Highlight
– Whole‑genome data from 416,085 UK Biobank participants show that polygenic risk, a composite rare‑variant gene set, and somatic clonal hematopoiesis (CHIP) each independently increase 5‑year incident atrial fibrillation (AF) risk.
– Individuals carrying all three genetic drivers had at least a two‑fold higher 5‑year AF cumulative incidence versus those with a single driver.
– Integrating the integrated genomic model for AF (IGM‑AF) with a clinical risk tool (CHARGE‑AF) raised the C‑statistic to 0.80 and produced a modest net reclassification improvement (NRI 0.08), supporting additive predictive value of comprehensive genomic profiling.
Background
Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia and a leading cause of stroke, heart failure, and health‑care utilization. Genetic predisposition to AF is complex and includes contributions from common polygenic variation, rare high‑impact coding variants, and acquired somatic mutations in hematopoietic cells (clonal hematopoiesis of indeterminate potential, CHIP). Prior genome‑wide association studies have identified many loci that modestly influence AF risk, and isolated reports link rare pathogenic variants (for example, in cardiomyopathy or ion channel genes) to AF. Meanwhile, CHIP—age‑related expansion of hematopoietic clones carrying somatic mutations in genes such as DNMT3A, TET2, and ASXL1—has been associated with cardiovascular disease and may mechanistically promote arrhythmogenesis via inflammation or altered immunity. How these three genetic layers combine to shape incident AF risk at the population level, and whether integrating them improves risk prediction beyond established clinical models, has not been well defined.
Study design
This cohort analysis used whole‑genome sequencing (WGS) from 416,085 participants of the UK Biobank (recruited 2006–2010, age 40–69 at baseline). Follow‑up ascertained incident AF through hospital records, death registries, and self‑report. Investigators derived an AF polygenic risk score (PRS) from published GWAS signals, constructed a composite rare‑variant gene set (AFgeneset) aggregating rare functional variants in genes implicated in AF or atrial cardiomyopathy, and identified somatic CHIP variants from WGS data. Clinical AF risk was estimated using the CHARGE‑AF score. The primary outcome was 5‑year incident AF. Associations were quantified as hazard ratios (HRs) per SD or per exposure category; predictive performance was assessed with C‑statistics and net reclassification indices. Analyses included tests for interactions among PRS, rare variants, and CHIP and sensitivity analyses adjusting for conventional risk factors and ancestry covariates.
Key findings
Population and outcome summary: The cohort comprised 416,085 individuals (mean age 56.6 ± 8.0 years; 54.0% female) with 30,797 AF cases identified during follow‑up.
Independent associations
– Polygenic risk score (PRS): Each 1‑SD increase in PRS was associated with substantially higher 5‑year AF risk (HR 1.65; 95% CI 1.63–1.67; P < 1 × 10‑8). This magnitude is consistent with prior work showing that individuals in the top strata of polygenic risk have markedly elevated AF incidence relative to the population mean.
– Rare variant gene set (AFgeneset): Carriers of qualifying rare variants in the AFgeneset had elevated AF risk (HR 1.63; 95% CI 1.52–1.75; P = 1.46 × 10‑42). This aggregated signal suggests that, although individual rare variants are uncommon, grouping functionally related variants into a gene set reveals a meaningful population attributable risk.
– Somatic CHIP variants: Presence of CHIP was associated with higher AF risk (HR 1.26; 95% CI 1.15–1.38; P = 1.41 × 10‑6), supporting a link between acquired hematopoietic mutations and arrhythmia susceptibility.
Combinatorial effects and absolute risk
Individuals carrying all three genetic drivers—high PRS, AFgeneset rare variant carrier status, and CHIP—experienced a ≥2‑fold greater 5‑year cumulative AF incidence compared with individuals with only one driver. This suggests additive (and possibly multiplicative) effects of inherited and acquired genomic risk factors on AF development. Although absolute incidence rates were not quoted here in full detail, the study emphasizes that combinations of genomic risks stratify population risk substantially.
Predictive performance and clinical integration
IGM‑AF (the combined genomic model comprising PRS, AFgeneset, and CHIP) improved discrimination and risk classification when combined with the CHARGE‑AF clinical score. Specifically:
– Combined IGM‑AF + CHARGE‑AF: C‑statistic 0.80 (95% CI 0.80–0.80).
– Addition of IGM‑AF to CHARGE‑AF produced a net reclassification index (NRI) of 0.08 (95% CI 0.07–0.09), indicating modest but statistically significant improvement in assigning individuals to higher or lower risk categories for incident AF.
These results imply that integrating comprehensive genomic data with established clinical predictors can materially enhance AF risk prediction beyond either data type alone.
Robustness and subgroup analyses
The primary associations persisted after adjustment for conventional cardiovascular risk factors and ancestry principal components. The authors report that the three genetic components provided complementary information rather than redundant signals, consistent with distinct biological pathways (polygenic susceptibility, high‑impact rare variants affecting atrial structure/function, and systemic effects of clonal hematopoiesis).
Expert commentary and interpretation
This study is notable for leveraging WGS at population scale to simultaneously evaluate common, rare, and somatic genetic contributors to AF. The findings support a multifaceted genetic architecture in which inherited polygenic susceptibility, rare monogenic effects, and acquired somatic mutations each additively influence AF risk.
Biological plausibility: The PRS likely indexes subtle lifelong differences in atrial electrophysiology, conduction, and structural remodeling. Rare deleterious variants in atrial or structural genes can produce more pronounced perturbations in atrial substrate. CHIP may promote AF via systemic inflammation, altered cytokine signaling, and endothelial dysfunction, mechanisms implicated in atrial remodeling and thrombogenicity.
Clinical implications: The demonstrated incremental predictive value (C=0.80 with combined model) suggests potential utility for genomic‑augmented risk stratification in AF screening programs. For example, individuals with high genomic risk could be prioritized for prolonged rhythm monitoring, earlier risk factor modification, or enrollment in prevention trials. However, translation into practice requires careful evaluation of clinical consequences (e.g., downstream testing, anticoagulation thresholds), cost‑effectiveness, and acceptability to patients and clinicians.
Limitations and generalizability: Several caveats merit emphasis. The UK Biobank cohort is not fully representative of the broader population and is enriched for European ancestry, possibly limiting transferability of PRS and rare‑variant findings to other ancestries. Ascertainment of AF primarily via routine records may miss subclinical paroxysmal AF. CHIP calling from WGS has evolving sensitivity and specificity depending on variant allele fraction thresholds. The rare‑variant gene set definition and aggregation strategy influence effect estimates and require replication. Finally, predictive gains while statistically significant were modest (NRI 0.08); whether such improvements change management or outcomes is unproven.
Clinical and research implications
For clinicians: Genomic profiling that extends beyond common variant PRS to include rare variant screening and somatic CHIP assessment can better stratify AF risk in a research setting, but routine clinical implementation is premature. Clinicians should consider genomic risk as one component of a multifactorial assessment that includes age, hypertension, diabetes, body mass index, and structural heart disease.
For researchers and policymakers: Key next steps include external validation in diverse ancestries, prospective trials assessing whether genomic‑guided screening improves outcomes (e.g., earlier AF detection, stroke prevention), cost‑effectiveness analyses, and development of evidence‑based pathways for returning results and acting on high genomic risk. Mechanistic studies to dissect how CHIP variants promote atrial remodeling would inform targeted interventions (for example, anti‑inflammatory strategies).
Ethical and implementation considerations: Widespread genomic screening raises issues around informed consent, data privacy, variant interpretation (particularly for rare variants), potential insurance implications, and equitable access. Any deployment should incorporate genetic counseling pathways and robust governance.
Conclusion
This large WGS cohort study demonstrates that common polygenic variation, aggregated rare pathogenic variants, and somatic CHIP mutations each independently predict incident AF and exert additive effects on 5‑year risk. Integrating these genomic layers with a validated clinical risk model (CHARGE‑AF) increases discriminatory ability and modestly improves risk classification. While promising for precision prevention and targeted screening strategies, translation to practice requires replication across ancestries, outcome‑focused intervention trials, and careful evaluation of clinical utility, cost, and ethical implications.
Funding and clinicaltrials.gov
Funding: See the original publication for detailed funding and disclosures (Zhang et al., JAMA Cardiology 2025). ClinicalTrials.gov: Not applicable (observational cohort analysis).
References
1. Zhang R, Kim MS, Yin W, et al. Contributions of Common, Rare, and Somatic Genetic Variants to Incidence of Atrial Fibrillation. JAMA Cardiol. 2025 Oct 8:e253664. doi:10.1001/jamacardio.2025.3664.
2. Hindricks G, Potpara T, Dagres N, et al. 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation. Eur Heart J. 2021;42(5):373–498. (Guideline that frames clinical risk management of AF.)
3. Bycroft C, Freeman C, Petkova D, et al. The UK Biobank resource with deep phenotyping and genomic data. Nature. 2018;562(7726):203–209.
4. Jaiswal S, Fontanillas P, Flannick J, et al. Age‑related clonal hematopoiesis associated with adverse outcomes. N Engl J Med. 2014;371(26):2488–2498. (Seminal description of CHIP and cardiovascular associations.)