Association of Genetic Variants With Rivaroxaban Pharmacokinetics and Pharmacodynamics in Older Patients With Atrial Fibrillation: A Prospective Cohort Study

Background Rivaroxaban is frequently used for stroke prevention in atrial fibrillation. Clinicians increasingly encounter complex decision‐making regarding oral anticoagulant prescription in older patients with atrial fibrillation. Genetic contributors to variability in rivaroxaban exposure and clinical outcomes remain uncertain. Methods A total of 168 eligible participants were consecutively recruited in this prospective observational study. Rivaroxaban dose regimens were accounted for using dose‐specific standardization, dose‐stratified analyses, and a sensitivity analysis excluding the small 5‐mg and 20‐mg dose groups. Associations of targeted genetic variants with rivaroxaban concentration, anti‐FXa (factor Xa) activity, and overall bleeding were evaluated. Exploratory protein‐structure modeling was performed to generate preliminary mechanistic hypotheses. Results Median age was 76 years (interquartile range, 71–81), and 10 mg was the most frequently prescribed dose. Aldo‐keto reductase family 7 member A3 rs1738023 and rs1738025 were nominally associated with trough concentration (both P=0.020). Trough anti‐FXa activity was nominally associated with ATP‐binding cassette subfamily G member 2 rs2622604 (P=0.042) and ATP‐binding cassette transporter B1 rs1128503 (P=0.0074). Overall bleeding occurred in 23 patients (13.7%). In Firth logistic regression adjusted for age and prior bleeding, ATP‐binding cassette transporter B1 rs1045642 AG genotype was nominally associated with higher bleeding odds than AA (adjusted odds ratio [aOR], 4.65 [95% CI, 1.02–44.61]; P=0.047), whereas ATP‐binding cassette transporter B1 rs4148738 TT was nominally associated with lower odds than CC/CT (aOR, 0.34 [95% CI, 0.10–0.96]; P=0.042). No association met the Bonferroni‐corrected significance threshold. Conclusions Selected ATP‐binding cassette transporter B1, ATP‐binding cassette subfamily G member 2, and aldo‐keto reductase family 7 member A3 variants showed exploratory associations with rivaroxaban pharmacokinetics, pharmacodynamics, and overall bleeding. As none of these associations remained statistically significant after correction for multiple testing, the findings require validation in larger, ethnically diverse cohorts.