Ten years of direct oral anticoagulants: lessons learned and challenges ahead

In the field of medicine, groundbreaking changes occur when compelling evidence on novel treatments, diagnostic procedures, or theories shatters a longstanding tradition embraced as an unwavering mantra. In the realm of antithrombotic treatments, direct oral anticoagulants (DOACs) represented a Copernican revolution in the last decade, disrupting the 50-year dominance of warfarin. In fact, warfarin, with its numerous pharmacological and food interactions and the mandatory need for regular laboratory monitoring, is unlikely to pass even phase 1 of the current extensive clinical drug development process. Nevertheless, it represents an institution in the treatment of atrial fibrillation (AF). In the last 10 years, DOACs have played a triple task: (i) providing a viable alternative and later a preferred choice over warfarin, (ii) challenging the mental taboo of requiring regular laboratory testing to guarantee the constant effectiveness of anticoagulation, and (iii) surpassing the prescriptive constraints set up by regulatory authorities. The RE-LY trial on dabigatran in 2009 started a new era of stroke prevention in patients with AF. In the early 2010s, owing to other positive large randomized trials of rivaroxaban, apixaban, and edoxaban, European guidelines stipulated to consider DOACs (2012) and then prefer them (2016) over vitamin K antagonists (VKAs) in the thromboembolic prophylaxis of patients with AF (Figure 1).1 Effect of standard dose direct oral anticoagulants vs. warfarin on gastrointestinal bleedings, ischaemic and haemorrhagic strokes, and death in patients with atrial fibrillation. Patient-level meta-analysis of four randomized trials by Carnicelli et al.1 The uptake rate of DOACs varied widely in Europe and globally after their launch in the 2010s. The main barrier to the early and broad introduction of DOACs was the marked budget impact of the wide use of DOACs that led to restrictive reimbursement policies in many countries. In England, DOACs accounted for 16% of dispensed OACs and 85% of the total cost of OACs, which was more than £190 million in 2015. The uptake of DOACs was clearly slower in Finland than in other Nordic countries (Figure 2).2 At the patient level, social determinants of health, i.e. lower socioeconomic status and belonging to a black minority, have been associated with a lower probability of receiving more expensive DOACs.3,4 The lack of an antidote and simple laboratory test to check the DOAC effect and fears of uncontrollable bleeding and rare unexpected adverse events caused concerns in the early years. These concerns were redundant, and reversal agents for DOACs (idarucizumab and andexanet alfa) became available during the latter half of the 2010s. After numerous reassuring real-life studies, DOACs have now been fully integrated into clinical practice, replacing VKAs in the management of non-valvular AF. Temporal rise in the uptake of direct oral anticoagulants over warfarin as the initial oral anticoagulant in patients with atrial fibrillation in Denmark (left panel) and Finland (middle panel). The uptake of direct oral anticoagulants was associated with increase in the initiation of anticoagulation in Finland (right panel) and in other Nordic countries in patients with CHA2DS2VA score ≥ 1. Based on Frost et al.2 Fixed dosing without laboratory monitoring saves patients’ time and travel and monitoring costs. The simple use of DOACs has facilitated the uptake of anticoagulants and helped improve the coverage of the recommended anticoagulation use (Figure 2). Direct oral anticoagulants have also helped mitigate disparities in OAC use between males and females. Furthermore, the superior efficacy and safety profile of DOACs have lowered the threshold of ischaemic stroke risk, above which OAC should be prescribed and below which anticoagulant therapy should be withheld. Direct oral anticoagulants have made life easier in many clinical scenarios. Up to 15%–20% of patients receiving chronic oral anticoagulation will require elective surgery or invasive procedures each year. However, perioperative management with VKA anticoagulation is problematic. The interruption and resumption of VKAs are complicated by the slow and variable onset and vanishing of the anticoagulative effect. Mixing heparin bridging to VKA pause causes extra hassle and is associated with increased bleeding risks. Clinical practices have been variable, and outcomes have been suboptimal, as 6% of the cases of ischaemic stroke and 2.9% of the cases of intracranial bleeding were preceded by invasive intervention during the VKA era.5 The rapid and predictable offset and onset of action of DOACs obviate the need for heparin bridging and help tailor perioperative anticoagulation according to the bleeding risk of the operation and patient characteristics. The use of current clear-cut recommendations has led to acceptable risks of perioperative bleeding and thromboembolism, as shown in the PAUSE trial.6 The slow onset of action of VKAs is also problematic when the treatment is initiated in the setting of acute cardioversion of AF and must be addressed with heparins. Direct oral anticoagulants offer a straightforward and user-friendly alternative with almost immediate effective periprocedural anticoagulation.7 In elective cardioversion, frustrating delays and costly rescheduling of cardioversions caused by subtherapeutic INRs can be avoided using DOACs. On the basis of extensive research, DOACs are currently the mainstream of long-term oral anticoagulation for patients with AF but are not a safe alternative for VKAs in the anticoagulation of patients with mechanical heart valves, left ventricular assist device, or moderate-to-severe mitral stenosis and those with antiphospholipid syndrome. Furthermore, no randomized head-to-head comparisons have been made to guide the selection among the four DOACs. A wide gap exists in the understanding of the safe use of DOACs among patients with end-stage renal failure. Similarly, the safety and dosing of different DOACs are still uncertain in advanced liver disease, and more data are needed in very old and frail patients with multiple comorbidities. Patients at extremes of weight also represent a group for whom fixed dosing raises concerns. Suboptimal DOAC adherence is a problem favoured by lack of regular clinical follow-up.8 Direct oral anticoagulants have lowered the ‘tipping point’ of stroke risk for starting anticoagulation in AF. Subclinical AF is a common finding carrying a low risk of ischaemic stroke, and more data are needed when DOACs are of net benefit for these patients.9 Even if the risk of intracerebral bleeding is lower when using DOACs than when using VKAs, gastrointestinal bleeding is a major concern during long-term treatment (Figure 1). The evaluation of bleeding risk based on the HAS-BLED score, including labile international normalized ratio (INR) from the VKA era, is suboptimal, and tailoring a better scoring system for the bleeding risk of patients receiving DOACs is needed. FXIa inhibitors are now on the horizon, but phase 3 trials are needed to document their efficacy and safety, especially in AF patients with high bleeding risks.10 All authors declare no disclosure of interest for this contribution.