The evolution of anticoagulant therapy

Abstract

Arterial and venous thromboembolism are leading causes of morbidity and mortality around the world. For almost 70 years, heparins (unfractionated heparin and low molecular weight heparins) and vitamin K antagonists have been the leading therapeutic medical options for the treatment and prevention of thromboembolic disorders. Nevertheless, the many limitations of these traditional anticoagulants have fuelled the search for novel agents over the past 15 years, and a new class of oral anticoagulants that specifically target activated factor X and thrombin has been developed and is now commercially available. In this narrative review, the evolution of anticoagulant therapy is summarised, with a focus on newer oral anticoagulants.

Introduction

Thromboembolic diseases are the leading cause of death and disability in high-income countries, and their incidence is also dramatically increasing in middle- and low-income countries¹. While arterial clots (platelet-rich and fibrin-poor, the so-called white clots) are usually generated at sites of vascular injury under high shear rates and are responsible for myocardial infarction and stroke, venous clots (fibrin- and red blood cell-rich and platelet-poor, the so-called red clots) cause venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE)^2,3.

The burden of thromboembolic diseases imposes primary prevention as the most important community goal, through lifestyle interventions on a number of modifiable risk factors such as smoking, hypertension, abdominal obesity, physical inactivity and inappropriate dietary habits. Despite the positive outcome associated with these measures, many people still develop clinical manifestations of thromboembolism, so that a pharmacological approach to prevention and treatment is necessary to reduce morbidity and mortality. Since blood hypercoagulability plays a pivotal role in thrombogenesis, it is reasonable that anticoagulant agents should be regarded as an essential therapeutic tool in the management of these patients⁴.

Heparins (unfractionated heparin [UFH] first, and low molecular weight heparins [LMWH] subsequently) and vitamin K antagonists (VKA: warfarin, phenprocoumon, acenocoumarol) have been used for decades for the treatment and prevention of thromboembolism^5,6. Over the past 15 years, however, the interest in anticoagulants has grown dramatically, as shown by the increasing number of drugs in both preclinical and clinical development as well as by the vast array of anticoagulants currently licensed (Table I). In particular, investigators are concentrating research on the so-called new direct oral anticoagulants (DOAC), which selectively target specific steps of the coagulation cascade^7–10. This narrative review summarises the evolution of anticoagulant therapy, focusing on advantages and disadvantages of traditional and newer anticoagulants.

Table I.

Currently available anticoagulant agents.

Anticoagulants
Parenteral	Oral
Heparin (UFH, LMWH)	VKA
Fondaparinux	Thrombin inhibitors (dabigatran)
Thrombin inhibitors (bivalirudin, argatroban)	Factor Xa inhibitors (rivaroxaban, apixaban, edoxaban)

UFH: unfractionated heparin; LMWH: low molecular weight heparin; VKA: vitamin K antagonists.

Search methods

We reviewed the medical literature for published clinical trials evaluating the efficacy and safety of anticoagulants for the prevention and treatment of thromboembolism. The MEDLINE^® electronic database was searched without temporal limits using an English language restriction. The Medical Subject Heading and keywords used were the following: “anticoagulants”, “heparin”, “unfractionated heparin”, “low molecular weight heparins”, “vitamin K antagonists”, “warfarin”, “new oral anticoagulants”, “novel oral anticoagulants”, “direct oral anticoagulants”, “target specific oral anticoagulants”, “non-vitamin K antagonist oral anticoagulants”, “dabigratan”, “rivaroxaban”, “apixaban”, “edoxaban”, “myocardial infarction”, “stroke”, “atrial fibrillation”, “venous thromboembolism”, “pulmonary embolism”, “deep vein thrombosis”, “primary prophylaxis”, “secondary prophylaxis”, “therapy”, “bleeding”, “survival”, “death”. We also screened the reference lists of the most relevant reviews for further eligible studies not captured in our initial literature search. Search terms were also applied to abstracts from the latest international haematology congresses on haemostasis and thrombosis.

Conventional anticoagulant agents

Conventional anticoagulant agents are mainly heparins (UFH and LMWH) and VKA (Figure 1 for their mechanisms of action).

Figure 1.

Biological targets of anticoagulant agents.

FVII: factor VII; TF: tissue factor; FXI: factor XI; FXIa: activated factor XI; FIX: factor IX; FIXa: activated factor IX; FX: factor X; FXa: activated factor X; VKA: vitamin K antagonist; F: factor; LMWH: low molecular weight heparin; TF: tissue factor; UFH: unfractionated heparin; FII: factor II; FIIa: activated factor II.

Unfractionated heparin and low molecular weight heparins

UFH was discovered a century ago, in 1914, by the medical student Jay McLean at John Hopkins in Baltimore, and was then introduced in clinical practice in the 1940s (Table II). UFH is a glycosaminoglycan that requires a cofactor to produce anticoagulant activity (indirect-acting anticoagulant)^11,12. UFH, which is administered subcutaneously or intravenously, binds to antithrombin and increases this latter’s ability to inactivate thrombin, factor Xa and factor IXa to a minor extent¹¹. Only UFH chains of at least 18 saccharide units (corresponding to a molecular weight of about 5,400 Daltons) can facilitate the interaction between antithrombin and thrombin. The activated partial thromboplastin time must be monitored in order to assess the anticoagulant effect of UFH, which is associated with an increased risk of developing heparin-induced thrombocytopenia and osteoporosis in patients undergoing long-term therapy¹³. LMWH are derived from UFH by various chemical or enzymatic depolymerisation processes, and have a mean molecular weight of about one-third that of UFH¹⁴. The currently licensed LMWH, which are administered subcutaneously, include enoxaparin, dalteparin, nadroparin, tinzaparin, certoparin, reviparin, ardeparin and bemiparin¹⁵. UFH and LMWH have been consistently found to reduce VTE complications in hip or knee arthroplasty and in the setting of high-risk medical conditions (heart failure, acute inflammatory diseases, prolonged immobilisation in bed) by approximately 60%. They are also administered, along with dual antiplatelet therapy (e.g., aspirin and clopidogrel) in patients with acute coronary syndromes, whether or not managed with revascularisation treatment. Since their introduction in the 1980s, LMWH have gradually replaced UFH for most clinical indications due to several advantages over other anticoagulant agents (see below)¹⁵. In addition, thanks to their high efficacy and safety profile, LMWH are currently recommended as the treatment of choice for acute and long-term management of cancer-associated VTE^16,17. Indeed, besides reducing mortality and morbidity related to VTE in cancer patients, accumulating experimental and clinical data suggest that LMWH significantly improve overall survival by a direct effect on the development and metastatisation of cancer itself^18,19.

Table II.

The history of evolution of anticoagulant therapy.

Year	Anticoagulant drug
1940s	Unfractionated heparin
1950s	Warfarin
1980s	Low molecular weight heparins
1990s	Parenteral direct thrombin inhibitors
2002	Fondaparinux
2010	Dabigatran
2011	Rivaroxaban
2012	Apixaban
2014	Edoxaban

Vitamin K antagonists

Warfarin, the most commonly used VKA, is an oral drug that exerts its anticoagulant activity by interfering with a post-translational modification of several coagulation (factors II, VII, IX, X) and anticoagulation proteins (protein C and S)^20,21. VKA have a narrow therapeutic range, frequently interact with food and other drugs, and their metabolism is genetically determined (by two gene polymorphisms: VKORC1, involved in the vitamin K cycle, and CYP2C9, involved in warfarin metabolism)²¹. Frequent laboratory monitoring of the international normalised ratio (INR) and dose adjustments are, therefore, needed when using these drugs. The benefits of warfarin therapy in a wide array of thromboembolic disorders have been well established. VKA, which were clinically developed more than 60 years ago, are effective at reducing the recurrence of VTE by more than 90%, and cardioembolic stroke in non-valvular atrial fibrillation by approximately 60%^21,22. In addition, a recent systematic review and meta-analysis reported that genotype-guided initial VKA dosing is effective at decreasing the risk of major bleeding events, the main drawback of VKA therapy, in approximately 50% of cases²³.

Other anticoagulants

Fondaparinux, originally developed in 2002, is a synthetic analogue of the antithrombin-binding pentasaccharide found in UFH and LMWH which selectively inhibits factor Xa in an antithrombin-dependent manner due to is small size^24,25. This highly effective drug with a good safety profile is currently licensed for the prophylaxis and treatment of VTE. Thanks to its complete bioavailability after subcutaneous injection and a plasma half-life of approximately 17 hours, fondaparinux is administered once daily at a fixed dose with no need for laboratory monitoring^10,25. Parenterally administered direct thrombin inhibitors, which do not require antithrombin for inhibition since they directly inactivate both free and fibrin-bound thrombin, include argatroban and bivalirudin²⁶. Argatroban, a small competitive inhibitor of thrombin, has been approved for treatment of patients with heparin-induced thrombocytopenia, whereas bivalirudin, an analogue of hirudin, has been licensed as an alternative to UFH in patients undergoing percutaneous coronary intervention^11,27.

Limitations of conventional anticoagulants

In spite of the excellent clinical results obtained with traditional anticoagulants, there is some wriggle room for the improvement of thromboembolism therapy. The advantages of LMWH over UH include a longer biological half-life and more predictable dose-response, which allows weight-adjusted fixed dosages, less strict requirements for routine coagulation monitoring, and lower binding to platelet factor 4 and bone cells, which results in decreased risks of heparin-induced thrombocytopenia and osteoporosis²⁸. Due to the combination of similar (or even greater) efficacy and safety coupled with a number of other advantages, LMWH have progressively replaced UFH in clinical practice. Nonetheless, the use of LMWH is still associated with a risk of heparin-induced thrombocytopenia (albeit lower than that seen with UFH), and the need for parenteral administration limits their long-term use in the outpatient setting²⁹. Similarly, although the benefits of VKA are well established in a wide spectrum of thromboembolic disorders⁴, their use is hampered by several drawbacks, such as delayed onset and offset of action, suboptimal adherence to therapy, narrow therapeutic range of clinical effectiveness, genetic variations of metabolism plus food and drug interactions, which necessitate frequent monitoring and dose adjustment³⁰.

New oral anticoagulant drugs

The limitations of heparins and warfarin led to the development of new anticoagulant agents selectively targeting specific steps in the coagulation cascade which, besides having a high efficacy and safety profile, have the advantage of being orally administered at fixed dosages with a lower need for laboratory monitoring. Two types of DOAC are currently licensed for use in thromboembolic disorders: factor Xa inhibitors (apixaban, edoxaban and rivaroxaban) and the thrombin inhibitor dabigatran³¹. The characteristics and mechanisms of action of these agents are described in Table III and Figure 1, respectively. The main results of phase III randomised trials with DOAC are reported in Table IV.

Table III.

Characteristics of novel oral anticoagulants.

Characteristics		Direct thrombin inhibitor		Factor Xa inhibitors
		Dabigratan	Apixaban	Edoxaban	Rivaroxaban
Bioavailability (%)		3–7	50	62	80
Time to peak concentration (hours)		1–3	1–3	1–3	2–4
Half-life (hours)		12–17	8–15	8–10	7–13
Renal clearance (%)		80	25	35	33
Dosing regimen		110–150 mg twice daily	2.5–5 mg twice daily	15–30 mg once daily	10–20 mg once daily; 15 mg once or twice daily
Metabolism		P-glycoprotein	P-glycoprotein, CYP3A4	P-glycoprotein, CYP3A4	P-glycoprotein, CYP3A4
Approved indications	Non-valvular AF	North America, Europe	North America, Europe	United States	North America, Europe
	VTE treatment	United States, Europe	United States, Europe	Unites States	North America, Europe
	VTE prevention	Canada, Europe	Canada, Europe	Japan	North America, Europe

AF: atrial fibrillation; VTE: venous thromboembolism.

Table IV.

Main results of the phase III trials with new oral anticoagulants.

Drug	Studyref	Indication	Pts	Study arms (drug vs comparator)	Primary outcome
					Efficacy	Safetya
Dabigatran	RE-MODEL34	VTE prophylaxis after TKR	2,076	DAB 150 mg od or 220 mg od, 6–10 days	DAB 150 mg 40.5% (p=0.017)b,c	DAB 150 mg 1.3% (p=1.0)
				ENX 40 mg od, 6–10 days	DAB 220 mg 36.4% (p=0.0003)b	DAB 220 mg 1.5% (p=0.82)
					ENX 37.7%	ENX 1.3%
	RE-NOVATE35	VTE prophylaxis after THR	3,494	DAB 150 mg od or 220 mg od, 28–35 days	DAB 150 mg 8.6% (p<0.0001)b,c	DAB 150 mg 1.3% (p=0.60)
				ENX 40 mg od, 28–35 days	DAB 220 mg 6.0% (p<0.0001)b	DAB 220 mg 2.0% (p=0.44)
					ENX 6.7%	ENX 1.6%
	RE-MOBILIZE36	VTE prophylaxis after TKR	2,715	DAB 150 mg od or 220 mg od, 28–35 days	DAB 150 mg 33.7% (p=0.0009)c,d	DAB 150 mg 0.6% (p=NS)
				ENX 30 mg td, 12–15 days	DAB 220 mg 31.7% (p=0.002)d	DAB 220 mg 0.6% (p=NS)
					ENX 25.3%	ENX 1.4%
	RE-COVER37	VTE treatment	2,564	DAB 150 mg td, 6 months	DAB 2.4% (HR: 1.10, CI: 0.65–0.84)b,e	DAB 1.6% (HR: 0.83, CI: 0.45–1.48)
				WARf, 6 months	WAR 2.1%	WAR 1.9%
	RE-COVER II38	VTE treatment	2,568	DAB 150 mg td, 6 months	DAB 2.3% (HR: 1.08, CI: 0.64–1.8)b,e	DAB 1.2% (HR: 0.69, CI: 0.36–1.32)
				WARf, 6 months	WAR 2.2%	WAR 1.7%
	RE-MEDY39	VTE treatment	2,856	DAB 150 mg td, 36 months	DAB 1.8% (p=0.03)b,e	DAB 0.9% (p=0.06)
				WARf, 36 months	WAR 1.3%	WAR 1.8%
	RE-SONATE39	VTE treatment	1,343	DAB 150 mg td, 6 months	DAB 0.4% (p<0.0001)e	DAB 0.3% (p=0.996)
				Placebo, 6 months	Placebo 5.6%	Placebo 0%
	RE-LY40	Stroke prophylaxis NAF	18,113	DAB 110 mg td or 150 mg td, 2 years	DAB 110 mg 1.5%/year (p<0.001)b,g	DAB 110 mg 2.7%/year (p=0.003)
				WARf, 2 years	DAB 150 mg 1.1%/year (p<0.001)h	DAB 150 mg 3.1%/year (p=0.31)
					WAR 1.7%/year	WAR 3.4%/year
	RE-LYABLE43	Stroke prophylaxis NAF	5,851	DAB 110 mg td, 2.3 years	DAB 110 mg 1.6%/year (HR: 0.91, CI: 0.69–1.20)g	DAB 110 mg 3.0%/year (HR: 1.26, CI: 1.04–1.53)
				DAB 150 mg td, 2.3 years	DAB 150 mg 1.5%/year	DAB 150 mg 3.7%/year
Apixaban	ADVANCE-146	VTE prophylaxis after TKR	3,195	APX 2.5 mg td, 10–14 days	APX 9.0% (p=0.06)b,c	APX 0.7% (p=0.03)
				ENX 30 mg td, 10–14 days	ENX 8.8%	ENX 1.4%
	ADVANCE-247	VTE prophylaxis after TKR	3,057	APX 2.5 mg td, 10–14 days	APX 15.0% (p<0.0001)h	APX 0.6% (p=0.3014)
				ENX 40 mg od, 10–14 days	ENX 24.0%	ENX 0.9%
	ADVANCE-348	VTE prophylaxis after THR	5,407	APX 2.5 mg td, 32–38 days	APX 1.4% (p<0.001)h	APX 0.8% (p=0.54)
				ENX 40 mg od, 32–38 days	ENX 3.9%	ENX 0.7%
	AMPLIFY50	VTE treatment	5,385	APX 10 mg td (7 d), 5 mg td (6 months)	APX 2.3% (p<0.001)b,e	APX 0.6% (p<0.001)
				ENX 1 mg/kg td (>5 d), WAR (6 months)f	ENX/WAR 2.7%	ENX/WAR 1.8%
	AMPLIFY-EXT51	VTE treatment	2,486	APX 2.5 mg td or 5 mg td (12 months)	APX 1.7% (p<0.001)e,h	APX 2.5 mg 0.2%, APX 5 mg 0.1%
				Placebo	Placebo 8.8%	Placebo 0.5% (p=NS)
	ARISTOTLE52	Stroke prophylaxis NAF	18,201	APX 5 mg td, 1.8 years	APX 1.3%/year (p=0.01)g,h	APX 2.1%/year (p<0.001)
				WARf, 1.8 years	WAR 1.6%/year	WAR 3.1%/year
	AVERROES53	Stroke prophylaxis NAF	5,599	APX 5 mg td, 1.1 years	APX 1.6%/year (p=0.001)g,h	APX 1.4%/year (p=0.57)
				Aspirin (81–324 mg/d), 1.1 years	WAR 3.7%/year	WAR 1.2%/year
Edoxaban	STARS-E355	VTE prophylaxis after TKR	716	EDX 30 mg od, 11–14 days	EDX 7.4% (p=0.010)c,h	EDX 1.1% (p=0.373)
				ENX 20 mg td, 11–14 dats	ENX 13.9%	ENX 0.3%
	STARS-J556	VTE prophylaxis after THR	264	EDX 15 mg od or 30 mg od, 11–14 days	EDX 15 mg 3.8% (p=1.000)c	EDX 15 mg 2.2% (p=1.000)
				ENX 20 mg td, 11–14 days	EDX 30 mg 2.8%	EDX 30 mg 1.2%
					ENX 4.1%	ENX 2.3%
	Hokusai-VTE57	VTE treatment	8,292	EDX 60 mg od or 30 mg od, 3–12 months	EDX 3.2% (p<0.001)b,e	EDX 8.5% (p=0.004)h
				WARf, 3–12 months	WAR 3.5%	WAR 10.3%
	ENGAGE-AF-TIMI 4858	Stroke prophylaxis NAF	21,105	EDX 60 mg od or 30 mg od, 2.8 years	EDX 60 mg 1.6%/year (p<0.001)b,g	EDX 60 mg 2.7%/year (p<0.001)
				WARf, 2.8 years	EDX 30 mg 1.2%/year (p=0.005)b	EDX 30 mg 1.6%/year (p<0.001)
					WAR 1.5%/year	WAR 3.4%/year
Rivaroxaban	RECORD-1 59	VTE prophylaxis after THR	4,541	RVX 10 mg od, 35 days	RVX 1.1% (p<0.001)c,h	RVX 0.3% (p=0.18)
				ENX 40 mg od, 35 days	ENX 3.7%	ENX 0.1%
	RECORD-2 60	VTE prophylaxis after THR	2,509	RVX 10 mg od, 35 days	RVX 2.0% (p<0.001)c,h	RVX 0.08% (p=NS)
				ENX 40 mg od, 10–14 days	ENX 9.3%	ENX 0.08%
	RECORD-3 61	VTE prophylaxis after TKR	2,531	RVX 10 mg od, 10–14 days	RVX 9.6% (p<0.001)c,h	RVX 0.6% (p=0.77)
				ENX 40 mg od, 10–14 days	ENX 18.9%	ENX 0.5%
	RECORD-4 62	VTE prophylaxis after TKR	3,148	RVX 10 mg od, 10–14 days	RVX 6.9% (p=0.012)c,h	RVX 0.7% (p=0.11)
				ENX 30 mg td, 10–14 days	ENX 10.1%	ENX 0.3%
	EINSTEIN-DVT64	VTE treatment	3,449	RVX 15 mg td (3 weeks), 20 mg od (3.6 or 12 months)	RVX 2.1% (p<0.001)b,e	RVX 8.1% (p=0.77)
				ENX 1 mg/kg td (≥5 days), WAR (3.6 or 12 months)f	ENX/WAR 3.0%	ENX/WAR 8.1%
	EINSTEIN-PE65	VTE treatment	4,832	RVX 15 mg td (3 weeks), 20 mg od (3.6 or 12 months)	RVX 2.1% (p=0.003)b,e	RVX 1.1% (p=0.003)
				ENX 1 mg/kg td (>5 days), WAR (3.6 or12 months)f	ENX/WAR 1.8%	ENX/WAR 2.2%
	EINSTEIN-Extension64	VTE treatment	1,197	RVX 20 mg od, 6–12 months	RVX 1.3% (p<0.001)c,h	RVX 0.7% (p=0.11)
				Placebo, 6–12 months	Placebo 7.1%	Placebo 0%
	ROCKET-AF67	Stroke prophylaxis NAF	14,264	RVX 20 mg od, 20 months	RVX 1.7%/year (p<0.001)b,g	RVX 5.6%/year (p=0.576)
				WARf, 20 months	WAR 2.2%/year	WAR 5.4%/year

VTE: venous thromboembolism; od: once daily; td: twice daily; NS: not significant; TKR: total knee replacement; THR: total hip replacement; DAB: dabigatran; ENX: enoxaparin; WAR: warfarin; HR: hazard ratio; CI: confidence interval; NAF: non-valvular atrial fibrillation; APX: apixaban; EDX: edoxaban; HFR: hip fracture surgery; RVX: rivaroxaban.

^amajor or clinically relevant non-major bleeding;

^bnon-inferiority;

^cVTE or related death;

^dinferiority;

^erecurrent VTE;

^ftarget INR 2.0–3.0;

^gstroke or systemic embolism;

^hsuperiority.

Direct thrombin inhibitors

Dabigatran etexilate, the only oral direct thrombin inhibitor currently licensed, is a prodrug that is rapidly converted into the active form dabigatran once absorbed from the gastrointestinal tract³². A pooled analysis of three trials³³, RE-MOBILIZE, RE-MODEL and RENOVATE^34–36, showed that dabigatran was at least as effective as enoxaparin for thromboprophylaxis after hip and knee replacement, and was associated with a similar incidence of major bleeding (1.4% in the enoxaparin group vs 1.4% in the dabigatran 220 mg group and 1.1% in the dabigatran 150 mg group)³³. The administration of this drug also produced similar effects as warfarin for acute management (RE-COVER and RE-COVER II trials)^37,38 and extended maintenance therapy of VTE (RE-MEDY trial)³⁸. A pooled analysis of the RE-COVER and RE-COVER II trials showed that the incidence of any bleeding, as well as that of major or clinically relevant non-major bleeding, was significantly lower in the dabigatran group than in the warfarin group³⁸. In the RE-SONATE trial³⁹, which assessed safety and efficacy of dabigatran vs placebo for extended treatment of VTE, a 92% relative risk (RR) reduction of recurrent VTE was observed in favour of dabigatran, with a similarly low bleeding risk. Dabigatran was non-inferior (110 mg twice daily) or superior (150 mg twice daily) to warfarin for stroke prevention in atrial fibrillation (RE-LY trial)⁴⁰. These four randomised trials (i.e., RE-COVER, RE-COVER II, RE-MEDY and RE-LY) along with the PETRO trial⁴¹ (which evaluated the efficacy of dabigatran with or without aspirin vs warfarin alone in patients with non-valvular atrial fibrillation) were included in a recent meta-analysis⁴², which reported that the risk of any bleeding with dabigatran was lower than with warfarin across all the five randomised trials, with a pooled RR of 0.77 (95% confidence interval [95% CI]: 0.64–0.93). A long-term, multicentre extension of dabigatran treatment in patients who completed RE-LY (RELY-ABLE) reported no significant difference in stroke or mortality with the two dabigatran doses (150 mg twice daily vs 110 mg twice daily), although a higher rate of major bleeding was found with the higher dabigatran dose during the additional 2.3 years of treatment⁴³. Finally, a Cochrane systematic review and meta-analysis including eight randomised controlled trial involving a total of 27,557 patients with non-valvular atrial fibrillation reported that dabigatran was non-inferior or superior (150 mg twice daily) with regards to the composite outcome of vascular mortality and ischaemic events with fewer major haemorrhagic events⁴⁴.

Factor Xa inhibitor

Apixaban acts by reversibly blocking factor X at the active site (Table III)⁴⁵.

A meta-analysis of three large phase III trials on the prevention of VTE after orthopaedic surgery (ADVANCE-1, ADVANCE-2 and ADVANCE-3)^46–48 showed that apixaban 2.5 mg twice daily was associated with a significant reduction in the rate of total VTE, all-cause mortality and a significantly lower risk of clinically relevant bleeding compared to enoxaparin⁴⁹. Apixaban (10 mg twice daily for 7 days followed by 5 mg twice daily for 6 months) was also non-inferior to conventional therapy with enoxaparin/warfarin for the treatment of acute VTE in the AMPLIFY trial⁵⁰, and was associated with a significant reduction in major bleeding. One-year extended anticoagulation with apixaban (2.5 mg and 5 mg twice daily) lowered the risk of recurrent VTE compared with placebo, without increasing the incidence of major bleeding (AMPLIFY-EXT)⁵¹. A phase III trial (ARISTOTLE) compared apixaban (5 mg twice daily) with warfarin for cardioembolic prophylaxis in patients with atrial fibrillation⁵², showing that the former drug was superior to warfarin for the prevention of stroke or systemic embolism, caused less bleeding and was ultimately associated with lower mortality. In the AVERROES trial, patients with atrial fibrillation who had failed or were unsuitable for VKA treatment were randomised to aspirin or apixaban (5 mg twice daily)⁵³. Apixaban was associated with a greater reduction of stroke, whereas the rate of major bleeding was similar for the two groups. Edoxaban inhibits factor Xa activity following rapid absorption from the gastrointestinal tract (Table III)⁵⁴. Two recently published phase III randomised trials comparing edoxaban vs enoxaparin for thromboprophylaxis after total knee (STARS-E3)⁵⁵ or hip (STARS-J5)⁵⁶ replacement surgery demonstrated that edoxaban had a similar (STARS-J5) or superior (STARS-E3) efficacy to enoxaparin, while displaying a comparable safety profile. The Hokusai-VTE was the largest phase III study ever conducted in acute symptomatic VTE, with 8,292 patients randomised to receive edoxaban (60 or 30 mg once daily) or warfarin⁵⁷. The study showed that edoxaban, administered once daily after initial heparin was non-inferior to standard therapy with warfarin after initial heparin treatment, with significantly less major and clinically relevant non-major bleeding. Finally, the antithrombotic effect of edoxaban (30 mg and 60 mg once daily) vs warfarin was explored in the phase III trial ENGAGE AF-TIMI 48⁵⁸. Both once-daily regimens of edoxaban were non-inferior to warfarin for the prevention of stroke or systemic embolism, and were also associated with significantly lower rates of bleeding and death from cardiovascular causes.

Rivaroxaban is a selective oral direct factor Xa inhibitor partially excreted (33%) by the kidneys (Table III). Four phase III randomised studies compared oral rivaroxaban (10 mg once daily) with enoxaparin (40 mg once daily or 30 mg twice daily) for the prevention of VTE after total hip (RECORD-1 and RECORD-2)^59,60 or knee (RECORD-3 and RECORD-4)^61,62 arthroplasty. A pooled analysis of these four trials showed that rivaroxaban significantly reduced the incidence of the composite end-point of VTE and all-cause mortality compared to enoxaparin-based regimens, with no statistical evidence for differences in bleeding events⁶³. As regards the role of rivaroxaban in acute VTE treatment, a pooled analysis of EINSTEIN-DVT⁶⁴ and EINSTEIN-PE⁶⁵ trials, which compared rivaroxaban to enoxaparin/warfarin for prevention of recurrent VTE, showed the non-inferiority of rivaroxaban to enoxaparin/warfarin and a 41% reduction in the RR of major bleeding⁶⁶. Extended prophylaxis with rivaroxaban (EINSTEIN-Extension)⁶⁴ reduced the incidence of symptomatic recurrent VTE to a greater extent than placebo, with a non-significant increase in the incidence of clinically relevant bleeding. In the ROCKET-AF trial, patients with atrial fibrillation and additional risk factors for stroke were randomised to receive either rivaroxaban or dose-adjusted warfarin⁶⁷. Rivaroxaban was non-inferior to warfarin for the prevention of stroke or systemic embolism, with no significant differences in rates of major and clinically relevant non-major bleeding between the two study groups.

Several systematic reviews and meta-analyses have investigated the primary outcomes of DOAC recipients in various clinical settings by pooling the results from the different trials. In a recent meta-analysis (54,875 patients in 12 phase II and phase III studies)⁶⁸ of the safety and efficacy of DOAC for stroke prevention in non-valvular atrial fibrillation, novel anticoagulants were associated with reduced total mortality (5.61 vs 6.02%; RR: 0.89, 95% CI: 0.83–0.96) and stroke/systemic embolism (2.40 vs 3.13%; RR: 0.77, 95% CI: 0.70–0.86) in comparison to VKA. A trend toward reduced major bleeding was also noticed (RR: 0.86, 95% CI: 0.72–1.02), with a significant reduction in intracranial haemorrhage (RR: 0.46, 95% CI: 0.39–0.56) in patients receiving DOAC. Another recent meta-analysis conducted in acute VTE patients (24,455 patients in 5 phase III randomised trials) found a similar risk of recurrent VTE (RR: 0.88, 95% CI: 0.74–1.05) and overall mortality (RR: 0.97, 95% CI: 0.83–1.14) in patients treated with DOAC or VKA⁶⁹. The DOAC were, however, associated with a significantly lower risk of major bleeding (RR: 0.60, 95% CI: 0.41–0.88). The risks of fatal bleeding (RR: 0.60, 95% CI: 0.46–0.77) and major bleeding (RR: 0.80, 95% CI: 0.63–1.01) bleeding were lower in DOAC-treated patients than in VKA recipients in a systematic review pooling data from the six most important randomised controlled trials on the management of atrial fibrillation and VTE⁷⁰. These results were replicated in a more recent systematic review of 12 randomised controlled trials involving 102,607 patients with VTE or atrial fibrillation⁷¹. There is, however, some evidence from post-marketing studies that some DOAC (particularly dabigatran, but also rivaroxaban) increase the rate of gastrointestinal bleeding in comparison with VKA^72–74, probably because they remain unchanged in the gastrointestinal tract. Caution should, therefore, be used when prescribing such novel oral anticoagulants to patients, particularly elderly ones, at increased risk of gastrointestinal bleeding.

Conclusions

After several decades without novelty in the armamentarium for long-term anticoagulation management of venous and selected arterial thromboembolic diseases, new classes of oral anticoagulants have emerged during the last decade, and have been subjected to extensive study of clinical effectiveness. Besides the undeniable advantages of oral administration at fixed dosages with less stringent need for laboratory monitoring, the results of the large phase III licensing trials provide solid evidence that DOAC are at least as effective as VKA for the prevention and treatment of thromboembolism and it is, therefore, easily predictable that these novel anticoagulant agents will be increasingly used in the near future in clinical practice worldwide. However, before recommending widespread use of DOAC, there are some important issues that should be addressed, such as their use in very elderly patients (i.e., aged 80 years or older), in those with impaired renal function, in cancer-associated thrombosis, as well as in patients at the extremes of body weight or under dual anti-platelet therapy^75–77. There are very few data from published phase III trials regarding these categories of patients. In addition, the lack of specific antidotes suggests the need for particular caution, especially in patients at increased risk of bleeding. Thus, although results from the first real-life studies are encouraging⁷⁸, further post-marketing long-term trials are needed for definitive assessment of their safety and efficacy with respect to VKA therapy. A final issue regards laboratory monitoring. Although DOAC require less extensive therapeutic monitoring, specific testing may be still required under certain circumstances, such as in the case of acute impairment of renal or liver function, unexpected bleeding or thrombosis, uncertain compliance to therapy, as well as combined administration with certain drugs such as antibiotics^79,80. In these circumstances, an appropriate combination of routine and second-line haemostasis testing may still be required to establish the anticoagulant effects of DOAC and adjust or optimise the therapeutic regimen⁸¹.