Systemic sclerosis (SSc; scleroderma) is a complex, immune-mediated disease associated with high mortality rates (1, 2). Although SSc has vascular, immunologic, and fibrotic components that are pathologically interconnected (1, 2), one hypothesis suggests that endothelial damage and vascular dysfunction may be some of the earliest pathogenetic alterations (3, 4). Functional abnormalities of the blood vessels, e.g., vasoconstriction, and structural changes, including intimal proliferation and obstruction, are expressed clinically as Raynaud's phenomenon, digital ulcers, renal and myocardial disease, and pulmonary hypertension (5).
Endothelial damage in SSc, whether caused by immunologic stimuli, ischemia-reperfusion injury, or other pathways, has many consequences, including increased production of endothelin (1, 2, 6). This endothelium-derived peptide, which is involved in the regulation of vascular function under normal physiologic conditions (7), plays a key role in vascular pathologies by exerting various deleterious effects, including hypertrophy of the vascular smooth muscle cells, cellular proliferation and fibrosis, increased vascular permeability, activation of leukocytes, and induction of cytokine and adhesion molecule expression (6, 8). Moreover, endothelin is the most potent naturally occurring vasoconstrictive mediator and when exogenously administered to healthy volunteers, a marked dose-dependent reduction of the forearm blood flow occurs (9). The effects of endothelin are transmitted upon binding to 2 cognate receptors, ETA and ETB, which are mainly expressed on endothelial cells (ETB), smooth muscle cells, and fibroblasts (6, 8).
Bosentan is a specific, orally active, dual endothelin receptor antagonist (10) that has recently been approved for the treatment of pulmonary hypertension. Two randomized, controlled studies have shown that bosentan reduces symptoms, improves exercise capacity and hemodynamics, and delays clinical worsening in patients with pulmonary hypertension (11, 12). A subgroup analysis of these trials and their open-label extensions in patients with connective tissue diseases showed that bosentan treatment may have a positive effect on outcome (13). Additional retrospective studies also suggest that bosentan treatment is clinically beneficial in SSc patients with pulmonary hypertension (14, 15), including patients with restrictive lung disease (15). Moreover, in a randomized, placebo-controlled trial of bosentan for SSc-related digital ulcers, although there was no difference between treatment groups in the healing of existing ulcers, bosentan prevented new digital ulcers and improved hand function (16).
Elevated circulating levels of endothelin have been repeatedly observed in SSc (for review, see ref.6), as well as in various other pathologies of the vascular endothelium (17), but whether increased endothelin activity is a causal factor of endothelial dysfunction and/or damage or an epiphenomenon remains unclear (17–19). Experimental studies have shown an improvement in the endothelial function of large arteries, assessed ex vivo, following short-term administration of endothelin receptor antagonists in animal models of hypertension (20, 21) and atherosclerosis (22), suggesting that some of the endothelin-mediated deleterious effects on the vasculature are reversible.
To the best of our knowledge, the direct or indirect effects of oral bosentan treatment on vascular endothelial dysfunction have not been previously addressed. To further study the effects of this treatment approach in SSc, we tested the hypothesis that bosentan, administered for pulmonary hypertension or digital ulcers in these patients, would improve endothelial function assessed by means of noninvasive, high-resolution brachial artery ultrasound.
- Top of page
- PATIENTS AND METHODS
- AUTHOR CONTRIBUTIONS
After the vascular hypothesis of SSc pathogenesis was introduced (3), endothelial cell function became the focus of research aimed at a better understanding of the pathogenesis of SSc, leading to more effective treatments (4, 31). Indeed, drugs currently used to treat the vascular complications of SSc, including bosentan for pulmonary arterial hypertension, have increased survival rates and can truly be classified as disease-modifying compounds (2). The present study was designed specifically to examine whether endothelial dysfunction of medium-sized conduit arteries, which are readily accessible for measurement, is affected by oral bosentan treatment. Endothelium-dependent and -independent vascular function was assessed by measuring brachial artery FMD% and NTG%, i.e., the change in diameter during hand hyperemia and after sublingual glyceryl trinitrate administration, respectively. This technique is attractive because it is noninvasive and allows for repeated measurements (24, 25).
FMD% is nitric oxide (NO)–dependent in humans and it is considered to be a reliable marker of endothelial function (32). As previous studies from our laboratory have shown (33, 34), this method can be applied in SSc patients without technical difficulties related to the thickened skin, while decreased brachial artery FMD% is a common finding in these patients (33–35). Using this approach, a highly significant improvement of endothelial function to a level comparable to that in healthy individuals, which was not due to changes in the degree of reactive hyperemia, was found. Moreover, all medications had to be discontinued for 12 hours before the vascular studies, thus implying that the bosentan-induced beneficial effect was probably due to rather long-lasting changes.
In contrast, bosentan treatment had no apparent effect on NTG%, which is considered to be NO-independent (24, 25, 32). Although the mechanism of bosentan-induced effect was not directly addressed in the present study, the results suggest that the improvement in endothelial function was mediated by enhanced NO production. Since increased endothelin activity inhibits NO synthesis (36), which has been found to be impaired in SSc (37), the bosentan-induced increase in FMD% is not surprising. Accordingly, using endothelial NO synthase–deficient mice, Gonon et al (38) showed that bosentan preserves endothelial and cardiac contractile function during ischemia and reperfusion via a mechanism dependent on endothelial NO production.
Improved endothelial dysfunction in the brachial artery may thus imply that inhibition of the endothelin/endothelin receptor system promotes NO bioavailability in different organ systems. Along this line, Girgis et al (39) have found that decreased exhaled NO in patients with pulmonary arterial hypertension is reversed with bosentan treatment. An improvement in endothelial function within the lungs induced by bosentan might therefore contribute to the beneficial effects of this treatment in pulmonary hypertension (11, 12), which is consistent with the inverse correlation between pulmonary pressure and FMD% changes observed in our patients.
Treatment with bosentan at a dosage of 125 mg/day had no apparent effect on systemic vasodilation, since arterial blood pressure was not significantly affected. This is consistent with previous reports, in which daily administration of 100 mg of bosentan did not affect the sympathetic nervous system or the renin– angiotensin–aldosterone system (40). Also, venous occlusion plethysmography failed to reveal an apparent effect of bosentan on peripheral flow reserve, probably because irreversible structural changes had already occurred in the microvasculature of patients with advanced, severe SSc (5).
Alternatively, the daily dose of 125 mg may be insufficient to reach the maximal effect, particularly on structural changes. This dose is efficacious, but the maximal effect of decreasing pulmonary pressure in patients with pulmonary hypertension is achieved with a daily dose of 250 mg (11, 12). Although the gold standard noninvasive measure of arterial stiffness, i.e., the radial–femoral or carotid–femoral pulse wave velocity, was not available, the parallel measurements of AI, an indirect marker of arterial stiffness, and peripheral vascular resistance (29), were similar before and after treatment, suggesting that bosentan exerted a direct effect on endothelial function.
We also determined, by measuring soluble ICAM-1 and E-selectin concentrations, whether the effect of bosentan on NO-dependent endothelial dysfunction is related to alterations in proinflammatory processes. Other findings suggest that increased concentrations of ICAM-1 in the serum, which are commonly found in patients with SSc (41), may contribute to impaired vasomotor function in forearm microvessels (42), but may also be related to FMD% in healthy individuals, independently of cardiovascular risk factors (43). Moreover, Xu et al (44) have shown that endothelin up-regulates ICAM-1 expression in normal and SSc-derived fibroblasts. In contrast to other molecules with wider cell-type expression, soluble E-selectin is shed only by activated endothelium, and elevated circulating levels have been found in various diseases, including SSc (for review, see ref.45). We found that treatment with bosentan for 4 weeks had no apparent effect on the increased baseline levels of either soluble ICAM-1 or E-selectin, which remained elevated for as long as 4 months after treatment.
Moreover, we observed that bosentan did not significantly modify serum concentrations of either VEGF or ET-1, which are commonly elevated in SSc (6, 46), at least after 4 weeks of treatment. Increased tissue expression of both these angiogenic molecules were shown to be prevented by bosentan in an animal model of pulmonary arterial hypertension (47), but relevant studies in humans are lacking. Taken together, these results suggest that the observed improvement of endothelial function can be attributed to the endothelin receptor level, since it is independent of effects related to deactivation of the endothelium or to decreases in proinflammatory processes. Such effects are shared by cyclophosphamide and prednisolone and may explain the mechanism by which immunosuppressive regimens act favorably on endothelial cell function in patients with early diffuse SSc (48).
The bosentan-induced improvement of FMD% in these patients was observed as early as the end of the first month of treatment and continued during the following months. Again, at the end of 4 months of treatment, bosentan had no significant effect on endothelium-independent vascular function, arterial blood pressure, arterial stiffness, and peripheral flow reserve. A similarly sustained effect was observed regardless of the dose, which was increased in 5 patients according to the standard dose for pulmonary hypertension. Finally, we had the opportunity to repeat the vascular studies at 12 months from baseline in 10 patients who received bosentan. FMD% values were considered normal at this point in 6 patients who continued bosentan treatment, whereas 4 patients who had discontinued bosentan treatment prior to this point had profoundly impaired FMD%, indicative of relapsed endothelial dysfunction, further suggesting a direct role of endothelin in SSc-associated vascular injury. Nevertheless, there is an obvious need for caution in applying these noncontrolled results (data not shown).
In conclusion, a rapid and sustained improvement in endothelial function in the brachial artery was evident following the administration of bosentan to patients with SSc. This improvement was not associated with hemodynamic changes, proinflammatory processes, or activated-endothelium effects, but rather, was due to an enhancement of NO production following inhibition of endothelin action, as seen in pulmonary hypertension (4, 39). These findings suggest that the endothelin receptor system is an important molecular pathway that is directly involved in certain reversible aspects of SSc-associated vascular injury. It remains to be determined whether a bosentan-induced improvement in vascular endothelial function may possibly translate into a better long-term clinical outcome, regardless of the presence of pulmonary hypertension (49).
Although due to the specific design of this study patients with normal FMD% were excluded, it would be interesting to further examine whether bosentan may prevent deterioration in patients with no apparent endothelial dysfunction of the brachial artery. Thus, taken together with findings suggesting that endothelin is also directly involved in lung fibrosis development (50, 51), a controlled trial may be warranted to examine the potentially global clinical benefit of endothelin receptor blockade in patients with early SSc, using small doses of bosentan or similar drugs.
- Top of page
- PATIENTS AND METHODS
- AUTHOR CONTRIBUTIONS
Dr. Sfikakis had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study design. Sfikakis, Papamichael, Stefanadis, Mavrikakis.
Acquisition of data. Papamichael, Stamatelopoulos, Tousoulis, Fragiadaki, Katsichti, Stefanadis.
Analysis and interpretation of data. Sfikakis, Papamichael, Tousoulis, Stefanadis, Mavrikakis.
Manuscript preparation. Sfikakis, Stamatelopoulos, Tousoulis, Mavrikakis.
Statistical analysis. Sfikakis, Stamatelopoulos.