Pulmonary arterial hypertension (PAH) is a progressive and debilitating disease with limited treatment options. Although some patients do well with calcium channel blockers, most ultimately need more advanced therapy, such as prostanoids. Recently, a new class of therapeutic agents has been developed to treat these patients: the endothelin receptor antagonists (ERAs). Although they are not offering a complete cure, ERAs represent a dramatic breakthrough in the treatment of PAH patients, improving hemodynamics, as well as functional and clinical outcomes. This article describes the preclinical foundations for the development of ERAs, presents the current clinical trial evidence, and discusses future directions for research and debate.
The discoveries of endothelial cell-derived vasoconstricting compounds and extraordinarily potent vasoconstrictor peptides from the venom of Atractaspis engaddensis (Israeli burrowing asp), known as sarafotoxins, prepared the foundation for the isolation and characterization of a novel peptide called endothelin (ET).1,2 ET is homologous to sarafotoxin and an extraordinarily potent vasoconstrictor, synthesized predominantly by endothelial cells.3 The initial gene product is preproendothelin-1, which is serially processed to the 21 amino acid end-product ET-1. Endothelial cells release over 75% of ET abluminally into the muscular media, consistent with an autocrine/paracrine mechanism of action. ET signals via two main types of seven transmembrane-spanning G-protein-coupled receptors, known as ET receptor subtype A (ETA) and ET receptor subtype B (ETB). ETA receptors are found predominantly in vascular smooth muscle cells (VSMCs) and cardiac myocytes, while the ETB receptors are found in endothelial cells and VSMCs. Stimulation of VSMC ETA and ETB receptors produces profound and long-lasting vasoconstriction,4,5 and increases mitogenic activity. Stimulation of endothelial ETB receptors results in increased nitric oxide (NO) and prostacyclin release, producing minor vasodilation. Endothelial ETB receptors also act as a major clearance mechanism for circulating ET-1 in the pulmonary and renal circulations. In addition to causing pulmonary vascular vasoconstriction and VSMC hyperplasia and hypertrophy, ET receptor activation can cause fibrosis through fibroblast proliferation and extracellular matrix deposition, increased pulmonary vascular permeability, stimulation of inflammation through increased neutrophil and mast cell activation, production of inflammatory signaling peptides and cytokines, and increased expression of integrins. These effects are all believed to contribute to the development of PAH.
The exact role that the ET system plays in the pathogenesis of PAH remains unclear, but the fact that it is important in the disease has been established through multiple pre-clinical and clinical studies.3 Patients with PAH have increased local tissue and elevated circulating ET-1 concentrations that correlate with the severity of disease.6-8 Increases in pulmonary vascular ET converting enzyme-1,9 ETA, and ETB receptors have also been noted,10 suggesting that every level of the ET system is altered in PAH. Furthermore, the clinical efficacy of ERAs has established the importance of ET in PAH.
Clinical Studies of Dual ERAs in PAH
Bosentan is the only ERA currently approved for the treatment of patients. It is an orally available, dual (ETA and ETB) receptor blocker that has been shown to have significant beneficial effects in animal models of PAH.11 The first randomized double-blind placebo-controlled study with bosentan in patients with PAH (either idiopathic or associated with scleroderma) was performed in 32 patients.12 The patients enrolled in this pilot study were World Health Organization (WHO) functional class III/IV with a baseline six-minute walk test distance of between 150m and 500m, a mean pulmonary artery pressure (PAP) of more than 25mm of mercury (mmHg), a pulmonary capillary wedge pressure (PCWP) of less than 15mmHg, and a pulmonary vascular resistance (PVR) of more than 240dyn x sec x cm-5, despite prior therapies. Patients could not be receiving epoprostenol. Patients were evaluated by another six-minute walk test after 12 weeks of randomized therapy, and bosentantreated patients had a 76m (95% confidence interval (CI) 12-139m; p=0.021) longer walking distance compared with placebo-treated patients. This was associated with significant improvements in cardiac index, PAP,PVR,and PCWP. There was also a significant improvement in functional class and a delay in time to clinical worsening, with a borderline improvement in the Borg dyspnea index in bosentan-treated patients. Although bosentan may cause reversible increases in serum transaminases due to inhibition of hepatocyte bile salt secretion, only two patients developed raised concentrations and in both levels returned to normal without discontinuation or change in dose.This pilot study provided strong support for continuing development of bosentan through the larger Bosentan Randomized Trial of Endothelin Antagonist Therapy (BREATHE-1).
BREATHE-1 was a large phase 3 study investigating the efficacy of bosentan in a patient population similar to the one in the pilot study.13 These PAH patients (either idiopathic or associated with scleroderma or systemic lupus erythematosus) were WHO functional class III/IV with a baseline six-minute walk test distance of between 150m and 450m, a mean PAP of more than 25mmHg, a PCWP of less than 15mmHg, and a PVR of more than 240dyn x sec x cm-5, despite prior therapies. The trial randomized 213 patients to placebo or one of two bosentan doses (titrated up to 125mg or 250mg twice-daily) for 16 weeks, followed by an open-label period. Patients were evaluated through a six-minute walk test after 16 weeks of randomized therapy, and bosentan-treated patients had a 44m (95% CI: 21-67m; p<0.001) longer walking distance compared with placebo-treated patients without evidence of dose effect, associated with significant improvements in Borg dyspnea index, a significant delay in time to clinical worsening (p=0.002), and borderline improvement in WHO functional class. An adverse event of abnormal hepatic function was noted in 3% of placebo-treated patients, 4% of 125mg twice-daily bosentan-treated and 14% of 250mg twice-daily bosentan-treated patients, but there were no cases of acute hepatitis or liver-related deaths. These efficacy results supported the approval of bosentan for PAH treatment of and opened a new era of therapeutics for this disease.
As the largest PAH trial, BREATHE-1 provided significant information on patients with this disease and a number of ancillary studies and sub-studies were performed. An echocardiographic sub-study demonstrated that bosentan treatment significantly improved the Doppler-derived cardiac index,14 left ventricular (LV) early diastolic filling velocity, LV end-systolic area, and measures of right ventricular (RV) remodeling and function. These echocardiographic structural and functional improvements in the hearts of patients treated with bosentan were associated with improvements in walking distance and other functional measures. Although large mortality trials in this patient population would be practically prohibitive to perform, survival analysis of the bosentan-treated patients from the two studies noted above demonstrated a marked survival benefit when compared with historical controls.15 In this analysis, bosentan-treated patients had a Kaplan-Meier survival estimate of 96% at 12 months and 89% at 24 months, compared with a predicted survival of 69% and 57%, respectively, suggesting that first-line bosentan therapy improved survival in patients with advanced PAH.
Furthermore, another analysis of first-line oral bosentan treatment compared with a historical cohort of patients treated with intravenous epoprostenol suggested that, even when adjusted for baseline factors, epoprostenol-treated patients had a greater probability of death (hazard ratio: 2.2; 95% CI: 1.2-4.0) compared with bosentan-treated patients.16 Although limited by the use of historical controls, these analyses suggest that there is a clear role for bosentan as a first-line agent in the treatment of patients with moderate to severe PAH.
Clinical Studies of ETA Receptor Antagonists in PAH
Sitaxsentan is a highly selective ETA receptor antagonist (6,500:1 selective for ETA versus ETB) that has also been studied in randomized studies. The initial open-label study of sitaxsentan in 20 patients using doses ranging from 100mg to 500mg twice-daily for 12 weeks suggested improvements in six-minute walk test distance and mean PAP, but no difference in cardiac index. Of the 20 patients receiving sitaxsentan, 11 patients had an adverse event related to increased prothrombin time/international normalized ratio (PT/INR). This adverse effect is anticipated, given that sitaxsentan is a potent inhibitor of CYP2C9 P450 hepatic enzyme (unpublished data), which is the principal enzyme involved in metabolism of warfarin. Although this adverse effect is clinically significant, given that most patients with PAH are treated with warfarin, dose adjustments and careful monitoring should be able to compensate. Seven patients (35%) developed elevations in serum transaminase levels and two patients developed serious complications, one of whom developed acute hepatitis, while the second developed fatal acute fulminant hepatitis with central bridging necrosis. After careful analysis of the pharmacokinetic data, lower sitaxsentan doses were selected for the subsequent randomized studies.
The Sitaxsentan to Relieve Impaired Exercise (STRIDE-1) trial evaluated the efficacy of 100mg or 300mg sitaxsentan once-daily compared with placebo in 178 patients with PAH (due to idiopathic, connective tissue disease, or congenital systemic-to-pulmonary shunts).18 In addition, patients were required to have a peak VO2 between 25% and 75% predicted, a mean PAP of more than 25mmHg, a PCWP of less than 15mmHg, and a PVR of more than 240dyn x sec x cm-5, despite prior therapies. Patients could not have received bosentan, prostaglandin analog, or epoprostenol within 30 days of study.
The primary end-point was per cent predicted peak VO2 at week 12, which demonstrated no significant difference in the 100mg sitaxsentan group, but a 3.1% (p<0.01; or p=0.22, when adjusted for multiple comparisons) improvement in the 300mg group. Improvements in the six-minute walk test were noted in both doses (placebo-corrected: 35m for 100mg, p<0.01; and 33m for the 300mg dose group, p<0.01). Also, hemodynamics and New York Heart Association (NYHA) functional class were improved. There were no significant differences in time to worsening of heart failure between the two groups. Adverse events included 19% of sitaxsentan-treated patients experiencing increased PT/INR (p<0.02) and estimates of amino-transferase elevations of more than three times the upper limit of normal to be 8% for the 100mg group and 26% for the 300mg group at six months.
Ambrisentan is a moderately selective (260:1) ETA receptor antagonist that has recently been studied in an uncontrolled double-blind dose-ranging study enrolling 64 patients with PAH (idiopathic or associated with collagen vascular disease, anorexigen use, or HIV infection). The patients enrolled in this pilot study had a baseline six-minute walk test distance of between 150m and 450 m, a mean PAP of more than or equal to 25 mmHg, a PCWP of less than 15 mmHg, and a PVR of more than 240dyn x sec x cm-5, despite prior therapies. Patients could not be receiving chronic prostanoid or ERA therapy within four weeks of study entry. Patients were randomized to ambisentan 1mg, 2.5mg, 5mg, or 10mg orally once-daily and evaluated after 12 weeks of randomized therapy. In the absence of a placebo control, the results are difficult to interpret, but there did appear to be an improvement in six-minute walk test distance (36m compared with baseline), Borg dyspnea index, WHO class (associated with reductions in mean PAP and PVR), and increased cardiac index. Two patients (3.2%) had increases in aminotransferase concentrations during the study. This dose-ranging study suggests that ambrisentan may also be a candidate for further study as a treatment for PAH.
The Choice between Dual ERAs or Selective ETA Receptor Antagonists
While the dual ERA bosentan has demonstrated clear clinical utility, the development of selective ETA receptor antagonists, such as sitaxsentan, encourages comparison of the two classes of ERA.19 Obviously, the only way to make a choice between the two is a well-designed, unbiased head-to-head comparison of the agents. However, until such a study is completed, other data sources will be drawn upon to inform the decision.
The pre-clinical data do not necessarily resolve the issue. Some argue that inhibition of the ETB receptor results in deleterious increases in circulating ET-1 by blocking the clearance receptors, while others have argued that complete blockade is necessary due to the important role that ETB receptors have in mediating pulmonary vasoconstriction, as well as the inflammatory, proliferative, and pro-fibrotic effects of ET-1. In a rat monocrotaline model of PAH,20 rats treated with a dual ERA had a 67% five-week survival compared with 56% in the selective ETA receptor antagonist and 35% in the control groups. Severe RV hypertophy was also inhibited by the dual ERA, but not by the selective ETA receptor antagonist, despite similar reductions in RV systolic pressure.
The clinical trial data may be more helpful. Bosentan demonstrated not only improvements in six-minute walk duration (76m in pilot study; 44m in BREATHE-1), but also a delay in time to clinical worsening in both placebo-controlled trials. The STRIDE-1 trial did not meet the pre-specified primary end-point adjusted for multiple comparisons of a significant change in peak VO2 and no significant impact on time to clinical worsening, but did demonstrate an increase in the six-minute walk test distance (35m and 33m for the 100mg and 300mg groups, respectively). Some investigators have pointed to the numerically greater increase in exercise duration in the bosentan studies as support for the superiority of dual ERA therapy. However, the STRIDE-1 investigators analyzed a subgroup of their study, selected to be similar to the patients enrolled in BREATHE-1, and found a 65m increase in six-minute exercise distance.21,22 While the authors suggest that this post hoc, non-adjusted subgroup analysis from a trial that failed to meet its primary end-point supports the superiority of selective ETA receptor antagonists, such conclusions are highly flawed given the small sizes of the studies and the inability to control for variations in patient characteristics. The STRIDE-2 trial (recently reported at the European Society of Cardiology 2005 Congress) enrolled 246 PAH patients (idiopathic or associated with connective tissue disease or congenital heart disease), randomized 1:1:1:1 to placebo, sitaxsentan 50mg, sitaxsentan 100mg, or open-label bosentan treatment for 18 weeks. Preliminary data showed that placebo-corrected improvements were noted in the sitaxsentan 100mg (31.4m; p<0.05) and bosentan (29.5m; p=0.05) groups, but not for the sitaxsentan 50mg group.
The occurrence of liver function test abnormalities of more than three times the upper limit of normal was 6.5% in placebo, 4.9% in sitaxsentan 50mg, 3.2% in sitaxsentan 100mg, and 11.5% in open-label bosentan groups. Two bleeding episodes occurred in the presence of elevated INR in sitaxsentan-treated patients. Comparisons with the bosentan group are severely confounded by the open-label nature of the randomization, but the full results are eagerly anticipated.
Combination Therapy with ERAs in Patients with PAH
The availability of bosentan in the context of patients treated with prostanoids has encouraged investigation into the effects of combined therapy. BREATHE-2 was a double-blind placebo-controlled study in 33 patients with PAH who started epoprostenol therapy and two days later were randomized to either placebo or bosentan.23 Epoprostenol was up-titrated during the ensuing 16 weeks, at which time patients were evaluated. In this study of acute co-administration, there were no significant differences in hemodynamics or clinical outcomes between the placebo and bosentan-treated patients. In another study, where prostanoid therapy had been initiated for a median of 16 months prior to entry into the study,24 20 patients who received bosentan for three months showed significant improvements in the six-minute walk test distance and maximal oxygen consumption on cardiopulmonary exercise testing. Similar improvements were noted in another uncontrolled open-label study.25 The open-label and non-randomized nature of some of these studies limits interpretation, but there seems to be potential for significant additive effect with bosentan and prostanoids. Studies with sitaxsentan combination therapy are on-going.
The promise suggested by the pre-clinical studies of the ERAs has been fulfilled in the recent clinical trials. Bosentan, a dual ERA, has been approved by the US Food and Drug Administration (FDA) for treatment of patients with PAH, and a new drug application (NDA) has been submitted to the FDA for sitaxsentan for a similar indication. Future studies will need to focus on establishing efficacy of new therapies and comparing their relative benefits and risks.