TL;DR
This study investigated how genetic variations influence the effectiveness and bleeding risk of rivaroxaban, a common anticoagulant for nonvalvular atrial fibrillation (NVAF). Researchers found that specific genetic markers are linked to both how well rivaroxaban works (measured by anti-FXa levels) and the likelihood of experiencing bleeding events, paving the way for more personalized treatment strategies.
What Problem Does This Paper Address?
Rivaroxaban is a widely used anticoagulant for conditions like nonvalvular atrial fibrillation (NVAF), but its effectiveness and safety can vary significantly between individuals. This variability is a major concern, as it can lead to either insufficient anticoagulation (increasing the risk of stroke) or excessive anticoagulation (increasing the risk of bleeding). While factors like age, weight, and other medical conditions are known to influence these outcomes, the role of an individual's genetic makeup in their response to rivaroxaban has not been fully elucidated. Specifically, understanding how genetic polymorphisms affect rivaroxaban's pharmacodynamics (how the drug affects the body) and its associated bleeding risk is crucial for optimizing patient care and minimizing adverse events.
Methods
The study employed a multicenter prospective cohort design, enrolling 257 patients with NVAF who were receiving rivaroxaban.
- Pharmacodynamics Assessment: The primary measure of rivaroxaban's effectiveness was its peak anti-Factor Xa (anti-FXa) level, determined 3 hours after drug administration. This is a standard method for assessing the anticoagulant effect of direct oral anticoagulants (DOACs).
- Genetic Testing: Whole-exome sequencing (WES) was performed on patient samples to identify single nucleotide polymorphisms (SNPs). Rigorous filtering criteria were applied to select relevant SNPs, including a missingness rate of less than 10%, a minor allele frequency of more than 5%, and a Hardy-Weinberg equilibrium p-value greater than 10⁻⁶. This resulted in the analysis of 75,853 SNPs.
- Statistical Analysis:
- Association analyses were conducted using PLINK v1.09 and R software to link genetic variations with peak anti-FXa levels and 12-month bleeding events. Linear regression was used for continuous variables (peak anti-FXa), and logistic regression for categorical variables (bleeding events), with Bonferroni adjustment for multiple comparisons.
- SNPs association analyses were adjusted for relevant covariates such as sex, age, rivaroxaban dose, and mean platelet volume (MPV) for anti-FXa levels, and MPV and HAS-BLED score for bleeding events.
- A relaxed p-value threshold (p < 0.05) was used for screening suggestive genetic variations, with further analysis using a more stringent threshold (p < 1.0×10⁻⁴) for identifying significant associations.
- A gene-based prediction model was developed using logistic regression, incorporating key predictors, and its performance was evaluated using Receiver Operating Characteristic (ROC) curves.
- Expression quantitative trait loci (eQTL) analyses were performed using the GTEx database to explore the functional relevance of identified SNPs.
Datasets
The study utilized data from 257 patients with NVAF enrolled between June 2017 and July 2019. This cohort served as the primary dataset for genetic analysis, pharmacodynamic measurements, and clinical outcome assessment. No external datasets were used for training or evaluation; the study generated its own data through prospective enrollment and testing.
Key Insights and Ideas
- Peak anti-FXa levels are directly linked to bleeding risk: This study reinforces previous findings that higher peak anti-FXa levels, indicative of stronger anticoagulation, are associated with an increased risk of bleeding events. This highlights the importance of monitoring and understanding factors that influence these levels.
- SUSD3 rs76292544 is a potential predictor of bleeding events: The study identified a specific SNP in the SUSD3 gene (rs76292544) that was suggestively associated with an increased risk of bleeding events over a 12-month period. Interestingly, this SNP did not appear to significantly affect peak anti-FXa levels, suggesting it might influence bleeding propensity through mechanisms independent of direct anticoagulant effect.
- Multiple SNPs influence rivaroxaban's pharmacodynamics: Five SNPs in genes like NCMAP, PRF1, PRKAG2, and POU2F3 were found to be suggestively associated with variations in peak anti-FXa levels. This indicates that an individual's genetic makeup can influence how effectively rivaroxaban inhibits Factor Xa.
- A gene-based prediction model shows promise: The development of a gene-based prediction model incorporating MPV and specific SNPs (rs76292544, rs2255317, and rs6728818) demonstrated a good predictive accuracy for 12-month bleeding events (AUC of 0.784), outperforming the traditional HAS-BLED score in this cohort. This suggests a potential for genetic information to enhance risk stratification.
- "East Paradox" and genetic influence: The study touches upon the "East Paradox," where East Asian populations may have a higher bleeding risk despite a lower ischemic risk. While not the primary focus, the identification of genetic variants contributing to these differences is a crucial step in understanding this phenomenon.
Key Results
- A total of 257 patients with NVAF were enrolled.
- Peak anti-FXa levels were significantly related to 12-month bleeding events (p = .027).
- SUSD3 rs76292544 was suggestively associated with 12-month bleeding events (OR: 4.20, 95% CI: 2.17–8.14, p = 6.43×10⁻⁵).
- Five SNPs (NCMAP rs4553122, PRF1 rs885821, PRKAG2 rs12703159, PRKAG2 rs13224758, and POU2F3 rs2298579) were suggestively associated with peak anti-FXa level (p values ranging from 2.29×10⁻⁵ to 8.70×10⁻⁵).
- The gene-based prediction model for 12-month bleeding events achieved an AUC of 0.784, with a sensitivity of 82.8% and specificity of 62.4%.
Limitations and Future Work
The authors acknowledge several limitations:
- Limited sample size for major bleeding events: The incidence of major bleeding events was low, preventing detailed pharmacogenomic analysis for this specific outcome.
- Exploratory nature of SNP associations: The identified SNP associations are considered suggestive and require validation in larger, independent cohorts of rivaroxaban-treated patients.
- Focus on anti-FXa levels: While anti-FXa levels are a good surrogate, direct measurement of rivaroxaban concentration could provide further insights.
- Need for broader FXa inhibitor study: The findings related to genetic polymorphisms influencing anti-FXa levels and bleeding risk should be further investigated in other FXa inhibitors like apixaban and edoxaban.
Future work should focus on validating these findings in larger cohorts, exploring the functional mechanisms of the identified genetic variants, and potentially integrating genetic risk scores into clinical decision-making tools for personalized rivaroxaban therapy. The authors also suggest that the gene-based prediction model needs further validation and optimization for clinical use, considering the cost-effectiveness of genetic testing.
Summary updated