Selecting appropriate DOACs for patients with AF and multiple comorbidities

Prof. Robert Giugliano
Giugliano of Brigham and Women’s Hospital
Harvard Medical School, Boston, Massachusetts, US
11 Jul 2022
Selecting appropriate DOACs for patients with AF and multiple comorbidities

Treating patients with nonvalvular atrial fibrillation (AF) and multiple comorbidities present special challenges, such as polypharmacy, and elevated risk of bleeding, stroke or systemic embolism (SE). In a recent webinar, principal investigator of the ENGAGE AF-TIMI 48 trial, Professor Robert Giugliano of Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts, US, discussed how to choose appropriate direct oral anticoagulants (DOACs; eg, edoxaban) for vulnerable patient groups on the basis of trial-level meta-analyses and drug-drug interactions (DDIs) between DOACs and commonly used drugs.

Choosing DOACs for vulnerable patients
The concept of vulnerable patients is centred around older age and multiple comorbidities, such as prior stroke or transient ischaemic attack (TIA), coronary artery disease (CAD), diabetes, and chronic kidney disease (CKD). Vulnerable patients with AF represent a particularly challenging group due to the simultaneously increased risks of thrombotic events and bleeding. [Eur Heart J Suppl 2022;24:A1-A10; Giugliano R, 2022]

Patients aged 75 years, or even older
“Most relevant to the elderly, ENGAGE AF-TIMI 48 demonstrated a strong graded relationship between increasing age and higher event rates in warfarin-treated patients. Compared with patients aged <65 years, there was a doubling in incidence of ischaemic stroke, a tripling in incidence of major bleeding, and a four-fold increase in incidence of intracranial haemorrhage [ICH] among patients ≥75 years of age [all p≤0.001],” reported Giugliano.“Clearly, patients aged ≥75 years are more prone to adverse clinical outcomes on warfarin.” [J Am Heart Assoc 2016;5:e003432]

A meta-analysis in patients aged 75 years showed similar efficacy of four DOACs in preventing stroke and SE vs warfarin, without heterogeneity across landmark studies (overall relative risk [RR], 0.70; 95 percent confidence interval [CI], 0.610.80; I2=0 percent). [Arch Gerontol Geriatr 2019;81:209-214; Arch Gerontol Geriatr 2019;83:338]

However, it showed a high degree of heterogeneity in major bleeding risk across different DOACs [I2=86 percent]. Elderly patients experienced significantly fewer major bleeding events with edoxaban 60/30 mg [RR, 0.77; 95 percent CI, 0.640.92] and apixaban 5/2.5 mg [RR, 0.74; 95 percent CI, 0.610.89] vs warfarin, noted Giugliano. In contrast, neither rivaroxaban 20/15 mg [RR, 0.95; 95 percent CI, 0.771.16] nor dabigatran [110 mg: RR, 1.45; 95 percent CI, 1.211.75; 150 mg: RR, 1.67; 95 percent CI, 1.401.99] significantly reduced major bleeding events [in this age group].

Similar safety outcomes were observed in patients aged 80 years. Dabigatran 150 mg was associated with a significantly increased risk of major bleeding (RR, 1.41; 95 percent CI, 1.021.94) while a trend toward more bleeding was observed with dabigatran 110 mg (RR, 1.18; 95 percent CI, 0.841.65) in patients aged 8084 years. In contrast, edoxaban and apixaban were associated with significantly reduced risk of major bleeding in patients aged 80 years (edoxaban 60/30 mg: RR, 0.75; 95 percent CI, 0.580.98; edoxaban 30/15 mg: RR, 0.42; 95 percent CI, 0.310.56) (apixaban 5/2.5 mg: RR, 0.66; 95 percent CI, 0.480.90). [Heart 2017;103:1015-1023; J Am Heart Assoc 2016;5:e003432; Eur Heart J 2014;35:1864-1872]

Although we have limited data in patients aged 85 years for any of the DOACs… an even greater risk reduction in major bleeding was seen with edoxaban in patients aged 85 years vs 80 years [60/30 mg: RR, 0.58; 95 percent CI, 0.350.94; 30/15 mg: RR, 0.36; 95 percent CI, 0.200.64], highlighted Giugliano. ENGAGE AF-TIMI 48 went on for nearly 3 years. If patients were 87 years of age when they entered the trial, they would be 90 years old at the end of the trial. Therefore, for very elderly patients, edoxaban is a very good choice. Even at a reduced dose, edoxaban is quite effective without significantly increasing the risk of major bleeding.

Prior ischaemic stroke or TIA
In patients with a history of TIA or stroke, all DOACs were associated with a significant reduction in stroke and SE vs warfarin, without significant heterogeneity between trials (odds ratio [OR], 0.86; 95 percent CI, 0.75–0.98; I2=0 percent). [Int J Stroke 2017;12:589-596]

However, heterogeneity in major bleeding was seen between the trials [I2=11 percent]. There was a greater reduction in risk of major bleeding with edoxaban 60/30 mg [OR, 0.82; 95 percent CI, 0.651.04] and apixaban 5/2.5 mg [OR, 0.74; 95 percent CI, 0.550.99], but the ORs for rivaroxaban 20/15 mg and dabigatran 150 mg were very close to 1.0, noted Giugliano.

Presence of CAD
In patients with AF and a history of CAD, edoxaban 60/30 mg was the only DOAC that significantly reduced the risk of stroke or SE (hazard ratio [HR], 0.65; 95 percent CI, 0.46–0.92) and cardiovascular mortality (HR, 0.69; 95 percent CI, 0.49–0.98) vs warfarin. [Eur Heart J Acute Cardiovasc Care 2019;8:554-561]

All DOACs showed a trend toward risk reduction in major bleeding, except rivaroxaban 20/15 mg (HR, 1.32; 95 percent CI, 0.961.81). Overall, edoxaban showed very good clinical outcomes in patients with concomitant CAD vs the other three DOACs, commented Giugliano.

DOACs overall significantly reduced the risk of stroke or SE vs warfarin in patients with diabetes (HR, 0.80; 95 percent CI, 0.69–0.93; I2=3.90 percent). However, when each DOAC was analyzed separately, this reduction reached statistical significance only with dabigatran 150 mg (HR, 0.61; 95 percent CI, 0.41–0.91). [Eur Heart J Cardiovasc Pharmacother 2021;7:f40-f49]

Interestingly, only edoxaban 60/30 mg significantly reduced major bleeding vs warfarin [HR, 0.79; 95 percent CI, 0.650.96] in patients with diabetes,” added Giugliano. The good safety profile of edoxaban in diabetes could be explained by a significant risk reduction in intraocular bleeding vs warfarin [HR, 0.62; 95 percent CI, 0.400.96], but not with other DOACs.[Eur Heart J Cardiovasc Pharmacother 2021;7:f40-f49; JAMA Ophthalmol 2015;133:834-839]

Moderate CKD
Patients with CKD represent a group at high risk of bleeding with warfarin. In patients with moderate CKD (ie, creatinine clearance [CrCl] 30–50 mL/min), all DOACs were at least noninferior to warfarin, but edoxaban and apixaban were the only DOACs that significantly decreased the risk of major bleeding vs warfarin (edoxaban 60/30 mg: HR, 0.76; 95 percent CI, 0.58–0.98) (apixaban 5/2.5 mg: HR, 0.50; 95 percent CI, 0.38–0.66). [J Am Coll Cardiol 2016;67:2888-2899; Circulation 2016;134:24-36; Eur Heart J 2012;33:2821-2830]

Adding polypharmacy to the mix
“Patients with AF, especially older patients, often present with multiple comorbidities in clinical practice. One of the key considerations for them is polypharmacy. It is important to be aware of DDIs as they may affect the metabolism of DOACs,” reminded Giugliano.

The 2021 European Heart Rhythm Association (EHRA) Practical Guide includes tables for DDIs with DOACs to help guide treatment decisions. After summarizing the data, Giugliano provided a simplified table of DDIs between DOACs and 19 commonly used drugs. (Table) [Europace 2021;23:1612-1676]


Among four DOACs, apixaban had the most potential concern on DDIs [OK to use or caution required, 8 out of 19; avoid or no data, 11 out of 19], while edoxaban had the least DDIs [OK to use or caution required, 13 out of 19; avoid or no data, 6 out of 19], said Giugliano. (Table)

Both apixabans and rivaroxabans metabolism relies heavily on the cytochrome P450 pathway [15 percent and 32 percent, respectively]. Thus, caution is required when prescribing these agents to patients on strong cytochrome P450 inhibitors,advised Giugliano. In contrast, no dabigatran and very little edoxaban [<4 percent] are metabolized through the cytochrome P450 system.

In addition, all DOACs interact with the P-glycoprotein (P-gp) transport system. Competitive inhibition of the P-gp pathway will result in increased DOAC levels, which needs to be considered as many drugs used in AF patients are P-gp inhibitors. Of note, only edoxaban was studied using concomitant drugs [ie, potent P-gp inhibitors, such as verapamil, quinidine, and dronedarone] as a dose reduction criterion in a phase III trial,” added Giugliano. [Europace 2021;23:1612-1676]

Why does polypharmacy matter?
DDIs could lead to higher levels of DOACs, and thus increase the risk of bleeding. [Pharmaceutics 2022;14:1120] “Analyses showed that greater number of comedications was associated with less protection from major bleeding with apixaban [pinteraction=0.017] and rivaroxaban [pinteraction=0.0074],” highlighted Giugliano. “However, this pattern was not observed with edoxaban. The risk of major bleeding was steady and tended to be reduced with edoxaban vs warfarin, regardless of the number of comedications [pinteraction=0.83].” (Figure)


“Unlike apixaban and rivaroxaban, there is no significant metabolism of edoxaban by the cytochrome P450 system. Besides, edoxaban is the only DOAC that offers dose adjustment recommendations for strong cardiac P-gp inhibitors, which may explain the consistent reduction in bleeding risk with edoxaban vs warfarin, regardless of the class of comedication used,” Giugliano explained.

“In patients with high bleeding risk, edoxaban and apixaban would be my first choices,” said Giugliano. Only these two agents were associated with lower risks of major bleeding vs warfarin in patients with AF and other conditions, including older age, prior stroke or TIA, CAD, and CKD, whereas no such trends were observed with rivaroxaban and dabigatran. Importantly, only edoxaban significantly reduced the risk of major bleeding vs warfarin in patients with diabetes.

“When selecting DOACs with similar efficacy and safety profile for vulnerable patients with AF, DDI is an important factor to consider [as polypharmacy is common],” said Giugliano. “Edoxaban maintains its safety profile under polypharmacy and has the least DDIs, which may be attributed to its minimal interaction with the cytochrome P450 system and unique dose reduction criteria for strong P-gp inhibitors. In addition, edoxaban is recommended for patients with a strong preference for once-daily dosing.”

Related MIMS Drugs

Editor's Recommendations