Abstract
Background
Heart failure (HF) has multiple etiologies, but genetic testing remains mainly recommended for selected subgroups with suspected inherited cardiomyopathy.
Objectives
This study identified and classified genetic variants in patients within a HF cohort with reduced ejection fraction or mildly reduced ejection fraction from diverse etiologies in an admixed population.
Methods
The authors analyzed whole-genome sequencing data from 1,013 unrelated individuals in the GENIUS-HF (Genetic and ElectroNic medIcal records to predict oUtcomeS in Heart Failure patients) cohort. Rare variants in 123 ClinGen-curated cardiovascular genes were classified according to American College of Medical Genetics and Genomics (ACMG) criteria. Local ancestry inference evaluated associations with variant distribution.
Results
Overall, 6.5% of individuals carried at least one pathogenic/likely pathogenic variant considered causal (no significant difference between HF cohort with reduced ejection fraction and HF with mildly reduced ejection fraction, P = 0.500), 77.7% had at least 1 variant of uncertain significance, and 15.8% had only benign/likely benign variants observed. TTN variants accounted for nearly half of the causal findings, including 13 novel variants, with additional causal variants in MYBPC3, FLNC, BAG3, and DSP. Idiopathic dilated cardiomyopathy showed the highest yield (9.9%), but pathogenic variants were also detected in ischemic (4.6%), Chagasic HF (4.9%), and hypertensive (3.8%). Local ancestry analyses showed that variant of uncertain significance were significantly more frequent in regions with African and Native American ancestry.
Conclusions
These findings replicate a significant diagnostic yield of molecular genetic testing in individuals with dilated cardiomyopathy but extend the relevance of genetic testing beyond traditionally selected HF subgroups. They also underscore the importance of including non-European and admixed populations in cardiovascular genomics to reduce interpretation bias and improve equity in precision medicine.
Key words: admixed population, cardiogenetics, heart failure, local ancestry, pathogenic variants, uncertain variants
Central Illustration
Heart failure (HF) is a progressive disorder that is characterized by structural and/or functional impairment of the heart, leading to insufficient cardiac output, which results in high hospitalization and mortality rates.1 In 2023, it affected approximately 64 million people worldwide, including 2 million in Brazil.2 HF can have diverse etiologies, including coronary artery disease, hypertension, Chagas disease, valvopathy, and cardiomyopathies.3,4
The genetic contribution to HF is heterogeneous and complex. Increasing evidence shows that causes previously thought to be purely environmental are, in reality, determined by a complex interplay between genetic and nongenetic causes.5, 6, 7 Previous studies have shown that the genetic diagnostic yield ranges from 3% to 35% across HF cohorts, depending on the cohort design, inclusion criteria, etiology, and ancestral background, reflecting the genetic heterogeneity of HF.8,9 However, most information on the prevalence of pathogenic (P) genetic variants in HF individuals is mostly restricted to cohorts of primary cardiomyopathies, such as hypertrophic and dilated cardiomyopathy (DCM), and it is still unknown what role, if any, genetic testing has in HF from different etiologies.
Moreover, allele frequency and variant effect differ across diverse ancestry groups, as current genomic knowledge has been derived primarily from European individuals, leading to an underrepresentation of admixed individuals in genetic databases and literature.10,11
Here, we conducted a comprehensive analysis to identify and classify rare and potentially P variants in 1,013 admixed patients with HF with reduced ejection fraction or mildly reduced ejection fraction of various etiologies. We focused on genes previously implicated in cardiac disease, and evaluated the frequency of P and uncertain variants across ancestries.
Methods
Study design and patients
This is a retrospective cohort study based on the analysis of genetic and clinical data previously collected from the GENIUS-HF (Genetic and ElectroNic medIcal records to predict oUtcomeS in Heart Failure patients) cohort,12 conducted at the Heart Institute (Instituto do Coraçao do Hospital das Clínicas da Faculdade de Medicina da Universidade de Sao Paulo) and part of the DNA do Brasil study.13
We included 1,013 Brazilian adults, unrelated, between 18-80 years of age, with HF of different etiologies and ejection fraction ≤50%. We excluded patients with impaired cognition, without telephone access, or who refused to participate, as previously published in the GENIUS-HF baseline description.12 All enrolled patients were followed and treated at the Heart Institute (São Paulo, Brazil).
HF etiology was the one considered at enrollment in the protocol. HF etiology was determined based on clinical records and confirmed by 3 board-certified cardiologists. An etiological definition was made after all diagnostic tests deemed initially necessary were available. Chagas disease etiology was confirmed by serological evidence of immunoglobulin G antibodies to Trypanosoma cruzi by immunoenzymatic and indirect immunofluorescence assays. Ischemic HF etiology was defined as a history of myocardial infarction, coronary artery bypass graft surgery, or obstructive coronary artery disease. Patients classified as idiopathic were considered to have DCM of unknown cause. Medication dosage was recorded from the most recent prescriptions at enrollment. Patients signed an informed consent form, and the study was approved by the local ethics committees (Comitê de Ética e Pesquisa HC-FMUSP CAAE #85633324.0.0000.0068).
Whole-genome sequencing, variant calling, and annotation
We extracted DNA from whole blood samples, prepared whole-genome sequencing libraries following the prespecified high-coverage (>35X) protocol for DNA polymerase chain reaction–free by Illumina, and sequenced using the NovaSeq system. After sequencing, as described elsewhere,13 we processed sequence data with Genome Analysis Toolkit (GATK) Docker v4.1.8.0. Per-sample variants were called with HaplotypeCaller followed by Variant Quality Score Recalibration according to GATK Best Practices to obtain a cohort joint Variant Call Format (VCF) file. Variants failing GATK’s PASS filter were excluded. In addition, genotype calls with a read depth <10× were removed from the analyses. Variants were annotated with the Ensembl Variant Effect Predictor14 v113 using the Genome Reference Consortium Human Build 38 (GRCh38). After annotation, variants with intrastudy minor allele frequency <0.01 were retained using bcftools. Only the variants predicted by Variant Effect Predictor as having high and moderate impact on the transcript were retained (ie, splice sites, stop gained, transcript ablation, frameshift, stop lost, start lost, transcript amplification, inframe insertion, inframe deletion, protein altering, and missense variants).14
Gene selection and rare variant analysis
Cardiovascular genes were selected according to different ClinGen Cardiovascular Curation Expert Panels.15, 16, 17, 18, 19, 20 We selected only the genes presented in cardiovascular panels with supportive evidence, defined as “limited,” “moderate,” “strong,” and “definitive,” excluding the genes defined as “no evidence” or “contradictory evidence.” The list of selected genes (123 in total) is provided in Supplemental Table 1.
All variants identified were interpreted and classified according to the ACMG (American College of Medical Genetics and Genomics)21 following the ACGS (Association for Clinical Genomic Science)22 specifications and ClinGen gene-specific Variant Curation Expert Panel recommendations, when available. General population data were obtained using the total allele frequencies from the gnomAD database (version 4.1), which contains data from different populations, and from ABraOM (version 2.1), a repository containing population data only from Brazilian elderly individuals.23 In silico prediction for missense variants was assessed using the REVEL (Rare Exome Variant Ensemble Learner) meta-predictor.24 We applied the PP3 (supporting evidence for pathogenicity) criteria for a REVEL score of ≥0.64 and the BP4 (supporting evidence for benign effect) for a REVEL score of ≤0.29, as recommended by ClinGen.25
After evaluating each criterion from the American College of Medical Genetics and Genomics guideline, the variants were classified as benign (B), likely benign (LB), uncertain (variant of uncertain significance [VUS]), likely pathogenic (LP), and P. A second analyst independently reviewed variant classifications to ensure consistency and accuracy.
We defined LP/P findings as causal when they occurred in genes with definitive, strong, and moderate evidence levels and curated as autosomal dominant (AD) or x-linked by ClinGen, or when biallelic in autosomal recessive (AR) genes. Heterozygous LP/P variants in genes with AR evidence were treated as carrier findings and LP/P variants in genes classified as limited by ClinGen were considered secondary findings, and, therefore, were excluded from the diagnostic yield.
Novel variants were defined as absent from ClinVar, gnomAD v4.1, ABraOM v.2.1, and dbSNP at the time of analysis (December 2025).
Local ancestry inference
We used genomic ancestry information to evaluate its influence on HF etiology and the distribution of genetic variants. As previously described,13 we inferred local ancestry using GNOMIX, a method that infers local ancestry by training a supervised machine learning model on reference panels to classify genomic segments along phased haplotypes, leveraging linkage patterns and neighboring markers to assign ancestry at a fine scale across the genome.26
Local ancestry was derived from the GNOMIX output file (.msp), which encodes ancestry assignments along each phased haplotype across the genome. To determine local ancestry at the variant level, we integrated a phased VCF file, generated using Shapeit5,27 with corresponding ancestry segments by mapping each variant position to its respective genomic interval in the msp file. This allowed assignment of ancestry to each allele at each variant site.
After the variant classification, we evaluated the distribution of local ancestry across LP/P and VUS variants. We calculated the proportion of ancestry-specific segments overlapping the variants and compared observed vs expected ancestry frequencies across variant classes (the observed proportion corresponds to the ancestry distribution among variant-containing segments, and the expected proportion reflects the ancestry distribution across all genomic segments).
Statistical analysis
Categorical variables were described as frequencies and compared using the chi-square test. Quantitative variables were described using the median (Q1-Q3), and comparisons were made using the Kruskal-Wallis rank-sum test. CIs for diagnostic yield estimates were calculated using the exact binomial method.
We set a statistically significant threshold at P < 0.05 for these analyses. All the analyses were performed using R, version 4.5.1.
Results
Individuals
We performed a whole-genome analysis in 1,013 Brazilian adults with HF (Table 1), who were categorized based on genetic findings. The median age was 57 years, and 35% of individuals were female. Median genomic local ancestry proportions revealed a predominant European component (62.3%), followed by African ancestry (23.8%), Native American (9.4%), and East Asian (0.4%). Regarding cardiac function, the median ejection fraction was 30%, and most patients were classified as having idiopathic (considered here as DCM) HF etiology (29.9%), followed by ischemic (21.2%), hypertensive (18.1%), and Chagasic (16.1%). Ancestry analysis demonstrated that, although European ancestry was predominant across all etiology groups, its proportion was higher in ischemic and idiopathic HF and lower in alcoholic and hypertensive HF, which showed relatively higher African ancestry proportion (Supplemental Table 5).
Table 1.
Individuals' Clinical and Demographic Characteristics
| All Individuals (N = 1,013) | Causal Variant (n = 66) | Uncertain (n = 787) | No Evidence for Causal Variant (n = 160) | P Valuea | |
|---|---|---|---|---|---|
| Age, y | 57 (48-65) | 52 (43-61) | 57 (48-65) | 57 (50-66) | 0.010 |
| Female | 356 (35) | 25 (38) | 276 (35) | 55 (34) | 0.9 |
| HF classification | 0.5 | ||||
| HFmrEF | 178 (18) | 9 (14) | 137 (17) | 32 (20) | |
| HFrEF | 834 (82) | 57 (86) | 650 (83) | 127 (80) | |
| LVEF, % | 30 (25-40) | 30 (25-35) | 31 (25-39) | 33 (28-40) | 0.2 |
| Local ancestry proportions, % | |||||
| European | 62.2 (47.6-77.2) | 63.2 (51.9-74.8) | 61.3 (46.6-77.2) | 66.4 (53.7-79.2) | 0.061 |
| African | 23.7 (11-39.5) | 22.0 (13.6-36.6) | 24.9 (11.1-40.9) | 21.5 (9.4-37.3) | 0.15 |
| Native American | 9.4 (5.8-12.9) | 9.0 (6.2-13.4) | 9.2 (5.6-12.7) | 9.7 (6.6-13.2) | 0.6 |
| East Asian | 0.4 (0-0.9) | 0.5 (0.04-1.0) | 0.4 (0.0-0.9) | 0.4 (0.0-0.8) | 0.6 |
| Obesity | 0.2 | ||||
| No | 591 (59) | 39 (62) | 469 (60) | 83 (53) | |
| Grade I | 215 (22) | 11 (17) | 171 (22) | 33 (21) | |
| Grade II | 134 (13) | 7 (11) | 98 (13) | 29 (18) | |
| Grade III | 57 (5.7) | 6 (9.5) | 39 (5) | 12 (7.6) | |
| BMI, kg/m2 | 26.8 (23.7-30.4) | 26.5 (24.3-30.2) | 26.7 (23.6-30.1) | 27.4 (24.4-32.3) | 0.040 |
| Systolic BP, mm Hg | 120 (107-140) | 118 (103-137) | 120 (106-140) | 122 (110-140) | 0.2 |
| Diastolic BP, mm Hg | 74 (65-85) | 70 (62-86) | 75 (65-84) | 75 (66-88) | 0.6 |
| Dyslipidemia | 0.15 | ||||
| Yes | 616 (62) | 32 (51) | 481 (62) | 103 (65) | |
| Arrhythmias | 0.2 | ||||
| No | 845 (84) | 52 (80) | 660 (85) | 133 (84) | |
| Atrial fibrillation | 140 (14) | 11 (17) | 109 (14) | 20 (13) | |
| Flutter | 9 (0.9) | 2 (3.1) | 5 (0.6) | 2 (1.3) | |
| VT/NSVT | 8 (0.8) | 0 (0) | 5 (0.6) | 3 (1.9) | |
| Myocardial infarction | 0.090 | ||||
| Yes | 257 (25) | 10 (15) | 201 (26) | 46 (29) | |
| HF etiology | 0.030 | ||||
| Idiopathic | 303 (29.9) | 30 (45) | 235 (30) | 38 (23.8) | |
| Ischemic | 215 (21.2) | 10 (15) | 171 (22) | 34 (21.2) | |
| Hypertensive | 183 (18.1) | 7 (11) | 142 (18) | 34 (21.3) | |
| Chagasic | 163 (16.1) | 8 (12) | 130 (17) | 25 (15.6) | |
| Alcoholic | 87 (8.6) | 3 (4.5) | 68 (8.6) | 16 (10) | |
| Other | 62 (6.1) | 8 (12) | 41 (5.2) | 13 (8.1) | |
| Cardiac disease family historyb | 0.034 | ||||
| Yes | 564 (56) | 41 (62) | 421 (54) | 102 (64) |
Values are median (Q1-Q3) and tested with the Kruskal-Wallis rank-sum test or n (%) and compared using Pearson chi-squared test, unless otherwise indicated. Bold indicates statistically significant differences (P < 0.05). A total of 1,013 Brazilians with heart failure were part of the study.
BMI = body mass index; BP = blood pressure; HF = heart failure; HFmrEF = heart failure with mildly reduced ejection fraction; HFrEF = heart failure with reduced ejection fraction; LVEF = left ventricular ejection fraction; NSVT = nonsustained ventricular tachycardia; VT = ventricular tachycardia.
P values correspond to the statistical test comparing the groups (Causal Variant, Uncertain, and No Evidence for Causal Variant).
Family history: myocardial infarction, HF, sudden cardiac death, or other cardiovascular causes in first-degree relatives.
Rare variant analysis
After excluding low-quality variants, we assessed and classified 3,286 single-nucleotide variants. Most variants were classified as LB or B (54%), or as a VUS (44%), with only 2% classified as LP or P. A summary of variant classification is shown in Supplemental Table 2, and the complete list of variants and their final classifications is provided in a Supplemental data file.
Among the 1,013 individuals, 76 carried at least one LP/P variant. Of these, 3 harbored heterozygous variants in genes with an AR inheritance pattern and were therefore considered carriers only. Five individuals carried LP/P variants in genes with limited evidence (OBSCN and TGFB3), and 2 carried LP/P variants in a gene primarily associated with long-QT syndrome (KCNQ1). These findings were considered secondary and were not included in the analysis (Supplemental Table 6). The remaining 66 individuals (6.51%; 95% CI: 5.1%-8.2%) met the criteria for causal LP/P findings in genes with definitive disease association and with an AD inheritance pattern. One of these individuals carried 2 distinct variants, yielding 67 causal LP/P occurrences in the cohort. Among the HF etiologies, the highest diagnostic yield was observed in idiopathic HF (9.9%; 95% CI: 6.8%-13.8%), followed by Chagasic (4.9%; 95% CI: 2.1%-9.4%), ischemic (4.7%; 95% CI: 2.3%-8.4%), hypertensive (3.8%; 95% CI: 1.6%-7.7%), and alcoholic HF (3.4%; 95% CI: 0.7%-9.7%) (Supplemental Table 10).
In total, these occurrences corresponded to 52 distinct causal variants, summarized in Table 2. Of the 52 distinct causal variants, 28 were identified in TTN, of which nearly half were novel (not present in ClinVar, gnomAD, ABraOM, or dbSNP) (Figure 1). All causal variants in TTN were truncating variants located in exons with a percentage spliced in (PSI) of 100%, according to the Cardiodb.28 Two additional truncating TTN variants were identified; however, because they were located in exons with PSI values of 62% and 75% they were classified as VUS (Supplemental Table 3). Additional novel causal variants were also identified in other genes, such as BAG3, DSP, FLNC, and MYBPC3.
Table 2.
Distinct Causal Variants (Pathogenic or Likely Pathogenic in Genes With Definitive Evidence-Level and With AD Inheritance) Identified
| Gene | ClinGen Phenotype | ClinGen Evidence Level | Novel | Reference SNP ID (rsID) | Chromosome | Genomic Position (GRCh38) | Nucleotide Change | AA Change | gnomAD Total AF | ABraOM Total AF | ClinVar | ACMG Classification | N |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| BAG3 | DCM | Definitive | Yes | — | 10 | 119669894 | c.228_237del | p.Pro77Glyfs∗131 | Absent | Absent | Absent | LP | 1 |
| BAG3 | DCM | Definitive | No | rs387906875 | 10 | 119670037 | c.367C>T | p.Arg123∗ | 0.0003% | Absent | LP/P | P | 1 |
| BAG3 | DCM | Definitive | Yes | — | 10 | 119676863 | c.1309C>T | p.Gln437∗ | Absent | Absent | Absent | LP | 1 |
| DES | DCM | Definitive | No | — | 2 | 219419041 | c.578+1G>A | — | 0.0002% | Absent | Absent | P | 1 |
| DSC2 | ARVC | Definitive | No | rs2510951125 | 18 | 31086588 | c.929delinsTT | p.Gln310Hisfs∗8 | Absent | Absent | LP | LP | 1 |
| DSG2 | ARVC | Definitive | No | rs1187924885 | 18 | 31524752 | c.882dup | p.Val295Serfs∗6 | 0.0003% | Absent | LP/P | P | 1 |
| DSP | ARVC | Definitive | Yes | — | 6 | 7574655 | c.2298-2A>G | — | Absent | Absent | Absent | LP | 1 |
| FLNC | DCM | Definitive | Yes | — | 7 | 128848961 | c.4908del | p.Ser1637Alafs∗29 | Absent | Absent | Absent | LP | 1 |
| FLNC | DCM | Definitive | No | — | 7 | 128850484 | c.5398+1G>A | — | 0.0000006% | Absent | Absent | LP | 1 |
| FLNC | DCM | Definitive | No | rs1808781579 | 7 | 128850924 | c.5520T>A | p.Tyr1840∗ | 0.0000006% | Absent | P | P | 1 |
| FLNC | DCM | Definitive | No | rs748416758 | 7 | 128854661 | c.6976C>T | p.Arg2326∗ | 0.0001% | Absent | P | P | 1 |
| MYBPC3 | HCM | Definitive | No | rs727504287 | 11 | 47342686 | c.1513_1515del | p.Lys505del | 0.0009% | Absent | LP | LP | 3 |
| MYBPC3 | HCM | Definitive | No | rs200411226 | 11 | 47342718 | c.1484G>A | p.Arg495Gln | 0.0024% | Absent | LP/P | P | 1 |
| MYBPC3 | HCM | Definitive | Yes | — | 11 | 47343490 | c.1223+2T>A | — | Absent | Absent | Absent | P | 1 |
| PKP2 | ARVC | Definitive | No | rs78897684 | 12 | 32824163 | c.1557-1G>C | — | 0.0005% | Absent | P | P | 1 |
| PLN | INTCM | Definitive | No | rs754782171 | 6 | 118558947 | c.26G>A | p.Arg9His | 0.0007% | Absent | Conflicting | LP | 1 |
| RBM20 | DCM | Definitive | No | rs796734066 | 10 | 110812297 | c.1900C>T | p.Arg634Trp | Absent | Absent | P | P | 1 |
| SCN5A | DCM; BrS; LQTS | Definitive | No | rs199473605 | 3 | 38560374 | c.4015G>A | p.Val1339Ile | 0.0023% | Absent | Conflicting | LP | 1 |
| SCN5A | DCM; BrS; LQTS | Definitive | No | rs199473153 | 3 | 38597737 | c.2254G>A | p.Gly752Arg | 0.0004% | Absent | LP/P | P | 1 |
| SCN5A | DCM; BrS; LQTS | Definitive | No | rs199473556 | 3 | 38630342 | c.361C>T | p.Arg121Trp | 0.0001% | Absent | LP/P | LP | 1 |
| TNNT2 | DCM | Definitive | No | rs397516456 | 1 | 201365298 | c.304C>T | p.Arg102Trp | 0.0005% | Absent | P | P | 1 |
| TPM1 | HCM | Definitive | No | rs730881151 | 15 | 63042891 | c.62G>T | p.Arg21Leu | 0.0009% | Absent | Conflicting | P | 1 |
| TTN | DCM | Definitive | Yes | — | 2 | 178560461 | c.85670del | p.Leu28557Trpfs∗6 | Absent | Absent | Absent | LP | 1 |
| TTN | DCM | Definitive | Yes | — | 2 | 178533818 | c.102795_102796dup | p.Asn34266Ilefs∗7 | Absent | Absent | Absent | LP | 1 |
| TTN | DCM | Definitive | No | rs752697861 | 2 | 178534092 | c.102523C>T | p.Arg34175∗ | 0.0004% | Absent | LP/P | P | 1 |
| TTN | DCM | Definitive | No | rs2468686204 | 2 | 178537825 | c.99381del | p.Arg33127Serfs∗34 | Absent | Absent | LP | LP | 1 |
| TTN | DCM | Definitive | No | — | 2 | 178548444 | c.93181del | p.Arg31061Valfs∗33 | Absent | Absent | LP | LP | 4 |
| TTN | DCM | Definitive | No | — | 2 | 178552137 | c.90762del | p.Gly30255Glufs∗35 | Absent | Absent | LP | LP | 1 |
| TTN | DCM | Definitive | Yes | — | 2 | 178552622 | c.90277del | p.Lys30094Argfs∗18 | Absent | Absent | Absent | LP | 1 |
| TTN | DCM | Definitive | Yes | — | 2 | 178553152 | c.89725_89747del | p.Lys29909Trpfs∗4 | Absent | Absent | Absent | LP | 1 |
| TTN | DCM | Definitive | No | rs762411631 | 2 | 178553933 | c.89178C>G | p.Tyr29726∗ | Absent | Absent | LP | LP | 3 |
| TTN | DCM | Definitive | No | — | 2 | 178557326 | c.87931_87935dup | p.Tyr29313Glyfs∗90 | Absent | Absent | LP | LP | 1 |
| TTN | DCM | Definitive | No | rs1060500525 | 2 | 178559797 | c.86335C>T | p.Arg28779∗ | 0.0001% | Absent | LP/P | LP | 1 |
| TTN | DCM | Definitive | No | rs756499458 | 2 | 178560639 | c.85493G>A | p.Trp28498∗ | 0.0001% | 0.0427% | LP | LP | 1 |
| TTN | DCM | Definitive | No | rs770038577 | 2 | 178561042 | c.85090C>T | p.Arg28364∗ | 0.0001% | Absent | LP/P | P | 1 |
| TTN | DCM | Definitive | Yes | — | 2 | 178566243 | c.79888del | p.Glu26630Lysfs∗14 | Absent | Absent | Absent | LP | 1 |
| TTN | DCM | Definitive | Yes | — | 2 | 178569070 | c.77062C>T | p.Gln25688∗ | Absent | Absent | Absent | LP | 1 |
| TTN | DCM | Definitive | No | rs794729281 | 2 | 178577085 | c.69250C>T | p.Arg23084∗ | 0.0000006% | Absent | LP/P | LP | 1 |
| TTN | DCM | Definitive | Yes | — | 2 | 178580535 | c.66843dup | p.Tyr22282Ilefs∗16 | Absent | Absent | Absent | LP | 1 |
| TTN | DCM | Definitive | Yes | — | 2 | 178588565 | c.63160A>T | p.Arg21054∗ | Absent | Absent | Absent | LP | 1 |
| TTN | DCM | Definitive | Yes | — | 2 | 178588606 | c.63118del | p.Ala21040Glnfs∗23 | Absent | Absent | Absent | LP | 1 |
| TTN | DCM | Definitive | No | rs769912484 | 2 | 178599063 | c.56648-1G>A | — | 0.0001% | Absent | Conflicting | LP | 1 |
| TTN | DCM | Definitive | No | rs1343120755 | 2 | 178609301 | c.52009C>T | p.Arg17337∗ | 0.0001% | Absent | LP/P | P | 1 |
| TTN | DCM | Definitive | No | rs2057454001 | 2 | 178617003 | c.47885dup | p.Met15963Asnfs∗7 | Absent | Absent | LP | LP | 1 |
| TTN | DCM | Definitive | No | rs2060187635 | 2 | 178634576 | c.42205C>T | p.Arg14069∗ | Absent | Absent | LP | LP | 1 |
| TTN | DCM | Definitive | No | — | 2 | 178635462 | c.41861dup | p.Tyr13954∗ | 0.0000006% | Absent | Absent | LP | 1 |
| TTN | DCM | Definitive | Yes | — | 2 | 178738353 | c.14099dup | p.Val4701Serfs∗22 | Absent | Absent | Absent | LP | 1 |
| TTN | DCM | Definitive | Yes | — | 2 | 178741889 | c.11344G>T | p.Glu3782∗ | Absent | Absent | Absent | LP | 1 |
| TTN | DCM | Definitive | Yes | — | 2 | 178768880 | c.8955del | p.Ile2986Leufs∗5 | Absent | Absent | Absent | LP | 1 |
| TTN | DCM | Definitive | Yes | — | 2 | 178777811 | c.4371_4372del | p.Val1458Leufs∗19 | Absent | Absent | Absent | LP | 1 |
| TTR | ATTR | Definitive | No | rs28933979 | 18 | 31592974 | c.148G>A | p.Val50Met | 0.0057% | Absent | P | P | 1 |
| TTR | ATTR | Definitive | No | rs76992529 | 18 | 31598655 | c.424G>A | p.Val142Ile | 0.0887% | 0.3416% | LP/P | P | 9 |
AA = amino acid; ACMG = American College of Medical Genetics and Genomics; AD = autosomal dominant; AF = allele frequency; ARVC = arrhythmogenic right ventricular cardiomyopathy; ATTR = transthyretin amyloidosis; BrS = Brugada syndrome; DCM = dilated cardiomyopathy; HCM = hypertrophic cardiomyopathy; INTCM = intrinsic cardiomyopathy; LP = likely pathogenic; LQTS = long QT syndrome; N = number of individuals; P = pathogenic.
Figure 1.
Distribution of Causal Variants
Distribution of distinct causal variants across genes. Bars indicate the number of pathogenic or likely pathogenic variants identified in each gene. Colors denote novelty status, with novel variants defined as those absent from ClinVar, gnomAD, ABraOM, and dbSNP.
Figure 2 summarizes the distribution of the causal variants across genes. TTN was the most frequently affected gene, followed by MYBPC3 and FLNC.
Figure 2.
Causal Variants Across HF Etiologies
Distribution of causal variants across heart failure etiologies. Each bar represents the number of causal variants identified within each etiology stratified by gene. Etiology-specific sample sizes are shown in parentheses as the number of individuals carrying at least one causal variant relative to the total number of individuals in that etiological group. HF = heart failure.
As several variants were observed in more than one individual, we made a pairwise kinship analysis using KING,29 which showed no evidence of relatedness up to the third degree among carriers of recurrent variants in KCNQ1, MYBPC3, TTN, and TTR (Supplemental Table 9).
Groups
We categorized the individuals according to the variant’s classification. The group in which a VUS was identified was named the “Uncertain” group (77.7%). Those with only B or LB variants were in the “No Evidence for Causal Variant” group (15.8%). Individuals with at least one P or LP variant in AD genes and with definitive association by ClinGen were classified as “Causal Variant,” representing 6.5% of the total. After categorization, we tested the statistical differences between clinical and demographic variables (Table 1). There was no statistical difference in sex, genomic ancestry, HF cohort with reduced ejection fraction or HF with mildly reduced ejection fraction, or other clinical measurements. Yet, we did find statistical differences in age (P = 0.010), body mass index (P = 0.040), HF etiology (P = 0.030), and family history of cardiac disease (P = 0.034). A detailed familial history comparison is provided in Supplemental Table 8.
Local ancestry inference—genomic intervals and variant level analysis
We evaluated the local ancestry distribution within genomic intervals harboring LP/P and VUS variants. Across all categories, European ancestry showed the highest mean proportion compared to African and Native American ancestries. No significant difference was observed between expected and observed frequencies of LP/P variants across ancestry groups (chi-square test 0.146, P = 0.929). In contrast, genomic regions containing uncertain variants showed a significant enrichment of African and Native American ancestries (chi-square test: 60.453, P < 0.001) (Figure 3).
Figure 3.
Local Ancestry Distribution Across Intervals Containing Pathogenic/Likely Pathogenic and VUS Variants
Local ancestry proportions were derived using inferred segments from GNOMIX (.msp file output). (A) Proportion of segments assigned to each ancestry group stratified by variant classification (ie, all, pathogenic/likely pathogenic, and variant of uncertain significance). (B) Observed-to-expected ratio of local ancestry proportions in segments containing pathogenic/likely pathogenic and variant of uncertain significance. The observed proportion corresponds to the ancestry distribution within segments containing variants, whereas the expected proportion corresponds to the ancestry distribution across all genomic segments. The dashed line indicates a ratio = 1, with no enrichment. P values were obtained using a chi-square test using the ancestry distribution across all variants as a reference. AFR = African; EUR = European; LP = likely pathogenic; NAM = Native American; P = Pathogenic; VUS = variant of uncertain significance.
At the individual variant level, we evaluated the local ancestry of the haplotype carrying the alternative allele. Among the 67 LP/P variants, 39 mutated alleles had European local ancestry (58.2%), 20 had African ancestry (29.85%), 6 had Native American ancestry (8.95%), and 2 cases were missing (Supplemental Table 4).
Discussion
We identified 52 distinct causal variants across cardiac disease–associated genes in 1,013 Brazilian unrelated adult individuals with ejection fraction ≤50% from different etiologies, 18 of which were novel. As expected, the distribution of HF etiologies differed between groups, with a higher proportion of causal variants observed among individuals classified as having idiopathic DCM (P = 0.030) (Central Illustration).
Central Illustration.
Genetics of Heart Failure Across Diverse Etiologies in an Admixed Population
A total of 1,013 individuals with HFrEF from diverse etiologies underwent rare variant analysis in 123 ClinGen-curated cardiovascular genes and local ancestry inference. Pathogenic or likely pathogenic variants were identified across multiple heart failure etiologies, with the highest frequency observed among individuals with idiopathic heart failure. Variants of uncertain significance were more frequently observed in genomic regions with African and Native American local ancestries compared with the genome-wide ancestry distribution, highlighting the challenges of variant interpretation in underrepresented populations. AFR = African; B = benign; EUR = European; HFrEF = heart failure with reduced ejection fraction; LB = likely benign; LVEF = left ventricular ejection fraction; LP = likely pathogenic; NAM = Native American; P = pathogenic; VUS = variant of uncertain significance.
At the gene level, TTN accounted for nearly 50% of LP/P findings, all of which were loss-of-function (LoF) variants. Several previous studies have demonstrated that TTN truncating variants are the most common cause of DCM,30, 31, 32 and a study published in 202533 showed that individuals with TTN LoF variants have a 21-fold increased risk of developing DCM when compared with TTN-negative individuals. Moreover, all TTN LP/P were located in constitutive exons, with a PSI of 100%. Two additional truncating TTN variants were identified; however, these variants were located in exons with PSI values of 62% and 75%. Previous studies have shown that TTN truncating variants associated with DCM are enriched in highly expressed exons (commonly defined as PSI >90%), whereas truncating variants in exons with lower PSI are more frequently observed in the general population.28,34 Therefore, based on the PSI values, these 2 variants were here classified as VUS. Notably, 71% of TTN LP/P variants were absent from the gnomAD database, and 46% were novel, reflecting the high mutational burden of TTN and the expected enrichment of rare variants in cardiomyopathy cohorts.35
Novel variants were also identified in MYBPC3, FLNC, BAG3, and DSP, genes previously implicated in cardiomyopathies. MYBPC3 is a well-established gene associated with hypertrophic cardiomyopathy (HCM), accounting for approximately 50% of the P variants identified in HCM cases.36 FLNC and BAG3 are both linked to DCM. FLNC encodes a cytoskeletal protein that maintains the structural integrity of the cardiac sarcomere,37 while BAG3 plays an important role in myofibrillogenesis and is widely associated with DCM, often with variable age of onset and phenotypic expressivity.38, 39, 40 Together, these findings reinforce that sarcomeric and cytoskeletal disruption are contributing mechanisms in the genetic susceptibility to HF.
P TTR variants were identified in 10 individuals, including 9 carriers of the p.Val142Ile variant. Although none had a diagnosis of cardiac amyloidosis at enrollment, it is important to highlight that the disease presentation is age-dependent. Indeed, one of our participants carrying the p.Val142Ile variant was diagnosed as an amyloidosis patient after the development of new symptoms pointed to this etiology (Supplemental Table 7). It remains to be determined if individuals with a different HF etiology than amyloidosis might benefit from the availability of disease-modifying therapies for amyloidosis, but our study provides an empirical estimate of this prevalence in a diverse HF cohort.
Other findings, although detected in individuals with a HF diagnosis, were considered secondary, as they do not influence HF directly. TGFB3, for example, currently has limited gene-disease evidence for arrhythmogenic right ventricular cardiomyopathy,15 though it is an established gene associated with Syndromic Aortic Aneurysms.41 Likewise, KCNQ1 is a major gene for Long QT Syndrome cases, and its association with HF or cardiomyopathies is rare.42 Five LP/P variants were identified in the OBSCN, a gene primarily involved in sarcomere assembly with a LoF mechanism and an AR inheritance pattern.43 Although the ClinGen Expert Panels classifies OBSCN as having limited evidence for AD hypertrophic and DCM,17,18 recent studies have shown an enrichment of truncating variants in individuals with HCM and DCM.44,45 These data suggest that OBSCN LoF variants may act as a contributing factor, often with variable or late penetrance. However, the biological mechanisms in this gene remain unclear, and further studies are needed to clarify the role of OBSCN in cardiomyopathy and HF. Thus, P variants in OBSCN, TGFB3, and KCNQ1 were considered secondary findings.
Interestingly, some patients classified as having nongenetic etiologies, such as Chagasic and hypertensive cardiomyopathy, also carried LP/P variants in well-established cardiomyopathy genes. It is important to note that we cannot determine whether the clinically assigned etiology caused HF or whether these individuals had an underlying genetic cardiomyopathy with a coexisting condition, such as positive serology for Chagas or hypertension. Although our data do not support observation of a gene-environment interaction, a “second-hit” mechanism has been proposed in other contexts, such as viral myocarditis.46
Our results showed that a causal variant was found in 6.5% of the individuals. This overall diagnostic yield is lower than that reported in cohorts enriched for genetic cardiomyopathies or advanced HF. For instance, Linnér et al9 identified a higher yield of P variants (about 30%) in a Swedish population with different HF etiologies. However, the subjects in that study consisted of individuals with early-onset HF who were eligible for a heart transplantation. Similarly, Dellefave-Castillo et al47 identified causal variants in 20% of the individuals with cardiomyopathy and arrhythmia, etiologies associated with a higher prevalence of monogenic disease. In contrast, Povysil et al8 reported a diagnostic yield of nearly 4% in a cohort of individuals with midrange and reduced HF of different etiologies, mostly ischemic. These differences highlight that the genetic yield clearly depends on the cohort composition. Our study included individuals with diverse HF etiologies, including ischemic, hypertensive, and Chagasic, many of which are predominantly multifactorial and in which the monogenic contribution is less pronounced, if ever considered. Consequently, the monogenetic contribution in this type of cohort is inherently lower than in studies focused on familial or early-onset cardiomyopathies. When we restricted the analysis to patients with idiopathic/dilated HF, the diagnostic yield increased to approximately 10% (30/303), compared with about 5% in nonidiopathic etiologies. As such, we propose that the presence of non-DCM HF together with a family history of HF should be seen as a risk predictor for DCM (or of HF associated with a P/LP genetic variant) and, thus, a candidate for genetic testing.
We did not observe statistical differences in local genetic ancestry distribution in regions containing LP/P variants, suggesting no evidence of ancestry-specific enrichment for P variants. However, a difference was observed in regions harboring uncertain variants, with lower rates in European ancestry. Other studies have also shown that non-European individuals have a higher rate of uncertain variants than Europeans, highlighting a knowledge bias in genomic data from different ancestries, as non-Europeans and admixed individuals remain underrepresented in genomic databases.48,49 This is expected, and we extend this observation to admixed individuals, even if they self-report as from European ancestry. These results underscore the need to increase the representation of diverse populations in genomic studies, which can improve the accuracy of genetic tests.50 Our results also highlight the role of local ancestry interrogation in population studies from admixed individuals, since self-referred ancestry might not completely capture the bias in available genetic information in public datasets.
Study limitations
First, it was conducted at a single tertiary hospital in São Paulo state, which may have biased the sample toward more severe HF phenotypes. The sample size remains modest for rare-variant association and ancestry testing. Finally, we did not analyze copy number variants or large structural variants, which can significantly contribute to HF. Other analyses, including noncoding variants or regulatory regions, should be performed. Also, functional assays such as RNA-Seq or Mini-gene editing can help determine the impact of some variants, especially those classified as VUS, which might be unequally distributed among patients with different types of HF.
Conclusions
In a Brazilian HF cohort with diverse etiologies and ancestries, we observed a significant monogenic contribution to HF in 6.5% of the cohort. TTN accounted for the largest number of LP/P variants, with an enrichment among idiopathic cases, consistent with a stronger monogenic component in this subgroup. We also identified previously unreported variants across other established cardiomyopathy genes. Local ancestry analyses showed a higher proportion of VUS in genomic regions of non-European ancestry, highlighting disparities in variant interpretation. These results raise the discussion on the use of genetic testing in HF from different etiologies and illustrate the expected diagnostic yield in cohorts with varied etiologies. It reinforces the need to increase representation of diverse populations in genomic studies of HF.
Perspectives.
COMPETENCY IN MEDICAL KNOWLEDGE: Pathogenic or likely pathogenic variants are present in a subset of patients with heart failure with an ejection fraction ≤50%, including those with nonidiopathic etiologies, supporting a role for genetic evaluation beyond classical cardiomyopathies.
TRANSLATIONAL OUTLOOK: The implementation of genetic testing in heterogeneous heart failure populations remains limited by challenges in variant interpretation, particularly in admixed individuals, where a higher proportion of variants of uncertain significance persists. Increasing the representation of non-European and admixed populations in genomic databases and studies could improve the interpretation of rare variants in these populations.
Data Availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Funding support and author disclosures
This study was financed, in part, by the São Paulo Research Foundation (FAPESP), Brasil (process number #2025/17744-0). Ms Lima has received a scholarship from Fundação Coordenação de Aperfeiçoamento de Pessoal de Nível Superior, Brazil (CAPES) (#88887.531079/2020-00). Drs L.V. Pereira, Hünemeier, and A.C. Pereira have received funding from Ministerio da Saude do Brasil (Programa Genomas Brasil–888379/2019). Dr Castro e Silva has received funding from the European Union's Horizon 2020 Research and Innovation Programme/Marie Skłodowska-Curie COFUND/EUTOPIA 945380 and grant JDC2022 to 049175-I funded by MICIU/AEI/10.13039/501100011033 and by European Union NextGenerationEU/PRTR. Dr L.V. Pereira is a founder of gen-t Science. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Footnotes
The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the Author Center.
Appendix
For an expanded Methods section as well as supplemental tables, please see the online version of this paper.
Supplementary material
References
- 1.Heidenreich P.A., Bozkurt B., Aguilar D., et al. 2022 AHA/ACC/HFSA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. J Am Coll Cardiol. 2022;79(17):1757–1780. doi: 10.1016/j.jacc.2021.12.011. [DOI] [PubMed] [Google Scholar]
- 2.Gioli-Pereira L., Marcondes-Braga F.G., Bernardez-Pereira S., et al. Predictors of one-year outcomes in chronic heart failure: the portrait of a middle income country. BMC Cardiovasc Disord. 2019;19(1):251. doi: 10.1186/s12872-019-1226-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Rohde L.E.P., Montera M.W., Bocchi E.A., et al. Diretriz brasileira de insuficiência cardíaca crônica e aguda. Arq Bras Cardiol. 2018 doi: 10.5935/abc.20180190. [DOI] [PubMed] [Google Scholar]
- 4.Czepluch F.S., Wollnik B., Hasenfuß G. Genetic determinants of heart failure: facts and numbers. ESC Heart Fail. 2018;5(3):211–217. doi: 10.1002/ehf2.12267. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Domínguez F., Adler E., García-pavía P. Alcoholic cardiomyopathy: an update. Eur Heart J. 2024;45(26):2294–2305. doi: 10.1093/eurheartj/ehae362. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Russell F., Hopewell S., Nqoba N. Peripartum cardiomyopathy: a comprehensive and contemporary review European Society of Cardiology. Heart Fail Rev. 2024;29(6):1261–1278. doi: 10.1007/s10741-024-10435-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ouarhache M., Marquet S., Frade A.F., et al. Rare pathogenic variants in mitochondrial and inflammation-associated genes may lead to inflammatory cardiomyopathy in Chagas disease. J Clin Immunol. 2021:1048–1063. doi: 10.1007/s10875-021-01000-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Povysil G., Chazara O., Carss K.J., et al. Assessing the role of rare genetic variation in patients with heart failure. JAMA Cardiol. 2021;6(4):379–386. doi: 10.1001/jamacardio.2020.6500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Linnér E., Czuba T., Gidlöf O., et al. Whole genome sequencing in early onset advanced heart failure. Sci Rep. 2025;15(1):1–9. doi: 10.1038/s41598-025-88465-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Petrovski S., Goldstein D.B. Unequal representation of genetic variation across ancestry groups creates healthcare inequality in the application of precision medicine. Genome Biol. 2016;17(1):16–18. doi: 10.1186/s13059-016-1016-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gudmundsson S., Singer-Berk M., Watts N.A., et al. Variant interpretation using population databases: lessons from gnomAD. Hum Mutat. 2022;43(8):1012–1030. doi: 10.1002/humu.24309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Gioli-Pereira L., Bernardez-Pereira S., Goulart Marcondes-Braga F., et al. Genetic and ElectroNic medIcal records to predict oUtcomeS in Heart Failure patients (GENIUS-HF) - design and rationale. BMC Cardiovasc Disord. 2014;14:1–5. doi: 10.1186/1471-2261-14-32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Nunes K., Araújo Castro e Silva M., Rodrigues M.R., et al. Admixture’s impact on Brazilian population evolution and health. Science. 2025;388(6748) doi: 10.1126/science.adl3564. [DOI] [PubMed] [Google Scholar]
- 14.McLaren W., Gil L., Hunt S.E., et al. The ensembl variant effect predictor. Genome Biol. 2016;17(1):1–14. doi: 10.1186/s13059-016-0974-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.James C.A., Jongbloed J.D.H., Hershberger R.E., et al. International evidence based reappraisal of genes associated with arrhythmogenic right ventricular cardiomyopathy using the clinical genome resource framework. Circ Genomic Precis Med. 2021;14(3) doi: 10.1161/CIRCGEN.120.003273. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Hosseini S.M., Kim R., Udupa S., et al. Reappraisal of reported genes for sudden arrhythmic death: Evidence-based evaluation of gene validity for Brugada syndrome. Circulation. 2018;138(12):1195–1205. doi: 10.1161/CIRCULATIONAHA.118.035070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Jordan E., Peterson L., Ai T., et al. Evidence-based assessment of genes in dilated cardiomyopathy. Circulation. 2021;144(1):7–19. doi: 10.1161/CIRCULATIONAHA.120.053033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Hespe S., Waddell A., Asatryan B., et al. Genes associated with hypertrophic cardiomyopathy: a reappraisal by the ClinGen hereditary cardiovascular disease gene curation expert panel. J Am Coll Cardiol. 2025;85(7):727–740. doi: 10.1016/j.jacc.2024.12.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ingles J., Goldstein J., Thaxton C., et al. Evaluating the clinical validity of hypertrophic cardiomyopathy genes. Circ Genomic Precis Med. 2019;12(2):57–64. doi: 10.1161/CIRCGEN.119.002460. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Adler A., Novelli V., Amin A.S., et al. An international, multicentered, evidence-based reappraisal of genes reported to cause congenital long QT syndrome. Circulation. 2020;141(6):418–428. doi: 10.1161/CIRCULATIONAHA.119.043132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Richards S., Aziz N., Bale S., et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405–424. doi: 10.1038/gim.2015.30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Durkie M., Cassidy E.J., Berry I., et al. ACGS best practice guidelines for variant classification in rare disease 2024. Assoc Clin Genomic Sci. 2024;v1(2):1–54. [Google Scholar]
- 23.Naslavsky M.S., Scliar M.O., Yamamoto G.L., et al. Whole-genome sequencing of 1,171 elderly admixed individuals from Brazil. Nat Commun. 2022;13(1):1004. doi: 10.1038/s41467-022-28648-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ioannidis N.M., Rothstein J.H., Pejaver V., et al. REVEL: an ensemble method for predicting the pathogenicity of rare missense variants. Am J Hum Genet. 2016;99(4):877–885. doi: 10.1016/j.ajhg.2016.08.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Pejaver V., Byrne A.B., jian F.B., et al. Calibration of computational tools for missense variant pathogenicity classification and ClinGen recommendations for PP3/BP4 criteria. Am J Hum Genet. 2022:2163–2177. doi: 10.1016/j.ajhg.2022.10.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Hilmarsson H., Kumar A.S., Rastogi R., Bustamante C.D., Montserrat D.M., Ioannidis A.G. High resolution ancestry deconvolution for next generation genomic data. bioRxiv. 2021 doi: 10.1101/2021.09.19.460980. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Hofmeister R.J., Ribeiro D.M., Rubinacci S., Delaneau O. Accurate rare variant phasing of whole-genome and whole-exome sequencing data in the UK Biobank. Nat Genet. 2023;55(7):1243–1249. doi: 10.1038/s41588-023-01415-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Roberts A.M., Ware J.S., Herman D.S., et al. Integrated allelic, transcriptional, and phenomic dissection of the cardiac effects of titin truncations in health and disease. Sci Transl Med. 2015;7(270) doi: 10.1126/scitranslmed.3010134. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Manichaikul A., Mychaleckyj J.C., Rich S.S., Daly K., Sale M., Chen W.M. Robust relationship inference in genome-wide association studies. Bioinformatics. 2010;26(22):2867–2873. doi: 10.1093/bioinformatics/btq559. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Herman D.S., Lam L., Taylor M.R.G., et al. Truncations of titin causing dilated cardiomyopathy. N Engl J Med. 2012;366(7):619–628. doi: 10.1056/nejmoa1110186. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Akinrinade O., Alastalo T.P., Koskenvuo J.W. Relevance of truncating titin mutations in dilated cardiomyopathy. Clin Genet. 2016;90(1):49–54. doi: 10.1111/cge.12741. [DOI] [PubMed] [Google Scholar]
- 32.Gerull B., Gramlich M., Atherton J., et al. Mutations of TTN, encoding the giant muscle filament titin, cause familial dilated cardiomyopathy. Nat Genet. 2002;30(2):201–204. doi: 10.1038/ng815. [DOI] [PubMed] [Google Scholar]
- 33.Johnson R., Fletcher R.A., Peters S., et al. Titin-related familial dilated cardiomyopathy: factors associated with disease onset. Eur Heart J. 2025:1–18. doi: 10.1093/eurheartj/ehaf380. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Schafer S., De Marvao A., Adami E., et al. Titin-truncating variants affect heart function in disease cohorts and the general population. Nat Genet. 2017;49(1):46–53. doi: 10.1038/ng.3719. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Francesca M., Feo D., Paramonov I., et al. The burden of TTN variants in the genomic era: analysis of 18,462 individuals from the Solve-RD consortium and general recommendations. Genet Med. 2025 doi: 10.1016/j.gim.2025.101649. [DOI] [PubMed] [Google Scholar]
- 36.Cirino A.L., Channaoui N., Ho C. In: GeneReviews [Internet] Adam M.P., Bick S., Mirzaa G.M., Pagon R.A., Wallace S.E., Amemiya A., editors. University of Washington, Seattle; Seattle (WA): 2008. Nonsyndromic Hypertrophic Cardiomyopathy Overview; pp. 1993–2026. [PubMed] [Google Scholar]
- 37.Olivetti N., Pagotti M., Santos B.M., et al. Filamin C and cardiomyopathies: linking genetic variants to their phenotypic impact. ABC Hear Fail Cardiomyop. 2025;5(2) doi: 10.36660/abchf.20240059i. [DOI] [Google Scholar]
- 38.Toro R., Pérez-Serra A., Campuzano O., et al. Familial dilated cardiomyopathy caused by a novel frameshift in the BAG3 gene. PLoS One. 2016;11(7):1–11. doi: 10.1371/journal.pone.0158730. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Feldman A.M., Begay R.L., Knezevic T., et al. Decreased levels of BAG3 in a family with a rare variant and in idiopathic dilated cardiomyopathy. J Cell Physiol. 2014;229(11):1697–1702. doi: 10.1002/jcp.24615. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Qu H.Q., Feldman A.M., Hakonarson H. Genetics of BAG3: a paradigm for developing precision therapies for dilated cardiomyopathies. J Am Heart Assoc. 2022;11(23) doi: 10.1161/JAHA.122.027373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Bertoli-Avella A.M., Gillis E., Morisaki H., et al. Mutations in a TGF-β ligand, TGFB3, cause syndromic aortic aneurysms and dissections. J Am Coll Cardiol. 2015;65(13):1324–1336. doi: 10.1016/j.jacc.2015.01.040. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Allen K.Y., Vetter V.L., Shah M.J., O’Connor M.J. Familial long QT syndrome and late development of dilated cardiomyopathy in a child with a KCNQ1 mutation: a case report. Hear Case Reports. 2016;2(2):128–131. doi: 10.1016/j.hrcr.2015.10.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Young P., Ehler E., Gautel M. Obscurin, a giant sarcomeric Rho guanine nucleotide exchange factor protein involved in sarcomere assembly. J Cell Biol. 2001;154(1):123–136. doi: 10.1083/jcb.200102110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Akinrinade O., Lesurf R., Ambrose J.C., et al. Age and sex differences in the genetics of cardiomyopathy. J Cardiovasc Transl Res. 2023;16(6):1287–1302. doi: 10.1007/s12265-023-10411-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Wu G., Liu J., Liu M., et al. Truncating variants in OBSCN gene associated with disease-onset and outcomes of hypertrophic cardiomyopathy. Circ Genomic Precis Med. 2021;14(5) doi: 10.1161/CIRCGEN.121.003401. [DOI] [PubMed] [Google Scholar]
- 46.Ammirati E., Kontorovich A.R., Cooper L.T. Illuminating a hidden risk: the genetic contribution to acute Myocarditis. JACC Heart Fail. 2024;12(6):1112–1114. doi: 10.1016/j.jchf.2024.03.015. [DOI] [PubMed] [Google Scholar]
- 47.Dellefave-Castillo L.M., Cirino A.L., Callis T.E., et al. Assessment of the diagnostic yield of combined cardiomyopathy and arrhythmia genetic testing. JAMA Cardiol. 2022;7(9):966–974. doi: 10.1001/jamacardio.2022.2455. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Chen E., Facio F.M., Aradhya K.W., et al. Rates and classification of variants of uncertain significance in hereditary disease genetic testing. JAMA Netw Open. 2023;6(10) doi: 10.1001/jamanetworkopen.2023.39571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Roberts M.E., Susswein L.R., Janice Cheng W., et al. Ancestry-specific hereditary cancer panel yields: moving toward more personalized risk assessment. J Genet Couns. 2020;29(4):598–606. doi: 10.1002/jgc4.1257. [DOI] [PubMed] [Google Scholar]
- 50.Hindorff L.A., Bonham V.L., Brody L.C., et al. Prioritizing diversity in human genomics research. Nat Rev Genet. 2018;19(3):175–185. doi: 10.1038/nrg.2017.89. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.