**1. Introduction**

Thromboembolism and its associated complications remain a huge healthcare burden worldwide. The central role of thrombosis is observable in a variety of cardiovascular disorders, most notably in coronary artery disease (CAD), atrial fibrillation and stroke, peripheral arterial disease (PAD), and venous thromboembolism (VTE). Ischemic heart disease and stroke collectively are responsible for nearly 25% of all deaths worldwide [1], whereas estimates for the incidence rate for VTE, comprising deep vein thrombosis (DVT) and pulmonary embolism (PE), range from 115 to 269 per 100,000 people worldwide [2]. The impact of thrombosis extends beyond the cardiovascular arena and is increasingly being encountered in pathologies as diverse as cancer, immunological diseases, and even psychiatric disorders. Even among patients with human immunodeficiency virus (HIV) infection, who are living longer thanks to improvements in antiretroviral treatment, there is evidence of increased thrombosis [3], which contributes to their increasing morbidity and mortality from cardiovascular causes (6–15% of total mortality) [4,5].

The high human and financial cost of thromboembolic events underscore the need for newer and better therapeutic options for the management of thrombotic disorders. The challenge is in developing an agent that has potent antithrombotic effects but minimal bleeding risk, as it requires a very fine balancing act in modulating the hemostatic processes. Current anticoagulant options for clinical treatment of thrombotic disorders include antithrombin activators (unfractionated heparin), low molecular weight heparins (LMWHs and fondaparinux), vitamin K antagonists (VKA—warfarin), direct inhibitors of activated factor X (rivaroxaban, apixaban, edoxaban, betrixaban), and direct inhibitors of thrombin (hirudins, argatroban, and dabigatran). Although heparins and warfarin are low-cost options with a high degree of efficacy, both are associated with drawbacks that limit their clinical use. Heparin-induced thrombocytopenia, although infrequent, can be potentially lethal, and the development of osteoporosis and risk of contamination

**Citation:** Badimon, J.J.; Escolar, G.; Zafar, M.U. Factor XI/XIa Inhibition: The Arsenal in Development for a New Therapeutic Target in Cardioand Cerebrovascular Disease. *J. Cardiovasc. Dev. Dis.* **2022**, *9*, 437. https://doi.org/10.3390/ jcdd9120437

Academic Editor: Aleš Blinc

Received: 14 October 2022 Accepted: 3 December 2022 Published: 6 December 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

91

are additional factors to consider when using unfractionated heparin [6]. Some of these shortcomings have been reduced by LMWH and fondaparinux [6]. Warfarin presents the limitation of a narrow therapeutic window and major food- and drug-interactions [7]. The significant intra- and inter-patient variability of response of VKA makes frequent blood testing for dose-adjustment a cumbersome necessity. The last few years have provided a much-improved treatment option in direct oral anticoagulants (DOACs) that are convenient in administration while being potent and equally effective to VKA, often with a lower risk of bleeding [8–11]. Even so, the annual rate of major bleeding in patients on DOAC treatment remains significant [12], approximately 5% in elderly patients with atrial fibrillation (AF) [13]. This is partly why an unacceptably high proportion of AF patients—nearly one-third—do not receive the prophylactic anticoagulation they require. Even among those that do receive anticoagulation therapy, nearly half do not receive the proper doses [14]. The need for newer, safer anticoagulants is therefore high, and novel targets for therapeutic intervention are constantly under investigation.

#### **2. Distinguishing Physiological Hemostasis from Pathological Thrombosis**

Hemostasis is the normal, physiological process by which the clotting cascade seals up vascular damage to limit blood loss following injury. Thrombosis, on the other hand, encompasses various pathological conditions where the normally physiological clotting processes end up generating blood clot(s) inside the vascular lumen that are disruptive to the normal flow of blood. Thrombin generation and fibrin formation are the culminating steps in both hemostasis and thrombosis, but with important differences in the pathways involved.

Hemostasis is commonly triggered when tissue factor (TF) within the adventitial layer of blood vessels gets exposed to blood. Injury to vasculature that can lead to bleeding activates a series of soluble plasma proteins that act together in a cascade of enzyme activation events and culminate in the formation of platelet-fibrin clot(s). Because of the relatively high concentration of TF in such scenarios, the generation of thrombin is rapid and intense, quickly forming a hemostatic plug that seals the inciting TF away from blood. This disrupts the amplification of the coagulation processes through feedback mechanisms to the point of becoming pathological.

The concentration of TF in thrombosis is lower relative to hemostasis, but its duration of contact with blood components often lasts longer. Whether triggered by TF from disruption of an atherosclerotic plaque or activated monocytes/macrophages recruited to the site of injury or inflammation, or by implanted medical devices or neutrophil extracellular traps (NETs), these scenarios depend on the feedback mechanisms of the coagulation cascade for the growth and stabilization of the thrombus. This clot or thrombus can impede the flow of blood to the distal tissues and organs, leading to ischemia and necrosis, manifesting as clinical events including acute coronary syndrome, stroke, or deep vein thrombosis.

#### **3. A Brief Review of the Classic Coagulation Cascade**

The two major pathways for triggering blood clotting cascade are well known; (1) the tissue factor pathway and (2) the contact pathway. Both pathways trigger a series of cascading events that generate a blood clot (Figure 1) with the purpose to separate and seal the triggering agent from blood, thereby preventing its further contact with plasma components and arresting the thrombotic process.

#### *3.1. Tissue Factor Pathway*

This pathway is also known as the 'Extrinsic' pathway, as it is triggered by plasma components coming into contact with an agent that is extrinsic to blood (i.e., TF). The contact may happen when TF, normally embedded in the vascular wall, is exposed to blood due to rupture of a plaque, or when TF is expressed on the surface of cells active in inflammatory and immunological processes (e.g., monocytes and macrophages).

**Figure 1.** The classic model of the coagulation cascade with the Tissue Factor/extrinsic, the Contact/intrinsic and the common pathways. Triggering factors for both pathways are shown in square boxes. NETs: neutrophil extracellular traps, ECMO: extracorporeal membrane oxygenation.

Tissue Factor is an integral cell-membrane protein that forms a complex with the coagulation factor VIIa (FVIIa), normally present in plasma in the inactive zymogen form FVII. The TF:FVIIa complex is a potent activator of coagulation and converts factors IX (FIX) and X (FX) to the active forms FIXa and FXa, respectively (Figure 1). Each of these active enzymes assembles with its protein cofactor (FVIIIa and FVa, respectively) on suitable membrane surfaces to further propagate the coagulation cascade. The end result is a large burst of thrombin, the last serine protease in the clotting cascade. Thrombin not only converts fibrinogen into fibrin via limited proteolysis—which in turn assembles into a fibrin clot—but is also one of the most potent activators of platelets. The activation and aggregation of platelets contributes to the formation of a hemostatic plug. Additionally, thrombin also activates FV, FVIII, and FXI, the latter two of which are part of the contact pathway. Thus, the initial thrombin generated by the TF pathway can lead to activation of the contact pathway.

#### *3.2. Contact Pathway*

Also known as the 'Intrinsic' pathway, this pathway is triggered when blood comes into contact with anionic surfaces, such as extracellular DNA, RNA from activated or dying cells including neutrophil extracellular traps (NETs) released by activated neutrophils [15], or polyphosphates from the dense granules of activated platelets or microorganisms [16], or those on artificial surfaces [17]. This leads to a change in the conformation of plasma factor XII (FXII) into the active factor XII (FXIIa) [18,19]. FXIIa activates Prekallikrein to Kallikrein, which in turn reciprocally activates FXII to FXIIa in a positive feedback loop [20]. Downstream, FXIIa activates FXI to FXIa, which in turn leads to proteolysis of factor IX (FIX) to the active form (FIXa). The complex of FIXa and FVIIIa then activates FX to FXa at the point where the TF and contact pathways converge to form the final common pathway (Figure 1). The end result of all these interactions again is thrombin generation and formation of a blood clot.

The hemostatic process is kept in check by various inhibitory mechanisms that shut down the coagulation pathways, thereby localizing the hemostatic plug. These inhibitory mechanisms include proteins such as the tissue factor pathway inhibitor (TFPI) that inhibits FXa [21], activated protein C (APC), which degrades FVa and FVIIIa [22], and antithrombin (AT), which, in addition to factors IIa, Xa, and IXa [23], can also inhibit FVIIa and FXIa [24,25].

#### Significance of Contact Pathway in Thrombosis

Our understanding of the coagulation system in thrombotic pathophysiology has improved significantly in recent years, with the classical cascade being superseded by the cell-based model of coagulation (Figure 2). The TF pathway is understood to play a larger role in the 'initiation' and 'propagation' phases of coagulation, functioning more in normal hemostasis than in thrombosis. The contact pathway is more important in 'amplifying' the coagulation response, and despite its important role in clot formation in vitro, may contribute minimally to hemostasis in vivo, as supported by the lack of bleeding tendencies in patients deficient in FXII [26]. The contact pathway does appear, however, to have an important role in thrombotic disorders. Increased activity of plasma FXII, FXI, or kallikrein has been associated with atherosclerosis [27] and myocardial infarction [28,29], whereas severe FXI deficiency has been associated with reduced risk of stroke and deep vein thrombosis [30,31]. Deficiency in FXII in animal models has been reported to be protective against arterial thrombosis [32] and ischemic brain injury [33].

**Figure 2.** Cell-based coagulation model illustrating the Initiation, Amplification, and Propagation of the coagulation process. In the initiation phase, a small amount of thrombin and FIXa is generated on the surface of the tissue factor (TF)-bearing cell that then diffuse away towards platelets. In the amplification phase, this thrombin activates platelets (releasing factor Va from α granules), acts on vWF-VIII to release vWF and activated factor VIIIa, and generates activated factor XIa. The role of factor XI/XIa is primarily in the amplification stage where it activates factor IX and, in a feedback loop, promotes further activation of zymogen factor XI to active factor XIa. The propagation phase involves assembly of the various enzymes generated earlier to advance the process towards fibrin generation and clot formation. Components of the classic TF-, contact- and common-pathway are shown in blue, dark red, and green, respectively, along with triggers (black boxes) for each pathway.

Selective modulation of the contact pathway theoretically should lower the risk of thrombosis without increasing bleeding. Development of drugs that act by inhibiting components of the contact pathway is currently in high gear, with factor XI (FXI), and to a lesser degree factor XII (FXII), being the most prominent targets [34–36]. Evidence from epidemiological studies supporting their role in thrombosis is stronger for FXI than it is for FXII [37].

#### **4. Factor XI as a Therapeutic Target**

Factor XI is a blood coagulation zymogen produced by the liver that is part of the early phase of the contact pathway [38]. It is converted to the active serine protease FXIa by thrombin, FXIIa, and by FXIa itself and in turn activates FIX to further advance the coagulation process [38]. FXI plays an important part in blood coagulation because its feedback activation amplifies in vivo thrombin generation and fibrin formation [39]. The additional thrombin formed via the FXI feedback loop also promotes the activation of Thrombin Activatable Fibrinolysis Inhibitor (TAFI), which increases the clot's resistance to fibrinolysis, thereby helping to stabilize the formed clot.

The greater role of FXIa in thrombosis compared to hemostasis is evident from several epidemiological and genetic studies. Higher levels of circulating FXI levels are associated with increased risk for venous and arterial thrombosis, including stroke [40,41]. Deficiency of FXI (Hemophilia C, Plasma Thromboplastin Antecedent Deficiency, Rosenthal Syndrome) is rare and characterized by little to no bleeding tendency. Bleeding risk with factor XI deficiency selectively increases in tissues with high fibrinolytic activity (e.g., following dental surgery, tonsillectomy, and prostate surgery) [42]. Most frequent presentations involve nosebleeds or bleeding after tooth extractions. In fact, patients suffering from congenital FXI deficiency appear to have some degree of protection from thrombotic events, with lower rates of ischemic stroke and venous thromboembolism [30,43]. Moreover, hemorrhaging does not correlate with the levels of FXI in blood, i.e., bleeding is not restricted to patients with severe deficiency, and individuals with similar levels of FXI can experience different degrees of bleeding.

#### **5. Pharmacologic Strategies for Factor XI Inhibition**

Given the larger role FXI is thought to play in thrombosis than in hemostasis, novel approaches to inhibit its generation and activity are being explored as new therapeutic strategies (Figure 3). These include: (a) Antisense Oligonucleotides (ASOs) that act on the liver to knockdown hepatic synthesis of FXI, (b) small molecules that target the FXI active site or the heparin allosteric site on FXIa, (c) monoclonal antibodies that act by blocking the activation or inhibiting the activity, and (d) Aptamers.

In addition to their varying mechanisms of action, these strategies also differ in their routes of administration (oral vs. parenteral), the onset of action, and the duration of effect. Parenteral administration is a requirement for ASOs, aptamers and monoclonal antibodies, whereas small molecule agents offer the option of either parenteral or oral administration. The varied onset and duration of action may present a broad set of treatment options depending on the pathology at hand; acute thrombotic events requiring quick-acting agents whereas longer-acting options, such as antibodies, would be more suitable for chronic prophylactic and preventative measures. Similarly, for conditions presenting a high risk of bleeding complications such as trauma or surgery, shorter-acting agents would be preferable.

Inhibition of FXIa as a therapeutic option may also allow the possibility to easily reverse the effects of treatment, as has been tested successfully in animal models. In a rabbit AV-shunt model of thrombosis, the antithrombotic effects of a small molecule FXIa inhibitor (71.3 ± 5.2% lowering of thrombus weight vs. vehicle) were completely abolished by non-specific reversal agents (222% and 64% increase in thrombus weight vs. vehicle with FEIBA and NovoSeven, respectively) [44]. In another rabbit study, the addition of a specific reversal agent fully normalized the 210% prolongation in APTT produced by an

anti-FXIa antibody [45]. This availability of reversing strategies for FXIa inhibition would be a significant advantage for this class of drugs, similar to the one available for some DOACs. Inability to reverse treatment effects can magnify the concerns about bleeding risks associated with any antithrombotic agent, thereby hampering its proper clinical utilization. As an example, although it is still possible to reverse the effects of antiplatelet drugs [46–48], the lack of a convenient and simple reversal strategy mandates that bleeding risk be always at the forefront of any discussion involving antiplatelet drugs.

**Figure 3.** Sites of action (represented by the yellow circles) of the factor XI/XIa inhibitory drugs currently at different stages of clinical development. The anti-FXI ASOs (Antisense Oligonucleotides) block mRNA transcription of FXI in the hepatocytes, thus inhibiting its synthesis. Some monoclonal antibodies attach to the catalytic domain of FXI and block its FXIIa-mediated conversion to the active FXIa form, thus locking it in the inactive zymogen state. Most of the currently in development anti-FXIa antibodies act similar to the small molecule FXIa inhibitors and bind to the active site(s) on FXIa, thereby blocking its activity.

A summary of FXIa inhibitors in more advanced stages of clinical development is presented in Table 1.
