Use of Dabigatran With Antiretrovirals

J Perram; E O'Dwyer; C Holloway


HIV Medicine. 2019;20(5):344-346. 

Objectives: Theoretical and untested interactions between antiretrovirals and direct-acting oral anticoagulants have limited the use of this new class of anticoagulant in people with HIV infection. This case series, the first of its kind, reports on the successful concurrent use of the direct-acting oral anticoagulant dabigatran and antiretroviral therapy.

Methods: This series involved 14 patients requiring anticoagulation for management of atrial fibrillation, who were either unable or unwilling to take warfarin, and who were receiving concurrent treatment for HIV infection. Participants were treated with dabigatran with dose monitoring to establish the safety and efficacy of concurrent use with antiretrovirals. All were commenced on 110 mg twice daily, increased to 150 mg twice daily if the trough level was < 69.3 ng/mL.

Results: In the 14 patients treated with dabigatran and antiretrovirals, there were no thromboembolic or bleeding complications. Dabigatran treatment was discontinued in one patient because of undetectable dabigatran levels despite dose escalation. Dabigatran levels fell within the fivefold variance seen in the Randomized Evaluation of Long-Term Anticoagulation Therapy (RE-LY) study at a dose of either 110 or 150 mg twice daily.

Conclusions: This case series represents the largest published population to date successfully receiving antiretroviral and direct-acting oral anticoagulant therapy. Given the significant health care burden faced by people living with HIV, the availability of safe anticoagulant therapy without the requirement for monitoring is an important option in this patient population.

A large US study published in 2013 confirmed the incident rate of atrial fibrillation (AF) to be 2.1% in people newly diagnosed with HIV infection, similar to rates of AF in the general population.[1,2] A subsequent analysis of the US Veterans Aging Cohort Study data (n = 76 835) confirmed that the risk of ischaemic stroke is higher among people living with HIV,[3] suggesting a greater need for treatment in this population. Indeed, stroke is the second most common cause of mortality among those with HIV infection.[4] Until recently, warfarin was the mainstay of management to reduce the risk of cardioembolic stroke; however, there has been significant evolution in treatment options over the past two decades.

Since the first direct-acting oral anticoagulants (DOACs) were licensed, they have been increasingly prescribed in first-line treatment as anticoagulants for stroke prevention in individuals with nonvalvular AF.[5] Many benefits of DOACs over vitamin K antagonists (VKAs) are espoused, most notably freedom from therapeutic monitoring, but also lower rates of serious bleeding and fixed dosing. There are of course disadvantages as well – less familiarity with adverse effects, drug interactions, lack of reversibility (which is being overcome with the development of antidotes) and cost.[6]

Drug interactions are a recurrent issue in the management of chronic diseases in people taking antiretroviral medications. Some registration trials for investigational new drugs exclude patients taking cytochrome P450 (CYP450) inducers/inhibitors, while others do not analyse the effects of concurrent use. The metabolism of warfarin relies on several CYP450 enzymes, including CYP2C9, CYP2C19 and CYP3A4. This has been a source of a plethora of drug interactions with antiretroviral agents, as well as other medication classes important in the management of HIV-associated complications, such as anti-fungals and antibiotics.[7] This issue has, however, been mitigated by the ability to monitor and titrate warfarin using the international normalized prothrombin ratio (INR).

Egan et al. published an evaluation of likely interactions between DOACs and antiretrovirals in 2014. This review predicted that dabigatran was the least likely of the DOACs to interact with common antiretroviral combinations, as a consequence of the fact that it is predominantly renally cleared rather than relying on the CYP450 system for metabolism.[8] By contrast, the factor Xa inhibitors rivaroxaban and apixaban are substrates of CYP3A4, and thus are expected to interact with protease inhibitors (PIs) and nonnucleoside reverse transcriptase inhibitors (NNRTIs).[8] Based on this theoretical information, and following our early experiences,[9] we have now managed 14 patients on antiretroviral therapy with dabigatran.

Our cases series involved 14 male patients with an average age of 64 years (range 50–81 years) and a mean a clinical prediction tool used to guide the need for anticoagulation in people with atrial fibrillation (CHADS-VASc) score of 2.5 (range 0–6). Viral load was below the limit of detection in nine patients, and CD4 count was > 400 cells/μL in nine of the 12 patients for whom results were available. All patients had normal liver function tests (defined as transaminases within twofold of the upper limit of normal). Three patients had moderate renal impairment, with an estimated glomerular filtration rate (eGFR) of 45–60 mL/min/1.73 m2, with the remainder having an eGFR > 60 mL/min/1.73 m2.

Patients taking dabigatran 110 mg twice daily (bid) had a mean dabigatran trough level (at 20–24 h post dose) of 48 ng/mL (range 9–222 ng/mL), compared to 65 ng/mL in the Randomized Evaluation of Long-Term Anticoagulation Therapy (RE-LY) study. The mean peak dabigatran level at 4 h post dose was 125 ng/mL (range 20–417 ng/mL), compared to 126 ng/mL in RE-LY[10] (Figure 1). Timing of collection of peak and trough levels after dosing was later in our patients than in the RE-LY study; however, given that drug levels varied by up to fivefold in RE-LY, our results sit well within the wide range of normal seen in registration trial data.

Figure 1.

Comparison between mean dabigatran levels in the Randomized Evaluation of Long-Term Anticoagulation Therapy (RE-LY) cohort and in our series of patients on antiretroviral therapy (ART). bid, twice daily.

Based on available evidence at the time, the dose was escalated to 150 mg bid in the five patients with a trough dabigatran level < 69.3 ng/mL.[11] Despite escalation, the trough concentration remained < 40 ng/mL in 60% of this group, with a mean peak level of 76 ng/mL.

All participants were receiving treatment with antiretroviral medications. The majority were on regimens including nucleoside reverse transcriptase inhibitors (NRTIs), NNRTIs, integrase inhibitors (IIs), PIs and/or C-C chemokine receptor type 5 (CCR5) receptor blockers. Two patients were using cobicistat to boost elvitegravir. A theoretical minor interaction exists between the CYP3A inhibitor cobicistat and dabigatran, which is minimally metabolized via CYP3A. In one of two patients taking this combination, a dose reduction was required from 110 mg bid (trough 222 ng/mL; peak 338 ng/mL) to 75 mg bid (trough 46 ng/mL; peak 192 ng/mL). The second patient taking cobicistat had an undetectable plasma dabigatran level despite escalation to 150 mg bid. The bottom 10th centile of people in registration trials had low dabigatran levels (< 28.2 ng/mL at 110 mg bid and < 39.8 ng/mL at 150 mg bid), with a 50% increase in stroke or embolic event at 28 ng/mL compared with the mean trough of 59 ng/mL.[10]

The value of dabigatran levels is limited, with the finding of a fivefold variance in level within that study population. The importance of determining dose in the context of patient age and renal function, rather than in response to drug level, was highlighted at the time of publication of the RE-LY data.[10] This has been an area of some controversy, given the correlation between level and bleeding events; however, on balance, regulators have agreed that drug levels should not generally direct patient management.

Prolonged thrombin time can be used as a surrogate for the presence of dabigatran in the acute setting; however, the actual result is poorly correlated with the anticoagulant effect and dabigatran level. Among our cohort taking dabigatran 110 mg bid, thrombin time was prolonged for all patients with a result (n = 13), ranging from 13 to 129 s (upper limit of normal 12 s).

In this 14 patient series, no adverse bleeding or thromboembolic events have occurred to date, with an average follow-up of 12 months (range 2–40 months). While numbers treated remain small, this represents the largest series to date treated concurrently with a DOAC and antiretroviral therapy. Although dabigatran dose monitoring is not recommended in the general population, the absence of data on pharmacological interactions makes it a useful tool in populations at risk of drug interactions. In this series, dabigatran levels remained within the population expected range in patients taking a variety of antiretroviral medications, adding weight to previous suggestions that dabigatran may be worthy of consideration as a warfarin alternative in patients with nonvalvular AF requiring anticoagulation who are concurrently taking antiretroviral agents.