Articles
Treatment of patients with mildly symptomatic pulmonary arterial hypertension with bosentan (EARLY study): a double-blind, randomised controlled trial N Galiè, L J Rubin, M M Hoeper, P Jansa, H Al-Hiti, G M B Meyer, E Chiossi, A Kusic-Pajic, G Simonneau
Summary Background Treatments for pulmonary arterial hypertension have been mainly studied in patients with advanced disease (WHO functional class [FC] III and IV). This study was designed to assess the effect of the dual endothelin receptor antagonist bosentan in patients with WHO FC II pulmonary arterial hypertension. Methods Patients with WHO FC II pulmonary arterial hypertension aged 12 years or over with 6-min walk distance of less than 80% of the normal predicted value or less than 500 m associated with a Borg dyspnoea index of 2 or greater were enrolled in this double-blind, placebo-controlled, multicentre trial. 185 patients were randomly assigned to receive bosentan (n=93) or placebo (n=92) for the 6-month double-blind treatment period via a centralised integrated voice recognition system. Primary endpoints were pulmonary vascular resistance at month 6 expressed as percentage of baseline and change from baseline to month 6 in 6-min walk distance. Analyses of the primary endpoints were done with all randomised patients who had a valid baseline assessment and an assessment or an imputed value for month 6. This trial was registered with ClinicalTrials.gov, number NCT00091715. Findings Analyses were done with 168 patients (80 in the bosentan group, 88 in the placebo group) for pulmonary vascular resistance and with 177 (86 and 91) for 6-min walking distance. At month 6, geometric mean pulmonary vascular resistance was 83·2% (95% CI 73·8–93·7) of the baseline value in the bosentan group and 107·5% (97·6–118·4) of the baseline value in the placebo group (treatment effect −22·6%, 95% CI −33·5 to −10·0; p<0·0001). Mean 6-min walk distance increased from baseline in the bosentan group (11·2 m, 95% CI –4·6 to 27·0) and decreased in the placebo group (–7·9 m, –24·3 to 8·5), with a mean treatment effect of 19·1 m (95% CI 3·6–41·8; p=0·0758). 12 (13%) patients in the bosentan group and eight (9%) in the placebo group reported serious adverse events, the most common of which were syncope in the bosentan group and right ventricular failure in the placebo group. Interpretation Bosentan treatment could be beneficial for patients with WHO FC II pulmonary arterial hypertension. Funding Actelion Pharmaceuticals Ltd.
Introduction Pulmonary arterial hypertension is a devastating, progressive disease with increasingly debilitating symptoms.1,2 Increased pulmonary vascular resistance results in extensive heart structural changes, limits patients exercise capacity and, eventually, leads to right heart failure and death.2 Endothelin is a 21-aminoacid peptide that has a key role in the pathobiology of pulmonary arterial hypertension,3,4 exerting vasoconstrictor and mitogenic effects by binding to two distinct receptor isoforms in the pulmonary vascular smooth muscle cells: endothelin A and B receptors.5 Bosentan is an orally active dual (A and B) endothelin receptor antagonist that has been shown to improve exercise capacity, haemodynamics, and clinical worsening in two pivotal clinical trials.6,7 In these studies, as well as in other trials of pulmonary arterial hypertension with bosentan8,9 and other compounds,10–22 most enrolled patients were in an advanced symptomatic state as testified by WHO functional class (WHO FC) at baseline. In fact, the proportion of enrolled patients in WHO www.thelancet.com Vol 371 June 21, 2008
FC III and IV in these trials ranged from 60% to 100%.6–22 Observational studies suggest that there could be a clinical advantage to earlier initiation of treatment for pulmonary arterial hypertension.23,24 However, the effect of pulmonary arterial hypertension treatments has never been explored exclusively in less compromised individuals (ie, those in WHO FC I and II). We present the results of the endothelin antagonist trial in mildly symptomatic pulmonary arterial hypertension patients (EARLY) study, a randomised controlled trial designed to assess the efficacy of the dual endothelin receptor antagonist bosentan in patients with WHO FC II pulmonary arterial hypertension.
Lancet 2008; 371: 2093–100 See Comment page 2061 Institute of Cardiology, University of Bologna, Bologna, Italy (Prof N Galiè MD); University of California, San Diego, USA (Prof L J Rubin MD); Medizinische Hochschule, Hannover, Germany (Prof M M Hoeper MD); Charles University, 1st Faculty of Medicine, 2nd Medical Department, Clinical Department of Cardiology and Angiology, Prague, Czech Republic (P Jansa MD); Institut Clinic and Experimental Medicine, Clinic of Cardiology, Prague, Czech Republic (H Al-Hiti MD); Complexo Hospitalar Sta Casa de Porto Alegre, Porto Alegre, Brazil (G M B Meyer MD); Actelion Pharmaceuticals Ltd, Allschwil, Switzerland (E Chiossi, A Kusic-Pajic MD); and Hôpital Antoine Béclère, Université Paris-Sud, Clamart, France (Prof G Simonneau MD) Correspondence to: Prof N Galiè, Institute of Cardiology, University of Bologna, Via Massarenti 9, 40138-Bologna, Italy nazzareno.galie@unibo.it
Methods Patients EARLY was a prospective, randomised, double-blind, multicentre, parallel group study done in 52 sites in 21 countries. Patients aged 12 years or over were enrolled if diagnosed with WHO FC II idiopathic pulmonary arterial hypertension, familial pulmonary arterial hypertension, or 2093
Articles
Procedures 476 patients with pulmonary arterial hypertension initially screened
215 patients did not meet inclusion criteria 139 not WHO functional class II 57 unspecified 19 already treated with medications for pulmonary arterial hypertension
261 patients with WHO functional class II pulmonary arterial hypertension
76 patients ineligible due to exclusion criteria
185 patients randomised and treated
93 assigned to bosentan
92 assigned to placebo
12 patients discontinued 9 due to adverse events (1 death) 1 withdrew consent 2 protocol violations
10 patients discontinued 9 due to adverse events (1 death) 1 withdrew consent
81 patients completed study
82 patients completed study
Figure 1: Trial profile
pulmonary arterial hypertension associated with HIV infection, anorexigen use, atrial septal defect of less than 2 cm in diameter, ventricular septal defect of less than 1 cm in diameter, patent ductus arteriosus, or connective tissue or auto-immune diseases. Additionally, participants were required to either have a baseline 6-min walk distance of less than 80% of the normal predicted value25 or less than 500 m in combination with a Borg dyspnoea index of 2 points or more, and pulmonary vascular resistance of 320 dyn/s/cm⁵ or greater. Treatments approved for pulmonary arterial hypertension, including prostanoids and other endothelin receptor antagonist drugs, were prohibited. Because the oral phosphodiesterase-V inhibitor sildenafil received approval for treatment of pulmonary arterial hypertension during the period that this trial was ongoing, the protocol was amended to allow its use. Concomitant use of anticoagulants was allowed, as was use of calcium channel blockers, provided that they were present at least 1 month before randomisation and the dose remained steady throughout the study. Local institutional review boards or independent ethics committees approved the protocol and written informed consent was obtained from all patients. 2094
Patients were randomly assigned in a one to one ratio to receive either bosentan at an initial dose of 62·5 mg twice daily, up-titrating to 125 mg twice daily after 4 weeks (or remaining at 62·5 mg twice daily if bodyweight <40 kg) or placebo for a 6-month double-blind treatment period via a centralised integrated voice recognition system (Fisher Clinical Services, Allschwil, Switzerland and Allentown, PA, USA). This system assigned a unique randomisation number to each patient and designated the correct blinded study medication to be dispensed, both at the start of study treatment and at each scheduled visit. This code was accessible only to authorised individuals who were not involved in the conduct or analysis of the study, until the time of unblinding. Bosentan and its matching placebo were indistinguishable and were identically packaged. Treatment assignment was done in a blinded manner. All participants were to remain blinded to study treatment until closure of the double-blind clinical database. At the end of the randomised phase, patients had the option to enter an open-label extension phase. The primary efficacy endpoints were designed to test the effect of bosentan on improvement of cardiopulmonary haemodynamics and, as subordinate, on exercise capacity in patients with WHO FC II pulmonary arterial hypertension. Two co-primary endpoints were tested sequentially: first, pulmonary vascular resistance at rest at month 6, expressed as a percentage of the baseline value, and second, change from baseline to month 6 in 6-min walk distance. Pulmonary vascular resistance was included as an endpoint for its prognostic relevance23,24 and because haemodynamic changes in mildly symptomatic patients are expected to be more extensive than symptoms and exercise capacity, and was calculated as follows: pulmonary pulmonary capillary ×80 (mean arterial pressure – wedge pressure ( cardiac output Secondary endpoints included: time to clinical worsening; and change from baseline to month 6 in WHO FC, Borg dyspnoea index, total pulmonary resistance, mean pulmonary arterial pressure, cardiac index, and mixed venous oxygen saturation. Clinical worsening was defined as death of any cause (during the treatment period or as the outcome of a treatment-emergent adverse event that led to permanent discontinuation of study treatment), hospitalisation due to pulmonary arterial hypertension complications, or symptomatic progression of pulmonary arterial hypertension. Symptomatic progression of pulmonary arterial hypertension was defined as the presence of one of the following: appearance or worsening of right heart failure (as assessed by the investigator); decrease of 10% or more from baseline in two 6-min walk tests done 2 weeks or more apart; or 5% or greater decrease from www.thelancet.com Vol 371 June 21, 2008
Articles
baseline in two 6-min walk tests done 2 weeks or more apart associated with a 2-point or greater increase in Borg dyspnoea index. Exploratory endpoints included change from baseline to month 6 in N-terminal-prohormone-brain natriuretic peptide (NT-proBNP) and quality of life, assessed by the short-form health survey (SF-36).26 NT-proBNP assessments were done only in a prespecified subgroup of centres. Post-hoc analysis included assessment of the frequency of the individual components of clinical worsening. Safety information was collected during the double-blind treatment period and up to 28 days after the permanent discontinuation of study treatment. A data and safety monitoring committee was not established because no new safety signals have been identified since the approval of bosentan.27
Statistical analysis A sample size of 85 patients per treatment group was calculated on the basis of a reduction in the geometric mean for pulmonary vascular resistance of 20% or more and an increase of 35 m or more in the mean 6-min walk distance. A significant difference between the active and the placebo group could be determined for pulmonary vascular resistance with greater than 99% power and in 6-min walk distance with 91% power. The percentage of baseline at month 6 in pulmonary vascular resistance was assumed to be log-normally distributed with SD equal to 0·280 and the change from baseline to month 6 in 6-min walk distance was assumed to be normally distributed with SD equal to 65 m. The two primary endpoints were assessed hierarchically, with the 6-min walk distance endpoint to be tested only if the change in pulmonary vascular resistance was statistically significant. Both co-primary endpoints were tested at a two-sided type-I error of 0·05. This analysis strategy guarantees a study-wise type-I error of less than 0·05 despite multiple testing.28,29 The randomisation was stratified according to sildenafil use at enrolment. The main analysis on the primary endpoints was to be done on the all-randomised analysis set, which included all randomised patients. Analyses of pulmonary vascular resistance and 6-min walk distance included all randomised patients who had a valid baseline assessment and an assessment or an imputed value for month 6. Missing values at month 6 of primary and secondary endpoints were replaced with the last observation carried forward or, in case of clinical worsening, the worst substitution rule. For pulmonary vascular resistance, the worst value was calculated with either the highest absolute value observed in all patients or by applying the highest absolute increase observed to baseline, whichever was higher. For 6-min walk distance the worst value was calculated by applying the greatest percentage decrease observed to baseline, in the case of clinical worsening or with 0 m if death occurred. Patients who did not worsen and did not have a valid on-treatment value were excluded www.thelancet.com Vol 371 June 21, 2008
from the relevant analysis. These rules were defined before breaking the code. In the event of a significant result, several robustness analyses were done, applying different substitution rules. Correction for multiplicity of testing was only done for the two co-primary endpoints. The null hypothesis of no difference between bosentan and placebo was tested with the two-sided Wilcoxon rank-sum test (for the main analysis) and the two-sided t test (for supportive analyses). For the assessment of treatment effect on pulmonary vascular resistance, the ratio of the geometric means with its 95% CI was used, since the percentage of baseline is log-normally distributed. For clarity, the percentage difference between bosentan and placebo was reported with the following formula: (ratio–1)×100. The analysis of the co-primary endpoints of the subgroup of patients on sildenafil treatment was pre-specified. When comparing patients stratified according to sildenafil use, the 6-min walk distance was also presented as a median to reduce the effect of extreme substituted values in the small sample of patients on sildenafil. Bosentan group (n=93)
Placebo group (n=92)
Sex (male/female)
22 (24%)/71 (76%)
34 (37%)/58 (63%)
Age (years)
45·2 (17·9)
44·2 (16·5)
Weight (kg)
67·2 (16·3)
69·0 (15·7)
Ethnic origin (white/black/Asian)
87 (94%)/2 (2%)/4 (4%)
81 (88%)/3 (3%)/8 (9%)
Time from diagnosis (years)
2·9 (5·5)
3·7 (6·5)
Idiopathic pulmonary arterial hypertension
54 (58%)
58 (63%)
Congenital heart disease
16 (17%)
16 (17%)
Connective tissue disease
18 (19%)
15 (16%)
Systemic sclerosis
9 (10%)
6 (7%)
Lupus
7 (8%)
4 (4%)
Sarcoidosis
1 (1%)
0
Mixed connective tissue disease
0
3 (3%)
Sjögren’s syndrome
0
1 (1%)
Takayasu’s arteritis
0
1 (1%)
Other (not specified)
1 (1%)
0
HIV
5 (5%)
2 (2%)
Other
0
1 (1%)
Cause
Concomitant treatments Oral anticoagulants
55 (59%)
60 (65%)
Calcium channel blockers
29 (31%)
38 (41%)
Sildenafil
14 (15%)
15 (16%)
6-min walk distance (m)
438 (86)
431 (91)
PVR (dyn/sec/cm5)
839 (531)
805 (369)
Right atrial pressure (mm Hg) Pulmonary artery pressure (mm Hg) Cardiac index (L/min/m2) Mixed venous oxygen saturation (SVO2; %)
6·9 (4·5)
7·5 (5·1)*
52·5 (18·9)
52·3 (16·0)
2·7 (0·8)† 66·3 (8·5)‡
2·7 (0·6) 68·0 (7·7)§
Data are n (%) or mean (SD). *n=91; †n=92; ‡n=84; §n=83.
Table 1: Patient demographics and baseline characteristics (all-randomised set)
2095
Articles
Secondary endpoints were analysed on the allrandomised set. The percentage of worsening-free patients at each timepoint was estimated with the Kaplan-Meier method; treatment effect was reported with the hazard ratio of the bosentan versus the placebo group and tested with the log-rank test. The proportions of patients who improved or worsened in WHO FC and SF-36 health transition index were compared with relative risks (bosentan vs placebo) and the treatment effect was tested with Fisher’s exact test. All safety data were summarised descriptively, and safety analyses were done on the safety set, including all patients who were randomised, received at least one dose of study medication, and had at least one post-baseline safety assessment. This trial is registered with ClinicalTrials.gov, number NCT00091715. All statistical analyses were done with SAS software (version 8.2).
Role of the funding source The study sponsor was involved in the study design; and in the collection, analysis, and interpretation of data. All authors had full access to study data and had final responsibility for the decision to submit for publication.
Results
See Online for webappendix
The trial profile is shown in figure 1. Patient demographics, disease characteristics, exercise capacity, and haemodynamics were much the same between the treatment groups, except for a higher proportion of women in the bosentan group (table 1). Two patients were not uptitrated to the target dose of 125 mg twice daily: one died before reaching week 4 of treatment and the other had high concentrations of liver aminotransferases at baseline. Only one patient, with a concomitant drug addiction, was identified as non-
115 110
30
105
Mean change in 6-min walking distance (m)
Percentage of baseline pulmonary vascular resistance at month 6
120
100 95 90 85 80 75 70
Bosentan (n=80) Placebo (n=88)
Figure 2: Percentage of baseline pulmonary vascular resistance at 6 months Error bars are 95% CI.
2096
compliant but improved after this was recognised. Overall, 22 patients discontinued the study prematurely (figure 1). The most common reason for discontinuation in the bosentan group was adverse events, mainly abnormal liver function (six patients). Worsening pulmonary arterial hypertension was the most common reason for discontinuation in the placebo group (six patients). At 20 weeks, 84 (90%) patients in the bosentan group were still on study medication; 80 (86%) were still receiving the drug at 24 weeks. Due to the pre-specified rules of the primary analysis the following patients were not included in the primary analysis set: 13 patients from pulmonary vascular resistance and seven patients from 6-min walk distance analyses in the bosentan group, and four patients from pulmonary vascular resistance and one from 6-min walk distance analyses in the placebo group. Patients were excluded because they had missing baseline values, missing post-baseline values, or post-baseline values collected outside of the treatment period. The primary analysis set for pulmonary vascular resistance thus consisted of 168 patients (80 in the bosentan group, 88 in the placebo group). At month 6, geometric mean pulmonary vascular resistance was 83·2% (95% CI 73·8–93·7) of the baseline value in the bosentan group and 107·5% (97·6–118·4) of the baseline value in the placebo group (treatment effect –22·6%, 95% CI –33·5 to –10·0; p<0·0001, in favour of bosentan; figure 2). These results were confirmed by subsequent robustness analyses (webappendix). A similar treatment effect was seen in the strata of patients with or without concomitant sildenafil at baseline (with sildenafil: −20·4%, 95% CI −43·9 to 13·0, p=0·0478, 13 patients in bosentan group, 15 in placebo group; without sildenafil: −23·1%, −35·1 to –8·9, p<0·0001, 67 patients in bosentan group, 73 in placebo group). The primary analysis set for 6-min walk distance consisted of 177 patients (86 in the bosentan group, 91 in the placebo group). Mean 6-min walk distance increased in the bosentan group (11·2 m, 95% CI –4·6 to 27·0)
25
Bosentan (n=86) Placebo (n=91)
20 15 10 5 0 –5 –10 –15 –20 –25
3 months
6 months
Figure 3: 6-min walking distance Error bars are 95% CI.
www.thelancet.com Vol 371 June 21, 2008
and decreased in the placebo group (–7·9 m –24·3 to 8·5). However, the treatment effect on the change from baseline in 6-min walk distance was not statistically significant at month 6 (mean 19·1 m, 95% CI –3·6 to 41·8; p=0·0758, in favour of bosentan; figure 3). In patients with sildenafil, the mean bosentan treatment effect was –17·3 m (95% CI –105·7 to 71·1; p=0·8551; 13 patients in the bosentan group; 15 in the placebo group) and 25·7 m (3·8–47·6; p=0·0795; 73 patients in the bosentan group, 76 in the placebo group) in those without sildenafil. The median treatment effect with sildenafil was 5·0 m (95% CI –43·1 to 53·9) and 15·0 m (–1·6 to 32·2) without. The mean and median results differ because one patient on concomitant sildenafil in the bosentan group died and was thus deemed to have a result of 0 m. Similar treatment effects on pulmonary vascular resistance and 6-min walk distance were seen in patients with idiopathic pulmonary arterial hypertension as compared with the other causes of pulmonary arterial hypertension (data not shown), although the study was not powered to investigate subgroups. There was a delay in time to clinical worsening with bosentan compared with placebo (figure 4; hazard ratio 0·227, 95% CI 0·065–0·798; p=0·0114). Fewer patients in the bosentan group reached at least one of the components of clinical worsening by month 6 than did those in the placebo group (table 2). All 14 of the patients who survived at the first event of clinical worsening showed one or more of the following signs of deterioration: premature discontinuation from the trial, subsequent hospitalisation, worsening haemodynamics, or worsening exercise or functional capacity by the end of the study. Bosentan treatment was associated with a lower incidence of worsening functional class compared with placebo (3 [3·4%] patients in the bosentan group vs 12 [13·2%] in the placebo group, p=0·0285). Improvements in other haemodynamic variables were also noted (table 3). No significant changes were seen in Borg dyspnoea index (mean treatment effect –0·4, 95% CI –1·0 to 0·1; p=0·2599). A reduction in plasma concentrations of NT-pro-BNP was seen in the bosentan-treated group (n=65), compared with placebo (n=71; mean treatment effect –471 ng/L, 95% CI −749 to −192; p=0·0003). 43 (57%) patients on bosentan and 31 (38%) on placebo felt that their condition had improved, as assessed by SF-36 health transition index (p=0·0244, relative risk 1·50, 95% CI 1·07–2·10). The treatment effect of bosentan relative to placebo on the individual domains of the SF-36 questionnaire was as follows: physical functioning (3·1, 95% CI –2·5 to 8·7; p=0·2563); role-physical (7·7, –1·3 to 16·7; p=0·1285); pain index (1·6, –7·3 to 10·5; p=0·9281); general health perceptions (5·5, 0·2 to 10·8; p=0·0674); vitality (4·0, –2·2 to 10·2; p=0·1822); social functioning (1·4, –7·3 to 10·2; p=0·6883); mental health www.thelancet.com Vol 371 June 21, 2008
Patients with no clinical worsening (%)
Articles
100 80 60 40 20
Bosentan Placebo
0 0
Number at risk Placebo 92 Bosentan 93
4
8
12
16
20
24
28
32
18 27
9 15
Time from start of treatment (weeks) 90 92
89 87
86 85
84 84
83 83
77 80
Figure 4: Time to clinical worsening
Bosentan group Placebo group (n=93) (n=92) Clinical worsening
3 (3%)
13 (14%)
Symptomatic progression of PAH*† 1 (1%)
9 (10%)
Hospitalisation for PAH
1 (1%)
3 (3%)
Death
1 (1%)
1 (1%)
Data are n (%). PAH=pulmonary arterial hypertension. *Excluding hospitalisation. †Symptomatic progression defined as the presence of one of the following: appearance or worsening of right heart failure (as assessed by the investigator); decrease ≥10% from baseline in two 6-minute walk tests done ≥2 weeks apart; or ≥5% decrease from baseline in two 6-minute walk tests done ≥2 weeks apart associated with a ≥2-point increase in Borg dyspnoea index.
Table 2: Incidence of clinical worsening components
(4·3, –1·9 to 10·5; p=0·1577); and role-emotional (–0·4, –9·1 to 8·3; p=0·9383). 65 (70%) patients in the bosentan group had at least one adverse event compared with 60 (65%) in the placebo group. Overall, the number of adverse events was similar between groups during the 6-month study. The most commonly reported adverse events in the bosentan group were nasopharyngitis and abnormal liver function tests (table 4). The incidence of oedema was similar in the two groups. Laboratory tests identified increases in aminotransferases of more than three times the upper limit of normality in 12 (13%) patients on bosentan compared with two (2%) patients on placebo. Of the 12 instances of patients with aminotransferases over three times the upper limit of normality in the bosentan group, ten occurred during the first 20 weeks of treatment. These increases returned towards baseline values for all patients, either without intervention, on discontinuation, or after dose reduction of bosentan treatment. 20 patients reported serious adverse events, including those deemed to be unrelated to treatment: 12 (13%) in the bosentan group and eight (9%) in the placebo group. Two deaths occurred during the study period (including 28-day follow-up period after the end of treatment), one in each study group. The death recorded in the bosentan 2097
Articles
Mean (95% CI) change from baseline at month 6 Bosentan Right atrial pressure (mm Hg) Pulmonary artery pressure (mm Hg)
Treatment effect (95% CI; Wilcoxon p value)
Placebo
0·5 (–0·5 to 1·5)
1·1 (0·0 to 2·1)
–0·6 (–2·0 to 0·9; 0·662)
–2·7 (–6·4 to 1·1)
3·0 (–0·1 to 6·0)
–5·7 (–10·4 to –0·9; <0·0001)
Cardiac index (L/min/m2)
0·09 (–0·07 to 0·25)
–0·15 (–0·29 to –0·00)
0·24 (0·02 to 0·45; 0·025)
Mixed venous oxygen saturation (%)
1·2 (–0·8 to 3·3)
–3·5 (–5·5 to –1·6)
4·8 (1·9 to 7·6; 0·002)
Table 3: Treatment effects of bosentan at month 6 on other haemodynamic measures
group occurred on day 101 of treatment and was sudden, following a fall with subsequent headache, possibly due to the patient’s underlying antiphospholipid syndrome. In the placebo group, one patient died on treatment day 31 as a result of spontaneous spinal bleeding. Both patients had systemic lupus erythematosus. No individual serious adverse event was reported by more than two patients in either treatment group. Events reported by two patients included syncope in the bosentan group, and right ventricular failure in the placebo group. 157 patients (77 from the bosentan group and 80 from the placebo group) were included in the open-label extension study, in which all patients from both groups were given bosentan treatment. Over a median follow-up of 19·5 (range 0·1–37·9) months, seven of these patients experienced serious adverse events that were categorised as related or possibly related to bosentan treatment. These included one case of cardiac failure and anaemia, one case of pulmonary arterial hypertension progression, and three cases of abnormal liver function tests. All cases of abnormal liver function tests resolved after discontinuation of treatment. In the other two cases, the relation to trial drug was judged by the investigator to be unlikely: one death after seizures was caused by suspected reactivated vasculitis; the other case was an episode of toxic epidermal necrolysis and acute hepatitis in a patient on concomitant allopurinol. Bosentan (n=93)
Placebo (n=92)
Nasopharyngitis
7 (8%)
8 (9%)
Abnormal liver function test
7 (8%)
3 (3%)
Oedema (peripheral)
6 (6%)
7 (8%)
Nausea
5 (5%)
8 (9%)
Dizziness
5 (5%)
5 (5%)
Chest pain
5 (5%)
4 (4%)
Cough
4 (4%)
7 (8%)
Diarrhoea
2 (2%)
7 (8%)
Rhinitis
2 (2%)
6 (7%)
Worsening pulmonary hypertension* 1 (1%)
7 (8%)
Headache
9 (10%)
4 (4%)
*Individual investigator’s judgment.
Table 4: Adverse events that occurred in 5% or more of patients in either group
2098
Discussion This study, done exclusively in patients with mildly symptomatic (WHO FC II) pulmonary arterial hypertension, indicates that bosentan is associated with improvements in pulmonary vascular resistance compared with placebo. Fewer patients who received bosentan worsened clinically, including a worsening of functional class, than did those who received placebo. However, the drug had no significant effect on exercise capacity. The EARLY study differs from previous randomised controlled studies with bosentan6,7 in that it encompassed a longer period of randomised treatment (26 weeks vs 12–16 weeks) and because it was focused specifically on patients with mildly symptomatic pulmonary arterial hypertension. The baseline exercise capacity characteristics of the patient population enrolled in the EARLY study are different from those reported in previous pulmonary arterial hypertension studies, confirming that they were indeed less functionally compromised. In fact, baseline 6-min walk distance ranged from 314 m to 398 m in previous trials,11,18 compared with 435 m here. By contrast, mean pulmonary arterial pressure was remarkably similar to that seen in other pulmonary arterial hypertension trials.8–22 Despite the inclusion criteria in EARLY being designed to encourage the enrolment of less compromised patients, deterioration was still seen in the haemodynamic, hormonal, and clinical variables in the placebo-treated patients after 6 months (table 2 and figures 2, 3, and 4). Thus, the EARLY study provides valuable information on the natural history of pulmonary arterial hypertension and illustrates that it is a rapidly progressive disease, even in less advanced stages. The haemodynamic improvement in patients treated with bosentan, as shown by change in pulmonary vascular resistance (figure 2), was the result of an observed mean treatment effect in both mean pulmonary arterial pressure and cardiac index (table 3). This effect is of clinical relevance because the extent of the reduction of pulmonary vascular resistance after pulmonary arterial hypertension treatments is a predictor of survival in this population,23,24 and is comparable with the reduction seen in other pulmonary arterial hypertension trials enrolling more compromised patients.6–22 The effect of bosentan on haemodynamics is also reflected in the reduction of NT-proBNP plasma concentration in the bosentan group, www.thelancet.com Vol 371 June 21, 2008
Articles
compared with the increase seen in the placebo group, although these data were only collected in a subgroup of centres. A decrease in plasma concentration of brain natriuretic peptide after targeted pulmonary arterial hypertension therapy is also considered a predictive factor for patient survival.30,31 The favourable effect of bosentan treatment on pulmonary vascular resistance was also confirmed in the subgroup of patients treated with the phosphodiesterase-V inhibitor, sildenafil. This finding might indicate that use of sequential combination therapy in patients with pulmonary arterial hypertension classified as WHO FC II could provide advantages over monotherapy and warrants further study. The effect of bosentan on exercise capacity as assessed by 6-min walk distance at month 6 was not statistically significant (figure 3). The reason for these findings is not clear. One explanation could be that the higher baseline values of 6-min walk distance in the EARLY trial, compared with those in other trials, left less opportunity for improvement. In fact, previous studies in pulmonary arterial hypertension have shown that a lower baseline exercise capacity predicted a greater improvement after treatment, suggesting that the more compromised patients are, the greater the benefit of targeted therapies with regards to 6-min walk distance.14,20,32,33 Appropriate comparison with the treatment effects on 6-min walk distance in patients with WHO FC II pulmonary arterial hypertension observed in other trials is difficult because no baseline data for this specific subgroup are provided.14,15,17–22 Changes in WHO FC and time to clinical worsening (figure 4) showed that more patients remained stable without signs of deterioration in the bosentan group than in the placebo group. If the individual components of time to clinical worsening are analysed (table 2), both symptomatic progression of the disease and hospitalisations for pulmonary arterial hypertension were seen to be less frequent in the bosentan group than in the placebo group, while one death was observed in each group. All 14 patients who survived the first worsening event showed further signs of deterioration by the end of the study, testifying to the reliability of the definition of clinical worsening adopted in this trial. Prevention of clinical deterioration can be considered an important treatment goal, especially in patients with mildly symptomatic pulmonary arterial hypertension such as those enrolled in the EARLY study. Improvements in clinical worsening have also been seen in the two other controlled studies with bosentan done in more compromised pulmonary arterial hypertension patient populations6,7 as well as in other studies with different therapies.11,16,22 The results of the EARLY study suggest that clinical worsening could be used as a primary endpoint in future clinical trials of pulmonary arterial hypertension, in particular if mildly symptomatic patients are included. The improvement in the quality of life index seen in www.thelancet.com Vol 371 June 21, 2008
this study has been reported only sporadically in previous pulmonary arterial hypertension trials.14,16 Even with the recent development of pulmonary arterial hypertensionspecific questionnaires,34 the results achieved in the mildly symptomatic population of the EARLY study by the SF-36 could be particularly encouraging. The safety profile seen here was consistent with that reported with bosentan in previous pulmonary arterial hypertension studies,6,7,9 and the rate of adverse events was similar in both groups (table 4). Increases of liver aminotransferase concentrations were the most common adverse events with bosentan; such events were invariably reversible on reduction of the dose or permanent discontinuation of the drug. A more comprehensive assessment of the safety profile of bosentan will require a longer period of observation in a larger group of patients. The main limitation of this study was that the doubleblind study period might not have been long enough to detect a difference between bosentan and placebo in some endpoints, because of the less severe disease in these patients at baseline. A longer follow-up might be needed for this patient population. Additionally, and in keeping with other trials, the study was not powered to prospectively perform subgroup analyses, such as investigation of the effect of the aetiology of pulmonary arterial hypertension. In conclusion, the EARLY study shows that, if left untreated, mildly symptomatic pulmonary arterial hypertension can progressively deteriorate both clinically and in terms of haemodynamics, despite the maintenance of exercise capacity. Bosentan treatment in this patient population is associated with improvements in haemodynamics and prevention of clinical deterioration. These findings indicate that treatment with bosentan might be of benefit to patients with WHO FC II pulmonary arterial hypertension. Contributors NG and GS contributed to study design, study conduct, and data analysis, and were directly involved in the recruitment and care of the participants, and data collection. LR, MH, EC, and AK-P contributed to study design, study conduct, and data analysis. PJ, HA-H, and GMBM were principal investigators directly involved in the recruitment and care of the participants, and in data collection. All authors were involved in the writing of the manuscript, and saw and approved the final version. Conflict of interest statement NG, MH, LR, and GS were members of the EARLY steering committee. NG has participated in advisory board activities for Actelion, Pfizer, United Therapeutics, Eli-Lilly, Bayer-Schering, Encysive, and GlaxoSmithKline, given paid lectures for Actelion, Pfizer, Bayer-Schering, and Encysive, and his institution has received research grants from Actelion, Pfizer, United Therapeutics, Eli-Lilly, Bayer-Schering, Encysive, and GlaxoSmithKline. LR has received consulting fees for service as a steering committee member and adviser, and honoraria for lectures from Actelion Pharmaceuticals. MH has received speaker fees and honoraria for consultations from Actelion Pharmaceuticals as well as from Encysive Pharmaceuticals, Pfizer, and Bayer-Schering. PJ has received honoraria from Actelion Pharmaceuticals and fees from Actelion Pharmaceuticals, Pfizer, GlaxoSmithKline, and AOP Orphan Pharmaceuticals for consultancies. HA-H is a consultant for Actelion Pharmaceuticals and has received ad hoc honoraria for presentations from Pfizer, Schering,
2099
Articles
Bayer Pharma, Abbott, GlaxoSmithKline, Medtronic and AOP. GMBM has received travel grants to the 4th, 5th and 6th International Pulmonary Hypertension Forums from Actelion Pharmaceuticals. GS has received honoraria for consultation and/or lecture fees from Pfizer, GlaxoSmithKline, Actelion, Lilly, United Therapeutics, and Bayer, and industry sponsored grants from Pfizer, GlaxoSmithKline, Actelion, Lilly, and United Therapeutics. EC and AK-P are employees of Actelion Pharmaceuticals. Acknowledgments We thank Tom Newton for medical writing and editorial assistance during the preparation of this manuscript. References 1 Rubin LJ. Primary pulmonary hypertension. N Engl J Med 1997; 336: 111–17. 2 Galie N, Rubin L, eds. Pulmonary arterial hypertension. Epidemiology, pathobiology, assessment and therapy. J Am Coll Cardiol 2004; 43: S1–90. 3 Galie N, Manes A, Branzi A. The endothelin system in pulmonary arterial hypertension. Cardiovasc Res 2004; 61: 227–37. 4 Farber HW, Loscalzo J. Pulmonary arterial hypertension. N Engl J Med 2004; 351: 1655–65. 5 Davie N, Haleen SJ, Upton PD, et al. ET(A) and ET(B) receptors modulate the proliferation of human pulmonary artery smooth muscle cells. Am J Respir Crit Care Med 2002; 165: 398–405. 6 Channick RN, Simonneau G, Sitbon O, et al. Effects of the dual endothelin-receptor antagonist bosentan in patients with pulmonary hypertension: a randomised placebo-controlled study. Lancet 2001; 358: 1119–23. 7 Rubin LJ, Badesch DB, Barst RJ, et al. Bosentan therapy for pulmonary arterial hypertension. N Engl J Med 2002; 346: 896–903. 8 Humbert M, Barst RJ, Robbins IM, et al. Combination of bosentan with epoprostenol in pulmonary arterial hypertension: BREATHE-2. Eur Respir J 2004; 24: 353–59. 9 Galie N, Beghetti M, Gatzoulis MA, et al. Bosentan therapy in patients with Eisenmenger syndrome: a multicenter, double-blind, randomized, placebo-controlled study. Circulation 2006; 114: 48–54. 10 Rubin LJ, Mendoza J, Hood M, et al. Treatment of primary pulmonary hypertension with continuous intravenous prostacyclin (epoprostenol). Results of a randomized trial. Ann Intern Med 1990; 112: 485–91. 11 Barst RJ, Rubin LJ, Long WA, et al. A comparison of continuous intravenous epoprostenol (prostacyclin) with conventional therapy for primary pulmonary hypertension. N Engl J Med 1996; 334: 296–302. 12 Badesch DB, Tapson VF, McGoon MD, et al. Continuous intravenous epoprostenol for pulmonary hypertension due to the scleroderma spectrum of disease. A randomized, controlled trial. Ann Intern Med 2000; 132: 425–34. 13 Langleben D, Christman BW, Barst RJ, et al. Effects of the thromboxane synthetase inhibitor and receptor antagonist terbogrel in patients with primary pulmonary hypertension. Am Heart J 2002; 143: E4. 14 Simonneau G, Barst RJ, Galie N, et al. Continuous subcutaneous infusion of treprostinil, a prostacyclin analogue, in patients with pulmonary arterial hypertension: a double-blind, randomized, placebo-controlled trial. Am J Respir Crit Care Med 2002; 165: 800–04. 15 Galie N, Humbert M, Vachiery JL, et al. Effects of beraprost sodium, an oral prostacyclin analogue, in patients with pulmonary arterial hypertension: a randomized, double-blind, placebo-controlled trial. J Am Coll Cardiol 2002; 39: 1496–502.
2100
16 17
18
19
20 21
22
23
24
25 26 27
28 29
30
31
32
33
34
Olschewski H, Simonneau G, Galie N, et al. Inhaled iloprost for severe pulmonary hypertension. N Engl J Med 2002; 347: 322–29. Barst RJ, McGoon M, McLaughlin V, et al. Beraprost therapy for pulmonary arterial hypertension. J Am Coll Cardiol 2003; 41: 2119–25. Barst RJ, Langleben D, Frost A, et al. Sitaxsentan therapy for pulmonary arterial hypertension. Am J Respir Crit Care Med 2004; 169: 441–47. Sastry BK, Narasimhan C, Reddy NK, Raju BS. Clinical efficacy of sildenafil in primary pulmonary hypertension: a randomized, placebo-controlled, double-blind, crossover study. J Am Coll Cardiol 2004; 43: 1149–53. Galie N, Ghofrani HA, Torbicki A, et al. Sildenafil citrate therapy for pulmonary arterial hypertension. N Engl J Med 2005; 353: 2148–57. Barst RJ, Langleben D, Badesch D, et al. Treatment of pulmonary arterial hypertension with the selective endothelin-A receptor antagonist sitaxsentan. J Am Coll Cardiol 2006; 47: 2049–56. McLaughlin VV, Oudiz RJ, Frost A, et al. Randomized study of adding inhaled iloprost to existing bosentan in pulmonary arterial hypertension. Am J Respir Crit Care Med 2006; 174: 1257–63. Sitbon O, Humbert M, Nunes H, et al. Long-term intravenous epoprostenol infusion in primary pulmonary hypertension: prognostic factors and survival. J Am Coll Cardiol 2002; 40: 780–88. McLaughlin VV, Shillington A, Rich S. Survival in primary pulmonary hypertension: the impact of epoprostenol therapy. Circulation 2002; 106: 1477–82. Enright PL, Sherrill DL. Reference equations for the six-minute walk in healthy adults. Am J Respir Crit Care Med 1998; 158: 1384–87. Ware JEJ, Kosinski M, Gandek B. SF-36 health survey: manual & interpretation guide. Lincoln, RI: Qualitymetric Inc, 2000. Humbert M, Segal ES, Kiely DG, Carlsen J, Schwierin B, Hoeper MM. Results of European post-marketing surveillance of bosentan in pulmonary hypertension. Eur Respir J 2007; 30: 338–44. Chi GYH. Multiple testings: multiple comparisons and multiple endpoints. Drug Inf J 1998; 32: 1347–62S. European Agency for the Evaluation of Medicinal Products (EMEA). Points to consider on multiplicity issues in clinical trials (CPMP/EWP/908/99). http://www.emea.europa.eu/pdfs/ human/ewp/090899en.pdf (accessed Feb 13, 2008). Nagaya N, Nishikimi T, Uematsu M, et al. Plasma brain natriuretic peptide as a prognostic indicator in patients with primary pulmonary hypertension. Circulation 2000; 102: 865–70. Williams MH, Handler CE, Akram R, et al. Role of N-terminal brain natriuretic peptide (N-TproBNP) in scleroderma-associated pulmonary arterial hypertension. Eur Heart J 2006; 27: 1485–94. Barst RJ, Rubin LJ, McGoon MD, Caldwell EJ, Long WA, Levy PS. Survival in primary pulmonary hypertension with long-term continuous intravenous prostacyclin. Ann Intern Med 1994; 121: 409–15. Frost AE, Langleben D, Oudiz R, et al. The 6-min walk test (6MW) as an efficacy endpoint in pulmonary arterial hypertension clinical trials: demonstration of a ceiling effect. Vascul Pharmacol 2005; 43: 36–39. McKenna SP, Doughty N, Meads DM, Doward LC, Pepke-Zaba J. The Cambridge Pulmonary Hypertension Outcome Review (CAMPHOR): a measure of health-related quality of life and quality of life for patients with pulmonary hypertension. Qual Life Res 2006; 15: 103–15.
www.thelancet.com Vol 371 June 21, 2008