Background
Scleroderma (systemic sclerosis, SSc) is a rare autoimmune disease characterized by excessive extracellular matrix deposition, fibrosis and vascular alterations[
1,
2]. The disorder can affect almost any organ, including the kidneys, the gastrointestinal tract, lungs or heart, and most notably the skin, and may lead to severe dysfunction up to complete organ failure[
3]. The two major forms of SSc are localized scleroderma and systemic scleroderma. Localized scleroderma is the more common form of the disease and only affects the skin without any internal organ involvement. By contrast, systemic scleroderma or systemic sclerosis is characterized by cutaneous and non-cutaneous involvement and can be further subdivided into limited cutaneous scleroderma (lcSSc) and diffuse cutaneous scleroderma (dcSSc). The latter types are defined with regard to the extent of skin tightening, the number of affected inner organs as well as their typical autoantibody profile. Any combination of SSc and a rheumatologic disease such as lupus erythematosus, polymyositis, rheumatoid arthritis or Sjögren’s syndrome, is referred to as overlap syndrome[
2‐
4].
Clinically, digital ulcers and gangrene are a frequent and chronically recurrent complication of SSc and may result in considerable disability[
5‐
7]. Of note, the incidence for finger amputation was reported to be as high as 1.2% per patient-year in patients with SSc affected by digital ulcers[
5]. The main causes of digital ulcers are SSc-associated vascular alterations[
8,
9]. Vascular disease involves the microcirculation and arterioles and comprises swelling of the intima, intimal proliferation in the arterioles and distortion of the capillaries with occasional capillary necrosis. Endothelial apoptosis has been recognized as an important component of the vascular disease[
10,
11]. The resulting capillary destruction leads to a reduced size of microvascular beds, followed by decreased organ blood flow, eventually resulting in chronic ischemia. In addition, patients with SSc demonstrate a vascular dysfunction that is characterized by vascular permeability, a deregulated control of the vascular tone as well as an activation of the platelets and the coagulation systems[
10,
12].
Endothelin is a potent vasoconstrictor that is released by fibroblasts[
13]. Endothelin overexpression has been associated with various medium- and long-term physiologic processes, such as mitogenesis, fibrosis, vascular hypertrophy, inflammation, and tissue remodeling[
14,
15]. Additionally, there is now accumulating evidence that endothelin-1 (ET-1) is a key mediator in the regulation of the vascular tone. In SSc, the endothelin production is significantly enhanced, leading to vasoconstriction, vessel remodeling, local ischemia and formation of ulcers of the fingertips[
16,
17]. So far, bosentan represents the only approved drug for the treatment of SSc-related symptoms, namely digital ulcers.
The treatment of SSc includes the following objectives: reduction of vasospastic phenomena, improvement of vascular permeability, counteracting endothelial dysfunction and antiplatelet action, prevention of visceral involvement, and improvement in quality of life (QOL)[
18‐
21]. New specific therapies have been developed targeting prostacyclin and endothelin, two major mediators governing endothelial function, leading to endothelial dysfunction[
1]. In this context, stable analogs of prostacyclin, like iloprost, have shown efficacy and improved life expectancy in patients with SSc[
21‐
23]. The main pharmacological effects of iloprost are inhibition of platelet aggregation and vasodilatation. Both effects are mediated by an activity of adenylate cyclase/cAMP complex, activation of fibrinolysis, and reduced release of free oxygen radicals[
24,
25].
Bosentan is a dual endothelin receptor antagonist. It competes with ET-1 by binding to the receptors ET-A and ET-B, which are localized in the endothelial and muscle layers of the blood vessel walls. The contribution of ET-1 to the development of digital ulcers and the efficacy of bosentan therapy in patients with SSc was assessed in clinical studies by Korn
et al. and Matucci-Cerinic
et al.[
26,
27]. RAPIDS-2 (RAndomized, Placebo-controlled study on the prevention of Ischemic Digital ulcers secondary to Scleroderma) demonstrated a reduced incidence of new digital ulcers in those patients who already had ulcers, whereas bosentan did not exert any effect on the healing of ulcers[
27]. The results of RAPIDS-2 are also included in a recent meta-analysis of the healing and prevention of digital ulcers in patients with SSc by Tingey
et al. Notably, results from studies on iloprost have been similar, demonstrating no statistically significant effects on the healing or improvement of digital ulcers in patients with SSc, while intravenous iloprost was reported to prevent new ulcers[
28]. In spite of the aforementioned positive effects of both bosentan and iloprost in the treatment of SSc, few common, non-serious adverse effects have to be mentioned, including vasodilatation leading to flush, headache, gastrointestinal symptoms, hypotensive reactions, bradycardia or paresthesia. Moreover, bosentan therapy has been associated with an elevation of the liver aminotransferases (ALT and AST) as well as bilirubin[
29‐
31].
Our own clinical observations suggest that the frequency of centrofacial telangiectasia (TAE) may be increased in patients with SSc treated with bosentan. Here, we sought to assess the frequency on TAE in patients with SSc treated with bosentan or iloprost. Results may point toward a hitherto little-known, in some cases stigmatizing adverse effect of bosentan therapy.
Discussion
The ET-1 receptor antagonist bosentan and the prostacyclin analog iloprost are well established in the management of Raynaud’s phenomenon and ischemic ulcers in patients with SSc[
3,
25,
27,
28,
32]. The most important documented adverse effects of iloprost include flushing, photosensitivity, jaw pain, headaches, diarrhea, nausea and vomiting[
28,
33]. Documented adverse effects of bosentan comprise pruritus, urticaria, leukocytoclastic vasculitis, indurated erythema, flushing, peripheral edema, elevated aminotransferases, headache, dizziness, cough, nasal congestion and a potential worsening of symptoms in heart failure patients[
27,
28,
34‐
37]. Interestingly, even though flushing is a known adverse effect of bosentan, persistent alterations of the facial vasculature such as TAE, have remained largely unnoticed.
TAE are a characteristic feature of connective tissue diseases such as SSc, dermatomyositis and overlap syndromes[
4,
38]. Indeed, they reflect one of the cardinal symptoms of CREST syndrome (Calcinosis, Raynaud’s phenomenon, Esophageal dysmotility, Sclerodactyly, TAE). Whereas some authors state that the acronym CREST is obsolete, others still considered it to be a form of a limited cutaneous SSc (lcSSc)[
2,
39]. Hence, the observation of TAE in our cohort of patients with SSc is not surprising. Yet, while it cannot be ruled out that the increase in the number of TAE is a consequence of a worsening of the disease over the course of the therapy, the frequency and rapid progression of TAE in patients with SSc treated with bosentan is remarkable. This hypothesis is supported by the fact that patients treated with bosentan suspected that the onset of TAE correlated to the administration of the drug, whereas no such correlations were suspected by patients treated with iloprost. A further limitation of our study is the small number of patients included. Yet, our results are in line with a recent case report by Tong and Kumarasinghe in a 76-year-old woman treated with bosentan for four years. In this patient, a prominent flushing gradually progressed to persistent redness and TAE[
37].
The molecular and cellular mechanisms governing the development of TAE in SSc, as well as the mechanisms by which bosentan may induce persistent vascular alterations, have remained largely elusive. It has been proposed that TAE in SSc develop as a response to endothelial injury. This concept is supported by micro-capillaroscopic analyses that reveal an extensive derangement and destruction of the microvasculature in a variety of organ systems[
38]. Interestingly, the distribution and appearance of TAE in SSc correspond to TAE in patients with hereditary hemorrhagic telangiectasia (HHT; Osler-Weber-Rendu syndrome), pointing toward similar pathogenetic mechanisms. HHT is an autosomal dominant disorder of the vasculature development characterized by TAE and arteriovenous malformations[
40]. Abnormal TGF-β signaling has been shown to play a crucial role in the pathogenesis of HHT[
41]. Moreover, TGF-β signaling has been recognized as a key regulator of wound healing and fibrosis and exerts a variety of effects on the biology of endothelial cells and vascular tissue[
1]. Likewise van Royen
et al. could show that exogenous TGF-β stimulated the angiogenesis in the peripheral circulation in an
in vivo rabbit model[
42]. Interestingly, patients with SSc show elevated serum levels of connective tissue growth factor (CTGF), a downstream target of TGF-β, and scleroderma fibroblasts show an increased expression of the TGF-β receptor[
43,
44]. Therefore, it is tempting to speculate that TGF-β signaling may also play a role in the pathogenesis of TAE in patients with SSc[
38].
With regard to ET-1 antagonists, a recent case report demonstrated a significant alteration of the macrovascular involvement by bosentan in a 50-year-old Japanese patient with SSc. Magnetic resonance angiography showed an attenuation of a stenosis of the ulnar artery. The authors concluded that bosentan, besides reversing the vasoconstrictive effects of ET-1, also exerts remodeling effects on the vasculature[
45]. Accordingly, ET-1 has been shown to contribute to the mitogenic activity of fibroblasts and smooth muscle cells
in vitro[
46,
47]. Hence, the promotion of TAE development by bosentan in patients with SSc may be the result of vasodilatatory and/or direct vascular remodeling effects.
Competing interest
PAG and SM have received travel/meeting support by Actelion Ltd., Allschwil, Switzerland.
SM has received research funding by Actelion Ltd., Allschwil, Switzerland.
Authors’ contributions
SH collected the data. BAB, HS, EB, SM, PAG and BH performed data analysis and interpretation. KK performed statistical analyses. BAB, PAG and BH wrote the manuscript. All authors read and approved the final manuscript.