 of these (64.3%) had been previously unreported. The most common type of genetic variant identified were missense mutations observed in 9 patients (64.3%), followed by 4 deletions (28.6%), and a splice‑site mutation in 1 person (7.1%) (Table 1).

Among these cases, 2 unrelated young men (aged 28 and 32 years) with no family history of DCM were found to have the same variant (c.62337_62340del) in the TTN gene. Both were classified as New York Heart Association class I, and presented either normal or slightly decreased physical tolerance, achieving 630 and 480 meters, respectively, on the 6‑minute walking test (6MWT). Both patients had significantly enlarged LV end‑diastolic diameter (65 and 75 mm, respectively) and reduced EF of 32.5% and 25%, respectively. CMR imaging showed no late gadolinium enhancement (LGE), indicating an absence of replacement fibrosis; however, the older patient had an apical thrombus. The younger patient had N‑terminal pro–B‑type natriuretic peptide (NT‑proBNP) level at the upper limit of normal (125 pg/ml; upper limit of normal <⁠125 pg/ml), while the older one exhibited a significantly elevated NT‑proBNP level of 3192 pg/ml. Holter ECG revealed paroxysmal atrial fibrillation (AF) in the younger patient, while the older one had ventricular arrhythmias: ventricular extrasystoles (VES) 2500/24‑hour, and runs of nonsustained ventricular tachycardia (nsVT).

Mutations in the FLNC gene were identified in 4 patients (16.7%), 2 of which were missense mutations and 1 was a splice‑site mutation occurring in 2 patients. The new variant in the FLNC gene (c2389+2T>G) was detected in 2 unrelated individuals. Both patients were approximately 40 years old, with very similar phenotypes of DCM: LV end‑diastolic diameter of 68 mm and EF of 18%. Neither had a family history of DCM, and CMR showed no LGE.

Single pathogenic mutations were detected in the sarcomeric genes MYBPC3 and MYH7, while other missense variants were found in the SCN5A, MYPN, BAG3, and DES genes. Additionally, a splice‑site mutation (c.1552‑3C>G) was identified in the GAA gene. Notably, 2 possibly pathogenic variants in DES (c.322G>A and c.635G>C) were observed in 1 patient. Interestingly, no mutation in the LMNA gene was detected.

A mutation in the Bcl‑2–associated athanogene 3 (BAG3) was detected in a 47‑year‑old man with mildly enlarged LV (LV end‑diastolic diameter, 59 mm) and moderately reduced EF (42%). CMR showed no LGE, and 24‑hour Holter identified benign supraventricular arrhythmias. The patient demonstrated good physical activity tolerance (555 m on 6MWT) and normal NT‑proBNP levels. His mother died suddenly at the age of 49.

In a 24‑year‑old man with a strikingly positive family history—2 brothers diagnosed with DCM, 1 of whom died in childhood and the other underwent heart transplantation at the age of 18 years—2 different pathogenic variants in DES were identified. The variant c.322G>A had been previously associated with DCM, while the variant c.635G>C is reported for the first time.

Discussion

To the best of our knowledge, this is the first, relatively large‑scale study investigating the genetic etiology of DCM conducted in southeastern Poland. Consistent with the existing literature, the most common genetic variants were identified within the TTN gene. However, the unexpectedly high number of novel TTN variants detected in our cohort was particularly notable.^1,4,6,8 According to previous studies, the prevalence of TTN gene mutations ranges from 15% to 25% among all genetic cases of DCM, and up to 45% in familial DCM.^4,8,9

The TTN gene, encoding titin, the largest protein in the human body, is an essential sarcomeric component that connects the Z‑disk to the M‑line, providing structural support and playing a critical role in signal transduction.¹⁰ The majority of TTN mutations are truncating variants (TTNtvs), and they account for approximately 80% of all TTN mutations. These truncating variants are associated not only with LV dilatation and low EF but also with a higher risk of arrhythmias.¹¹ In our study, 40% of the identified TTN variants can cause protein truncation; however, the relatively small sample size limits broader generalization. Among our patients with TTNtvs, a significant burden of ventricular arrhythmias was observed, affecting 50% of cases, while 33% experienced new‑onset AF. This aligns with the findings by Barret et al,⁷ who reported that certain TTNtvs are associated with early‑onset AF, typically before the age of 60 years. Additionally, we identified several novel missense variants in TTN that have not been previously reported in the literature. Using a combination of clinical presentation, diagnostic tests, and the analysis of publicly available genetic databases (ClinVar and VarSome platform), we concluded that these variants are likely pathogenic and contribute to development of the DCM phenotype. However, further in‑depth studies are necessary to confirm their pathogenicity.

Interestingly, despite reports indicating that LMNA mutations are the second most common genetic cause of DCM (with a prevalence ranging from 5% to 13%),^4,8,13 no pathogenic LMNA variants were identified in our cohort. This finding is unexpected, given the known association of LMNA variants with conduction disturbances and arrhythmias, both of which were prevalent in our patients. This discrepancy highlights the need for more extensive GT in Polish DCM populations, as their genetic background may differ from that of Western Europe or the United States, where most genetic data on DCM etiology have been generated.

While TTN and LMNA mutations are responsible for a great majority of genetic DCM cases, pathogenic variants in BAG3 are rare, representing less than 0.3% of all cases.^1,4,8 In our cohort, we identified 1 patient with this rare pathogenic variant (c.367C>T). BAG3 mutations were first reported in relation to the autosomal dominant DCM gene in 2011, and at least 16 unique variants including missense, in‑frame, nonsense, and frameshift variants, have been reported in humans. BAG3 is essential for chaperone activity (heat‑shock protein 70) and it regulates numerous biological processes including stabilization of the sarcomere, activation of autophagy, and inhibition of apoptosis.¹⁴ Our variant (c.367C>T), previously described in a family from the United States, contains a premature stop codon in the encoded protein (p.Arg123stop), and is present in patients with more favorable and late onset of cardiac involvement.¹⁵ However, our patient had atrioventricular (AV) conduction abnormalities with an advanced AV block (2:1, 3:1) and ventricular arrhythmia (nsVT and VES), which are also reported in the literature.¹⁵ In contrast, other BAG3 variants, such as c.626C>T (p.Pro209Leu), have been linked to dramatically different phenotypes, including childhood‑onset muscular dystrophy, myofibrillar myopathy, neuropathy, and restrictive or hypertrophic cardiomyopathy.¹⁶

A unique patient in our study was a 24‑year‑old man with a rare double mutation within the DES gene. He carried 2 variants: c.322G>A, classified as a variant of uncertain significance (ClinVar 2021), and c.635G>C identified as a novel, likely pathogenic variant. Mutations in DES, which encodes desmin, lead to desmin‑related myofibrillar myopathy, a heritable condition affecting skeletal and cardiac muscles, with a prevalence of 1%–2% of all genetic DCM cases.^1,4,17 Both variants in our patient were missense mutations that resulted in amino acid changes in the encoded protein. Although DES mutations typically cause right ventricular arrhythmogenic cardiomyopathy, our patient presented with rapidly progressive DCM and severe HF.¹⁸ Ultimately, heart transplantation was necessary for this patient and his younger brother.

Another interesting observation was the presence of simultaneous double mutations in 3 patients, each involving 2 different genes. Two patients had combined mutations in TTN and FLNC, while 1 had mutations in the TTN and MYPN genes. Of note, both variants in TTN (c.404T>A in the first patient and c.7324G>A in the second one) were described as of uncertain significance (ClinVar 2024). Nevertheless, emerging evidence suggests that these TTN variants may indeed have a pathogenic potential.¹⁹ According to the literature, the incomplete penetrance of TTN mutations often requires the presence of additional genetic variants to modulate the disease expression and phenotypic outcomes.¹⁹ This appears to be the case in these patients, where the co‑occurrence of multiple variants may have influenced severity of the disease. Both patients from our cohort presented with DCM and markedly reduced EF, suggesting that the variants of uncertain significance may act as pathogenic in clinically relevant contexts.²⁰ These findings underscore the importance of considering the cumulative impact of multiple genetic alterations when evaluating genotype‑phenotype correlations in DCM.