Is anticoagulative therapy in systemic sclerosis to be reconsidered?

Abstract

Systemic sclerosis is a rare disease with a high mortality rate. It is a multisystem connective tissue disease due to endothelial autoimmune activation along with tissue and vascular fibrosis, inducing vasculopathy, with an angiogenesis wasting. The endothelial damage provokes platelet activation and immune cell adhesion. The detachment of endothelial cells leads to the interaction of platelets and collagen present in the exposed subendothelial layer. This provokes the activation of several coagulative factors, inducing a pro-thrombotic condition by thrombin generation, which converts fibrinogen into fibrin. Moreover, thrombin has other functions, such as the induction of hyperplasia in smooth muscle cells and fibroblasts, thereby favouring fibrosis. An increased risk of venous thromboembolism has been found in systemic sclerosis, whereas pulmonary hypertension may be due to the obstruction of small pulmonary arteries. Pulmonary veno-occlusive disease may also occur. Warfarin showed inconsistent results, while the outcomes of a randomised, placebo-controlled clinical trial on apixaban versus placebo are still awaited. A new anticoagulation strategy based on anti-factor XI drugs is being developed, with the aim of achieving optimal anticoagulation along with a low risk of bleeding. The molecule types under investigation in this category include monoclonal antibodies, small molecules, natural inhibitors, antisense oligonucleotides, and aptamers. Patients with systemic sclerosis may be ideal candidates for clinical trials planned to analyse the efficacy and safety of these molecules.

Introduction

Systemic sclerosis (SSc) is a rare disease with an annual incidence of 0.6–5.6 per 100,000 adults and a prevalence of 7.2–44.3 per 100,000 adults. ¹ This disease has a high mortality rate; however, the cumulative 5-year survival has improved to 84% in diffuse cutaneous SSc and 93%–96% in limited cutaneous SSc. ² SSc is a multisystem connective tissue disease sustained by autoantibodies against the endothelial cells (ECs) along with tissue and vascular fibrosis, which consequent vasculopathy and an angiogenesis waisting. ³ In the early phase of the disease, the endothelial damage provokes an inflammatory perivascular infiltrate, resulting in tissue hypoxia and multiple organ involvement. ⁴ Subsequent alterations in the endothelial membrane induce platelet activation and immune cell adhesion, and chemokine production through endothelial spaces. This process may also induce cell detachment, indicated by an increase of circulating ECs. ⁵ The detachment of the ECs provokes the interaction between platelets and collagen of the exposed subendothelial layer. This process, in turn, activates several coagulative factors, leading to a pro-thrombotic condition. ⁶ In particular, platelets adhere to collagen via glycoprotein VI ⁷ and then aggregate by means of the von Willebrand factor (vWf). ⁸ Activated platelets are characterised by a procoagulant activity, due to the exposure of phospholipids, such as phosphotidylserine, on their membrane, which enable the adsorption of clotting factors which in turn generate thrombin, thus amplifying the coagulative cascade. ⁹ Tissue factor (TF) plays a crucial role in the activation of blood coagulation, as it is placed in the subendothelial layer exposed because of EC detachment. ¹⁰ TF forms a complex with factor VII which activates both the extrinsic pathway of the coagulation cascade and factor IX. The latter is speeded up by factor VIII. The amplification of thrombin generation induces an increased fibrin deposition, as thrombin converts fibrinogen into fibrin. An imbalance between the activation and inactivation of the haemostatic system is the final results. ¹¹ Thrombin exerts other effects beyond haemostasis: it enhances the role of blood coagulation in inducing tissue fibrosis in multiple organs. ¹² Moreover, thrombin drives fibroblast and neutrophil migration and adhesion, ¹³ while inducing the activity of profibrogenic factors in smooth muscle cells and epithelial cells. ¹⁴ Moreover, the role of thrombin activity in promoting lung fibrogenesis has been demonstrated. ¹⁵ Notably, Chambers and Scotton ¹² reported that coagulation zymogens are synthesised by the hyperplastic alveolar epithelium in pulmonary fibrosis, thereby completing a vicious circle, which further induces collagen deposition in the lungs. In vitro studies on fibroblasts showed that the activation of proteinase activated receptors (PAR-1) with both thrombin and FXa provokes their proliferation via the autocrine induction of platelet-derived growth factor-A (Figure 1). ¹⁶ Finally, in patients with SSc, thrombin concentrations in broncho alveolar lavage fluid (BALF) were higher than those in controls (14.6 vs 3.6 nmol/L, P < 0.02). ¹⁷ All these data clearly indicate that thrombin plays a crucial role in the pathogenesis of vasculopathy in SSc.

Figure 1.

SSc physiopathology. The different actions of thrombin: platelet activation, smooth muscle cell hyperplasia, fibroblasts proliferation with secondary fibrosis and fibrin deposition. Thrombin formation is induced by an endothelial damage with exposition of the sub-endothelium collagen which leads to platelet adhesion and aggregation, neutrophil and monocytes adhesion by adessive proteins (VCAM and ICAM). Monocytes in turn expose tissue factor, the trigger of blood coagulation, with consequent further thrombin production.

SSc and vascular complications

Several consequences of vascular complications in SSc exist. First, an increased risk of venous thromboembolism (VTE) has been shown in SSc. ¹⁸ Moreover, the pathogenesis of pulmonary hypertension (PH), which accounts for 30%–40% of deaths in SSc, ¹⁹ should be reviewed with attention. Launay et al. ²⁰ highlighted that PH in SSc may be the consequence of pulmonary veno-occlusive disease and obstruction of the small pulmonary arteries. Recently, the possible development of pulmonary thrombosis has been envisaged in severe acute respiratory syndrome coronacirus 2 (SARS CoV-2) infection, ²¹ pneumonia, asthma, chronic obstructive pulmonary disease, sickle cell disease and Gaucher disease. ²² In 214 SSc patients, high D-dimer levels were found to be associated with both microvascular and macrovascular lesions. After a follow-up of 2.3 years [range: 1.1–3.3], new macrovascular complications occurred only in patients with high D-dimer levels, ²³ thus indicating the presence of a pro-thrombotic background in SSc.

SSc and anticoagulation

Several years ago, it became clear that anticoagulation therapy could be useful in SSc. In 2012, a pilot study on 117 SSc patients with PAH found that warfarin (hazard ratio (HR) = 0.20, 95% confidence interval (CI): 0.05–0.78, P = 0.02) and combination PAH therapy (HR = 0.20, 95% CI: 0.05–0.83, P = 0.03) were successful during 2.6 ± 1.8 years of follow-up. ²⁴ Therefore, anticoagulation therapy was taken into account for patients with SSc and PAH. ²⁵ Nevertheless, Palazzini et al. ²⁶ stated that anticoagulation was not effective in the treatment of PAH in patients with SSc, after having considered two studies, which employed antivitamin K drugs. It is worth to note that warfarin needs frequent control of the prothrombin time; therefore, venipunctures are required, thus further impairing the quality of life of these patients. In 2016, a multicentre randomised placebo-controlled trial of 2.5 mg apixaban twice daily in SSc-PAH was planned in Australia. ²⁷ Apibaxan is a direct oral anticoagulant (DOAC), which does not require laboratory monitoring. ²⁸ A 3-year follow-up was envisioned, with the primary outcomes being the time to death, the worsening of PAH, the quality of life and adverse events. Finally, a total sample size of 170 patients (85 per arm) was identified as a correct sample size. The trial has been completed; however, the results have not yet been published. Although apixaban possesses a good safety and efficacy profile and is being used for the prophylaxis and treatment of atrial fibrillation ²⁹ and VTE, ³⁰ major bleeding occurs in a percentage around 2% year while gastrointestinal bleeding accounts for 0.76%/year, as reported by the Aristotle trial on atrial fibrillation. ²⁹ A high risk of gastrointestinal bleeding may be experienced, if other direct anticoagulants are employed. ³¹ This may be matter of concern in patients with SSc, as they may develop intestinal or gastric telangiectasias ³² Nevertheless, Bogatkevich et al., ³³ based on their studies on the anti-fibrotic properties of dabigatran, conducted a small trial in SSc-associated interstitial lung disease using dabigatran 75 mg/twice/day, ³⁴ to investigate the safety and tolerability of this direct thrombin anticoagulant. ³⁵ After 6 months, dabigatran was found to be safe and well tolerated. In the meantime, a new anticoagulant strategy based on anti-factor XI drugs is being developed. ³⁶ The aim is to achieve optimal anticoagulation with a further reduction of bleeding.

Factor XI is a new potential target, as its deficiency or inhibition offers protection against thrombosis with no or little bleeding, as shown by several studies. ³⁷ Upon activation by factor XII, factor XI activates factor IX, which, along with factor VIII, acts upon factor X, the core of the blood coagulation cascade. Notably, thrombin further stimulates factor XI, thereby self-amplifying its production by means of this positive feedback. ³⁸ A number of molecules is under investigation in Phase II studies: monoclonal antibodies, small molecules, natural inhibitors, antisense oligonucleotide (ASO) and aptamers. In particular, ASOs and monoclonal antibodies can be administrated either subcutaneously weekly or monthly, providing advantage in terms of quality of life. The characteristics of these molecules are summarised in Table 1, while the key results of several phase I and II trials have been reported in excellent reviews recently published.^39,40 Finally, all the targets of the anticoagulants are depicted in Figure 2.

Table 1.

Characteristics of the anti-XI anticogaulants.

Characterisics	ASOs, IONIS-FXIRX, IONIS-FXI-LRX, FXI-LICA (2nd generation)	Monoclonal, Antibodies, Osocimab, Abelacimab, Xisomab 3G3	Small molecules, up to 900 Da, Milvexian, Asundexian, ONO-7684, SHR2285, EP-7041, BMS-962212	Natural Inhibitors acaNAP10 (nematodes) fasxiator (snakes) desmolaris (bat) boophilin Ixodes ricinus (ticks)	Aptamers single-stranded oligonucleotides
Mechanism	Target FXI mRNA in the liver	Specificity and affinity for FXI	High selectivity against FXIa	High selectivity against FXIa	Target FXI and XII
Administration	SC	IV and SC	Oral or IV	SC or IV	SC or IV
Adm. frequency	Weekly–montly	Montly	Daily	Daily	Daily
Start of action	Weeks	Hours	3–4 hours	Minutes	Minutes–hours
End of action	Weeks	~4 weeks	8–14 hours	Hours	Minutes–hours
Renal removal	No	No	Yes	Unclear	No
CYP metab.	No	No	Yes	No	No
Studies	Phase II	Phase I	Phase I, II, and one RCT	Phase I	No

ASO: antisense oligonucleotide; SC: subcutaneousl; IV: intravenously, RCT: randomised controlled trials; CYP: Cytochrome P450.

Figure 2.

The target of the all anticoagulants is depicted: Warfarin and Acenocoumarol (AVK), Direct oral anticoagulants (Apixaban, Rivaroxban, Edoxaban and Dabigatran) and upcoming anti-factor XI.

Conclusion

In conclusion, we believe that inhibiting thrombin activity by targeting factors, such as factor XI, in the coagulation cascade may be a good approach in anticoagulation therapy as since it could reduce fibrin deposition and the profibrotic effects described above. In our opinion, efforts are required from rheumatologists in proposing randomised clinical trials to achieve a reduction in thrombin activity on one hand, while avoiding bleeding as much as possible on the other.