DOAC drug interactions management resource

Background Over the past decade, direct oral anticoagulants (DOACs; apixaban, dabigatran, edoxaban and rivaroxaban) have offered many advantages over traditional therapy with warfarin ± low-molecular-weight heparins (LMWHs). The DOACs have established dosing without the need for coagulation monitoring as well as a quick onset (C max at 1-4 hours) and offset (half-lives ranging from 9-14 hours for patients with normal renal function), thereby eliminating the need for bridging with LMWHs (Figure 1). Moreover, DOACs have fewer drugdrug interactions (DDIs) relative to warfarin; however, as the use of DOACs continues to increase in clinical practice, more information surrounding DOAC DDIs is necessary to make timely clinical decisions. Pathways relevant to DOAC DDIs encompass the cytochrome P450 system (focusing on 3A4), as well as the Pglycoprotein (P-gp) transport system. Rivaroxaban and apixaban are substrates for P-gp and (in part) metabolized by CYP 3A4. Subsequently, rivaroxaban and apixaban DDIs must strongly affect both P-gp and CYP 3A4; the clinician should ensure a patient is not on 2 concomitant drugs that affect CYP 3A4 and P-gp separately, as these combined DDIs could cause significant changes in DOAC concentrations. In contrast, dabigatran and edoxaban are affected only by strong inhibitors/ inducers of P-gp, as they lack metabolism by the CYP enzyme. The P-gp impact is within the gastrointestinal tract; hence, to minimize the P-gp DDI, dabigatran or edoxaban may be administered 2 hours prior to the interacting agent. Notably, all DOACs have a component of renal elimination (dabigatran > edoxaban > rivaroxaban > apixaban), and while progressive renal dysfunction will result in elevated DOAC concentrations, this elimination is not a direct mechanism of DDIs. At this time, there is limited clinical pharmacokinetic (PK)/ pharmacodynamic (PD) data to quantify the clinical impact of specific DOAC DDIs. DDIs of this nature (i.e., P-gp or CYP 450) are highly variable because of the timing of the induction/ inhibition turnover as well as the strength (mild, moderate or strong) of the interaction. In addition, there is inherent intersubject variability of 30% for concentrations of dabigatran, edoxaban and apixaban, with rivaroxaban reaching 40% for PK parameters. In addition, reported ranges of DOAC concentrations assessed in subgroups of clinical trials demonstrate variability in peak/trough ratios of nearly 1.6-fold. With this in mind, DDIs that alter DOAC concentrations of 30% to 40% often still result in DOAC concentrations falling within these reported concentration ranges. Subsequently, when regulators consider providing advice surrounding DDIs, within the context of high PK/PD variability, general recommendations are often to avoid these combinations; specifically, regulators contraindicate DOACs for DDIs with inducing agents (for fear of thrombotic events) and recommend use with caution and assess other factors that may warrant avoidance when an inhibitor is the interacting culprit. Limited, if any, data provide a comparison of DDIs between the DOACs. Unique to edoxaban are recommendations for dose reduction (from 60 to 30 mg daily) in the presence of P-gp inhibitors (except amiodarone and verapamil), with certain drugs listed based on clinical trial protocols or product monograph content. As the front-line clinician continues to manage more complex clinical scenarios with consideration of DOAC use, a summary of available literature specific to DOAC DDIs is necessary, given there may be no or conflicting information for drug interactions. As such, our purpose is to provide a tool that differentiates DDIs across the 4 DOACs specific to agents commonly prescribed for patients with cardiovascular disease, with a description of available data to support the same.


Background
Over the past decade, direct oral anticoagulants (DOACs; apixaban, dabigatran, edoxaban and rivaroxaban) have offered many advantages over traditional therapy with warfarin ± low-molecular-weight heparins (LMWHs). The DOACs have established dosing without the need for coagulation monitoring as well as a quick onset (C max at 1-4 hours) and offset (half-lives ranging from 9-14 hours for patients with normal renal function), thereby eliminating the need for bridging with LMWHs ( Figure 1). [1][2][3][4][5] Moreover, DOACs have fewer drugdrug interactions (DDIs) relative to warfarin; however, as the use of DOACs continues to increase in clinical practice, more information surrounding DOAC DDIs is necessary to make timely clinical decisions.
Pathways relevant to DOAC DDIs encompass the cytochrome P450 system (focusing on 3A4), as well as the Pglycoprotein (P-gp) transport system. 7 Rivaroxaban and apixaban are substrates for P-gp and (in part) metabolized by CYP 3A4. Subsequently, rivaroxaban and apixaban DDIs must strongly affect both P-gp and CYP 3A4; the clinician should ensure a patient is not on 2 concomitant drugs that affect CYP 3A4 and P-gp separately, as these combined DDIs could cause significant changes in DOAC concentrations. In contrast, dabigatran and edoxaban are affected only by strong inhibitors/ inducers of P-gp, as they lack metabolism by the CYP enzyme. The P-gp impact is within the gastrointestinal tract; hence, to minimize the P-gp DDI, dabigatran or edoxaban may be administered 2 hours prior to the interacting agent. 4 Notably, all DOACs have a component of renal elimination (dabigatran > edoxaban > rivaroxaban > apixaban), and while progressive renal dysfunction will result in elevated DOAC concentrations, this elimination is not a direct mechanism of DDIs. [2][3][4][5] At this time, there is limited clinical pharmacokinetic (PK)/ pharmacodynamic (PD) data to quantify the clinical impact of specific DOAC DDIs. DDIs of this nature (i.e., P-gp or CYP 450) are highly variable because of the timing of the induction/ inhibition turnover as well as the strength (mild, moderate or strong) of the interaction. 8 In addition, there is inherent intersubject variability of 30% for concentrations of dabigatran, edoxaban and apixaban, with rivaroxaban reaching 40% for PK parameters. 9 In addition, reported ranges of DOAC concentrations assessed in subgroups of clinical trials demonstrate variability in peak/trough ratios of nearly 1.6-fold. [2][3][4][5] With this in mind, DDIs that alter DOAC concentrations of 30% to 40% often still result in DOAC concentrations falling within these reported concentration ranges. Subsequently, when regulators consider providing advice surrounding DDIs, within the context of high PK/PD variability, general recommendations are often to avoid these combinations; specifically, regulators contraindicate DOACs for DDIs with inducing agents (for fear of thrombotic events) and recommend use with caution and assess other factors that may warrant avoidance when an inhibitor is the interacting culprit.
Limited, if any, data provide a comparison of DDIs between the DOACs. Unique to edoxaban are recommendations for dose reduction (from 60 to 30 mg daily) in the presence of P-gp inhibitors (except amiodarone and verapamil), with certain drugs listed based on clinical trial protocols or product monograph content. 5,10 As the front-line clinician continues to manage more complex clinical scenarios with consideration of DOAC use, a summary of available literature specific to DOAC DDIs is necessary, given there may be no or conflicting information for drug interactions. As such, our purpose is to provide a tool that differentiates DDIs across the 4 DOACs specific to agents commonly prescribed for patients with cardiovascular disease, with a description of available data to support the same.

Development of the practice tool
To create the practice tool, a systematic approach was used to collate data from both product monographs and peer-reviewed PRACTICE TOOL literature available for DDIs with the DOACs. As conflicting information was identified across multiple sources, we streamlined our approach. First, a general table of drugs known to be CYP 3A4 and P-gp inhibitors and inducers was created using data from LexiComp and was cross-checked using the Food and Drug Administration (FDA) database where inconsistencies arose. 11,12 Following this, all possible medication interactions were entered into the Lexi-Interact database-the one most commonly used by our clinical pharmacists. 13 As most information was general in nature and based on a theoretical interaction, a formal search of the literature was then completed using the OVID database searching both MEDLINE (back to 1946) and Embase (back to 1974) on May 14, 2021, using the following search strategy: search term 1: "Dabigatran or Pradaxa or Apixaban or Eliquis or Rivaroxaban or Xarelto or Edoxaban or Lixiana or DOAC* or direct oral acting anticoagulant* or NOAC* or novel oral acting anticoagulant*" and search term 2: "Drug interaction* or Drug-drug interaction* or medication interaction*". A total of 182 articles were identified and included if they demonstrated area under the curve (AUC) data or any clinical evidence (either drug concentrations or clinical outcomes) of a DDI. Among included articles, citations were also reviewed for relevant literature. Based on available data, recommendations for concomitant use with a DOAC (Table 1) were classified as follows: • Green: No interaction or clinically nonsignificant interaction-no effect on pharmacokinetics • Green/yellow: Use together with caution; limited data suggest either increased major bleeding or altered drug concentrations • Yellow: Use with caution as either: product monograph recommendation to avoid or contraindicate, implies expected drug concentrations exceed the observed and acceptable variability Inclusion of all actual or potential DDIs with DOACs was beyond the scope of our tool. As this tool was created for use by practitioners within an anticoagulation clinic having a thrombosis/cardiology-based practice, herbal supplements and drug   No interaction or clinically nonsignificant interaction-no effect on pharmacokinetics Use together with caution; limited data suggest either increased major bleeding or altered drug concentrations Use with caution as either: • a theoretical/documented interaction that would affect DOAC concentration yet in an allowable quantity, • product monograph recommendation to use with caution or • for edoxaban, recommendation to reduce dose (signified with ↓ dose) Concomitant use is not recommended; limited data may support use Avoid combination, may use only if DOAC concentrations are assessed as either: • theoretical/documented interaction that affects DOAC concentration or • product monograph recommendation to avoid or contraindicate implying expected drug concentrations due to the interaction exceed the observed and acceptable variability DOAC, direct oral anticoagulant; DDI, drug-drug interaction; R, rivaroxaban; A, apixaban; D, dabigatran; E, edoxaban; P-gp, P-glycoprotein; MB, major bleeding; GIB, gastrointestinal bleeding; PM, product monograph; ICH, intracerebral haemorrhage; SSRI, selective serotonin reuptake inhibitor; SNRI, serotonin-norepinephrine reuptake inhibitor. interaction details:

Clinical management of DOAC DDIs
To effectively manage a potential/actual DDI with a DOAC, the clinician should consider individual patient characteristics and how these may have an impact on anticipated DOAC concentrations. For patients prescribed anticoagulants, the clinician should assess the risk of clotting vs bleeding to provide a basis for comfort in having the patient's anticipated DOAC concentration on the higher vs lower end. Risk for clotting is specific to the indication for anticoagulant use; for some indications, validated risk scores are available (e.g., CHADS 2 score for nonvalvular atrial fibrillation), whereas for others, such as venous thromboembolism, clinical factors such as the proximity/extensiveness of the clot are more helpful. Specific to bleeding risk, the clinician should contemplate factors that encompass patient history of bleeding, diseases of note (e.g., esophageal varices, diffuse diverticulitis) or drugs increasing risk (e.g., concomitant antiplatelet therapy). Knowledge of renal dysfunction and the impact on DOAC concentration should also be integrated into this assessment. Once done, the clinician should extrapolate a preference for having the DOAC concentration on the high end (assuming clot risk trumps bleeding risk) or the low end (assuming the opposite).

Conclusion
This tool has been developed to assist clinicians in making decisions surrounding DOAC use. The clinician is encouraged to review the basis of the recommendation with available literature described, all drugs being administered and renal function to gauge the overall impact on DOAC concentration. With this in mind, clinical judgement should dictate practice. ■