Poor Correlation of Rivaroxaban Concentration with the Routine Coagulation Screening Test in Chinese Patients with Atrial Fibrillation

He, Jiake; Zhu, Bo; Shen, Yang; Hu, Jianping; Hong, Kui

doi:https://doi.org/10.1155/2023/9962812

Journal of Clinical Pharmacy and Therapeutics

On this page

Abstract Introduction Materials and Methods Results Discussion Conclusion Abbreviations Data Availability Consent Conflicts of Interest Acknowledgments Supplementary Materials References Copyright Related Articles

Research Article | Open Access

Volume 2023 | Article ID 9962812 | https://doi.org/10.1155/2023/9962812

Poor Correlation of Rivaroxaban Concentration with the Routine Coagulation Screening Test in Chinese Patients with Atrial Fibrillation

Jiake He,^1,2Bo Zhu,¹Yang Shen,^3,4Jianping Hu,⁴and Kui Hong^1,3,4

Academic Editor: Nizar Tlili

Received02 Feb 2023

Revised09 Aug 2023

Accepted13 Oct 2023

Published31 Oct 2023

Abstract

Aims. The aim of this study is to assess the relationship between rivaroxaban plasma concentration quantified by the gold standard and anticoagulant activities measured by routine coagulation assays in Chinese atrial fibrillation (AF) patients. Whether the normal results of these tests were reliable to rule out clinically relevant rivaroxaban levels at various thresholds was also explored. The effect of clinical drug-drug interactions (DDIs) on the exposure and anticoagulant effect of rivaroxaban were further evaluated. Methods. 116 patients receiving rivaroxaban for the management of nonvalvular AF were recruited. Rivaroxaban concentrations and coagulation tests were measured by ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) and a blood coagulation analyzer, respectively. Results. The correlation of trough concentration (C_trough) and prothrombin time (PT) or international normalized ratio (INR) was moderate with Spearman’s correlation coefficient of 0.495 and 0.506, respectively. A normal PT/INR was unable to rule out C_trough levels of >30 ng/mL and >50 ng/mL, but the negative predictive value reached 100% to exclude C_trough of >100 ng/mL. C_trough showed a small correlation with activated partial thromboplastin time (aPTT) (Spearman’s correlation coefficient: 0.241) and no correlation with thrombin time (TT) (Spearman’s correlation coefficient: 0.074). Neither aPTT nor TT accurately predicted C_trough at any concentration. Peak concentration (C_peak) did not correlate with any coagulation parameters. The presence of digoxin and febuxostat significantly increased rivaroxaban C_trough by 2.18 fold and prolonged PT and INR by 44.16% and 43.60%, respectively. Conclusions. Normal routine coagulation assays were insufficient to monitor therapy with rivaroxaban. Poor correlations between rivaroxaban concentration and routine coagulation assays were observed in Chinese AF patients. The use of digoxin/febuxostat alone had no effect on rivaroxaban concentrations; however, combined strong breast cancer resistance protein inhibitor (febuxostat) and P-glycoprotein probe (digoxin) in patients with renal impairment is likely to cause clinically significant DDI with rivaroxaban. More studies are needed to establish routine therapeutic drug monitoring of rivaroxaban in clinical practice.

1. Introduction

Atrial fibrillation (AF), the most common sustained cardiac arrhythmia, affects 1.8% of Chinese adults [1]. AF is associated with an increased risk of myocardial infarction, all-cause mortality, heart failure, and ischemic stroke [2, 3]. The initiation of direct oral anticoagulants (DOACs) such as rivaroxaban is now recommended as the first drug of choice as an alternative or in preference to vitamin K antagonists (VKAs) in the guidelines for AF management [4, 5].

Rivaroxaban (Xarelto), a direct factor Xa inhibitor, has been approved by the China Food and Drug Administration for the prevention of stroke and systemic embolism in patients with nonvalvular AF in 2015. It potently inhibits prothrombinase activity and endogenous factor Xa activity with IC_50s of 2.1 ± 0.4 nM and 21 ± 1 nM, respectively [6, 7]. Rivaroxaban is rapidly absorbed with an absolute bioavailability of about 80–100% for the 10 mg dosage. The maximal plasma concentration (C_max) is achieved at 2–4 h post-dose. Approximately, two-thirds of rivaroxaban is metabolized and one-third of the dose is excreted as an unchanged drug in the urine [8]. Cytochrome P450 (CYP) 3A4/5, CYP2J2, and non-CYP-associated hydrolysis are responsible for 18%, 14%, and 14% of total rivaroxaban elimination, respectively. P-glycoprotein (P-gp) and breast cancer resistance protein (BCRP) are also involved in the disposition of rivaroxaban [9].

The clinical use of rivaroxaban is generally safe and effective despite the risk of bleeding. Routine monitoring of coagulation or frequent dose adjustment is not required. However, the evaluation of rivaroxaban exposure and anticoagulant effect may help clinicians in ensuring adherence, switching between different anticoagulant therapies, in patients with compromised renal or hepatic function, in the presence of drug-drug interaction (DDI), as well as in emergencies such as bleeding, urgent procedures, antidote requirement, or an acute stroke [5, 10].

A surgically relevant rivaroxaban plasma concentration was less than 50 ng/mL [11]. Anti-Xa activity correlated well with rivaroxaban plasma concentrations in a range between 50 and 200 ng/mL [12]. However, these assays systematically overestimated rivaroxaban concentration. Ultra-high performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) has been considered as the gold standard to quantitatively assess the concentration of rivaroxaban. However, it usually requires highly skilled personnel and has a long turnaround time and is not readily available to smaller medical institutions. Point-of-care tests are not yet available for Chinese patients receiving DOACs. Therefore, routine coagulation screening tests, such as prothrombin time (PT), activated partial thromboplastin time (aPTT), international normalized ratio (INR), and thrombin time (TT) remain widely used in emergencies or urgent clinical scenarios to determine the presence, absence, supratherapeutic, or subtherapeutic of the anticoagulation effect.

However, there is an ongoing debate on the usefulness of routine coagulation assay to evaluate DOACs. Although studies have reported that rivaroxaban prolonged PT and aPTT and elevated INR in patients, routine test itself suffers from large variability, considerably due to different responsiveness to increasing concentrations, limited specificity, as well as sensitivity [13–16]. The Chinese were found to have less prothrombotic calibrated automated thrombogram parameters, longer aPTT, as well as lower protein C and S levels than Caucasians [17]. Furthermore, real-life data on rivaroxaban’s effect on routine coagulation parameters in Chinese patients is scarce. The aim of our study was to assess the relationship between rivaroxaban plasma concentration quantified by UPLC-MS/MS and anticoagulant activity measured by routine coagulation screening tests in a real clinical setting. Whether the normal results of these tests were reliable to rule out clinically relevant rivaroxaban levels at various thresholds were also explored. Then, the study determined whether concomitant medication or the presence of transporter/metabolism-related DDI had a clinically significant impact on the exposure and anticoagulant effect of rivaroxaban in Chinese AF patients.

2. Materials and Methods

2.1. Design

This was a single-center, prospective, observational study.

2.2. Ethical Statement

The study protocol was reviewed and approved by the medical ethics committee of the second affiliated hospital of Nanchang University ((2017) no. 099). This study was in accordance with the requirements of the Declaration of Helsinki of the World Medical Association. Written and informed consent was obtained from each patient or the family.

2.3. Patients and Treatment

The inclusion criteria were patients over 18 years of age receiving rivaroxaban for the management of nonvalvular AF, as well as collection of blood samples at trough and peak concentrations. In this case, the included patients were inpatients. The dosages of rivaroxaban (10–20 mg/day) and concomitant drugs were prescribed at the discretion of the attending physician based on both clinical characteristics and renal function.

At baseline, patient’s demographic data (age, sex, and body weight (bw)); hematologic tests, including hemoglobin (HGB), platelets (PLT), hematocrit (HCT), renal function tests, including serum creatine (Cre) and glomerular filtration rate (eGFR) using a simplified modification of diet in renal disease (MDRD) formula, and liver function tests, including liver enzymes, bilirubin (BIL), total protein (TP), and albumin (ALB); comorbidities, and concomitant medications were obtained (Supplemental Table 1).

2.4. Blood Sampling and Processing

Blood sampling was performed at least 3 days after initiating rivaroxaban for AF. A venous blood sample was collected in blood collection tubes with sodium citrate just before (trough, 24 ± 2 h after last pill intake) and 2 h (peak, 2–2.5 h after last pill intake) after rivaroxaban administration. Plasma samples were obtained by centrifugation and all samples were stored at −80°C until analysis by UPLC-MS/MS.

2.5. Quantification of Rivaroxaban Plasma Concentrations in Patients

Biological sample preparation and analysis were performed according to our previously established method with minor modifications [18]. A Shimadzu LC-30AD liquid chromatography system (Shimadzu, Kyoto, Japan) coupled with API 4000 triple quadruple tandem mass spectrometer (Applied Biosystems/Sciex, Framingham, MA, USA) was used. In brief, multiple reaction monitoring in positive mode was applied. Ion transitions at m/z 436.1⟶m/z 145.1 for rivaroxaban and m/z 440.2⟶m/z 145.0 for d4-rivaroxaban (internal standard) were used for the UPLC-MS/MS analysis, respectively. The optimized MS/MS parameters were as follows: dwell time, 100 ms; ion spray voltage, 5.5 kV; curtain gas, gas 1 and gas 2 (all nitrogen): 172.4 kPa, 448.2 kPa, and 379.2 kPa, respectively; source temperature, 450°C; declustering potential, 95 V for rivaroxaban and 120 V for d4-rivaroxaban; and collision energy, 40 eV for rivaroxaban and 43 eV for d4-rivaroxaban. The chromatographic separation was performed with a Thermo Hypersil Gold C18 analytical column (column size: 2.1 × 100 mm, 1.9 μm, and no. 25002-102130) at 40°C. The mobile phase consisted of water (A) and acetonitrile (B) containing 0.1% (v/v) of formic acid and 5 mmol/L of ammonium acetate. The flow rate was 0.4 mL/min and the mobile phase was programmed to linearity change as follows: 10% (B) at 0–0.4 min, 10%–80% (B) at 0.4–0.8 min, 80% (B) at 0.8–2.0 min, 80%–10% (B) at 2.0-2.1 min, and 10% (B) at 2.1–3 min. The injection volume was 5 μL.

2.6. Coagulation Screening Test

Thromborel S, Dade Actin FSL Activated PTT reagent, and Test Thrombin reagent (Siemens Healthcare Diagnostics, Marburg, Germany) were used in the measurement of PT, aPTT, and TT by Sysmex CS-5100 Automated Blood Coagulation Analyzer, respectively. The INR was calculated through the manufacturer’s international sensitivity index.

2.7. Sample Size

The sample size calculation was 116 patients, which was based upon the need to recruit in a timely fashion and the expectation that this sample size would provide reasonable estimates of the correlation between the rivaroxaban level and the coagulation screening test (assuming correlation coefficient = 0.24, significance level = 0.95, power = 0.80, and 10% loss during the study) (R, ver. 4.1.0, R Foundation for Statistical Computing).

2.8. Statistical Analysis

Categorical data were reported as frequency (n) and percent (%) and numerical data as mean and standard deviation (SD). Plasma concentration of rivaroxaban (C_trough and C_peak) and aPTT were logarithmically transformed and PT and INR were reciprocally transformed to fit a normal distribution.

Normalized concentrations were used for dose correction, thus reducing variation. The correlation between rivaroxaban concentration and coagulation parameters was evaluated using Spearman’s rank correlation coefficient. Spearman’s coefficient absolute values were interpreted as follows: between 0.1 and 0.3 as a small correlation, greater than or equal to 0.3 and less than 0.6 as a moderate correlation, and greater than or equal to 0.6 as a strong correlation. The entire range of C_trough in patients was subdivided into 5 groups according to the cutoff values for surgical procedures [11, 19–22]: ≤5 ng/mL (low concentration without a potent systemic anticoagulation effect), 5–30 ng/mL (clinically relevant concentration for surgery with a high bleeding risk), 30–50 ng/mL (clinically relevant concentration without increasing bleeding risk in general patients), 50–100 ng/mL (clinically relevant concentration warranting antidote administration in patients with serious bleeding or requiring an urgent intervention associated with a high risk of bleeding), and >100 ng/mL (extremely high concentration with a potent systemic anticoagulation effect and preclusion of intravenous thrombolysis). These concentrations were then plotted (y-axis) against the corresponding coagulation parameters (x-axis), and the results were compared with the upper limit of the reference interval. Statistical analysis was performed using SPSS ver. 22.0 (IBM, Chicago, IL, USA). Nonparametric Kruskal–Wallis one-way analysis was used in those data that were not normally distributed. After testing for homogeneity of variance, statistical analysis was performed with one-way ANOVA followed by LSD or Dunnett’s T3 post hoc test, which was used to compare rivaroxaban concentrations and coagulation parameters between patient groups receiving rivaroxaban and co-medications. Chi-square statistics were used to compare categorical data. A two-tailed probability (p) of less than 0.05 was statistically significant. Figures were designed using software (R, ver. 4.1.0, R Foundation for Statistical Computing, and SPSS ver. 22.0, IBM).

3. Results

3.1. Patient Characteristics

A total of 116 patients were enrolled and 232 blood samples were tested. Baseline patient characteristics are shown in Supplemental Table 1. 48.3% of the patients were male and 51.7% were female, with an average age of 68.0 ± 12.5 years. Renal function was mildly impaired with a mean eGFR of 77.9 ± 23.1 mL/min and a mean Cre level of 86.5 ± 32.2 μM. Of the 116 patients in this study, 7 (6.0%) received the rivaroxaban dose of 10 mg qd, 17 (14.7%) received the rivaroxaban dose of 15 mg qd, and 92 (79.3%) received the standard dose of rivaroxaban (20 mg qd).

On comparison of patient characteristics on admission, four factors indicated substantial differences (Supplemental Table 1). Dose and eGFR in the febuxostat group were lower than those of the rivaroxaban alone group ( = 0.005 and = 0.048, respectively). Body weight (bw) in the digoxin group was higher than that of the rivaroxaban alone group ( = 0.005). HCT in the digoxin plus febuxostat group and statin group was lower than that of the rivaroxaban alone group ( = 0.001 and = 0.012, respectively). Subgroup analysis of concomitant medication was performed according to the influence of drug metabolism enzymes and transporters on the pharmacokinetics of rivaroxaban (Supplemental Table 1).

3.2. Relationship of Rivaroxaban Plasma Concentration vs. Routine Coagulation Parameters

The correlation of rivaroxaban C_trough and PT was moderate (spearman’s correlation coefficient 0.495, < 0.001, Table 1, and Supplemental Figure 1A). A normal PT was unable to rule out rivaroxaban C_trough levels of >30 and >50 ng/mL (negative predictive value (NPV): 79.5% and 95% confidence intervals (CI): 68.1–87.7%; NPV: 97.3% and 95% CI: 89.6–99.5%, respectively). NPV reached 100% (95% CI: 93.8–100%) for C_trough of > 100 ng/mL (Table 2, Figure 1). Similarly, the correlation of rivaroxaban C_trough with INR was moderate (Spearman’s correlation coefficient: 0.506, < 0.001, Table 1, and Supplemental Figure 1B). A normal INR was unable to rule out rivaroxaban C_trough levels of >30 and >50 ng/mL (NPV: 76.7% and 95% CI: 66.2–84.9%; NPV: 97.7% and 95% CI: 91.1–99.6%, respectively). NPV reached 100% (95% CI: 94.7–100%) for C_trough of > 100 ng/mL (Table 2, Figure 1). The aPTT showed a small correlation with C_trough (Spearman’s correlation coefficient: 0.241, = 0.009, Table 1, and Supplemental Figure 1C). However, there was no correlation between C_trough and TT (Spearman’s correlation coefficient: 0.074, = 0.431, Table 1, and Supplemental Figure 1D). The aPTT and TT were unable to accurately predict rivaroxaban C_trough at any concentration (Table 2 and Figure 1). C_peak did not correlate with any routine coagulation parameters (Table 1).

(a)

(b)

(c)

(d)

3.3. Rivaroxaban Plasma Concentration in the Presence of Drug Interactions

The accuracy (bias, %) or precision (CV, %) in the assay of rivaroxaban, based on quality control (QC) samples that spanned the calibration range, was <15% (Table 2).

The presence of digoxin and febuxostat significantly increased rivaroxaban C_trough by 2.18 fold ( = 0.037) (Table 3). Although concomitant use of statin, amiodarone, and febuxostat increased rivaroxaban C_trough by 7.3%, 79.9%, and 22.2%, respectively, and increased normalized C_trough (defined as C_trough/daily dose) by 11.3%, 79.8%, and 66.9%, respectively, they failed to elicit significant changes (Table 3).

Rivaroxaban C_peak was significantly higher than C_trough ( < 0.001). There were no significant differences in rivaroxaban C_peak (Table 3) among various groups.

3.4. Coagulation Parameters in Patients in the Presence of Drug Interactions

The presence of digoxin and febuxostat significantly prolonged PT by 44.16% ( = 0.001) and INR by 43.60%, respectively ( = 0.001 and Table 3). Concomitant use of digoxin or febuxostat increased PT and INR by 7.21% and 6.17% and 7.34% and 6.32% respectively, while they missed statistical significance (Table 3). There were no significant differences in aPTT and TT among these groups. The quality controls of coagulation screening tests are shown in Supplemental Table 3.

4. Discussion

As a potent DOAC with predictable pharmacokinetic and pharmacodynamic profiles, rivaroxaban is widely indicated to reduce the risk of stroke and systemic embolism in patients with nonvalvular AF [8, 9]. There is an ongoing debate regarding the quantitative laboratory monitoring of DOACs [13, 15, 16, 23]. However, given the complexities in real-life clinical practice, assessment of DOACs’ anticoagulant activity by a fast and reliable assay could be helpful in decision-making, especially under urgent clinical situations. Despite the development of specific anti-Xa, anti-IIa, and LC-MS/MS assays for DOACs, [16, 24] routine coagulation screening test (PT, INR, aPTT, and TT) remains the most readily available method.

The PT test measures the activity of clotting factors I, II, V, VII, and X and the INR is a mathematical standardization of the PT. The aPTT is a measure of the activity and presence of factors II, V, and VIII-XII. Dose-dependent effects on PT and aPTT have been observed for rivaroxaban [25, 26]. In healthy Chinese volunteers, there was a strong correlation between rivaroxaban plasma concentrations and prolongation of PT (r = 0.931, a median prolongation of 1.51 times baseline) [27]. In this study, we noticed a maximum of 1.43–1.47 folds increase in PT. Li et al. found PT and aPTT correlated with the plasma concentration of rivaroxaban in Chinese patients with deep venous thrombosis (r = 0.827 and r = 0.807, respectively) [28]. However, we observed a statistically significant (but moderate) correlation of PT and INR with rivaroxaban C_trough. The difference between recruited study population (deep venous thrombosis vs. AF), sample size (39 vs. 116), concomitant medication (unknown vs. Supplemental Table 1), and used PT reagents (unknown vs. Thromborel S) could be the complicating factors. Normal PT was unreliable to exclude rivaroxaban plasma levels of >30 ng/mL and >50 ng/mL, but the NPV reached 100% to rule out C_trough of > 100 ng/mL. Thus, the PT provided some quantitative information on rivaroxaban exposure and anticoagulant effects, which was in accordance with the findings of Jabet et al. [29]. It is suggested each clinical center establishes its own institutional interdisciplinary standard operating procedure or performs a dose-response study for better data interpretation. Generally, the INR is used for VKA assessment and is not a viable option for the evaluation of factor Xa inhibitory activity. However, we found the NPV of a normal INR was 100% (46.3%–100%) to exclude rivaroxaban of >100 ng/mL, suggesting that INR must be applied with caution in the clinic. Similarly, Ofek et al. reported that INR was significantly elevated in patients receiving rivaroxaban and apixaban therapies [14]. Normal aPTT was not able to rule out rivaroxaban level at any thresholds. In addition, limited sensitivity, variability in reagents, and paradoxical response at low concentrations make aPTT not suitable for measuring rivaroxaban blood concentrations. Clinicians should be aware that the prolongation of PT/INR or aPTT may be due to other factors other than the presence of rivaroxaban, such as compromised liver function, antibiotic use, vitamin K deficiency, and lupus anticoagulant. The TT measured thrombin activity in plasma. It showed no correlation with rivaroxaban C_trough and cannot be used for any meaningful evaluation of rivaroxaban. Moreover, the mechanism of factor Xa inhibition with rivaroxaban also makes the TT an undesirable assay. Overall, we confirmed that routine coagulation assays were not sufficient to exclude a clinically relevant rivaroxaban plasma concentration [29–31].

For patients taking DOACs, surgical procedures were sometimes unavoidable, especially in the presence of ischemic or hemorrhagic strokes. The lower limit of quantification for most anti-Xa assays was set at 20 ng/mL–30 ng/mL, therefore this concentration was considered to be clinically safe and was accepted as C_trough cutoff values in many clinical centers. A preoperative rivaroxaban concentration of less than 30 ng/mL was recommended for surgery with high bleeding risk, while an empirical cutoff value of 50 ng/mL has been adopted by the University Hospital of Zurich [11, 19, 20]. However, definite clinical thresholds for DOACs in Chinese AF patients have not been established yet. In this study, most of the patients (89.5%) were found to have rivaroxaban C_trough of < 50 ng/mL, suggesting a remote likelihood of antidote requirement. In case of surgery such as AF ablation, additional intervention is likely to be unnecessary in patients (64.7%) with a C_trough of < 30 ng/mL. However, we should keep in mind that plasma concentration alone may not provide clinicians with sufficient information for decision-making. When interpreting drug concentrations, it is important to consider the timing of the last dose relative to blood sampling and other factors.

Large interindividual variability exists in rivaroxaban exposure in real-life patients, causing broad ranges of C_trough and C_peak [15, 32]. A higher-than-expected interindividual variabilities were observed (Supplemental Table 4). This large variation could result in less predictable pharmacokinetics and pharmacodynamics of rivaroxaban in a real-world clinical setting. In another aspect, it may diminish the effect of comedications on the disposition of rivaroxaban. Both intrinsic (i.e., age, renal impairment, body weight, or genetics) and extrinsic (i.e., comorbidity or comedications and environment) factors may also have an impact on the pharmacokinetics of rivaroxaban [33]. Rivaroxaban absorption was dependent on the site of drug release in the gastrointestinal tract [34]. Delay in rivaroxaban absorption is likely to result in interpatient variability as seen in this study. In addition, rivaroxaban is a substrate of ATP-binding cassette transporter of subfamily B, members 1 (ABCB1 and P-gp) and ATP-binding cassette transporter of subfamily G, members 2 (ABCG2 and BCRP), [32] which might play a role on rivaroxaban’s disposition. No clinically relevant effect of age on rivaroxaban pharmacokinetics and pharmacodynamics was observed in healthy, older adults [35]. It is well established that systemic rivaroxaban concentrations or exposures increased as the renal function declined [36]. Rivaroxaban labeling also recommended dose adjustment for renally impaired patients according to the degree of impairment. Body weight has little influence on the pharmacokinetics or pharmacodynamics of rivaroxaban in healthy subjects, in patients for prevention or treatment of venous thromboembolism, and in Thai patients with nonvalvular AF [33, 37, 38]. A previous study reported that the risk of major bleeding for patients receiving 20 mg qd rivaroxaban was 5.3% and increased with increasing exposure (AUC_ss or C_max,ss) [39, 40]. In the current study, the maximum C_peak was 708 ng/mL which is significantly high and could lead to potential bleeding risk. The C_trough for the same patient was as high as 94.9 ng/mL (Supplemental Table 4). However, we did not observe any bleeding events. An increased risk does not mean a bleeding event. Zhang et al. failed to identify a statistically significant association between exposure and the risk of major bleeding [40]. A shallow exposure-response relationship with no clear threshold for acceleration of bleeding risk has been found in patients receiving rivaroxaban for nonvalvular AF, for venous thromboembolism prophylaxis after hip/knee replacement surgery, and for acute coronary syndrome [40–42]. However, carefully examining the signs or symptoms of blood loss and promptly evaluating the risk of bleeding in patients with clinically relevant increases in exposure due to intrinsic and extrinsic factors would potentially bring clinical benefits.

The presence of polypharmacy has increased the risk of DDIs and associated adverse effects. In this study, 91.4% of AF patients received polypharmacotherapy. It is important that clinicians must be aware of the potential for DDI between rivaroxaban and specific drugs and must take measures to prevent it. As a substrate of CYP3A4/5, CYP2J2, P-gp, and BCRP, rivaroxaban is likely to act as a victim of metabolism and transporter-related DDI. Digoxin is a known P-gp probe and amiodarone is a moderate P-gp inhibitor. Both drugs were commonly prescribed in AF patients for cardiac rhythm control [43]. Consistent with previous reports, 0.125 mg of digoxin once daily has no effect on rivaroxaban concentrations. Steffel et al. reported a minor effect of amiodarone on rivaroxaban plasma concentration [5]. We observed a statistically insignificant increase in rivaroxaban C_trough (79.9%) and normalized C_trough (79.8%), respectively. In patients without any renal impairment, it seems no precautions are necessary when on concomitant use of amiodarone, as the change in exposure is unlikely to affect the bleeding risk. In line with a previous study, coadministration of rivaroxaban with other substrates of CYP3A4, CYP2C9, P-gp, BCRP, or all (e.g., atorvastatin, simvastatin, rosuvastatin, and fluvastatin) does not significantly alter the plasma levels and routine coagulation test of rivaroxaban. Potent uricosuric agents, such as febuxostat and benzbromarone, were also identified as comedications with the frequency of 14.7% in AF patients receiving rivaroxaban. Febuxostat and benzbromarone were inhibitors of BCRP with IC₅₀ values of 0.35 μM and 0.238 μM, respectively [44, 45]. At the dose of 40 mg febuxostat, the mechanistic static model predicted that the (I₂)/IC_50 BCRP and (I_max,u)/IC_50 BCRP for febuxostat were approximately 1145.7 and 0.243 (>cutoff value 10 and 0.1), indicating a high likelihood of a potential DDI through BCRP inhibition (Supplemental Table 5). 50 mg of benzbromarone was also predicted to cause a potential DDI through BCRP inhibition ((I₂)/IC_50 BCRP ratios >10 and (I_max,u)/IC_50 BCRP >0.1; Supplemental Table 5). Despite compromised renal function, the significant increase in rivaroxaban C_trough, prolonged PT, and INR after concomitant use of digoxin and febuxostat was a surprise (Table 3). The mechanism implicated is inhibition of BCRP in the gut, liver, and/or kidney, as well as in P-gp competition. It also suggests that patients taking concomitant BCRP inhibitors (such as febuxostat) and P-gp probes (such as digoxin) may be at high risk for super-therapeutic concentrations and subsequently, bleeding risk and thus can benefit from rivaroxaban concentration monitoring to prophylactically identify this risk. Dosage adjustment in certain patients could be a possible strategy.

This study has limitations. First, statistical analysis was limited by the small sample size and large variability. However, this allows us to perform an in-depth analysis of rivaroxaban at the individual level. Second, possible genetic polymorphisms, such as ABCB1 and ABCG2, and their impacts on rivaroxaban exposures and coagulation parameters in each patient were not assessed. However, implementation of this evaluation should be better performed by a new study.

5. Conclusion

Normal routine coagulation assays were insufficient to monitor therapy with rivaroxaban. Poor correlations between rivaroxaban concentration and routine coagulation screening tests were observed in Chinese AF patients. The use of digoxin/febuxostat alone had no effect on rivaroxaban concentrations; however, combined strong BCRP inhibitor (febuxostat) and P-gp probe (digoxin) in patients with renal impairment is likely to cause clinically significant pharmacokinetic and pharmacodynamic DDI with rivaroxaban. More studies are needed to establish routine therapeutic drug monitoring of rivaroxaban in clinical practice.

Abbreviations

AF:	Atrial fibrillation
DOACs:	Direct oral anticoagulants
VKAs:	Vitamin K antagonists
C_max:	Maximal plasma concentration
CYP:	Cytochrome P450
P-gp:	P-glycoprotein
BCRP:	Breast cancer resistance protein
DDI:	Drug-drug interaction
UPLC-MS/MS:	Ultra-performance liquid chromatography-tandem mass spectrometry
PT:	Prothrombin time
aPTT:	Activated partial thromboplastin time
INR:	International normalized ratio
TT:	Thrombin time
BW:	Body weight
HGB:	Hemoglobin
PLT:	Platelets
HCT:	Hematocrit
Cre:	Serum creatine
eGFR:	Glomerular filtration rate
MDRD:	Modification of diet in renal disease
TBIL:	Total bilirubin
DBIL:	Direct bilirubin
TP:	Total protein
ALB:	Albumin
C_trough:	Trough concentration
C_peak:	Peak concentration
NPV:	Negative predictive value
PPV:	Positive predictive value
QC:	Quality control
T_max:	Time to reach maximal plasma concentration
ABCB1:	ATP-binding cassette transporter of subfamily B, members 1
ABCG2:	ATP-binding cassette transporter of subfamily G, members 2
AUC_ss:	Area under the curve at steady-state
CI:	Confidence intervals
SD:	Standard deviation
CV:	Coefficient of variation
I₂:	Maximal theoretical gastrointestinal concentration
I_max,u:	Unbound steady-state plasma peak concentration
IC₅₀:	Half maximal inhibitory concentration.

Data Availability

The data used to support the findings of the study are available from the corresponding author upon request.

All patients were informed and signed the consent form.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was supported by the National Nature Science Foundation of China (82160707), the Natural Science Foundation of Jiangxi Province (20202ACB216005), the Natural Science Foundation for Academic and Technical Leaders of Jiangxi Province (20204BCJL23053), and the Jiangxi Postdoctoral Science Foundation (2017KY24).

Supplementary Materials

Supplemental Table 1: patient characteristics. Supplemental Table 2: precision and accuracy of rivaroxaban quantification in patients’ plasma by ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS). Supplemental Table 3: quality control for coagulation screening tests. Supplemental Table 4: specific patient characteristics. Supplemental Table 5: summary of inhibitory properties and pharmacokinetic parameters of febuxostat and benzbromarone at clinically relevant doses. Supplemental Table 6: the plasma concentrations and coagulation parameters in other medication with the mixed effect group. Supplemental Figure 1: correlation of rivaroxaban trough concentration (C_trough) level and routine coagulation parameters: (A) prothrombin time (PT), (B) international normalized ratio (INR), (C) activated partial thromboplastin time (aPTT), and (D) thrombin time (TT). (Supplementary Materials)

References

X. Du, L. Guo, S. Xia et al., “Atrial fibrillation prevalence, awareness and management in a nationwide survey of adults in China,” Heart, vol. 107, no. 7, pp. 535–541, 2021.
View at: Publisher Site | Google Scholar
B. Freedman, T. S. Potpara, and G. Y. Lip, “Stroke prevention in atrial fibrillation,” The Lancet, vol. 388, no. 10046, pp. 806–817, 2016.
View at: Publisher Site | Google Scholar
V. Ruddox, I. Sandven, J. Munkhaugen, J. Skattebu, T. Edvardsen, and J. E. Otterstad, “Atrial fibrillation and the risk for myocardial infarction, all-cause mortality and heart failure: a systematic review and meta-analysis,” European Journal of Preventive Cardiology, vol. 24, no. 14, pp. 1555–1566, 2017.
View at: Publisher Site | Google Scholar
C. T. January, L. S. Wann, H. Calkins et al., “AHA/ACC/HRS focused update of the 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American college of cardiology/American heart association task force on clinical practice guidelines and the heart rhythm society,” Journal of the American College of Cardiology, vol. 74, no. 1, pp. 104–132, 2019.
View at: Publisher Site | Google Scholar
J. Steffel, P. Verhamme, T. S. Potpara et al., “The 2018 European Heart Rhythm Association Practical Guide on the use of non-vitamin K antagonist oral anticoagulants in patients with atrial fibrillation,” European Heart Journal, vol. 39, no. 16, pp. 1330–1393, 2018.
View at: Publisher Site | Google Scholar
E. Perzborn, J. Strassburger, A. Wilmen et al., “In vitro and in vivo studies of the novel antithrombotic agent BAY 59-7939--an oral, direct Factor Xa inhibitor,” Journal of Thrombosis and Haemostasis, vol. 3, no. 3, pp. 514–521, 2005.
View at: Publisher Site | Google Scholar
D. Kubitza, M. Becka, B. Voith, M. Zuehlsdorf, and G. Wensing, “Safety, pharmacodynamics, and pharmacokinetics of single doses of BAY 59-7939, an oral, direct factor Xa inhibitor,” Clinical Pharmacology and Therapeutics (St. Louis), vol. 78, no. 4, pp. 412–421, 2005.
View at: Publisher Site | Google Scholar
W. Mueck, J. Stampfuss, D. Kubitza, and M. Becka, “Clinical pharmacokinetic and pharmacodynamic profile of rivaroxaban,” Clinical Pharmacokinetics, vol. 53, no. 1, pp. 1–16, 2014.
View at: Publisher Site | Google Scholar
T. Kvasnicka, I. Malikova, Z. Zenahlikova et al., “Rivaroxaban metabolism, pharmacologic properties and drug interactions,” Current Drug Metabolism, vol. 18, no. 7, pp. 636–642, 2017.
View at: Publisher Site | Google Scholar
A. Gundlund, L. Staerk, E. L. Fosbol et al., “Initiation of anticoagulation in atrial fibrillation: which factors are associated with choice of anticoagulant?” Journal of Internal Medicine, vol. 282, no. 2, pp. 164–174, 2017.
View at: Publisher Site | Google Scholar
G. Pernod, P. Albaladejo, A. Godier et al., “Management of major bleeding complications and emergency surgery in patients on long-term treatment with direct oral anticoagulants, thrombin or factor-Xa inhibitors: proposals of the working group on perioperative haemostasis (GIHP) march 2013,” Archives of Cardiovascular Diseases, vol. 106, no. 6-7, pp. 382–393, 2013.
View at: Publisher Site | Google Scholar
J. D. Studt, L. Alberio, A. Angelillo-Scherrer et al., “Accuracy and consistency of anti-Xa activity measurement for determination of rivaroxaban plasma levels,” Journal of Thrombosis and Haemostasis, vol. 15, no. 8, pp. 1576–1583, 2017.
View at: Publisher Site | Google Scholar
S. E. Conway, A. Y. Hwang, C. D. Ponte, and J. G. Gums, “Laboratory and clinical monitoring of direct acting oral anticoagulants: what clinicians need to know,” Pharmacotherapy: The Journal of Human Pharmacology and Drug Therapy, vol. 37, no. 2, pp. 236–248, 2017.
View at: Publisher Site | Google Scholar
F. Ofek, S. Bar Chaim, N. Kronenfeld, T. Ziv-Baran, and M. Berkovitch, “International normalized ratio is significantly elevated with rivaroxaban and apixaban drug therapies: a retrospective study,” Clinical Therapeutics, vol. 39, no. 5, pp. 1003–1010, 2017.
View at: Publisher Site | Google Scholar
B. T. Samuelson, A. Cuker, D. M. Siegal, M. Crowther, and D. A. Garcia, “Laboratory assessment of the anticoagulant activity of direct oral anticoagulants: a systematic review,” Chest, vol. 151, no. 1, pp. 127–138, 2017.
View at: Publisher Site | Google Scholar
J. Douxfils, W. Ageno, C. M. Samama et al., “Laboratory testing in patients treated with direct oral anticoagulants: a practical guide for clinicians,” Journal of Thrombosis and Haemostasis, vol. 16, no. 2, pp. 209–219, 2018.
View at: Publisher Site | Google Scholar
P. Ho, C. Ng, J. Rigano et al., “Significant age, race and gender differences in global coagulation assays parameters in the normal population,” Thrombosis Research, vol. 154, pp. 80–83, 2017.
View at: Publisher Site | Google Scholar
J. K. C. Z. He, R. Wan, M. Y. Rao, A. Z. Xiong, and K. Hong, “Determination of rivaroxaban in human plasma by ultra-performance liquid chromatography-tandem mass spectrometry,” Herald of Medicine, vol. 38, no. 3, pp. 384–387, 2019.
View at: Google Scholar
P. Billoir, V. Barbay, L. M. Joly, M. Fresel, M. H. Chrétien, and V. Le Cam Duchez, “Anti-xa oral anticoagulant plasma concentration assay in real life: rivaroxaban and apixaban quantification in emergency with LMWH calibrator,” The Annals of Pharmacotherapy, vol. 53, no. 4, pp. 341–347, 2019.
View at: Publisher Site | Google Scholar
A. Kaserer, G. E. Kiavialaitis, J. Braun et al., “Impact of rivaroxaban plasma concentration on perioperative red blood cell loss,” Transfusion, vol. 60, no. 1, pp. 197–205, 2020.
View at: Publisher Site | Google Scholar
D. J. Seiffge, C. Traenka, A. A. Polymeris et al., “Intravenous thrombolysis in patients with stroke taking rivaroxaban using drug specific plasma levels: experience with a standard operation procedure in clinical practice,” J Stroke, vol. 19, no. 3, pp. 347–355, 2017.
View at: Publisher Site | Google Scholar
T. Steiner, M. Bohm, M. Dichgans et al., “Recommendations for the emergency management of complications associated with the new direct oral anticoagulants (DOACs), apixaban, dabigatran and rivaroxaban,” Clinical Research in Cardiology, vol. 102, no. 6, pp. 399–412, 2013.
View at: Publisher Site | Google Scholar
A. Cuker, D. M. Siegal, M. A. Crowther, and D. A. Garcia, “Laboratory measurement of the anticoagulant activity of the non-vitamin K oral anticoagulants,” Journal of the American College of Cardiology, vol. 64, no. 11, pp. 1128–1139, 2014.
View at: Publisher Site | Google Scholar
E. Lindhoff-Last, M. M. Samama, T. L. Ortel, J. I. Weitz, and T. E. Spiro, “Assays for measuring rivaroxaban: their suitability and limitations,” Therapeutic Drug Monitoring, vol. 32, no. 6, pp. 673–679, 2010.
View at: Publisher Site | Google Scholar
I. G. Girgis, M. R. Patel, G. R. Peters et al., “Population pharmacokinetics and pharmacodynamics of rivaroxaban in patients with non-valvular atrial fibrillation: results from ROCKET AF,” The Journal of Clinical Pharmacology, vol. 54, no. 8, pp. 917–927, 2014.
View at: Publisher Site | Google Scholar
W. Mueck, A. W. Lensing, G. Agnelli, H. Decousus, P. Prandoni, and F. Misselwitz, “Rivaroxaban: population pharmacokinetic analyses in patients treated for acute deep-vein thrombosis and exposure simulations in patients with atrial fibrillation treated for stroke prevention,” Clinical Pharmacokinetics, vol. 50, no. 10, pp. 675–686, 2011.
View at: Publisher Site | Google Scholar
X. Zhao, P. Sun, Y. Zhou et al., “Safety, pharmacokinetics and pharmacodynamics of single/multiple doses of the oral, direct Factor Xa inhibitor rivaroxaban in healthy Chinese subjects,” British Journal of Clinical Pharmacology, vol. 68, no. 1, pp. 77–88, 2009.
View at: Publisher Site | Google Scholar
Y. Li, L. Du, X. Tang, Y. Chen, and D. Mei, “Laboratory monitoring of rivaroxaban in Chinese patients with deep venous thrombosis: a preliminary study,” BMC Pharmacol Toxicol, vol. 21, no. 1, p. 38, 2020.
View at: Publisher Site | Google Scholar
A. Jabet, A. Stepanian, J. L. Golmard et al., “Are screening tests reliable to rule out direct oral anticoagulant plasma levels at various thresholds (30, 50, or 100 ng/mL) in emergency situations?” Chest, vol. 153, no. 1, pp. 288–290, 2018.
View at: Publisher Site | Google Scholar
A. Kaserer, A. Schedler, B. Seifert, D. R. Spahn, J. D. Studt, and P. Stein, “Standard coagulation assays alone are not sufficient to exclude surgically relevant rivaroxaban plasma concentrations,” Perioperative Medicine, vol. 8, no. 1, p. 15, 2019.
View at: Publisher Site | Google Scholar
S. Testa, C. Legnani, A. Tripodi et al., “Poor comparability of coagulation screening test with specific measurement in patients receiving direct oral anticoagulants: results from a multicenter/multiplatform study,” Journal of Thrombosis and Haemostasis, vol. 14, no. 11, pp. 2194–2201, 2016.
View at: Publisher Site | Google Scholar
A. L. Sennesael, A. S. Larock, J. Douxfils et al., “Rivaroxaban plasma levels in patients admitted for bleeding events: insights from a prospective study,” Thrombosis Journal, vol. 16, no. 1, p. 28, 2018.
View at: Publisher Site | Google Scholar
N. Singkham, A. Phrommintikul, P. Pacharasupa et al., “Population pharmacokinetics and dose optimization based on renal function of rivaroxaban in Thai patients with non-valvular atrial fibrillation,” Pharmaceutics, vol. 14, no. 8, p. 1744, 2022.
View at: Publisher Site | Google Scholar
H. A. Hakeam and N. Al-Sanea, “Effect of major gastrointestinal tract surgery on the absorption and efficacy of direct acting oral anticoagulants (DOACs),” Journal of Thrombosis and Thrombolysis, vol. 43, no. 3, pp. 343–351, 2017.
View at: Publisher Site | Google Scholar
K. T. Moore, P. Wong, L. Zhang, G. Pan, and J. Foody, “Influence of age on the pharmacokinetics, pharmacodynamics, efficacy, and safety of rivaroxaban,” Current Medical Research and Opinion, vol. 34, no. 12, pp. 2053–2061, 2018.
View at: Publisher Site | Google Scholar
V. Ashton, S. Kerolus-Georgi, and K. T. Moore, “The pharmacology, efficacy, and safety of rivaroxaban in renally impaired patient populations,” The Journal of Clinical Pharmacology, vol. 61, no. 8, pp. 1010–1026, 2021.
View at: Publisher Site | Google Scholar
S. J. Barsam, J. P. Patel, L. N. Roberts et al., “The impact of body weight on rivaroxaban pharmacokinetics,” Research and Practice in Thrombosis and Haemostasis, vol. 1, no. 2, pp. 180–187, 2017.
View at: Publisher Site | Google Scholar
D. Kubitza, M. Becka, M. Zuehlsdorf, and W. Mueck, “Body weight has limited influence on the safety, tolerability, pharmacokinetics, or pharmacodynamics of rivaroxaban (BAY 59-7939) in healthy subjects,” The Journal of Clinical Pharmacology, vol. 47, no. 2, pp. 218–226, 2007.
View at: Publisher Site | Google Scholar
A. Solms, S. Willmann, I. Reinecke et al., “Associations between model-predicted rivaroxaban exposure and patient characteristics and efficacy and safety outcomes in the treatment of venous thromboembolism,” Journal of Thrombosis and Thrombolysis, vol. 50, no. 1, pp. 1–11, 2020.
View at: Publisher Site | Google Scholar
L. Zhang, X. Yan, K. A. A. Fox et al., “Associations between model-predicted rivaroxaban exposure and patient characteristics and efficacy and safety outcomes in patients with non-valvular atrial fibrillation,” Journal of Thrombosis and Thrombolysis, vol. 50, no. 1, pp. 20–29, 2020.
View at: Publisher Site | Google Scholar
I. Reinecke, A. Solms, S. Willmann et al., “Associations between model-predicted rivaroxaban exposure and patient characteristics and efficacy and safety outcomes in the prevention of venous thromboembolism,” Journal of Thrombosis and Thrombolysis, vol. 50, no. 1, pp. 12–19, 2020.
View at: Google Scholar
L. Zhang, X. Yan, P. Nandy et al., “Influence of model-predicted rivaroxaban exposure and patient characteristics on efficacy and safety outcomes in patients with acute coronary syndrome,” Therapeutic Advances in Cardiovascular Disease, vol. 13, 2019.
View at: Publisher Site | Google Scholar
K. V. Ledwitch and A. G. Roberts, “Cardiovascular ion channel inhibitor drug-drug interactions with P-glycoprotein,” The AAPS Journal, vol. 19, no. 2, pp. 409–420, 2017.
View at: Publisher Site | Google Scholar
M. Lehtisalo, J. E. Keskitalo, A. Tornio et al., “Febuxostat, but not allopurinol, markedly raises the plasma concentrations of the breast cancer resistance protein substrate rosuvastatin,” Clinical and Translational Science, vol. 13, no. 6, pp. 1236–1243, 2020.
View at: Publisher Site | Google Scholar
P. Tatrai, F. Erdo, G. Dornyei, and P. Krajcsi, “Modulation of urate transport by drugs,” Pharmaceutics, vol. 13, no. 6, p. 899, 2021.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2023 Jiake He et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

242

Downloads

171

Citations