Genetic Testing in Atrial Fibrillation

Definition

Genetic testing in atrial fibrillation refers to the clinical or research-grade sequencing of cardiomyopathy and arrhythmia gene panels in patients with early-onset AF to identify disease-associated (pathogenic/likely pathogenic) variants that may signal an underlying inherited syndrome, guide additional diagnostic evaluation, and inform family screening.

Key Concepts

Epidemiology and Diagnostic Yield

• ~10% of patients with AF diagnosed before age 66 carry a disease-associated variant on a 145-gene commercial arrhythmia/cardiomyopathy panel — rising to ~17% in those diagnosed before age 30. This is comparable to yields reported in other heritable cardiac conditions where testing is guideline-endorsed. (sources/eoaf-jama-2021 — high)
• Prospective yield in a dedicated AF genetics clinic: 20%. Of 264 patients referred with AF before age 60, 242 completed a cardiomyopathy/arrhythmia panel and 48 (20%) carried a P/LP variant — twice the ~10% from unselected retrospective cohorts. Only 6 of the 48 had a cardiomyopathy or ventricular arrhythmia syndrome diagnosed before referral, so the clinic made the diagnosis rather than confirming it. (sources/genetic-eoaf-ehj-2026, rating: high)
• Prospective yield in AF <40 without structural heart disease: 11.6% P/LP, 39.1% VUS. In 122 patients with AF before age 40 excluding SHD (mean age 31.2 years), P/LP variants were predominantly cardiomyopathy genes: TTN ×4, LMNA ×2, PKP2 ×2, TNNI3 ×1. All 8 familial AF patients who tested harboured at least one variant (3 P/LP, 5 VUS). (sources/eoaf-riskfactor-ehj-2026 — medium)
• Structural normalcy at AF presentation does not exclude a cardiomyopathy genotype. LMNA, TTN, and PKP2 P/LP variants were found in patients with normal CMR — confirming that a cardiomyopathy gene may be active before overt structural disease develops. (sources/eoaf-riskfactor-ehj-2026 — medium)
• Cross-disease diagnoses are common. In a real-world combined cardiomyopathy + arrhythmia panel cohort (n=4,782), 27 of 75 additional diagnoses gained over disease-specific panels (36%) involved an arrhythmia referral indication with a cardiomyopathy gene finding, or vice versa. 10% of EOAF patients harbour variants in cardiomyopathy genes (Yoneda et al.), and 5% of DCM patients carry variants in both ion channel and cardiomyopathy genes (Li et al.). This supports combined panel testing rather than arrhythmia-only testing in AF. (sources/genetic-yield-jama-card-2022 — high)
• Current guideline status: AHA/ACC 2023 provides the only formal AF-specific recommendation: genetic testing is Class IIb / B-NR in AF ≤45 years without obvious risk factors. ESC 2024 AF guidelines do not include a specific recommendation. Major AF guidelines prior to 2023 (AHA/ACC/HRS 2014; HRS/EHRA 2011) did not recommend routine genetic testing for AF — clinical practice has outpaced formal guideline adoption. (sources/AF-AHA-2023 — very high; sources/eoaf-jama-2021 — high)

Who to Test — Patient Selection Frameworks

Three-Tier Clinical Framework (Roberts et al., CJC 2024)

A Canadian CCS/CJC White Paper stratifies AF patients into three tiers for genetic testing decisions (sources/genetic-af-cjc-2024 — high):

Tier 1 — AF as presentation of known cardiogenetic disease:

• Genetic testing is usually already indicated for the underlying cardiogenetic condition (DCM, HCM, ARVC, laminopathy, BrS, muscular dystrophy)
• Clinical clues: family history of HF/VA/SCD; AV block of any degree; spontaneously slow ventricular rate in AF; coexisting NSVT/high PVC burden; unexplained syncope; skeletal muscle disease
• Key action: even when AF is the first manifestation, careful clinical screening for the underlying cardiogenetic syndrome must precede or accompany genetic testing

Tier 2 — Isolated early-onset AF:

• Genetic testing is reasonable; 2022 HRS/EHRA/LAHRS/APHRS: "genetic testing may be performed in all index patients in whom familial AF (age <60) is established" (weak recommendation)
• Yield is lower than often reported when restricted to high-evidence genes and truly isolated AF
• No consensus on ideal gene panel; false-positive risk significant when low-evidence genes are included
• Cost-effectiveness and clinical outcome data remain unavailable
• Cascade screening in family members may motivate lifestyle modification even if genotype affects atrial phenotype only

Tier 3 — AF with conventional risk factors (age >65, hypertension, HF, valvular disease):

• Genetic contribution is low; yield of rare variant testing is limited
• PRS adds only modest prediction benefit (ΔC-statistic 0.009–0.017 over CHARGE-AF score alone)
• Not currently recommended for genetic testing

5-Criterion Patient Selection Framework (Pensa et al., JCE 2022)

Genetic testing should be considered when any of the following are present (sources/genetic-af-dxmx-jce-2022 — medium):

1. Lone AF: AF diagnosis <66 years in the absence of structural heart disease
2. No classical AF risk factors: absence of hypertension, CKD, diabetes
3. Family history of ICD implantation
4. Family history of sudden cardiac death
5. Multigenerational family history of AF

A positive family history of AF in a first-degree relative ≤65 years carries ~2× relative risk of AF in unaffected relatives and is associated with increased arrhythmia recurrence post-ablation and higher likelihood of ICD/pacemaker requirement. A thorough 3-generation family history should be obtained at every new AF diagnosis.

Prospective Pre-Test Probability (EHJ 2026, n=242)

The first prospectively measured odds ratios for a positive AF panel, adjusted for age at diagnosis, sex and time to testing: (sources/genetic-eoaf-ehj-2026, rating: high)

• Family history of AF did not predict a positive result, nor did AF type or cardioversion count.
• HARMS2-AF did not discriminate: no difference in aggregated clinical risk between genotype-positive and genotype-negative patients (P=.76), so risk-factor burden cannot be used to ration testing.
• T1 mapping beat LGE as a predictor — consistent with T1 detecting diffuse fibrosis that LGE misses.
• The authors caution that no single variable is strong enough to gate referral; a multivariable model is proposed but not derived.

Pre-Test Likelihood Factors (EHJ 2024)

Factors that increase pre-test likelihood (sources/genetic-eoaf-ehj-2024 — high):

• AF onset <45 years; family history of EOAF or cardiomyopathy <65 years
• Structural/ECG abnormalities: borderline LVH, bundle branch block, AV block, T-wave inversion
• Clinical syndrome features: LMNA (AF + conduction disease/VA); KCNQ1 (AF + QT abnormality)

Factors that reduce pre-test likelihood (sources/genetic-eoaf-ehj-2024 — high):

• Older age; classical modifiable risk factors (obesity, hypertension, diabetes, sleep apnoea, alcohol, smoking)
• Endurance sport-triggered AF; hormonal causes (hyperthyroidism); SVT-induced AF

How to Test — Gene Panel Composition and ClinGen Curation

• No AF-specific gene panel exists. Current practice uses commercial panels designed for DCM, HCM, ARVC, LQTS, BrS, and CPVT — genes validated for ventricular disease are applied to an atrial phenotype. Variant classification and penetrance estimates from cardiomyopathy cohorts may not be directly transferable to AF-centric presentations. (sources/eoaf-jama-2021 — high)
• Gene panel composition: Commercial panels typically include 145+ genes across cardiomyopathy and arrhythmia syndromes. The ESC 2021 minimum DCM panel (TTN, LMNA, MYH7, MYBPC3, TNNT2, RBM20, PLN, SCN5A, BAG3, and others) substantially overlaps with the genes identified in EOAF cohorts — validating the use of cardiomyopathy panels in AF. (sources/HF-ESC-2021 — very high)
• ClinGen evidence appraisal of AF genes: Applying the ClinGen framework to 12 AF candidate genes, only 3 had sufficient evidence to justify AF panel inclusion: SCN5A, KCNA5, TTN. Genes with growing but not definitive evidence: KCNQ1, GJA5, LMNA. The 2022 HRS/EHRA/LAHRS/APHRS expert consensus recommended SCN5A, KCNQ1, MYL4, TTN for an AF gene panel. Most genes on commercial 145+ gene panels lack sufficient AF-specific ClinGen evidence — variants in these genes carry high risk of being incorrectly attributed to AF, potentially causing harm. (sources/genetic-af-cjc-2024 — high)
• Preferred testing approach: WES or WGS over targeted panels. WES/WGS enables re-interrogation as gene classifications evolve and allows PRS calculation from WGS data. Panel testing should focus on high-confidence ClinGen-curated genes to limit VUS burden. High-confidence actionable ACMG genes in EOAF include: TTN (cardiac expressed exons), LMNA, MYBPC3, KCNQ1, DSP, SCN5A, PKP2 — accounting for ~4% of EOAF patients in the largest study. (sources/genetic-eoaf-ehj-2024 — high)
• VUS burden: ~60% of patients tested for EOAF receive only a VUS result. In the prospective <40 cohort, 39.1% carried VUS; the EOAF-JAMA 2021 cohort reported ~63% VUS. Restricting panels to high-confidence genes and applying ACMG/ClinGen-adherent curation reduces but does not eliminate VUS findings. Patients must be counselled that a VUS is the most likely single result from EOAF genetic testing. (sources/genetic-eoaf-ehj-2024 — high; sources/eoaf-riskfactor-ehj-2026 — medium; sources/eoaf-jama-2021 — high)
• ACMG variant classification: Variants are classified as benign, likely benign, VUS, likely pathogenic, or pathogenic. VUS results require expert genetic counselling to avoid undue patient anxiety or inappropriate clinical action. (sources/eoaf-jama-2021 — high)

Gene-Specific Management When P/LP Found

• Clinical action when P/LP found: Detection of a disease-associated variant triggers syndrome-specific evaluation — echocardiography for DCM/HCM, cardiac MRI for ARVC, Holter monitoring for LMNA-related conduction disease — as well as cascade family screening. (sources/eoaf-jama-2021 — high)
• CMR preferred over echocardiography for variant carriers: Structural cardiomyopathy can be detectable on CMR when echocardiography is still normal, as shown in 5/13 EOAF variant carriers in one cohort. (sources/genetic-eoaf-ehj-2024 — high)
• LMNA / FLNC + subclinical cardiomyopathy: Early surveillance and ICD consideration per ESC cardiomyopathy guidelines, even before LVEF falls to ≤35%. AF and conduction disease may precede structural cardiomyopathy by years to decades. (sources/genetic-eoaf-ehj-2024 — high)
• PKP2: Avoidance of competitive high-level exercise; surveillance for ARVC even when initial presentation is isolated AF. Large PKP2 deletions may be missed by standard Sanger sequencing — NGS required. (sources/genetic-eoaf-ehj-2024 — high; sources/genetic-af-cjc-2024 — high)
• SCN5A LOF (Brugada-associated): Caution with sodium channel blockers used in AF management. (sources/genetic-eoaf-ehj-2024 — high)
• KCNQ1 / KCNH2 LOF: Caution with potassium channel blockers. (sources/genetic-eoaf-ehj-2024 — high)
• TTN: Longitudinal cardiomyopathy surveillance (CMR preferred) even when initial ventricular function is normal; AF may precede cardiomyopathy in ~50% of TTNtv carriers who eventually develop both. (sources/genetic-eoaf-ehj-2024 — high)
• Proposed workflow: Early-onset AF → inherited heart disease clinic → genetic counselling → combined cardiomyopathy/arrhythmia panel → results management with cardiology and genetics → cascade family screening if P/LP found. (sources/eoaf-jama-2021 — high)
• Interdisciplinary team required: AF cardiologist, cardiomyopathy cardiologist, cardiovascular geneticist, genetic counsellor, and bioinformatician — particularly important given the complexity of variant interpretation in an atrial phenotype context. (sources/genetic-eoaf-ehj-2024 — high)

Measured Impact on Management

Genetic testing in AF is usually justified as prognostic. The prospective clinic cohort shows it is also therapeutic: management changed in 52% of genotype-positive patients (27 of 52). (sources/genetic-eoaf-ehj-2026, rating: high)

• Seven new ICDs and two pacemaker-to-ICD upgrades, from four distinct indications — genetic cardiomyopathy with LVEF <45% plus risk factors; ACM with syncope and ventricular arrhythmia; new HCM with individualised primary prevention; and a pacing indication in a high-ventricular-arrhythmia-risk genotype, where an ICD with pacing capability is preferred.
• Anticoagulation beyond the score: seven patients with a CHA₂DS₂-VA ≤1 were newly diagnosed with HCM or ACM and started on oral anticoagulation. This is the most direct example in the AF literature of a genetic result overriding a conventional thrombotic risk score.
• GDMT started or optimised in genotype-positive patients with any history of LV systolic dysfunction, including TTN carriers whose dysfunction had recovered with rhythm control — the authors treat these as cardiomyopathy rather than arrhythmia-induced dysfunction.
• Drug avoidance: three SCN5A carriers advised against Class I antiarrhythmics with fever precautions; one KCNQ1 carrier advised against QT-prolonging and Class III agents.
• Exercise restriction to ≤6 METs for four ACM participants with desmosomal variants.
• The change rate tracks syndrome penetrance rather than the mere presence of a variant: 33% in DCM-phenotype patients, 92% in ACM, 100% in HCM.
• A gene-elusive overlap syndrome also changes management — 11% (7 of 22) — so phenotyping has therapeutic value independent of genotype.

Genotype-Informed Anticoagulation Considerations

Standard CHA2DS2-VASc/CHADS-65 frameworks may underestimate stroke risk in certain genetic AF subtypes (sources/genetic-af-cjc-2024 — high):

• OAC regardless of CHA2DS2-VASc score: HCM; certain muscular dystrophies (myotonic dystrophy)
• Lower anticoagulation threshold may be warranted: LMNA and MYL4 (prominent atrial fibrosis and atrial mechanical dysfunction); SCN5A-p.D1275N (atrial standstill with immobile atrial myocardium)

Genotype-Guided Pharmacotherapy — GENETIC-AF Trial

• The GENETIC-AF trial is the first prospective RCT of genotype-guided AF pharmacotherapy (sources/genetic-af-dxmx-jce-2022 — medium):
  • Population: HFrEF patients with symptomatic AF/AFL
  • Intervention: Bucindolol (non-selective beta-blocker) vs. metoprolol succinate
  • Genotype: ADRB1 Arg389Arg (homozygous wildtype at beta-1 adrenergic receptor locus)
  • Secondary endpoint result: Bucindolol → 55% reduction in AF burden; 32% reduction in need for rhythm control strategies
  • Significance: Proof-of-concept for genotype-directed AF pharmacotherapy in HFrEF
• Note — conflicting trial characterisation: JCE 2022 reports the 55% AF burden reduction as the key finding; CJC 2024 characterises the trial as having failed to detect a difference in time to first AF event (the primary endpoint). See Contradictions below. (sources/genetic-af-dxmx-jce-2022 — medium; sources/genetic-af-cjc-2024 — high)

Population-Level Risk — PRS + Rare Variants (UK Biobank)

• 6 pLOF genes significantly associated with AF at population scale (n=403,990 WES): TTN, RPL3L, PKP2, CTNNA3, KDM5B, C10orf71 (5 replicated). Carrier prevalence 1.27% in the incident AF-free population. (sources/Biobank-AF-JAMA-2024 — high)
• Combined top-quintile PRS + pLOF variant → OR 7.08 (95% CI 6.03–8.28) for AF — the largest multiplicative genetic risk estimate for AF to date. This dose-response relationship provides the strongest evidence that both common and rare genetic variation should be integrated in AF risk stratification. (sources/Biobank-AF-JAMA-2024 — high)
• 10-year absolute risk: Males >60 years + top PRS quintile + pLOF variant: 24%; equivalent females: 16%. Modifiable risk factors (BMI ≥30 + hypertension + top-PRS) confer comparable absolute risk to pLOF carriers — illustrating that genetic and lifestyle risk are additive. (sources/Biobank-AF-JAMA-2024 — high)
• AF as first manifestation of cardiomyopathy — pLOF evidence: pLOF carriers (predominantly TTNtv) had cardiomyopathy HR 3.13 before AF diagnosis and HR 2.98 after — both largely TTN-driven. The AF PRS did not predict cardiomyopathy (HR 0.99). This distinguishes two biologically separate AF pathways: rare pLOF-positive AF is a cardiomyopathy-linked phenotype requiring cardiomyopathy surveillance; high-PRS AF does not carry equivalent CM risk. (sources/Biobank-AF-JAMA-2024 — high)
• Non-European applicability is limited: All UK Biobank data are from European ancestry individuals. OR 7.08 combined risk estimate cannot be extrapolated to non-European AF patients. AF PRS weights derived from European GWAS underperform in non-European populations. (sources/Biobank-AF-JAMA-2024 — high)

Contradictions / Open Questions

• Guidelines lag clinical evidence: As of 2021, major AF guidelines did not recommend genetic testing for AF. The AHA 2023 Class IIb recommendation (AF ≤45 years) is the only formal guideline adoption — but clinical practice in many centres has already extended beyond this threshold. ESC 2024 AF guidelines still have no specific recommendation. The gap between evidence and guidelines risks creating unequal access depending on centre and jurisdiction. (sources/eoaf-jama-2021 — high; sources/AF-AHA-2023 — very high)
• GENETIC-AF trial — positive vs. negative: conflicting interpretations: JCE 2022 characterises GENETIC-AF as showing a 55% reduction in AF burden with bucindolol vs. metoprolol in ADRB1 Arg389Arg HFrEF patients. CJC 2024 characterises the same trial as having failed to detect a difference in time to first AF event — the primary endpoint. The discrepancy reflects reporting of secondary (AF burden) vs. primary (time-to-first-event) endpoints; both agree GENETIC-AF is an important prototype requiring further validation. (sources/genetic-af-dxmx-jce-2022 — medium; sources/genetic-af-cjc-2024 — high)
• No AF-specific gene panel — testing leverages cardiomyopathy/arrhythmia overlap: No dedicated AF gene panel exists. Variant classification and penetrance estimates from cardiomyopathy cohorts may not be directly transferable to an AF-centric presentation. ClinGen appraisal confirms only 3 genes have sufficient AF-specific evidence (SCN5A, KCNA5, TTN), yet commercial panels include 145+ genes — a mismatch that creates high VUS burden and risk of misattribution. (sources/eoaf-jama-2021 — high; sources/genetic-af-cjc-2024 — high)
• VUS in familial AF: In 5 of 8 familial AF patients who tested positive in the prospective <40 cohort, only a VUS was found — not a P/LP variant. The clinical value of segregation analyses in relatives for VUS reclassification in familial young AF is promising but not standardised. (sources/eoaf-riskfactor-ehj-2026 — medium)
• TTN drives cardiomyopathy signal — other pLOF genes may not predict CM progression: When TTN variants are excluded from the UK Biobank analysis, the association of pLOF variants with cardiomyopathy and HF becomes non-significant. This raises the question of whether PKP2, CTNNA3, KDM5B, and RPL3L pLOF carriers face the same cardiomyopathy surveillance imperative as TTN carriers. Gene-specific surveillance strategies may be more appropriate than a uniform "any pLOF = CM surveillance" approach. (sources/Biobank-AF-JAMA-2024 — high)
• 5-Criterion JCE 2022 framework broader than AHA 2023 guideline — unvalidated prospectively: The 5-criterion framework includes SCD family history and ICD family history regardless of patient age — broader than the AHA 2023 Class IIb recommendation limited to AF ≤45 years. Neither framework has been prospectively validated in an outcome-driven study. (sources/genetic-af-dxmx-jce-2022 — medium; sources/AF-AHA-2023 — very high)
• The guideline's exclusion criterion is contradicted by the prospective data. AHA/ACC 2023 offers testing for AF ≤45 years without obvious risk factors, and selection frameworks treat classical risk factors as reasons to lower suspicion. But HARMS2-AF risk burden did not differ between genotype-positive and genotype-negative patients (P=.76), and 14% of patients with OSA — the strongest negative predictor — still carried a pathogenic variant. If AF risk factors act as second hits in a genetically susceptible atrium rather than as alternative explanations, then "obvious risk factors" is the wrong exclusion criterion and the guideline is excluding the patients it should be testing. (sources/genetic-eoaf-ehj-2026, rating: high; sources/AF-AHA-2023, rating: very high)
• Prognostic test or therapeutic test? The 52% management-change figure is presented as vindication, but no outcome was measured — seven ICDs, seven anticoagulation starts and a set of drug-avoidance recommendations have not yet been shown to prevent a stroke, an arrhythmic death or a heart-failure admission in this cohort. The case for testing currently rests on the quality of the decisions it enables. (sources/genetic-eoaf-ehj-2026)
• Non-European underrepresentation throughout: Population yield estimates (~10%), OR estimates (OR 7.08), and VUS classification rates are all derived predominantly from European ancestry cohorts. Variant misclassification risk is higher in non-European patients. No genetic testing framework has been validated in non-European AF populations. (sources/eoaf-jama-2021 — high; sources/Biobank-AF-JAMA-2024 — high; sources/genetic-eoaf-ehj-2024 — high)

Connections

• Related to concepts/Early-Onset-Atrial-Fibrillation
• Related to concepts/Cascade-Family-Screening
• Related to concepts/Genetic-Testing-in-Cardiomyopathy
• Related to concepts/Cardiogenetic-Centers
• Related to concepts/Variant-Reclassification
• Related to entities/Atrial-Fibrillation
• Related to entities/TTN
• Related to entities/LMNA
• Related to entities/PKP2
• Related to entities/SCN5A
• Related to concepts/Arrhythmogenic-Cardiomyopathy
• Related to entities/ATTR-Amyloidosis — a treatable incidental finding on an AF panel
• Related to concepts/CHA2DS2-VA — genotype can override the score

Sources

• sources/genetic-eoaf-ehj-2026
• sources/AF-AHA-2023
• sources/Biobank-AF-JAMA-2024
• sources/HF-ESC-2021
• sources/eoaf-jama-2021
• sources/eoaf-riskfactor-ehj-2026
• sources/genetic-af-cjc-2024
• sources/genetic-af-dxmx-jce-2022
• sources/genetic-eoaf-ehj-2024
• sources/genetic-yield-jama-card-2022